jiazhizhong/higress
1
0
Fork 0
mirror of https://github.com/alibaba/higress.git synced 2026-02-06 15:10:54 +08:00
Code Issues Packages Projects Releases Wiki Activity

update plugins doc (#1305)

Browse Source
This commit is contained in:
澄潭
2024-09-12 21:48:40 +08:00
committed by GitHub
parent 0f9113ed82
commit c7c4ae1da2
80 changed files with 7373 additions and 2368 deletions
Download Patch File Download Diff File Expand all files Collapse all files

106
plugins/wasm-cpp/extensions/basic_auth/README.md
View File

@@ -1,16 +1,30 @@
<p>
<a href="README_EN.md">English</a> | 中文
</p>
---
title: Basic 认证
keywords: [higress,basic auth]
description: Basic 认证插件配置参考
---
# 功能说明
## 功能说明
`basic-auth`插件实现了基于 HTTP Basic Auth 标准进行认证鉴权的功能
# 配置字段
## 运行属性
| 名称 | 数据类型 | 填写要求 | 默认值 | 描述 |
| ----------- | --------------- | -------- | ------ | ---------------------------------------------------- |
| `consumers` | array of object | 必填 | - | 配置服务的调用者，用于对请求进行认证 |
| `_rules_` | array of object | 选填 | - | 配置特定路由或域名的访问权限列表，用于对请求进行鉴权 |
插件执行阶段：`认证阶段`
插件执行优先级：`320`
## 配置字段
**注意：**
- 在一个规则里，鉴权配置和认证配置不可同时存在
- 对于通过认证鉴权的请求，请求的header会被添加一个`X-Mse-Consumer`字段，用以标识调用者的名称。
### 认证配置
| 名称 | 数据类型 | 填写要求 | 默认值 | 描述 |
| ----------- | --------------- | -------- | ------ | ---------------------------------------------------- |
| `global_auth` | bool | 选填（**仅实例级别配置**） | - | 只能在实例级别配置，若配置为true，则全局生效认证机制; 若配置为false，则只对做了配置的域名和路由生效认证机制，若不配置则仅当没有域名和路由配置时全局生效（兼容老用户使用习惯）。 |
| `consumers` | array of object | 必填 | - | 配置服务的调用者，用于对请求进行认证 |
`consumers`中每一项的配置字段说明如下：
@@ -19,51 +33,49 @@
| `credential` | string | 必填 | - | 配置该consumer的访问凭证 |
| `name` | string | 必填 | - | 配置该consumer的名称 |
`_rules_` 中每一项的配置字段说明如下：
### 鉴权配置（非必需）
| 名称 | 数据类型 | 填写要求 | 默认值 | 描述 |
| ---------------- | --------------- | ------------------------------------------------- | ------ | -------------------------------------------------- |
| `_match_route_` | array of string | 选填，`_match_route_`，`_match_domain_`中选填一项 | - | 配置要匹配的路由名称 |
| `_match_domain_` | array of string | 选填，`_match_route_`，`_match_domain_`中选填一项 | - | 配置要匹配的域名 |
| `allow` | array of string | 必填 | - | 对于符合匹配条件的请求，配置允许访问的consumer名称 |
**注意：**
- 若不配置`_rules_`字段，则默认对当前网关实例的所有路由开启认证；
- 对于通过认证鉴权的请求，请求的header会被添加一个`X-Mse-Consumer`字段，用以标识调用者的名称。
## 配置示例
# 配置示例
## 对特定路由或域名开启认证和鉴权
### 全局配置认证和路由粒度进行鉴权
以下配置将对网关特定路由或域名开启 Basic Auth 认证和鉴权，注意凭证信息中的用户名和密码之间使用":"分隔，`credential`字段不能重复
在实例级别做如下插件配置：
```yaml
# 使用 _rules_ 字段进行细粒度规则配置
consumers:
- credential: 'admin:123456'
name: consumer1
- credential: 'guest:abc'
name: consumer2
_rules_:
# 规则一：按路由名称匹配生效
- _match_route_:
- route-a
- route-b
allow:
- consumer1
# 规则二：按域名匹配生效
- _match_domain_:
- "*.example.com"
- test.com
allow:
- consumer2
global_auth: false
```
此例 `_match_route_` 中指定的 `route-a` 和 `route-b` 即在创建网关路由时填写的路由名称，当匹配到这两个路由时，将允许`name`为`consumer1`的调用者访问，其他调用者不允许访问；
对 route-a 和 route-b 这两个路由做如下配置：
此例 `_match_domain_` 中指定的 `*.example.com` 和 `test.com` 用于匹配请求的域名，当发现域名匹配时，将允许`name`为`consumer2`的调用者访问，其他调用者不允许访问。
```yaml
allow:
- consumer1
```
### 根据该配置，下列请求可以允许访问：
对 *.example.com 和 test.com 在这两个域名做如下配置:
```yaml
allow:
- consumer2
```
若是在控制台进行配置，此例指定的 `route-a` 和 `route-b` 即在控制台创建路由时填写的路由名称，当匹配到这两个路由时，将允许`name`为`consumer1`的调用者访问，其他调用者不允许访问；
此例指定的 `*.example.com` 和 `test.com` 用于匹配请求的域名，当发现域名匹配时，将允许`name`为`consumer2`的调用者访问，其他调用者不允许访问。
根据该配置，下列请求可以允许访问：
**请求指定用户名密码**
@@ -77,7 +89,7 @@ curl -H 'Authorization: Basic YWRtaW46MTIzNDU2' http://xxx.hello.com/test
认证鉴权通过后，请求的header中会被添加一个`X-Mse-Consumer`字段，在此例中其值为`consumer1`，用以标识调用方的名称
### 下列请求将拒绝访问：
下列请求将拒绝访问：
**请求未提供用户名密码，返回401**
```bash
@@ -93,22 +105,10 @@ curl -u admin:abc http://xxx.hello.com/test
curl -u guest:abc http://xxx.hello.com/test
```
## 网关实例级别开启
## 相关错误码
以下配置未指定`_rules_`字段，因此将对网关实例级别开启 Basic Auth 认证
```yaml
consumers:
- credential: 'admin:123456'
name: consumer1
- credential: 'guest:abc'
name: consumer2
```
# 相关错误码
| HTTP 状态码 | 出错信息 | 原因说明 |
| ----------- | ------------------------------------------------------------------------------ | ---------------------- |
| HTTP 状态码 | 出错信息 | 原因说明 |
| ----------- |--------------------------------------------------------------------------------| ---------------------- |
| 401 | Request denied by Basic Auth check. No Basic Authentication information found. | 请求未提供凭证 |
| 401 | Request denied by Basic Auth check. Invalid username and/or password | 请求凭证无效 |
| 403 | Request denied by Basic Auth check. Unauthorized consumer | 请求的调用方无访问权限 |
| 401 | Request denied by Basic Auth check. Invalid username and/or password. | 请求凭证无效 |
| 403 | Request denied by Basic Auth check. Unauthorized consumer. | 请求的调用方无访问权限 |

144
plugins/wasm-cpp/extensions/basic_auth/README_EN.md
View File

@@ -1,117 +1,99 @@
<p>
English | <a href="README.md">中文</a>
</p>
---
title: Basic Authentication
keywords: [higress,basic auth]
description: Basic authentication plugin configuration reference
---
## Function Description
The `basic-auth` plugin implements authentication and authorization based on the HTTP Basic Auth standard.
# Description
`basic-auth` plugin implements the function of authentication based on the HTTP Basic Auth standard.
# Configuration Fields
| Name | Type | Requirement | Default Value | Description |
| ----------- | --------------- | -------- | ------ | ---------------------------------------------------- |
| `consumers` | array of object | Required | - | Caller of the service for authentication of requests |
| `_rules_` | array of object | Optional | - | Configure access permission list for specific routes or domains to authenticate requests |
Filed descriptions of `consumers` items:
| Name | Type | Requirement | Default Value | Description |
| ------------ | ------ | ----------- | ------------- | ------------------------------------- |
| `credential` | string | Required | - | Credential for this consumer's access |
| `name` | string | Required | - | Name of this consumer |
Configuration field descriptions for each item in `_rules_` are as follows:
| Field Name | Data Type | Requirement | Default | Description |
| ---------------- | --------------- | ------------------------------------------------- | ------ | -------------------------------------------------- |
| `_match_route_` | array of string | One of `_match_route_` or `_match_domain_` | - | Configure the routes to match for request authorization |
| `_match_domain_` | array of string | One of `_match_route_` , `_match_domain_` | - | Configure the domains to match for request authorization |
| `allow` | array of string | Required | - | Configure the consumer names allowed to access requests that match the match condition |
## Operation Attributes
Plugin execution stage: `Authentication Phase`
Plugin execution priority: `320`
## Configuration Fields
**Note:**
- In one rule, authentication configurations and authorization configurations cannot coexist.
- For requests that pass authentication, the request header will include an `X-Mse-Consumer` field to identify the caller's name.
- If the `_rules_` field is not configured, authentication is enabled for all routes of the current gateway instance by default;
- For authenticated requests, `X-Mse-Consumer` field will be added to the request header to identify the name of the caller.
### Authentication Configuration
| Name | Data Type | Requirements | Default Value | Description |
| ------------- | ---------------- | ------------------------------- | -------------- | ------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------- |
| `global_auth` | bool | Optional (**instance-level only**) | - | Can only be configured at the instance level. If set to true, the authentication mechanism will take effect globally; if set to false, it will only take effect for the configured domains and routes. If not configured, it will only take effect globally when there are no domain and route configurations (compatible with old user habits). |
| `consumers` | array of object | Required | - | Configures the service callers for request authentication. |
# Configuration Samples
## Enable Authentication and Authorization for specific routes or domains
The following configuration will enable Basic Auth authentication and authorization for specific routes or domains of the gateway. Note that the username and password in the credential information are separated by a ":", and the `credential` field cannot be repeated.
Each configuration field in `consumers` is described as follows:
| Name | Data Type | Requirements | Default Value | Description |
| ------------ | --------- | ------------ | ------------- | ------------------------------- |
| `credential` | string | Required | - | Configures the access credentials for this consumer. |
| `name` | string | Required | - | Configures the name of this consumer. |
### Authorization Configuration (Optional)
| Name | Data Type | Requirements | Default Value | Description |
| ---------------- | ---------------- | ---------------------------------------------------- | -------------- | -------------------------------------------------------- |
| `allow` | array of string | Required | - | Configures the consumer names allowed to access for matching requests. |
## Configuration Example
### Global Authentication and Route Granularity Authorization
The following configuration will enable Basic Auth authentication and authorization for specific routes or domains of the gateway. Note that the username and password in the credential information are separated by ":", and the `credential` field cannot be duplicated.
Make the following plugin configuration at the instance level:
```yaml
# use the _rules_ field for fine-grained rule configuration.
consumers:
- credential: 'admin:123456'
name: consumer1
- credential: 'guest:abc'
name: consumer2
_rules_:
# rule 1: match by the route name.
- _match_route_:
- route-a
- route-b
allow:
- consumer1
# rule 2: match by the domain.
- _match_domain_:
- "*.example.com"
- test.com
allow:
- consumer2
global_auth: false
```
In this sample, `route-a` and `route-b` specified in `_match_route_` are the route names filled in when creating gateway routes. When these two routes are matched, the caller with `name` as `consumer1` is allowed to access, and other callers are not allowed to access.
The `*.example.com` and `test.com` specified in `_match_domain_` are used to match the domain name of the request. When the domain name is matched, the caller with `name` as `consumer2` is allowed to access, and other callers are not allowed to access.
For routes `route-a` and `route-b`, configure as follows:
```yaml
allow:
- consumer1
```
For the domains `*.example.com` and `test.com`, configure as follows:
```yaml
allow:
- consumer2
```
### According to this configuration, the following requests are allowed:
If configured in the console, the specified `route-a` and `route-b` refer to the route names filled in when creating the routes in the console. When matching these two routes, callers with the name `consumer1` will be allowed access, while other callers will not.
**Requests with specified username and password**
The specified `*.example.com` and `test.com` are used to match the request domain. When a match is found, callers with the name `consumer2` will be allowed access, while other callers will not.
Based on this configuration, the following requests may be allowed access:
**Request with specified username and password**
```bash
# Assuming the following request will match with route-a
# Use -u option of curl to specify the credentials
# Assuming the following request matches the route-a route
# Using curl's -u parameter to specify
curl -u admin:123456 http://xxx.hello.com/test
# Or specify the Authorization request header directly with the credentials in base64 encoding
# Or directly specify the Authorization request header with the username and password encoded in base64
curl -H 'Authorization: Basic YWRtaW46MTIzNDU2' http://xxx.hello.com/test
```
A `X-Mse-Consumer` field will be added to the headers of the request, and its value in this example is `consumer1`, used to identify the name of the caller when passed authentication and authorization.
After successful authentication, the request header will have an added `X-Mse-Consumer` field, which in this case is `consumer1` to identify the caller's name.
### The following requests will be denied:
**Requests without providing username and password, returning 401**
The following requests will be denied access:
**Request without username and password, returns 401**
```bash
curl http://xxx.hello.com/test
```
**Requests with incorrect username or password, returning 401**
**Request with incorrect username and password, returns 401**
```bash
curl -u admin:abc http://xxx.hello.com/test
```
**Requests matched with a caller who has no access permission, returning 403**
**Caller matched by username and password has no access, returns 403**
```bash
# consumer2 is not in the allow list of route-a
# consumer2 is not in the allow list for route-a
curl -u guest:abc http://xxx.hello.com/test
```
## Enable basic auth for gateway instance
The following configuration does not specify the `_rules_` field, so Basic Auth authentication will be effective for the whole gateway instance.
```yaml
consumers:
- credential: 'admin:123456'
name: consumer1
- credential: 'guest:abc'
name: consumer2
```
# Error Codes
| HTTP Status Code | Error Info | Reason |
| ----------- | ------------------------------------------------------------------------------ | ---------------------- |
| 401 | Request denied by Basic Auth check. No Basic Authentication information found. | Credentials not provided in the request |
| 401 | Request denied by Basic Auth check. Invalid username and/or password | Invalid username and/or password |
| 403 | Request denied by Basic Auth check. Unauthorized consumer | Unauthorized consumer |
## Related Error Codes
| HTTP Status Code | Error Message | Reason Description |
| ---------------- | ------------------------------------------------------------------------------------- | -------------------------------- |
| 401 | Request denied by Basic Auth check. No Basic Authentication information found. | Request did not provide credentials. |
| 401 | Request denied by Basic Auth check. Invalid username and/or password. | Request credentials are invalid. |
| 403 | Request denied by Basic Auth check. Unauthorized consumer. | The caller making the request does not have access. |

47
plugins/wasm-cpp/extensions/bot_detect/README.md
View File

@@ -1,11 +1,21 @@
<p>
<a href="README_EN.md"> English </a> | 中文
</p>
---
title: Bot 拦截
keywords: [higress,bot detect]
description: Bot 拦截插件配置参考
---
## 功能说明
# 功能说明
`bot-detect`插件可以用于识别并阻止互联网爬虫对站点资源的爬取
# 配置字段
## 运行属性
插件执行阶段：`授权阶段`
插件执行优先级：`310`
## 配置字段
| 名称 | 数据类型 | 填写要求 | 默认值 | 描述 |
| -------- | -------- | -------- | -------- | -------- |
@@ -33,9 +43,9 @@
(CSimpleSpider|Cityreview Robot|CrawlDaddy|CrawlFire|Finderbots|Index crawler|Job Roboter|KiwiStatus Spider|Lijit Crawler|QuerySeekerSpider|ScollSpider|Trends Crawler|USyd-NLP-Spider|SiteCat Webbot|BotName\/\$BotVersion|123metaspider-Bot|1470\.net crawler|50\.nu|8bo Crawler Bot|Aboundex|Accoona-[A-z]{1,30}-Agent|AdsBot-Google(?:-[a-z]{1,30}|)|altavista|AppEngine-Google|archive.{0,30}\.org_bot|archiver|Ask Jeeves|[Bb]ai[Dd]u[Ss]pider(?:-[A-Za-z]{1,30})(?:-[A-Za-z]{1,30}|)|bingbot|BingPreview|blitzbot|BlogBridge|Bloglovin|BoardReader Blog Indexer|BoardReader Favicon Fetcher|boitho.com-dc|BotSeer|BUbiNG|\b\w{0,30}favicon\w{0,30}\b|\bYeti(?:-[a-z]{1,30}|)|Catchpoint(?: bot|)|[Cc]harlotte|Checklinks|clumboot|Comodo HTTP\(S\) Crawler|Comodo-Webinspector-Crawler|ConveraCrawler|CRAWL-E|CrawlConvera|Daumoa(?:-feedfetcher|)|Feed Seeker Bot|Feedbin|findlinks|Flamingo_SearchEngine|FollowSite Bot|furlbot|Genieo|gigabot|GomezAgent|gonzo1|(?:[a-zA-Z]{1,30}-|)Googlebot(?:-[a-zA-Z]{1,30}|)|Google SketchUp|grub-client|gsa-crawler|heritrix|HiddenMarket|holmes|HooWWWer|htdig|ia_archiver|ICC-Crawler|Icarus6j|ichiro(?:/mobile|)|IconSurf|IlTrovatore(?:-Setaccio|)|InfuzApp|Innovazion Crawler|InternetArchive|IP2[a-z]{1,30}Bot|jbot\b|KaloogaBot|Kraken|Kurzor|larbin|LEIA|LesnikBot|Linguee Bot|LinkAider|LinkedInBot|Lite Bot|Llaut|lycos|Mail\.RU_Bot|masscan|masidani_bot|Mediapartners-Google|Microsoft .{0,30} Bot|mogimogi|mozDex|MJ12bot|msnbot(?:-media {0,2}|)|msrbot|Mtps Feed Aggregation System|netresearch|Netvibes|NewsGator[^/]{0,30}|^NING|Nutch[^/]{0,30}|Nymesis|ObjectsSearch|OgScrper|Orbiter|OOZBOT|PagePeeker|PagesInventory|PaxleFramework|Peeplo Screenshot Bot|PlantyNet_WebRobot|Pompos|Qwantify|Read%20Later|Reaper|RedCarpet|Retreiver|Riddler|Rival IQ|scooter|Scrapy|Scrubby|searchsight|seekbot|semanticdiscovery|SemrushBot|Simpy|SimplePie|SEOstats|SimpleRSS|SiteCon|Slackbot-LinkExpanding|Slack-ImgProxy|Slurp|snappy|Speedy Spider|Squrl Java|Stringer|TheUsefulbot|ThumbShotsBot|Thumbshots\.ru|Tiny Tiny RSS|Twitterbot|WhatsApp|URL2PNG|Vagabondo|VoilaBot|^vortex|Votay bot|^voyager|WASALive.Bot|Web-sniffer|WebThumb|WeSEE:[A-z]{1,30}|WhatWeb|WIRE|WordPress|Wotbox|www\.almaden\.ibm\.com|Xenu(?:.s|) Link Sleuth|Xerka [A-z]{1,30}Bot|yacy(?:bot|)|YahooSeeker|Yahoo! Slurp|Yandex\w{1,30}|YodaoBot(?:-[A-z]{1,30}|)|YottaaMonitor|Yowedo|^Zao|^Zao-Crawler|ZeBot_www\.ze\.bz|ZooShot|ZyBorg)(?:[ /]v?(\d+)(?:\.(\d+)(?:\.(\d+)|)|)|)
```
# 配置示例
## 配置示例
## 放行原本命中爬虫规则的请求
### 放行原本命中爬虫规则的请求
```yaml
allow:
- ".*Go-http-client.*"
@@ -44,7 +54,7 @@ allow:
若不作该配置，默认的 Golang 网络库请求会被视做爬虫，被禁止访问
## 增加爬虫判断
### 增加爬虫判断
```yaml
deny:
- "spd-tools.*"
@@ -56,24 +66,3 @@ deny:
curl http://example.com -H 'User-Agent: spd-tools/1.1'
curl http://exmaple.com -H 'User-Agent: spd-tools'
```
## 对特定路由或域名开启
```yaml
# 使用 _rules_ 字段进行细粒度规则配置
_rules_:
# 规则一：按路由名称匹配生效
- _match_route_:
- route-a
- route-b
# 规则二：按域名匹配生效
- _match_domain_:
- "*.example.com"
- test.com
allow:
- ".*Go-http-client.*"
```
此例 `_match_route_` 中指定的 `route-a` 和 `route-b` 即在创建网关路由时填写的路由名称，当匹配到这两个路由时，将使用此段配置；
此例 `_match_domain_` 中指定的 `*.example.com` 和 `test.com` 用于匹配请求的域名，当发现域名匹配时，将使用此段配置；
配置的匹配生效顺序，将按照 `_rules_` 下规则的排列顺序，匹配第一个规则后生效对应配置，后续规则将被忽略。

66
plugins/wasm-cpp/extensions/bot_detect/README_EN.md
View File

@@ -1,22 +1,26 @@
<p>
English | <a href="README.md">中文</a>
</p>
---
title: Bot Detect
keywords: [higress, bot detect]
description: Bot detect plugin configuration reference
---
## Function Description
The `bot-detect` plugin can be used to identify and block internet crawlers from accessing site resources.
# Description
`bot-detect` plugin can be used to identify and prevent web crawlers from crawling websites.
## Running Properties
Plugin Execution Phase: `Authorization Phase`
Plugin Execution Priority: `310`
# Configuration Fields
## Configuration Fields
| Name | Data Type | Required | Default Value | Description |
| ----------------- | ------------------- | --------------| --------------| ---------------------------------------------------------- |
| allow | array of string | Optional | - | Regular expressions to match the User-Agent request header; requests matching will be allowed to access. |
| deny | array of string | Optional | - | Regular expressions to match the User-Agent request header; requests matching will be blocked. |
| blocked_code | number | Optional | 403 | HTTP status code returned when a request is blocked. |
| blocked_message | string | Optional | - | HTTP response body returned when a request is blocked. |
| Name | Type | Requirement | Default Value | Description |
| -------- | -------- | -------- | -------- | -------- |
| allow | array of string | Optional | - | A regular expression to match the User-Agent request header and will allow access if the match hits |
| deny | array of string | Optional | - | A regular expression to match the User-Agent request header and will block the request if the match hits |
| blocked_code | number | Optional | 403 | The HTTP status code returned when a request is blocked |
| blocked_message | string | Optional | - | The HTTP response Body returned when a request is blocked |
The `allow` and `deny` fields can both be left unconfigured, in which case the default crawler identification logic will be executed. Configuring the `allow` field can allow requests that would otherwise hit the default crawler identification logic. Configuring the `deny` field can add additional crawler identification logic.
If field `allow` and field `deny` are not configured at the same time, the default logic to identify crawlers will be executed. By configuring the `allow` field, requests that would otherwise hit the default logic can be allowed. The judgement can be extended by configuring the `deny` field
The default set of crawler judgment regular expressions is as follows：
The default crawler identification regular expression set is as follows:
```bash
# Bots General matcher 'name/0.0'
@@ -33,45 +37,23 @@ The default set of crawler judgment regular expressions is as follows：
(CSimpleSpider|Cityreview Robot|CrawlDaddy|CrawlFire|Finderbots|Index crawler|Job Roboter|KiwiStatus Spider|Lijit Crawler|QuerySeekerSpider|ScollSpider|Trends Crawler|USyd-NLP-Spider|SiteCat Webbot|BotName\/\$BotVersion|123metaspider-Bot|1470\.net crawler|50\.nu|8bo Crawler Bot|Aboundex|Accoona-[A-z]{1,30}-Agent|AdsBot-Google(?:-[a-z]{1,30}|)|altavista|AppEngine-Google|archive.{0,30}\.org_bot|archiver|Ask Jeeves|[Bb]ai[Dd]u[Ss]pider(?:-[A-Za-z]{1,30})(?:-[A-Za-z]{1,30}|)|bingbot|BingPreview|blitzbot|BlogBridge|Bloglovin|BoardReader Blog Indexer|BoardReader Favicon Fetcher|boitho.com-dc|BotSeer|BUbiNG|\b\w{0,30}favicon\w{0,30}\b|\bYeti(?:-[a-z]{1,30}|)|Catchpoint(?: bot|)|[Cc]harlotte|Checklinks|clumboot|Comodo HTTP\(S\) Crawler|Comodo-Webinspector-Crawler|ConveraCrawler|CRAWL-E|CrawlConvera|Daumoa(?:-feedfetcher|)|Feed Seeker Bot|Feedbin|findlinks|Flamingo_SearchEngine|FollowSite Bot|furlbot|Genieo|gigabot|GomezAgent|gonzo1|(?:[a-zA-Z]{1,30}-|)Googlebot(?:-[a-zA-Z]{1,30}|)|Google SketchUp|grub-client|gsa-crawler|heritrix|HiddenMarket|holmes|HooWWWer|htdig|ia_archiver|ICC-Crawler|Icarus6j|ichiro(?:/mobile|)|IconSurf|IlTrovatore(?:-Setaccio|)|InfuzApp|Innovazion Crawler|InternetArchive|IP2[a-z]{1,30}Bot|jbot\b|KaloogaBot|Kraken|Kurzor|larbin|LEIA|LesnikBot|Linguee Bot|LinkAider|LinkedInBot|Lite Bot|Llaut|lycos|Mail\.RU_Bot|masscan|masidani_bot|Mediapartners-Google|Microsoft .{0,30} Bot|mogimogi|mozDex|MJ12bot|msnbot(?:-media {0,2}|)|msrbot|Mtps Feed Aggregation System|netresearch|Netvibes|NewsGator[^/]{0,30}|^NING|Nutch[^/]{0,30}|Nymesis|ObjectsSearch|OgScrper|Orbiter|OOZBOT|PagePeeker|PagesInventory|PaxleFramework|Peeplo Screenshot Bot|PlantyNet_WebRobot|Pompos|Qwantify|Read%20Later|Reaper|RedCarpet|Retreiver|Riddler|Rival IQ|scooter|Scrapy|Scrubby|searchsight|seekbot|semanticdiscovery|SemrushBot|Simpy|SimplePie|SEOstats|SimpleRSS|SiteCon|Slackbot-LinkExpanding|Slack-ImgProxy|Slurp|snappy|Speedy Spider|Squrl Java|Stringer|TheUsefulbot|ThumbShotsBot|Thumbshots\.ru|Tiny Tiny RSS|Twitterbot|WhatsApp|URL2PNG|Vagabondo|VoilaBot|^vortex|Votay bot|^voyager|WASALive.Bot|Web-sniffer|WebThumb|WeSEE:[A-z]{1,30}|WhatWeb|WIRE|WordPress|Wotbox|www\.almaden\.ibm\.com|Xenu(?:.s|) Link Sleuth|Xerka [A-z]{1,30}Bot|yacy(?:bot|)|YahooSeeker|Yahoo! Slurp|Yandex\w{1,30}|YodaoBot(?:-[A-z]{1,30}|)|YottaaMonitor|Yowedo|^Zao|^Zao-Crawler|ZeBot_www\.ze\.bz|ZooShot|ZyBorg)(?:[ /]v?(\d+)(?:\.(\d+)(?:\.(\d+)|)|)|)
```
# Configuration Samples
## Release Requests that would otherwise Hit the Crawler Rules
## Configuration Example
### Allowing Requests That Hit the Crawler Rules
```yaml
allow:
- ".*Go-http-client.*"
```
Without this configuration, the default Golang web library request will be treated as a crawler and access will be denied.
If this configuration is not made, requests from the default Golang network library will be treated as crawlers and blocked.
## Add Crawler Judgement
### Adding Crawler Identification
```yaml
deny:
- "spd-tools.*"
```
According to this configuration, the following requests will be denied:
With this configuration, the following requests will be blocked:
```bash
curl http://example.com -H 'User-Agent: spd-tools/1.1'
curl http://exmaple.com -H 'User-Agent: spd-tools'
```
## Only Enabled for Specific Routes or Domains
```yaml
# Use _rules_ field for fine-grained rule configurations
_rules_:
# Rule 1: Match by route name
- _match_route_:
- route-a
- route-b
# Rule 2: Match by domain
- _match_domain_:
- "*.example.com"
- test.com
allow:
- ".*Go-http-client.*"
```
In the rule sample of `_match_route_`, `route-a` and `route-b` are the route names provided when creating a new gateway route. When the current route names matches the configuration, the rule following shall be applied.
In the rule sample of `_match_domain_`, `*.example.com` and `test.com` are the domain names used for request matching. When the current domain name matches the configuration, the rule following shall be applied.
All rules shall be checked following the order of items in the `_rules_` field, The first matched rule will be applied. All remained will be ignored.

48
plugins/wasm-cpp/extensions/custom_response/README.md
View File

@@ -1,11 +1,19 @@
<p>
<a href="README_EN.md"> English </a> | 中文
</p>
---
title: 自定义应答
keywords: [higress,customn response]
description: 自定义应答插件配置参考
---
# 功能说明
## 功能说明
`custom-response`插件支持配置自定义的响应，包括自定义 HTTP 应答状态码、HTTP 应答头，以及 HTTP 应答 Body。可以用于 Mock 响应，也可以用于判断特定状态码后给出自定义应答，例如在触发网关限流策略时实现自定义响应。
# 配置字段
## 运行属性
插件执行阶段：`认证阶段`
插件执行优先级：`910`
## 配置字段
| 名称 | 数据类型 | 填写要求 | 默认值 | 描述 |
| -------- | -------- | -------- | -------- | -------- |
@@ -14,9 +22,9 @@
| body | string | 选填 | - | 自定义 HTTP 应答 Body |
| enable_on_status | array of number | 选填 | - | 匹配原始状态码，生成自定义响应，不填写时，不判断原始状态码 |
# 配置示例
## 配置示例
## Mock 应答场景
### Mock 应答场景
```yaml
status_code: 200
@@ -38,7 +46,7 @@ Content-Length: 17
{"hello":"world"}
```
## 触发限流时自定义响应
### 触发限流时自定义响应
```yaml
enable_on_status:
@@ -58,27 +66,3 @@ Location: https://example.com
从而实现基于浏览器 302 重定向机制，将限流后的用户引导到其他页面，比如可以是一个 CDN 上的静态页面。
如果希望触发限流时，正常返回其他应答，参考 Mock 应答场景配置相应的字段即可。
## 对特定路由或域名开启
```yaml
# 使用 _rules_ 字段进行细粒度规则配置
_rules_:
# 规则一：按路由名称匹配生效
- _match_route_:
- route-a
- route-b
body: "{\"hello\":\"world\"}"
# 规则二：按域名匹配生效
- _match_domain_:
- "*.example.com"
- test.com
enable_on_status:
- 429
status_code: 200
headers:
- Content-Type=application/json
body: "{\"errmsg\": \"rate limited\"}"
```
此例 `_match_route_` 中指定的 `route-a` 和 `route-b` 即在创建网关路由时填写的路由名称，当匹配到这两个路由时，将使用此段配置；
此例 `_match_domain_` 中指定的 `*.example.com` 和 `test.com` 用于匹配请求的域名，当发现域名匹配时，将使用此段配置；
配置的匹配生效顺序，将按照 `_rules_` 下规则的排列顺序，匹配第一个规则后生效对应配置，后续规则将被忽略。

74
plugins/wasm-cpp/extensions/custom_response/README_EN.md
View File

@@ -1,84 +1,54 @@
<p>
English | <a href="README.md">中文</a>
</p>
---
title: Custom Response
keywords: [higress, custom response]
description: Custom response plugin configuration reference
---
## Function Description
The `custom-response` plugin supports the configuration of custom responses, including custom HTTP response status codes, HTTP response headers, and HTTP response bodies. It can be used for Mock responses or for providing custom responses based on specific status codes, such as implementing custom responses when triggering the gateway rate-limiting policy.
# Description
`custom-response` plugin implements a function of sending custom responses, including custom HTTP response status codes, HTTP response headers and HTTP response body, which can be used in the scenarios of response mocking and sending a custom response for specific status codes, such as customizing the response for rate-limited requests.
## Running Attributes
Plugin Execution Phase: `Authentication Phase`
# Configuration Fields
Plugin Execution Priority: `910`
| Name | Type | Requirement | Default Value | Description |
## Configuration Fields
| Name | Data Type | Requirements | Default Value | Description |
| -------- | -------- | -------- | -------- | -------- |
| status_code | number | Optional | 200 | Custom HTTP response status code |
| headers | array of string | Optional | - | Custom HTTP response header. Key and value shall be separated using `=`. |
| headers | array of string | Optional | - | Custom HTTP response headers, keys and values separated by `=` |
| body | string | Optional | - | Custom HTTP response body |
| enable_on_status | array of number | Optional | - | The original response status code to match. Generate the custom response only the actual status code matches the configuration. Ignore the status code match if left unconfigured. |
# Configuration Samples
## Mock Responses
| enable_on_status | array of number | Optional | - | Match original status codes to generate custom responses; if not specified, the original status code is not checked |
## Configuration Example
### Mock Response Scenario
```yaml
status_code: 200
headers:
- Content-Type=application/json
- Hello=World
body: "{\"hello\":\"world\"}"
```
According to the configuration above, all the requests will get the following custom response:
With this configuration, the request will return the following custom response:
```text
HTTP/1.1 200 OK
Content-Type: application/json
Hello: World
Content-Length: 17
{"hello":"world"}
```
## Send a Custom Response when Rate-Limited
### Custom Response on Rate Limiting
```yaml
enable_on_status:
enable_on_status:
- 429
status_code: 302
headers:
- Location=https://example.com
```
When rate-limited, normally gateway will return a status code of `429` . Now, rate-limited requests will get the following custom response:
When the gateway rate limiting is triggered, it generally returns the `429` status code, and the request will return the following custom response:
```text
HTTP/1.1 302 Found
Location: https://example.com
```
This achieves the goal of redirecting users who have been rate-limited to another page based on the browser's 302 redirect mechanism, which could be a static page on a CDN.
So based on the 302 redirecting mechanism provided by browsers, this can redirect rate-limited users to other pages, for example, a static page hosted on CDN.
If you'd like to send other responses when rate-limited, please add other fields into the configuration, referring to the Mock Responses scenario.
## Only Enabled for Specific Routes or Domains
```yaml
# Use _rules_ field for fine-grained rule configurations
_rules_:
# Rule 1: Match by route name
- _match_route_:
- route-a
- route-b
body: "{\"hello\":\"world\"}"
# Rule 2: Match by domain
- _match_domain_:
- "*.example.com"
- test.com
enable_on_status:
- 429
status_code: 200
headers:
- Content-Type=application/json
body: "{\"errmsg\": \"rate limited\"}"
```
In the rule sample of `_match_route_`, `route-a` and `route-b` are the route names provided when creating a new gateway route. When the current route names matches the configuration, the rule following shall be applied.
In the rule sample of `_match_domain_`, `*.example.com` and `test.com` are the domain names used for request matching. When the current domain name matches the configuration, the rule following shall be applied.
All rules shall be checked following the order of items in the `_rules_` field, The first matched rule will be applied. All remained will be ignored.
If you wish to return other responses normally when rate limiting is triggered, just refer to the Mock response scenario to configure the relevant fields accordingly.

119
plugins/wasm-cpp/extensions/hmac_auth/README.md
View File

@@ -1,13 +1,32 @@
# 功能说明
---
title: HMAC 认证
keywords: [higress,hmac auth]
description: HMAC 认证插件配置参考
---
## 功能说明
`hmac-auth`插件实现了基于 HMAC 算法为 HTTP 请求生成不可伪造的签名，并基于签名实现身份认证和鉴权
# 配置字段
## 运行属性
插件执行阶段：`认证阶段`
插件执行优先级：`330`
## 配置字段
**注意：**
- 在一个规则里，鉴权配置和认证配置不可同时存在
- 对于通过认证鉴权的请求，请求的header会被添加一个`X-Mse-Consumer`字段，用以标识调用者的名称。
### 认证配置
| 名称 | 数据类型 | 填写要求 | 默认值 | 描述 |
| ------------- | --------------- | -------- | ------ | ------------------------------------------------------------------------------------------------------------------- |
| `global_auth` | bool | 选填（**仅实例级别配置**） | - | 只能在实例级别配置，若配置为true，则全局生效认证机制; 若配置为false，则只对做了配置的域名和路由生效认证机制，若不配置则仅当没有域名和路由配置时全局生效（兼容老用户使用习惯）。 |
| `consumers` | array of object | 必填 | - | 配置服务的调用者，用于对请求进行认证 |
| `date_offset` | number | 选填 | - | 配置允许的客户端最大时间偏移，单位为秒，根据请求头`Date`解析客户端 UTC 时间，可用于避免请求重放；未配置时，不做校验 |
| `_rules_` | array of object | 选填 | - | 配置特定路由或域名的访问权限列表，用于对请求进行鉴权 |
`consumers`中每一项的配置字段说明如下：
@@ -17,24 +36,21 @@
| `secret` | string | 必填 | - | 配置用于生成签名的secret |
| `name` | string | 必填 | - | 配置该consumer的名称 |
`_rules_` 中每一项的配置字段说明如下：
### 鉴权配置（非必需）
| 名称 | 数据类型 | 填写要求 | 默认值 | 描述 |
| ---------------- | --------------- | ------------------------------------------------- | ------ | -------------------------------------------------- |
| `_match_route_` | array of string | 选填，`_match_route_`，`_match_domain_`中选填一项 | - | 配置要匹配的路由名称 |
| `_match_domain_` | array of string | 选填，`_match_route_`，`_match_domain_`中选填一项 | - | 配置要匹配的域名 |
| `allow` | array of string | 必填 | - | 对于符合匹配条件的请求，配置允许访问的consumer名称 |
| 名称 | 数据类型 | 填写要求 | 默认值 | 描述 |
| ----------- | --------------- | ------------------------------------------- | ------ | ----------------------------------------------------------- |
| `allow` | array of string | 选填(**非实例级别配置**) | - | 只能在路由或域名等细粒度规则上配置，对于符合匹配条件的请求，配置允许访问的 consumer，从而实现细粒度的权限控制 |
**注意：**
- 若不配置`_rules_`字段，则默认对当前网关实例的所有路由开启认证；
- 对于通过认证鉴权的请求，请求的header会被添加一个`X-Mse-Consumer`字段，用以标识调用者的名称。
# 配置示例
## 配置示例
以下配置将对网关特定路由或域名开启 Hmac Auth 认证和鉴权，注意`key`字段不能重复
### 全局配置认证和路由粒度进行鉴权
在实例级别做如下插件配置, 注意`key`字段不能重复:
## 对特定路由或域名开启
```yaml
global_auth: false
consumers:
- key: appKey-example-1
secret: appSecret-example-1
@@ -42,34 +58,33 @@ consumers:
- key: appKey-example-2
secret: appSecret-example-2
name: consumer-2
# 使用 _rules_ 字段进行细粒度规则配置
_rules_:
# 规则一：按路由名称匹配生效
- _match_route_:
- route-a
- route-b
allow:
- consumer-1
# 规则二：按域名匹配生效
- _match_domain_:
- "*.example.com"
- test.com
allow:
- consumer-2
```
每条匹配规则下的`allow`字段用于指定该匹配条件下允许访问的调用者列表；
此例 `_match_route_` 中指定的 `route-a` 和 `route-b` 即在创建网关路由时填写的路由名称，当匹配到这两个路由时，将允许`name`为`consumer-1`的调用者访问，其他调用者不允许访问；
此例 `_match_domain_` 中指定的 `*.example.com` 和 `test.com` 用于匹配请求的域名，当发现域名匹配时，将允许`name`为`consumer-2`的调用者访问，其他调用者不允许访问；
认证成功后，请求的header中会被添加一个`X-Mse-Consumer`字段，其值为调用方的名称，例如`consumer-1`。
## 网关实例级别开启
以下配置将对网关实例级别开启 Hamc Auth 认证
route-a和route-b两个路由做如下插件配置：
```yaml
allow:
- consumer1
```
在*.example.com和test.com两个域名做如下插件配置：
```yaml
allow:
- consumer2
```
若是在控制台进行配置，此例指定的route-a和route-b即在创建网关路由时填写的路由名称，当匹配到这两个路由时，将允许name为consumer1的调用者访问，其他调用者不允许访问。
此例指定的*.example.com和test.com用于匹配请求的域名，当发现域名匹配时，将允许name为consumer2的调用者访问，其他调用者不被允许访问。
### 网关实例级别开启
以下配置将对网关实例级别开启 Hamc Auth 认证，所有请求均需要经过认证后才能访问。
```yaml
global_auth: true
consumers:
- key: appKey-example-1
secret: appSecret-example-1
@@ -80,18 +95,18 @@ consumers:
```
# 签名机制说明
## 签名机制说明
## 配置准备
### 配置准备
如上指引，在插件配置中配置生成和验证签名需要用的凭证配置
- key: 用于请求头 `x-ca-key` 中设置
- secret: 用于生成请求签名
## 客户端签名生成方式
### 客户端签名生成方式
### 流程简介
#### 流程简介
客户端生成签名一共分三步处理：
@@ -104,7 +119,7 @@ consumers:
如下图所示：
![](https://help-static-aliyun-doc.aliyuncs.com/assets/img/zh-CN/1745707061/p188113.png)
### 签名串提取流程
#### 签名串提取流程
客户端需要从Http请求中提取出关键数据，组合成一个签名串，生成的签名串的格式如下：
@@ -160,7 +175,7 @@ Path + "?" + Key1 + "=" + Value1 + "&" + Key2 + "=" + Value2 + ... "&" + KeyN +
4. Query和Form存在数组参数时（key相同，value不同的参数） ，取第一个Value参与签名计算
### 签名串提取示例
#### 签名串提取示例
初始的HTTP请求：
```text
@@ -190,7 +205,7 @@ x-ca-timestamp:1525872629832
/http2test/test?param1=test&password=123456789&username=xiaoming
```
### 签名计算流程
#### 签名计算流程
客户端从HTTP请求中提取出关键数据组装成签名串后，需要对签名串进行加密及编码处理，形成最终的签名
@@ -206,7 +221,7 @@ String sign = Base64.encodeBase64String(result);
总结一下，就是将 `stringToSign` 使用UTF-8解码后得到Byte数组，然后使用加密算法对Byte数组进行加密，然后使用Base64算法进行编码，形成最终的签名。
### 添加签名流程
#### 添加签名流程
客户端需要将以下四个Header放在HTTP请求中传输给API网关，进行签名校验：
@@ -238,9 +253,9 @@ content-length:33
username=xiaoming&password=123456789
```
## 服务端签名验证方式
### 服务端签名验证方式
### 流程简介
#### 流程简介
服务器验证客户端签名一共分四步处理：
@@ -256,7 +271,7 @@ username=xiaoming&password=123456789
![](https://help-static-aliyun-doc.aliyuncs.com/assets/img/zh-CN/1745707061/p188116.png)
## 签名排错方法
### 签名排错方法
网关签名校验失败时，会将服务端的签名串（StringToSign）放到HTTP Response的Header中返回到客户端，Key为：X-Ca-Error-Message，用户只需要将本地计算的签名串（StringToSign）与服务端返回的签名串进行对比即可找到问题；
@@ -269,7 +284,7 @@ X-Ca-Error-Message: Server StringToSign:`GET#application/json##application/json
```
# 相关错误码
## 相关错误码
| HTTP 状态码 | 出错信息 | 原因说明 |
| ----------- | ---------------------- | -------------------------------------------------------------------------------- |
@@ -281,5 +296,3 @@ X-Ca-Error-Message: Server StringToSign:`GET#application/json##application/json
| 413 | Request Body Too Large | 请求 Body 超过限制大小：32 MB |
| 413 | Payload Too Large | 请求 Body 超过全局配置 DownstreamConnectionBufferLimits |
| 403 | Unauthorized Consumer | 请求的调用方无访问权限 |

301
plugins/wasm-cpp/extensions/hmac_auth/README_EN.md
View File

@@ -1,76 +1,71 @@
# Function Description
The `hmac-auth` plugin implements the generation of tamper-proof signatures for HTTP requests based on HMAC algorithm, and uses the signature for identity authentication and authorization.
---
title: HMAC Authentication
keywords: [higress,hmac auth]
description: HMAC Authentication plugin configuration reference
---
## Function Description
The `hmac-auth` plugin implements the generation of tamper-proof signatures for HTTP requests based on the HMAC algorithm, and performs authentication and authorization based on the signature.
# Configuration Fields
## Running Attributes
Plugin execution phase: `Authentication phase`
Plugin execution priority: `330`
| Name | Data Type | Required | Default | Description |
| ------------- | --------------- | -------------| ------ | ------------------------------------------------------------------------------------------------------------------------------------------------- |
| `consumers` | array of object | Required | - | Configures the caller of the service to authenticate the request. |
| `date_offset` | number | Optional | - | Configures the maximum allowed time deviation of the client, in seconds. It is used to parse the client's UTC time from `the Date` header of the request, and can be used to prevent replay attacks. If not configured, no validation is performed. |
| `_rules_` | array of object | Optional | - | Configures the access control list for specific routes or domains, used for authorization of requests. |
## Configuration Fields
**Note:**
- In a rule, authentication and authorization configurations cannot coexist.
- For requests that pass authentication and authorization, the request header will be added with an `X-Mse-Consumer` field to identify the caller's name.
The configuration fields for each item in `consumers` are as follows :
### Authentication Configuration
| Name | Data Type | Requirement | Default Value | Description |
| ------------- | ---------------- | ------------------- | ------------- | ----------------------------------------------------------------------------------------------------------------------- |
| `global_auth` | bool | Optional (**Instance level configuration only**) | - | Can only be configured at the instance level. If set to true, it acts globally; if false, only applies to configured domains and routes. If not configured, it will apply globally only when there are no domain and route configurations (to accommodate old user habits). |
| `consumers` | array of object | Mandatory | - | Configures the callers of the service for request authentication. |
| `date_offset` | number | Optional | - | Configures the maximum allowed client time offset, in seconds; parsed based on the request header `Date`; can be used to prevent request replay; no validation is performed if not configured. |
| Name | Data Type| Required | Default| Description |
| -------- | -------- | ------------ | ------ | ----------------------------------------------------------------------- |
| `key` | string | Required | - | Configures the key extracted from the `x-ca-key` header of the request. |
| `secret` | string | Required | - | Configures the secret used to generate the signature. |
| `name` | string | Required | - | Configures the name of the consumer. |
The configuration fields for each item in `consumers` are as follows:
| Name | Data Type | Requirement | Default Value | Description |
| -------- | --------- | ----------- | ------------- | ------------------------------------------- |
| `key` | string | Mandatory | - | Configures the key extracted from the `x-ca-key` header of the request. |
| `secret` | string | Mandatory | - | Configures the secret used to generate the signature. |
| `name` | string | Mandatory | - | Configures the name of the consumer. |
The configuration fields for each item in `_rules_` are as follows:
### Authorization Configuration (Optional)
| Name | Data Type | Requirement | Default Value | Description |
| ----------- | ---------------- | --------------------------------------------- | ------------- | --------------------------------------------------------------------------------------------------------------------------------------------------------------------- |
| `allow` | array of string | Optional (**Non-instance level configuration**) | - | Can only be configured on granular rules such as routes or domains. For requests that match the conditions, configure the allowed consumers to achieve fine-grained permission control. |
| Name | Data Type | Required | Default | Description |
| ---------------- | --------------- | ------------------------------------------------- | ---------------------------- | -------------------------------------------------- |
| `_match_route_` | array of string | Optional, either `_match_route_` or `_match_domain_` must be provided | - | Configures the name of the route to match. |
| `_match_domain_` | array of string | Optional, either `_match_route_` or `_match_domain_` must be provided | - | Configures the name of the domain to match. |
| `allow` | array of string | Required | - | Configures the name of the consumer to allow for requests that match the specified route or domain. |
**Note：**
- If `_rules_` is not configured, authentication is enabled for all routes on the current gateway instance by default ；
- For requests that pass authentication and authorization, a `X-Mse-Consumer` header will be added to the request headers to identify the name of the consumer.
# Configuration Example
The following configuration enables Hmac Auth authentication and authorization for specific routes or domains on the gateway. Note that the `key` field should not be duplicated.
## Enabling for specific routes or domains
## Configuration Example
### Global Configuration Authentication and Route Granular Authorization
Configure the following plugin settings at the instance level. Note that the `key` field cannot be duplicated:
```yaml
consumers:
global_auth: false
consumers:
- key: appKey-example-1
secret: appSecret-example-1
name: consumer-1
- key: appKey-example-2
secret: appSecret-example-2
name: consumer-2
# Configuring Fine-Grained Rules using _rules_ Field
_rules_:
# Rule 1: Matching by route name.
- _match_route_:
- route-a
- route-b
allow:
- consumer-1
# Rule 2: Applies based on domain name matching.
- _match_domain_:
- "*.example.com"
- test.com
allow:
- consumer-2
```
The `allow` field under each matching rule specifies the list of callers allowed to access under that matching condition;
In this example, `route-a` and `route-b` specified in `_match_route_` are the route names filled in when creating the gateway route. When either of these routes is matched, it will allow access to the caller named `consumer-1`, while denying access to other callers；
In` _match_domain_`, `*.example.com` and `test.com` are used to match the requested domain name. When a match is found, it will allow access to the caller named `consumer-2`, while denying access to other callers；
Upon successful authentication, the `X-Mse-Consumer` field will be added to the request header with the value set to the caller's name, such as `consumer-1`.。
## Enable at the Gateway Instance Level
The following configuration enables HMAC authentication at the gateway instance level.
For route-a and route-b, configure the plugin as follows:
```yaml
consumers:
allow:
- consumer1
```
For the two domains *.example.com and test.com, configure as follows:
```yaml
allow:
- consumer2
```
If configured in the console, the specified route names route-a and route-b correspond to the route names filled in when creating the gateway routes. When matched to these two routes, access will be allowed for the caller named consumer1, while other callers will not be allowed access.
The specified *.example.com and test.com are used to match the domains of the requests. When a domain match is found, access will be allowed for the caller named consumer2, while other callers will not be allowed access.
### Gateway Instance Level Activation
The following configuration will enable HMAC Auth authentication at the gateway instance level, requiring all requests to undergo authentication before access.
```yaml
global_auth: true
consumers:
- key: appKey-example-1
secret: appSecret-example-1
name: consumer-1
@@ -79,34 +74,24 @@ consumers:
name: consumer-2
```
## Signature Mechanism Description
### Configuration Preparation
As mentioned above, configure the credentials required for generating and verifying signatures in the plugin settings.
- key: to be set in the request header `x-ca-key`.
- secret: used for generating request signatures.
# Description of Signing Mechanism
### Client Signature Generation Process
#### Overview
The client generates a signature through three main steps:
1. Extract key data from the original request to create a string for signing.
2. Encrypt the key data signing string using the algorithm and the configured `secret` to obtain the signature.
3. Include all relevant headers for the signature into the original HTTP request to form the final HTTP request.
## Configuration Preparation
As mentioned in the guide above, configure the credential settings required for generating and validating signatures in the plugin configuration.
- key: Used for setting in the request header `x-ca-key`.
- secret: Used for generating the request signature.
## Client Signature Generation Method
### Overview of the Process
The process for generating a signature on the client side consists of three steps:
1. Extracting key data from the original request to obtain a string to be signed.
2. Using encryption algorithms and the configured `secret` to encrypt the key data signing string and obtain a signature.
3. Adding all headers related to the signature to the original HTTP request to obtain the final HTTP request.
As shown below :
As shown in the figure below:
![](https://help-static-aliyun-doc.aliyuncs.com/assets/img/zh-CN/1745707061/p188113.png)
### Process for Extracting Signing String
To generate a signature, the client needs to extract key data from the HTTP request and combine it into a signing string. The format of the generated signing string is as follows:
#### Signing String Extraction Process
The client needs to extract key data from the HTTP request, combine it into a signing string, which has the following format:
```text
HTTPMethod
Accept
@@ -116,54 +101,38 @@ Date
Headers
PathAndParameters
```
The signing string consists of the above 7 fields separated by \n. If Headers is empty, no \n is needed. If other fields are empty, the \n should still be retained. The signature is case-sensitive. Below are the rules for extracting each field:
- HTTPMethod: The HTTP method used in the request, in all capital letters, such as POST.
- Accept: The value of the Accept header in the request, which can be empty. It is recommended to explicitly set the Accept header. When Accept is empty, some HTTP clients will set the default value of `*/*`, which may cause signature verification to fail.
- Content-MD5: The value of the Content-MD5 header in the request, which can be empty. It is only calculated when there is a non-form body in the request. The following is a reference calculation method for Content-MD5 values in ：
The seven fields above constitute the entire signing string, separated by newline characters `\n`. If Headers is empty, no newline is needed; other fields should retain `\n` if empty. The signature is case-sensitive. Below are the extraction rules for each field:
- HTTPMethod: The HTTP method, all uppercase (e.g., POST).
- Accept: The value of the Accept header in the request, can be empty. It is recommended to explicitly set the Accept Header. When Accept is empty, some HTTP clients may set a default value of `*/*`, resulting in a signature verification failure.
- Content-MD5: The value of the Content-MD5 header in the request, can be empty. It is calculated only if there is a Body in the request and it is not in Form format. Here’s a reference calculation method for the Content-MD5 value in Java:
```java
String content-MD5 = Base64.encodeBase64(MD5(bodyStream.getbytes("UTF-8")));
String content-MD5 = Base64.encodeBase64(MD5(bodyStream.getBytes("UTF-8")));
```
- Content-Type: The value of the Content-Type header in the request, which can be empty.
- Date: The value of the Date header in the request. When the` date_offset` configuration is not enabled, it can be empty. Otherwise, it will be used for time offset verification.
- Headers: Users can select specific headers to participate in the signature. There are the following rules for concatenating the signature string with headers:
- The keys of the headers participating in the signature calculation are sorted in alphabetical order and concatenated as follows:
- Content-Type: The value of the Content-Type header in the request, can be empty.
- Date: The value of the Date header in the request. If the `date_offset` configuration is not turned on, it can be empty; otherwise, it will be used for time offset verification.
- Headers: Users can select specific headers to participate in the signature. The rules for concatenating the signing header string are as follows:
- The Keys of the headers participating in the signature calculation are concatenated after being sorted lexicographically, as follows:
```text
HeaderKey1 + ":" + HeaderValue1 + "\n"\+
HeaderKey2 + ":" + HeaderValue2 + "\n"\+
HeaderKey1 + ":" + HeaderValue1 + "\n" +
HeaderKey2 + ":" + HeaderValue2 + "\n" +
...
HeaderKeyN + ":" + HeaderValueN + "\n"
```
- If the value of a header is empty, it will participate in the signature with the `HeaderKey+":"+"\n"` only, and the key and english colon should be retained.
- The set of keys for all headers participating in the signature is separated by a comma and placed in the `X-Ca-Signature-Headers header`.
- If the Value of a certain header is empty, use HeaderKey + ":" + "\n" to participate in the signature, retaining the Key and the colon.
- The collection of all participating header Keys is placed in the Header with the key X-Ca-Signature-Headers, separated by commas.
- The following headers are not included in the header signature calculation: X-Ca-Signature, X-Ca-Signature-Headers, Accept, Content-MD5, Content-Type, Date.
- PathAndParameters: This field contains all parameters in the path, query, and form. The specific format is as follows:
- PathAndParameters: This field includes Path, Query, and all parameters in Form, specifically organized as follows:
```text
Path + "?" + Key1 + "=" + Value1 + "&" + Key2 + "=" + Value2 + ... "&" + KeyN + "=" + ValueN
```
Note:
1. The Key of Query and Form parameters should be sorted lexicographically before being concatenated as above.
2. If Query and Form parameters are empty, just use Path without adding `?`.
3. If the Value of parameters is empty, only the Key should be retained in the signature, the equal sign does not need to be added.
4. In the case of array parameters (parameters with the same key but different values), only the first Value should be used for signature calculation.
Notes:
1. The keys of the query and form parameter pairs are sorted alphabetically, and the same format as above is used for concatenation.
2. If there are no query and form parameters, use the path directly without adding `?` .
3. If the value of a parameter is empty, only the key will be included in the signature. The equal sign should not be included in the signature.
4. If there are array parameters in the query or form (parameters with the same key but different values), only the first value should be included in the signature calculation.
### Example of Extracting Signing String
The initial HTTP request :
#### Signing String Extraction Example
Initial HTTP request:
```text
POST /http2test/test?param1=test HTTP/1.1
host:api.aliyun.com
@@ -177,8 +146,7 @@ x-ca-nonce:c9f15cbf-f4ac-4a6c-b54d-f51abf4b5b44
content-length:33
username=xiaoming&password=123456789
```
The correct generated signature string is :
The generated correct signing string is:
```text
POST
application/json; charset=utf-8
@@ -190,13 +158,10 @@ x-ca-signature-method:HmacSHA256
x-ca-timestamp:1525872629832
/http2test/test?param1=test&password=123456789&username=xiaoming
```
#### Signature Calculation Process
After the client assembles the key data extracted from the HTTP request into a signing string, it needs to encrypt the signing string and encode it to form the final signature.
### Signature Calculation Process
After extracting the key data from the HTTP request and assembling it into a signature string, the client needs to encrypt and encode the signature string to form the final signature.
The specific encryption format is as follows, where `stringToSign` is the extracted signature string, `secret` is the one filled in the plugin configuration, and `sign` is the final generated signature:
The specific encryption form is as follows, where `stringToSign` is the extracted signing string, `secret` is the one filled in the plugin configuration, and `sign` is the final generated signature:
```java
Mac hmacSha256 = Mac.getInstance("HmacSHA256");
byte[] secretBytes = secret.getBytes("UTF-8");
@@ -204,23 +169,16 @@ hmacSha256.init(new SecretKeySpec(secretBytes, 0, secretBytes.length, "HmacSHA25
byte[] result = hmacSha256.doFinal(stringToSign.getBytes("UTF-8"));
String sign = Base64.encodeBase64String(result);
```
To summarize, the `stringToSign` is decoded using UTF-8 to obtain a Byte array, then the encryption algorithm is applied to the Byte array, and finally, the Base64 algorithm is used for encoding, forming the final signature.
In summary, the `stringToSign` is decoded using UTF-8 to obtain a Byte array. Then, an encryption algorithm is used to encrypt the Byte array, and finally, the Base64 algorithm is used to encode the encrypted data, resulting in the final signature.
### The Process of Adding a Signature
The client needs to include the following four headers in the HTTP request to be transmitted to the API gateway for signature verification:
- x-ca-key: The value is the APP Key and is required.
- x-ca-signature-method: The signature algorithm, the value can be HmacSHA256 or HmacSHA1, optional. The default value is HmacSHA256.
- x-ca-signature-headers: The collection of keys for all signature headers, separated by commas. Optional.
- x-ca-signature: The signature and it is required.
Here is an example of a complete HTTP request with a signature :
#### Adding the Signature Process
The client needs to include the following four headers in the HTTP request to transmit to the API gateway for signature verification:
- x-ca-key: The APP Key, mandatory.
- x-ca-signature-method: The signature algorithm, can be HmacSHA256 or HmacSHA1, optional, default is HmacSHA256.
- x-ca-signature-headers: The collection of all signature header Keys, separated by commas, optional.
- x-ca-signature: The signature, mandatory.
Below is an example of the entire HTTP request carrying the signature:
```text
POST /http2test/test?param1=test HTTP/1.1
host:api.aliyun.com
@@ -239,48 +197,35 @@ content-length:33
username=xiaoming&password=123456789
```
## Server-side Signature Verification Method
### Server-side Signature Verification Method
#### Overview
The server verifies the client signature through four main steps:
1. Extract key data from the received request to create a signing string.
2. Read the `key` from the received request and query the corresponding `secret`.
3. Encrypt the key data signing string using the algorithm and the `secret` to obtain the signature.
4. Read the client signature from the received request and compare the server-side signature with the client-side signature for consistency.
### Overview of the Process
The server-side signature verification of the client's request involves four steps :
1. Extract crucial data from the received request to obtain a string for signing.
2. Retrieve the `key` from the received request and use it to query its corresponding `secret`.
3. Encrypt the string for signing using the encryption algorithm and `secret`.
4. Retrieve the client's signature from the received request, and compare the consistency of the server-side signature with the client's signature.
As shown below :
As shown in the figure below:
![](https://help-static-aliyun-doc.aliyuncs.com/assets/img/zh-CN/1745707061/p188116.png)
### Signature Troubleshooting Method
When the gateway signature verification fails, the server's signing string (StringToSign) will be returned in the HTTP Response header to the client, with the key: X-Ca-Error-Message. The user only needs to compare the locally computed signing string (StringToSign) with the signing string returned by the server to find the issue.
## Troubleshooting Signature Errors
When the gateway signature verification fails, the server-side signing string (StringToSign) will be returned to the client in the HTTP Response Header. The key is X-Ca-Error-Message. Users only need to compare the locally calculated signing string with the server-side signing string returned to locate the problem;
If the StringToSign on the server side is consistent with that on the client side, please check whether the APP Secret used for signature calculation is correct；
Because line breaks cannot be represented in HTTP headers, all line breaks in the StringToSign are replaced with #, as shown below:
If the StringToSign from the server and the client are consistent, please check whether the APP Secret used for signature calculation is correct.
Since HTTP headers cannot express line breaks, the line breaks in the StringToSign have been replaced with `#`, as shown below:
```text
X-Ca-Error-Message: Server StringToSign:`GET#application/json##application/json##X-Ca-Key:200000#X-Ca-Timestamp:1589458000000#/app/v1/config/keys?keys=TEST`
```
# Related Error Codes
| HTTP Status Code | Error Message | Reason |
| ----------- | ---------------------- | -------------------------------------------------------------------------------- |
| 401 | Invalid Key | The x-ca-key request header is not provided or is invalid. |
| 401 | Empty Signature | The x-ca-signature request header does not contain a signature. |
| 400 | Invalid Signature | The x-ca-signature request header contains a signature that does not match the server-calculated signature. |
| 400 | Invalid Content-MD5 | The content-md5 request header is incorrect. |
| 400 | Invalid Date | The time offset calculated based on the date request header exceeds the configured date_offset. |
| 413 | Request Body Too Large | The request body exceeds the size limit of 32 MB. |
| 413 | Payload Too Large | The request body exceeds the DownstreamConnectionBufferLimits global configuration. |
| 403 | Unauthorized Consumer | The requesting party does not have access permission. |
## Related Error Codes
| HTTP Status Code | Error Message | Reasoning |
| ---------------- | --------------------- | --------------------------------------------- |
| 401 | Invalid Key | The request header did not provide x-ca-key, or x-ca-key is invalid. |
| 401 | Empty Signature | The request header did not provide the x-ca-signature signing string. |
| 400 | Invalid Signature | The x-ca-signature signing string in the request header does not match the signature calculated by the server. |
| 400 | Invalid Content-MD5 | The Content-MD5 header in the request is incorrect. |
| 400 | Invalid Date | The time offset calculated based on the Date header in the request exceeds the configured date_offset. |
| 413 | Request Body Too Large| The request Body exceeds the maximum size of 32 MB. |
| 413 | Payload Too Large | The request Body exceeds the global configured DownstreamConnectionBufferLimits. |
| 403 | Unauthorized Consumer | The calling party does not have access permissions for the request. |

409
plugins/wasm-cpp/extensions/jwt_auth/README.md
View File

@@ -1,17 +1,216 @@
# 功能说明
---
title: JWT 认证
keywords: [higress,jwt auth]
description: JWT 认证插件配置参考
---
## 功能说明
`jwt-auth`插件实现了基于JWT(JSON Web Tokens)进行认证鉴权的功能，支持从HTTP请求的URL参数、请求头、Cookie字段解析JWT，同时验证该Token是否有权限访问。
本插件和`安全能力->认证鉴权`中JWT认证的区别是，额外提供了调用方身份识别的能力，支持对不同调用方配置不同的JWT凭证。
# 详细说明
## 运行属性
## 1、基于token的认证
插件执行阶段：`认证阶段`
插件执行优先级：`340`
### 1.1 简介
## 配置字段
**注意：**
- 在一个规则里，鉴权配置和认证配置不可同时存在
- 对于通过认证鉴权的请求，请求的header会被添加一个`X-Mse-Consumer`字段，用以标识调用者的名称。
### 认证配置
| 名称 | 数据类型 | 填写要求 | 默认值 | 描述 |
| ----------- | --------------- | ------------------------------------------- | ------ | ----------------------------------------------------------- |
| `global_auth` | bool | 选填（**仅实例级别配置**） | - | 只能在实例级别配置，若配置为true，则全局生效认证机制; 若配置为false，则只对做了配置的域名和路由生效认证机制，若不配置则仅当没有域名和路由配置时全局生效（兼容老用户使用习惯）。 |
| `consumers` | array of object | 必填 | - | 配置服务的调用者，用于对请求进行认证 |
`consumers`中每一项的配置字段说明如下：
| 名称 | 数据类型 | 填写要求 | 默认值 | 描述 |
| ----------------------- | ----------------- | -------- | ------------------------------------------------- | ------------------------ |
| `name` | string | 必填 | - | 配置该consumer的名称 |
| `jwks` | string | 必填 | - | https://www.rfc-editor.org/rfc/rfc7517 指定的json格式字符串，是由验证JWT中签名的公钥（或对称密钥）组成的Json Web Key Set |
| `issuer` | string | 必填 | - | JWT的签发者，需要和payload中的iss字段保持一致 |
| `claims_to_headers` | array of object | 选填 | - | 抽取JWT的payload中指定字段，设置到指定的请求头中转发给后端 |
| `from_headers` | array of object | 选填 | {"name":"Authorization","value_prefix":"Bearer "} | 从指定的请求头中抽取JWT |
| `from_params` | array of string | 选填 | access_token | 从指定的URL参数中抽取JWT |
| `from_cookies` | array of string | 选填 | - | 从指定的cookie中抽取JWT |
| `clock_skew_seconds` | number | 选填 | 60 | 校验JWT的exp和iat字段时允许的时钟偏移量，单位为秒 |
| `keep_token` | bool | 选填 | ture | 转发给后端时是否保留JWT |
**注意：**
- 只有当`from_headers`,`from_params`,`from_cookies`均未配置时，才会使用默认值
`from_headers` 中每一项的配置字段说明如下：
| 名称 | 数据类型 | 填写要求| 默认值 | 描述 |
| ---------------- | --------------- | ------- | ------ | --------------------------------------------------------- |
| `name` | string | 必填 | - | 抽取JWT的请求header |
| `value_prefix` | string | 必填 | - | 对请求header的value去除此前缀，剩余部分作为JWT |
`claims_to_headers` 中每一项的配置字段说明如下：
| 名称 | 数据类型 | 填写要求| 默认值 | 描述 |
| ---------------- | --------------- | ------- | ------ | --------------------------------------------------------- |
| `claim` | string | 必填 | - | JWT payload中的指定字段，要求必须是字符串或无符号整数类型 |
| `header` | string | 必填 | - | 从payload取出字段的值设置到这个请求头中，转发给后端 |
| `override` | bool | 选填 | true | true时，存在同名请求头会进行覆盖；false时，追加同名请求头 |
### 鉴权配置（非必需）
| 名称 | 数据类型 | 填写要求 | 默认值 | 描述 |
| ----------- | --------------- | ------------------------------------------- | ------ | ----------------------------------------------------------- |
| `allow` | array of string | 选填(**非实例级别配置**) | - | 只能在路由或域名等细粒度规则上配置，对于符合匹配条件的请求，配置允许访问的 consumer，从而实现细粒度的权限控制 |
## 配置示例
### 全局配置认证和路由粒度进行鉴权
注意如果一个JWT能匹配多个`jwks`，则按照配置顺序命中第一个匹配的`consumer`
在实例级别做如下插件配置：
```yaml
global_auth: false
consumers:
- name: consumer1
issuer: abcd
jwks: |
{
"keys": [
{
"kty": "oct",
"kid": "123",
"k": "hM0k3AbXBPpKOGg__Ql2Obcq7s60myWDpbHXzgKUQdYo7YCRp0gUqkCnbGSvZ2rGEl4YFkKqIqW7mTHdj-bcqXpNr-NOznEyMpVPOIlqG_NWVC3dydBgcsIZIdD-MR2AQceEaxriPA_VmiUCwfwL2Bhs6_i7eolXoY11EapLQtutz0BV6ZxQQ4dYUmct--7PLNb4BWJyQeWu0QfbIthnvhYllyl2dgeLTEJT58wzFz5HeNMNz8ohY5K0XaKAe5cepryqoXLhA-V-O1OjSG8lCNdKS09OY6O0fkyweKEtuDfien5tHHSsHXoAxYEHPFcSRL4bFPLZ0orTt1_4zpyfew",
"alg": "HS256"
}
]
}
- name: consumer2
issuer: abc
jwks: |
{
"keys": [
{
"kty": "RSA",
"e": "AQAB",
"use": "sig",
"kid": "123",
"alg": "RS256",
"n": "i0B67f1jggT9QJlZ_8QL9QQ56LfurrqDhpuu8BxtVcfxrYmaXaCtqTn7OfCuca7cGHdrJIjq99rz890NmYFZuvhaZ-LMt2iyiSb9LZJAeJmHf7ecguXS_-4x3hvbsrgUDi9tlg7xxbqGYcrco3anmalAFxsbswtu2PAXLtTnUo6aYwZsWA6ksq4FL3-anPNL5oZUgIp3HGyhhLTLdlQcC83jzxbguOim-0OEz-N4fniTYRivK7MlibHKrJfO3xa_6whBS07HW4Ydc37ZN3Rx9Ov3ZyV0idFblU519nUdqp_inXj1eEpynlxH60Ys_aTU2POGZh_25KXGdF_ZC_MSRw"
}
]
}
```
对 route-a 和 route-b 这两个路由做如下配置：
```yaml
allow:
- consumer1
```
对 *.example.com 和 test.com 在这两个域名做如下配置:
```yaml
allow:
- consumer2
```
**说明：**
此例指定的route-a和route-b即在创建网关路由时填写的路由名称，当匹配到这两个路由时，将允许name为consumer1的调用者访问，其他调用者不允许访问。
此例指定的*.example.com和test.com用于匹配请求的域名，当发现域名匹配时，将允许name为consumer2的调用者访问，其他调用者不被允许访问。
根据该配置，下列请求可以允许访问：
假设以下请求会匹配到route-a这条路由
**将 JWT 设置在 url 参数中**
```bash
curl 'http://xxx.hello.com/test?access_token=eyJhbGciOiJIUzI1NiIsInR5cCI6IkpXVCIsImtpZCI6IjEyMyJ9.eyJpc3MiOiJhYmNkIiwic3ViIjoidGVzdCIsImlhdCI6MTY2NTY2MDUyNywiZXhwIjoxODY1NjczODE5fQ.-vBSV0bKeDwQcuS6eeSZN9dLTUnSnZVk8eVCXdooCQ4'
```
**将 JWT 设置在 http 请求头中**
```bash
curl http://xxx.hello.com/test -H 'Authorization: Bearer eyJhbGciOiJIUzI1NiIsInR5cCI6IkpXVCIsImtpZCI6IjEyMyJ9.eyJpc3MiOiJhYmNkIiwic3ViIjoidGVzdCIsImlhdCI6MTY2NTY2MDUyNywiZXhwIjoxODY1NjczODE5fQ.-vBSV0bKeDwQcuS6eeSZN9dLTUnSnZVk8eVCXdooCQ4'
```
认证鉴权通过后，请求的header中会被添加一个`X-Mse-Consumer`字段，在此例中其值为`consumer1`，用以标识调用方的名称
下列请求将拒绝访问：
**请求未提供JWT，返回401**
```bash
curl http://xxx.hello.com/test
```
**根据请求提供的JWT匹配到的调用者无访问权限，返回403**
```bash
# consumer1不在*.example.com的allow列表里
curl 'http://xxx.example.com/test' -H 'Authorization: Bearer eyJhbGciOiJIUzI1NiIsInR5cCI6IkpXVCIsImtpZCI6IjEyMyJ9.eyJpc3MiOiJhYmNkIiwic3ViIjoidGVzdCIsImlhdCI6MTY2NTY2MDUyNywiZXhwIjoxODY1NjczODE5fQ.-vBSV0bKeDwQcuS6eeSZN9dLTUnSnZVk8eVCXdooCQ4'
```
#### 网关实例级别开启
以下配置将对网关实例级别开启 JWT Auth 认证，所有请求均需要经过认证后才能访问。
```yaml
global_auth: true
consumers:
- name: consumer1
issuer: abcd
jwks: |
{
"keys": [
{
"kty": "oct",
"kid": "123",
"k": "hM0k3AbXBPpKOGg__Ql2Obcq7s60myWDpbHXzgKUQdYo7YCRp0gUqkCnbGSvZ2rGEl4YFkKqIqW7mTHdj-bcqXpNr-NOznEyMpVPOIlqG_NWVC3dydBgcsIZIdD-MR2AQceEaxriPA_VmiUCwfwL2Bhs6_i7eolXoY11EapLQtutz0BV6ZxQQ4dYUmct--7PLNb4BWJyQeWu0QfbIthnvhYllyl2dgeLTEJT58wzFz5HeNMNz8ohY5K0XaKAe5cepryqoXLhA-V-O1OjSG8lCNdKS09OY6O0fkyweKEtuDfien5tHHSsHXoAxYEHPFcSRL4bFPLZ0orTt1_4zpyfew",
"alg": "HS256"
}
]
}
- name: consumer2
issuer: abc
jwks: |
{
"keys": [
{
"kty": "RSA",
"e": "AQAB",
"use": "sig",
"kid": "123",
"alg": "RS256",
"n": "i0B67f1jggT9QJlZ_8QL9QQ56LfurrqDhpuu8BxtVcfxrYmaXaCtqTn7OfCuca7cGHdrJIjq99rz890NmYFZuvhaZ-LMt2iyiSb9LZJAeJmHf7ecguXS_-4x3hvbsrgUDi9tlg7xxbqGYcrco3anmalAFxsbswtu2PAXLtTnUo6aYwZsWA6ksq4FL3-anPNL5oZUgIp3HGyhhLTLdlQcC83jzxbguOim-0OEz-N4fniTYRivK7MlibHKrJfO3xa_6whBS07HW4Ydc37ZN3Rx9Ov3ZyV0idFblU519nUdqp_inXj1eEpynlxH60Ys_aTU2POGZh_25KXGdF_ZC_MSRw"
}
]
}
```
## 常见错误码说明
| HTTP 状态码 | 出错信息 | 原因说明 |
| ----------- | ---------------------- | -------------------------------------------------------------------------------- |
| 401 | Jwt missing | 请求头未提供JWT |
| 401 | Jwt expired | JWT已经过期 |
| 401 | Jwt verification fails | JWT payload校验失败，如iss不匹配 |
| 403 | Access Denied | 无权限访问当前路由 |
## 详细说明
### 1、基于token的认证
#### 1.1 简介
很多对外开放的API需要识别请求者的身份，并据此判断所请求的资源是否可以返回给请求者。token就是一种用于身份验证的机制，基于这种机制，应用不需要在服务端保留用户的认证信息或者会话信息，可实现无状态、分布式的Web应用授权，为应用的扩展提供了便利。
### 1.2 流程描述
#### 1.2 流程描述
![](https://help-static-aliyun-doc.aliyuncs.com/assets/img/zh-CN/2336348951/p135822.png)
@@ -35,13 +234,13 @@
在这个整个过程中, 网关利用token认证机制，实现了用户使用自己的用户体系对自己API进行授权的能力。下面我们就要介绍网关实现token认证所使用的结构化令牌Json Web Token(JWT)。
### 1.3 JWT
#### 1.3 JWT
#### 1.3.1 简介
##### 1.3.1 简介
Json Web Toke（JWT），是为了在网络应用环境间传递声明而执行的一种基于JSON的开放标准RFC7519。JWT一般可以用作独立的身份验证令牌，可以包含用户标识、用户角色和权限等信息，以便于从资源服务器获取资源，也可以增加一些额外的其它业务逻辑所必须的声明信息，特别适用于分布式站点的登录场景。
#### 1.3.2 JWT的构成
##### 1.3.2 JWT的构成
`eyJhbGciOiJIUzI1NiIsInR5cCI6IkpXVCJ9.eyJzdWIiOiIxMjM0NTY3ODkwIiwibmFtZSI6IkpvaG4gRG9lIiwiYWRtaW4iOnRydWV9.TJVA95OrM7E2cBab30RMHrHDcEfxjoYZgeFONFh7HgQ`
@@ -119,11 +318,11 @@ var signature = HMACSHA256(encodedString, '$secret');
将这三部分用 . 连接成一个完整的字符串，就构成了 1.3.2 节最开始的JWT示例。
#### 1.3.3 时效
##### 1.3.3 时效
网关会验证token中的exp字段，一旦这个字段过期了，网关会认为这个token无效而将请求直接打回。过期时间这个值必须设置。
#### 1.3.4 JWT的几个特点
##### 1.3.4 JWT的几个特点
1. JWT 默认是不加密，不能将秘密数据写入 JWT。
2. JWT 不仅可以用于认证，也可以用于交换信息。有效使用 JWT，可以降低服务器查询数据库的次数。
@@ -131,9 +330,9 @@ var signature = HMACSHA256(encodedString, '$secret');
4. JWT 本身包含了认证信息，一旦泄露，任何人都可以获得该令牌的所有权限。为了减少盗用，JWT 的有效期应该设置得比较短。对于一些比较重要的权限，使用时应该再次对用户进行认证。
5. 为了减少盗用，JWT 不应该使用 HTTP 协议明码传输，要使用HTTPS 协议传输。
## 2、用户系统如何应用JWT插件保护API
### 2、用户系统如何应用JWT插件保护API
### 2.1 生成一对JWK（JSON Web 密钥）
#### 2.1 生成一对JWK（JSON Web 密钥）
**方法一、在线生成：**
@@ -162,7 +361,7 @@ final String publicKeyString = rsaJsonWebKey.toJson(JsonWebKey.OutputControlLeve
final String privateKeyString = rsaJsonWebKey.toJson(JsonWebKey.OutputControlLevel.INCLUDE_PRIVATE);
```
### 2.2 使用JWK中的私钥实现颁发token 的认证服务
#### 2.2 使用JWK中的私钥实现颁发token 的认证服务
需要使用2.1节中在线生成的 Keypair JSON字符串（三个方框内的第一个）或者本地生成的 privateKeyString JSON字符串作为私钥来颁发token，用于授权可信的用户访问受保护的API，具体实现可以参考下方示例。 向客户颁发token的形式由用户根据具体的业务场景决定，可以将颁发token的功能部署到生产环境，配置成普通API后由访问者通过用户名密码获得，也可以直接在本地环境生成token 后，直接拷贝给指定用户使用。
@@ -216,187 +415,3 @@ public class GenerateJwtDemo {
}
}
```
# 插件配置说明
## 配置字段
| 名称 | 数据类型 | 填写要求 | 默认值 | 描述 |
| ----------- | --------------- | ------------------------------------------- | ------ | ----------------------------------------------------------- |
| `consumers` | array of object | 必填 | - | 配置服务的调用者，用于对请求进行认证 |
| `_rules_` | array of object | 选填 | - | 配置特定路由或域名的访问权限列表，用于对请求进行鉴权 |
`consumers`中每一项的配置字段说明如下：
| 名称 | 数据类型 | 填写要求 | 默认值 | 描述 |
| ----------------------- | ----------------- | -------- | ------------------------------------------------- | ------------------------ |
| `name` | string | 必填 | - | 配置该consumer的名称 |
| `jwks` | string | 必填 | - | https://www.rfc-editor.org/rfc/rfc7517 指定的json格式字符串，是由验证JWT中签名的公钥（或对称密钥）组成的Json Web Key Set |
| `issuer` | string | 必填 | - | JWT的签发者，需要和payload中的iss字段保持一致 |
| `claims_to_headers` | array of object | 选填 | - | 抽取JWT的payload中指定字段，设置到指定的请求头中转发给后端 |
| `from_headers` | array of object | 选填 | {"name":"Authorization","value_prefix":"Bearer "} | 从指定的请求头中抽取JWT |
| `from_params` | array of string | 选填 | access_token | 从指定的URL参数中抽取JWT |
| `from_cookies` | array of string | 选填 | - | 从指定的cookie中抽取JWT |
| `clock_skew_seconds` | number | 选填 | 60 | 校验JWT的exp和iat字段时允许的时钟偏移量，单位为秒 |
| `keep_token` | bool | 选填 | ture | 转发给后端时是否保留JWT |
**注意：**
- 只有当`from_headers`,`from_params`,`from_cookies`均未配置时，才会使用默认值
`from_headers` 中每一项的配置字段说明如下：
| 名称 | 数据类型 | 填写要求| 默认值 | 描述 |
| ---------------- | --------------- | ------- | ------ | --------------------------------------------------------- |
| `name` | string | 必填 | - | 抽取JWT的请求header |
| `value_prefix` | string | 必填 | - | 对请求header的value去除此前缀，剩余部分作为JWT |
`claims_to_headers` 中每一项的配置字段说明如下：
| 名称 | 数据类型 | 填写要求| 默认值 | 描述 |
| ---------------- | --------------- | ------- | ------ | --------------------------------------------------------- |
| `claim` | string | 必填 | - | JWT payload中的指定字段，要求必须是字符串或无符号整数类型 |
| `header` | string | 必填 | - | 从payload取出字段的值设置到这个请求头中，转发给后端 |
| `override` | bool | 选填 | true | true时，存在同名请求头会进行覆盖；false时，追加同名请求头 |
`_rules_` 中每一项的配置字段说明如下：
| 名称 | 数据类型 | 填写要求 | 默认值 | 描述 |
| ---------------- | --------------- | ------------------------------------------------- | ------ | -------------------------------------------------- |
| `_match_route_` | array of string | 选填，`_match_route_`，`_match_domain_`中选填一项 | - | 配置要匹配的路由名称 |
| `_match_domain_` | array of string | 选填，`_match_route_`，`_match_domain_`中选填一项 | - | 配置要匹配的域名 |
| `allow` | array of string | 必填 | - | 对于符合匹配条件的请求，配置允许访问的consumer名称 |
**注意：**
- 若不配置`_rules_`字段，则默认对当前网关实例的所有路由开启认证；
- 对于通过认证鉴权的请求，请求的header会被添加一个`X-Mse-Consumer`字段，用以标识调用者的名称。
## 配置示例
### 对特定路由或域名开启
以下配置将对网关特定路由或域名开启 Jwt Auth 认证和鉴权，注意如果一个JWT能匹配多个`jwks`，则按照配置顺序命中第一个匹配的`consumer`
```yaml
consumers:
- name: consumer1
issuer: abcd
jwks: |
{
"keys": [
{
"kty": "oct",
"kid": "123",
"k": "hM0k3AbXBPpKOGg__Ql2Obcq7s60myWDpbHXzgKUQdYo7YCRp0gUqkCnbGSvZ2rGEl4YFkKqIqW7mTHdj-bcqXpNr-NOznEyMpVPOIlqG_NWVC3dydBgcsIZIdD-MR2AQceEaxriPA_VmiUCwfwL2Bhs6_i7eolXoY11EapLQtutz0BV6ZxQQ4dYUmct--7PLNb4BWJyQeWu0QfbIthnvhYllyl2dgeLTEJT58wzFz5HeNMNz8ohY5K0XaKAe5cepryqoXLhA-V-O1OjSG8lCNdKS09OY6O0fkyweKEtuDfien5tHHSsHXoAxYEHPFcSRL4bFPLZ0orTt1_4zpyfew",
"alg": "HS256"
}
]
}
- name: consumer2
issuer: abc
jwks: |
{
"keys": [
{
"kty": "RSA",
"e": "AQAB",
"use": "sig",
"kid": "123",
"alg": "RS256",
"n": "i0B67f1jggT9QJlZ_8QL9QQ56LfurrqDhpuu8BxtVcfxrYmaXaCtqTn7OfCuca7cGHdrJIjq99rz890NmYFZuvhaZ-LMt2iyiSb9LZJAeJmHf7ecguXS_-4x3hvbsrgUDi9tlg7xxbqGYcrco3anmalAFxsbswtu2PAXLtTnUo6aYwZsWA6ksq4FL3-anPNL5oZUgIp3HGyhhLTLdlQcC83jzxbguOim-0OEz-N4fniTYRivK7MlibHKrJfO3xa_6whBS07HW4Ydc37ZN3Rx9Ov3ZyV0idFblU519nUdqp_inXj1eEpynlxH60Ys_aTU2POGZh_25KXGdF_ZC_MSRw"
}
]
}
# 使用 _rules_ 字段进行细粒度规则配置
_rules_:
# 规则一：按路由名称匹配生效
- _match_route_:
- route-a
- route-b
allow:
- consumer1
# 规则二：按域名匹配生效
- _match_domain_:
- "*.example.com"
- test.com
allow:
- consumer2
```
此例 `_match_route_` 中指定的 `route-a` 和 `route-b` 即在创建网关路由时填写的路由名称，当匹配到这两个路由时，将允许`name`为`consumer1`的调用者访问，其他调用者不允许访问；
此例 `_match_domain_` 中指定的 `*.example.com` 和 `test.com` 用于匹配请求的域名，当发现域名匹配时，将允许`name`为`consumer2`的调用者访问，其他调用者不允许访问。
#### 根据该配置，下列请求可以允许访问：
假设以下请求会匹配到route-a这条路由
**将 JWT 设置在 url 参数中**
```bash
curl 'http://xxx.hello.com/test?access_token=eyJhbGciOiJIUzI1NiIsInR5cCI6IkpXVCIsImtpZCI6IjEyMyJ9.eyJpc3MiOiJhYmNkIiwic3ViIjoidGVzdCIsImlhdCI6MTY2NTY2MDUyNywiZXhwIjoxODY1NjczODE5fQ.-vBSV0bKeDwQcuS6eeSZN9dLTUnSnZVk8eVCXdooCQ4'
```
**将 JWT 设置在 http 请求头中**
```bash
curl http://xxx.hello.com/test -H 'Authorization: Bearer eyJhbGciOiJIUzI1NiIsInR5cCI6IkpXVCIsImtpZCI6IjEyMyJ9.eyJpc3MiOiJhYmNkIiwic3ViIjoidGVzdCIsImlhdCI6MTY2NTY2MDUyNywiZXhwIjoxODY1NjczODE5fQ.-vBSV0bKeDwQcuS6eeSZN9dLTUnSnZVk8eVCXdooCQ4'
```
认证鉴权通过后，请求的header中会被添加一个`X-Mse-Consumer`字段，在此例中其值为`consumer1`，用以标识调用方的名称
#### 下列请求将拒绝访问：
**请求未提供JWT，返回401**
```bash
curl http://xxx.hello.com/test
```
**根据请求提供的JWT匹配到的调用者无访问权限，返回403**
```bash
# consumer1不在*.example.com的allow列表里
curl 'http://xxx.example.com/test' -H 'Authorization: Bearer eyJhbGciOiJIUzI1NiIsInR5cCI6IkpXVCIsImtpZCI6IjEyMyJ9.eyJpc3MiOiJhYmNkIiwic3ViIjoidGVzdCIsImlhdCI6MTY2NTY2MDUyNywiZXhwIjoxODY1NjczODE5fQ.-vBSV0bKeDwQcuS6eeSZN9dLTUnSnZVk8eVCXdooCQ4'
```
### 网关实例级别开启
以下配置未指定`_rules_`字段，因此将对网关实例级别开启 JWT Auth 认证
```yaml
consumers:
- name: consumer1
issuer: abcd
jwks: |
{
"keys": [
{
"kty": "oct",
"kid": "123",
"k": "hM0k3AbXBPpKOGg__Ql2Obcq7s60myWDpbHXzgKUQdYo7YCRp0gUqkCnbGSvZ2rGEl4YFkKqIqW7mTHdj-bcqXpNr-NOznEyMpVPOIlqG_NWVC3dydBgcsIZIdD-MR2AQceEaxriPA_VmiUCwfwL2Bhs6_i7eolXoY11EapLQtutz0BV6ZxQQ4dYUmct--7PLNb4BWJyQeWu0QfbIthnvhYllyl2dgeLTEJT58wzFz5HeNMNz8ohY5K0XaKAe5cepryqoXLhA-V-O1OjSG8lCNdKS09OY6O0fkyweKEtuDfien5tHHSsHXoAxYEHPFcSRL4bFPLZ0orTt1_4zpyfew",
"alg": "HS256"
}
]
}
- name: consumer2
issuer: abc
jwks: |
{
"keys": [
{
"kty": "RSA",
"e": "AQAB",
"use": "sig",
"kid": "123",
"alg": "RS256",
"n": "i0B67f1jggT9QJlZ_8QL9QQ56LfurrqDhpuu8BxtVcfxrYmaXaCtqTn7OfCuca7cGHdrJIjq99rz890NmYFZuvhaZ-LMt2iyiSb9LZJAeJmHf7ecguXS_-4x3hvbsrgUDi9tlg7xxbqGYcrco3anmalAFxsbswtu2PAXLtTnUo6aYwZsWA6ksq4FL3-anPNL5oZUgIp3HGyhhLTLdlQcC83jzxbguOim-0OEz-N4fniTYRivK7MlibHKrJfO3xa_6whBS07HW4Ydc37ZN3Rx9Ov3ZyV0idFblU519nUdqp_inXj1eEpynlxH60Ys_aTU2POGZh_25KXGdF_ZC_MSRw"
}
]
}
```
# 常见错误码说明
| HTTP 状态码 | 出错信息 | 原因说明 |
| ----------- | ---------------------- | -------------------------------------------------------------------------------- |
| 401 | Jwt missing | 请求头未提供JWT |
| 401 | Jwt expired | JWT已经过期 |
| 401 | Jwt verification fails | JWT payload校验失败，如iss不匹配 |
| 403 | Access Denied | 无权限访问当前路由 |

578
plugins/wasm-cpp/extensions/jwt_auth/README_EN.md
View File

@@ -1,170 +1,306 @@
# Description
The `jwt-auth` plugin implements authentication and authorization based on JWT (JSON Web Token), supports parsing JWTs from URL parameters, request headers, and Cookie fields from HTTP requests, and verifies whether the token has permission to access.
---
title: JWT Authentication
keywords: [higress,jwt auth]
description: JWT Authentication plugin configuration reference
---
The difference between this plugin and the JWT authentication in `Security Capabilities->Authentication and Authorization` is that it provides additional capabilities for identifying the caller's identity, supporting the configuration of different JWT credentials for different callers.
## Function Description
`jwt-auth` plugin implements authentication and authorization based on JWT (JSON Web Tokens). It supports extracting JWT from HTTP request URL parameters, request headers, and Cookie fields, while verifying whether the Token has the necessary permissions to access the resource.
# Detailed Description
The difference between this plugin and the JWT authentication in `Security Capability -> Authentication and Authorization` is that it additionally provides the capability of identifying the caller's identity, supporting different JWT credentials for different callers.
## 1. Token-based authentication
## Runtime Properties
Plugin execution phase: `Authentication Phase`
Plugin execution priority: `340`
### 1.1 Introduction
Many open APIs need to identify the identity of the caller and determine whether the requested resource can be returned to the caller based on this identity. Token is a mechanism used for identity verification. Based on this mechanism, the application does not need to retain the user's authentication information or session information on the server, which can realize stateless and distributed web application authorization and provide convenience for application extension.
## Configuration Fields
**Note:**
- In one rule, authentication configuration and authorization configuration cannot coexist.
- For requests authenticated through authentication and authorization, the request header will be augmented with an `X-Mse-Consumer` field to identify the caller's name.
### 1.2 Process Description
### Authentication Configuration
| Name | Data Type | Requirements | Default Value | Description |
| ----------- | --------------- | ---------------------------------------------- | ------------- | ----------------------------------------------------------- |
| `global_auth` | bool | Optional (**instance-level configuration only**) | - | Can only be configured at the instance level. If set to true, it will globally enable the authentication mechanism; if set to false, it will only apply to the configured domain names and routes. If not configured, it will only globally take effect when no domain names and routes are configured (to be compatible with old user habits). |
| `consumers` | array of object | Required | - | Configure service consumers for request authentication |
![](process.png)
The configuration fields for each item in `consumers` are as follows:
| Name | Data Type | Requirements | Default Value | Description |
| ----------------------- | ------------------ | ------------ | -------------------------------------------------- | ------------------------------- |
| `name` | string | Required | - | The name of the consumer |
| `jwks` | string | Required | - | JSON format string specified by https://www.rfc-editor.org/rfc/rfc7517, consisting of the public key (or symmetric key) used to verify the JWT signature. |
| `issuer` | string | Required | - | The issuer of the JWT, must match the `iss` field in the payload. |
| `claims_to_headers` | array of object | Optional | - | Extract the specified fields from the JWT payload and set them in the specified request headers to forward to the backend. |
| `from_headers` | array of object | Optional | {"name":"Authorization","value_prefix":"Bearer "} | Extract JWT from the specified request headers. |
| `from_params` | array of string | Optional | access_token | Extract JWT from the specified URL parameters. |
| `from_cookies` | array of string | Optional | - | Extract JWT from the specified cookies. |
| `clock_skew_seconds` | number | Optional | 60 | The allowed clock skew when validating the `exp` and `iat` fields of the JWT, measured in seconds. |
| `keep_token` | bool | Optional | true | Whether to retain the JWT when forwarding to the backend. |
The above figure is the business process sequence diagram when the gateway uses JWT for authentication, and the following we will describe the steps marked in the figure in detail in words:
**Note:**
- The default values will only be used when `from_headers`, `from_params`, and `from_cookies` are not all configured.
The configuration fields for each item in `from_headers` are as follows:
| Name | Data Type | Requirements | Default Value | Description |
| --------------- | ---------------- | ------------ | ------------- | ----------------------------------------------- |
| `name` | string | Required | - | Extract JWT from the request header. |
| `value_prefix` | string | Required | - | Remove the prefix from the request header value, with the remaining part serving as the JWT. |
1. The client initiates an authentication request to the API gateway, usually carrying the end user's username and password in the request;
The configuration fields for each item in `claims_to_headers` are as follows:
| Name | Data Type | Requirements | Default Value | Description |
| --------------- | ---------------- | ------------ | ------------- | --------------------------------------------- |
| `claim` | string | Required | - | The specified field in the JWT payload, must be a string or unsigned integer type. |
| `header` | string | Required | - | The value of the field extracted from the payload is set to this request header and forwarded to the backend. |
| `override` | bool | Optional | true | If true, existing headers with the same name will be overridden; if false, they will be appended. |
2. The gateway forwards the request directly to the backend service;
### Authorization Configuration (Optional)
| Name | Data Type | Requirements | Default Value | Description |
| ----------- | --------------- | ---------------------------------------------- | ------------- | ----------------------------------------------------------------------------------------------------- |
| `allow` | array of string | Optional (**not instance-level configuration**) | - | Can only be configured on fine-grained rules such as routes or domain names, allowing specified consumers to access matching requests for fine-grained permission control. |
3. The backend service reads the authentication information (such as the username and password) in the request for verification. After the verification is passed, it uses the private key to generate a standard token and returns it to the gateway;
## Configuration Examples
### Global Configuration for Authentication and Route-Level Authorization
Note: If a JWT matches multiple `jwks`, the first matching consumer will be applied according to the configuration order.
4. The gateway returns the response with the token to the client, and the client needs to cache this token locally;
5. The client sends a business request to the API gateway, carrying the token in the request;
6. The gateway uses the public key set by the user to verify the token in the request. After the verification is passed, the request is passed through to the backend service;
7. The backend service processes the business and responds;
8. The gateway returns the business response to the client.
In this entire process, the gateway uses the token authentication mechanism to realize the ability of users to use their own user system to authorize their API. Next, we will introduce the structured token JSON Web Token (JWT) used by the gateway to implement token authentication.
### 1.3 JWT
#### 1.3.1 Introduction
JSON Web Token (JWT) is an open standard RFC7519 based on JSON for executing a type of claim to pass between network applications. JWT can generally be used as an independent identity verification token, which can contain user identification, user roles, and permissions information, making it easier to obtain resources from the resource server, and can also add some other necessary declarative information for other business logic, especially suitable for the login scenario of distributed sites.
#### 1.3.2 Composition of JWT
`eyJhbGciOiJIUzI1NiIsInR5cCI6IkpXVCJ9.eyJzdWIiOiIxMjM0NTY3ODkwIiwibmFtZSI6IkpvaG4gRG9lIiwiYWRtaW4iOnRydWV9.TJVA95OrM7E2cBab30RMHrHDcEfxjoYZgeFONFh7HgQ`
As shown in the example above, JWT is a string consisting of three parts:
- Header
- Payload
- Signature
**Header**
The header of the JWT carries two pieces of information:
- The type of the token, which is JWT
- The algorithm used for encryption
The gateway supports the following encryption algorithms:
```text
ES256, ES384, ES512,
HS256, HS384, HS512,
RS256, RS384, RS512,
PS256, PS384, PS512,
EdDSA
Configure the plugin at the instance level as follows:
```yaml
global_auth: false
consumers:
- name: consumer1
issuer: abcd
jwks: |
{
"keys": [
{
"kty": "oct",
"kid": "123",
"k": "hM0k3AbXBPpKOGg__Ql2Obcq7s60myWDpbHXzgKUQdYo7YCRp0gUqkCnbGSvZ2rGEl4YFkKqIqW7mTHdj-bcqXpNr-NOznEyMpVPOIlqG_NWVC3dydBgcsIZIdD-MR2AQceEaxriPA_VmiUCwfwL2Bhs6_i7eolXoY11EapLQtutz0BV6ZxQQ4dYUmct--7PLNb4BWJyQeWu0QfbIthnvhYllyl2dgeLTEJT58wzFz5HeNMNz8ohY5K0XaKAe5cepryqoXLhA-V-O1OjSG8lCNdKS09OY6O0fkyweKEtuDfien5tHHSsHXoAxYEHPFcSRL4bFPLZ0orTt1_4zpyfew",
"alg": "HS256"
}
]
}
- name: consumer2
issuer: abc
jwks: |
{
"keys": [
{
"kty": "RSA",
"e": "AQAB",
"use": "sig",
"kid": "123",
"alg": "RS256",
"n": "i0B67f1jggT9QJlZ_8QL9QQ56LfurrqDhpuu8BxtVcfxrYmaXaCtqTn7OfCuca7cGHdrJIjq99rz890NmYFZuvhaZ-LMt2iyiSb9LZJAeJmHf7ecguXS_-4x3hvbsrgUDi9tlg7xxbqGYcrco3anmalAFxsbswtu2PAXLtTnUo6aYwZsWA6ksq4FL3-anPNL5oZUgIp3HGyhhLTLdlQcC83jzxbguOim-0OEz-N4fniTYRivK7MlibHKrJfO3xa_6whBS07HW4Ydc37ZN3Rx9Ov3ZyV0idFblU519nUdqp_inXj1eEpynlxH60Ys_aTU2POGZh_25KXGdF_ZC_MSRw"
}
]
}
```
The complete header looks like the following JSON:
Configure the following for routes `route-a` and `route-b`:
```yaml
allow:
- consumer1
```
Configure the following for domain names `*.example.com` and `test.com`:
```yaml
allow:
- consumer2
```
**Explanation:**
The specified `route-a` and `route-b` refer to the routing names filled in when creating the gateway route. When these two routes are matched, the caller with the name consumer1 will be allowed access, while others will not be permitted.
The specified `*.example.com` and `test.com` are used to match the request domain names. When a matching domain name is found, the caller with the name consumer2 will be allowed access, while others will not.
Based on this configuration, the following requests will be allowed:
Suppose the following request matches the route-a.
**Setting the JWT in URL parameters**
```bash
curl 'http://xxx.hello.com/test?access_token=eyJhbGciOiJIUzI1NiIsInR5cCI6IkpXVCIsImtpZCI6IjEyMyJ9.eyJpc3MiOiJhYmNkIiwic3ViIjoidGVzdCIsImlhdCI6MTY2NTY2MDUyNywiZXhwIjoxODY1NjczODE5fQ.-vBSV0bKeDwQcuS6eeSZN9dLTUnSnZVk8eVCXdooCQ4'
```
**Setting the JWT in HTTP request headers**
```bash
curl http://xxx.hello.com/test -H 'Authorization: Bearer eyJhbGciOiJIUzI1NiIsInR5cCI6IkpXVCIsImtpZCI6IjEyMyJ9.eyJpc3MiOiJhYmNkIiwic3ViIjoidGVzdCIsImlhdCI6MTY2NTY2MDUyNywiZXhwIjoxODY1NjczODE5fQ.-vBSV0bKeDwQcuS6eeSZN9dLTUnSnZVk8eVCXdooCQ4'
```
After successful authentication and authorization, the request's header will include an `X-Mse-Consumer` field, in this example with the value `consumer1`, to identify the caller's name.
The following requests will be denied:
**Request without JWT returns 401**
```bash
curl http://xxx.hello.com/test
```
**Caller matching from the provided JWT has no access permission, returning 403**
```bash
# consumer1 is not in the allow list for *.example.com
curl 'http://xxx.example.com/test' -H 'Authorization: Bearer eyJhbGciOiJIUzI1NiIsInR5cCI6IkpXVCIsImtpZCI6IjEyMyJ9.eyJpc3MiOiJhYmNkIiwic3ViIjoidGVzdCIsImlhdCI6MTY2NTY2MDUyNywiZXhwIjoxODY1NjczODE5fQ.-vBSV0bKeDwQcuS6eeSZN9dLTUnSnZVk8eVCXdooCQ4'
```
#### Enable at Gateway Instance Level
The following configuration will enable JWT Auth authentication at the instance level, requiring all requests to be authenticated before accessing.
```yaml
global_auth: true
consumers:
- name: consumer1
issuer: abcd
jwks: |
{
"keys": [
{
"kty": "oct",
"kid": "123",
"k": "hM0k3AbXBPpKOGg__Ql2Obcq7s60myWDpbHXzgKUQdYo7YCRp0gUqkCnbGSvZ2rGEl4YFkKqIqW7mTHdj-bcqXpNr-NOznEyMpVPOIlqG_NWVC3dydBgcsIZIdD-MR2AQceEaxriPA_VmiUCwfwL2Bhs6_i7eolXoY11EapLQtutz0BV6ZxQQ4dYUmct--7PLNb4BWJyQeWu0QfbIthnvhYllyl2dgeLTEJT58wzFz5HeNMNz8ohY5K0XaKAe5cepryqoXLhA-V-O1OjSG8lCNdKS09OY6O0fkyweKEtuDfien5tHHSsHXoAxYEHPFcSRL4bFPLZ0orTt1_4zpyfew",
"alg": "HS256"
}
]
}
- name: consumer2
issuer: abc
jwks: |
{
"keys": [
{
"kty": "RSA",
"e": "AQAB",
"use": "sig",
"kid": "123",
"alg": "RS256",
"n": "i0B67f1jggT9QJlZ_8QL9QQ56LfurrqDhpuu8BxtVcfxrYmaXaCtqTn7OfCuca7cGHdrJIjq99rz890NmYFZuvhaZ-LMt2iyiSb9LZJAeJmHf7ecguXS_-4x3hvbsrgUDi9tlg7xxbqGYcrco3anmalAFxsbswtu2PAXLtTnUo6aYwZsWA6ksq4FL3-anPNL5oZUgIp3HGyhhLTLdlQcC83jzxbguOim-0OEz-N4fniTYRivK7MlibHKrJfO3xa_6whBS07HW4Ydc37ZN3Rx9Ov3ZyV0idFblU519nUdqp_inXj1eEpynlxH60Ys_aTU2POGZh_25KXGdF_ZC_MSRw"
}
]
}
```
## Common Error Codes
| HTTP Status Code | Error Message | Reason Description |
| ---------------- | ----------------------------- | ---------------------------------------------------------------------------- |
| 401 | Jwt missing | The request header did not provide a JWT |
| 401 | Jwt expired | The JWT has expired |
| 401 | Jwt verification fails | JWT payload verification failed, such as mismatched `iss` |
| 403 | Access Denied | No permission to access the current route |
## Detailed Description
### 1. Token-based Authentication
#### 1.1 Introduction
Many open APIs need to identify the requester's identity and determine whether the requested resource can be returned. A token is a mechanism used for identity verification. With this mechanism, applications do not need to retain user authentication information or session information on the server, allowing for stateless, distributed web application authorization, facilitating application scaling.
#### 1.2 Process Description
![](https://help-static-aliyun-doc.aliyuncs.com/assets/img/zh-CN/2336348951/p135822.png)
The above image shows the entire business process sequence diagram for gateway authentication using JWT. Below, we will detail the steps indicated in the diagram:
1. The client initiates an authentication request to the API gateway, generally carrying the terminal user's username and password.
2. The gateway forwards the request directly to the backend service.
3. The backend service reads the verification information in the request (such as username and password) for validation. Upon successful verification, it generates a standard token using a private key and returns it to the gateway.
4. The gateway returns a response containing the token to the client, who must cache this token locally.
5. The client sends a business request to the API gateway, carrying the token in the request.
6. The gateway validates the token using the user's set public key. Upon successful validation, it forwards the request to the backend service.
7. The backend service processes the business and responds.
8. The gateway returns the business response to the client.
Throughout this process, the gateway utilizes the token authentication mechanism, enabling the user to authorize their API using their user system. Next, we will introduce the structured token used by the gateway for token authentication: JSON Web Token (JWT).
#### 1.3 JWT
##### 1.3.1 Introduction
JSON Web Token (JWT) is an open standard (RFC7519) that defines a compact and self-contained way for securely transmitting information between parties as a JSON object. JWT can be used as a stand-alone authentication token, capable of containing user identity, user roles, permissions, and other information, aiding in resource retrieval from resource servers and adding any additional claims required by business logic, particularly suitable for login scenarios for distributed sites.
##### 1.3.2 JWT Structure
`eyJhbGciOiJIUzI1NiIsInR5cCI6IkpXVCJ9.eyJzdWIiOiIxMjM0NTY3ODkwIiwibmFtZSI6IkpvaG4gRG9lIiwiYWRtaW4iOnRydWV9.TJVA95OrM7E2cBab30RMHrHDcEfxjoYZgeFONFh7HgQ`
As shown above, JWT is a string composed of three parts:
- Header
- Payload
- Signature
**Header**
The header of the JWT carries two pieces of information:
- Token type, which is JWT
- Signing algorithm
The supported signing algorithms by the gateway are as follows:
```text
ES256, ES384, ES512,
HS256, HS384, HS512,
RS256, RS384, RS512,
PS256, PS384, PS512,
EdDSA
```
The complete header looks like the following JSON:
```js
{
'typ': 'JWT',
'alg': 'HS256'
}
```
The header is then Base64-encoded (this encoding can be symmetrically decoded), forming the first part:
`eyJ0eXAiOiJKV1QiLCJhbGciOiJIUzI1NiJ9`
**Payload**
The payload is where the actual information is stored. The details are defined below:
```
Then the header is Base64 encoded (this encoding is symmetrically decodable), forming the first part.
`eyJ0eXAiOiJKV1QiLCJhbGciOiJIUzI1NiJ9`
**Payload**
The payload is where valid information is stored. Its details are defined as follows:
```text
iss: The issuer of the token. This indicates who created the token and is a string value.
sub: The subject identifier. This is the unique identifier for the end user provided by the issuer, and is no longer than 255 ASCII characters, and is case-sensitive within the issuer's scope.
aud: The audience(s) of the token, which is an array of case-sensitive strings.
exp: The expiration time of the token, after which the token will be invalidated, is an integer declaration representing the Unix timestamp in seconds.
iat: The issuance time of the token, is an integer declaration representing the Unix timestamp in seconds.
jti: The unique identifier of the token, and the value is unique for every token created by the issuer. It is usually a cryptographically random value to prevent conflicts. This component adds a random entropy that an attacker cannot obtain to the structured token, making it more difficult for the token to be guessed or replayed.
```
Custom fields for a user feature can also be added, such as the example below adding a "name" field for the user's nickname:
iss: Token issuer. Indicates who created the token, this claim is a string.
sub: Subject Identifier, an identifier provided by the issuer for its end users, unique within the issuer's scope, up to 255 ASCII characters, case sensitive.
aud: Audience(s), the audience of the token, an array of strings that are case-sensitive.
exp: Expiration time, a timestamp of the token's expiration. Tokens expired beyond this time will be invalid. This claim is an integer, representing the number of seconds since January 1, 1970.
iat: Issuance time of the token, this claim is an integer, representing the number of seconds since January 1, 1970.
jti: Unique identifier for the token, the value of this claim must be unique for each token created by the token issuer to prevent collisions. It is typically a cryptographically random value. This value adds a random entropy component to the structured token that is not accessible to an attacker, helping prevent token guess and replay attacks.
```
Custom fields necessary for the user system can also be added, for example, the following example adds a nickname `name`:
```js
{
"sub": "1234567890",
"name": "John Doe"
}
```
The payload is then Base64-encoded, forming the second part of the JWT:
`JTdCJTBBJTIwJTIwJTIyc3ViJTIyJTNBJTIwJTIyMTIzNDU2Nzg5MCUyMiUyQyUwQSUyMCUyMCUyMm5hbWUlMjIlM0ElMjAlMjJKb2huJTIwRG9lJTIyJTBBJTdE`
**Signature**
This part is a string that consists of the Base64-encoded header and Base64-encoded payload concatenated with a period, followed by the encryption of the resulting string using the algorithm specified in the header (where $secret represents the user's private key).
```
Then encode it in Base64 to obtain the second part of the JWT:
`JTdCJTBBJTIwJTIwJTIyc3ViJTIyJTNBJTIwJTIyMTIzNDU2Nzg5MCUyMiUyQyUwQSUyMCUyMCUyMm5hbWUlMjIlM0ElMjAlMjJKb2huJTIwRG9lJTIyJTBBJTdE`
**Signature**
This part requires the Base64-encoded Header and Base64-encoded Payload to be connected by a period, and then encrypted using the signing method declared in the Header (where `$secret` represents the user's private key) to form the third part of the JWT.
```js
var encodedString = base64UrlEncode(header) + '.' + base64UrlEncode(payload);
var signature = HMACSHA256(encodedString, '$secret');
```
```
Connecting these three parts with a period creates a complete string that forms the JWT example at the beginning of section 1.3.2.
These three parts are then concatenated using periods to form the complete JWT string as shown in the example at the beginning of this section.
##### 1.3.3 Validity Period
The gateway will validate the `exp` field in the token. Once this field expires, the gateway will consider this token invalid and directly reject the request. The expiration time must be set.
#### 1.3.3 Time validity
##### 1.3.4 Characteristics of JWT
1. JWT is not encrypted by default; do not write secret data into JWT.
2. JWT can be used for both authentication and information exchange. Effectively using JWT can reduce the number of queries to the database on the server.
3. The biggest drawback of JWT is that since the server does not maintain session state, it cannot revoke a token during use or change the permissions of said token. In other words, once a JWT is issued, it will remain valid until expiration, unless the server implements additional logic.
4. JWT itself contains authentication information, and once leaked, anyone can gain all permissions of that token. To minimize theft, the validity period of JWT should be set to be relatively short. For some critical permissions, users should be re-authenticated.
5. To reduce theft, JWT should not be transmitted in plain text over HTTP but should use HTTPS for transmission.
The gateway will verify the exp field in the token. Once this field has expired, the gateway will consider the token invalid and reject the request directly. The expiration time value must be set.
#### 1.3.4 Several Characteristics of JWT
1. By default, JWT is not encrypted and cannot write secret data into JWT.
2. JWT can not only be used for authentication but also for exchanging information. Using JWT effectively can reduce the number of times servers query the database.
3. The biggest drawback of JWT is that the server cannot invalidate a token during use, or change the token's permission, because the server does not keep the session state. That is, once JWT is issued, it will always be valid before it expires, unless the server deploys additional logic.
4. JWT contains authentication information itself. Once leaked, anyone can obtain all permissions of the token. To reduce theft, the expiration time of JWT should be set relatively short. For some more important permissions, users should be authenticated again when using them.
5. To reduce theft, JWT should not be transmitted in plaintext using the HTTP protocol, and the HTTPS protocol should be used for transmission.
## 2. How to apply the JWT plugin to protect the API of the user system
### 2.1 Generate a pair of JWK (JSON Web Key)
**Method 1: Generate online**
Users can generate the private and public keys used for token generation and verification on this website https://mkjwk.org. The private key is used for the authorization service to issue JWT, and the public key is configured into the JWT plugin for the gateway to verify the signature of the request. Note that the JWKs format configuration used by the gateway requires the public key in the figure below to be placed in the keys structure, such as: `{"keys":[{"kty":"RSA","e":"AQAB",...}]}`
<img src="https://help-static-aliyun-doc.aliyuncs.com/assets/img/zh-CN/2336348951/p135823.png" style="zoom:50%" />
**Method 2: Generate locally**
This article uses a Java example to illustrate, and users of other languages can also find relevant tools to generate key pairs. Create a new Maven project and add the following dependencies:
### 2. How User Systems Apply the JWT Plugin to Protect APIs
#### 2.1 Generating a Pair of JWKs (JSON Web Keys)
**Method 1: Online Generation:**
Users can generate the private and public keys used for token generation and verification at this site https://mkjwk.org. The private key is used by the authorization service to issue JWTs, and the public key is configured into the JWT plugin for the gateway to verify requests. Pay attention to the jwks format configuration used by the gateway. In the image below, the Public Key should be placed into the keys structure, for example: `{"keys":[{"kty":"RSA","e":"AQAB",...}]}`
<img src="https://help-static-aliyun-doc.aliyuncs.com/assets/img/zh-CN/2336348951/p135823.png" style="zoom:50%" />
**Method 2: Local Generation:**
This article demonstrates using Java; users of other languages can find related tools to generate key pairs. Create a new Maven project and include the following dependency:
```xml
<dependency>
<groupId>org.bitbucket.b_c</groupId>
<artifactId>jose4j</artifactId>
<version>0.7.0</version>
</dependency>
```
Use the following code to generate a pair of RSA keys:
```
Use the following code to generate a pair of RSA keys:
```java
RsaJsonWebKey rsaJsonWebKey = RsaJwkGenerator.generateJwk(2048);
final String publicKeyString = rsaJsonWebKey.toJson(JsonWebKey.OutputControlLevel.PUBLIC_ONLY);
final String privateKeyString = rsaJsonWebKey.toJson(JsonWebKey.OutputControlLevel.INCLUDE_PRIVATE);
```
### 2.2 Implement the token issuance authentication service using the private key in JWK
The Keypair JSON string generated online in Section 2.1 (the first of the three boxes) or privateKeyString JSON string generated locally needs to be used as the private key to issue tokens for trusted users to access protected APIs. The specific implementation can refer to the following example. The form of issuing tokens to customers is determined by the user according to the specific business scenario. The function of issuing tokens can be deployed in the production environment and configured as a normal API for visitors to obtain through username and password, or tokens can be generated directly in the local environment and copied to designated users for use.
```
#### 2.2 Using the Private Key in JWK to Implement the Token Issuance Authentication Service
You will need to use the Keypair JSON string (the first inside the three boxes) generated online in section 2.1 or the locally generated privateKeyString JSON string as the private key to issue tokens for authorizing trusted users to access protected APIs. The specific implementation can refer to the example below. The form of issuing tokens to customers can be determined by the user based on the specific business scenario; it can be deployed in the production environment, configured to be a normal API that visitors can access through username and password, or tokens can be generated locally and directly copied for specified users to use.
```java
import java.security.PrivateKey;
import java.security.PrivateKey;
import org.jose4j.json.JsonUtil;
import org.jose4j.jwk.RsaJsonWebKey;
import org.jose4j.jwk.RsaJwkGenerator;
@@ -173,32 +309,29 @@ import org.jose4j.jws.JsonWebSignature;
import org.jose4j.jwt.JwtClaims;
import org.jose4j.jwt.NumericDate;
import org.jose4j.lang.JoseException;
public class GenerateJwtDemo {
public static void main(String[] args) throws JoseException {
String keyId = "uniq_key";
//Use the Keypair generated in section 2.1 of this article
// Use the Keypair generated in section 2.1
String privateKeyJson = "{\n"
+ " \"kty\": \"RSA\",\n"
+ " \"d\": "
+
"\"O9MJSOgcjjiVMNJ4jmBAh0mRHF_TlaVva70Imghtlgwxl8BLfcf1S8ueN1PD7xV6Cnq8YenSKsfiNOhC6yZ_fjW1syn5raWfj68eR7cjHWjLOvKjwVY33GBPNOvspNhVAFzeqfWneRTBbga53Agb6jjN0SUcZdJgnelzz5JNdOGaLzhacjH6YPJKpbuzCQYPkWtoZHDqWTzCSb4mJ3n0NRTsWy7Pm8LwG_Fd3pACl7JIY38IanPQDLoighFfo-Lriv5z3IdlhwbPnx0tk9sBwQBTRdZ8JkqqYkxUiB06phwr7mAnKEpQJ6HvhZBQ1cCnYZ_nIlrX9-I7qomrlE1UoQ\",\n"
+ " \"d\": \"O9MJSOgcjjiVMNJ4jmBAh0mRHF_TlaVva70Imghtlgwxl8BLfcf1S8ueN1PD7xV6Cnq8YenSKsfiNOhC6yZ_fjW1syn5raWfj68eR7cjHWjLOvKjwVY33GBPNOvspNhVAFzeqfWneRTBbga53Agb6jjN0SUcZdJgnelzz5JNdOGaLzhacjH6YPJKpbuzCQYPkWtoZHDqWTzCSb4mJ3n0NRTsWy7Pm8LwG_Fd3pACl7JIY38IanPQDLoighFfo-Lriv5z3IdlhwbPnx0tk9sBwQBTRdZ8JkqqYkxUiB06phwr7mAnKEpQJ6HvhZBQ1cCnYZ_nIlrX9-I7qomrlE1UoQ\",\n"
+ " \"e\": \"AQAB\",\n"
+ " \"alg\": \"RS256\",\n"
+ " \"n\": \"vCuB8MgwPZfziMSytEbBoOEwxsG7XI3MaVMoocziP4SjzU4IuWuE_DodbOHQwb_thUru57_Efe"
+
"--sfATHEa0Odv5ny3QbByqsvjyeHk6ZE4mSAV9BsHYa6GWAgEZtnDceeeDc0y76utXK2XHhC1Pysi2KG8KAzqDa099Yh7s31AyoueoMnrYTmWfEyDsQL_OAIiwgXakkS5U8QyXmWicCwXntDzkIMh8MjfPskesyli0XQD1AmCXVV3h2Opm1Amx0ggSOOiINUR5YRD6mKo49_cN-nrJWjtwSouqDdxHYP-4c7epuTcdS6kQHiQERBd1ejdpAxV4c0t0FHF7MOy9kw\"\n"
+ " \"n\": \"vCuB8MgwPZfziMSytEbBoOEwxsG7XI3MaVMoocziP4SjzU4IuWuE_DodbOHQwb_thUru57_Efe--sfATHEa0Odv5ny3QbByqsvjyeHk6ZE4mSAV9BsHYa6GWAgEZtnDceeeDc0y76utXK2XHhC1Pysi2KG8KAzqDa099Yh7s31AyoueoMnrYTmWfEyDsQL_OAIiwgXakkS5U8QyXmWicCwXntDzkIMh8MjfPskesyli0XQD1AmCXVV3h2Opm1Amx0ggSOOiINUR5YRD6mKo49_cN-nrJWjtwSouqDdxHYP-4c7epuTcdS6kQHiQERBd1ejdpAxV4c0t0FHF7MOy9kw\"\n"
+ "}";
JwtClaims claims = new JwtClaims();
claims.setGeneratedJwtId();
claims.setIssuedAtToNow();
//Expiration time must be set
// Expiration time must be set
NumericDate date = NumericDate.now();
date.addSeconds(120*60);
claims.setExpirationTime(date);
claims.setNotBeforeMinutesInThePast(1);
claims.setSubject("YOUR_SUBJECT");
claims.setAudience("YOUR_AUDIENCE");
//Add custom parameters, use String type for all values
// Add custom parameters, all values should be String type
claims.setClaim("userId", "1213234");
claims.setClaim("email", "userEmail@youapp.com");
JsonWebSignature jws = new JsonWebSignature();
@@ -206,194 +339,9 @@ public class GenerateJwtDemo {
jws.setKeyIdHeaderValue(keyId);
jws.setPayload(claims.toJson());
PrivateKey privateKey = new RsaJsonWebKey(JsonUtil.parseJson(privateKeyJson)).getPrivateKey();
jws.setKey(privateKey);
String jwtResult = jws.getCompactSerialization();
System.out.println("Generate Json Web token , result is " + jwtResult);
}
}
```
# Plugin Configuration Guide
## Configuration Fields
| Name | Data Type | Required | Default | Description |
| ----------- | --------------- | ------------------------------------------- | ------ | ----------------------------------------------------------- |
| `consumers` | array of object | Yes | - | Configures callers of the service for authenticating requests |
| `_rules_` | array of object | Optional | - | Configures access control lists for specific routes or domains for authorizing requests |
The configuration field descriptions for each item in consumers are as follows:
| Name | Data Type | Required | Default | Description|
| ----------------------- | ----------------- | -------- | ------------------------------------------------- | ------------------------ |
| `name` | string | Yes | - | Configures the name of this consumer |
| `jwks` | string | Yes | - | Specifies a JSON Web Key Set, as defined in https://www.rfc-editor.org/rfc/rfc7517, consisting of public keys (or symmetric keys) used to verify the signature of JWT |
| `issuer` | string | Yes | - | The issuer of the JWT, which should be consistent with the iss field in the payload |
| `claims_to_headers` | array of object | Optional | - | Extracts the specified fields from the JWT's payload and sets them to the specified request headers for forwarding to the backend |
| `from_headers` | array of object | Optional | {"name":"Authorization","value_prefix":"Bearer "} | Extracts the JWT from the specified request headers |
| `from_params` | array of string | Optional | access_token | Extracts the JWT from the specified URL parameters |
| `from_cookies` | array of string | Optional | - | Extracts the JWT from the specified cookie(s) |
| `clock_skew_seconds` | number | Optional | 60 | The amount of clock skew, in seconds, that is allowed when verifying the exp and iat fields of the JWT |
| `keep_token` | bool | Optional | ture | Whether to keep the JWT when forwarding it to the backend |
**Note:**
- The default value is used only if neither `from_headers`, `from_params`, nor `from_cookies` are configured.
The configuration field descriptions for each item in `from_headers` are as follows:
| Name | Data Type | Required| Default | Description |
| ---------------- | --------------- | ------- | ------ | --------------------------------------------------------- |
| `name` | string | Yes | - | Specifies the request header to extract the JWT from |
| `value_prefix` | string | Yes | - | Removes the specified prefix from the request header's value, leaving the rest as the JWT |
The configuration field descriptions for each item in `claims_to_headers` are as follows:
| Name | Data Type | Required| Default | Description |
| ---------------- | --------------- | ------- | ------ | --------------------------------------------------------- |
| `claim` | string | Yes | - | The name of the field in the JWT payload, which must be a string or unsigned integer |
| `header` | string | Yes | - | Sets the value of the specified request header to the value of the specified field in the payload, for forwarding to the backend |
| `override` | bool | Optional | true | If true, overrides an existing header with the same name; if false, appends the header to the existing headers |
The configuration field descriptions for each item in `_rules_` are as follows:
| Name | Data Type | Required| Default | Description |
| ---------------- | --------------- | ------------------------------------------------- | ------ | -------------------------------------------------- |
| `_match_route_` | array of string | Optional, select either `_match_route_` or `_match_domain_` | - | Configures the route names to match|
| `_match_domain_` | array of string | Optional, select either `_match_route_` or `_match_domain_` | - | Configures the domains to match |
| `allow` | array of string | Required | - | Configures the consumer names allowed to access the matched requests |
**Note:**
- If the `_rules_` field is not configured, authentication and authorization are enabled for all routes of the current gateway instance by default;
- For authenticated and authorized requests, a `X-Mse-Consumer` field is added to the request header to identify the caller's name.
## Configuration Example
### Enable for Specific Routes or Domains
The following configuration enables Jwt Auth authentication and authorization for specific routes or domains of the gateway. If a JWT can match multiple `jwks`, the first matching `consumer` is hit according to the configuration order.
```yaml
consumers:
- name: consumer1
issuer: abcd
jwks: |
{
"keys": [
{
"kty": "oct",
"kid": "123",
"k": "hM0k3AbXBPpKOGg__Ql2Obcq7s60myWDpbHXzgKUQdYo7YCRp0gUqkCnbGSvZ2rGEl4YFkKqIqW7mTHdj-bcqXpNr-NOznEyMpVPOIlqG_NWVC3dydBgcsIZIdD-MR2AQceEaxriPA_VmiUCwfwL2Bhs6_i7eolXoY11EapLQtutz0BV6ZxQQ4dYUmct--7PLNb4BWJyQeWu0QfbIthnvhYllyl2dgeLTEJT58wzFz5HeNMNz8ohY5K0XaKAe5cepryqoXLhA-V-O1OjSG8lCNdKS09OY6O0fkyweKEtuDfien5tHHSsHXoAxYEHPFcSRL4bFPLZ0orTt1_4zpyfew",
"alg": "HS256"
}
]
}
- name: consumer2
issuer: abc
jwks: |
{
"keys": [
{
"kty": "RSA",
"e": "AQAB",
"use": "sig",
"kid": "123",
"alg": "RS256",
"n": "i0B67f1jggT9QJlZ_8QL9QQ56LfurrqDhpuu8BxtVcfxrYmaXaCtqTn7OfCuca7cGHdrJIjq99rz890NmYFZuvhaZ-LMt2iyiSb9LZJAeJmHf7ecguXS_-4x3hvbsrgUDi9tlg7xxbqGYcrco3anmalAFxsbswtu2PAXLtTnUo6aYwZsWA6ksq4FL3-anPNL5oZUgIp3HGyhhLTLdlQcC83jzxbguOim-0OEz-N4fniTYRivK7MlibHKrJfO3xa_6whBS07HW4Ydc37ZN3Rx9Ov3ZyV0idFblU519nUdqp_inXj1eEpynlxH60Ys_aTU2POGZh_25KXGdF_ZC_MSRw"
}
]
}
# Use the _rules_ field for fine-grained rule configuration
_rules_:
# Rule 1: Effective when matched by route name
- _match_route_:
- route-a
- route-b
allow:
- consumer1
# Rule 2: Effective when matched by domain name
- _match_domain_:
- "*.example.com"
- test.com
allow:
- consumer2
```
In this example, the `route-a` and `route-b` specified in `_match_route_` are the names of the routes filled in when creating the gateway route. When these two routes are matched, access will be allowed for the caller with the `name` of `consumer1`, and other callers will not be allowed to access.
The `*.example.com` and `test.com` specified in `_match_domain_` are used to match the domain names of the requests. When a domain name match is found, access will be allowed for the caller with the `name` of `consumer2`, and other callers will not be allowed to access.
#### According to this configuration, the following requests are allowed:
Assuming the following requests will match the route-a route:
**JWT is set in URL parameter**
```bash
curl 'http://xxx.hello.com/test?access_token=eyJhbGciOiJIUzI1NiIsInR5cCI6IkpXVCIsImtpZCI6IjEyMyJ9.eyJpc3MiOiJhYmNkIiwic3ViIjoidGVzdCIsImlhdCI6MTY2NTY2MDUyNywiZXhwIjoxODY1NjczODE5fQ.-vBSV0bKeDwQcuS6eeSZN9dLTUnSnZVk8eVCXdooCQ4'
```
**JWT is set in HTTP request header**
```bash
curl http://xxx.hello.com/test -H 'Authorization: Bearer eyJhbGciOiJIUzI1NiIsInR5cCI6IkpXVCIsImtpZCI6IjEyMyJ9.eyJpc3MiOiJhYmNkIiwic3ViIjoidGVzdCIsImlhdCI6MTY2NTY2MDUyNywiZXhwIjoxODY1NjczODE5fQ.-vBSV0bKeDwQcuS6eeSZN9dLTUnSnZVk8eVCXdooCQ4'
```
After authentication and authorization, a `X-Mse-Consumer` field will be added in the request header with a value of `consumer1` in this example, to identify the name of the caller.
#### The following requests will be denied:
**Request without JWT provided, returns 401**
```bash
curl http://xxx.hello.com/test
```
**The consumer matched by the provided JWT in the request does not have access, returns 403**
```bash
# consumer1 is not in the allow list of *.example.com
curl 'http://xxx.example.com/test' -H 'Authorization: Bearer eyJhbGciOiJIUzI1NiIsInR5cCI6IkpXVCIsImtpZCI6IjEyMyJ9.eyJpc3MiOiJhYmNkIiwic3ViIjoidGVzdCIsImlhdCI6MTY2NTY2MDUyNywiZXhwIjoxODY1NjczODE5fQ.-vBSV0bKeDwQcuS6eeSZN9dLTUnSnZVk8eVCXdooCQ4'
```
### Enabling at Gateway Instance Level
The following configuration does not specify the `_rules_` field, so JWT authentication will be enabled at the gateway instance level:
```yaml
consumers:
- name: consumer1
issuer: abcd
jwks: |
{
"keys": [
{
"kty": "oct",
"kid": "123",
"k": "hM0k3AbXBPpKOGg__Ql2Obcq7s60myWDpbHXzgKUQdYo7YCRp0gUqkCnbGSvZ2rGEl4YFkKqIqW7mTHdj-bcqXpNr-NOznEyMpVPOIlqG_NWVC3dydBgcsIZIdD-MR2AQceEaxriPA_VmiUCwfwL2Bhs6_i7eolXoY11EapLQtutz0BV6ZxQQ4dYUmct--7PLNb4BWJyQeWu0QfbIthnvhYllyl2dgeLTEJT58wzFz5HeNMNz8ohY5K0XaKAe5cepryqoXLhA-V-O1OjSG8lCNdKS09OY6O0fkyweKEtuDfien5tHHSsHXoAxYEHPFcSRL4bFPLZ0orTt1_4zpyfew",
"alg": "HS256"
}
]
}
- name: consumer2
issuer: abc
jwks: |
{
"keys": [
{
"kty": "RSA",
"e": "AQAB",
"use": "sig",
"kid": "123",
"alg": "RS256",
"n": "i0B67f1jggT9QJlZ_8QL9QQ56LfurrqDhpuu8BxtVcfxrYmaXaCtqTn7OfCuca7cGHdrJIjq99rz890NmYFZuvhaZ-LMt2iyiSb9LZJAeJmHf7ecguXS_-4x3hvbsrgUDi9tlg7xxbqGYcrco3anmalAFxsbswtu2PAXLtTnUo6aYwZsWA6ksq4FL3-anPNL5oZUgIp3HGyhhLTLdlQcC83jzxbguOim-0OEz-N4fniTYRivK7MlibHKrJfO3xa_6whBS07HW4Ydc37ZN3Rx9Ov3ZyV0idFblU519nUdqp_inXj1eEpynlxH60Ys_aTU2POGZh_25KXGdF_ZC_MSRw"
}
]
}
```
# Common Error Codes
| HTTP Status Code | Error Message | Reason Description|
|------------------| ---------------------- | -------------------------------------------------------------------------------- |
| 401 | JWT missing | The JWT is not provided in the request header. |
| 401 | JWT expired | The JWT has expired. |
| 401 | JWT verification fails | The JWT payload verification failed, such as the iss mismatch. |
| 403 | Access denied | Access to the current route is denied. |

113
plugins/wasm-cpp/extensions/key_auth/README.md
View File

@@ -1,15 +1,32 @@
# 功能说明
---
title: Key 认证
keywords: [higress,key auth]
description: Key 认证插件配置参考
---
## 功能说明
`key-auth`插件实现了基于 API Key 进行认证鉴权的功能，支持从 HTTP 请求的 URL 参数或者请求头解析 API Key，同时验证该 API Key 是否有权限访问。
# 配置字段
## 运行属性
| 名称 | 数据类型 | 填写要求 | 默认值 | 描述 |
| ----------- | --------------- | ------------------------------------------- | ------ | ----------------------------------------------------------- |
| `consumers` | array of object | 必填 | - | 配置服务的调用者，用于对请求进行认证 |
| `keys` | array of string | 必填 | - | API Key 的来源字段名称，可以是 URL 参数或者 HTTP 请求头名称 |
| `in_query` | bool | `in_query` 和 `in_header` 至少有一个为 true | true | 配置 true 时，网关会尝试从 URL 参数中解析 API Key |
| `in_header` | bool | `in_query` 和 `in_header` 至少有一个为 true | true | 配置 true 时，网关会尝试从 HTTP 请求头中解析 API Key |
| `_rules_` | array of object | 选填 | - | 配置特定路由或域名的访问权限列表，用于对请求进行鉴权 |
插件执行阶段：`认证阶段`
插件执行优先级：`310`
## 配置字段
**注意：**
- 在一个规则里，鉴权配置和认证配置不可同时存在
- 对于通过认证鉴权的请求，请求的header会被添加一个`X-Mse-Consumer`字段，用以标识调用者的名称。
### 认证配置
| 名称 | 数据类型 | 填写要求 | 默认值 | 描述 |
| ----------- | --------------- | ------------------------------------------- | ------ | ----------------------------------------------------------- |
| `global_auth` | bool | 选填（**仅实例级别配置**） | - | 只能在实例级别配置，若配置为true，则全局生效认证机制; 若配置为false，则只对做了配置的域名和路由生效认证机制，若不配置则仅当没有域名和路由配置时全局生效（兼容老用户使用习惯）。 |
| `consumers` | array of object | 必填 | - | 配置服务的调用者，用于对请求进行认证 |
| `keys` | array of string | 必填 | - | API Key 的来源字段名称，可以是 URL 参数或者 HTTP 请求头名称 |
| `in_query` | bool | `in_query` 和 `in_header` 至少有一个为 true | true | 配置 true 时，网关会尝试从 URL 参数中解析 API Key |
| `in_header` | bool | `in_query` 和 `in_header` 至少有一个为 true | true | 配置 true 时，网关会尝试从 HTTP 请求头中解析 API Key |
`consumers`中每一项的配置字段说明如下：
@@ -18,25 +35,22 @@
| `credential` | string | 必填 | - | 配置该consumer的访问凭证 |
| `name` | string | 必填 | - | 配置该consumer的名称 |
`_rules_` 中每一项的配置字段说明如下：
### 鉴权配置（非必需）
| 名称 | 数据类型 | 填写要求 | 默认值 | 描述 |
| ---------------- | --------------- | ------------------------------------------------- | ------ | -------------------------------------------------- |
| `_match_route_` | array of string | 选填，`_match_route_`，`_match_domain_`中选填一项 | - | 配置要匹配的路由名称 |
| `_match_domain_` | array of string | 选填，`_match_route_`，`_match_domain_`中选填一项 | - | 配置要匹配的域名 |
| `allow` | array of string | 必填 | - | 对于符合匹配条件的请求，配置允许访问的consumer名称 |
| 名称 | 数据类型 | 填写要求 | 默认值 | 描述 |
| ----------- | --------------- | ------------------------------------------- | ------ | ----------------------------------------------------------- |
| `allow` | array of string | 选填(**非实例级别配置**) | - | 只能在路由或域名等细粒度规则上配置，对于符合匹配条件的请求，配置允许访问的 consumer，从而实现细粒度的权限控制 |
**注意：**
- 若不配置`_rules_`字段，则默认对当前网关实例的所有路由开启认证；
- 对于通过认证鉴权的请求，请求的header会被添加一个`X-Mse-Consumer`字段，用以标识调用者的名称。
## 配置示例
# 配置示例
### 全局配置认证和路由粒度进行鉴权
## 对特定路由或域名开启
以下配置将对网关特定路由或域名开启Key Auth认证和鉴权。credential字段不能重复。
以下配置将对网关特定路由或域名开启 Key Auth 认证和鉴权，注意`credential`字段不能重复
在实例级别做如下插件配置：
```yaml
global_auth: false
consumers:
- credential: 2bda943c-ba2b-11ec-ba07-00163e1250b5
name: consumer1
@@ -44,31 +58,33 @@ consumers:
name: consumer2
keys:
- apikey
in_query: true
# 使用 _rules_ 字段进行细粒度规则配置
_rules_:
# 规则一：按路由名称匹配生效
- _match_route_:
- route-a
- route-b
allow:
- consumer1
# 规则二：按域名匹配生效
- _match_domain_:
- "*.example.com"
- test.com
allow:
- consumer2
- x-api-key
```
此例 `_match_route_` 中指定的 `route-a` 和 `route-b` 即在创建网关路由时填写的路由名称，当匹配到这两个路由时，将允许`name`为`consumer1`的调用者访问，其他调用者不允许访问；
对 route-a 和 route-b 这两个路由做如下配置：
此例 `_match_domain_` 中指定的 `*.example.com` 和 `test.com` 用于匹配请求的域名，当发现域名匹配时，将允许`name`为`consumer2`的调用者访问，其他调用者不允许访问。
```yaml
allow:
- consumer1
```
### 根据该配置，下列请求可以允许访问：
对 *.example.com 和 test.com 在这两个域名做如下配置:
```yaml
allow:
- consumer2
```
**说明：**
此例指定的route-a和route-b即在创建网关路由时填写的路由名称，当匹配到这两个路由时，将允许name为consumer1的调用者访问，其他调用者不允许访问。
此例指定的*.example.com和test.com用于匹配请求的域名，当发现域名匹配时，将允许name为consumer2的调用者访问，其他调用者不被允许访问。
根据该配置，下列请求可以允许访问：
假设以下请求会匹配到route-a这条路由
n
**将 API Key 设置在 url 参数中**
```bash
curl http://xxx.hello.com/test?apikey=2bda943c-ba2b-11ec-ba07-00163e1250b5
@@ -80,7 +96,7 @@ curl http://xxx.hello.com/test -H 'x-api-key: 2bda943c-ba2b-11ec-ba07-00163e125
认证鉴权通过后，请求的header中会被添加一个`X-Mse-Consumer`字段，在此例中其值为`consumer1`，用以标识调用方的名称
### 下列请求将拒绝访问：
下列请求将拒绝访问：
**请求未提供 API Key，返回401**
```bash
@@ -97,11 +113,12 @@ curl http://xxx.hello.com/test?apikey=926d90ac-ba2e-11ec-ab68-00163e1250b5
curl http://xxx.hello.com/test?apikey=c8c8e9ca-558e-4a2d-bb62-e700dcc40e35
```
## 网关实例级别开启
### 网关实例级别开启
以下配置未指定`_rules_`字段，因此将对网关实例级别开启 Key Auth 认证
以下配置将对网关实例级别开启 Basic Auth 认证，所有请求均需要经过认证后才能访问。
```yaml
global_auth: true
consumers:
- credential: 2bda943c-ba2b-11ec-ba07-00163e1250b5
name: consumer1
@@ -109,13 +126,15 @@ consumers:
name: consumer2
keys:
- apikey
in_query: true
- x-api-key
```
# 相关错误码
## 相关错误码
| HTTP 状态码 | 出错信息 | 原因说明 |
| ----------- | --------------------------------------------------------- | ----------------------- |
| 401 | No API key found in request | 请求未提供 API Key |
| 401 | Request denied by Key Auth check. Muti API key found in request | 请求提供多个 API Key |
| 401 | Request denied by Key Auth check. No API key found in request | 请求未提供 API Key |
| 401 | Request denied by Key Auth check. Invalid API key | 不允许当前 API Key 访问 |
| 403 | Request denied by Basic Auth check. Unauthorized consumer | 请求的调用方无访问权限 |
| 403 | Request denied by Key Auth check. Unauthorized consumer | 请求的调用方无访问权限 |

144
plugins/wasm-cpp/extensions/key_auth/README_EN.md
View File

@@ -1,43 +1,47 @@
# Features
The `key-auth` plug-in implements the authentication function based on the API Key, supports parsing the API Key from the URL parameter or request header of the HTTP request, and verifies whether the API Key has permission to access.
---
title: Key Authentication
keywords: [higress,key auth]
description: Key Authentication Plugin Configuration Reference
---
## Function Description
The `key-auth` plugin implements authentication based on API Key, supporting the parsing of the API Key from HTTP request URL parameters or request headers, while also verifying whether the API Key has permission to access the resource.
# Configuration field
## Runtime Properties
Plugin Execution Phase: `Authentication Phase`
Plugin Execution Priority: `310`
| Name | Data Type | Parameter requirements | Default| Description |
| ----------- | --------------- | -------------------------------------------------------- | ------ | --------------------------------------------------------------------------------------------------------- |
| `consumers` | array of object | Required | - | Configure the caller of the service to authenticate the request. |
| `keys` | array of string | Required | - | The name of the source field of the API Key, which can be a URL parameter or an HTTP request header name. |
| `in_query` | bool | At least one of `in_query` and `in_header` must be true. | true | When configured true, the gateway will try to parse the API Key from the URL parameters. |
| `in_header` | bool | The same as above. | true | The same as above. |
| `_rules_` | array of object | Optional | - | Configure the access list of a specific route or domain name for authenticating requests. |
## Configuration Fields
**Note:**
- Authentication and authorization configurations cannot coexist within a single rule.
- For requests that are authenticated, a header field `X-Mse-Consumer` will be added to identify the caller's name.
### Authentication Configuration
| Name | Data Type | Requirements | Default Value | Description |
| ------------- | ---------------- | ----------------------------------------------- | ------------- | ---------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------- |
| `global_auth` | bool | Optional (**Instance-Level Configuration Only**) | - | Can only be configured at the instance level; if set to true, the authentication mechanism takes effect globally; if set to false, it only applies to the configured hostnames and routes. If not configured, it will only take effect globally when no hostname and route configurations are present (to maintain compatibility with older user habits). |
| `consumers` | array of object | Required | - | Configures the service callers for request authentication. |
| `keys` | array of string | Required | - | Source field names for the API Key, which can be URL parameters or HTTP request header names. |
| `in_query` | bool | At least one of `in_query` and `in_header` must be true | true | When configured as true, the gateway will attempt to parse the API Key from URL parameters. |
| `in_header` | bool | At least one of `in_query` and `in_header` must be true | true | When configured as true, the gateway will attempt to parse the API Key from HTTP request headers. |
The configuration fields of each item in `consumers` are described as follows:
The configuration field descriptions for each item in `consumers` are as follows:
| Name | Data Type | Requirements | Default Value | Description |
| ------------ | --------- | ------------ | ------------- | ------------------------------ |
| `credential` | string | Required | - | Configures the access credential for this consumer. |
| `name` | string | Required | - | Configures the name for this consumer. |
| Name | Data Type | Parameter requirements | Default | Description |
| ------------ | --------- | -----------------------| ------ | ------------------------------------------- |
| `credential` | string | Required | - | Configure the consumer's access credentials. |
| `name` | string | Required | - | Configure the name of the consumer. |
### Authorization Configuration (Optional)
| Name | Data Type | Requirements | Default Value | Description |
| ----------- | ---------------- | ----------------------------------------------- | ------------- | --------------------------------------------------------------------------------------------------------------------------------------------------------------------- |
| `allow` | array of string | Optional (**Non-Instance Level Configuration**) | - | Can only be configured on fine-grained rules such as routes or hostnames; specifies the allowed consumers for matching requests, allowing for fine-grained permission control. |
The configuration fields of each item in `_rules_` are described as follows:
| Name | Data Type | Parameter requirements | Default| Description |
| ---------------- | --------------- | --------------------------------------------------------------------- | ------ | -------------------------------------------------- |
| `_match_route_` | array of string | Optional，Optionally fill in one of `_match_route_`, `_match_domain_`. | - | Configure the route name to match. |
| `_match_domain_` | array of string | Optional，Optionally fill in one of `_match_route_`, `_match_domain_`. | - | Configure the domain name to match. |
| `allow` | array of string | Required | - | For requests that meet the matching conditions, configure the name of the consumer that is allowed to access. |
**Warning：**
- If the `_rules_` field is not configured, authentication will be enabled for all routes of the current gateway instance by default;
- For a request that passes authentication, an `X-Mse-Consumer` field will be added to the request header to identify the name of the caller.
# Example configuration
## Enabled for specific routes or domains
The following configuration will enable Key Auth authentication and authentication for gateway-specific routes or domain names. Note that the `credential` field can not be repeated.
## Configuration Example
### Global Configuration for Authentication and Granular Route Authorization
The following configuration will enable Key Auth authentication and authorization for specific routes or hostnames in the gateway. The `credential` field must not repeat.
At the instance level, do the following plugin configuration:
```yaml
global_auth: false
consumers:
- credential: 2bda943c-ba2b-11ec-ba07-00163e1250b5
name: consumer1
@@ -45,64 +49,63 @@ consumers:
name: consumer2
keys:
- apikey
in_query: true
# Use the _rules_ field for fine-grained rule configuration
_rules_:
# Rule 1: Match by route name to take effect
- _match_route_:
- route-a
- route-b
allow:
- consumer1
# Rule 2: Take effect by domain name matching
- _match_domain_:
- "*.example.com"
- test.com
allow:
- consumer2
- x-api-key
```
The `route-a` and `route-b` specified in `_match_route_` in this example are the route names filled in when creating the gateway route. When these two routes are matched, calls whose `name` is `consumer1` will be allowed Access by callers, other callers are not allowed to access;
For routes route-a and route-b, do the following configuration:
```yaml
allow:
- consumer1
```
`*.example.com` and `test.com` specified in `_match_domain_` in this example are used to match the domain name of the request. When the domain name matches, the caller whose `name` is `consumer2` will be allowed to access, and other calls access is not allowed.
For the hostnames *.example.com and test.com, do the following configuration:
```yaml
allow:
- consumer2
```
### Depending on this configuration, the following requests would allow access：
**Note:**
The routes route-a and route-b specified in this example refer to the route names filled in when creating the gateway routes. When matched with these two routes, requests from the caller named consumer1 will be allowed while others will be denied.
Assume that the following request will match the route-a route:
The specified hostnames *.example.com and test.com are used to match the request's domain name. When a domain name is matched, callers named consumer2 will be allowed while others will be denied.
**Set the API Key in the url parameter**
Based on this configuration, the following requests will be allowed:
Assuming the following request matches route-a:
**Setting API Key in URL Parameters**
```bash
curl http://xxx.hello.com/test?apikey=2bda943c-ba2b-11ec-ba07-00163e1250b5
```
**Set the API Key in the http request header**
**Setting API Key in HTTP Request Headers**
```bash
curl http://xxx.hello.com/test -H 'x-api-key: 2bda943c-ba2b-11ec-ba07-00163e1250b5'
```
After the authentication is passed, an `X-Mse-Consumer` field will be added to the header of the request. In this example, its value is `consumer1`, which is used to identify the name of the caller.
After successful authentication and authorization, the request's header will have an added `X-Mse-Consumer` field with the value `consumer1`, to identify the name of the caller.
### The following requests will deny access：
**The request does not provide an API Key, return 401**
The following requests will be denied access:
**Request without an API Key returns 401**
```bash
curl http://xxx.hello.com/test
```
**The API Key provided by the request is not authorized to access, return 401**
**Request with an invalid API Key returns 401**
```bash
curl http://xxx.hello.com/test?apikey=926d90ac-ba2e-11ec-ab68-00163e1250b5
```
**The caller matched according to the API Key provided in the request has no access rights, return 403**
**Caller matched with provided API Key has no access rights, returns 403**
```bash
# consumer2 is not in the allow list of route-a
curl http://xxx.hello.com/test?apikey=c8c8e9ca-558e-4a2d-bb62-e700dcc40e35
```
## Gateway instance level enabled
The following configuration does not specify the `_rules_` field, so Key Auth authentication will be enabled at the gateway instance level.
### Enabling at the Instance Level
The following configuration will enable Basic Auth authentication at the instance level for the gateway, requiring all requests to pass authentication before accessing.
```yaml
global_auth: true
consumers:
- credential: 2bda943c-ba2b-11ec-ba07-00163e1250b5
name: consumer1
@@ -110,14 +113,13 @@ consumers:
name: consumer2
keys:
- apikey
in_query: true
- x-api-key
```
# Error code
| HTTP status code | Error information | Reason |
| ---------------- | --------------------------------------------------------- | -------------------------------------------- |
| 401 | No API key found in request. | API not provided by request Key. |
| 401 | Request denied by Key Auth check. Invalid API key. | Current API Key access is not allowed. |
| 403 | Request denied by Basic Auth check. Unauthorized consumer. | The requested caller does not have access. |
## Related Error Codes
| HTTP Status Code | Error Message | Reason Explanation |
| ---------------- | ---------------------------------------------------------- | --------------------------------- |
| 401 | Request denied by Key Auth check. Multiple API keys found in request | Multiple API Keys provided in the request. |
| 401 | Request denied by Key Auth check. No API key found in request | API Key not provided in the request. |
| 401 | Request denied by Key Auth check. Invalid API key | The current API Key is not authorized for access. |
| 403 | Request denied by Key Auth check. Unauthorized consumer | The caller does not have access permissions. |

49
plugins/wasm-cpp/extensions/key_rate_limit/README.md
View File

@@ -1,11 +1,18 @@
<p>
<a href="README_EN.md"> English </a> | 中文
</p>
---
title: 基于 Key 的本地限流
keywords: [higress,key rate limit]
description: Key 本地限流插件配置参考
---
# 功能说明
## 功能说明
`key-rate-limit`插件实现了基于特定键值实现限流，键值来源可以是 URL 参数、HTTP 请求头
# 配置字段
## 运行属性
插件执行阶段：`默认阶段`
插件执行优先级：`10`
## 配置字段
| 名称 | 数据类型 | 填写要求 | 默认值 | 描述 |
| -------- | -------- | -------- | -------- | -------- |
@@ -23,9 +30,9 @@
| query_per_hour | number | 选填，`query_per_second`,`query_per_minute`,`query_per_hour`,`query_per_day` 中选填一项 | - | 允许每小时请求次数 |
| query_per_day | number | 选填，`query_per_second`,`query_per_minute`,`query_per_hour`,`query_per_day` 中选填一项 | - | 允许每天请求次数 |
# 配置示例
## 配置示例
## 识别请求参数 apikey，进行区别限流
### 识别请求参数 apikey，进行区别限流
```yaml
limit_by_param: apikey
limit_keys:
@@ -35,7 +42,7 @@ limit_keys:
query_per_minute: 100
```
## 识别请求头 x-ca-key，进行区别限流
### 识别请求头 x-ca-key，进行区别限流
```yaml
limit_by_header: x-ca-key
limit_keys:
@@ -45,29 +52,3 @@ limit_keys:
query_per_hour: 10
```
## 对特定路由或域名开启
```yaml
# 使用 _rules_ 字段进行细粒度规则配置
_rules_:
# 规则一：按路由名称匹配生效
- _match_route_:
- route-a
- route-b
limit_by_header: x-ca-key
limit_keys:
- key: 102234
query_per_second: 10
# 规则二：按域名匹配生效
- _match_domain_:
- "*.example.com"
- test.com
limit_by_header: x-ca-key
limit_keys:
- key: 102234
query_per_second: 100
```
此例 `_match_route_` 中指定的 `route-a` 和 `route-b` 即在创建网关路由时填写的路由名称，当匹配到这两个路由时，将使用此段配置；
此例 `_match_domain_` 中指定的 `*.example.com` 和 `test.com` 用于匹配请求的域名，当发现域名匹配时，将使用此段配置；
配置的匹配生效顺序，将按照 `_rules_` 下规则的排列顺序，匹配第一个规则后生效对应配置，后续规则将被忽略。

74
plugins/wasm-cpp/extensions/key_rate_limit/README_EN.md
View File

@@ -1,31 +1,36 @@
<p>
English | <a href="README.md">中文</a>
</p>
---
title: Key-based Local Rate Limiting
keywords: [higress,key rate limit]
description: Configuration reference for Key local rate limiting plugin
---
# Description
`key-rate-limit` plugin implements a rate-limiting function based on specific key-values. The key-values may come from URL parameters or HTTP headers.
## Functional Description
The `key-rate-limit` plugin implements rate limiting based on specific key values, which can originate from URL parameters or HTTP request headers.
# Configuration Fields
## Running Properties
Plugin execution phase: `default phase`
Plugin execution priority: `10`
| Name | Type | Requirement | Default Value | Description |
| -------- | -------- | -------- | -------- | -------- |
| limit_by_header | string | Optional. Choose one from following: `limit_by_header`, `limit_by_param`. | - | The name of HTTP header used to obtain key-value used in rate-limiting. |
| limit_by_param | string | Optional. Choose one from following: `limit_by_header`, `limit_by_param`. | - | The name of URL parameter used to obtain key-value used in rate-limiting. |
| limit_keys | array of object | Required | - | Rate-limiting thresholds when matching specific key-values |
## Configuration Fields
Field descriptions of `limit_keys` items:
| Name | Data Type | Required | Default Value | Description |
|-----------------|-----------------|---------------------------------------------------------------|---------------|----------------------------------------------------------------------------------------|
| limit_by_header | string | Optional, choose one from `limit_by_header`, `limit_by_param` | - | Configuration for the source of the rate limiting key value (HTTP request header name) |
| limit_by_param | string | Optional, choose one from `limit_by_header`, `limit_by_param` | - | Configuration for the source of the rate limiting key value (URL parameter name) |
| limit_keys | array of object | Required | - | Configuration for the rate limiting frequency based on matched key values |
| Name | Type | Requirement | Default Value | Description |
| -------- | -------- | -------- | -------- | -------- |
| key | string | Required | - | Value to match of the specific key |
| query_per_second | number | Optional. Choose one from following: `query_per_second`, `query_per_minute`, `query_per_hour`, `query_per_day`. | - | Number of requests allowed per second |
| query_per_minute | number | Optional. Choose one from following: `query_per_second`, `query_per_minute`, `query_per_hour`, `query_per_day`. | - | Number of requests allowed per minute |
| query_per_hour | number | Optional. Choose one from following: `query_per_second`, `query_per_minute`, `query_per_hour`, `query_per_day`. | - | Number of requests allowed per hour |
| query_per_day | number | Optional. Choose one from following: `query_per_second`, `query_per_minute`, `query_per_hour`, `query_per_day`. | - | Number of requests allowed per day |
Explanation of each configuration field in `limit_keys`
# Configuration Samples
| Name | Data Type | Required | Default Value | Description |
|------------------|-----------|-----------------------------------------------------------------------------------------------------|---------------|---------------------------------------|
| key | string | Required | - | Matched key value |
| query_per_second | number | Optional, choose one from `query_per_second`, `query_per_minute`, `query_per_hour`, `query_per_day` | - | Allowed number of requests per second |
| query_per_minute | number | Optional, choose one from `query_per_second`, `query_per_minute`, `query_per_hour`, `query_per_day` | - | Allowed number of requests per minute |
| query_per_hour | number | Optional, choose one from `query_per_second`, `query_per_minute`, `query_per_hour`, `query_per_day` | - | Allowed number of requests per hour |
| query_per_day | number | Optional, choose one from `query_per_second`, `query_per_minute`, `query_per_hour`, `query_per_day` | - | Allowed number of requests per day |
## Use query parameter `apikey` for rate-limiting
## Configuration Examples
### Identify request parameter apikey for differentiated rate limiting
```yaml
limit_by_param: apikey
limit_keys:
@@ -35,7 +40,7 @@ limit_keys:
query_per_minute: 100
```
## Use HTTP header parameter `x-ca-key` for rate-limiting
### Identify request header x-ca-key for differentiated rate limiting
```yaml
limit_by_header: x-ca-key
limit_keys:
@@ -44,28 +49,3 @@ limit_keys:
- key: 308239
query_per_hour: 10
```
## Enable rate-limiting for specific routes or domains
```yaml
# Use _rules_ field for fine-grained rule configurations
_rules_:
# Rule 1: Match by route name
- _match_route_:
- route-a
- route-b
limit_by_header: x-ca-key
limit_keys:
- key: 102234
query_per_second: 10
# Rule 2: Match by domain
- _match_domain_:
- "*.example.com"
- test.com
limit_by_header: x-ca-key
limit_keys:
- key: 102234
query_per_second: 100
```
In the rule sample of `_match_route_`, `route-a` and `route-b` are the route names provided when creating a new gateway route. When the current route names matches the configuration, the rule following shall be applied.
In the rule sample of `_match_domain_`, `*.example.com` and `test.com` are the domain names used for request matching. When the current domain name matches the configuration, the rule following shall be applied.
All rules shall be checked following the order of items in the `_rules_` field, The first matched rule will be applied. All remained will be ignored.

161
plugins/wasm-cpp/extensions/oauth/README.md
View File

@@ -1,21 +1,32 @@
# 功能说明
---
title: OAuth2 认证
keywords: [higress,oauth2]
description: OAuth2 认证插件配置参考
---
## 功能说明
`OAuth2`插件实现了基于JWT(JSON Web Tokens)进行OAuth2 Access Token签发的能力, 遵循[RFC9068](https://datatracker.ietf.org/doc/html/rfc9068)规范
# 插件配置说明
## 运行属性
插件执行阶段：`认证阶段`
插件执行优先级：`350`
## 配置字段
| 名称 | 数据类型 | 填写要求 | 默认值 | 描述 |
| ----------- | --------------- | ------------------------------------------- | ------ | ----------------------------------------------------------- |
| `consumers` | array of object | 必填 | - | 配置服务的调用者，用于对请求进行认证 |
| `_rules_` | array of object | 选填 | - | 配置特定路由或域名的访问权限列表，用于对请求进行鉴权 |
| `issuer` | string | 选填 | Higress-Gateway | 用于填充JWT中的issuer |
| `auth_path` | string | 选填 | /oauth2/token | 指定路径后缀用于签发Token，路由级配置时，要确保首先能匹配对应的路由 |
| `global_credentials` | bool | 选填 | ture | 是否开启全局凭证，即允许路由A下的auth_path签发的Token可以用于访问路由B |
| `auth_header_name` | string | 选填 | Authorization | 用于指定从哪个请求头获取JWT |
| `token_ttl` | number | 选填 | 7200 | token从签发后多久内有效，单位为秒 |
| `clock_skew_seconds` | number | 选填 | 60 | 校验JWT的exp和iat字段时允许的时钟偏移量，单位为秒 |
| `keep_token` | bool | 选填 | ture | 转发给后端时是否保留JWT |
### 认证配置
| 名称 | 数据类型 | 填写要求 | 默认值 | 描述 |
| ----------- | --------------- | ------------------------------------------- | ------ | ----------------------------------------------------------- |
| `consumers` | array of object | 必填 | - | 配置服务的调用者，用于对请求进行认证 |
| `issuer` | string | 选填 | Higress-Gateway | 用于填充JWT中的issuer |
| `auth_path` | string | 选填 | /oauth2/token | 指定路径后缀用于签发Token，路由级配置时，要确保首先能匹配对应的路由, 使用 API 管理时，需要创建相同路径的接口 |
| `global_credentials` | bool | 选填 | ture | 在通过 consumer 认证的前提下，允许任意路由签发的凭证访问 |
| `auth_header_name` | string | 选填 | Authorization | 用于指定从哪个请求头获取JWT |
| `token_ttl` | number | 选填 | 7200 | token从签发后多久内有效，单位为秒 |
| `clock_skew_seconds` | number | 选填 | 60 | 校验JWT的exp和iat字段时允许的时钟偏移量，单位为秒 |
| `keep_token` | bool | 选填 | ture | 转发给后端时是否保留JWT |
| `global_auth` | array of string | 选填(**仅实例级别配置**) | - | 只能在实例级别配置，若配置为true，则全局生效认证机制; 若配置为false，则只对做了配置的域名和路由生效认证机制; 若不配置则仅当没有域名和路由配置时全局 生效（兼容老用户使用习惯） |
`consumers`中每一项的配置字段说明如下：
@@ -25,27 +36,53 @@
| `client_id` | string | 必填 | - | OAuth2 client id |
| `client_secret` | string | 必填 | - | OAuth2 client secret |
`_rules_` 中每一项的配置字段说明如下：
**注意：**
- 对于开启该配置的路由，如果路径后缀和`auth_path`匹配，则该路由不会到原目标服务，而是用于生成Token
- 如果关闭`global_credentials`,请确保启用此插件的路由不是精确匹配路由，此时若存在另一条前缀匹配路由，则可能导致预期外行为
- 对于通过认证鉴权的请求，请求的header会被添加一个`X-Mse-Consumer`字段，用以标识调用者的名称。
| 名称 | 数据类型 | 填写要求 | 默认值 | 描述 |
| ---------------- | --------------- | ------------------------------------------------- | ------ | -------------------------------------------------- |
| `_match_route_` | array of string | 选填，`_match_route_`，`_match_domain_`中选填一项 | - | 配置要匹配的路由名称 |
| `_match_domain_` | array of string | 选填，`_match_route_`，`_match_domain_`中选填一项 | - | 配置要匹配的域名 |
| `allow` | array of string | 必填 | - | 对于符合匹配条件的请求，配置允许访问的consumer名称 |
### 鉴权配置（非必需）
| 名称 | 数据类型 | 填写要求 | 默认值 | 描述 |
| ----------- | --------------- | ------------------------------------------- | ------ | ----------------------------------------------------------- |
| `allow` | array of string | 选填(**非实例级别配置**) | - | 只能在路由或域名等细粒度规则上配置，对于符合匹配条件的请求，配置允许访问的 consumer，从而实现细粒度的权限控制 |
**注意：**
- 对于开启该配置的路由，如果路径后缀和`auth_path`匹配，则该路由到原目标服务，而是用于生成Token
- 如果关闭`global_credentials`,请确保启用此插件的路由不是精确匹配路由，此时若存在另一条前缀匹配路由，则可能导致预期外行为
- 若不配置`_rules_`字段，则默认对当前网关实例的所有路由开启认证；
- 对于通过认证鉴权的请求，请求的header会被添加一个`X-Mse-Consumer`字段，用以标识调用者的名称。
- 在一个规则里，鉴权配置和认证配置不可同时存在
## 配置示例
### 对特定路由或域名开启
### 路由粒度配置认证
在`route-a`和`route-b`两个路由做如下插件配置：
```yaml
consumers:
- name: consumer1
client_id: 12345678-xxxx-xxxx-xxxx-xxxxxxxxxxxx
client_secret: abcdefgh-xxxx-xxxx-xxxx-xxxxxxxxxxxx
```
此时虽然使用同一份配置，但`route-a` 下签发的凭证无法用于访问 `route-b`，反之亦然。
如果希望同一份配置共享凭证访问权限，可以做如下配置:
```yaml
global_credentials: true
consumers:
- name: consumer1
client_id: 12345678-xxxx-xxxx-xxxx-xxxxxxxxxxxx
client_secret: abcdefgh-xxxx-xxxx-xxxx-xxxxxxxxxxxx
```
### 全局配置认证，路由粒度进行鉴权
以下配置将对网关特定路由或域名开启 Jwt Auth 认证和鉴权，注意如果一个JWT能匹配多个`jwks`，则按照配置顺序命中第一个匹配的`consumer`
在实例级别做如下插件配置：
```yaml
global_auth: false
consumers:
- name: consumer1
client_id: 12345678-xxxx-xxxx-xxxx-xxxxxxxxxxxx
@@ -53,37 +90,54 @@ consumers:
- name: consumer2
client_id: 87654321-xxxx-xxxx-xxxx-xxxxxxxxxxxx
client_secret: hgfedcba-xxxx-xxxx-xxxx-xxxxxxxxxxxx
# 使用 _rules_ 字段进行细粒度规则配置
_rules_:
# 规则一：按路由名称匹配生效
- _match_route_:
- route-a
- route-b
allow:
- consumer1
# 规则二：按域名匹配生效
- _match_domain_:
- "*.example.com"
- test.com
allow:
- consumer2
```
此例 `_match_route_` 中指定的 `route-a` 和 `route-b` 即在创建网关路由时填写的路由名称，当匹配到这两个路由时，将允许`name`为`consumer1`的调用者访问，其他调用者不允许访问；
在`route-a`和`route-b`两个路由做如下插件配置：
此例 `_match_domain_` 中指定的 `*.example.com` 和 `test.com` 用于匹配请求的域名，当发现域名匹配时，将允许`name`为`consumer2`的调用者访问，其他调用者不允许访问。
```yaml
allow:
- consumer1
```
#### 使用 Client Credential 授权模式
在`*.exmaple.com`和`test.com`两个域名做如下插件配置：
**获取 AccessToken**
```yaml
allow:
- consumer2
```
此例指定的 `route-a` 和 `route-b` 即在创建网关路由时填写的路由名称，当匹配到这两个路由时，将允许`name`为`consumer1`的调用者访问，其他调用者不允许访问；
此例指定的 `*.example.com` 和 `test.com` 用于匹配请求的域名，当发现域名匹配时，将允许`name`为`consumer2`的调用者访问，其他调用者不允许访问。
### 网关实例级别开启
以下配置将对网关实例级别开启 OAuth2 认证，所有请求均需要经过认证后才能访问
```yaml
global_auth: true
consumers:
- name: consumer1
client_id: 12345678-xxxx-xxxx-xxxx-xxxxxxxxxxxx
client_secret: abcdefgh-xxxx-xxxx-xxxx-xxxxxxxxxxxx
- name: consumer2
client_id: 87654321-xxxx-xxxx-xxxx-xxxxxxxxxxxx
client_secret: hgfedcba-xxxx-xxxx-xxxx-xxxxxxxxxxxx
```
# 请求示例
## 使用 Client Credential 授权模式
### 获取 AccessToken
```bash
# 通过 GET 方法获取
# 通过 GET 方法获取（推荐）
curl 'http://test.com/oauth2/token?grant_type=client_credentials&client_id=12345678-xxxx-xxxx-xxxx-xxxxxxxxxxxx&client_secret=abcdefgh-xxxx-xxxx-xxxx-xxxxxxxxxxxx'
# 通过 POST 方法获取 (需要先匹配到有真实目标服务的路由)
# 通过 POST 方法获取（需要先匹配到有真实目标服务的路由，否则网关不会读取请求 Body）
curl 'http://test.com/oauth2/token' -H 'content-type: application/x-www-form-urlencoded' -d 'grant_type=client_credentials&client_id=12345678-xxxx-xxxx-xxxx-xxxxxxxxxxxx&client_secret=abcdefgh-xxxx-xxxx-xxxx-xxxxxxxxxxxx'
@@ -96,29 +150,13 @@ curl 'http://test.com/oauth2/token' -H 'content-type: application/x-www-form-url
```
**使用 AccessToken 请求**
### 使用 AccessToken 请求
```bash
curl 'http://test.com' -H 'Authorization: Bearer eyJhbGciOiJIUzI1NiIsInR5cCI6ImFwcGxpY2F0aW9uXC9hdCtqd3QifQ.eyJhdWQiOiJkZWZhdWx0IiwiY2xpZW50X2lkIjoiMTIzNDU2NzgteHh4eC14eHh4LXh4eHgteHh4eHh4eHh4eHh4IiwiZXhwIjoxNjg3OTUxNDYzLCJpYXQiOjE2ODc5NDQyNjMsImlzcyI6IkhpZ3Jlc3MtR2F0ZXdheSIsImp0aSI6IjEwOTU5ZDFiLThkNjEtNGRlYy1iZWE3LTk0ODEwMzc1YjYzYyIsInN1YiI6ImNvbnN1bWVyMSJ9.NkT_rG3DcV9543vBQgneVqoGfIhVeOuUBwLJJ4Wycb0'
```
因为 test.com 仅授权了 consumer2，但这个 Access Token 是基于 consumer1 的 `client_id`，`client_secret` 获取的，因此将返回 `403 Access Denied`
### 网关实例级别开启
以下配置未指定`_rules_`字段，因此将对网关实例级别开启 OAuth2 认证
```yaml
consumers:
- name: consumer1
client_id: 12345678-xxxx-xxxx-xxxx-xxxxxxxxxxxx
client_secret: abcdefgh-xxxx-xxxx-xxxx-xxxxxxxxxxxx
- name: consumer2
client_id: 87654321-xxxx-xxxx-xxxx-xxxxxxxxxxxx
client_secret: hgfedcba-xxxx-xxxx-xxxx-xxxxxxxxxxxx
```
# 常见错误码说明
@@ -126,4 +164,3 @@ consumers:
| ----------- | ---------------------- | -------------------------------------------------------------------------------- |
| 401 | Invalid Jwt token | 请求头未提供JWT, 或者JWT格式错误，或过期等原因 |
| 403 | Access Denied | 无权限访问当前路由 |

138
plugins/wasm-cpp/extensions/oauth/README_EN.md Normal file
View File

@@ -0,0 +1,138 @@
---
title: OAuth2 Authentication
keywords: [higress,oauth2]
description: OAuth2 authentication plugin configuration reference
---
## Function Description
`OAuth2` plugin implements the capability of issuing OAuth2 Access Tokens based on JWT (JSON Web Tokens), complying with the [RFC9068](https://datatracker.ietf.org/doc/html/rfc9068) specification.
## Runtime Properties
Plugin execution phase: `Authentication Phase`
Plugin execution priority: `350`
## Configuration Fields
### Authentication Configuration
| Name | Data Type | Requirement | Default Value | Description |
| -------------------- | ---------------- | ------------------------------------------- | ---------------- | --------------------------------------------------------------------------------------------------------------------------------------------------------------------------------- |
| `consumers` | array of object | Required | - | Configures the callers of the service for request authentication |
| `issuer` | string | Optional | Higress-Gateway | Used to fill the issuer in the JWT |
| `auth_path` | string | Optional | /oauth2/token | Specifies the path suffix for issuing Tokens. When configured at the routing level, ensure it matches the corresponding route first. When using API management, create an interface with the same path. |
| `global_credentials` | bool | Optional | true | Allows any route to issue credentials for access under the condition of authentication through the consumer. |
| `auth_header_name` | string | Optional | Authorization | Specifies which request header to retrieve the JWT from |
| `token_ttl` | number | Optional | 7200 | The time duration in seconds for which the token is valid after issuance. |
| `clock_skew_seconds` | number | Optional | 60 | Allowed clock skew when verifying the exp and iat fields of the JWT, in seconds. |
| `keep_token` | bool | Optional | true | Indicates whether to keep the JWT when forwarding to the backend. |
| `global_auth` | array of string | Optional (**Instance-level configuration only**) | - | Can only be configured at the instance level. If set to true, the global authentication mechanism takes effect; if false, the authentication mechanism only takes effect for configured domains and routes; if not configured, global effect occurs only when there are no domain and route configurations (compatible with legacy user habits). |
The configuration fields for each item in `consumers` are as follows:
| Name | Data Type | Requirement | Default Value | Description |
| ----------------------- | ------------------| ----------- | ------------------------------------------------- | ---------------------------------- |
| `name` | string | Required | - | Configures the name of the consumer. |
| `client_id` | string | Required | - | OAuth2 client id |
| `client_secret` | string | Required | - | OAuth2 client secret |
**Note:**
- For routes with this configuration enabled, if the path suffix matches `auth_path`, the route will not forward to the original target service but will be used to generate a Token.
- If `global_credentials` is disabled, please ensure that the routes enabling this plugin do not precisely match routes. If there is another prefix-matching route, it may lead to unexpected behavior.
- For requests authenticated and authorized, the request header will have an `X-Mse-Consumer` field added to identify the caller's name.
### Authorization Configuration (Optional)
| Name | Data Type | Requirement | Default Value | Description |
| ----------- | ---------------- | ---------------------------------------------- | -------------- | ------------------------------------------------------------------------------------------------------------------------------------------------------------------- |
| `allow` | array of string | Optional (**Non-instance-level configuration**) | - | Can only be configured on fine-grained rules such as routes or domains, allowing specified consumers to access requests that meet the matching conditions for fine-grained permission control. |
**Note:**
- Authentication and authorization configurations cannot coexist in one rule.
## Configuration Example
### Route Granularity Configuration Authentication
For the two routes `route-a` and `route-b`, do the following plugin configuration:
```yaml
consumers:
- name: consumer1
client_id: 12345678-xxxx-xxxx-xxxx-xxxxxxxxxxxx
client_secret: abcdefgh-xxxx-xxxx-xxxx-xxxxxxxxxxxx
```
At this time, although using the same configuration, the credentials issued under `route-a` cannot be used to access `route-b`, and vice versa.
If you want the same configuration to share credential access permissions, you can configure as follows:
```yaml
global_credentials: true
consumers:
- name: consumer1
client_id: 12345678-xxxx-xxxx-xxxx-xxxxxxxxxxxx
client_secret: abcdefgh-xxxx-xxxx-xxxx-xxxxxxxxxxxx
```
### Global Configuration Authentication, Route Granularity Authorization
The following configuration will enable Jwt Auth for specific routes or domains on the gateway. Note that if a JWT matches multiple `jwks`, it will hit the first matching `consumer` in the order of configuration.
At the instance level, do the following plugin configuration:
```yaml
global_auth: false
consumers:
- name: consumer1
client_id: 12345678-xxxx-xxxx-xxxx-xxxxxxxxxxxx
client_secret: abcdefgh-xxxx-xxxx-xxxx-xxxxxxxxxxxx
- name: consumer2
client_id: 87654321-xxxx-xxxx-xxxx-xxxxxxxxxxxx
client_secret: hgfedcba-xxxx-xxxx-xxxx-xxxxxxxxxxxx
```
For the routes `route-a` and `route-b`, do the following plugin configuration:
```yaml
allow:
- consumer1
```
For the domains `*.example.com` and `test.com`, do the following plugin configuration:
```yaml
allow:
- consumer2
```
In this example, route names `route-a` and `route-b` refer to the route names filled in when creating the gateway route. When these two routes are matched, it will allow access for the caller with `name` as `consumer1`, and other callers will not be allowed to access.
In this example, the domains `*.example.com` and `test.com` are used to match request domains. When a matching domain is found, it will allow access for the caller with `name` as `consumer2`, while other callers will not be allowed to access.
### Enable at Gateway Instance Level
The following configuration will enable OAuth2 authentication at the gateway instance level, requiring all requests to be authenticated before access.
```yaml
global_auth: true
consumers:
- name: consumer1
client_id: 12345678-xxxx-xxxx-xxxx-xxxxxxxxxxxx
client_secret: abcdefgh-xxxx-xxxx-xxxx-xxxxxxxxxxxx
- name: consumer2
client_id: 87654321-xxxx-xxxx-xxxx-xxxxxxxxxxxx
client_secret: hgfedcba-xxxx-xxxx-xxxx-xxxxxxxxxxxx
```
# Request Example
## Using Client Credential Authorization Mode
### Get AccessToken
```bash
# Get via GET method (recommended)
curl 'http://test.com/oauth2/token?grant_type=client_credentials&client_id=12345678-xxxx-xxxx-xxxx-xxxxxxxxxxxx&client_secret=abcdefgh-xxxx-xxxx-xxxx-xxxxxxxxxxxx'
# Get via POST method (requires matching a route with a real target service first, or the gateway will not read the request Body)
curl 'http://test.com/oauth2/token' -H 'content-type: application/x-www-form-urlencoded' -d 'grant_type=client_credentials&client_id=12345678-xxxx-xxxx-xxxx-xxxxxxxxxxxx&client_secret=abcdefgh-xxxx-xxxx-xxxx-xxxxxxxxxxxx'
# Simply get the access_token field from the response:
{
"token_type": "bearer",
"access_token": "eyJhbGciOiJIUzI1NiIsInR5cCI6ImFwcGxpY2F0aW9uXC9hdCtqd3QifQ.eyJhdWQiOiJkZWZhdWx0IiwiY2xpZW50X2lkIjoiMTIzNDU2NzgteHh4eC14eHh4LXh4eHgteHh4eHh4eHh4eHh4IiwiZXhwIjoxNjg3OTUxNDYzLCJpYXQiOjE2ODc5NDQyNjMsImlzcyI6IkhpZ3Jlc3MtR2F0ZXdheSIsImp0aSI6IjEwOTU5ZDFiLThkNjEtNGRlYy1iZWE3LTk0ODEwMzc1YjYzYyIsInN1YiI6ImNvbnN1bWVyMSJ9.NkT_rG3DcV9543vBQgneVqoGfIhVeOuUBwLJJ4Wycb0",
"expires_in": 7200
}
```
### AccessToken Request
```bash
curl 'http://test.com' -H 'Authorization: Bearer eyJhbGciOiJIUzI1NiIsInR5cCI6ImFwcGxpY2F0aW9uXC9hdCtqd3QifQ.eyJhdWQiOiJkZWZhdWx0IiwiY2xpZW50X2lkIjoiMTIzNDU2NzgteHh4eC14eHh4LXh4eHgteHh4eHh4eHh4eHh4IiwiZXhwIjoxNjg3OTUxNDYzLCJpYXQiOjE2ODc5NDQyNjMsImlzcyI6IkhpZ3Jlc3MtR2F0ZXdheSIsImp0aSI6IjEwOTU5ZDFiLThkNjEtNGRlYy1iZWE3LTk0ODEwMzc1YjYzYyIsInN1YiI6ImNvbnN1bWVyMSJ9.NkT_rG3DcV9543vBQgneVqoGfIhVeOuUBwLJJ4Wycb0'
```
# Common Error Code Description
| HTTP Status Code | Error Message | Explanation |
| ---------------- | ----------------------| --------------------------------------------------------------------------- |
| 401 | Invalid Jwt token | JWT not provided in request header, or JWT format is incorrect, or expired, etc. |
| 403 | Access Denied | No permission to access the current route. |

49
plugins/wasm-cpp/extensions/request_block/README.md
View File

@@ -1,11 +1,18 @@
<p>
<a href="README_EN.md"> English </a> | 中文
</p>
---
title: 请求屏蔽
keywords: [higress,request block]
description: 请求屏蔽插件配置参考
---
# 功能说明
## 功能说明
`request-block`插件实现了基于 URL、请求头等特征屏蔽 HTTP 请求，可以用于防护部分站点资源不对外部暴露
# 配置字段
## 运行属性
插件执行阶段：`鉴权阶段`
插件执行优先级：`320`
## 配置字段
| 名称 | 数据类型 | 填写要求 | 默认值 | 描述 |
| -------- | -------- | -------- | -------- | -------- |
@@ -16,9 +23,9 @@
| blocked_message | string | 选填 | - | 配置请求被屏蔽时返回的 HTTP 应答 Body |
| case_sensitive | bool | 选填 | true | 配置匹配时是否区分大小写，默认区分 |
# 配置示例
## 配置示例
## 屏蔽请求 url 路径
### 屏蔽请求 url 路径
```yaml
block_urls:
- swagger.html
@@ -33,7 +40,7 @@ curl http://example.com?foo=Bar
curl http://exmaple.com/Swagger.html
```
## 屏蔽请求 header
### 屏蔽请求 header
```yaml
block_headers:
- example-key
@@ -47,7 +54,7 @@ curl http://example.com -H 'example-key: 123'
curl http://exmaple.com -H 'my-header: example-value'
```
## 屏蔽请求 body
### 屏蔽请求 body
```yaml
block_bodies:
- "hello world"
@@ -61,30 +68,8 @@ curl http://example.com -d 'Hello World'
curl http://exmaple.com -d 'hello world'
```
## 对特定路由或域名开启
```yaml
# 使用 _rules_ 字段进行细粒度规则配置
_rules_:
# 规则一：按路由名称匹配生效
- _match_route_:
- route-a
- route-b
block_bodies:
- "hello world"
# 规则二：按域名匹配生效
- _match_domain_:
- "*.example.com"
- test.com
block_urls:
- "swagger.html"
block_bodies:
- "hello world"
```
此例 `_match_route_` 中指定的 `route-a` 和 `route-b` 即在创建网关路由时填写的路由名称，当匹配到这两个路由时，将使用此段配置；
此例 `_match_domain_` 中指定的 `*.example.com` 和 `test.com` 用于匹配请求的域名，当发现域名匹配时，将使用此段配置；
配置的匹配生效顺序，将按照 `_rules_` 下规则的排列顺序，匹配第一个规则后生效对应配置，后续规则将被忽略。
# 请求 Body 大小限制
## 请求 Body 大小限制
当配置了 `block_bodies` 时，仅支持小于 32 MB 的请求 Body 进行匹配。若请求 Body 大于此限制，并且不存在匹配到的 `block_urls` 和 `block_headers` 项时，不会对该请求执行屏蔽操作
当配置了 `block_bodies` 时，若请求 Body 超过全局配置 DownstreamConnectionBufferLimits，将返回 `413 Payload Too Large`

80
plugins/wasm-cpp/extensions/request_block/README_EN.md
View File

@@ -1,24 +1,28 @@
<p>
English | <a href="README.md">中文</a>
</p>
---
title: Request Blocking
keywords: [higress,request block]
description: Request blocking plugin configuration reference
---
## Function Description
The `request-block` plugin implements HTTP request blocking based on features such as URL, request headers, etc. It can be used to protect certain site resources from being exposed to the outside.
# Description
`request-block` plugin implements a request blocking function based on request characteristics such as URL and request header. It can be used to protect internal resources from unauthorized access.
## Running Attributes
Plugin Execution Stage: `Authentication Stage`
# Configuration Fields
Plugin Execution Priority: `320`
| Name | Type | Requirement | Default Value | Description |
| -------- | -------- | -------- | -------- | -------- |
| block_urls | array of string | Optional. Choose one from following: `block_urls`, `block_headers`, `block_bodies` | - | HTTP URLs to be blocked. |
| block_headers | array of string | Optional. Choose one from following: `block_urls`, `block_headers`, `block_bodies` | - | HTTP request headers to be blocked. |
| block_bodies | array of string | Optional. Choose one from following: `block_urls` ,`block_headers`, `block_bodies` | - | HTTP request bodies to be blocked. |
| blocked_code | number | Optional | 403 | HTTP response status code to be sent when corresponding request is blocked. |
| blocked_message | string | Optional | - | HTTP response body to be sent when corresponding request is blocked. |
| case_sensitive | bool | Optional | true | Whether to use case-senstive comparison when matching. Enabled by default. |
## Configuration Fields
| Name | Data Type | Fill Requirement | Default Value | Description |
|--------------------|--------------------|---------------------------------------------------------|---------------|------------------------------------------------------------|
| block_urls | array of string | Optional, at least one of `block_urls`, `block_headers`, `block_bodies` must be filled | - | Configure strings for matching URLs that need to be blocked |
| block_headers | array of string | Optional, at least one of `block_urls`, `block_headers`, `block_bodies` must be filled | - | Configure strings for matching request headers that need to be blocked |
| block_bodies | array of string | Optional, at least one of `block_urls`, `block_headers`, `block_bodies` must be filled | - | Configure strings for matching request bodies that need to be blocked |
| blocked_code | number | Optional | 403 | Configure the HTTP status code returned when a request is blocked |
| blocked_message | string | Optional | - | Configure the HTTP response body returned when a request is blocked |
| case_sensitive | bool | Optional | true | Configure whether matching is case-sensitive, default is case-sensitive |
# Configuration Samples
## Block Specific Request URLs
## Configuration Example
### Blocking Request URL Paths
```yaml
block_urls:
- swagger.html
@@ -26,65 +30,39 @@ block_urls:
case_sensitive: false
```
According to the configuration above, following requests will be blocked:
Based on this configuration, the following requests will be denied access:
```bash
curl http://example.com?foo=Bar
curl http://exmaple.com/Swagger.html
```
## Block Specific Request Headers
### Blocking Request Headers
```yaml
block_headers:
- example-key
- example-value
```
According to the configuration above, following requests will be blocked:
Based on this configuration, the following requests will be denied access:
```bash
curl http://example.com -H 'example-key: 123'
curl http://exmaple.com -H 'my-header: example-value'
```
## Block Specific Request Bodies
### Blocking Request Bodies
```yaml
block_bodies:
- "hello world"
case_sensitive: false
```
According to the configuration above, following requests will be blocked:
Based on this configuration, the following requests will be denied access:
```bash
curl http://example.com -d 'Hello World'
curl http://exmaple.com -d 'hello world'
```
## Only Enable for Specific Routes or Domains
```yaml
# Use _rules_ field for fine-grained rule configurations
_rules_:
# Rule 1: Match by route name
- _match_route_:
- route-a
- route-b
block_bodies:
- "hello world"
# Rule 2: Match by domain
- _match_domain_:
- "*.example.com"
- test.com
block_urls:
- "swagger.html"
block_bodies:
- "hello world"
```
In the rule sample of `_match_route_`, `route-a` and `route-b` are the route names provided when creating a new gateway route. When the current route names matches the configuration, the rule following shall be applied.
In the rule sample of `_match_domain_`, `*.example.com` and `test.com` are the domain names used for request matching. When the current domain name matches the configuration, the rule following shall be applied.
All rules shall be checked following the order of items in the `_rules_` field, The first matched rule will be applied. All remained will be ignored.
## Request Body Size Limit
When `block_bodies` is configured, only request bodies smaller than 32 MB are supported for matching. If the request body exceeds this limit and there are no matching `block_urls` or `block_headers`, the blocking operation will not be executed for that request.
# Maximum Request Body Size Limitation
When `block_bodies` is configured, body matching shall only be performed when its size is smaller than 32MB. If not, and no `block_urls` or `block_headers` configuration is matched, the request won't be blocked.
When `block_bodies` is configured, if the size of request body exceeds the global configuration of DownstreamConnectionBufferLimits, a ``413 Payload Too Large`` response will be returned.
When `block_bodies` is configured and the request body exceeds the global configuration DownstreamConnectionBufferLimits, it will return `413 Payload Too Large`.

9
plugins/wasm-cpp/extensions/sni_misdirect/README.md
View File

@@ -1,11 +1,14 @@
# 功能说明
## 功能说明
**此插件已经废弃，从1.4版本开始，Higress 从根本上解决了这一问题，不再需要此插件**
`http2-misdirect`插件用于解决网关开启 HTTP2 时，因为浏览器复用连接导致访问出现 404 等问题。
# 插件原理
## 插件原理
HTTP2 协议允许两个不同域名的请求，在域名解析到相同 IP，并且使用了相同证书的情况下，复用同一条连接。这在一些特殊场景会导致复用连接的请求发送给了错误的 Virtual Host 进行处理，从而导致出现 404 等问题。
本插件基于`HTTP/2 RFC 7540`的`9.1.1`和`9.1.2`章节描述，在发现请求 SNI 与当前 Virtual Host 不匹配时，发送 HTTP 421 状态码，强制浏览器新建连接，并根据当前请求域名生成匹配的 SNI，从而让网关能正确处理路由。
# 浏览器兼容性
## 浏览器兼容性
`Safari` 浏览器 `15.1` 版本以下不支持 HTTP 421 状态码，若有此类客户端访问场景，建议对相应域名关闭 HTTP2 的 ALPN

6
plugins/wasm-go/extensions/ai-agent/README.md
View File

@@ -9,6 +9,10 @@ description: AI Agent插件配置参考
agent流程图如下：
![ai-agent](https://github.com/user-attachments/assets/b0761a0c-1afa-496c-a98e-bb9f38b340f8)
## 运行属性
插件执行阶段：`默认阶段`
插件执行优先级：`20`
## 配置字段
@@ -347,4 +351,4 @@ curl 'http://<这里换成网关公网IP>/api/openai/v1/chat/completions' \
```json
{"id":"65dcf12c-61ff-9e68-bffa-44fc9e6070d5","choices":[{"index":0,"message":{"role":"assistant","content":" “九头蛇万岁!”的德语翻译为“Hoch lebe Hydra!”。"},"finish_reason":"stop"}],"created":1724043865,"model":"qwen-max-0403","object":"chat.completion","usage":{"prompt_tokens":908,"completion_tokens":52,"total_tokens":960}}
```
```

328
plugins/wasm-go/extensions/ai-agent/README_EN.md Normal file
View File

@@ -0,0 +1,328 @@
---
title: AI Agent
keywords: [ AI Gateway, AI Agent ]
description: AI Agent plugin configuration reference
---
## Functional Description
A customizable API AI Agent that supports configuring HTTP method types as GET and POST APIs. Currently, it only supports non-streaming mode.
The agent flow chart is as follows:
![ai-agent](https://github.com/user-attachments/assets/b0761a0c-1afa-496c-a98e-bb9f38b340f8)
## Runtime Properties
Plugin execution phase: `Default Phase`
Plugin execution priority: `20`
## Configuration Fields
### Basic Configuration
| Name | Data Type | Requirement | Default Value | Description |
|------------------|-----------|-------------|---------------|----------------------------------|
| `llm` | object | Required | - | Configuration information for AI service provider |
| `apis` | object | Required | - | Configuration information for external API service provider |
| `promptTemplate` | object | Optional | - | Configuration information for Agent ReAct template |
The configuration fields for `llm` are as follows:
| Name | Data Type | Requirement | Default Value | Description |
|--------------------|-----------|-------------|---------------|-------------------------------------|
| `apiKey` | string | Required | - | Token for authentication when accessing large model services. |
| `serviceName` | string | Required | - | Name of the large model service |
| `servicePort` | int | Required | - | Port of the large model service |
| `domain` | string | Required | - | Domain for accessing the large model service |
| `path` | string | Required | - | Path for accessing the large model service |
| `model` | string | Required | - | Model name for accessing the large model service |
| `maxIterations` | int | Required | 15 | Maximum steps before ending the execution loop |
| `maxExecutionTime` | int | Required | 50000 | Timeout for each request to the large model, in milliseconds |
| `maxTokens` | int | Required | 1000 | Token limit for each request to the large model |
The configuration fields for `apis` are as follows:
| Name | Data Type | Requirement | Default Value | Description |
|-----------------|-----------|-------------|---------------|-------------------------------------|
| `apiProvider` | object | Required | - | Information about the external API service |
| `api` | string | Required | - | OpenAPI documentation of the tool |
The configuration fields for `apiProvider` are as follows:
| Name | Data Type | Requirement | Default Value | Description |
|-----------------|-----------|-------------|---------------|--------------------------------------------------|
| `apiKey` | object | Optional | - | Token for authentication when accessing external API services. |
| `serviceName` | string | Required | - | Name of the external API service |
| `servicePort` | int | Required | - | Port of the external API service |
| `domain` | string | Required | - | Domain for accessing the external API |
The configuration fields for `apiKey` are as follows:
| Name | Data Type | Requirement | Default Value | Description |
|-------------------|-----------|-------------|---------------|-------------------------------------------------------------------------------------|
| `in` | string | Optional | header | Whether the authentication token for accessing the external API service is in the header or in the query; default is header. |
| `name` | string | Optional | - | The name of the token for authentication when accessing the external API service. |
| `value` | string | Optional | - | The value of the token for authentication when accessing the external API service. |
The configuration fields for `promptTemplate` are as follows:
| Name | Data Type | Requirement | Default Value | Description |
|-----------------|-----------|-------------|---------------|----------------------------------------------------|
| `language` | string | Optional | EN | Language type of the Agent ReAct template, including CH and EN. |
| `chTemplate` | object | Optional | - | Agent ReAct Chinese template |
| `enTemplate` | object | Optional | - | Agent ReAct English template |
The configuration fields for `chTemplate` and `enTemplate` are as follows:
| Name | Data Type | Requirement | Default Value | Description |
|-----------------|-----------|-------------|---------------|---------------------------------------------------|
| `question` | string | Optional | - | The question part of the Agent ReAct template |
| `thought1` | string | Optional | - | The thought1 part of the Agent ReAct template |
| `actionInput` | string | Optional | - | The actionInput part of the Agent ReAct template |
| `observation` | string | Optional | - | The observation part of the Agent ReAct template |
| `thought2` | string | Optional | - | The thought2 part of the Agent ReAct template |
| `finalAnswer` | string | Optional | - | The finalAnswer part of the Agent ReAct template |
| `begin` | string | Optional | - | The begin part of the Agent ReAct template |
## Usage Example
**Configuration Information**
```yaml
llm:
apiKey: xxxxxxxxxxxxxxxxxx
domain: dashscope.aliyuncs.com
serviceName: dashscope.dns
servicePort: 443
path: /compatible-mode/v1/chat/completions
model: qwen-max-0403
maxIterations: 2
promptTemplate:
language: CH
apis:
- apiProvider:
domain: restapi.amap.com
serviceName: geo.dns
servicePort: 80
apiKey:
in: query
name: key
value: xxxxxxxxxxxxxxx
api: |
openapi: 3.1.0
info:
title: Amap
description: Get related information of POI
version: v1.0.0
servers:
- url: https://restapi.amap.com
paths:
/v5/place/text:
get:
description: Get latitude and longitude coordinates based on POI name
operationId: get_location_coordinate
parameters:
- name: keywords
in: query
description: POI name, must be in Chinese
required: true
schema:
type: string
- name: region
in: query
description: The name of the region where the POI is located, must be in Chinese
required: true
schema:
type: string
deprecated: false
/v5/place/around:
get:
description: Search for POI near the given coordinates
operationId: search_nearby_pois
parameters:
- name: keywords
in: query
description: Keywords for the target POI
required: true
schema:
type: string
- name: location
in: query
description: Latitude and longitude of the center point, separated by a comma
required: true
schema:
type: string
deprecated: false
components:
schemas: {}
- apiProvider:
domain: api.seniverse.com
serviceName: seniverse.dns
servicePort: 80
apiKey:
in: query
name: key
value: xxxxxxxxxxxxxxx
api: |
openapi: 3.1.0
info:
title: XZWeather
description: Get weather related information
version: v1.0.0
servers:
- url: https://api.seniverse.com
paths:
/v3/weather/now.json:
get:
description: Get weather conditions for a specified city
operationId: get_weather_now
parameters:
- name: location
in: query
description: The city to query
required: true
schema:
type: string
- name: language
in: query
description: Language used for the weather query results
required: true
schema:
type: string
default: zh-Hans
enum:
- zh-Hans
- en
- ja
- name: unit
in: query
description: Units of temperature, available in Celsius and Fahrenheit
required: true
schema:
type: string
default: c
enum:
- c
- f
deprecated: false
components:
schemas: {}
- apiProvider:
apiKey:
in: "header"
name: "DeepL-Auth-Key"
value: "73xxxxxxxxxxxxxxx:fx"
domain: "api-free.deepl.com"
serviceName: "deepl.dns"
servicePort: 443
api: |
openapi: 3.1.0
info:
title: DeepL API Documentation
description: The DeepL API provides programmatic access to DeepL’s machine translation technology.
version: v1.0.0
servers:
- url: https://api-free.deepl.com/v2
paths:
/translate:
post:
summary: Request Translation
operationId: translateText
requestBody:
required: true
content:
application/json:
schema:
type: object
required:
- text
- target_lang
properties:
text:
description: |
Text to be translated. Only UTF-8-encoded plain text is supported.
The parameter may be specified up to 50 times in a single request.
Translations are returned in the same order as they are requested.
type: array
maxItems: 50
items:
type: string
example: Hello, World!
target_lang:
description: The language into which the text should be translated.
type: string
enum:
- BG
- CS
- DA
- DE
- EL
- EN-GB
- EN-US
- ES
- ET
- FI
- FR
- HU
- ID
- IT
- JA
- KO
- LT
- LV
- NB
- NL
- PL
- PT-BR
- PT-PT
- RO
- RU
- SK
- SL
- SV
- TR
- UK
- ZH
- ZH-HANS
example: DE
components:
schemas: {}
```
This example configures three services demonstrating both GET and POST types of tools. The GET type tools include Amap and XZWeather, while the POST type tool is the DeepL translation. All three services need to be properly configured in the Higress service with DNS domain names and should be healthy.
Amap provides two tools, one for obtaining the coordinates of a specified location and the other for searching for points of interest near the coordinates. Document: https://lbs.amap.com/api/webservice/guide/api-advanced/newpoisearch
XZWeather provides one tool to get real-time weather conditions for a specified city, supporting results in Chinese, English, and Japanese, as well as representations in Celsius and Fahrenheit. Document: https://seniverse.yuque.com/hyper_data/api_v3/nyiu3t
DeepL provides one tool for translating given sentences, supporting multiple languages. Document: https://developers.deepl.com/docs/v/zh/api-reference/translate?fallback=true
Below are test cases. For stability, it is recommended to maintain a stable version of the large model. The example used here is qwen-max-0403:
**Request Example**
```shell
curl 'http://<replace with gateway public IP>/api/openai/v1/chat/completions' \
-H 'Accept: application/json, text/event-stream' \
-H 'Content-Type: application/json' \
--data-raw '{"model":"qwen","frequency_penalty":0,"max_tokens":800,"stream":false,"messages":[{"role":"user","content":"I want to have coffee near the Xinshi Building in Jinan, please recommend a few."}],"presence_penalty":0,"temperature":0,"top_p":0}'
```
**Response Example**
```json
{"id":"139487e7-96a0-9b13-91b4-290fb79ac992","choices":[{"index":0,"message":{"role":"assistant","content":" Near the Xinshi Building in Jinan, you can choose from the following coffee shops:\n1. luckin coffee 瑞幸咖啡(鑫盛大厦店), located in the lobby of Xinshi Building, No. 1299 Xinluo Avenue;\n2. 三庆齐盛广场挪瓦咖啡(三庆·齐盛广场店), located 60 meters southwest of the intersection of Xinluo Avenue and Yingxiu Road;\n3. luckin coffee 瑞幸咖啡(三庆·齐盛广场店), located at No. 1267 Yingxiu Road;\n4. 库迪咖啡(齐鲁软件园店), located in the commercial space of Building 4, Sanqing Qisheng Plaza, Xinluo Avenue;\n5. 库迪咖啡(美莲广场店), located at L117, Meilian Plaza, No. 1166 Xinluo Avenue, High-tech Zone; and a few other options. I hope these suggestions help!"},"finish_reason":"stop"}],"created":1723172296,"model":"qwen-max-0403","object":"chat.completion","usage":{"prompt_tokens":886,"completion_tokens":50,"total_tokens":936}}
```
**Request Example**
```shell
curl 'http://<replace with gateway public IP>/api/openai/v1/chat/completions' \
-H 'Accept: application/json, text/event-stream' \
-H 'Content-Type: application/json' \
--data-raw '{"model":"qwen","frequency_penalty":0,"max_tokens":800,"stream":false,"messages":[{"role":"user","content":"What is the current weather in Jinan?"}],"presence_penalty":0,"temperature":0,"top_p":0}'
```
**Response Example**
```json
{"id":"ebd6ea91-8e38-9e14-9a5b-90178d2edea4","choices":[{"index":0,"message":{"role":"assistant","content":" The current weather condition in Jinan is overcast, with a temperature of 31°C. This information was last updated on August 9, 2024, at 15:12 (Beijing time)."},"finish_reason":"stop"}],"created":1723187991,"model":"qwen-max-0403","object":"chat.completion","usage":{"prompt_tokens":890,"completion_tokens":56,"total_tokens":946}}
```
**Request Example**
```shell
curl 'http://<replace with gateway public IP>/api/openai/v1/chat/completions' \
-H 'Accept: application/json, text/event-stream' \
-H 'Content-Type: application/json' \
--data-raw '{"model":"qwen","frequency_penalty":0,"max_tokens":800,"stream":false,"messages":[{"role":"user","content":"What is the current weather in Jinan? Please indicate in Fahrenheit and respond in Japanese."}],"presence_penalty":0,"temperature":0,"top_p":0}'
```
**Response Example**
```json
{"id":"ebd6ea91-8e38-9e14-9a5b-90178d2edea4","choices":[{"index":0,"message":{"role":"assistant","content":" 現在の济南の天気は曇りで、気温は88°Fです。この情報は2024年8月9日15時12分（東京時間）に更新されました。"},"finish_reason":"stop"}],"created":1723187991,"model":"qwen-max-0403","object":"chat.completion","usage":{"prompt_tokens":890,"completion_tokens":56,"total_tokens":946}}
```
**Request Example**
```shell
curl 'http://<replace with gateway public IP>/api/openai/v1/chat/completions' \
-H 'Accept: application/json, text/event-stream' \
-H 'Content-Type: application/json' \
--data-raw '{"model":"qwen","frequency_penalty":0,"max_tokens":800,"stream":false,"messages":[{"role":"user","content":"Help me translate the following sentence into German: \"Hail Hydra!\""}],"presence_penalty":0,"temperature":0,"top_p":0}'
```
**Response Example**
```json
{"id":"65dcf12c-61ff-9e68-bffa-44fc9e6070d5","choices":[{"index":0,"message":{"role":"assistant","content":" The German translation of \"Hail Hydra!\" is \"Hoch lebe Hydra!\"."},"finish_reason":"stop"}],"created":1724043865,"model":"qwen-max-0403","object":"chat.completion","usage":{"prompt_tokens":908,"completion_tokens":52,"total_tokens":960}}
```

16
plugins/wasm-go/extensions/ai-cache/README.md
View File

@@ -1,13 +1,19 @@
## 简介
---
title: AI 缓存
keywords: [higress,ai cache]
description: AI 缓存插件配置参考
---
**Note**
> 需要数据面的proxy wasm版本大于等于0.2.100
> 编译时，需要带上版本的tag，例如：`tinygo build -o main.wasm -scheduler=none -target=wasi -gc=custom -tags="custommalloc nottinygc_finalizer proxy_wasm_version_0_2_100" ./`
## 功能说明
LLM 结果缓存插件，默认配置方式可以直接用于 openai 协议的结果缓存，同时支持流式和非流式响应的缓存。
## 运行属性
插件执行阶段：`认证阶段`
插件执行优先级：`10`
## 配置说明
| Name | Type | Requirement | Default | Description |

41
plugins/wasm-go/extensions/ai-cache/README_EN.md Normal file
View File

@@ -0,0 +1,41 @@
---
title: AI Cache
keywords: [higress,ai cache]
description: AI Cache Plugin Configuration Reference
---
## Function Description
LLM result caching plugin, the default configuration can be directly used for result caching under the OpenAI protocol, and it supports caching of both streaming and non-streaming responses.
## Runtime Properties
Plugin Execution Phase: `Authentication Phase`
Plugin Execution Priority: `10`
## Configuration Description
| Name | Type | Requirement | Default | Description |
| -------- | -------- | -------- | -------- | -------- |
| cacheKeyFrom.requestBody | string | optional | "messages.@reverse.0.content" | Extracts a string from the request Body based on [GJSON PATH](https://github.com/tidwall/gjson/blob/master/SYNTAX.md) syntax |
| cacheValueFrom.responseBody | string | optional | "choices.0.message.content" | Extracts a string from the response Body based on [GJSON PATH](https://github.com/tidwall/gjson/blob/master/SYNTAX.md) syntax |
| cacheStreamValueFrom.responseBody | string | optional | "choices.0.delta.content" | Extracts a string from the streaming response Body based on [GJSON PATH](https://github.com/tidwall/gjson/blob/master/SYNTAX.md) syntax |
| cacheKeyPrefix | string | optional | "higress-ai-cache:" | Prefix for the Redis cache key |
| cacheTTL | integer | optional | 0 | Cache expiration time in seconds, default value is 0, which means never expire |
| redis.serviceName | string | required | - | The complete FQDN name of the Redis service, including the service type, e.g., my-redis.dns, redis.my-ns.svc.cluster.local |
| redis.servicePort | integer | optional | 6379 | Redis service port |
| redis.timeout | integer | optional | 1000 | Timeout for requests to Redis, in milliseconds |
| redis.username | string | optional | - | Username for logging into Redis |
| redis.password | string | optional | - | Password for logging into Redis |
| returnResponseTemplate | string | optional | `{"id":"from-cache","choices":[%s],"model":"gpt-4o","object":"chat.completion","usage":{"prompt_tokens":0,"completion_tokens":0,"total_tokens":0}}` | Template for returning HTTP response, with %s marking the part to be replaced by cache value |
| returnStreamResponseTemplate | string | optional | `data:{"id":"from-cache","choices":[{"index":0,"delta":{"role":"assistant","content":"%s"},"finish_reason":"stop"}],"model":"gpt-4o","object":"chat.completion","usage":{"prompt_tokens":0,"completion_tokens":0,"total_tokens":0}}\n\ndata:[DONE]\n\n` | Template for returning streaming HTTP response, with %s marking the part to be replaced by cache value |
## Configuration Example
```yaml
redis:
serviceName: my-redis.dns
timeout: 2000
```
## Advanced Usage
The current default cache key is based on the GJSON PATH expression: `messages.@reverse.0.content`, meaning to get the content of the first item after reversing the messages array;
GJSON PATH supports conditional syntax, for instance, if you want to take the content of the last role as user as the key, it can be written as: `messages.@reverse.#(role=="user").content`;
If you want to concatenate all the content with role as user into an array as the key, it can be written as: `messages.@reverse.#(role=="user")#.content`;
It also supports pipeline syntax, for example, if you want to take the second role as user as the key, it can be written as: `messages.@reverse.#(role=="user")#.content|1`.
For more usage, you can refer to the [official documentation](https://github.com/tidwall/gjson/blob/master/SYNTAX.md) and use the [GJSON Playground](https://gjson.dev/) for syntax testing.

14
plugins/wasm-go/extensions/ai-history/README.md
View File

@@ -7,15 +7,17 @@ description: AI 历史对话插件配置参考
## 功能说明
`AI 历史对话` 基于请求头实现用户身份识别，并自动缓存对应用户的历史对话,且在后续对话中自动填充到上下文。同时支持用户主动查询历史对话。
**Note**
> 需要数据面的proxy wasm版本大于等于0.2.100
> 编译时，需要带上版本的tag，例如：
`tinygo build -o main.wasm -scheduler=none -target=wasi -gc=custom -tags="custommalloc nottinygc_finalizer proxy_wasm_version_0_2_100" ./`
> 路径后缀匹配 `ai-history/query` 时，会返回历史对话
## 运行属性
插件执行阶段：`默认阶段`
插件执行优先级：`650`
## 配置字段
| 名称 | 数据类型 | 填写要求 | 默认值 | Description |
@@ -169,4 +171,4 @@ curl 'http://example.com/api/openai/v1/chat/completions/ai-history/query?cnt=3'
]
```
返回三个历史对话,如果未传入 cnt 默认返回所有缓存历史对话。
返回三个历史对话,如果未传入 cnt 默认返回所有缓存历史对话。

158
plugins/wasm-go/extensions/ai-history/README_EN.md Normal file
View File

@@ -0,0 +1,158 @@
---
title: AI History Dialogue
keywords: [ AI Gateway, AI History Dialogue ]
description: AI History Dialogue Plugin Configuration Reference
---
## Functional Description
`AI History Dialogue` implements user identity recognition based on request headers and automatically caches the historical dialogues of corresponding users, which are then automatically filled into the context in subsequent dialogues. It also supports users to actively query historical dialogues.
**Note**
> When the path suffix matches `ai-history/query`, it will return the historical dialogues.
## Runtime Properties
Plugin Execution Phase: `Default Phase`
Plugin Execution Priority: `650`
## Configuration Fields
| Name | Data Type | Required | Default Value | Description |
|-------------------|---------|----------|-----------------------|---------------------------------------------------------------------------|
| identityHeader | string | optional | "Authorization" | The request header for identity resolution, can be Authorization, X-Mse-Consumer, etc. |
| fillHistoryCnt | integer | optional | 3 | Default number of historical dialogues to be filled. |
| cacheKeyPrefix | string | optional | "higress-ai-history:" | Prefix for Redis cache key. |
| cacheTTL | integer | optional | 0 | Cache expiration time in seconds, default value is 0, meaning it never expires. |
| redis.serviceName | string | required | - | Redis service name, full FQDN name with service type, e.g., my-redis.dns, redis.my-ns.svc.cluster.local |
| redis.servicePort | integer | optional | 6379 | Redis service port. |
| redis.timeout | integer | optional | 1000 | Timeout for requests to Redis, in milliseconds. |
| redis.username | string | optional | - | Username for logging into Redis. |
| redis.password | string | optional | - | Password for logging into Redis. |
## Usage Example
### Configuration Information
```yaml
redis:
serviceName: my-redis.dns
timeout: 2000
```
### Request Example
**Auto-fill Request Example:**
First Round Request:
```
curl 'http://example.com/api/openai/v1/chat/completions?fill_history_cnt=3' \
-H 'Accept: application/json, text/event-stream' \
-H 'Content-Type: application/json' \
-H 'Authorization: Bearer sk-Nzf7RtkdS4s0zFyn5575124129254d9bAf9473A5D7D06dD3'
--data-raw '{"model":"qwen-long","frequency_penalty":0,"max_tokens":800,"stream":false,"messages":[
{
"role": "user",
"content": "Can Higress replace Nginx?"
}
],"presence_penalty":0,"temperature":0.7,"top_p":0.95}'
```
After Request Fill:
> First round request, no fill. Consistent with the original request.
First Round Response:
```json
{
"id": "02f4c621-820e-97d4-a905-1e3d0d8f59c6",
"choices": [
{
"index": 0,
"message": {
"role": "assistant",
"content": "While both Higress and Nginx have gateway functionalities, their design philosophies and application scenarios differ. Nginx is better known as a high-performance HTTP and reverse proxy server, while Higress is a cloud-native gateway that integrates many cloud-native features such as service mesh, observability, and security management in addition to basic routing capabilities.\n\nTherefore, if you want to deploy applications in a cloud-native environment and wish to obtain advanced features required for modern applications, such as service governance, gray release, circuit breaker and rate limiting, and security authentication, then Higress can be a good alternative to Nginx. However, if it's a relatively simple static website or only requires basic reverse proxy functionality, traditional Nginx configurations may be simpler and more direct."
},
"finish_reason": "stop"
}
],
"created": 1724077770,
"model": "qwen-long",
"object": "chat.completion",
"usage": {
"prompt_tokens": 7316,
"completion_tokens": 164,
"total_tokens": 7480
}
}
```
Second Round Request:
```
curl 'http://example.com/api/openai/v1/chat/completions?fill_history_cnt=3' \
-H 'Accept: application/json, text/event-stream' \
-H 'Content-Type: application/json' \
-H 'Authorization: Bearer sk-Nzf7RtkdS4s0zFyn5575124129254d9bAf9473A5D7D06dD3'
--data-raw '{"model":"qwen-long","frequency_penalty":0,"max_tokens":800,"stream":false,"messages":[
{
"role": "user",
"content": "What about Spring Cloud GateWay?"
}
],"presence_penalty":0,"temperature":0.7,"top_p":0.95}'
```
After Request Fill:
> Second round request, automatically filled with the historical dialogue from the previous round.
```
curl 'http://example.com/api/openai/v1/chat/completions?fill_history_cnt=3' \
-H 'Accept: application/json, text/event-stream' \
-H 'Content-Type: application/json' \
-H 'Authorization: Bearer sk-Nzf7RtkdS4s0zFyn5575124129254d9bAf9473A5D7D06dD3'
--data-raw '{"model":"qwen-long","frequency_penalty":0,"max_tokens":800,"stream":false,"messages":[
{
"role": "user",
"content": "Can Higress replace Nginx?"
},
{
"role": "assistant",
"content": "While both Higress and Nginx have gateway functionalities, their design philosophies and application scenarios differ. Nginx is better known as a high-performance HTTP and reverse proxy server, while Higress is a cloud-native gateway that integrates many cloud-native features such as service mesh, observability, and security management in addition to basic routing capabilities.\n\nTherefore, if you want to deploy applications in a cloud-native environment and wish to obtain advanced features required for modern applications, such as service governance, gray release, circuit breaker and rate limiting, and security authentication, then Higress can be a good alternative to Nginx. However, if it's a relatively simple static website or only requires basic reverse proxy functionality, traditional Nginx configurations may be simpler and more direct."
},
{
"role": "user",
"content": "What about Spring Cloud GateWay?"
}
],"presence_penalty":0,"temperature":0.7,"top_p":0.95}'
```
Each round of requests only needs to include the current question and the number of historical dialogues to fill in, enabling automatic filling of historical dialogues.
**Example of Retrieving Historical Data:**
```
curl 'http://example.com/api/openai/v1/chat/completions/ai-history/query?cnt=3' \
-H 'Accept: application/json, text/event-stream' \
-H 'Content-Type: application/json' \
-H 'Authorization: Bearer sk-Nzf7RtkdS4s0zFyn5575124129254d9bAf9473A5D7D06dD3'
```
Response Example:
```json
[
{
"role": "user",
"content": "Can Higress replace Nginx?"
},
{
"role": "assistant",
"content": "While both Higress and Nginx have gateway functionalities, their design philosophies and application scenarios differ. Nginx is better known as a high-performance HTTP and reverse proxy server, while Higress is a cloud-native gateway that integrates many cloud-native features such as service mesh, observability, and security management in addition to basic routing capabilities.\\n\\nTherefore, if you want to deploy applications in a cloud-native environment and wish to obtain advanced features required for modern applications, such as service governance, gray release, circuit breaker and rate limiting, and security authentication, then Higress can be a good alternative to Nginx. However, if it's a relatively simple static website or only requires basic reverse proxy functionality, traditional Nginx configurations may be simpler and more direct."
},
{
"role": "user",
"content": "What about Spring Cloud GateWay?"
},
{
"role": "assistant",
"content": "Compared to Spring Cloud Gateway, Higress is also an API gateway, but there are some key differences between them:\\n\\n- **Design Philosophy**: Spring Cloud Gateway mainly targets service-to-service communication and routing in microservice architectures. As part of the Spring Cloud ecosystem, it is more focused on scenarios for Java developers in microservices. In contrast, Higress, as a cloud-native gateway, not only focuses on service communication but also offers a range of cloud-native features such as service mesh, observability, security management, etc.\\n- **Deployment Method**: Spring Cloud Gateway typically runs as part of microservice applications on application servers, whereas Higress usually deploys as an independent microservice or containerized service in Kubernetes environments, suitable for modern cloud-native deployment models.\\n- **Scalability and Integration**: Higress provides wider integration and support, for example, deep integration with ecosystems like Istio, Kubernetes, etc., making it better suited for complex cloud-native environments.\\n\\nTherefore, if your application is built on Spring Cloud and you want a lightweight, easy-to-integrate service gateway, then Spring Cloud Gateway might be a suitable choice. However, if you are building or refactoring cloud-native applications and require more powerful routing rules, service governance, observability, etc., then Higress is likely a better choice."
},
{
"role": "user",
"content": "Can Higress replace Nginx?"
},
{
"role": "assistant",
"content": "While both Higress and Nginx have gateway functionalities, their design philosophies and application scenarios differ. Nginx is better known as a high-performance HTTP and reverse proxy server, while Higress is a cloud-native gateway that integrates many cloud-native features such as service mesh, observability, and security management in addition to basic routing capabilities.\\n\\nTherefore, if you want to deploy applications in a cloud-native environment and wish to obtain advanced features required for modern applications, such as service governance, gray release, circuit breaker and rate limiting, and security authentication, then Higress can be a good alternative to Nginx. However, if it's a relatively simple static website or only requires basic reverse proxy functionality, traditional Nginx configurations may be simpler and more direct."
}
]
```
Returns three historical dialogues. If the `cnt` parameter is not provided, it will default to returning all cached historical dialogues.

21
plugins/wasm-go/extensions/ai-intent/README.md
View File

@@ -1,15 +1,20 @@
## 简介
---
title: AI 意图识别
keywords: [ AI网关, AI意图识别 ]
description: AI 意图识别插件配置参考
---
**Note**
> 需要数据面的proxy wasm版本大于等于0.2.100
> 编译时，需要带上版本的tag，例如：`tinygo build -o main.wasm -scheduler=none -target=wasi -gc=custom -tags="custommalloc nottinygc_finalizer proxy_wasm_version_0_2_100" ./`
## 功能说明
LLM 意图识别插件，能够智能判断用户请求与某个领域或agent的功能契合度，从而提升不同模型的应用效果和用户体验
## 运行属性
插件执行阶段：`默认阶段`
插件执行优先级：`700`
## 配置说明
> 1.该插件的优先级要高于ai-cache、ai-proxy等后续使用意图的插件，后续插件可以通过proxywasm.GetProperty([]string{"intent_category"})方法获取到意图主题，按照意图主题去做不同缓存库或者大模型的选择
> 1.该插件的优先级高于ai-proxy等后续使用意图的插件，后续插件可以通过proxywasm.GetProperty([]string{"intent_category"})方法获取到意图主题，按照意图主题去做不同缓存库或者大模型的选择
> 2.需新建一条higress的大模型路由，供该插件访问大模型,如：路由以 /intent 作为前缀，服务选择大模型服务，为该路由开启ai-proxy插件
@@ -19,7 +24,7 @@ LLM 意图识别插件，能够智能判断用户请求与某个领域或agent
| 名称 | 数据类型 | 填写要求 | 默认值 | 描述 |
| -------------- | --------------- | -------- | ------ | ------------------------------------------------------------ |
| `scene.category` | string | 必填 | - | 预设场景类别，以"|"分割，如："金融|电商|法律|Higress"|
| `scene.category` | string | 必填 | - | 预设场景类别，以`|`分割，如：`金融|电商|法律|Higress`|
| `scene.prompt` | string | 非必填 | 你是一个智能类别识别助手，负责根据用户提出的问题和预设的类别，确定问题属于哪个预设的类别，并给出相应的类别。用户提出的问题为:%s,预设的类别为%s，直接返回一种具体类别，如果没有找到就返回'NotFound'。 | llm请求prompt模板 |
| `llm.proxyServiceName` | string | 必填 | - | 新建的higress服务，指向大模型 (取higress中的 FQDN 值)|
| `llm.proxyUrl` | string | 必填 | - | 大模型路由请求地址全路径，可以是网关自身的地址，也可以是其他大模型的地址（openai协议），例如：http://127.0.0.1:80/intent/compatible-mode/v1/chat/completions |

45
plugins/wasm-go/extensions/ai-intent/README_EN.md Normal file
View File

@@ -0,0 +1,45 @@
---
title: AI Intent Recognition
keywords: [ AI Gateway, AI Intent Recognition ]
description: AI Intent Recognition Plugin Configuration Reference
---
## Function Description
LLM Intent Recognition plugin can intelligently determine the alignment between user requests and the functionalities of a certain domain or agent, thereby enhancing the application effectiveness of different models and user experience.
## Execution Attributes
Plugin execution phase: `Default Phase`
Plugin execution priority: `700`
## Configuration Instructions
> 1. This plugin's priority is higher than that of plugins such as ai-proxy which follow up and use intent. Subsequent plugins can retrieve the intent category using the proxywasm.GetProperty([]string{"intent_category"}) method and make selections for different cache libraries or large models based on the intent category.
> 2. A new Higress large model route needs to be created to allow this plugin to access the large model. For example: the route should use `/intent` as a prefix, the service should select the large model service, and the ai-proxy plugin should be enabled for this route.
> 3. A fixed-address service needs to be created (for example, intent-service), which points to 127.0.0.1:80 (i.e., the gateway instance and port). The ai-intent plugin requires this service for calling to access the newly added route. The service name corresponds to llm.proxyServiceName (a DNS type service can also be created to allow the plugin to access other large models).
> 4. If using a fixed-address service to call the gateway itself, 127.0.0.1 must be added to the gateway's access whitelist.
| Name | Data Type | Requirement | Default Value | Description |
| -------------- | --------------- | ----------- | ------------- | --------------------------------------------------------------- |
| `scene.category` | string | Required | - | Preset scene categories, separated by "|", e.g.: "Finance|E-commerce|Law|Higress" |
| `scene.prompt` | string | Optional | You are a smart category recognition assistant responsible for determining which preset category a user’s question belongs to based on the question posed by the user and the preset categories, and returning the corresponding category. The user's question is: %s, the preset categories are %s, directly return a specific category; if not found, return 'NotFound'. | llm request prompt template |
| `llm.proxyServiceName` | string | Required | - | Newly created Higress service pointing to the large model (use the FQDN value from Higress) |
| `llm.proxyUrl` | string | Required | - | The full path to the large model route request address, which can be the gateway’s own address or the address of another large model (OpenAI protocol), for example: http://127.0.0.1:80/intent/compatible-mode/v1/chat/completions |
| `llm.proxyDomain` | string | Optional | Retrieved from proxyUrl | Domain of the large model service |
| `llm.proxyPort` | string | Optional | Retrieved from proxyUrl | Port number of the large model service |
| `llm.proxyApiKey` | string | Optional | - | API_KEY corresponding to the external large model service when using it |
| `llm.proxyModel` | string | Optional | qwen-long | Type of the large model |
| `llm.proxyTimeout` | number | Optional | 10000 | Timeout for calling the large model, unit ms, default: 10000ms |
## Configuration Example
```yaml
scene:
category: "Finance|E-commerce|Law|Higress"
prompt: "You are a smart category recognition assistant responsible for determining which preset category a user's question belongs to based on the question posed by the user and the preset categories, and returning the corresponding category. The user's question is: '%s', the preset categories are '%s', directly return a specific category; if not found, return 'NotFound'."
llm:
proxyServiceName: "intent-service.static"
proxyUrl: "http://127.0.0.1:80/intent/compatible-mode/v1/chat/completions"
proxyDomain: "127.0.0.1"
proxyPort: "80"
proxyModel: "qwen-long"
proxyApiKey: ""
proxyTimeout: "10000"
```

82
plugins/wasm-go/extensions/ai-json-resp/README.md
View File

@@ -1,15 +1,17 @@
## 简介
**Note**
> 需要数据面的proxy wasm版本大于等于0.2.100
>
> 编译时，需要带上版本的tag，例如：tinygo build -o main.wasm -scheduler=none -target=wasi -gc=custom -tags="custommalloc nottinygc_finalizer proxy_wasm_version_0_2_100" ./
---
title: AI JSON 格式化
keywords: [ AI网关, AI JSON 格式化 ]
description: AI JSON 格式化插件配置参考
---
## 功能说明
LLM响应结构化插件，用于根据默认或用户配置的Json Schema对AI的响应进行结构化，以便后续插件处理。注意目前只支持 `非流式响应`。
## 运行属性
插件执行阶段：`默认阶段`
插件执行优先级：`150`
### 配置说明
@@ -109,7 +111,6 @@ curl -X POST "http://localhost:8001/v1/chat/completions" \
### 支持openai接口的AI服务
以qwen为例，基本配置如下：
Yaml格式配置如下
```yaml
serviceName: qwen
serviceDomain: dashscope.aliyuncs.com
@@ -133,70 +134,15 @@ jsonSchema:
additionalProperties: false
```
JSON 格式配置
```json
{
"serviceName": "qwen",
"serviceUrl": "https://dashscope.aliyuncs.com/compatible-mode/v1/chat/completions",
"apiKey": "[Your API Key]",
"jsonSchema": {
"title": "ActionItemsSchema",
"type": "object",
"properties": {
"action_items": {
"type": "array",
"items": {
"type": "object",
"properties": {
"description": {
"type": "string",
"description": "Description of the action item."
},
"due_date": {
"type": ["string", "null"],
"description": "Due date for the action item, can be null if not specified."
},
"owner": {
"type": ["string", "null"],
"description": "Owner responsible for the action item, can be null if not specified."
}
},
"required": ["description", "due_date", "owner"],
"additionalProperties": false
},
"description": "List of action items from the meeting."
}
},
"required": ["action_items"],
"additionalProperties": false
}
}
```
### 本地网关服务
为了能复用已经配置好的服务，本插件也支持配置本地网关服务。例如，若网关已经配置好了[AI-proxy服务](../ai-proxy/README.md)，则可以直接配置如下：
1. 创建一个固定IP为127.0.0.1的服务，例如localservice.static
```yaml
- name: outbound|10000||localservice.static
connect_timeout: 30s
type: LOGICAL_DNS
dns_lookup_family: V4_ONLY
lb_policy: ROUND_ROBIN
load_assignment:
cluster_name: outbound|8001||localservice.static
endpoints:
- lb_endpoints:
- endpoint:
address:
socket_address:
address: 127.0.0.1
port_value: 10000
```
为了能复用已经配置好的服务，本插件也支持配置本地网关服务。例如，若网关已经配置好了AI-proxy服务，则可以直接配置如下：
1. 创建一个固定IP地址为127.0.0.1:80的服务，例如localservice.static
2. 配置文件中添加localservice.static的服务配置
```yaml
serviceName: localservice
serviceDomain: 127.0.0.1
servicePort: 10000
servicePort: 80
```
3. 自动提取请求的Path，Header等信息
插件会自动提取请求的Path，Header等信息，从而避免对AI服务的重复配置。

140
plugins/wasm-go/extensions/ai-json-resp/README_EN.md Normal file
View File

@@ -0,0 +1,140 @@
---
title: AI JSON Formatting
keywords: [ AI Gateway, AI JSON Formatting ]
description: AI JSON Formatting plugin configuration reference
---
## Function Description
LLM structured response plugin, used to structure AI responses according to the default or user-configured Json Schema for subsequent plugin processing. Note that only `non-streaming responses` are currently supported.
## Running Attributes
Plugin execution phase: `default phase`
Plugin execution priority: `150`
### Configuration Description
| Name | Type | Requirement | Default | **Description** |
| --- | --- | --- | --- | --- |
| serviceName | str | required | - | AI service or gateway service name that supports AI-Proxy |
| serviceDomain | str | optional | - | AI service or gateway service domain/IP address that supports AI-Proxy |
| servicePath | str | optional | '/v1/chat/completions' | AI service or gateway service base path that supports AI-Proxy |
| serviceUrl | str | optional | - | AI service or gateway service URL that supports AI-Proxy; the plugin will automatically extract domain and path to fill in the unconfigured serviceDomain or servicePath |
| servicePort | int | optional | 443 | Gateway service port |
| serviceTimeout | int | optional | 50000 | Default request timeout |
| maxRetry | int | optional | 3 | Number of retry attempts when the answer cannot be correctly extracted and formatted |
| contentPath | str | optional | "choices.0.message.content” | gpath path to extract the response result from the LLM answer |
| jsonSchema | str (json) | optional | - | The jsonSchema against which the request is validated; if empty, only valid Json format responses are returned |
| enableSwagger | bool | optional | false | Whether to enable the Swagger protocol for validation |
| enableOas3 | bool | optional | true | Whether to enable the Oas3 protocol for validation |
| enableContentDisposition | bool | optional | true | Whether to enable the Content-Disposition header; if enabled, the response header will include `Content-Disposition: attachment; filename="response.json"` |
> For performance reasons, the maximum supported Json Schema depth is 6 by default. Json Schemas exceeding this depth will not be used to validate responses; the plugin will only check if the returned response is a valid Json format.
### Request and Return Parameter Description
- **Request Parameters**: The request format for this plugin is the OpenAI request format, including the `model` and `messages` fields, where `model` is the AI model name and `messages` is a list of conversation messages, each containing `role` and `content` fields, with `role` being the message role and `content` being the message content.
```json
{
"model": "gpt-4",
"messages": [
{"role": "user", "content": "give me a api doc for add the variable x to x+5"}
]
}
```
Other request parameters should refer to the corresponding documentation of the configured AI service or gateway service.
- **Return Parameters**:
- Returns a `Json format response` that satisfies the constraints of the defined Json Schema.
- If no Json Schema is defined, returns a valid `Json format response`.
- If an internal error occurs, returns `{ "Code": 10XX, "Msg": "Error message" }`.
## Request Example
```bash
curl -X POST "http://localhost:8001/v1/chat/completions" \
-H "Content-Type: application/json" \
-d '{
"model": "gpt-4",
"messages": [
{"role": "user", "content": "give me a api doc for add the variable x to x+5"}
]
}'
```
## Return Example
### Normal Return
Under normal circumstances, the system should return JSON data validated by the JSON Schema. If no JSON Schema is configured, the system will return legally valid JSON data that complies with JSON standards.
```json
{
"apiVersion": "1.0",
"request": {
"endpoint": "/add_to_five",
"method": "POST",
"port": 8080,
"headers": {
"Content-Type": "application/json"
},
"body": {
"x": 7
}
}
}
```
### Exception Return
In case of an error, the return status code is `500`, and the return content is a JSON format error message. It contains two fields: error code `Code` and error message `Msg`.
```json
{
"Code": 1006,
"Msg": "retry count exceed max retry count"
}
```
### Error Code Description
| Error Code | Description |
| --- | --- |
| 1001 | The configured Json Schema is not in a valid Json format |
| 1002 | The configured Json Schema compilation failed; it is not a valid Json Schema format or depth exceeds jsonSchemaMaxDepth while rejectOnDepthExceeded is true |
| 1003 | Unable to extract valid Json from the response |
| 1004 | Response is an empty string |
| 1005 | Response does not conform to the Json Schema definition |
| 1006 | Retry count exceeds the maximum limit |
| 1007 | Unable to retrieve the response content; may be due to upstream service configuration errors or incorrect ContentPath path to get the content |
| 1008 | serviceDomain is empty; please note that either serviceDomain or serviceUrl cannot be empty at the same time |
## Service Configuration Description
This plugin requires configuration of upstream services to support automatic retry mechanisms in case of exceptions. Supported configurations mainly include `AI services supporting OpenAI interfaces` or `local gateway services`.
### AI Services Supporting OpenAI Interfaces
Taking Qwen as an example, the basic configuration is as follows:
```yaml
serviceName: qwen
serviceDomain: dashscope.aliyuncs.com
apiKey: [Your API Key]
servicePath: /compatible-mode/v1/chat/completions
jsonSchema:
title: ReasoningSchema
type: object
properties:
reasoning_steps:
type: array
items:
type: string
description: The reasoning steps leading to the final conclusion.
answer:
type: string
description: The final answer, taking into account the reasoning steps.
required:
- reasoning_steps
- answer
additionalProperties: false
```
### Local Gateway Services
To reuse already configured services, this plugin also supports configuring local gateway services. For example, if the gateway has already configured the AI-proxy service, it can be directly configured as follows:
1. Create a service with a fixed IP address of 127.0.0.1:80, for example, localservice.static.
2. Add the service configuration for localservice.static in the configuration file.
```yaml
serviceName: localservice
serviceDomain: 127.0.0.1
servicePort: 80
```
3. Automatically extract request Path, Header, and other information.
The plugin will automatically extract request Path, Header, and other information to avoid repetitive configuration for the AI service.

22
plugins/wasm-go/extensions/ai-prompt-decorator/README.md
View File

@@ -1,7 +1,19 @@
# 简介
AI提示词装饰器插件，支持在LLM的请求前后插入prompt。
---
title: AI 提示词
keywords: [ AI网关, AI提示词 ]
description: AI 提示词插件配置参考
---
# 配置说明
## 功能说明
AI提示词插件，支持在LLM的请求前后插入prompt。
## 运行属性
插件执行阶段：`默认阶段`
插件执行优先级：`450`
## 配置说明
| 名称 | 数据类型 | 填写要求 | 默认值 | 描述 |
|----------------|-----------------|------|-----|----------------------------------|
@@ -15,7 +27,7 @@ message object 配置说明：
| `role` | string | 必填 | - | 角色 |
| `content` | string | 必填 | - | 消息 |
# 示例
## 示例
配置示例如下：
@@ -69,7 +81,7 @@ curl http://localhost/test \
```
# 基于geo-ip插件的能力，扩展AI提示词装饰器插件携带用户地理位置信息
## 基于geo-ip插件的能力，扩展AI提示词装饰器插件携带用户地理位置信息
如果需要在LLM的请求前后加入用户地理位置信息，请确保同时开启geo-ip插件和AI提示词装饰器插件。并且在相同的请求处理阶段里，geo-ip插件的优先级必须高于AI提示词装饰器插件。首先geo-ip插件会根据用户ip计算出用户的地理位置信息，然后通过请求属性传递给后续插件。比如在默认阶段里，geo-ip插件的priority配置1000，ai-prompt-decorator插件的priority配置500。
geo-ip插件配置示例：

130
plugins/wasm-go/extensions/ai-prompt-decorator/README_EN.md Normal file
View File

@@ -0,0 +1,130 @@
---
title: AI Prompts
keywords: [ AI Gateway, AI Prompts ]
description: AI Prompts plugin configuration reference
---
## Function Description
The AI Prompts plugin allows inserting prompts before and after requests in LLM.
## Execution Properties
Plugin execution phase: `Default Phase`
Plugin execution priority: `450`
## Configuration Description
| Name | Data Type | Requirement | Default Value | Description |
|---------------|----------------------|-------------|---------------|--------------------------------------|
| `prepend` | array of message object | optional | - | Statements inserted before the initial input |
| `append` | array of message object | optional | - | Statements inserted after the initial input |
Message object configuration description:
| Name | Data Type | Requirement | Default Value | Description |
|-----------|-------------|-------------|---------------|-------------|
| `role` | string | required | - | Role |
| `content` | string | required | - | Message |
## Example
An example configuration is as follows:
```yaml
prepend:
- role: system
content: "Please answer the questions in English."
append:
- role: user
content: "After answering each question, try to ask a follow-up question."
```
Using the above configuration to initiate a request:
```bash
curl http://localhost/test \
-H "content-type: application/json" \
-d '{
"model": "gpt-3.5-turbo",
"messages": [
{
"role": "user",
"content": "Who are you?"
}
]
}
```
After processing through the plugin, the actual request will be:
```bash
curl http://localhost/test \
-H "content-type: application/json" \
-d '{
"model": "gpt-3.5-turbo",
"messages": [
{
"role": "system",
"content": "Please answer the questions in English."
},
{
"role": "user",
"content": "Who are you?"
},
{
"role": "user",
"content": "After answering each question, try to ask a follow-up question."
}
]
}
```
## Based on the geo-ip plugin's capabilities, extend AI Prompt Decorator plugin to carry user geographic location information.
If you need to include user geographic location information before and after the LLM's requests, please ensure both the geo-ip plugin and the AI Prompt Decorator plugin are enabled. Moreover, in the same request processing phase, the geo-ip plugin's priority must be higher than that of the AI Prompt Decorator plugin. First, the geo-ip plugin will calculate the user's geographic location information based on the user's IP, and then pass it to subsequent plugins via request attributes. For instance, in the default phase, the geo-ip plugin's priority configuration is 1000, while the ai-prompt-decorator plugin's priority configuration is 500.
Example configuration for the geo-ip plugin:
```yaml
ipProtocal: "ipv4"
```
An example configuration for the AI Prompt Decorator plugin is as follows:
```yaml
prepend:
- role: system
content: "The user's current geographic location is, country: ${geo-country}, province: ${geo-province}, city: ${geo-city}."
append:
- role: user
content: "After answering each question, try to ask a follow-up question."
```
Using the above configuration to initiate a request:
```bash
curl http://localhost/test \
-H "content-type: application/json" \
-H "x-forwarded-for: 87.254.207.100,4.5.6.7" \
-d '{
"model": "gpt-3.5-turbo",
"messages": [
{
"role": "user",
"content": "How is the weather today?"
}
]
}'
```
After processing through the plugin, the actual request will be:
```bash
curl http://localhost/test \
-H "content-type: application/json" \
-H "x-forwarded-for: 87.254.207.100,4.5.6.7" \
-d '{
"model": "gpt-3.5-turbo",
"messages": [
{
"role": "system",
"content": "The user's current geographic location is, country: China, province: Beijing, city: Beijing."
},
{
"role": "user",
"content": "How is the weather today?"
},
{
"role": "user",
"content": "After answering each question, try to ask a follow-up question."
}
]
}'
```

18
plugins/wasm-go/extensions/ai-prompt-template/README.md
View File

@@ -1,7 +1,19 @@
# 简介
---
title: AI 提示词模版
keywords: [ AI网关, AI提示词模版 ]
description: AI 提示词模版配置参考
---
## 功能说明
AI提示词模板，用于快速构建同类型的AI请求。
# 配置说明
## 运行属性
插件执行阶段：`默认阶段`
插件执行优先级：`500`
## 配置说明
| 名称 | 数据类型 | 填写要求 | 默认值 | 描述 |
|----------------|-----------------|------|-----|----------------------------------|
| `templates` | array of object | 必填 | - | 模板设置 |
@@ -45,4 +57,4 @@ templates:
"language": "python"
}
}
```
```

53
plugins/wasm-go/extensions/ai-prompt-template/README_EN.md Normal file
View File

@@ -0,0 +1,53 @@
---
title: AI Prompt Template
keywords: [ AI Gateway, AI Prompt Template ]
description: AI Prompt Template Configuration Reference
---
## Function Description
AI prompt templates are used to quickly build similar types of AI requests.
## Execution Properties
Plugin Execution Phase: `Default Phase`
Plugin Execution Priority: `500`
## Configuration Description
| Name | Data Type | Required | Default Value | Description |
|-----------------|-------------------|----------|---------------|-----------------------------------|
| `templates` | array of object | Required | - | Template settings |
Template object configuration description:
| Name | Data Type | Required | Default Value | Description |
|-----------------------|-------------------|----------|---------------|-----------------------------------|
| `name` | string | Required | - | Template name |
| `template.model` | string | Required | - | Model name |
| `template.messages` | array of object | Required | - | Input for large model |
Message object configuration description:
| Name | Data Type | Required | Default Value | Description |
|----------------|-------------------|----------|---------------|-----------------------------------|
| `role` | string | Required | - | Role |
| `content` | string | Required | - | Message |
Configuration example:
```yaml
templates:
- name: "developer-chat"
template:
model: gpt-3.5-turbo
messages:
- role: system
content: "You are a {{program}} expert, in {{language}} programming language."
- role: user
content: "Write me a {{program}} program."
```
Example request body using the above configuration:
```json
{
"template": "developer-chat",
"properties": {
"program": "quick sort",
"language": "python"
}
}
```

5
plugins/wasm-go/extensions/ai-proxy/README.md
View File

@@ -15,6 +15,11 @@ description: AI 代理插件配置参考
> 请求路径后缀匹配 `/v1/embeddings` 时，对应文本向量场景，会用 OpenAI 的文本向量协议解析请求 Body，再转换为对应 LLM 厂商的文本向量协议
## 运行属性
插件执行阶段：`默认阶段`
插件执行优先级：`100`
## 配置字段

1487
plugins/wasm-go/extensions/ai-proxy/README_EN.md Normal file
View File

File diff suppressed because it is too large Load Diff

25
plugins/wasm-go/extensions/ai-quota/README.md
View File

@@ -1,10 +1,21 @@
# 功能说明
---
title: AI 配额管理
keywords: [ AI网关, AI配额 ]
description: AI 配额管理插件配置参考
---
## 功能说明
`ai-qutoa` 插件实现给特定 consumer 根据分配固定的 quota 进行 quota 策略限流，同时支持 quota 管理能力，包括查询 quota 、刷新 quota、增减 quota。
`ai-quota` 插件需要配合 认证插件比如 `key-auth`、`jwt-auth` 等插件获取认证身份的 consumer 名称，同时需要配合 `ai-statatistics` 插件获取 AI Token 统计信息。
# 配置说明
## 运行属性
插件执行阶段：`默认阶段`
插件执行优先级：`750`
## 配置说明
| 名称 | 数据类型 | 填写要求 | 默认值 | 描述 |
|--------------------|-----------------|--------------------------------------| ---- |--------------------------------------------|
@@ -25,9 +36,9 @@
# 配置示例
## 配置示例
## 识别请求参数 apikey，进行区别限流
### 识别请求参数 apikey，进行区别限流
```yaml
redis_key_prefix: "chat_quota:"
admin_consumer: consumer3
@@ -39,19 +50,19 @@ redis:
```
## 刷新 quota
### 刷新 quota
如果当前请求 url 的后缀符合 admin_path，例如插件在 example.com/v1/chat/completions 这个路由上生效，那么更新 quota 可以通过
curl https://example.com/v1/chat/completions/quota/refresh -H "Authorization: Bearer credential3" -d "consumer=consumer1&quota=10000"
Redis 中 key 为 chat_quota:consumer1 的值就会被刷新为 10000
## 查询 quota
### 查询 quota
查询特定用户的 quota 可以通过 curl https://example.com/v1/chat/completions/quota?consumer=consumer1 -H "Authorization: Bearer credential3"
将返回： {"quota": 10000, "consumer": "consumer1"}
## 增减 quota
### 增减 quota
增减特定用户的 quota 可以通过 curl https://example.com/v1/chat/completions/quota/delta -d "consumer=consumer1&value=100" -H "Authorization: Bearer credential3"
这样 Redis 中 Key 为 chat_quota:consumer1 的值就会增加100，可以支持负数，则减去对应值。

54
plugins/wasm-go/extensions/ai-quota/README_EN.md Normal file
View File

@@ -0,0 +1,54 @@
---
title: AI Quota Management
keywords: [ AI Gateway, AI Quota ]
description: AI quota management plugin configuration reference
---
## Function Description
The `ai-quota` plugin implements quota rate limiting based on fixed quotas allocated to specific consumers. It also supports quota management capabilities, including querying quotas, refreshing quotas, and increasing or decreasing quotas. The `ai-quota` plugin needs to work with authentication plugins such as `key-auth`, `jwt-auth`, etc., to obtain the consumer name associated with the authenticated identity, and it needs to work with the `ai-statistics` plugin to obtain AI Token statistical information.
## Runtime Properties
Plugin execution phase: `default phase`
Plugin execution priority: `750`
## Configuration Description
| Name | Data Type | Required Conditions | Default Value | Description |
|---------------------|------------------|--------------------------------------------|---------------|---------------------------------------------------|
| `redis_key_prefix` | string | Optional | chat_quota: | Quota redis key prefix |
| `admin_consumer` | string | Required | | Consumer name for managing quota management identity |
| `admin_path` | string | Optional | /quota | Prefix for the path to manage quota requests |
| `redis` | object | Yes | | Redis related configuration |
Explanation of each configuration field in `redis`
| Configuration Item | Type | Required | Default Value | Explanation |
|---------------------|------------------|----------|---------------------------------------------------------|-----------------------------------------------|
| service_name | string | Required | - | Redis service name, full FQDN name with service type, e.g., my-redis.dns, redis.my-ns.svc.cluster.local |
| service_port | int | No | Default value for static service is 80; others are 6379 | Service port for the redis service |
| username | string | No | - | Redis username |
| password | string | No | - | Redis password |
| timeout | int | No | 1000 | Redis connection timeout in milliseconds |
## Configuration Example
### Identify request parameter apikey and apply rate limiting accordingly
```yaml
redis_key_prefix: "chat_quota:"
admin_consumer: consumer3
admin_path: /quota
redis:
service_name: redis-service.default.svc.cluster.local
service_port: 6379
timeout: 2000
```
### Refresh Quota
If the suffix of the current request URL matches the admin_path, for example, if the plugin is effective on the route example.com/v1/chat/completions, then the quota can be updated via:
curl https://example.com/v1/chat/completions/quota/refresh -H "Authorization: Bearer credential3" -d "consumer=consumer1&quota=10000"
The value of the key `chat_quota:consumer1` in Redis will be refreshed to 10000.
### Query Quota
To query the quota of a specific user, you can use:
curl https://example.com/v1/chat/completions/quota?consumer=consumer1 -H "Authorization: Bearer credential3"
The response will return: {"quota": 10000, "consumer": "consumer1"}
### Increase or Decrease Quota
To increase or decrease the quota of a specific user, you can use:
curl https://example.com/v1/chat/completions/quota/delta -d "consumer=consumer1&value=100" -H "Authorization: Bearer credential3"
This will increase the value of the key `chat_quota:consumer1` in Redis by 100, and negative values can also be supported, thus subtracting the corresponding value.

21
plugins/wasm-go/extensions/ai-rag/README.md
View File

@@ -1,9 +1,22 @@
# 简介
---
title: AI RAG
keywords: [ AI网关, AI RAG ]
description: AI RAG 插件配置参考
---
## 功能说明
通过对接阿里云向量检索服务实现LLM-RAG，流程如图所示：
<img src="https://img.alicdn.com/imgextra/i1/O1CN01LuRVs41KhoeuzakeF_!!6000000001196-0-tps-1926-1316.jpg" width=600>
# 配置说明
## 运行属性
插件执行阶段：`默认阶段`
插件执行优先级：`400`
## 配置说明
| 名称 | 数据类型 | 填写要求 | 默认值 | 描述 |
|----------------|-----------------|------|-----|----------------------------------------------------------------------------------|
| `dashscope.apiKey` | string | 必填 | - | 用于在访问通义千问服务时进行认证的令牌。 |
@@ -20,7 +33,7 @@
插件开启后，在使用链路追踪功能时，会在span的attribute中添加rag检索到的文档id信息，供排查问题使用。
# 示例
## 示例
```yaml
dashscope:
@@ -54,4 +67,4 @@ dashvector:
经过RAG插件处理后LLM返回的结果为：
```
海南追尾事故发生在海文高速公路文昌至海口方向37公里处。关于事故的具体原因，交警部门当时仍在进一步调查中，所以根据提供的信息无法确定事故的确切原因。人员伤亡情况是1人死亡（司机当场死亡），另有8人受伤（包括2名儿童和6名成人），所有受伤人员都被解救并送往医院进行治疗。
```
```

61
plugins/wasm-go/extensions/ai-rag/README_EN.md Normal file
View File

@@ -0,0 +1,61 @@
---
title: AI RAG
keywords: [ AI Gateway, AI RAG ]
description: AI RAG Plugin Configuration Reference
---
## Function Description
Implement LLM-RAG by integrating with Alibaba Cloud Vector Search Service, as shown in the figure below:
<img src="https://img.alicdn.com/imgextra/i1/O1CN01LuRVs41KhoeuzakeF_!!6000000001196-0-tps-1926-1316.jpg" width=600>
## Running Attributes
Plugin execution phase: `Default Phase`
Plugin execution priority: `400`
## Configuration Description
| Name | Data Type | Requirement | Default Value | Description |
|--------------------------|-----------|-------------|---------------|-------------------------------------------------------------------------------------------|
| `dashscope.apiKey` | string | Required | - | Token used for authentication when accessing Tongyi Qianwen service. |
| `dashscope.serviceFQDN` | string | Required | - | Tongyi Qianwen service name |
| `dashscope.servicePort` | int | Required | - | Tongyi Qianwen service port |
| `dashscope.serviceHost` | string | Required | - | Domain name for accessing Tongyi Qianwen service |
| `dashvector.apiKey` | string | Required | - | Token used for authentication when accessing Alibaba Cloud Vector Search Service. |
| `dashvector.serviceFQDN`| string | Required | - | Alibaba Cloud Vector Search service name |
| `dashvector.servicePort`| int | Required | - | Alibaba Cloud Vector Search service port |
| `dashvector.serviceHost`| string | Required | - | Domain name for accessing Alibaba Cloud Vector Search service |
| `dashvector.topk` | int | Required | - | Number of vectors to retrieve from Alibaba Cloud Vector Search |
| `dashvector.threshold` | float | Required | - | Vector distance threshold; documents above this threshold will be filtered out |
| `dashvector.field` | string | Required | - | Field name where documents are stored in Alibaba Cloud Vector Search |
Once the plugin is enabled, while using the tracing feature, the document ID information retrieved by RAG will be added to the span's attributes for troubleshooting purposes.
## Example
```yaml
dashscope:
apiKey: xxxxxxxxxxxxxxx
serviceFQDN: dashscope
servicePort: 443
serviceHost: dashscope.aliyuncs.com
dashvector:
apiKey: xxxxxxxxxxxxxxxxxxxx
serviceFQDN: dashvector
servicePort: 443
serviceHost: vrs-cn-xxxxxxxxxxxxxxx.dashvector.cn-hangzhou.aliyuncs.com
collection: xxxxxxxxxxxxxxx
topk: 1
threshold: 0.4
field: raw
```
The [CEC-Corpus](https://github.com/shijiebei2009/CEC-Corpus) dataset contains 332 news reports on emergency events, along with annotation data. The original news text is extracted, vectorized, and then added to Alibaba Cloud Vector Search Service. For text vectorization tutorials, you can refer to [“Implementing Semantic Search Based on Vector Search Service and Lingji”](https://help.aliyun.com/document_detail/2510234.html).
Below is an example enhanced using RAG, with the original request being:
```
Where did the rear-end collision in Hainan occur? What was the cause? How many casualties were there?
```
The result returned by LLM without RAG plugin processing was:
```
I'm sorry, as an AI model, I cannot retrieve and update specific information on news events in real time, including details such as location, cause, and casualties. For such specific events, it is recommended that you consult the latest news reports or official announcements for accurate information. You can visit mainstream media websites, use news applications, or follow announcements from relevant government departments to get dynamic updates.
```
After processing with RAG plugin, the result returned by LLM was:
```
The rear-end collision in Hainan occurred on the Haiven Expressway, 37 kilometers from Wenchang to Haikou. Regarding the specific cause of the accident, traffic police were still conducting further investigations at the time, so the exact cause of the accident cannot be determined based on the provided information. The casualty situation is 1 death (the driver died on the spot) and 8 injuries (including 2 children and 6 adults). All injured persons were rescued and sent to the hospital for treatment.
```

8
plugins/wasm-go/extensions/ai-security-guard/README.md
View File

@@ -1,6 +1,6 @@
# 简介
## 功能说明
# 配置说明
## 配置说明
| Name | Type | Requirement | Default | Description |
| :-: | :-: | :-: | :-: | :-: |
| serviceSource | string | requried | - | 服务来源，填dns |
@@ -11,7 +11,7 @@
| sk | string | requried | - | 阿里云SK |
# 配置示例
## 配置示例
```yaml
serviceSource: "dns"
serviceName: "safecheck"
@@ -19,4 +19,4 @@ servicePort: 443
domain: "green-cip.cn-shanghai.aliyuncs.com"
ak: "XXXXXXXXX"
sk: "XXXXXXXXXXXXXXX"
```
```

28
plugins/wasm-go/extensions/ai-token-ratelimit/README.md
View File

@@ -1,10 +1,20 @@
# 功能说明
---
title: AI Token限流
keywords: [ AI网关, AI token限流 ]
description: AI Token限流插件配置参考
---
## 功能说明
`ai-token-ratelimit`插件实现了基于特定键值实现token限流，键值来源可以是 URL 参数、HTTP 请求头、客户端 IP 地址、consumer 名称、cookie中 key 名称
## 运行属性
插件执行阶段：`默认阶段`
插件执行优先级：`600`
# 配置说明
## 配置说明
| 配置项 | 类型 | 必填 | 默认值 | 说明 |
| ----------------------- | ------ | ---- | ------ |---------------------------------------------------------------------------|
@@ -51,9 +61,9 @@
# 配置示例
## 配置示例
## 识别请求参数 apikey，进行区别限流
### 识别请求参数 apikey，进行区别限流
```yaml
rule_name: default_rule
@@ -81,7 +91,7 @@ redis:
## 识别请求头 x-ca-key，进行区别限流
### 识别请求头 x-ca-key，进行区别限流
```yaml
rule_name: default_rule
@@ -109,7 +119,7 @@ redis:
## 根据请求头 x-forwarded-for 获取对端IP，进行区别限流
### 根据请求头 x-forwarded-for 获取对端IP，进行区别限流
```yaml
rule_name: default_rule
@@ -129,7 +139,7 @@ redis:
service_name: redis.static
```
## 识别consumer，进行区别限流
### 识别consumer，进行区别限流
```yaml
rule_name: default_rule
@@ -157,7 +167,7 @@ redis:
## 识别cookie中的键值对，进行区别限流
### 识别cookie中的键值对，进行区别限流
```yaml
rule_name: default_rule
@@ -183,4 +193,4 @@ rejected_code: 200
rejected_msg: '{"code":-1,"msg":"Too many requests"}'
redis:
service_name: redis.static
```
```

170
plugins/wasm-go/extensions/ai-token-ratelimit/README_EN.md Normal file
View File

@@ -0,0 +1,170 @@
---
title: AI Token Rate Limiting
keywords: [ AI Gateway, AI Token Rate Limiting ]
description: AI Token Rate Limiting Plugin Configuration Reference
---
## Function Description
The `ai-token-ratelimit` plugin implements token rate limiting based on specific key values. The key values can come from URL parameters, HTTP request headers, client IP addresses, consumer names, or key names in cookies.
## Runtime Attributes
Plugin execution phase: `default phase`
Plugin execution priority: `600`
## Configuration Description
| Configuration Item | Type | Required | Default Value | Description |
| ----------------------- | ----------------- | -------- | ------------- | ----------------------------------------------------------------------------- |
| rule_name | string | Yes | - | Name of the rate limiting rule, used to assemble the redis key based on the rule name + rate limiting type + rate limiting key name + actual value corresponding to the rate limiting key |
| rule_items | array of object | Yes | - | Rate limiting rule items. After matching the first rule_item, subsequent rules will be ignored based on the order in `rule_items` |
| rejected_code | int | No | 429 | The HTTP status code returned when the request is rate limited |
| rejected_msg | string | No | Too many requests | The response body returned when the request is rate limited |
| redis | object | Yes | - | Redis related configuration |
Field descriptions for each item in `rule_items`
| Configuration Item | Type | Required | Default Value | Description |
| ------------------------ | ----------------- | --------------------------- | ------------- | ------------------------------------------------------------ |
| limit_by_header | string | No, optionally select one in `limit_by_*` | - | Configure the source HTTP header name for obtaining the rate limiting key value |
| limit_by_param | string | No, optionally select one in `limit_by_*` | - | Configure the source URL parameter name for obtaining the rate limiting key value |
| limit_by_consumer | string | No, optionally select one in `limit_by_*` | - | Rate limit by consumer name, no actual value needs to be added |
| limit_by_cookie | string | No, optionally select one in `limit_by_*` | - | Configure the source key name in cookies for obtaining the rate limiting key value |
| limit_by_per_header | string | No, optionally select one in `limit_by_*` | - | Match specific HTTP request headers according to rules and calculate rate limiting separately for each header. Configure the source HTTP header name for obtaining the rate limiting key value. Supports regular expressions or `*` when configuring `limit_keys` |
| limit_by_per_param | string | No, optionally select one in `limit_by_*` | - | Match specific URL parameters according to rules and calculate rate limiting separately for each parameter. Configure the source URL parameter name for obtaining the rate limiting key value. Supports regular expressions or `*` when configuring `limit_keys` |
| limit_by_per_consumer | string | No, optionally select one in `limit_by_*` | - | Match specific consumers according to rules and calculate rate limiting separately for each consumer. Rate limit by consumer name, no actual value needs to be added. Supports regular expressions or `*` when configuring `limit_keys` |
| limit_by_per_cookie | string | No, optionally select one in `limit_by_*` | - | Match specific cookies according to rules and calculate rate limiting separately for each cookie. Configure the source key name in cookies for obtaining the rate limiting key value. Supports regular expressions or `*` when configuring `limit_keys` |
| limit_by_per_ip | string | No, optionally select one in `limit_by_*` | - | Match specific IPs according to rules and calculate rate limiting separately for each IP. Configure the source IP parameter name for obtaining the rate limiting key value from request headers, `from-header-<header name>`, such as `from-header-x-forwarded-for`. Directly get the remote socket IP by configuring `from-remote-addr` |
| limit_keys | array of object | Yes | - | Configure the number of rate limit requests after matching keys |
Field descriptions for each item in `limit_keys`
| Configuration Item | Type | Required | Default Value | Description |
| ----------------------- | ----------------- | ------------------------------------------- | ------------- | ----------------------------------------------- |
| key | string | Yes | - | Matched key value. Types `limit_by_per_header`, `limit_by_per_param`, `limit_by_per_consumer`, `limit_by_per_cookie` support configuring regular expressions (beginning with regexp: followed by the regex) or `*` (representing all). Example regex: `regexp:^d.*` (all strings starting with d); `limit_by_per_ip` supports configuring IP addresses or IP segments |
| token_per_second | int | No, optionally select one in `token_per_second`, `token_per_minute`, `token_per_hour`, `token_per_day` | - | Allowed number of token requests per second |
| token_per_minute | int | No, optionally select one in `token_per_second`, `token_per_minute`, `token_per_hour`, `token_per_day` | - | Allowed number of token requests per minute |
| token_per_hour | int | No, optionally select one in `token_per_second`, `token_per_minute`, `token_per_hour`, `token_per_day` | - | Allowed number of token requests per hour |
| token_per_day | int | No, optionally select one in `token_per_second`, `token_per_minute`, `token_per_hour`, `token_per_day` | - | Allowed number of token requests per day |
Field descriptions for each item in `redis`
| Configuration Item | Type | Required | Default Value | Description |
| ----------------------- | ----------------- | -------- | --------------------------------------------------------------- | ----------------------------------------------- |
| service_name | string | Required | - | Full FQDN name of the redis service, including service type, e.g., my-redis.dns, redis.my-ns.svc.cluster.local |
| service_port | int | No | Default value for static addresses (static service) is 80; otherwise, it is 6379 | Input the service port of the redis service |
| username | string | No | - | Redis username |
| password | string | No | - | Redis password |
| timeout | int | No | 1000 | Redis connection timeout in milliseconds |
## Configuration Examples
### Identify request parameter apikey for differentiated rate limiting
```yaml
rule_name: default_rule
rule_items:
- limit_by_param: apikey
limit_keys:
- key: 9a342114-ba8a-11ec-b1bf-00163e1250b5
token_per_minute: 10
- key: a6a6d7f2-ba8a-11ec-bec2-00163e1250b5
token_per_hour: 100
- limit_by_per_param: apikey
limit_keys:
# Regular expression, matches all strings starting with a, each apikey corresponds to 10 qds
- key: "regexp:^a.*"
token_per_second: 10
# Regular expression, matches all strings starting with b, each apikey corresponds to 100 qd
- key: "regexp:^b.*"
token_per_minute: 100
# Fallback, matches all requests, each apikey corresponds to 1000 qdh
- key: "*"
token_per_hour: 1000
redis:
service_name: redis.static
```
### Identify request header x-ca-key for differentiated rate limiting
```yaml
rule_name: default_rule
rule_items:
- limit_by_header: x-ca-key
limit_keys:
- key: 102234
token_per_minute: 10
- key: 308239
token_per_hour: 10
- limit_by_per_header: x-ca-key
limit_keys:
# Regular expression, matches all strings starting with a, each apikey corresponds to 10 qds
- key: "regexp:^a.*"
token_per_second: 10
# Regular expression, matches all strings starting with b, each apikey corresponds to 100 qd
- key: "regexp:^b.*"
token_per_minute: 100
# Fallback, matches all requests, each apikey corresponds to 1000 qdh
- key: "*"
token_per_hour: 1000
redis:
service_name: redis.static
```
### Get the peer IP using the request header x-forwarded-for for differentiated rate limiting
```yaml
rule_name: default_rule
rule_items:
- limit_by_per_ip: from-header-x-forwarded-for
limit_keys:
# Exact IP
- key: 1.1.1.1
token_per_day: 10
# IP segment, matching IPs in this segment, each IP 100 qpd
- key: 1.1.1.0/24
token_per_day: 100
# Fallback, i.e., default each IP 1000 qpd
- key: 0.0.0.0/0
token_per_day: 1000
redis:
service_name: redis.static
```
### Identify consumer for differentiated rate limiting
```yaml
rule_name: default_rule
rule_items:
- limit_by_consumer: ''
limit_keys:
- key: consumer1
token_per_second: 10
- key: consumer2
token_per_hour: 100
- limit_by_per_consumer: ''
limit_keys:
# Regular expression, matches all strings starting with a, each consumer corresponds to 10 qds
- key: "regexp:^a.*"
token_per_second: 10
# Regular expression, matches all strings starting with b, each consumer corresponds to 100 qd
- key: "regexp:^b.*"
token_per_minute: 100
# Fallback, matches all requests, each consumer corresponds to 1000 qdh
- key: "*"
token_per_hour: 1000
redis:
service_name: redis.static
```
### Identify key-value pairs in cookies for differentiated rate limiting
```yaml
rule_name: default_rule
rule_items:
- limit_by_cookie: key1
limit_keys:
- key: value1
token_per_minute: 10
- key: value2
token_per_hour: 100
- limit_by_per_cookie: key1
limit_keys:
# Regular expression, matches all strings starting with a, each value in cookie corresponds to 10 qds
- key: "regexp:^a.*"
token_per_second: 10
# Regular expression, matches all strings starting with b, each value in cookie corresponds to 100 qd
- key: "regexp:^b.*"
token_per_minute: 100
# Fallback, matches all requests, each value in cookie corresponds to 1000 qdh
- key: "*"
token_per_hour: 1000
rejected_code: 200
rejected_msg: '{"code":-1,"msg":"Too many requests"}'
redis:
service_name: redis.static
```

20
plugins/wasm-go/extensions/ai-transformer/README.md
View File

@@ -1,7 +1,19 @@
# 简介
低代码开发插件，通过LLM对请求/响应的header以及body进行修改。
---
title: AI 请求响应转换
keywords: [higress,AI transformer]
description: AI 请求响应转换插件配置参考
---
# 配置说明
## 功能说明
AI 请求响应转换插件，通过LLM对请求/响应的header以及body进行修改。
## 运行属性
插件执行阶段：`认证阶段`
插件执行优先级：`410`
## 配置说明
| Name | Type | Requirement | Default | Description |
| :- | :- | :- | :- | :- |
| request.enable | bool | requried | - | 是否在request阶段开启转换 |
@@ -12,7 +24,7 @@
| provider.domain | string | requried | - | LLM服务域名 |
| provider.apiKey | string | requried | - | 阿里云dashscope服务的API Key |
# 配置示例
## 配置示例
```yaml
request:
enable: false

85
plugins/wasm-go/extensions/ai-transformer/README_EN.md Normal file
View File

@@ -0,0 +1,85 @@
---
title: AI Request-Response Transformation
keywords: [higress, AI transformer]
description: AI Request-Response Transformation plugin configuration reference
---
## Function Description
The AI Request-Response Transformation plugin modifies the header and body of requests/responses using LLM.
## Execution Attributes
Plugin execution phase: `Authentication Phase`
Plugin execution priority: `410`
## Configuration Description
| Name | Type | Requirement | Default | Description |
| :- | :- | :- | :- | :- |
| request.enable | bool | required | - | Whether to enable transformation in the request phase |
| request.prompt | string | required | - | Prompt used for transformation in the request phase |
| response.enable | bool | required | - | Whether to enable transformation in the response phase |
| response.prompt | string | required | - | Prompt used for transformation in the response phase |
| provider.serviceName | string | required | - | DNS type service name, currently only supports Qwen |
| provider.domain | string | required | - | LLM service domain |
| provider.apiKey | string | required | - | Alibaba Cloud Dashscope service API Key |
## Configuration Example
```yaml
request:
enable: false
prompt: "If the request path starts with /httpbin, please remove the /httpbin prefix and do not change anything else."
response:
enable: true
prompt: "Please modify the following HTTP response information with the requirements: 1. change content-type to application/json; 2. convert body from xml to json; 3. remove content-length."
provider:
serviceName: qwen
domain: dashscope.aliyuncs.com
apiKey: xxxxxxxxxxxxx
```
Accessing the original httpbin's /xml interface yields:
```
<?xml version='1.0' encoding='us-ascii'?>
<!-- A SAMPLE set of slides -->
<slideshow
title="Sample Slide Show"
date="Date of publication"
author="Yours Truly"
>
<!-- TITLE SLIDE -->
<slide type="all">
<title>Wake up to WonderWidgets!</title>
</slide>
<!-- OVERVIEW -->
<slide type="all">
<title>Overview</title>
<item>Why <em>WonderWidgets</em> are great</item>
<item/>
<item>Who <em>buys</em> WonderWidgets</item>
</slide>
</slideshow>
```
Using the above configuration, accessing the httpbin's /xml interface through the gateway yields:
```
{
"slideshow": {
"title": "Sample Slide Show",
"date": "Date of publication",
"author": "Yours Truly",
"slides": [
{
"type": "all",
"title": "Wake up to WonderWidgets!"
},
{
"type": "all",
"title": "Overview",
"items": [
"Why <em>WonderWidgets</em> are great",
"",
"Who <em>buys</em> WonderWidgets"
]
}
]
}
}
```

34
plugins/wasm-go/extensions/basic-auth/README.md
View File

@@ -14,12 +14,17 @@ description: Basic 认证插件配置参考
## 配置字段
### 全局配置
**注意：**
| 名称 | 数据类型 | 填写要求 | 默认值 | 描述 |
| ----------- | --------------- | -------- | ------ | ---------------------------------------------------- |
| `consumers` | array of object | 必填 | - | 配置服务的调用者，用于对请求进行认证 |
| `global_auth` | bool | 选填 | - | 若配置为true，则全局生效认证机制; 若配置为false，则只对做了配置的域名和路由生效认证机制; 若不配置则仅当没有域名和路由配置时全局生效（兼容机制） |
- 在一个规则里，鉴权配置和认证配置不可同时存在
- 对于通过认证鉴权的请求，请求的header会被添加一个`X-Mse-Consumer`字段，用以标识调用者的名称。
### 认证配置
| 名称 | 数据类型 | 填写要求 | 默认值 | 描述 |
| ----------- | --------------- | -------- | ------ | ---------------------------------------------------- |
| `global_auth` | bool | 选填（**仅实例级别配置**） | - | 只能在实例级别配置，若配置为true，则全局生效认证机制; 若配置为false，则只对做了配置的域名和路由生效认证机制，若不配置则仅当没有域名和路由配置时全局生效（兼容老用户使用习惯）。 |
| `consumers` | array of object | 必填 | - | 配置服务的调用者，用于对请求进行认证 |
`consumers`中每一项的配置字段说明如下：
@@ -28,22 +33,20 @@ description: Basic 认证插件配置参考
| `credential` | string | 必填 | - | 配置该consumer的访问凭证 |
| `name` | string | 必填 | - | 配置该consumer的名称 |
### 域名和路由级配置
### 鉴权配置（非必需）
| 名称 | 数据类型 | 填写要求 | 默认值 | 描述 |
| ---------------- | --------------- | ------------------------------------------------- | ------ | -------------------------------------------------- |
| `allow` | array of string | 必填 | - | 对于符合匹配条件的请求，配置允许访问的consumer名称 |
**注意：**
- 对于通过认证鉴权的请求，请求的header会被添加一个`X-Mse-Consumer`字段，用以标识调用者的名称。
## 配置示例
### 对特定路由或域名开启认证和鉴权
### 全局配置认证和路由粒度进行鉴权
以下配置将对网关特定路由或域名开启 Basic Auth 认证和鉴权，注意凭证信息中的用户名和密码之间使用":"分隔，`credential`字段不能重复
**全局配置**
在实例级别做如下插件配置：
```yaml
consumers:
@@ -54,9 +57,6 @@ consumers:
global_auth: false
```
**路由级配置**
对 route-a 和 route-b 这两个路由做如下配置：
```yaml
@@ -75,7 +75,7 @@ allow:
此例指定的 `*.example.com` 和 `test.com` 用于匹配请求的域名，当发现域名匹配时，将允许`name`为`consumer2`的调用者访问，其他调用者不允许访问。
#### 根据该配置，下列请求可以允许访问：
根据该配置，下列请求可以允许访问：
**请求指定用户名密码**
@@ -89,7 +89,7 @@ curl -H 'Authorization: Basic YWRtaW46MTIzNDU2' http://xxx.hello.com/test
认证鉴权通过后，请求的header中会被添加一个`X-Mse-Consumer`字段，在此例中其值为`consumer1`，用以标识调用方的名称
#### 下列请求将拒绝访问：
下列请求将拒绝访问：
**请求未提供用户名密码，返回401**
```bash
@@ -111,4 +111,4 @@ curl -u guest:abc http://xxx.hello.com/test
| ----------- |--------------------------------------------------------------------------------| ---------------------- |
| 401 | Request denied by Basic Auth check. No Basic Authentication information found. | 请求未提供凭证 |
| 401 | Request denied by Basic Auth check. Invalid username and/or password. | 请求凭证无效 |
| 403 | Request denied by Basic Auth check. Unauthorized consumer. | 请求的调用方无访问权限 |
| 403 | Request denied by Basic Auth check. Unauthorized consumer. | 请求的调用方无访问权限 |

140
plugins/wasm-go/extensions/basic-auth/README_EN.md
View File

@@ -1,119 +1,99 @@
---
title: Basic Auth
keywords: [higress, basic auth]
description: Basic authentication plug-in configuration reference
title: Basic Authentication
keywords: [higress,basic auth]
description: Basic authentication plugin configuration reference
---
## Function Description
The `basic-auth` plugin implements authentication and authorization based on the HTTP Basic Auth standard.
## Description
`basic-auth` plugin implements the function of authentication based on the HTTP Basic Auth standard.
## Operation Attributes
Plugin execution stage: `Authentication Phase`
Plugin execution priority: `320`
## Configuration Fields
| Name | Type | Requirement | Default Value | Description |
| ----------- | --------------- | -------- | ------ | ---------------------------------------------------- |
| `consumers` | array of object | Required | - | Caller of the service for authentication of requests |
| `_rules_` | array of object | Optional | - | Configure access permission list for specific routes or domains to authenticate requests |
Filed descriptions of `consumers` items:
| Name | Type | Requirement | Default Value | Description |
| ------------ | ------ | ----------- | ------------- | ------------------------------------- |
| `credential` | string | Required | - | Credential for this consumer's access |
| `name` | string | Required | - | Name of this consumer |
Configuration field descriptions for each item in `_rules_` are as follows:
| Field Name | Data Type | Requirement | Default | Description |
| ---------------- | --------------- | ------------------------------------------------- | ------ | -------------------------------------------------- |
| `_match_route_` | array of string | One of `_match_route_` or `_match_domain_` | - | Configure the routes to match for request authorization |
| `_match_domain_` | array of string | One of `_match_route_` , `_match_domain_` | - | Configure the domains to match for request authorization |
| `allow` | array of string | Required | - | Configure the consumer names allowed to access requests that match the match condition |
**Note:**
- In one rule, authentication configurations and authorization configurations cannot coexist.
- For requests that pass authentication, the request header will include an `X-Mse-Consumer` field to identify the caller's name.
- If the `_rules_` field is not configured, authentication is enabled for all routes of the current gateway instance by default;
- For authenticated requests, `X-Mse-Consumer` field will be added to the request header to identify the name of the caller.
### Authentication Configuration
| Name | Data Type | Requirements | Default Value | Description |
| ------------- | ---------------- | ------------------------------- | -------------- | ------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------- |
| `global_auth` | bool | Optional (**instance-level only**) | - | Can only be configured at the instance level. If set to true, the authentication mechanism will take effect globally; if set to false, it will only take effect for the configured domains and routes. If not configured, it will only take effect globally when there are no domain and route configurations (compatible with old user habits). |
| `consumers` | array of object | Required | - | Configures the service callers for request authentication. |
## Configuration Samples
### Enable Authentication and Authorization for specific routes or domains
The following configuration will enable Basic Auth authentication and authorization for specific routes or domains of the gateway. Note that the username and password in the credential information are separated by a ":", and the `credential` field cannot be repeated.
Each configuration field in `consumers` is described as follows:
| Name | Data Type | Requirements | Default Value | Description |
| ------------ | --------- | ------------ | ------------- | ------------------------------- |
| `credential` | string | Required | - | Configures the access credentials for this consumer. |
| `name` | string | Required | - | Configures the name of this consumer. |
### Authorization Configuration (Optional)
| Name | Data Type | Requirements | Default Value | Description |
| ---------------- | ---------------- | ---------------------------------------------------- | -------------- | -------------------------------------------------------- |
| `allow` | array of string | Required | - | Configures the consumer names allowed to access for matching requests. |
## Configuration Example
### Global Authentication and Route Granularity Authorization
The following configuration will enable Basic Auth authentication and authorization for specific routes or domains of the gateway. Note that the username and password in the credential information are separated by ":", and the `credential` field cannot be duplicated.
Make the following plugin configuration at the instance level:
```yaml
# use the _rules_ field for fine-grained rule configuration.
consumers:
- credential: 'admin:123456'
name: consumer1
- credential: 'guest:abc'
name: consumer2
_rules_:
# rule 1: match by the route name.
- _match_route_:
- route-a
- route-b
allow:
- consumer1
# rule 2: match by the domain.
- _match_domain_:
- "*.example.com"
- test.com
allow:
- consumer2
global_auth: false
```
In this sample, `route-a` and `route-b` specified in `_match_route_` are the route names filled in when creating gateway routes. When these two routes are matched, the caller with `name` as `consumer1` is allowed to access, and other callers are not allowed to access.
The `*.example.com` and `test.com` specified in `_match_domain_` are used to match the domain name of the request. When the domain name is matched, the caller with `name` as `consumer2` is allowed to access, and other callers are not allowed to access.
For routes `route-a` and `route-b`, configure as follows:
```yaml
allow:
- consumer1
```
For the domains `*.example.com` and `test.com`, configure as follows:
```yaml
allow:
- consumer2
```
#### According to this configuration, the following requests are allowed:
If configured in the console, the specified `route-a` and `route-b` refer to the route names filled in when creating the routes in the console. When matching these two routes, callers with the name `consumer1` will be allowed access, while other callers will not.
**Requests with specified username and password**
The specified `*.example.com` and `test.com` are used to match the request domain. When a match is found, callers with the name `consumer2` will be allowed access, while other callers will not.
Based on this configuration, the following requests may be allowed access:
**Request with specified username and password**
```bash
# Assuming the following request will match with route-a
# Use -u option of curl to specify the credentials
# Assuming the following request matches the route-a route
# Using curl's -u parameter to specify
curl -u admin:123456 http://xxx.hello.com/test
# Or specify the Authorization request header directly with the credentials in base64 encoding
# Or directly specify the Authorization request header with the username and password encoded in base64
curl -H 'Authorization: Basic YWRtaW46MTIzNDU2' http://xxx.hello.com/test
```
A `X-Mse-Consumer` field will be added to the headers of the request, and its value in this example is `consumer1`, used to identify the name of the caller when passed authentication and authorization.
After successful authentication, the request header will have an added `X-Mse-Consumer` field, which in this case is `consumer1` to identify the caller's name.
#### The following requests will be denied:
**Requests without providing username and password, returning 401**
The following requests will be denied access:
**Request without username and password, returns 401**
```bash
curl http://xxx.hello.com/test
```
**Requests with incorrect username or password, returning 401**
**Request with incorrect username and password, returns 401**
```bash
curl -u admin:abc http://xxx.hello.com/test
```
**Requests matched with a caller who has no access permission, returning 403**
**Caller matched by username and password has no access, returns 403**
```bash
# consumer2 is not in the allow list of route-a
# consumer2 is not in the allow list for route-a
curl -u guest:abc http://xxx.hello.com/test
```
### Enable basic auth for gateway instance
The following configuration does not specify the `_rules_` field, so Basic Auth authentication will be effective for the whole gateway instance.
```yaml
consumers:
- credential: 'admin:123456'
name: consumer1
- credential: 'guest:abc'
name: consumer2
```
## Error Codes
| HTTP Status Code | Error Info | Reason |
| ----------- |--------------------------------------------------------------------------------| ---------------------- |
| 401 | Request denied by Basic Auth check. No Basic Authentication information found. | Credentials not provided in the request |
| 401 | Request denied by Basic Auth check. Invalid username and/or password. | Invalid username and/or password |
| 403 | Request denied by Basic Auth check. Unauthorized consumer. | Unauthorized consumer |
## Related Error Codes
| HTTP Status Code | Error Message | Reason Description |
| ---------------- | ------------------------------------------------------------------------------------- | -------------------------------- |
| 401 | Request denied by Basic Auth check. No Basic Authentication information found. | Request did not provide credentials. |
| 401 | Request denied by Basic Auth check. Invalid username and/or password. | Request credentials are invalid. |
| 403 | Request denied by Basic Auth check. Unauthorized consumer. | The caller making the request does not have access. |

28
plugins/wasm-go/extensions/bot-detect/README.md
View File

@@ -1,11 +1,21 @@
<p>
<a href="README_EN.md"> English </a> | 中文
</p>
---
title: Bot 拦截
keywords: [higress,bot detect]
description: Bot 拦截插件配置参考
---
## 功能说明
# 功能说明
`bot-detect`插件可以用于识别并阻止互联网爬虫对站点资源的爬取
# 配置字段
## 运行属性
插件执行阶段：`授权阶段`
插件执行优先级：`310`
## 配置字段
| 名称 | 数据类型 | 填写要求 | 默认值 | 描述 |
| -------- | -------- | -------- | -------- | -------- |
@@ -33,9 +43,9 @@
(CSimpleSpider|Cityreview Robot|CrawlDaddy|CrawlFire|Finderbots|Index crawler|Job Roboter|KiwiStatus Spider|Lijit Crawler|QuerySeekerSpider|ScollSpider|Trends Crawler|USyd-NLP-Spider|SiteCat Webbot|BotName\/\$BotVersion|123metaspider-Bot|1470\.net crawler|50\.nu|8bo Crawler Bot|Aboundex|Accoona-[A-z]{1,30}-Agent|AdsBot-Google(?:-[a-z]{1,30}|)|altavista|AppEngine-Google|archive.{0,30}\.org_bot|archiver|Ask Jeeves|[Bb]ai[Dd]u[Ss]pider(?:-[A-Za-z]{1,30})(?:-[A-Za-z]{1,30}|)|bingbot|BingPreview|blitzbot|BlogBridge|Bloglovin|BoardReader Blog Indexer|BoardReader Favicon Fetcher|boitho.com-dc|BotSeer|BUbiNG|\b\w{0,30}favicon\w{0,30}\b|\bYeti(?:-[a-z]{1,30}|)|Catchpoint(?: bot|)|[Cc]harlotte|Checklinks|clumboot|Comodo HTTP\(S\) Crawler|Comodo-Webinspector-Crawler|ConveraCrawler|CRAWL-E|CrawlConvera|Daumoa(?:-feedfetcher|)|Feed Seeker Bot|Feedbin|findlinks|Flamingo_SearchEngine|FollowSite Bot|furlbot|Genieo|gigabot|GomezAgent|gonzo1|(?:[a-zA-Z]{1,30}-|)Googlebot(?:-[a-zA-Z]{1,30}|)|Google SketchUp|grub-client|gsa-crawler|heritrix|HiddenMarket|holmes|HooWWWer|htdig|ia_archiver|ICC-Crawler|Icarus6j|ichiro(?:/mobile|)|IconSurf|IlTrovatore(?:-Setaccio|)|InfuzApp|Innovazion Crawler|InternetArchive|IP2[a-z]{1,30}Bot|jbot\b|KaloogaBot|Kraken|Kurzor|larbin|LEIA|LesnikBot|Linguee Bot|LinkAider|LinkedInBot|Lite Bot|Llaut|lycos|Mail\.RU_Bot|masscan|masidani_bot|Mediapartners-Google|Microsoft .{0,30} Bot|mogimogi|mozDex|MJ12bot|msnbot(?:-media {0,2}|)|msrbot|Mtps Feed Aggregation System|netresearch|Netvibes|NewsGator[^/]{0,30}|^NING|Nutch[^/]{0,30}|Nymesis|ObjectsSearch|OgScrper|Orbiter|OOZBOT|PagePeeker|PagesInventory|PaxleFramework|Peeplo Screenshot Bot|PlantyNet_WebRobot|Pompos|Qwantify|Read%20Later|Reaper|RedCarpet|Retreiver|Riddler|Rival IQ|scooter|Scrapy|Scrubby|searchsight|seekbot|semanticdiscovery|SemrushBot|Simpy|SimplePie|SEOstats|SimpleRSS|SiteCon|Slackbot-LinkExpanding|Slack-ImgProxy|Slurp|snappy|Speedy Spider|Squrl Java|Stringer|TheUsefulbot|ThumbShotsBot|Thumbshots\.ru|Tiny Tiny RSS|Twitterbot|WhatsApp|URL2PNG|Vagabondo|VoilaBot|^vortex|Votay bot|^voyager|WASALive.Bot|Web-sniffer|WebThumb|WeSEE:[A-z]{1,30}|WhatWeb|WIRE|WordPress|Wotbox|www\.almaden\.ibm\.com|Xenu(?:.s|) Link Sleuth|Xerka [A-z]{1,30}Bot|yacy(?:bot|)|YahooSeeker|Yahoo! Slurp|Yandex\w{1,30}|YodaoBot(?:-[A-z]{1,30}|)|YottaaMonitor|Yowedo|^Zao|^Zao-Crawler|ZeBot_www\.ze\.bz|ZooShot|ZyBorg)(?:[ /]v?(\d+)(?:\.(\d+)(?:\.(\d+)|)|)|)
```
# 配置示例
## 配置示例
## 放行原本命中爬虫规则的请求
### 放行原本命中爬虫规则的请求
```yaml
allow:
- ".*Go-http-client.*"
@@ -44,7 +54,7 @@ allow:
若不作该配置，默认的 Golang 网络库请求会被视做爬虫，被禁止访问
## 增加爬虫判断
### 增加爬虫判断
```yaml
deny:
- "spd-tools.*"
@@ -55,4 +65,4 @@ deny:
```bash
curl http://example.com -H 'User-Agent: spd-tools/1.1'
curl http://exmaple.com -H 'User-Agent: spd-tools'
```
```

49
plugins/wasm-go/extensions/bot-detect/README_EN.md
View File

@@ -1,22 +1,26 @@
<p>
English | <a href="README.md">中文</a>
</p>
---
title: Bot Detect
keywords: [higress, bot detect]
description: Bot detect plugin configuration reference
---
## Function Description
The `bot-detect` plugin can be used to identify and block internet crawlers from accessing site resources.
# Description
`bot-detect` plugin can be used to identify and prevent web crawlers from crawling websites.
## Running Properties
Plugin Execution Phase: `Authorization Phase`
Plugin Execution Priority: `310`
# Configuration Fields
## Configuration Fields
| Name | Data Type | Required | Default Value | Description |
| ----------------- | ------------------- | --------------| --------------| ---------------------------------------------------------- |
| allow | array of string | Optional | - | Regular expressions to match the User-Agent request header; requests matching will be allowed to access. |
| deny | array of string | Optional | - | Regular expressions to match the User-Agent request header; requests matching will be blocked. |
| blocked_code | number | Optional | 403 | HTTP status code returned when a request is blocked. |
| blocked_message | string | Optional | - | HTTP response body returned when a request is blocked. |
| Name | Type | Requirement | Default Value | Description |
| -------- | -------- | -------- | -------- | -------- |
| allow | array of string | Optional | - | A regular expression to match the User-Agent request header and will allow access if the match hits |
| deny | array of string | Optional | - | A regular expression to match the User-Agent request header and will block the request if the match hits |
| blocked_code | number | Optional | 403 | The HTTP status code returned when a request is blocked |
| blocked_message | string | Optional | - | The HTTP response Body returned when a request is blocked |
The `allow` and `deny` fields can both be left unconfigured, in which case the default crawler identification logic will be executed. Configuring the `allow` field can allow requests that would otherwise hit the default crawler identification logic. Configuring the `deny` field can add additional crawler identification logic.
If field `allow` and field `deny` are not configured at the same time, the default logic to identify crawlers will be executed. By configuring the `allow` field, requests that would otherwise hit the default logic can be allowed. The judgement can be extended by configuring the `deny` field
The default set of crawler judgment regular expressions is as follows：
The default crawler identification regular expression set is as follows:
```bash
# Bots General matcher 'name/0.0'
@@ -33,26 +37,23 @@ The default set of crawler judgment regular expressions is as follows：
(CSimpleSpider|Cityreview Robot|CrawlDaddy|CrawlFire|Finderbots|Index crawler|Job Roboter|KiwiStatus Spider|Lijit Crawler|QuerySeekerSpider|ScollSpider|Trends Crawler|USyd-NLP-Spider|SiteCat Webbot|BotName\/\$BotVersion|123metaspider-Bot|1470\.net crawler|50\.nu|8bo Crawler Bot|Aboundex|Accoona-[A-z]{1,30}-Agent|AdsBot-Google(?:-[a-z]{1,30}|)|altavista|AppEngine-Google|archive.{0,30}\.org_bot|archiver|Ask Jeeves|[Bb]ai[Dd]u[Ss]pider(?:-[A-Za-z]{1,30})(?:-[A-Za-z]{1,30}|)|bingbot|BingPreview|blitzbot|BlogBridge|Bloglovin|BoardReader Blog Indexer|BoardReader Favicon Fetcher|boitho.com-dc|BotSeer|BUbiNG|\b\w{0,30}favicon\w{0,30}\b|\bYeti(?:-[a-z]{1,30}|)|Catchpoint(?: bot|)|[Cc]harlotte|Checklinks|clumboot|Comodo HTTP\(S\) Crawler|Comodo-Webinspector-Crawler|ConveraCrawler|CRAWL-E|CrawlConvera|Daumoa(?:-feedfetcher|)|Feed Seeker Bot|Feedbin|findlinks|Flamingo_SearchEngine|FollowSite Bot|furlbot|Genieo|gigabot|GomezAgent|gonzo1|(?:[a-zA-Z]{1,30}-|)Googlebot(?:-[a-zA-Z]{1,30}|)|Google SketchUp|grub-client|gsa-crawler|heritrix|HiddenMarket|holmes|HooWWWer|htdig|ia_archiver|ICC-Crawler|Icarus6j|ichiro(?:/mobile|)|IconSurf|IlTrovatore(?:-Setaccio|)|InfuzApp|Innovazion Crawler|InternetArchive|IP2[a-z]{1,30}Bot|jbot\b|KaloogaBot|Kraken|Kurzor|larbin|LEIA|LesnikBot|Linguee Bot|LinkAider|LinkedInBot|Lite Bot|Llaut|lycos|Mail\.RU_Bot|masscan|masidani_bot|Mediapartners-Google|Microsoft .{0,30} Bot|mogimogi|mozDex|MJ12bot|msnbot(?:-media {0,2}|)|msrbot|Mtps Feed Aggregation System|netresearch|Netvibes|NewsGator[^/]{0,30}|^NING|Nutch[^/]{0,30}|Nymesis|ObjectsSearch|OgScrper|Orbiter|OOZBOT|PagePeeker|PagesInventory|PaxleFramework|Peeplo Screenshot Bot|PlantyNet_WebRobot|Pompos|Qwantify|Read%20Later|Reaper|RedCarpet|Retreiver|Riddler|Rival IQ|scooter|Scrapy|Scrubby|searchsight|seekbot|semanticdiscovery|SemrushBot|Simpy|SimplePie|SEOstats|SimpleRSS|SiteCon|Slackbot-LinkExpanding|Slack-ImgProxy|Slurp|snappy|Speedy Spider|Squrl Java|Stringer|TheUsefulbot|ThumbShotsBot|Thumbshots\.ru|Tiny Tiny RSS|Twitterbot|WhatsApp|URL2PNG|Vagabondo|VoilaBot|^vortex|Votay bot|^voyager|WASALive.Bot|Web-sniffer|WebThumb|WeSEE:[A-z]{1,30}|WhatWeb|WIRE|WordPress|Wotbox|www\.almaden\.ibm\.com|Xenu(?:.s|) Link Sleuth|Xerka [A-z]{1,30}Bot|yacy(?:bot|)|YahooSeeker|Yahoo! Slurp|Yandex\w{1,30}|YodaoBot(?:-[A-z]{1,30}|)|YottaaMonitor|Yowedo|^Zao|^Zao-Crawler|ZeBot_www\.ze\.bz|ZooShot|ZyBorg)(?:[ /]v?(\d+)(?:\.(\d+)(?:\.(\d+)|)|)|)
```
# Configuration Samples
## Release Requests that would otherwise Hit the Crawler Rules
## Configuration Example
### Allowing Requests That Hit the Crawler Rules
```yaml
allow:
- ".*Go-http-client.*"
```
Without this configuration, the default Golang web library request will be treated as a crawler and access will be denied.
If this configuration is not made, requests from the default Golang network library will be treated as crawlers and blocked.
## Add Crawler Judgement
### Adding Crawler Identification
```yaml
deny:
- "spd-tools.*"
```
According to this configuration, the following requests will be denied:
With this configuration, the following requests will be blocked:
```bash
curl http://example.com -H 'User-Agent: spd-tools/1.1'
curl http://exmaple.com -H 'User-Agent: spd-tools'
```
```

17
plugins/wasm-go/extensions/cache-control/README.md
View File

@@ -1,14 +1,25 @@
# 功能说明
---
title: 浏览器缓存控制
keywords: [higress,浏览器缓存控制]
description: 浏览器缓存控制插件配置参考
---
## 功能说明
`cache-control`插件实现了基于 URL 文件后缀来为请求的响应头部添加 `Expires` 和 `Cache-Control` 头部，从而方便浏览器对特定后缀的文件进行缓存，例如 `jpg`、`png` 等图片文件。
# 配置字段
## 运行属性
插件执行阶段：`认证阶段`
插件执行优先级：`420`
## 配置字段
| 名称 | 数据类型 | 填写要求 | 默认值 | 描述 |
|---------|--------|-----------------------------------------------------------------------------------------------------|-|--------------------------|
| suffix | string | 选填，表示匹配的文件后缀名，例如 `jpg`、`png` 等。<br/>如果需要匹配多种后缀，需要用 `\|` 进行分割，例如 `png\|jpg`。<br/>如果不填写，表示匹配所有后缀 | - | 配置用于匹配的请求文件后缀 |
| expires | string | 必填，表示缓存的最长时间。<br/>当填入的字符串为数字时，单位为秒，例如需要缓存1小时，需填写 3600。<br/>另外，还可以填写 epoch 或 max<br/>，与 nginx 中语义相同。 | - | 配置缓存的最大时间 |
# 配置示例
## 配置示例
1. 缓存后缀为 `jpg`, `png`, `jpeg` 的文件，缓存时间为一小时
```yaml
suffix: jpg|png|jpeg

33
plugins/wasm-go/extensions/cache-control/README_EN.md Normal file
View File

@@ -0,0 +1,33 @@
---
title: Browser Cache Control
keywords: [higress, browser cache control]
description: Browser cache control plugin configuration reference
---
## Function Description
The `cache-control` plugin implements adding `Expires` and `Cache-Control` headers to the response based on the URL file extensions, making it easier for the browser to cache files with specific extensions, such as `jpg`, `png`, and other image files.
## Runtime Attributes
Plugin execution phase: `Authentication Phase`
Plugin execution priority: `420`
## Configuration Fields
| Name | Data Type | Requirements | Default Value | Description |
|-----------|-------------|----------------------------------------------------------------------------------------------------------|---------------|-----------------------------------|
| suffix | string | Optional, indicates the file extensions to match, such as `jpg`, `png`, etc.<br/>If multiple extensions are needed, separate them with `\|`, for example `png\|jpg`.<br/>If not specified, it matches all extensions. | - | Configures the request file extensions to match |
| expires | string | Required, indicates the maximum caching time.<br/>When the input string is a number, the unit is seconds; for example, if you want to cache for 1 hour, enter 3600.<br/>You can also enter epoch or max<br/>, with the same semantics as in nginx. | - | Configures the maximum caching time |
## Configuration Example
1. Cache files with extensions `jpg`, `png`, `jpeg`, with a caching time of one hour
```yaml
suffix: jpg|png|jpeg
expires: 3600
```
With this configuration, the following requests will have `Expires` and `Cache-Control` fields added to the response headers, with an expiration time of 1 hour later.
```bash
curl http://example.com/test.png
curl http://example.com/test.jpg
```
2. Cache all files, with a maximum caching time of `"Thu, 31 Dec 2037 23:55:55 GMT"`
```yaml
expires: max
```

5
plugins/wasm-go/extensions/cluster-key-rate-limit/README.md
View File

@@ -9,6 +9,11 @@ description: 基于 Key 集群限流插件配置参考
`cluster-key-rate-limit` 插件基于 Redis 实现集群限流，适用于需要跨多个 Higress Gateway 实例实现全局一致速率限制的场景。
限流所使用的 Key 可以来源于 URL 参数、HTTP 请求头、客户端 IP 地址、消费者名称或 Cookie 中的 Key。
## 运行属性
插件执行阶段：`默认阶段`
插件执行优先级：`20`
## 配置说明
| 配置项 | 类型 | 必填 | 默认值 | 说明 |

183
plugins/wasm-go/extensions/cluster-key-rate-limit/README_EN.md Normal file
View File

@@ -0,0 +1,183 @@
---
title: Key-Based Cluster Rate Limiting
keywords: [higress, rate-limit]
description: Configuration reference for the Key-Based Cluster Rate Limiting plugin
---
## Function Description
The `cluster-key-rate-limit` plugin implements cluster rate limiting based on Redis, suitable for scenarios that require global consistent rate limiting across multiple Higress Gateway instances.
The Key used for rate limiting can originate from URL parameters, HTTP request headers, client IP addresses, consumer names, or keys in cookies.
## Execution Attributes
Plugin Execution Phase: `default phase`
Plugin Execution Priority: `20`
## Configuration Description
| Configuration Item | Type | Required | Default Value | Description |
|---------------------------|---------------|----------|---------------|-------------------------------------------------------------------------------------------|
| rule_name | string | Yes | - | The name of the rate limiting rule. The Redis key is constructed using rule name + rate limit type + limit key name + actual value of the limit key. |
| rule_items | array of object| Yes | - | Rate limiting rule items. The first matching `rule_item` based on the order under `rule_items` will trigger the rate limiting, and subsequent rules will be ignored. |
| show_limit_quota_header | bool | No | false | Whether to display `X-RateLimit-Limit` (total requests allowed) and `X-RateLimit-Remaining` (remaining requests that can be sent) in the response headers. |
| rejected_code | int | No | 429 | HTTP status code returned when a request is rate limited. |
| rejected_msg | string | No | Too many requests | Response body returned when a request is rate limited. |
| redis | object | Yes | - | Redis related configuration. |
Description of configuration fields for each item in `rule_items`.
| Configuration Item | Type | Required | Default Value | Description |
|---------------------------|---------------|------------------------|---------------|-------------------------------------------------------------------------------------------------------|
| limit_by_header | string | No, one of `limit_by_*` | - | The name of the HTTP request header from which to retrieve the rate limiting key value. |
| limit_by_param | string | No, one of `limit_by_*` | - | The name of the URL parameter from which to retrieve the rate limiting key value. |
| limit_by_consumer | string | No, one of `limit_by_*` | - | Applies rate limiting based on consumer name without needing to add an actual value. |
| limit_by_cookie | string | No, one of `limit_by_*` | - | The name of the key in the Cookie from which to retrieve the rate limiting key value. |
| limit_by_per_header | string | No, one of `limit_by_*` | - | Matches specific HTTP request headers according to the rules and calculates rate limits for each header. The `limit_keys` configuration supports regular expressions or `*`. |
| limit_by_per_param | string | No, one of `limit_by_*` | - | Matches specific URL parameters according to the rules and calculates rate limits for each parameter. The `limit_keys` configuration supports regular expressions or `*`. |
| limit_by_per_consumer | string | No, one of `limit_by_*` | - | Matches specific consumers according to the rules and calculates rate limits for each consumer. The `limit_keys` configuration supports regular expressions or `*`. |
| limit_by_per_cookie | string | No, one of `limit_by_*` | - | Matches specific cookies according to the rules and calculates rate limits for each cookie. The `limit_keys` configuration supports regular expressions or `*`. |
| limit_by_per_ip | string | No, one of `limit_by_*` | - | Matches specific IPs according to the rules and calculates rate limits for each IP. Retrieve via IP parameter name from request headers, defined as `from-header-{header name}`, e.g., `from-header-x-forwarded-for`. To get the remote socket IP directly, use `from-remote-addr`. |
| limit_keys | array of object | Yes | - | Configures the limit counts after matching key values. |
Description of configuration fields for each item in `limit_keys`.
| Configuration Item | Type | Required | Default Value | Description |
|---------------------------|---------------|------------------------------------------------------------------|---------------|--------------------------------------------------------------------|
| key | string | Yes | - | Matched key value; types `limit_by_per_header`, `limit_by_per_param`, `limit_by_per_consumer`, `limit_by_per_cookie` support regular expression configurations (starting with regexp: followed by a regular expression) or `*` (representing all), e.g., `regexp:^d.*` (all strings starting with d); `limit_by_per_ip` supports configuring IP addresses or IP segments. |
| query_per_second | int | No, one of `query_per_second`, `query_per_minute`, `query_per_hour`, `query_per_day` is optional. | - | Allowed number of requests per second. |
| query_per_minute | int | No, one of `query_per_second`, `query_per_minute`, `query_per_hour`, `query_per_day` is optional. | - | Allowed number of requests per minute. |
| query_per_hour | int | No, one of `query_per_second`, `query_per_minute`, `query_per_hour`, `query_per_day` is optional. | - | Allowed number of requests per hour. |
| query_per_day | int | No, one of `query_per_second`, `query_per_minute`, `query_per_hour`, `query_per_day` is optional. | - | Allowed number of requests per day. |
Description of configuration fields for each item in `redis`.
| Configuration Item | Type | Required | Default Value | Description |
|---------------------------|---------------|----------|------------------------------------------------------------|---------------------------------------------------------------------------|
| service_name | string | Required | - | Full FQDN name of the Redis service, including service type, e.g., my-redis.dns, redis.my-ns.svc.cluster.local. |
| service_port | int | No | 80 for static services; otherwise 6379 | Service port for the Redis service. |
| username | string | No | - | Redis username. |
| password | string | No | - | Redis password. |
| timeout | int | No | 1000 | Redis connection timeout in milliseconds. |
## Configuration Examples
### Distinguish rate limiting based on the request parameter apikey
```yaml
rule_name: default_rule
rule_items:
- limit_by_param: apikey
limit_keys:
- key: 9a342114-ba8a-11ec-b1bf-00163e1250b5
query_per_minute: 10
- key: a6a6d7f2-ba8a-11ec-bec2-00163e1250b5
query_per_hour: 100
- limit_by_per_param: apikey
limit_keys:
# Regular expression, matches all strings starting with a, each apikey corresponds to 10qds.
- key: "regexp:^a.*"
query_per_second: 10
# Regular expression, matches all strings starting with b, each apikey corresponds to 100qd.
- key: "regexp:^b.*"
query_per_minute: 100
# As a fallback, matches all requests, each apikey corresponds to 1000qdh.
- key: "*"
query_per_hour: 1000
redis:
service_name: redis.static
show_limit_quota_header: true
```
### Distinguish rate limiting based on the header x-ca-key
```yaml
rule_name: default_rule
rule_items:
- limit_by_header: x-ca-key
limit_keys:
- key: 102234
query_per_minute: 10
- key: 308239
query_per_hour: 10
- limit_by_per_header: x-ca-key
limit_keys:
# Regular expression, matches all strings starting with a, each apikey corresponds to 10qds.
- key: "regexp:^a.*"
query_per_second: 10
# Regular expression, matches all strings starting with b, each apikey corresponds to 100qd.
- key: "regexp:^b.*"
query_per_minute: 100
# As a fallback, matches all requests, each apikey corresponds to 1000qdh.
- key: "*"
query_per_hour: 1000
redis:
service_name: redis.static
show_limit_quota_header: true
```
### Distinguish rate limiting based on the client IP from the request header x-forwarded-for
```yaml
rule_name: default_rule
rule_items:
- limit_by_per_ip: from-header-x-forwarded-for
limit_keys:
# Exact IP
- key: 1.1.1.1
query_per_day: 10
# IP segment, for IPs matching this segment, each IP corresponds to 100qpd.
- key: 1.1.1.0/24
query_per_day: 100
# As a fallback, defaults to 1000 qpd for each IP.
- key: 0.0.0.0/0
query_per_day: 1000
redis:
service_name: redis.static
show_limit_quota_header: true
```
### Distinguish rate limiting based on consumers
```yaml
rule_name: default_rule
rule_items:
- limit_by_consumer: ''
limit_keys:
- key: consumer1
query_per_second: 10
- key: consumer2
query_per_hour: 100
- limit_by_per_consumer: ''
limit_keys:
# Regular expression, matches all strings starting with a, each consumer corresponds to 10qds.
- key: "regexp:^a.*"
query_per_second: 10
# Regular expression, matches all strings starting with b, each consumer corresponds to 100qd.
- key: "regexp:^b.*"
query_per_minute: 100
# As a fallback, matches all requests, each consumer corresponds to 1000qdh.
- key: "*"
query_per_hour: 1000
redis:
service_name: redis.static
show_limit_quota_header: true
```
### Distinguish rate limiting based on key-value pairs in cookies
```yaml
rule_name: default_rule
rule_items:
- limit_by_cookie: key1
limit_keys:
- key: value1
query_per_minute: 10
- key: value2
query_per_hour: 100
- limit_by_per_cookie: key1
limit_keys:
# Regular expression, matches all strings starting with a, each cookie's value corresponds to 10qds.
- key: "regexp:^a.*"
query_per_second: 10
# Regular expression, matches all strings starting with b, each cookie's value corresponds to 100qd.
- key: "regexp:^b.*"
query_per_minute: 100
# As a fallback, matches all requests, each cookie's value corresponds to 1000qdh.
- key: "*"
query_per_hour: 1000
rejected_code: 200
rejected_msg: '{"code":-1,"msg":"Too many requests"}'
redis:
service_name: redis.static
show_limit_quota_header: true
```

59
plugins/wasm-go/extensions/cors/README.md
View File

@@ -1,27 +1,38 @@
# 功能说明
---
title: 跨域资源共享
keywords: [higress,cors]
description: 跨域资源共享插件配置参考
---
## 功能说明
`cors` 插件可以为服务端启用 CORS（Cross-Origin Resource Sharing，跨域资源共享）的返回 http 响应头。
# 配置字段
## 运行属性
| 名称 | 数据类型 | 填写要求 | 默认值 | 描述 |
|-----------------------|-----------------|-------|---------------------------------------------------------------------------------------------------------------------------|------------------------------------------------------------------------------------------------------------------------------------------------------------------------------|
| allow_origins | array of string | 选填 | * | 允许跨域访问的 Origin，格式为 scheme://host:port，示例如 http://example.com:8081。当 allow_credentials 为 false 时，可以使用 * 来表示允许所有 Origin 通过 |
| allow_origin_patterns | array of string | 选填 | - | 允许跨域访问的 Origin 模式匹配， 用 * 匹配域名或者端口， <br/>比如 http://*.example.com -- 匹配域名， http://*.example.com:[8080,9090] -- 匹配域名和指定端口， http://*.example.com:[*] -- 匹配域名和所有端口。单独 * 表示匹配所有域名和端口 |
| allow_methods | array of string | 选填 | GET, PUT, POST, DELETE, PATCH, OPTIONS | 允许跨域访问的 Method，比如：GET，POST 等。可以使用 * 来表示允许所有 Method。 |
| allow_headers | array of string | 选填 | DNT,X-CustomHeader,Keep-Alive,User-Agent,X-Requested-With，<br/>If-Modified-Since,Cache-Control,Content-Type,Authorization | 允许跨域访问时请求方携带哪些非 CORS 规范以外的 Header。可以使用 * 来表示允许任意 Header。 |
| expose_headers | array of string | 选填 | - | 允许跨域访问时响应方携带哪些非 CORS 规范以外的 Header。可以使用 * 来表示允许任意 Header。 |
| allow_credentials | bool | 选填 | false | 是否允许跨域访问的请求方携带凭据（如 Cookie 等）。根据 CORS 规范，如果设置该选项为 true，在 allow_origins 不能使用 *， 替换成使用 allow_origin_patterns * |
| max_age | number | 选填 | 86400秒 | 浏览器缓存 CORS 结果的最大时间，单位为秒。<br/>在这个时间范围内，浏览器会复用上一次的检查结果 |
插件执行阶段：`授权阶段`
插件执行优先级：`340`
## 配置字段
| 名称 | 数据类型 | 填写要求 | 默认值 | 描述 |
|-----------------------|-----------------|----------|----------------------------------------------------------------------------------------------------------------------------|----------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------|
| allow_origins | array of string | 选填 | * | 允许跨域访问的 Origin，格式为 scheme://host:port，示例如 http://example.com:8081。当 allow_credentials 为 false 时，可以使用 * 来表示允许所有 Origin 通过 |
| allow_origin_patterns | array of string | 选填 | - | 允许跨域访问的 Origin 模式匹配， 用 * 匹配域名或者端口， <br/>比如 http://*.example.com -- 匹配域名， http://*.example.com:[8080,9090] -- 匹配域名和指定端口， http://*.example.com:[*] -- 匹配域名和所有端口。单独 * 表示匹配所有域名和端口 |
| allow_methods | array of string | 选填 | GET, PUT, POST, DELETE, PATCH, OPTIONS | 允许跨域访问的 Method，比如：GET，POST 等。可以使用 * 来表示允许所有 Method。 |
| allow_headers | array of string | 选填 | DNT,X-CustomHeader,Keep-Alive,User-Agent,X-Requested-With，<br/>If-Modified-Since,Cache-Control,Content-Type,Authorization | 允许跨域访问时请求方携带哪些非 CORS 规范以外的 Header。可以使用 * 来表示允许任意 Header。 |
| expose_headers | array of string | 选填 | - | 允许跨域访问时响应方携带哪些非 CORS 规范以外的 Header。可以使用 * 来表示允许任意 Header。 |
| allow_credentials | bool | 选填 | false | 是否允许跨域访问的请求方携带凭据（如 Cookie 等）。根据 CORS 规范，如果设置该选项为 true，在 allow_origins 不能使用 *， 替换成使用 allow_origin_patterns * |
| max_age | number | 选填 | 86400秒 | 浏览器缓存 CORS 结果的最大时间，单位为秒。<br/>在这个时间范围内，浏览器会复用上一次的检查结果 |
> 注意
> * allow_credentials 是一个很敏感的选项，请谨慎开启。开启之后，allow_credentials 和 allow_origins 为 * 不能同时使用，同时设置时， allow_origins 值为 "*" 生效。
> * allow_origins 和 allow_origin_patterns 可以同时设置， 先检查 allow_origins 是否匹配，然后再检查 allow_origin_patterns 是否匹配
> * 非法 CORS 请求， HTTP 状态码返回是 403， 返回体内容为 "Invalid CORS request"
# 配置示例
## 配置示例
## 允许所有跨域访问, 不允许请求方携带凭据
### 允许所有跨域访问, 不允许请求方携带凭据
```yaml
allow_origins:
- '*'
@@ -35,7 +46,7 @@ allow_credentials: false
max_age: 7200
```
## 允许所有跨域访问,同时允许请求方携带凭据
### 允许所有跨域访问,同时允许请求方携带凭据
```yaml
allow_origin_patterns:
- '*'
@@ -49,7 +60,7 @@ allow_credentials: true
max_age: 7200
```
## 允许特定子域,特定方法，特定请求头跨域访问，同时允许请求方携带凭据
### 允许特定子域,特定方法，特定请求头跨域访问，同时允许请求方携带凭据
```yaml
allow_origin_patterns:
- http://*.example.com
@@ -69,9 +80,9 @@ allow_credentials: true
max_age: 7200
```
# 测试
## 测试
## 测试配置
### 测试配置
```yaml
apiVersion: networking.higress.io/v1
@@ -143,9 +154,9 @@ spec:
imagePullPolicy: Always
```
## 请求测试
### 请求测试
### 简单请求
#### 简单请求
```shell
curl -v -H "Origin: http://httpbin2.example.org:9090" -H "Host: httpbin.example.com" http://127.0.0.1/anything/get\?foo\=1
@@ -156,7 +167,7 @@ curl -v -H "Origin: http://httpbin2.example.org:9090" -H "Host: httpbin.example
> access-control-allow-credentials: true
```
### 预检请求
#### 预检请求
```shell
curl -v -X OPTIONS -H "Origin: http://httpbin2.example.org:9090" -H "Host: httpbin.example.com" -H "Access-Control-Request-Method: POST" -H "Access-Control-Request-Headers: Content-Type, Token" http://127.0.0.1/anything/get\?foo\=1
@@ -177,7 +188,7 @@ curl -v -X OPTIONS -H "Origin: http://httpbin2.example.org:9090" -H "Host: http
* Closing connection 0
```
### 非法 CORS Origin 预检请求
#### 非法 CORS Origin 预检请求
```shell
curl -v -X OPTIONS -H "Origin: http://httpbin2.example.org" -H "Host: httpbin.example.com" -H "Access-Control-Request-Method: GET" http://127.0.0.1/anything/get\?foo\=1
@@ -193,7 +204,7 @@ curl -v -X OPTIONS -H "Origin: http://httpbin2.example.org" -H "Host: httpbin.e
Invalid CORS request
```
### 非法 CORS Method 预检请求
#### 非法 CORS Method 预检请求
```shell
curl -v -X OPTIONS -H "Origin: http://httpbin2.example.org:9090" -H "Host: httpbin.example.com" -H "Access-Control-Request-Method: DELETE" http://127.0.0.1/anything/get\?foo\=1
@@ -209,7 +220,7 @@ curl -v -X OPTIONS -H "Origin: http://httpbin2.example.org:9090" -H "Host: http
Invalid CORS request
```
### 非法 CORS Header 预检请求
#### 非法 CORS Header 预检请求
```shell
curl -v -X OPTIONS -H "Origin: http://httpbin2.example.org:9090" -H "Host: httpbin.example.com" -H "Access-Control-Request-Method: GET" -H "Access-Control-Request-Headers: TokenView" http://127.0.0.1/anything/get\?foo\=1
@@ -225,6 +236,6 @@ Invalid CORS request
Invalid CORS request
```
# 参考文档
## 参考文档
- https://www.ruanyifeng.com/blog/2016/04/cors.html
- https://fetch.spec.whatwg.org/#http-cors-protocol

223
plugins/wasm-go/extensions/cors/README_EN.md Normal file
View File

@@ -0,0 +1,223 @@
---
title: Cross-Origin Resource Sharing
keywords: [higress,cors]
description: Cross-Origin Resource Sharing plugin configuration reference
---
## Function Description
The `cors` plugin can enable CORS (Cross-Origin Resource Sharing) HTTP response headers for the server.
## Execution Attributes
Plugin execution phase: `Authorization Phase`
Plugin execution priority: `340`
## Configuration Fields
| Name | Data Type | Required | Default Value | Description |
|-----------------------|------------------|----------|-----------------------------------------------------------------------------------------------------------------------------|---------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------|
| allow_origins | array of string | Optional | * | Allowed Origins for cross-origin access, formatted as scheme://host:port, for example, http://example.com:8081. When allow_credentials is false, * can be used to allow all Origins through. |
| allow_origin_patterns | array of string | Optional | - | Patterns for matching allowed Origins for cross-origin access, using * to match domain or port, <br/>for example http://*.example.com -- matches domain, http://*.example.com:[8080,9090] -- matches domain and specified ports, http://*.example.com:[*] -- matches domain and all ports. A single * indicates matching all domains and ports. |
| allow_methods | array of string | Optional | GET, PUT, POST, DELETE, PATCH, OPTIONS | Allowed Methods for cross-origin access, for example: GET, POST, etc. * can be used to indicate all Methods are allowed. |
| allow_headers | array of string | Optional | DNT,X-CustomHeader,Keep-Alive,User-Agent,X-Requested-With,<br/>If-Modified-Since,Cache-Control,Content-Type,Authorization | Allowed Headers for the requester to carry that are not part of CORS specifications during cross-origin access. * can be used to indicate any Header is allowed. |
| expose_headers | array of string | Optional | - | Allowed Headers for the responder to carry that are not part of CORS specifications during cross-origin access. * can be used to indicate any Header is allowed. |
| allow_credentials | bool | Optional | false | Whether to allow the requester to carry credentials (e.g. Cookies) during cross-origin access. According to CORS specifications, if this option is set to true, * cannot be used for allow_origins, replace it with allow_origin_patterns. |
| max_age | number | Optional | 86400 seconds | Maximum time for browsers to cache CORS results, in seconds. <br/>Within this time frame, browsers will reuse the previous inspection results. |
> Note
> * allow_credentials is a very sensitive option, please enable it with caution. Once enabled, allow_credentials and allow_origins cannot both be *, if both are set, the allow_origins value of "*" takes effect.
> * allow_origins and allow_origin_patterns can be set simultaneously. First, check if allow_origins matches, then check if allow_origin_patterns matches.
> * Illegal CORS requests will return HTTP status code 403, with the response body content as "Invalid CORS request".
## Configuration Examples
### Allow all cross-origin access, without allowing the requester to carry credentials
```yaml
allow_origins:
- '*'
allow_methods:
- '*'
allow_headers:
- '*'
expose_headers:
- '*'
allow_credentials: false
max_age: 7200
```
### Allow all cross-origin access, while allowing the requester to carry credentials
```yaml
allow_origin_patterns:
- '*'
allow_methods:
- '*'
allow_headers:
- '*'
expose_headers:
- '*'
allow_credentials: true
max_age: 7200
```
### Allow specific subdomains, specific methods, and specific request headers for cross-origin access, while allowing the requester to carry credentials
```yaml
allow_origin_patterns:
- http://*.example.com
- http://*.example.org:[8080,9090]
allow_methods:
- GET
- PUT
- POST
- DELETE
allow_headers:
- Token
- Content-Type
- Authorization
expose_headers:
- '*'
allow_credentials: true
max_age: 7200
```
## Testing
### Test Configuration
```yaml
apiVersion: networking.higress.io/v1
kind: McpBridge
metadata:
name: mcp-cors-httpbin
namespace: higress-system
spec:
registries:
- domain: httpbin.org
name: httpbin
port: 80
type: dns
---
apiVersion: networking.k8s.io/v1
kind: Ingress
metadata:
annotations:
higress.io/destination: httpbin.dns
higress.io/upstream-vhost: "httpbin.org"
higress.io/backend-protocol: HTTP
name: ingress-cors-httpbin
namespace: higress-system
spec:
ingressClassName: higress
rules:
- host: httpbin.example.com
http:
paths:
- backend:
resource:
apiGroup: networking.higress.io
kind: McpBridge
name: mcp-cors-httpbin
path: /
pathType: Prefix
---
apiVersion: extensions.higress.io/v1alpha1
kind: WasmPlugin
metadata:
name: wasm-cors-httpbin
namespace: higress-system
spec:
defaultConfigDisable: true
matchRules:
- config:
allow_origins:
- http://httpbin.example.net
allow_origin_patterns:
- http://*.example.com:[*]
- http://*.example.org:[9090,8080]
allow_methods:
- GET
- POST
- PATCH
allow_headers:
- Content-Type
- Token
- Authorization
expose_headers:
- X-Custom-Header
- X-Env-UTM
allow_credentials: true
max_age: 3600
configDisable: false
ingress:
- ingress-cors-httpbin
url: oci://higress-registry.cn-hangzhou.cr.aliyuncs.com/plugins/cors:1.0.0
imagePullPolicy: Always
```
### Request Testing
#### Simple Request
```shell
curl -v -H "Origin: http://httpbin2.example.org:9090" -H "Host: httpbin.example.com" http://127.0.0.1/anything/get\?foo\=1
< HTTP/1.1 200 OK
> x-cors-version: 1.0.0
> access-control-allow-origin: http://httpbin2.example.org:9090
> access-control-expose-headers: X-Custom-Header,X-Env-UTM
> access-control-allow-credentials: true
```
#### Preflight Request
```shell
curl -v -X OPTIONS -H "Origin: http://httpbin2.example.org:9090" -H "Host: httpbin.example.com" -H "Access-Control-Request-Method: POST" -H "Access-Control-Request-Headers: Content-Type, Token" http://127.0.0.1/anything/get\?foo\=1
< HTTP/1.1 200 OK
< x-cors-version: 1.0.0
< access-control-allow-origin: http://httpbin2.example.org:9090
< access-control-allow-methods: GET,POST,PATCH
< access-control-allow-headers: Content-Type,Token,Authorization
< access-control-expose-headers: X-Custom-Header,X-Env-UTM
< access-control-allow-credentials: true
< access-control-max-age: 3600
< date: Tue, 23 May 2023 11:41:28 GMT
< server: istio-envoy
< content-length: 0
<
* Connection #0 to host 127.0.0.1 left intact
* Closing connection 0
```
#### Illegal CORS Origin Preflight Request
```shell
curl -v -X OPTIONS -H "Origin: http://httpbin2.example.org" -H "Host: httpbin.example.com" -H "Access-Control-Request-Method: GET" http://127.0.0.1/anything/get\?foo\=1
HTTP/1.1 403 Forbidden
< content-length: 70
< content-type: text/plain
< x-cors-version: 1.0.0
< date: Tue, 23 May 2023 11:27:01 GMT
< server: istio-envoy
<
* Connection #0 to host 127.0.0.1 left intact
Invalid CORS request
```
#### Illegal CORS Method Preflight Request
```shell
curl -v -X OPTIONS -H "Origin: http://httpbin2.example.org:9090" -H "Host: httpbin.example.com" -H "Access-Control-Request-Method: DELETE" http://127.0.0.1/anything/get\?foo\=1
< HTTP/1.1 403 Forbidden
< content-length: 49
< content-type: text/plain
< x-cors-version: 1.0.0
< date: Tue, 23 May 2023 11:28:51 GMT
< server: istio-envoy
<
* Connection #0 to host 127.0.0.1 left intact
Invalid CORS request
```
#### Illegal CORS Header Preflight Request
```shell
curl -v -X OPTIONS -H "Origin: http://httpbin2.example.org:9090" -H "Host: httpbin.example.com" -H "Access-Control-Request-Method: GET" -H "Access-Control-Request-Headers: TokenView" http://127.0.0.1/anything/get\?foo\=1
< HTTP/1.1 403 Forbidden
< content-length: 52
< content-type: text/plain
< x-cors-version: 1.0.0
< date: Tue, 23 May 2023 11:31:03 GMT
< server: istio-envoy
<
* Connection #0 to host 127.0.0.1 left intact
Invalid CORS request
```
## Reference Documents
- https://www.ruanyifeng.com/blog/2016/04/cors.html
- https://fetch.spec.whatwg.org/#http-cors-protocol

48
plugins/wasm-go/extensions/custom-response/README.md
View File

@@ -1,11 +1,19 @@
<p>
<a href="README_EN.md"> English </a> | 中文
</p>
---
title: 自定义应答
keywords: [higress,customn response]
description: 自定义应答插件配置参考
---
# 功能说明
## 功能说明
`custom-response`插件支持配置自定义的响应，包括自定义 HTTP 应答状态码、HTTP 应答头，以及 HTTP 应答 Body。可以用于 Mock 响应，也可以用于判断特定状态码后给出自定义应答，例如在触发网关限流策略时实现自定义响应。
# 配置字段
## 运行属性
插件执行阶段：`认证阶段`
插件执行优先级：`910`
## 配置字段
| 名称 | 数据类型 | 填写要求 | 默认值 | 描述 |
| -------- | -------- | -------- | -------- | -------- |
@@ -14,9 +22,9 @@
| body | string | 选填 | - | 自定义 HTTP 应答 Body |
| enable_on_status | array of number | 选填 | - | 匹配原始状态码，生成自定义响应，不填写时，不判断原始状态码 |
# 配置示例
## 配置示例
## Mock 应答场景
### Mock 应答场景
```yaml
status_code: 200
@@ -38,7 +46,7 @@ Content-Length: 17
{"hello":"world"}
```
## 触发限流时自定义响应
### 触发限流时自定义响应
```yaml
enable_on_status:
@@ -58,27 +66,3 @@ Location: https://example.com
从而实现基于浏览器 302 重定向机制，将限流后的用户引导到其他页面，比如可以是一个 CDN 上的静态页面。
如果希望触发限流时，正常返回其他应答，参考 Mock 应答场景配置相应的字段即可。
## 对特定路由或域名开启
```yaml
# 使用 matchRules 字段进行细粒度规则配置
matchRules:
# 规则一：按 Ingress 名称匹配生效
- ingress:
- default/foo
- default/bar
body: "{\"hello\":\"world\"}"
# 规则二：按域名匹配生效
- domain:
- "*.example.com"
- test.com
enable_on_status:
- 429
status_code: 200
headers:
- Content-Type=application/json
body: "{\"errmsg\": \"rate limited\"}"
```
此例 `ingress` 中指定的 `default/foo` 和 `default/bar` 对应 default 命名空间下名为 foo 和 bar 的 Ingress，当匹配到这两个 Ingress 时，将使用此段配置；
此例 `domain` 中指定的 `*.example.com` 和 `test.com` 用于匹配请求的域名，当发现域名匹配时，将使用此段配置；
配置的匹配生效顺序，将按照 `matchRules` 下规则的排列顺序，匹配第一个规则后生效对应配置，后续规则将被忽略。

74
plugins/wasm-go/extensions/custom-response/README_EN.md
View File

@@ -1,84 +1,54 @@
<p>
English | <a href="README.md">中文</a>
</p>
---
title: Custom Response
keywords: [higress, custom response]
description: Custom response plugin configuration reference
---
## Function Description
The `custom-response` plugin supports the configuration of custom responses, including custom HTTP response status codes, HTTP response headers, and HTTP response bodies. It can be used for Mock responses or for providing custom responses based on specific status codes, such as implementing custom responses when triggering the gateway rate-limiting policy.
# Description
`custom-response` plugin implements a function of sending custom responses, including custom HTTP response status codes, HTTP response headers and HTTP response body, which can be used in the scenarios of response mocking and sending a custom response for specific status codes, such as customizing the response for rate-limited requests.
## Running Attributes
Plugin Execution Phase: `Authentication Phase`
# Configuration Fields
Plugin Execution Priority: `910`
| Name | Type | Requirement | Default Value | Description |
## Configuration Fields
| Name | Data Type | Requirements | Default Value | Description |
| -------- | -------- | -------- | -------- | -------- |
| status_code | number | Optional | 200 | Custom HTTP response status code |
| headers | array of string | Optional | - | Custom HTTP response header. Key and value shall be separated using `=`. |
| headers | array of string | Optional | - | Custom HTTP response headers, keys and values separated by `=` |
| body | string | Optional | - | Custom HTTP response body |
| enable_on_status | array of number | Optional | - | The original response status code to match. Generate the custom response only the actual status code matches the configuration. Ignore the status code match if left unconfigured. |
# Configuration Samples
## Mock Responses
| enable_on_status | array of number | Optional | - | Match original status codes to generate custom responses; if not specified, the original status code is not checked |
## Configuration Example
### Mock Response Scenario
```yaml
status_code: 200
headers:
- Content-Type=application/json
- Hello=World
body: "{\"hello\":\"world\"}"
```
According to the configuration above, all the requests will get the following custom response:
With this configuration, the request will return the following custom response:
```text
HTTP/1.1 200 OK
Content-Type: application/json
Hello: World
Content-Length: 17
{"hello":"world"}
```
## Send a Custom Response when Rate-Limited
### Custom Response on Rate Limiting
```yaml
enable_on_status:
enable_on_status:
- 429
status_code: 302
headers:
- Location=https://example.com
```
When rate-limited, normally gateway will return a status code of `429` . Now, rate-limited requests will get the following custom response:
When the gateway rate limiting is triggered, it generally returns the `429` status code, and the request will return the following custom response:
```text
HTTP/1.1 302 Found
Location: https://example.com
```
This achieves the goal of redirecting users who have been rate-limited to another page based on the browser's 302 redirect mechanism, which could be a static page on a CDN.
So based on the 302 redirecting mechanism provided by browsers, this can redirect rate-limited users to other pages, for example, a static page hosted on CDN.
If you'd like to send other responses when rate-limited, please add other fields into the configuration, referring to the Mock Responses scenario.
## Only Enabled for Specific Routes or Domains
```yaml
# Use matchRules field for fine-grained rule configurations
matchRules:
# Rule 1: Match by Ingress name
- ingress:
- default/foo
- default/bar
body: "{\"hello\":\"world\"}"
# Rule 2: Match by domain
- domain:
- "*.example.com"
- test.com
enable_on_status:
- 429
status_code: 200
headers:
- Content-Type=application/json
body: "{\"errmsg\": \"rate limited\"}"
```
In the rule sample of `ingress`, `default/foo` and `default/bar` are the Ingresses named foo and bar in the default namespace. When the current Ingress names matches the configuration, the rule following shall be applied.
In the rule sample of `domain`, `*.example.com` and `test.com` are the domain names used for request matching. When the current domain name matches the configuration, the rule following shall be applied.
All rules shall be checked following the order of items in the `matchRules` field, The first matched rule will be applied. All remained will be ignored.
If you wish to return other responses normally when rate limiting is triggered, just refer to the Mock response scenario to configure the relevant fields accordingly.

190
plugins/wasm-go/extensions/de-graphql/README.md
View File

@@ -1,97 +1,17 @@
# DeGraphQL
---
title: DeGraphQL
keywords: [higress, DeGraphQL]
description: DeGraphQL 插件配置参考
---
## GraphQL
## 功能说明
### GraphQL 端点
`de-graphql`插件通过将URIs映射到GraphQL查询，从而可以将GraphQL上游转换为传统服务进行访问
REST API 有多个端点，GraphQL API 只有一个端点。
## 运行属性
```shell
https://api.github.com/graphql
```
### 与 GraphQL 通信
由于 GraphQL 操作由多行 JSON 组成，可以使用 curl 或任何其他采用 HTTP 的库。
在 REST 中，HTTP 谓词确定执行的操作。 在 GraphQL 中，执行查询要提供 JSON 请求体，因此 HTTP 谓词为 POST。 唯一的例外是内省查询，它是一种简单的 GET 到终结点查询。
### GraphQL POST 请求参数
标准的 GraphQL POST 请求情况如下：
- 添加 HTTP 请求头： Content-Type: application/json
- 使用 JSON 格式的请求体
- JSON 请求体包含三个字段
- query：查询文档，必填
- variables：变量，选填
- operationName：操作名称，选填，查询文档有多个操作时必填
```json
{
"query": "{viewer{name}}",
"operationName": "",
"variables": {
"name": "value"
}
}
```
### GraphQL 基本参数类型
- 基本参数类型包含： String, Int, Float, Boolean
- [类型]代表数组，例如：[Int]代表整型数组
- GraphQL 基本参数传递
- 小括号内定义形参，注意：参数需要定义类型
- !（叹号）代表参数不能为空
```shell
query ($owner : String!, $name : String!) {
repository(owner: $owner, name: $name) {
name
forkCount
description
}
}
```
### GitHub GraphQL 测试
使用 curl 命令查询 GraphQL， 用有效 JSON 请求体发出 POST 请求。 有效请求体必须包含一个名为 query 的字符串。
```shell
curl https://api.github.com/graphql -X POST \
-H "Authorization: bearer <PAT>" \
-d "{\"query\": \"query { viewer { login }}\"}"
{
"data": {
"viewer": {
"login": "2456868764"
}
}
}
```
```shell
curl 'https://api.github.com/graphql' -X POST \
-H 'Authorization: bearer <PAT>' \
-d '{"query":"query ($owner: String!, $name: String!) {\n repository(owner: $owner, name: $name) {\n name\n forkCount\n description\n }\n}\n","variables":{"owner":"2456868764","name":"higress"}}'
{
"data": {
"repository": {
"name": "higress",
"forkCount": 149,
"description": "Next-generation Cloud Native Gateway | 下一代云原生网关"
}
}
}
```
## DeGraphQL 插件
插件执行阶段：`认证阶段`
插件执行优先级：`430`
### 参数配置
@@ -182,6 +102,96 @@ curl "http://localhost/api?owner=alibaba&name=higress" -H "Authorization: Bearer
}
```
## GraphQL介绍
### GraphQL 端点
REST API 有多个端点，GraphQL API 只有一个端点。
```shell
https://api.github.com/graphql
```
### 与 GraphQL 通信
由于 GraphQL 操作由多行 JSON 组成，可以使用 curl 或任何其他采用 HTTP 的库。
在 REST 中，HTTP 谓词确定执行的操作。 在 GraphQL 中，执行查询要提供 JSON 请求体，因此 HTTP 谓词为 POST。 唯一的例外是内省查询，它是一种简单的 GET 到终结点查询。
### GraphQL POST 请求参数
标准的 GraphQL POST 请求情况如下：
- 添加 HTTP 请求头： Content-Type: application/json
- 使用 JSON 格式的请求体
- JSON 请求体包含三个字段
- query：查询文档，必填
- variables：变量，选填
- operationName：操作名称，选填，查询文档有多个操作时必填
```json
{
"query": "{viewer{name}}",
"operationName": "",
"variables": {
"name": "value"
}
}
```
### GraphQL 基本参数类型
- 基本参数类型包含： String, Int, Float, Boolean
- [类型]代表数组，例如：[Int]代表整型数组
- GraphQL 基本参数传递
- 小括号内定义形参，注意：参数需要定义类型
- !（叹号）代表参数不能为空
```shell
query ($owner : String!, $name : String!) {
repository(owner: $owner, name: $name) {
name
forkCount
description
}
}
```
### GitHub GraphQL 测试
使用 curl 命令查询 GraphQL， 用有效 JSON 请求体发出 POST 请求。 有效请求体必须包含一个名为 query 的字符串。
```shell
curl https://api.github.com/graphql -X POST \
-H "Authorization: bearer <PAT>" \
-d "{\"query\": \"query { viewer { login }}\"}"
{
"data": {
"viewer": {
"login": "2456868764"
}
}
}
```
```shell
curl 'https://api.github.com/graphql' -X POST \
-H 'Authorization: bearer <PAT>' \
-d '{"query":"query ($owner: String!, $name: String!) {\n repository(owner: $owner, name: $name) {\n name\n forkCount\n description\n }\n}\n","variables":{"owner":"2456868764","name":"higress"}}'
{
"data": {
"repository": {
"name": "higress",
"forkCount": 149,
"description": "Next-generation Cloud Native Gateway | 下一代云原生网关"
}
}
}
```
## 参考文档
- https://github.com/graphql/graphql-spec

199
plugins/wasm-go/extensions/de-graphql/README_EN.md Normal file
View File

@@ -0,0 +1,199 @@
---
title: DeGraphQL
keywords: [higress, DeGraphQL]
description: DeGraphQL 插件配置参考
---
## 功能说明
`de-graphql`插件通过将URIs映射到GraphQL查询，从而可以将GraphQL上游转换为传统服务进行访问
## 运行属性
插件执行阶段：`认证阶段`
插件执行优先级：`430`
### 参数配置
| 参数 | 描述 | 默认 |
|:----------------|:------------------------|:-----------|
| `gql` | graphql 查询 | 不能为空 |
| `endpoint` | graphql 查询端点 | `/graphql` |
| `timeout` | 查询连接超时，单位毫秒 | `5000` |
| `domain` | 服务域名，当服务来源是dns配置 | |
### 插件使用
https://github.com/alibaba/higress/issues/268
- 测试配置
```yaml
apiVersion: networking.higress.io/v1
kind: McpBridge
metadata:
name: default
namespace: higress-system
spec:
registries:
- domain: api.github.com
name: github
port: 443
type: dns
---
apiVersion: networking.k8s.io/v1
kind: Ingress
metadata:
annotations:
higress.io/destination: github.dns
higress.io/upstream-vhost: "api.github.com"
higress.io/backend-protocol: HTTPS
name: github-api
namespace: higress-system
spec:
ingressClassName: higress
rules:
- http:
paths:
- backend:
resource:
apiGroup: networking.higress.io
kind: McpBridge
name: default
path: /api
pathType: Prefix
---
apiVersion: extensions.higress.io/v1alpha1
kind: WasmPlugin
metadata:
name: de-graphql-github-api
namespace: higress-system
spec:
matchRules:
- ingress:
- github-api
config:
timeout: 5000
endpoint: /graphql
domain: api.github.com
gql: |
query ($owner:String! $name:String!){
repository(owner:$owner, name:$name) {
name
forkCount
description
}
}
url: oci://higress-registry.cn-hangzhou.cr.aliyuncs.com/plugins/de-graphql:1.0.0
```
- 测试结果
```shell
curl "http://localhost/api?owner=alibaba&name=higress" -H "Authorization: Bearer some-token"
{
"data": {
"repository": {
"description": "Next-generation Cloud Native Gateway",
"forkCount": 149,
"name": "higress"
}
}
}
```
## GraphQL介绍
### GraphQL 端点
REST API 有多个端点，GraphQL API 只有一个端点。
```shell
https://api.github.com/graphql
```
### 与 GraphQL 通信
由于 GraphQL 操作由多行 JSON 组成，可以使用 curl 或任何其他采用 HTTP 的库。
在 REST 中，HTTP 谓词确定执行的操作。 在 GraphQL 中，执行查询要提供 JSON 请求体，因此 HTTP 谓词为 POST。 唯一的例外是内省查询，它是一种简单的 GET 到终结点查询。
### GraphQL POST 请求参数
标准的 GraphQL POST 请求情况如下：
- 添加 HTTP 请求头： Content-Type: application/json
- 使用 JSON 格式的请求体
- JSON 请求体包含三个字段
- query：查询文档，必填
- variables：变量，选填
- operationName：操作名称，选填，查询文档有多个操作时必填
```json
{
"query": "{viewer{name}}",
"operationName": "",
"variables": {
"name": "value"
}
}
```
### GraphQL 基本参数类型
- 基本参数类型包含： String, Int, Float, Boolean
- [类型]代表数组，例如：[Int]代表整型数组
- GraphQL 基本参数传递
- 小括号内定义形参，注意：参数需要定义类型
- !（叹号）代表参数不能为空
```shell
query ($owner : String!, $name : String!) {
repository(owner: $owner, name: $name) {
name
forkCount
description
}
}
```
### GitHub GraphQL 测试
使用 curl 命令查询 GraphQL， 用有效 JSON 请求体发出 POST 请求。 有效请求体必须包含一个名为 query 的字符串。
```shell
curl https://api.github.com/graphql -X POST \
-H "Authorization: bearer <PAT>" \
-d "{\"query\": \"query { viewer { login }}\"}"
{
"data": {
"viewer": {
"login": "2456868764"
}
}
}
```
```shell
curl 'https://api.github.com/graphql' -X POST \
-H 'Authorization: bearer <PAT>' \
-d '{"query":"query ($owner: String!, $name: String!) {\n repository(owner: $owner, name: $name) {\n name\n forkCount\n description\n }\n}\n","variables":{"owner":"2456868764","name":"higress"}}'
{
"data": {
"repository": {
"name": "higress",
"forkCount": 149,
"description": "Next-generation Cloud Native Gateway | 下一代云原生网关"
}
}
}
```
## 参考文档
- https://github.com/graphql/graphql-spec
- https://docs.github.com/zh/graphql/guides/forming-calls-with-graphql
- https://github.com/altair-graphql/altair

6
plugins/wasm-go/extensions/ext-auth/README.md
View File

@@ -8,6 +8,10 @@ description: Ext 认证插件实现了调用外部授权服务进行认证鉴权
`ext-auth` 插件实现了向外部授权服务发送鉴权请求，以检查客户端请求是否得到授权。该插件实现时参考了Envoy原生的[ext_authz filter](https://www.envoyproxy.io/docs/envoy/latest/configuration/http/http_filters/ext_authz_filter)，实现了原生filter中对接HTTP服务的部分能力
## 运行属性
插件执行阶段：`认证阶段`
插件执行优先级：`360`
## 配置字段
@@ -281,4 +285,4 @@ Content-Length: 0
| x-forwarded-method | 原始请求的方法，比如get/post/delete/patch |
| x-forwarded-host | 原始请求的host |
| x-forwarded-uri | 原始请求的path，包含路径参数，比如/v1/app?test=true |
| x-forwarded-for | 原始请求的客户端IP地址 |
| x-forwarded-for | 原始请求的客户端IP地址 |

242
plugins/wasm-go/extensions/ext-auth/README_EN.md Normal file
View File

@@ -0,0 +1,242 @@
---
title: External Authentication
keywords: [higress, auth]
description: The Ext authentication plugin implements the functionality to call external authorization services for authentication and authorization.
---
## Function Description
The `ext-auth` plugin implements the ability to send authorization requests to external authorization services to check whether client requests are authorized. This plugin is based on the [ext_authz filter](https://www.envoyproxy.io/docs/envoy/latest/configuration/http/http_filters/ext_authz_filter) of Envoy, and implements part of the capability of connecting HTTP services found in the native filter.
## Runtime Attributes
Plugin Execution Phase: `Authentication Phase`
Plugin Execution Priority: `360`
## Configuration Fields
| Name | Data Type | Required | Default Value | Description |
| ------------------------------- | --------- | -------- | ------------- | --------------------------------------------------------------------------------------------------------------------------------------------------------------- |
| `http_service` | object | Yes | - | Configuration for the external authorization service |
| `failure_mode_allow` | bool | No | false | When set to true, client requests will still be accepted even if communication with the authorization service fails, or if the authorization service returns an HTTP 5xx error. |
| `failure_mode_allow_header_add` | bool | No | false | When both `failure_mode_allow` and `failure_mode_allow_header_add` are set to true, if communication with the authorization service fails or if the authorization service returns an HTTP 5xx error, the request header will add `x-envoy-auth-failure-mode-allowed: true`. |
| `status_on_error` | int | No | 403 | Sets the HTTP status code returned to the client when the authorization service is unreachable or returns a status code of 5xx. The default status code is `403`. |
Description of each configuration field under `http_service`
| Name | Data Type | Required | Default Value | Description |
| ------------------------ | --------- | -------- | ------------- | -------------------------------------------- |
| `endpoint_mode` | string | No | envoy | Choose one from `envoy`, `forward_auth`. |
| `endpoint` | object | Yes | - | HTTP service information for sending authorization requests. |
| `timeout` | int | No | 1000 | Timeout for connecting to `ext-auth` service, in milliseconds. |
| `authorization_request` | object | No | - | Configuration for sending authorization requests. |
| `authorization_response` | object | No | - | Configuration for processing authorization responses. |
Description of each configuration field under `endpoint`
| Name | Data Type | Required | Default Value | Description |
|------------|-----------|----------|---------------|----------------------------------------------------------------------------------------------|
| `service_name` | string | Required | - | The full FQDN name of the authorization service, e.g., `ext-auth.dns`, `ext-auth.my-ns.svc.cluster.local`. |
| `service_port` | int | No | 80 | The service port of the authorization service. |
| `path_prefix` | string | Required if `endpoint_mode` is `envoy` | | When `endpoint_mode` is `envoy`, this is the request path prefix sent by the client to the authorization service. |
| `request_method` | string | No | GET | When `endpoint_mode` is `forward_auth`, this is the HTTP Method sent by the client to the authorization service. |
| `path` | string | Required if `endpoint_mode` is `forward_auth` | - | When `endpoint_mode` is `forward_auth`, this is the request path sent by the client to the authorization service. |
Description of each configuration field under `authorization_request`
| Name | Data Type | Required | Default Value | Description |
|-------------------------|------------------------|----------|---------------|------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------|
| `allowed_headers` | array of StringMatcher | No | - | When set, client request headers that match will be added to the request to the authorization service. In addition to user-defined header matching rules, the `Authorization` HTTP header will automatically be included in the authorization service request. (If `endpoint_mode` is `forward_auth`, the original request's path will be set to `X-Original-Uri`, and the original request's HTTP Method will be set to `X-Original-Method`.) |
| `headers_to_add` | `map[string]string` | No | - | Sets the list of request headers to include in the request to the authorization service. Note that headers with the same name from the client request will be overwritten. |
| `with_request_body` | bool | No | false | Buffers the client request body and sends it in the authorization request (does not take effect for HTTP Methods GET, OPTIONS, HEAD). |
| `max_request_body_bytes`| int | No | 10MB | Sets the maximum size of the client request body to be stored in memory. When the request body reaches the value set in this field, an HTTP 413 status code will be returned, and the authorization process will not be initiated. Note that this setting takes precedence over the configuration of `failure_mode_allow`. |
Description of each configuration field under `authorization_response`
| Name | Data Type | Required | Default Value | Description |
|---------------------------|------------------------|----------|---------------|-----------------------------------------------------------------------------------------|
| `allowed_upstream_headers` | array of StringMatcher | No | - | When set, response headers from the authorization request that match will be added to the original client request headers. Note that headers with the same name will be overwritten. |
| `allowed_client_headers` | array of StringMatcher | No | - | If not set, when a request is denied, all response headers from the authorization request will be added to the client response. When set, in the case of a denied request, response headers from the authorization request that match will be added to the client response. |
Configuration of the `StringMatcher` type, when using `array of StringMatcher`, will be configured in the order defined in the array
| Name | Data Type | Required | Default Value | Description |
|------------|-----------|------------------------------------------------------------------|---------------|------------------|
| `exact` | string | No, choose one from `exact`, `prefix`, `suffix`, `contains`, `regex` | - | Exact match |
| `prefix` | string | No, choose one from `exact`, `prefix`, `suffix`, `contains`, `regex` | - | Prefix match |
| `suffix` | string | No, choose one from `exact`, `prefix`, `suffix`, `contains`, `regex` | - | Suffix match |
| `contains` | string | No, choose one from `exact`, `prefix`, `suffix`, `contains`, `regex` | - | Contains |
| `regex` | string | No, choose one from `exact`, `prefix`, `suffix`, `contains`, `regex` | - | Regex match |
## Configuration Example
Assuming the `ext-auth` service in Kubernetes has a serviceName of `ext-auth`, port `8090`, path of `/auth`, and namespace of `backend`, it supports two types of `endpoint_mode`:
- When `endpoint_mode` is `envoy`, the authorization request will use the original HTTP Method and the configured `path_prefix` as a prefix combined with the original request path.
- When `endpoint_mode` is `forward_auth`, the authorization request will use the configured `request_method` as the HTTP Method and the configured `path` as the request path.
### Example for endpoint_mode being envoy
#### Example 1
Configuration of the `ext-auth` plugin:
```yaml
http_service:
endpoint_mode: envoy
endpoint:
service_name: ext-auth.backend.svc.cluster.local
service_port: 8090
path_prefix: /auth
timeout: 1000
```
Using the following request through the gateway, when the `ext-auth` plugin is enabled:
```shell
curl -X POST http://localhost:8082/users?apikey=9a342114-ba8a-11ec-b1bf-00163e1250b5 -X GET -H "foo: bar" -H "Authorization: xxx"
```
**Request to `ext-auth` service successful:**
The `ext-auth` service will receive the following authorization request:
```shell
POST /auth/users?apikey=9a342114-ba8a-11ec-b1bf-00163e1250b5 HTTP/1.1
Host: ext-auth
Authorization: xxx
Content-Length: 0
```
**Request to `ext-auth` service failed:**
When calling the `ext-auth` service gets a 5xx response, the client will receive an HTTP response code of 403 along with all response headers returned by the `ext-auth` service.
If the `ext-auth` service returns response headers like `x-auth-version: 1.0` and `x-auth-failed: true`, these will be passed to the client:
```shell
HTTP/1.1 403 Forbidden
x-auth-version: 1.0
x-auth-failed: true
date: Tue, 16 Jul 2024 00:19:41 GMT
server: istio-envoy
content-length: 0
```
When `ext-auth` is unreachable or returns a status code of 5xx, the client request will be denied with the status code configured in `status_on_error`. If the `ext-auth` service returns other HTTP status codes, the client request will be denied with the returned status code. If `allowed_client_headers` is configured, response headers with corresponding matching items will be added to the client response.
#### Example 2
Configuration of the `ext-auth` plugin:
```yaml
http_service:
authorization_request:
allowed_headers:
- exact: x-auth-version
headers_to_add:
x-envoy-header: true
authorization_response:
allowed_upstream_headers:
- exact: x-user-id
- exact: x-auth-version
endpoint_mode: envoy
endpoint:
service_name: ext-auth.backend.svc.cluster.local
service_port: 8090
path_prefix: /auth
timeout: 1000
```
Using the following request through the gateway, when the `ext-auth` plugin is enabled:
```shell
curl -X POST http://localhost:8082/users?apikey=9a342114-ba8a-11ec-b1bf-00163e1250b5 -X GET -H "foo: bar" -H "Authorization: xxx"
```
The `ext-auth` service will receive the following authorization request:
```shell
POST /auth/users?apikey=9a342114-ba8a-11ec-b1bf-00163e1250b5 HTTP/1.1
Host: ext-auth
Authorization: xxx
X-Auth-Version: 1.0
x-envoy-header: true
Content-Length: 0
```
If the response headers from the `ext-auth` service contain `x-user-id` and `x-auth-version`, these two headers will be included in the upstream request when the gateway calls upstream.
### Example for endpoint_mode being forward_auth
#### Example 1
Configuration of the `ext-auth` plugin:
```yaml
http_service:
endpoint_mode: forward_auth
endpoint:
service_name: ext-auth.backend.svc.cluster.local
service_port: 8090
path: /auth
request_method: POST
timeout: 1000
```
Using the following request through the gateway, when the `ext-auth` plugin is enabled:
```shell
curl -i http://localhost:8082/users?apikey=9a342114-ba8a-11ec-b1bf-00163e1250b5 -X GET -H "foo: bar" -H "Authorization: xxx"
```
**Request to `ext-auth` service successful:**
The `ext-auth` service will receive the following authorization request:
```shell
POST /auth HTTP/1.1
Host: ext-auth
Authorization: xxx
X-Original-Uri: /users?apikey=9a342114-ba8a-11ec-b1bf-00163e1250b5
X-Original-Method: GET
Content-Length: 0
```
**Request to `ext-auth` service failed:**
When calling the `ext-auth` service gets a 5xx response, the client will receive an HTTP response code of 403 along with all response headers returned by the `ext-auth` service.
If the `ext-auth` service returns response headers like `x-auth-version: 1.0` and `x-auth-failed: true`, these will be passed to the client:
```shell
HTTP/1.1 403 Forbidden
x-auth-version: 1.0
x-auth-failed: true
date: Tue, 16 Jul 2024 00:19:41 GMT
server: istio-envoy
content-length: 0
```
When `ext-auth` is unreachable or returns a status code of 5xx, the client request will be denied with the status code configured in `status_on_error`. If the `ext-auth` service returns other HTTP status codes, the client request will be denied with the returned status code. If `allowed_client_headers` is configured, response headers with corresponding matching items will be added to the client response.
#### Example 2
Configuration of the `ext-auth` plugin:
```yaml
http_service:
authorization_request:
allowed_headers:
- exact: x-auth-version
headers_to_add:
x-envoy-header: true
authorization_response:
allowed_upstream_headers:
- exact: x-user-id
- exact: x-auth-version
endpoint_mode: forward_auth
endpoint:
service_name: ext-auth.backend.svc.cluster.local
service_port: 8090
path: /auth
request_method: POST
timeout: 1000
```
Using the following request through the gateway, when the `ext-auth` plugin is enabled:
```shell
curl -i http://localhost:8082/users?apikey=9a342114-ba8a-11ec-b1bf-00163e1250b5 -X GET -H "foo: bar" -H "Authorization: xxx" -H "X-Auth-Version: 1.0"
```
The `ext-auth` service will receive the following authorization request:
```shell
POST /auth HTTP/1.1
Host: ext-auth
Authorization: xxx
X-Original-Uri: /users?apikey=9a342114-ba8a-11ec-b1bf-00163e1250b5
X-Original-Method: GET
X-Auth-Version: 1.0
x-envoy-header: true
Content-Length: 0
```
If the response headers from the `ext-auth` service contain `x-user-id` and `x-auth-version`, these two headers will be included in the upstream request when the gateway calls upstream.
#### x-forwarded-* header
When `endpoint_mode` is `forward_auth`, higress will automatically generate and send the following headers to the authorization service.
| Header | Description |
|--------------------|------------------------------------------|
| x-forwarded-proto | The original request scheme, e.g., http/https |
| x-forwarded-method | The original request method, e.g., get/post/delete/patch |
| x-forwarded-host | The original request host |
| x-forwarded-uri | The original request path, including path parameters, e.g., /v1/app?test=true |
| x-forwarded-for | The original request client IP address |

13
plugins/wasm-go/extensions/frontend-gray/README.md
View File

@@ -1,7 +1,18 @@
# frontend-gray 前端灰度插件
---
title: 前端灰度
keywords: [higress,frontend gray]
description: 前端灰度插件配置参考
---
## 功能说明
`frontend-gray`插件实现了前端用户灰度的的功能，通过此插件，不但可以用于业务`A/B实验`，同时通过`可灰度`配合`可监控`,`可回滚`策略保证系统发布运维的稳定性。
## 运行属性
插件执行阶段：`认证阶段`
插件执行优先级：`450`
## 配置字段
| 名称 | 数据类型 | 填写要求 | 默认值 | 描述 |
|----------------|--------------|----|-----|----------------------------------------------------------------------------------------------------|

162
plugins/wasm-go/extensions/frontend-gray/README_EN.md Normal file
View File

@@ -0,0 +1,162 @@
---
title: Frontend Gray
keywords: [higress, frontend gray]
description: Frontend gray plugin configuration reference
---
## Function Description
The `frontend-gray` plugin implements the functionality of user gray release on the frontend. Through this plugin, it can be used for business `A/B testing`, while the `gradual release` combined with `monitorable` and `rollback` strategies ensures the stability of system release operations.
## Runtime Attributes
Plugin execution phase: `Authentication Phase`
Plugin execution priority: `450`
## Configuration Fields
| Name | Data Type | Requirements | Default Value | Description |
|-----------------|-------------------|---------------|---------------|-------------------------------------------------------------------------------------------------------------|
| `grayKey` | string | Optional | - | The unique identifier of the user ID, which can be from Cookie or Header, such as userid. If not provided, uses `rules[].grayTagKey` and `rules[].grayTagValue` to filter gray release rules. |
| `graySubKey` | string | Optional | - | User identity information may be output in JSON format, for example: `userInfo:{ userCode:"001" }`, in the current example, `graySubKey` is `userCode`. |
| `rules` | array of object | Required | - | User-defined different gray release rules, adapted to different gray release scenarios. |
| `rewrite` | object | Required | - | Rewrite configuration, generally used for OSS/CDN frontend deployment rewrite configurations. |
| `baseDeployment`| object | Optional | - | Configuration of the Base baseline rules. |
| `grayDeployments` | array of object | Optional | - | Configuration of the effective rules for gray release, as well as the effective versions. |
`rules` field configuration description:
| Name | Data Type | Requirements | Default Value | Description |
|------------------|-------------------|---------------|---------------|--------------------------------------------------------------------------------------------|
| `name` | string | Required | - | Unique identifier for the rule name, associated with `deploy.gray[].name` for effectiveness. |
| `grayKeyValue` | array of string | Optional | - | Whitelist of user IDs. |
| `grayTagKey` | string | Optional | - | Label key for user classification tagging, derived from Cookie. |
| `grayTagValue` | array of string | Optional | - | Label value for user classification tagging, derived from Cookie. |
`rewrite` field configuration description:
> `indexRouting` homepage rewrite and `fileRouting` file rewrite essentially use prefix matching, for example, `/app1`: `/mfe/app1/{version}/index.html` represents requests with the prefix /app1 routed to `/mfe/app1/{version}/index.html` page, where `{version}` represents the version number, which will be dynamically replaced by `baseDeployment.version` or `grayDeployments[].version` during execution.
> `{version}` will be replaced dynamically during execution by the frontend version from `baseDeployment.version` or `grayDeployments[].version`.
| Name | Data Type | Requirements | Default Value | Description |
|------------------|-------------------|---------------|---------------|---------------------------------------|
| `host` | string | Optional | - | Host address, if OSS set to the VPC internal access address. |
| `notFoundUri` | string | Optional | - | 404 page configuration. |
| `indexRouting` | map of string to string | Optional | - | Defines the homepage rewrite routing rules. Each key represents the homepage routing path, and the value points to the redirect target file. For example, the key `/app1` corresponds to the value `/mfe/app1/{version}/index.html`. If the effective version is `0.0.1`, the access path is `/app1`, it redirects to `/mfe/app1/0.0.1/index.html`. |
| `fileRouting` | map of string to string | Optional | - | Defines resource file rewrite routing rules. Each key represents the resource access path, and the value points to the redirect target file. For example, the key `/app1/` corresponds to the value `/mfe/app1/{version}`. If the effective version is `0.0.1`, the access path is `/app1/js/a.js`, it redirects to `/mfe/app1/0.0.1/js/a.js`. |
`baseDeployment` field configuration description:
| Name | Data Type | Requirements | Default Value | Description |
|------------------|-------------------|---------------|---------------|-------------------------------------------------------------------------------------------|
| `version` | string | Required | - | The version number of the Base version, as a fallback version. |
`grayDeployments` field configuration description:
| Name | Data Type | Requirements | Default Value | Description |
|------------------|-------------------|---------------|---------------|----------------------------------------------------------------------------------------------|
| `version` | string | Required | - | Version number of the Gray version, if the gray rules are hit, this version will be used. If it is a non-CDN deployment, add `x-higress-tag` to the header. |
| `backendVersion` | string | Required | - | Gray version for the backend, which will add `x-mse-tag` to the header of `XHR/Fetch` requests. |
| `name` | string | Required | - | Rule name associated with `rules[].name`. |
| `enabled` | boolean | Required | - | Whether to activate the current gray release rule. |
## Configuration Example
### Basic Configuration
```yml
grayKey: userid
rules:
- name: inner-user
grayKeyValue:
- '00000001'
- '00000005'
- name: beta-user
grayKeyValue:
- '00000002'
- '00000003'
grayTagKey: level
grayTagValue:
- level3
- level5
baseDeployment:
version: base
grayDeployments:
- name: beta-user
version: gray
enabled: true
```
The unique identifier of the user in the cookie is `userid`, and the current gray release rule has configured the `beta-user` rule.
When the following conditions are met, the version `version: gray` will be used:
- `userid` in the cookie equals `00000002` or `00000003`
- Users whose `level` in the cookie equals `level3` or `level5`
Otherwise, use version `version: base`.
### User Information Exists in JSON
```yml
grayKey: appInfo
graySubKey: userId
rules:
- name: inner-user
grayKeyValue:
- '00000001'
- '00000005'
- name: beta-user
grayKeyValue:
- '00000002'
- '00000003'
grayTagKey: level
grayTagValue:
- level3
- level5
baseDeployment:
version: base
grayDeployments:
- name: beta-user
version: gray
enabled: true
```
The cookie contains JSON data for `appInfo`, which includes the field `userId` as the current unique identifier.
The current gray release rule has configured the `beta-user` rule.
When the following conditions are met, the version `version: gray` will be used:
- `userid` in the cookie equals `00000002` or `00000003`
- Users whose `level` in the cookie equals `level3` or `level5`
Otherwise, use version `version: base`.
### Rewrite Configuration
> Generally used in CDN deployment scenarios.
```yml
grayKey: userid
rules:
- name: inner-user
grayKeyValue:
- '00000001'
- '00000005'
- name: beta-user
grayKeyValue:
- '00000002'
- '00000003'
grayTagKey: level
grayTagValue:
- level3
- level5
rewrite:
host: frontend-gray.oss-cn-shanghai-internal.aliyuncs.com
notFoundUri: /mfe/app1/dev/404.html
indexRouting:
/app1: '/mfe/app1/{version}/index.html'
/: '/mfe/app1/{version}/index.html',
fileRouting:
/: '/mfe/app1/{version}'
/app1/: '/mfe/app1/{version}'
baseDeployment:
version: base
grayDeployments:
- name: beta-user
version: gray
enabled: true
```
`{version}` will be dynamically replaced with the actual version during execution.
#### indexRouting: Homepage Route Configuration
Accessing `/app1`, `/app123`, `/app1/index.html`, `/app1/xxx`, `/xxxx` will route to '/mfe/app1/{version}/index.html'.
#### fileRouting: File Route Configuration
The following file mappings are effective:
- `/js/a.js` => `/mfe/app1/v1.0.0/js/a.js`
- `/js/template/a.js` => `/mfe/app1/v1.0.0/js/template/a.js`
- `/app1/js/a.js` => `/mfe/app1/v1.0.0/js/a.js`
- `/app1/js/template/a.js` => `/mfe/app1/v1.0.0/js/template/a.js`

30
plugins/wasm-go/extensions/geo-ip/README.md
View File

@@ -1,8 +1,20 @@
# 功能说明
---
title: IP 地理位置
keywords: [higress,geo ip]
description: IP 地理位置插件配置参考
---
## 功能说明
`geo-ip`本插件实现了通过用户ip查询出地理位置信息，然后通过请求属性和新添加的请求头把地理位置信息传递给后续插件。
# 配置字段
## 运行属性
插件执行阶段：`认证阶段`
插件执行优先级：`440`
## 配置字段
| 名称 | 数据类型 | 填写要求 | 默认值 | 描述 |
| -------- | -------- | -------- | -------- | -------- |
| ip_protocol | string | 否 | ipv4 | 可选值：1. ipv4：只对ipv4用户请求查找地理位置信息，传递给后续插件。而ipv6用户的请求会跳过该插件，继续由后续插件处理。 2. ipv6：(未来实现后)只对ipv6用户查找地理位置信息，传递给后续插件。而ipv4用户的请求会跳过该插件，继续由后续插件处理。（目前是跳过插件，请求由后续插件处理。）
@@ -10,7 +22,7 @@
| ip_header_name | string | 否 | x-forwarded-for | 当`ip_source_type`为`header`时，指定自定义IP来源头 |
# 配置示例
## 配置示例
```yaml
ip_protocol: ipv4
@@ -18,7 +30,7 @@ ip_source_type: header
ip_header_name: X-Real-Ip
```
# 生成geoCidr.txt的说明
## 生成geoCidr.txt的说明
在generateCidr目录里包含的ip.merge.txt文件是github上ip2region项目的全世界的ip网段库。 ipRange2Cidr.go 是把ip网段转换成多个cidr的程序。转换出的cidr 和地理位置信息存在 /data/geoCidr.txt文件里。geo-ip插件会在Higress启动读配置阶段读取geoCidr.txt文件并且解析到radixtree数据结构的内存里，以便以后查询用户ip对应的地理位置信息。转换程序运行命令如下：
@@ -26,13 +38,5 @@ ip_header_name: X-Real-Ip
go run generateCidr/ipRange2Cidr.go
```
# property 的使用方式
## property 的使用方式
在geo-ip插件里调用proxywasm.SetProperty() 分别把country、city、province、isp设置进请求属性里，以便后续插件可以调用proxywasm.GetProperty()获取该请求的用户ip对应的地理信息。
# ip网段转换成cidr列表的单元测试
在 generateCidr 目录里的 ipRange2Cidr_test.go 是ip网段转换成cidr 列表的单元测试程序。在 generateCidr 目录里运行命令 go test 。通过的情况显示如下：
``bash
PASS
ok higress/plugins/wasm-go/extensions/geo-ip/generateCidr 0.018s
```

34
plugins/wasm-go/extensions/geo-ip/README_EN.md Normal file
View File

@@ -0,0 +1,34 @@
---
title: IP Geolocation
keywords: [higress,geo ip]
description: IP Geolocation Plugin Configuration Reference
---
## Function Description
The `geo-ip` plugin allows querying geographical location information based on the user's IP address, and then passes this geographical information to subsequent plugins through request attributes and newly added request headers.
## Runtime Properties
Plugin Execution Phase: `Authentication Phase`
Plugin Execution Priority: `440`
## Configuration Fields
| Name | Data Type | Requirement | Default Value | Description |
| -------- | ----------- | ----------- | ------------------ | ------------ |
| ip_protocol | string | No | ipv4 | Optional values: 1. ipv4: Only queries geographical location information for ipv4 user requests, passing it to subsequent plugins. Requests from ipv6 users will skip this plugin and be processed by later plugins. 2. ipv6: (To be implemented in the future) Only queries geographical location information for ipv6 users, passing it to subsequent plugins. Requests from ipv4 users will skip this plugin and be processed by later plugins. (Currently skips the plugin; requests are handled by subsequent plugins.) |
| ip_source_type | string | No | origin-source | Optional values: 1. Peer socket IP: `origin-source`; 2. Retrieved via header: `header` |
| ip_header_name | string | No | x-forwarded-for | When `ip_source_type` is `header`, specify the custom IP source header. |
## Configuration Example
```yaml
ip_protocol: ipv4
ip_source_type: header
ip_header_name: X-Real-Ip
```
## Explanation for Generating geoCidr.txt
The ip.merge.txt file included in the generateCidr directory is the global IP segment database from the ip2region project on GitHub. The ipRange2Cidr.go program converts IP segments into multiple CIDRs. The converted CIDRs and geographical location information are stored in the /data/geoCidr.txt file. The geo-ip plugin will read the geoCidr.txt file during the configuration stage when Higress starts and parse it into the radixtree data structure in memory for future queries of geographical location information corresponding to user IP addresses. The command to run the conversion program is as follows:
```bash
go run generateCidr/ipRange2Cidr.go
```
## Usage of Properties
In the geo-ip plugin, call proxywasm.SetProperty() to set country, city, province, and isp into request attributes so that subsequent plugins can use proxywasm.GetProperty() to obtain the geographical information corresponding to the user's IP for that request.

16
plugins/wasm-go/extensions/ip-restriction/README.md
View File

@@ -1,9 +1,21 @@
# 功能说明
---
title: IP 访问限制
keywords: [higress, ip restriction]
description: IP 访问限制插件配置参考
---
## 功能说明
`ip-restriction `插件可以通过将 IP 地址列入白名单或黑名单来限制对服务或路由的访问.支持对单个 IP 地址、多个 IP 地址和类似
10.10.10.0/24 的 CIDR范围的限制.
# 配置说明
## 运行属性
插件执行阶段：`认证阶段`
插件执行优先级：`210`
## 配置说明
| 配置项 | 类型 | 必填 | 默认值 | 说明 |
|----------------|--------|----|-----------------------------|------------------------------------------|

37
plugins/wasm-go/extensions/ip-restriction/README_EN.md Normal file
View File

@@ -0,0 +1,37 @@
---
title: IP Access Restriction
keywords: [higress, ip restriction]
description: IP access restriction plugin configuration reference
---
## Function Description
The `ip-restriction` plugin can restrict access to services or routes by whitelisting or blacklisting IP addresses. It supports restrictions on a single IP address, multiple IP addresses, and CIDR ranges like 10.10.10.0/24.
## Running Attributes
Plugin execution phase: `Authentication Phase`
Plugin execution priority: `210`
## Configuration Description
| Configuration Item | Type | Required | Default Value | Description |
|---------------------|---------|----------|---------------------------------|---------------------------------------------|
| ip_source_type | string | No | origin-source | Optional values: 1. Peer socket IP: `origin-source`; 2. Get from header: `header` |
| ip_header_name | string | No | x-forwarded-for | When `ip_source_type` is `header`, specify the custom IP source header |
| allow | array | No | [] | Whitelist |
| deny | array | No | [] | Blacklist |
| status | int | No | 403 | HTTP status code when access is denied |
| message | string | No | Your IP address is blocked. | Return message when access is denied |
```yaml
ip_source_type: origin-source
allow:
- 10.0.0.1
- 192.168.0.0/16
```
```yaml
ip_source_type: header
ip_header_name: x-real-iP
deny:
- 10.0.0.1
- 192.169.0.0/16
```

96
plugins/wasm-go/extensions/key-auth/README.md
View File

@@ -1,15 +1,32 @@
# 功能说明
---
title: Key 认证
keywords: [higress,key auth]
description: Key 认证插件配置参考
---
## 功能说明
`key-auth`插件实现了基于 API Key 进行认证鉴权的功能，支持从 HTTP 请求的 URL 参数或者请求头解析 API Key，同时验证该 API Key 是否有权限访问。
# 配置字段
## 运行属性
| 名称 | 数据类型 | 填写要求 | 默认值 | 描述 |
| ----------- | --------------- | ------------------------------------------- | ------ | ----------------------------------------------------------- |
| `global_auth` | bool | 选填 | - | 若配置为true，则全局生效认证机制; 若配置为false，则只对做了配置的域名和路由生效认证机制; 若不配置则仅当没有域名和路由配置时全局生效（兼容机制） |
| `consumers` | array of object | 必填 | - | 配置服务的调用者，用于对请求进行认证 |
| `keys` | array of string | 必填 | - | API Key 的来源字段名称，可以是 URL 参数或者 HTTP 请求头名称 |
| `in_query` | bool | `in_query` 和 `in_header` 至少有一个为 true | true | 配置 true 时，网关会尝试从 URL 参数中解析 API Key |
| `in_header` | bool | `in_query` 和 `in_header` 至少有一个为 true | true | 配置 true 时，网关会尝试从 HTTP 请求头中解析 API Key |
插件执行阶段：`认证阶段`
插件执行优先级：`310`
## 配置字段
**注意：**
- 在一个规则里，鉴权配置和认证配置不可同时存在
- 对于通过认证鉴权的请求，请求的header会被添加一个`X-Mse-Consumer`字段，用以标识调用者的名称。
### 认证配置
| 名称 | 数据类型 | 填写要求 | 默认值 | 描述 |
| ----------- | --------------- | ------------------------------------------- | ------ | ----------------------------------------------------------- |
| `global_auth` | bool | 选填（**仅实例级别配置**） | - | 只能在实例级别配置，若配置为true，则全局生效认证机制; 若配置为false，则只对做了配置的域名和路由生效认证机制，若不配置则仅当没有域名和路由配置时全局生效（兼容老用户使用习惯）。 |
| `consumers` | array of object | 必填 | - | 配置服务的调用者，用于对请求进行认证 |
| `keys` | array of string | 必填 | - | API Key 的来源字段名称，可以是 URL 参数或者 HTTP 请求头名称 |
| `in_query` | bool | `in_query` 和 `in_header` 至少有一个为 true | true | 配置 true 时，网关会尝试从 URL 参数中解析 API Key |
| `in_header` | bool | `in_query` 和 `in_header` 至少有一个为 true | true | 配置 true 时，网关会尝试从 HTTP 请求头中解析 API Key |
`consumers`中每一项的配置字段说明如下：
@@ -18,18 +35,22 @@
| `credential` | string | 必填 | - | 配置该consumer的访问凭证 |
| `name` | string | 必填 | - | 配置该consumer的名称 |
### 鉴权配置（非必需）
**注意：**
- 对于通过认证鉴权的请求，请求的header会被添加一个`X-Mse-Consumer`字段，用以标识调用者的名称。
| 名称 | 数据类型 | 填写要求 | 默认值 | 描述 |
| ----------- | --------------- | ------------------------------------------- | ------ | ----------------------------------------------------------- |
| `allow` | array of string | 选填(**非实例级别配置**) | - | 只能在路由或域名等细粒度规则上配置，对于符合匹配条件的请求，配置允许访问的 consumer，从而实现细粒度的权限控制 |
# 配置示例
## 配置示例
## 对特定路由或域名开启
### 全局配置认证和路由粒度进行鉴权
以下配置将对网关特定路由或域名开启 Key Auth 认证和鉴权，注意`credential`字段不能重复
以下配置将对网关特定路由或域名开启Key Auth认证和鉴权。credential字段不能重复。
在实例级别做如下插件配置：
```yaml
global_auth: true
global_auth: false
consumers:
- credential: 2bda943c-ba2b-11ec-ba07-00163e1250b5
name: consumer1
@@ -40,8 +61,6 @@ keys:
- x-api-key
```
**路由级配置**
对 route-a 和 route-b 这两个路由做如下配置：
```yaml
@@ -56,10 +75,16 @@ allow:
- consumer2
```
### 根据该配置，下列请求可以允许访问：
**说明：**
此例指定的route-a和route-b即在创建网关路由时填写的路由名称，当匹配到这两个路由时，将允许name为consumer1的调用者访问，其他调用者不允许访问。
此例指定的*.example.com和test.com用于匹配请求的域名，当发现域名匹配时，将允许name为consumer2的调用者访问，其他调用者不被允许访问。
根据该配置，下列请求可以允许访问：
假设以下请求会匹配到route-a这条路由
n
**将 API Key 设置在 url 参数中**
```bash
curl http://xxx.hello.com/test?apikey=2bda943c-ba2b-11ec-ba07-00163e1250b5
@@ -71,7 +96,7 @@ curl http://xxx.hello.com/test -H 'x-api-key: 2bda943c-ba2b-11ec-ba07-00163e125
认证鉴权通过后，请求的header中会被添加一个`X-Mse-Consumer`字段，在此例中其值为`consumer1`，用以标识调用方的名称
### 下列请求将拒绝访问：
下列请求将拒绝访问：
**请求未提供 API Key，返回401**
```bash
@@ -88,31 +113,24 @@ curl http://xxx.hello.com/test?apikey=926d90ac-ba2e-11ec-ab68-00163e1250b5
curl http://xxx.hello.com/test?apikey=c8c8e9ca-558e-4a2d-bb62-e700dcc40e35
```
## 网关实例级别开启
### 网关实例级别开启
以下配置未指定`matchRules`字段，因此将对网关实例级别开启全局 Key Auth 认证.
以下配置将对网关实例级别开启 Basic Auth 认证，所有请求均需要经过认证后才能访问。
```yaml
defaultConfig
consumers:
- credential: 2bda943c-ba2b-11ec-ba07-00163e1250b5
name: consumer1
- credential: c8c8e9ca-558e-4a2d-bb62-e700dcc40e35
name: consumer2
keys:
- apikey
in_query: true
global_auth: true
consumers:
- credential: 2bda943c-ba2b-11ec-ba07-00163e1250b5
name: consumer1
- credential: c8c8e9ca-558e-4a2d-bb62-e700dcc40e35
name: consumer2
keys:
- apikey
- x-api-key
```
开启`matchRules`方式如下：
```yaml
matchRules:
- config:
allow:
- consumer1
```
# 相关错误码
## 相关错误码
| HTTP 状态码 | 出错信息 | 原因说明 |
| ----------- | --------------------------------------------------------- | ----------------------- |

192
plugins/wasm-go/extensions/key-auth/README_EN.md
View File

@@ -1,31 +1,108 @@
# Features
The `key-auth` plug-in implements the authentication function based on the API Key, supports parsing the API Key from the URL parameter or request header of the HTTP request, and verifies whether the API Key has permission to access.
---
title: Key Authentication
keywords: [higress,key auth]
description: Key Authentication Plugin Configuration Reference
---
## Function Description
The `key-auth` plugin implements authentication based on API Key, supporting the parsing of the API Key from HTTP request URL parameters or request headers, while also verifying whether the API Key has permission to access the resource.
# Configuration field
## Runtime Properties
Plugin Execution Phase: `Authentication Phase`
Plugin Execution Priority: `310`
| Name | Data Type | Parameter requirements | Default| Description |
| ----------- | --------------- | -------------------------------------------------------- | ------ | --------------------------------------------------------------------------------------------------------- |
| `global_auth` | bool | Optional | - | If configured to true, the authentication mechanism will take effect globally; if configured to false, the authentication mechanism will only take effect for the configured domain names and routes; if not configured, the authentication mechanism will only take effect globally when no domain names and routes are configured (compatibility mechanism) |
| `consumers` | array of object | Required | - | Configure the caller of the service to authenticate the request. |
| `keys` | array of string | Required | - | The name of the source field of the API Key, which can be a URL parameter or an HTTP request header name. |
| `in_query` | bool | At least one of `in_query` and `in_header` must be true. | true | When configured true, the gateway will try to parse the API Key from the URL parameters. |
| `in_header` | bool | The same as above. | true | The same as above. |
## Configuration Fields
**Note:**
- Authentication and authorization configurations cannot coexist within a single rule.
- For requests that are authenticated, a header field `X-Mse-Consumer` will be added to identify the caller's name.
The configuration fields of each item in `consumers` are described as follows:
### Authentication Configuration
| Name | Data Type | Requirements | Default Value | Description |
| ------------- | ---------------- | ----------------------------------------------- | ------------- | ---------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------- |
| `global_auth` | bool | Optional (**Instance-Level Configuration Only**) | - | Can only be configured at the instance level; if set to true, the authentication mechanism takes effect globally; if set to false, it only applies to the configured hostnames and routes. If not configured, it will only take effect globally when no hostname and route configurations are present (to maintain compatibility with older user habits). |
| `consumers` | array of object | Required | - | Configures the service callers for request authentication. |
| `keys` | array of string | Required | - | Source field names for the API Key, which can be URL parameters or HTTP request header names. |
| `in_query` | bool | At least one of `in_query` and `in_header` must be true | true | When configured as true, the gateway will attempt to parse the API Key from URL parameters. |
| `in_header` | bool | At least one of `in_query` and `in_header` must be true | true | When configured as true, the gateway will attempt to parse the API Key from HTTP request headers. |
| Name | Data Type | Parameter requirements | Default | Description |
| ------------ | --------- | -----------------------| ------ | ------------------------------------------- |
| `credential` | string | Required | - | Configure the consumer's access credentials. |
| `name` | string | Required | - | Configure the name of the consumer. |
The configuration field descriptions for each item in `consumers` are as follows:
| Name | Data Type | Requirements | Default Value | Description |
| ------------ | --------- | ------------ | ------------- | ------------------------------ |
| `credential` | string | Required | - | Configures the access credential for this consumer. |
| `name` | string | Required | - | Configures the name for this consumer. |
**Warning：**
- For a request that passes authentication, an `X-Mse-Consumer` field will be added to the request header to identify the name of the caller.
### Authorization Configuration (Optional)
| Name | Data Type | Requirements | Default Value | Description |
| ----------- | ---------------- | ----------------------------------------------- | ------------- | --------------------------------------------------------------------------------------------------------------------------------------------------------------------- |
| `allow` | array of string | Optional (**Non-Instance Level Configuration**) | - | Can only be configured on fine-grained rules such as routes or hostnames; specifies the allowed consumers for matching requests, allowing for fine-grained permission control. |
# Example configuration
## Configuration Example
### Global Configuration for Authentication and Granular Route Authorization
The following configuration will enable Key Auth authentication and authorization for specific routes or hostnames in the gateway. The `credential` field must not repeat.
## Enabled for specific routes or domains
At the instance level, do the following plugin configuration:
```yaml
global_auth: false
consumers:
- credential: 2bda943c-ba2b-11ec-ba07-00163e1250b5
name: consumer1
- credential: c8c8e9ca-558e-4a2d-bb62-e700dcc40e35
name: consumer2
keys:
- apikey
- x-api-key
```
The following configuration will enable Key Auth authentication and authentication for gateway-specific routes or domain names. Note that the `credential` field can not be repeated.
For routes route-a and route-b, do the following configuration:
```yaml
allow:
- consumer1
```
For the hostnames *.example.com and test.com, do the following configuration:
```yaml
allow:
- consumer2
```
**Note:**
The routes route-a and route-b specified in this example refer to the route names filled in when creating the gateway routes. When matched with these two routes, requests from the caller named consumer1 will be allowed while others will be denied.
The specified hostnames *.example.com and test.com are used to match the request's domain name. When a domain name is matched, callers named consumer2 will be allowed while others will be denied.
Based on this configuration, the following requests will be allowed:
Assuming the following request matches route-a:
**Setting API Key in URL Parameters**
```bash
curl http://xxx.hello.com/test?apikey=2bda943c-ba2b-11ec-ba07-00163e1250b5
```
**Setting API Key in HTTP Request Headers**
```bash
curl http://xxx.hello.com/test -H 'x-api-key: 2bda943c-ba2b-11ec-ba07-00163e1250b5'
```
After successful authentication and authorization, the request's header will have an added `X-Mse-Consumer` field with the value `consumer1`, to identify the name of the caller.
The following requests will be denied access:
**Request without an API Key returns 401**
```bash
curl http://xxx.hello.com/test
```
**Request with an invalid API Key returns 401**
```bash
curl http://xxx.hello.com/test?apikey=926d90ac-ba2e-11ec-ab68-00163e1250b5
```
**Caller matched with provided API Key has no access rights, returns 403**
```bash
# consumer2 is not in the allow list of route-a
curl http://xxx.hello.com/test?apikey=c8c8e9ca-558e-4a2d-bb62-e700dcc40e35
```
### Enabling at the Instance Level
The following configuration will enable Basic Auth authentication at the instance level for the gateway, requiring all requests to pass authentication before accessing.
```yaml
global_auth: true
@@ -37,73 +114,12 @@ consumers:
keys:
- apikey
- x-api-key
in_query: true
```
The `route-a` and `route-b` specified in `_match_route_` in this example are the route names filled in when creating the gateway route. When these two routes are matched, calls whose `name` is `consumer1` will be allowed Access by callers, other callers are not allowed to access;
`*.example.com` and `test.com` specified in `_match_domain_` in this example are used to match the domain name of the request. When the domain name matches, the caller whose `name` is `consumer2` will be allowed to access, and other calls access is not allowed.
### Depending on this configuration, the following requests would allow access：
Assume that the following request will match the route-a route:
**Set the API Key in the url parameter**
```bash
curl http://xxx.hello.com/test?apikey=2bda943c-ba2b-11ec-ba07-00163e1250b5
```
**Set the API Key in the http request header**
```bash
curl http://xxx.hello.com/test -H 'x-api-key: 2bda943c-ba2b-11ec-ba07-00163e1250b5'
```
After the authentication is passed, an `X-Mse-Consumer` field will be added to the header of the request. In this example, its value is `consumer1`, which is used to identify the name of the caller.
### The following requests will deny access：
**The request does not provide an API Key, return 401**
```bash
curl http://xxx.hello.com/test
```
**The API Key provided by the request is not authorized to access, return 401**
```bash
curl http://xxx.hello.com/test?apikey=926d90ac-ba2e-11ec-ab68-00163e1250b5
```
**The caller matched according to the API Key provided in the request has no access rights, return 403**
```bash
# consumer2 is not in the allow list of route-a
curl http://xxx.hello.com/test?apikey=c8c8e9ca-558e-4a2d-bb62-e700dcc40e35
```
## Gateway instance level enabled
The following configuration does not specify the `matchRules` field, so Key Auth authentication will be enabled at the gateway instance level.
```yaml
consumers:
- credential: 2bda943c-ba2b-11ec-ba07-00163e1250b5
name: consumer1
- credential: c8c8e9ca-558e-4a2d-bb62-e700dcc40e35
name: consumer2
keys:
- apikey
in_query: true
```
configuration specify the `matchRules` field, like：
```yaml
matchRules:
- config:
allow:
- consumer1
```
# Error code
| HTTP status code | Error information | Reason |
| ---------------- | --------------------------------------------------------- | -------------------------------------------- |
| 401 | Muti API key found in request. | Muti API provided by request Key. |
| 401 | No API key found in request. | API not provided by request Key. |
| 401 | Request denied by Key Auth check. Invalid API key. | Current API Key access is not allowed. |
| 403 | Request denied by Key Auth check. Unauthorized consumer. | The requested caller does not have access. |
## Related Error Codes
| HTTP Status Code | Error Message | Reason Explanation |
| ---------------- | ---------------------------------------------------------- | --------------------------------- |
| 401 | Request denied by Key Auth check. Multiple API keys found in request | Multiple API Keys provided in the request. |
| 401 | Request denied by Key Auth check. No API key found in request | API Key not provided in the request. |
| 401 | Request denied by Key Auth check. Invalid API key | The current API Key is not authorized for access. |
| 403 | Request denied by Key Auth check. Unauthorized consumer | The caller does not have access permissions. |

276
plugins/wasm-go/extensions/oidc/README.md
View File

@@ -1,140 +1,19 @@
# OIDC Wasm 插件
---
title: OIDC 认证
keywords: [higress, oidc]
description: OIDC 认证插件配置参考
---
## 简介
## 功能说明
本仓库提供了一个高度可集成的Wasm插件，以支持OpenID Connect（OIDC）身份认证。同时，该插件强化了对跨站请求伪造（CSRF）攻击的防御能力，并支持OpenID Connect协议中的注销端点（Logout Endpoint）以及刷新令牌（Refresh Token）机制。在经过Wasm插件进行OIDC验证后的请求将携带 `Authorization` 头部，包含相应的访问令牌（Access Token）。
本插件支持OpenID Connect（OIDC）身份认证。同时，该插件强化了对跨站请求伪造（CSRF）攻击的防御能力，并支持OpenID Connect协议中的注销端点（Logout Endpoint）以及刷新令牌（Refresh Token）机制。在经过Wasm插件进行OIDC验证后的请求将携带 `Authorization` 头部，包含相应的访问令牌（Access Token）。
### OIDC 流程图
## 运行属性
<p align="center">
<img src="https://gw.alicdn.com/imgextra/i3/O1CN01TJSh9c1VwR61Q2nek_!!6000000002717-55-tps-1807-2098.svg" alt="oidc_process" width="600" />
</p>
插件执行阶段：`认证阶段`
插件执行优先级：`350`
### OIDC 流程解析
#### 用户未登录
1. 模拟用户访问对应服务 API
```shell
curl --url "foo.bar.com/headers"
```
2. Higress 重定向到 OIDC Provider 登录页同时携带 client_id、response_type、scope 等 OIDC 认证的参数并设置 csrf cookie 防御CSRF 攻击
```shell
curl --url "https://dev-o43xb1mz7ya7ach4.us.auth0.com/authorize"\
--url-query "approval_prompt=force" \
--url-query "client_id=YagFqRD9tfNIaac5BamjhsSatjrAnsnZ" \
--url-query "redirect_uri=http%3A%2F%2Ffoo.bar.com%2Foauth2%2Fcallback" \
--url-query "response_type=code" \
--url-query "scope=openid+email+offline_access" \
--url-query "state=nT06xdCqn4IqemzBRV5hmO73U_hCjskrH_VupPqdcdw%3A%2Ffoo" \
--header "Set-Cookie: _oauth2_proxy_csrf=LPruATEDgcdmelr8zScD_ObhsbP4zSzvcgmPlcNDcJpFJ0OvhxP2hFotsU-kZnYxd5KsIjzeIXGTOjf8TKcbTHbDIt-aQoZORXI_0id3qeY0Jt78223DPeJ1xBqa8VO0UiEOUFOR53FGxirJOdKFxaAvxDFb1Ok=|1718962455|V1QGWyjQ4hMNOQ4Jtf17HeQJdVqHdt5d65uraFduMIU=; Path=/; Expires=Fri, 21 Jun 2024 08:06:20 GMT; HttpOnly"
```
3. 重定向到登录页
![keycloak_login](https://gw.alicdn.com/imgextra/i4/O1CN01HLcl7r1boXwwnzGqA_!!6000000003512-0-tps-3840-2160.jpg)
4. 用户输入用户名密码登录完成
5. 携带授权重定向到 Higress 并携带了 state 参数用于验证 csrf cookie ，授权码用于交换 token
```shell
curl --url "http://foo.bar.com/oauth2/callback" \
--url-query "state=nT06xdCqn4IqemzBRV5hmO73U_hCjskrH_VupPqdcdw%3A%2Ffoo" \
--url-query "code=0bdopoS2c2lx95u7iO0OH9kY1TvaEdJHo4lB6CT2_qVFm"
```
6. 校验 csrf cookie 中加密存储的 state 值与 url 参数中的 state 值必须相同
7. 利用授权交换 id_token 和 access_token
```shell
curl -X POST \
--url "https://dev-o43xb1mz7ya7ach4.us.auth0.com/oauth/token" \
--data "grant_type=authorization_code" \
--data "client_id=YagFqRD9tfNIaac5BamjhsSatjrAnsnZ" \
--data "client_secret=ekqv5XoZuMFtYms1NszEqRx03qct6BPvGeJUeptNG4y09PrY16BKT9IWezTrrhJJ" \
--data "redirect_uri=http%3A%2F%2Ffoo.bar.com%2Foauth2%2Fcallback" \
--data "code=0bdopoS2c2lx95u7iO0OH9kY1TvaEdJHo4lB6CT2_qVFm" \
```
返回的请求里包含了 id_token, access_token，refresh_token 用于后续刷新 token
```json
{
"access_token": "eyJhbGciOiJkaXIiLCJlbmMiOiJBMjU2R0NNIiwiaXNzIjoiaHR0cHM6Ly9kZXYtbzQzeGIxbXo3eWE3YWNoNC51cy5hdXRoMC5jb20vIn0..WP_WRVM-y3fM1sN4.fAQqtKoKZNG9Wj0OhtrMgtsjTJ2J72M2klDRd9SvUKGbiYsZNPmIl_qJUf81D3VIjD59o9xrOOJIzXTgsfFVA2x15g-jBlNh68N7dyhXu9237Tbplweu1jA25IZDSnjitQ3pbf7xJVIfPnWcrzl6uT8G1EP-omFcl6AQprV2FoKFMCGFCgeafuttppKe1a8mpJDj7AFLPs-344tT9mvCWmI4DuoLFh0PiqMMJBByoijRSxcSdXLPxZng84j8JVF7H6mFa-dj-icP-KLy6yvzEaRKz_uwBzQCzgYK434LIpqw_PRuN3ClEsenwRgIsNdVjvKcoAysfoZhmRy9BQaE0I7qTohSBFNX6A.mgGGeeWgugfXcUcsX4T5dQ",
"refresh_token": "GrZ1f2JvzjAZQzSXmyr1ScWbv8aMFBvzAXHBUSiILcDEG",
"id_token": "eyJhbGciOiJSUzI1NiIsInR5cCI6IkpXVCIsImtpZCI6Imc1Z1ExSF9ZbTY0WUlvVkQwSVpXTCJ9.eyJlbWFpbCI6IjE2MDExNTYyNjhAcXEuY29tIiwiZW1haWxfdmVyaWZpZWQiOmZhbHNlLCJpc3MiOiJodHRwczovL2Rldi1vNDN4YjFtejd5YTdhY2g0LnVzLmF1dGgwLmNvbS8iLCJhdWQiOiJZYWdGcVJEOXRmTklhYWM1QmFtamhzU2F0anJBbnNuWiIsImlhdCI6MTcxOTE5ODYzOCwiZXhwIjoxNzE5MjM0NjM4LCJzdWIiOiJhdXRoMHw2NjVkNzFlNzRjMTMxMTc3YmU2NmU2MDciLCJzaWQiOiJjdDJVOF9ZUS16VDdFOGkwRTNNeUstejc5ZGlWUWhhVSJ9.gfzXKJ0FeqzYqOUDLQHWcUG19IOLqkpLN09xTmIat0umrlGV5VNSumgWH3XJmmwnhdb8AThH3Jf-7kbRJzu4rM-BbGbFTRBTzNHeUajFOFrIgld5VENQ_M_sXHkTp0psWKSr9vF24kmilCfSbvC5lBKjt878ljZ7-xteWuaUYOMUdcJb4DSv0-zjX01sonJxYamTlhji3M4TAW7VwhwqyZt8dBhVSNaRw1wUKj-M1JrBDLyx65sroZtSqVA0udIrqMHEbWYb2de7JjzlqG003HRMzwOm7OXgEd5ZVFqgmBLosgixOU5DJ4A26nlqK92Sp6VqDMRvA-3ym8W_m-wJ_A",
"scope": "openid email offline_access",
"expires_in": 86400,
"token_type": "Bearer"
}
```
8. 将获得的 id_token, access_token, refresh_token 加密存储在cookie _oauth2_proxy中
9. 重定向到用户访问的后端服务并设置 cookie，用于后续用户登录状态的验证，同时清除 cookie _oauth2_proxy_csrf
```json
"Set-Cookie": [
"_oauth2_proxy_csrf=; Path=/; Expires=Mon, 24 Jun 2024 02:17:39 GMT; HttpOnly",
"_oauth2_proxy=8zM_Pcfpp_gesKFe4SMg08o5Iv0A8WAOQOmG1-vZBbQ56UggYVC0Cu-gFMEoxJZU5q1O5vqRlVBizlLetgVjRCksGVbttwl8tQ7h5YiyIubbbtvF1T4JzLh3QfzUUrwbB-VznOkh8qLbjAhddocecjBt4rMiDyceKXqMr4eO5TUEMx4vHtJYnTYalMeTYhGXk5MNSyrdZX9NnQnkdrCjiOQM13ggwob2nYwhGWaAlgzFSWkgkdtBy2Cl_YMWZ8_gKk9rDX289-JrJyGpr5k9O9RzRhZoY2iE3Mcr8-Q37RTji1Ga22QO-XkAcSaGqY1Qo7jLdmgZTYKC5JvtdLc4rj3vcbveYxU7R3Pt2vEribQjKTh4Sqb0aA03p4cxXyZN4SUfBW1NAOm4JLPUhKJy8frqC9_E0nVqPvpvnacaoQs8WkX2zp75xHoMa3SD6KZhQ5JUiPEiNkOaUsyafLvht6lLkNDhgzW3BP2czoe0DCDBLnsot0jH-qQpMZYkaGr-ZnRKI1OPl1vHls3mao5juOAW1VB2A9aughgc8SJ55IFZpMfFMdHdTDdMqPODkItX2PK44GX-pHeLxkOqrzp3GHtMInpL5QIQlTuux3erm3CG-ntlUE7JBtN2T9LEb8XfIFu58X9_vzMun4JQlje2Thi9_taI_z1DSaTtvNNb54wJfSPwYCCl4OsH-BacVmPQhH6TTZ6gP2Qsm5TR2o1U2D9fuVkSM-OPCG9l3tILambIQwC3vofMW6X8SIFSmhJUDvN7NbwxowBiZ6Y7GJRZlAk_GKDkpsdrdIvC67QqczZFphRVnm6qi-gPO41APCbcO6fgTwyOhbP3RrZZKWSIqWJYhNE3_Sfkf0565H7sC7Hc8XUUjJvP3WnjKS9x7KwzWa-dsUjV3-Q-VNl-rXTguVNAIirYK-qrMNMZGCRcJqcLnUF0V_J2lVmFyVsSlE3t0sDw2xmbkOwDptXFOjQL5Rb4esUMYdCBWFajBfvUtcZEFtYhD0kb6VcbjXO3NCVW5qKh_l9C9SRCc7TG1vcRAqUQlRXHacTGWfcWsuQkCJ3Mp_oWaDxs1GRDykQYxAn5sTICovThWEU2C6o75grWaNrkj5NU-0eHh3ryvxLmGLBOXZV9OQhtKShWmUgywSWMxOHOuZAqdAPULc8KheuGFjXYp-RnCbFYWePJmwzfQw89kSkj1KUZgMYwKEjSz62z2qc9KLczomv76ortQzvo4Hv9kaW6xVuQj5R5Oq6_WMBOqsmUMzcXpxCIOGjcdcZRBc0Fm09Uy9oV1PRqvAE4PGtfyrCaoqILBix8UIww63B07YGwzQ-hAXDysBK-Vca2x7GmGdXsNXXcTgu00bdsjtHZPDBBWGfL3g_rMAXr2vWyvK4CwNjcaPAmrlF3geHPwbIePT0hskBboX1v1bsuhzsai7rGM4r53pnb1ZEoTQDa1B-HyokFgo14XiwME0zE1ifpNzefjpkz1YY2krJlqfCydNwoKaTit4tD2yHlnxAeFF9iIrxzSKErNUFpmyLa7ge7V33vhEH-6k5oBTLE2Q2BrC6aAkLCcPwU9xv_SzBDQPRY0MEYv3kGF03Swo1crRbGh-aifYX9NiHDsmG6r1vAnx0MAOw2Jzuz2x6SSdfBrzlcoWBlrwiZzd9kAKq75n1Uy9uzZ8SRnkBrEZySHBwEbu196VklkRE0jqwC-e3wWNNuviSOfwkVeX-7QdOoO10yw9VK2sW52lFvIEf4chv_ta7bGfAZOWBjpktG6ZLD81SE6A88zpqG2SysSyNMp9hl-umG-5sFsjCn_c9E8bDvwkUOUVb9bNqhBDsZgR0BNPawiOZjmyfhzmwmWf-zgFzfFSV6BvOwNRi3sCOHTsWcuk9NBQ_YK8CpNkVl3WeIBSDfidimuC_QV9UWKs1GPk35ZRkM4zKtLY2JsBFWKaDy_P80TcOzcMBoP8gIBClXZ-WUqfE8s1yyc4jrq-qL1_wJ24ef1O9FktsbyZiDKXw2vnqsT8-g_hCeG-unrT1ZFscf8oNdqczARHX-K4vKH2k3uIqEx1M=|1719199056|2rsgdUIClHNEpxBLlHOVRYup6e4oKensQfljtmn4B80=; Path=/; Expires=Mon, 01 Jul 2024 03:17:36 GMT; HttpOnly"
]
```
10. 校验是否存在 cookie 存储了用户的 token 信息同时查看是否过期
11. 使用含有 Authorization 头部存储用户的 access_token 访问相应的 API
```shell
curl --url "foo.bar.com/headers"
--header "Authorization: Bearer eyJhbGciOiJkaXIiLCJlbmMiOiJBMjU2R0NNIiwiaXNzIjoiaHR0cHM6Ly9kZXYtbzQzeGIxbXo3eWE3YWNoNC51cy5hdXRoMC5jb20vIn0..WP_WRVM-y3fM1sN4.fAQqtKoKZNG9Wj0OhtrMgtsjTJ2J72M2klDRd9SvUKGbiYsZNPmIl_qJUf81D3VIjD59o9xrOOJIzXTgsfFVA2x15g-jBlNh68N7dyhXu9237Tbplweu1jA25IZDSnjitQ3pbf7xJVIfPnWcrzl6uT8G1EP-omFcl6AQprV2FoKFMCGFCgeafuttppKe1a8mpJDj7AFLPs-344tT9mvCWmI4DuoLFh0PiqMMJBByoijRSxcSdXLPxZng84j8JVF7H6mFa-dj-icP-KLy6yvzEaRKz_uwBzQCzgYK434LIpqw_PRuN3ClEsenwRgIsNdVjvKcoAysfoZhmRy9BQaE0I7qTohSBFNX6A.mgGGeeWgugfXcUcsX4T5dQ"
```
12. 后端服务根据 access_token 获取用户授权信息并返回对应的 HTTP 响应
```json
{
"email": "******",
"email_verified": false,
"iss": "https://dev-o43xb1mz7ya7ach4.us.auth0.com/",
"aud": "YagFqRD9tfNIaac5BamjhsSatjrAnsnZ",
"iat": 1719198638,
"exp": 1719234638,
"sub": "auth0|665d71e74c131177be66e607",
"sid": "ct2U8_YQ-zT7E8i0E3MyK-z79diVQhaU"
}
```
#### 用户令牌刷新
1. 模拟用户访问对应服务 API
```shell
curl --url "foo.bar.com/headers"
```
2. 验证令牌的过期时间
3. 如果在 cookie 中检测到存在 refresh_token，则可以访问相应的接口以交换新的 id_token 和 access_token
```shell
curl -X POST \
--url "https://dev-o43xb1mz7ya7ach4.us.auth0.com/oauth/token" \
--data "grant_type=refresh_token" \
--data "client_id=YagFqRD9tfNIaac5BamjhsSatjrAnsnZ" \
--data "client_secret=ekqv5XoZuMFtYms1NszEqRx03qct6BPvGeJUeptNG4y09PrY16BKT9IWezTrrhJJ" \
--data "refresh_token=GrZ1f2JvzjAZQzSXmyr1ScWbv8aMFBvzAXHBUSiILcDEG"
```
4. 携带 Authorization 的标头对应 access_token 访问对应 API
5. 后端服务根据 access_token 获取用户授权信息并返回对应的 HTTP 响应
## 配置
### 配置项
## 配置字段
| Option | Type | Description | Default |
| ----------------------------- | ------------ | ------------------------------------------------------------ | ----------------- |
@@ -166,6 +45,8 @@ curl -X POST \
| match_rule_path | string | match rule path such as `/headers` | |
| match_rule_type | string | match rule type can be `exact` or `prefix` or `regex` | |
## 使用方式
### 生成 Cookie 密钥
``` python
@@ -413,4 +294,133 @@ match_list:
#### 登陆成功跳转到服务页面
![aliyun_result](https://gw.alicdn.com/imgextra/i3/O1CN015pGvi51eakt3pFS8Y_!!6000000003888-0-tps-3840-2160.jpg)
![aliyun_result](https://gw.alicdn.com/imgextra/i3/O1CN015pGvi51eakt3pFS8Y_!!6000000003888-0-tps-3840-2160.jpg)
### OIDC 流程图
<p align="center">
<img src="https://gw.alicdn.com/imgextra/i3/O1CN01TJSh9c1VwR61Q2nek_!!6000000002717-55-tps-1807-2098.svg" alt="oidc_process" width="600" />
</p>
### OIDC 流程解析
#### 用户未登录
1. 模拟用户访问对应服务 API
```shell
curl --url "foo.bar.com/headers"
```
2. Higress 重定向到 OIDC Provider 登录页同时携带 client_id、response_type、scope 等 OIDC 认证的参数并设置 csrf cookie 防御CSRF 攻击
```shell
curl --url "https://dev-o43xb1mz7ya7ach4.us.auth0.com/authorize"\
--url-query "approval_prompt=force" \
--url-query "client_id=YagFqRD9tfNIaac5BamjhsSatjrAnsnZ" \
--url-query "redirect_uri=http%3A%2F%2Ffoo.bar.com%2Foauth2%2Fcallback" \
--url-query "response_type=code" \
--url-query "scope=openid+email+offline_access" \
--url-query "state=nT06xdCqn4IqemzBRV5hmO73U_hCjskrH_VupPqdcdw%3A%2Ffoo" \
--header "Set-Cookie: _oauth2_proxy_csrf=LPruATEDgcdmelr8zScD_ObhsbP4zSzvcgmPlcNDcJpFJ0OvhxP2hFotsU-kZnYxd5KsIjzeIXGTOjf8TKcbTHbDIt-aQoZORXI_0id3qeY0Jt78223DPeJ1xBqa8VO0UiEOUFOR53FGxirJOdKFxaAvxDFb1Ok=|1718962455|V1QGWyjQ4hMNOQ4Jtf17HeQJdVqHdt5d65uraFduMIU=; Path=/; Expires=Fri, 21 Jun 2024 08:06:20 GMT; HttpOnly"
```
3. 重定向到登录页
![keycloak_login](https://gw.alicdn.com/imgextra/i4/O1CN01HLcl7r1boXwwnzGqA_!!6000000003512-0-tps-3840-2160.jpg)
4. 用户输入用户名密码登录完成
5. 携带授权重定向到 Higress 并携带了 state 参数用于验证 csrf cookie ，授权码用于交换 token
```shell
curl --url "http://foo.bar.com/oauth2/callback" \
--url-query "state=nT06xdCqn4IqemzBRV5hmO73U_hCjskrH_VupPqdcdw%3A%2Ffoo" \
--url-query "code=0bdopoS2c2lx95u7iO0OH9kY1TvaEdJHo4lB6CT2_qVFm"
```
6. 校验 csrf cookie 中加密存储的 state 值与 url 参数中的 state 值必须相同
7. 利用授权交换 id_token 和 access_token
```shell
curl -X POST \
--url "https://dev-o43xb1mz7ya7ach4.us.auth0.com/oauth/token" \
--data "grant_type=authorization_code" \
--data "client_id=YagFqRD9tfNIaac5BamjhsSatjrAnsnZ" \
--data "client_secret=ekqv5XoZuMFtYms1NszEqRx03qct6BPvGeJUeptNG4y09PrY16BKT9IWezTrrhJJ" \
--data "redirect_uri=http%3A%2F%2Ffoo.bar.com%2Foauth2%2Fcallback" \
--data "code=0bdopoS2c2lx95u7iO0OH9kY1TvaEdJHo4lB6CT2_qVFm" \
```
返回的请求里包含了 id_token, access_token，refresh_token 用于后续刷新 token
```json
{
"access_token": "eyJhbGciOiJkaXIiLCJlbmMiOiJBMjU2R0NNIiwiaXNzIjoiaHR0cHM6Ly9kZXYtbzQzeGIxbXo3eWE3YWNoNC51cy5hdXRoMC5jb20vIn0..WP_WRVM-y3fM1sN4.fAQqtKoKZNG9Wj0OhtrMgtsjTJ2J72M2klDRd9SvUKGbiYsZNPmIl_qJUf81D3VIjD59o9xrOOJIzXTgsfFVA2x15g-jBlNh68N7dyhXu9237Tbplweu1jA25IZDSnjitQ3pbf7xJVIfPnWcrzl6uT8G1EP-omFcl6AQprV2FoKFMCGFCgeafuttppKe1a8mpJDj7AFLPs-344tT9mvCWmI4DuoLFh0PiqMMJBByoijRSxcSdXLPxZng84j8JVF7H6mFa-dj-icP-KLy6yvzEaRKz_uwBzQCzgYK434LIpqw_PRuN3ClEsenwRgIsNdVjvKcoAysfoZhmRy9BQaE0I7qTohSBFNX6A.mgGGeeWgugfXcUcsX4T5dQ",
"refresh_token": "GrZ1f2JvzjAZQzSXmyr1ScWbv8aMFBvzAXHBUSiILcDEG",
"id_token": "eyJhbGciOiJSUzI1NiIsInR5cCI6IkpXVCIsImtpZCI6Imc1Z1ExSF9ZbTY0WUlvVkQwSVpXTCJ9.eyJlbWFpbCI6IjE2MDExNTYyNjhAcXEuY29tIiwiZW1haWxfdmVyaWZpZWQiOmZhbHNlLCJpc3MiOiJodHRwczovL2Rldi1vNDN4YjFtejd5YTdhY2g0LnVzLmF1dGgwLmNvbS8iLCJhdWQiOiJZYWdGcVJEOXRmTklhYWM1QmFtamhzU2F0anJBbnNuWiIsImlhdCI6MTcxOTE5ODYzOCwiZXhwIjoxNzE5MjM0NjM4LCJzdWIiOiJhdXRoMHw2NjVkNzFlNzRjMTMxMTc3YmU2NmU2MDciLCJzaWQiOiJjdDJVOF9ZUS16VDdFOGkwRTNNeUstejc5ZGlWUWhhVSJ9.gfzXKJ0FeqzYqOUDLQHWcUG19IOLqkpLN09xTmIat0umrlGV5VNSumgWH3XJmmwnhdb8AThH3Jf-7kbRJzu4rM-BbGbFTRBTzNHeUajFOFrIgld5VENQ_M_sXHkTp0psWKSr9vF24kmilCfSbvC5lBKjt878ljZ7-xteWuaUYOMUdcJb4DSv0-zjX01sonJxYamTlhji3M4TAW7VwhwqyZt8dBhVSNaRw1wUKj-M1JrBDLyx65sroZtSqVA0udIrqMHEbWYb2de7JjzlqG003HRMzwOm7OXgEd5ZVFqgmBLosgixOU5DJ4A26nlqK92Sp6VqDMRvA-3ym8W_m-wJ_A",
"scope": "openid email offline_access",
"expires_in": 86400,
"token_type": "Bearer"
}
```
8. 将获得的 id_token, access_token, refresh_token 加密存储在cookie _oauth2_proxy中
9. 重定向到用户访问的后端服务并设置 cookie，用于后续用户登录状态的验证，同时清除 cookie _oauth2_proxy_csrf
```json
"Set-Cookie": [
"_oauth2_proxy_csrf=; Path=/; Expires=Mon, 24 Jun 2024 02:17:39 GMT; HttpOnly",
"_oauth2_proxy=8zM_Pcfpp_gesKFe4SMg08o5Iv0A8WAOQOmG1-vZBbQ56UggYVC0Cu-gFMEoxJZU5q1O5vqRlVBizlLetgVjRCksGVbttwl8tQ7h5YiyIubbbtvF1T4JzLh3QfzUUrwbB-VznOkh8qLbjAhddocecjBt4rMiDyceKXqMr4eO5TUEMx4vHtJYnTYalMeTYhGXk5MNSyrdZX9NnQnkdrCjiOQM13ggwob2nYwhGWaAlgzFSWkgkdtBy2Cl_YMWZ8_gKk9rDX289-JrJyGpr5k9O9RzRhZoY2iE3Mcr8-Q37RTji1Ga22QO-XkAcSaGqY1Qo7jLdmgZTYKC5JvtdLc4rj3vcbveYxU7R3Pt2vEribQjKTh4Sqb0aA03p4cxXyZN4SUfBW1NAOm4JLPUhKJy8frqC9_E0nVqPvpvnacaoQs8WkX2zp75xHoMa3SD6KZhQ5JUiPEiNkOaUsyafLvht6lLkNDhgzW3BP2czoe0DCDBLnsot0jH-qQpMZYkaGr-ZnRKI1OPl1vHls3mao5juOAW1VB2A9aughgc8SJ55IFZpMfFMdHdTDdMqPODkItX2PK44GX-pHeLxkOqrzp3GHtMInpL5QIQlTuux3erm3CG-ntlUE7JBtN2T9LEb8XfIFu58X9_vzMun4JQlje2Thi9_taI_z1DSaTtvNNb54wJfSPwYCCl4OsH-BacVmPQhH6TTZ6gP2Qsm5TR2o1U2D9fuVkSM-OPCG9l3tILambIQwC3vofMW6X8SIFSmhJUDvN7NbwxowBiZ6Y7GJRZlAk_GKDkpsdrdIvC67QqczZFphRVnm6qi-gPO41APCbcO6fgTwyOhbP3RrZZKWSIqWJYhNE3_Sfkf0565H7sC7Hc8XUUjJvP3WnjKS9x7KwzWa-dsUjV3-Q-VNl-rXTguVNAIirYK-qrMNMZGCRcJqcLnUF0V_J2lVmFyVsSlE3t0sDw2xmbkOwDptXFOjQL5Rb4esUMYdCBWFajBfvUtcZEFtYhD0kb6VcbjXO3NCVW5qKh_l9C9SRCc7TG1vcRAqUQlRXHacTGWfcWsuQkCJ3Mp_oWaDxs1GRDykQYxAn5sTICovThWEU2C6o75grWaNrkj5NU-0eHh3ryvxLmGLBOXZV9OQhtKShWmUgywSWMxOHOuZAqdAPULc8KheuGFjXYp-RnCbFYWePJmwzfQw89kSkj1KUZgMYwKEjSz62z2qc9KLczomv76ortQzvo4Hv9kaW6xVuQj5R5Oq6_WMBOqsmUMzcXpxCIOGjcdcZRBc0Fm09Uy9oV1PRqvAE4PGtfyrCaoqILBix8UIww63B07YGwzQ-hAXDysBK-Vca2x7GmGdXsNXXcTgu00bdsjtHZPDBBWGfL3g_rMAXr2vWyvK4CwNjcaPAmrlF3geHPwbIePT0hskBboX1v1bsuhzsai7rGM4r53pnb1ZEoTQDa1B-HyokFgo14XiwME0zE1ifpNzefjpkz1YY2krJlqfCydNwoKaTit4tD2yHlnxAeFF9iIrxzSKErNUFpmyLa7ge7V33vhEH-6k5oBTLE2Q2BrC6aAkLCcPwU9xv_SzBDQPRY0MEYv3kGF03Swo1crRbGh-aifYX9NiHDsmG6r1vAnx0MAOw2Jzuz2x6SSdfBrzlcoWBlrwiZzd9kAKq75n1Uy9uzZ8SRnkBrEZySHBwEbu196VklkRE0jqwC-e3wWNNuviSOfwkVeX-7QdOoO10yw9VK2sW52lFvIEf4chv_ta7bGfAZOWBjpktG6ZLD81SE6A88zpqG2SysSyNMp9hl-umG-5sFsjCn_c9E8bDvwkUOUVb9bNqhBDsZgR0BNPawiOZjmyfhzmwmWf-zgFzfFSV6BvOwNRi3sCOHTsWcuk9NBQ_YK8CpNkVl3WeIBSDfidimuC_QV9UWKs1GPk35ZRkM4zKtLY2JsBFWKaDy_P80TcOzcMBoP8gIBClXZ-WUqfE8s1yyc4jrq-qL1_wJ24ef1O9FktsbyZiDKXw2vnqsT8-g_hCeG-unrT1ZFscf8oNdqczARHX-K4vKH2k3uIqEx1M=|1719199056|2rsgdUIClHNEpxBLlHOVRYup6e4oKensQfljtmn4B80=; Path=/; Expires=Mon, 01 Jul 2024 03:17:36 GMT; HttpOnly"
]
```
10. 校验是否存在 cookie 存储了用户的 token 信息同时查看是否过期
11. 使用含有 Authorization 头部存储用户的 access_token 访问相应的 API
```shell
curl --url "foo.bar.com/headers"
--header "Authorization: Bearer eyJhbGciOiJkaXIiLCJlbmMiOiJBMjU2R0NNIiwiaXNzIjoiaHR0cHM6Ly9kZXYtbzQzeGIxbXo3eWE3YWNoNC51cy5hdXRoMC5jb20vIn0..WP_WRVM-y3fM1sN4.fAQqtKoKZNG9Wj0OhtrMgtsjTJ2J72M2klDRd9SvUKGbiYsZNPmIl_qJUf81D3VIjD59o9xrOOJIzXTgsfFVA2x15g-jBlNh68N7dyhXu9237Tbplweu1jA25IZDSnjitQ3pbf7xJVIfPnWcrzl6uT8G1EP-omFcl6AQprV2FoKFMCGFCgeafuttppKe1a8mpJDj7AFLPs-344tT9mvCWmI4DuoLFh0PiqMMJBByoijRSxcSdXLPxZng84j8JVF7H6mFa-dj-icP-KLy6yvzEaRKz_uwBzQCzgYK434LIpqw_PRuN3ClEsenwRgIsNdVjvKcoAysfoZhmRy9BQaE0I7qTohSBFNX6A.mgGGeeWgugfXcUcsX4T5dQ"
```
12. 后端服务根据 access_token 获取用户授权信息并返回对应的 HTTP 响应
```json
{
"email": "******",
"email_verified": false,
"iss": "https://dev-o43xb1mz7ya7ach4.us.auth0.com/",
"aud": "YagFqRD9tfNIaac5BamjhsSatjrAnsnZ",
"iat": 1719198638,
"exp": 1719234638,
"sub": "auth0|665d71e74c131177be66e607",
"sid": "ct2U8_YQ-zT7E8i0E3MyK-z79diVQhaU"
}
```
#### 用户令牌刷新
1. 模拟用户访问对应服务 API
```shell
curl --url "foo.bar.com/headers"
```
2. 验证令牌的过期时间
3. 如果在 cookie 中检测到存在 refresh_token，则可以访问相应的接口以交换新的 id_token 和 access_token
```shell
curl -X POST \
--url "https://dev-o43xb1mz7ya7ach4.us.auth0.com/oauth/token" \
--data "grant_type=refresh_token" \
--data "client_id=YagFqRD9tfNIaac5BamjhsSatjrAnsnZ" \
--data "client_secret=ekqv5XoZuMFtYms1NszEqRx03qct6BPvGeJUeptNG4y09PrY16BKT9IWezTrrhJJ" \
--data "refresh_token=GrZ1f2JvzjAZQzSXmyr1ScWbv8aMFBvzAXHBUSiILcDEG"
```
4. 携带 Authorization 的标头对应 access_token 访问对应 API
5. 后端服务根据 access_token 获取用户授权信息并返回对应的 HTTP 响应

354
plugins/wasm-go/extensions/oidc/README_EN.md Normal file
View File

@@ -0,0 +1,354 @@
---
title: OIDC Authentication
keywords: [higress, oidc]
description: OIDC Authentication Plugin Configuration Reference
---
## Function Description
This plugin supports OpenID Connect (OIDC) identity authentication. Additionally, it enhances defenses against Cross-Site Request Forgery (CSRF) attacks and supports the Logout Endpoint and Refresh Token mechanism in the OpenID Connect protocol. Requests verified by the Wasm plugin after OIDC validation will carry the `Authorization` header, including the corresponding Access Token.
## Running Attributes
Plugin execution phase: `authentication phase`
Plugin execution priority: `350`
## Configuration Fields
| Option | Type | Description | Default |
| ----------------------------- | ------------ | ------------------------------------------------------------ | ----------------- |
| cookie_name | string | The name of the cookie that the oauth_proxy creates. Should be changed to use a [cookie prefix](https://developer.mozilla.org/en-US/docs/Web/HTTP/Cookies#cookie_prefixes) (`__Host-` or `__Secure-`) if `--cookie-secure` is set. | `"_oauth2_proxy"` |
| cookie_secret | string | The seed string for secure cookies (optionally base64 encoded) | |
| cookie_domains | string\|list | Optional cookie domains to force cookies to (e.g. `.yourcompany.com`). The longest domain matching the request's host will be used (or the shortest cookie domain if there is no match). | |
| cookie_path | string | An optional cookie path to force cookies to (e.g. `/poc/`) | `"/"` |
| cookie_expire | duration | Expire timeframe for cookie. If set to 0, cookie becomes a session-cookie which will expire when the browser is closed. | 168h0m0s |
| cookie_refresh | duration | Refresh the cookie after this duration; `0` to disable | |
| cookie_secure | bool | Set [secure (HTTPS only) cookie flag](https://owasp.org/www-community/controls/SecureFlag) | true |
| cookie_httponly | bool | Set HttpOnly cookie flag | true |
| cookie_samesite | string | Set SameSite cookie attribute (`"lax"`, `"strict"`, `"none"`, or `""`). | `""` |
| cookie_csrf_per_request | bool | Enable having different CSRF cookies per request, making it possible to have parallel requests. | false |
| cookie_csrf_expire | duration | Expire timeframe for CSRF cookie | 15m |
| client_id | string | The OAuth Client ID | |
| client_secret | string | The OAuth Client Secret | |
| provider | string | OAuth provider | oidc |
| pass_authorization_header | bool | Pass OIDC IDToken to upstream via Authorization Bearer header | true |
| oidc_issuer_url | string | The OpenID Connect issuer URL, e.g. `"https://dev-o43xb1mz7ya7ach4.us.auth0.com"` | |
| oidc_verifier_request_timeout | uint32 | OIDC verifier discovery request timeout | 2000(ms) |
| scope | string | OAuth scope specification | |
| redirect_url | string | The OAuth Redirect URL, e.g. `"https://internalapp.yourcompany.com/oauth2/callback"` | |
| service_name | string | Registered name of the OIDC service, e.g. `auth.dns`, `keycloak.static` | |
| service_port | int64 | Service port of the OIDC service | |
| service_host | string | Host of the OIDC service when type is static IP | |
| match_type | string | Match type (`whitelist` or `blacklist`) | `"whitelist"` |
| match_list | rule\|list | A list of (match_rule_domain, match_rule_path, and match_rule_type). | |
| match_rule_domain | string | Match rule domain, support wildcard pattern such as `*.bar.com` | |
| match_rule_path | string | Match rule path such as `/headers` | |
| match_rule_type | string | Match rule type can be `exact` or `prefix` or `regex` | |
## Usage
### Generate Cookie Secret
``` python
python -c 'import os, base64; print(base64.urlsafe_b64encode(os.urandom(32)).decode())'
```
Reference: [Oauth2-proxy Generating a Cookie Secret](https://oauth2-proxy.github.io/oauth2-proxy/configuration/overview#generating-a-cookie-secret)
### Whitelist and Blacklist Mode
Supports whitelist and blacklist configuration, defaulting to whitelist mode, which is empty, meaning all requests need to be validated. Domain matching supports wildcard domains like `*.bar.com`, and matching rules support exact match `exact`, prefix match `prefix`, and regex match `regex`.
* **Whitelist Mode**
```yaml
match_type: 'whitelist'
match_list:
- match_rule_domain: '*.bar.com'
match_rule_path: '/foo'
match_rule_type: 'prefix'
```
Requests matching the prefix `/foo` under the wildcard domain `*.bar.com` do not need validation.
* **Blacklist Mode**
```yaml
match_type: 'blacklist'
match_list:
- match_rule_domain: '*.bar.com'
match_rule_path: '/headers'
match_rule_type: 'prefix'
```
Only requests matching the prefix `/headers` under the wildcard domain `*.bar.com` need validation.
### Logout Users
To log out users, they must be redirected to the `/oauth2/sign_out` endpoint. This endpoint only removes cookies set by oauth2-proxy, meaning the user remains logged into the OIDC Provider and may be automatically logged back in upon re-accessing the application. Therefore, the `rd` query parameter must be used to redirect the user to the authentication provider's logout page, redirecting the user to a URL similar to the following (note URL encoding!):
```
/oauth2/sign_out?rd=https%3A%2F%2Fmy-oidc-provider.example.com%2Fsign_out_page
```
Alternatively, the redirect URL can be included in the `X-Auth-Request-Redirect` header:
```
GET /oauth2/sign_out HTTP/1.1
X-Auth-Request-Redirect: https://my-oidc-provider.example.com/sign_out_page
...
```
The redirect URL can include the `post_logout_redirect_uri` parameter to specify the page to redirect to after OIDC Provider logout; for example, the backend service's logout page. If this parameter is not included, it will default to redirect to the OIDC Provider's logout page. For details, see the examples below for Auth0 and Keycloak (if the OIDC Provider supports session management and discovery, then "sign_out_page" should be obtained from the [metadata](https://openid.net/specs/openid-connect-discovery-1_0.html#ProviderConfig) `end_session_endpoint`).
### OIDC Service HTTPS Protocol
If the OIDC Provider uses the HTTPS protocol, refer to Higress's documentation on [Configuring Backend Service Protocol: HTTPS](https://higress.io/docs/latest/user/annotation-use-case/#%E9%85%8D%E7%BD%AE%E5%90%8E%E7%AB%AF%E6%9C%8D%E5%8A%A1%E5%8D%8F%E8%AE%AEhttps%E6%88%96grpc), which states that requests forwarded to the backend service should use the HTTPS protocol by using the annotation `higress.io/backend-protocol: "HTTPS"`, as shown in the Auth0 example Ingress configuration.
## Configuration Example
### Auth0 Configuration Example
#### Step 1: Configure Auth0 Account
- Log in to the Developer Okta website [Developer Auth0 site](https://auth0.com/)
- Register a test web application
**Note**: You must fill in the Allowed Callback URLs, Allowed Logout URLs, Allowed Web Origins, etc., otherwise the OIDC Provider will consider the user's redirect URL or logout URL to be invalid.
#### Step 2: Higress Configure Service Source
* Create an Auth0 DNS source in Higress service sources.
![auth0 create](https://gw.alicdn.com/imgextra/i1/O1CN01p9y0jF1tfzdXTzNYm_!!6000000005930-0-tps-3362-670.jpg)
#### Step 3: OIDC Service HTTPS Configuration
```yaml
apiVersion: networking.k8s.io/v1
kind: Ingress
metadata:
name: auth0-ingress
annotations:
higress.io/destination: auth.dns
higress.io/backend-protocol: "HTTPS"
higress.io/ignore-path-case: "false"
spec:
ingressClassName: higress
rules:
- host: foo.bar.com
http:
paths:
- path: /
pathType: Prefix
backend:
resource:
apiGroup: networking.higress.io
kind: McpBridge
name: default
```
#### Step 4: Wasm Plugin Configuration
```yaml
redirect_url: 'http://foo.bar.com/oauth2/callback'
oidc_issuer_url: 'https://dev-o43xb1mz7ya7ach4.us.auth0.com/'
client_id: 'XXXXXXXXXXXXXXXX'
client_secret: 'XXXXXXXXXXXXXXXX'
scope: 'openid email offline_access'
cookie_secret: 'nqavJrGvRmQxWwGNptLdyUVKcBNZ2b18Guc1n_8DCfY='
service_name: 'auth.dns'
service_port: 443
match_type: 'whitelist'
match_list:
- match_rule_domain: '*.bar.com'
match_rule_path: '/foo'
match_rule_type: 'prefix'
```
**Note**: You must configure the service source first. The Wasm plugin needs to access the configured service to obtain OpenID configuration during initialization.
#### Access Service Page; Redirect if Not Logged In
![auth0_login](https://gw.alicdn.com/imgextra/i3/O1CN01hVNk0C1gkUWLwuC0N_!!6000000004180-0-tps-3840-2160.jpg)
#### Successful Login Redirects to Service Page
In the headers, you can see the `_oauth2_proxy` cookie carried for the next login access, and the Authorization corresponds to the IDToken used to get user information from the backend service.
![auth0 service](https://gw.alicdn.com/imgextra/i1/O1CN01vyrB6u1xPHep1RRqb_!!6000000006435-2-tps-3840-2160.png)
#### Access Logout Redirects to Logout Page
```
http://foo.bar.com/oauth2/sign_out?rd=https%3A%2F%2Fdev-o43xb1mz7ya7ach4.us.auth0.com%2Foidc%2Flogout
```
![auth0 logout](https://gw.alicdn.com/imgextra/i3/O1CN01UntF4x1UqC4StMqtT_!!6000000002568-0-tps-3840-2160.jpg)
#### Access Logout Redirects to Logout Page (with post_logout_redirect_uri Parameter to Redirect to Specified URI)
```
http://foo.bar.com/oauth2/sign_out?rd=https%3A%2F%2Fdev-o43xb1mz7ya7ach4.us.auth0.com%2Foidc%2Flogout%3Fpost_logout_redirect_uri%3Dhttp%3A%2F%2Ffoo.bar.com%2Ffoo
```
Note: The URI to which `post_logout_redirect_uri` redirects needs to be configured in the OIDC Provider Allowed URLs for a normal redirection.
![auth0 logout redirect](https://gw.alicdn.com/imgextra/i1/O1CN01AtZ2cd1JlBxsgyCjG_!!6000000001068-0-tps-3840-2160.jpg)
### Keycloak Configuration Example
#### Step 1: Get Started with Keycloak on Docker
<https://www.keycloak.org/getting-started/getting-started-docker>
**Note**: You must fill in Valid redirect URIs, Valid post logout URIs, Web origins, otherwise the OIDC Provider will consider the user's redirect URL or logout URL to be invalid.
#### Step 2: Higress Configure Service Source
* Create a Keycloak fixed address service in Higress service sources.
![keycloak create](https://gw.alicdn.com/imgextra/i1/O1CN01p9y0jF1tfzdXTzNYm_!!6000000005930-0-tps-3362-670.jpg)
#### Step 3: Wasm Plugin Configuration
```yaml
redirect_url: 'http://foo.bar.com/oauth2/callback'
oidc_issuer_url: 'http://127.0.0.1:9090/realms/myrealm'
client_id: 'XXXXXXXXXXXXXXXX'
client_secret: 'XXXXXXXXXXXXXXXX'
scope: 'openid email'
cookie_secret: 'nqavJrGvRmQxWwGNptLdyUVKcBNZ2b18Guc1n_8DCfY='
service_name: 'keycloak.static'
service_port: 80
service_host: '127.0.0.1:9090'
match_type: 'blacklist'
match_list:
- match_rule_domain: '*.bar.com'
match_rule_path: '/headers'
match_rule_type: 'prefix'
```
#### Access Service Page; Redirect if Not Logged In
![keycloak_login](https://gw.alicdn.com/imgextra/i4/O1CN01HLcl7r1boXwwnzGqA_!!6000000003512-0-tps-3840-2160.jpg)
#### Successful Login Redirects to Service Page
![keycloak service](https://gw.alicdn.com/imgextra/i1/O1CN01vyrB6u1xPHep1RRqb_!!6000000006435-2-tps-3840-2160.png)
#### Access Logout Redirects to Logout Page
```
http://foo.bar.com/oauth2/sign_out?rd=http%3A%2F%2F127.0.0.1:9090%2Frealms%2Fmyrealm%2Fprotocol%2Fopenid-connect%2Flogout
```
![keycloak logout](https://gw.alicdn.com/imgextra/i4/O1CN01kQwqB523OiroOWMgM_!!6000000007246-0-tps-3840-2160.jpg)
#### Access Logout Redirects to Logout Page (with post_logout_redirect_uri Parameter to Redirect to Specified URI)
```
http://foo.bar.com/oauth2/sign_out?rd=http%3A%2F%2F127.0.0.1:9090%2Frealms%2Fmyrealm%2Fprotocol%2Fopenid-connect%2Flogout%3Fpost_logout_redirect_uri%3Dhttp%3A%2F%2Ffoo.bar.com%2Ffoo
```
![keycloak logout redirect](https://gw.alicdn.com/imgextra/i1/O1CN01AtZ2cd1JlBxsgyCjG_!!6000000001068-0-tps-3840-2160.jpg)
### Aliyun Configuration Example
#### Step 1: Configure Aliyun OAuth Application
Refer to the [Web Application Login to Alibaba Cloud](https://help.aliyun.com/zh/ram/user-guide/access-alibaba-cloud-apis-from-a-web-application) process to configure the OAuth application.
#### Step 2: Higress Configure Service Source
* Create an Aliyun DNS service in Higress service sources.
![Aliyun service](https://gw.alicdn.com/imgextra/i3/O1CN01PMNGFS1mHXBtsEvEq_!!6000000004929-0-tps-3312-718.jpg)
#### Step 3: Wasm Plugin Configuration
```yaml
redirect_url: 'http://foo.bar.com/oauth2/callback'
provider: aliyun
oidc_issuer_url: 'https://oauth.aliyun.com/'
client_id: 'XXXXXXXXXXXXXXXX'
client_secret: 'XXXXXXXXXXXXXXXX'
scope: 'openid'
cookie_secret: 'nqavJrGvRmQxWwGNptLdyUVKcBNZ2b18Guc1n_8DCfY='
service_name: 'aliyun.dns'
service_port: 443
match_type: whitelist
match_list:
- match_rule_domain: 'foo.bar.com'
match_rule_path: /foo
match_rule_type: prefix
```
#### Access Service Page; Redirect if Not Logged In
![aliyun_login_1](https://gw.alicdn.com/imgextra/i1/O1CN01L379Uk1b2umAraylT_!!6000000003408-0-tps-3840-2160.jpg)
Directly login using a RAM user or click the main account login.
![aliyun_login_2](https://gw.alicdn.com/imgextra/i1/O1CN01pfdA3l27Dy2TL83NA_!!6000000007764-0-tps-3840-2160.jpg)
#### Successful Login Redirects to Service Page
![aliyun_result](https://gw.alicdn.com/imgextra/i3/O1CN015pGvi51eakt3pFS8Y_!!6000000003888-0-tps-3840-2160.jpg)
### OIDC Flow Diagram
<p align="center">
<img src="https://gw.alicdn.com/imgextra/i3/O1CN01TJSh9c1VwR61Q2nek_!!6000000002717-55-tps-1807-2098.svg" alt="oidc_process" width="600" />
</p>
### OIDC Flow Analysis
#### User Not Logged In
1. Simulate user accessing the corresponding service API
```shell
curl --url "foo.bar.com/headers"
```
2. Higress redirects to the OIDC Provider login page while carrying the client_id, response_type, scope, and other OIDC authentication parameters, setting a CSRF cookie to defend against CSRF attacks.
```shell
curl --url "https://dev-o43xb1mz7ya7ach4.us.auth0.com/authorize"\
--url-query "approval_prompt=force" \
--url-query "client_id=YagFqRD9tfNIaac5BamjhsSatjrAnsnZ" \
--url-query "redirect_uri=http%3A%2F%2Ffoo.bar.com%2Foauth2%2Fcallback" \
--url-query "response_type=code" \
--url-query "scope=openid+email+offline_access" \
--url-query "state=nT06xdCqn4IqemzBRV5hmO73U_hCjskrH_VupPqdcdw%3A%2Ffoo" \
--header "Set-Cookie: _oauth2_proxy_csrf=LPruATEDgcdmelr8zScD_ObhsbP4zSzvcgmPlcNDcJpFJ0OvhxP2hFotsU-kZnYxd5KsIjzeIXGTOjf8TKcbTHbDIt-aQoZORXI_0id3qeY0Jt78223DPeJ1xBqa8VO0UiEOUFOR53FGxirJOdKFxaAvxDFb1Ok=|1718962455|V1QGWyjQ4hMNOQ4Jtf17HeQJdVqHdt5d65uraFduMIU=; Path=/; Expires=Fri, 21 Jun 2024 08:06:20 GMT; HttpOnly"
```
3. Redirect to the login page.
![keycloak_login](https://gw.alicdn.com/imgextra/i4/O1CN01HLcl7r1boXwwnzGqA_!!6000000003512-0-tps-3840-2160.jpg)
4. The user enters the username and password to log in.
5. A redirect back to Higress occurs with the authorization code and state parameter for CSRF cookie validation.
```shell
curl --url "http://foo.bar.com/oauth2/callback" \
--url-query "state=nT06xdCqn4IqemzBRV5hmO73U_hCjskrH_VupPqdcdw%3A%2Ffoo" \
--url-query "code=0bdopoS2c2lx95u7iO0OH9kY1TvaEdJHo4lB6CT2_qVFm"
```
6. Verify that the encrypted state value stored in the CSRF cookie matches the state value in the URL parameters.
7. Use the authorization code to exchange for id_token and access_token.
```shell
curl -X POST \
--url "https://dev-o43xb1mz7ya7ach4.us.auth0.com/oauth/token" \
--data "grant_type=authorization_code" \
--data "client_id=YagFqRD9tfNIaac5BamjhsSatjrAnsnZ" \
--data "client_secret=ekqv5XoZuMFtYms1NszEqRx03qct6BPvGeJUeptNG4y09PrY16BKT9IWezTrrhJJ" \
--data "redirect_uri=http%3A%2F%2Ffoo.bar.com%2Foauth2%2Fcallback" \
--data "code=0bdopoS2c2lx95u7iO0OH9kY1TvaEdJHo4lB6CT2_qVFm" \
```
The response will include id_token, access_token, and refresh_token for future token refresh.
```json
{
"access_token": "eyJhbGciOiJkaXIiLCJlbmMiOiJBMjU2R0NNIiwiaXNzIjoiaHR0cHM6Ly9kZXYtbzQzeGIxbXo3eWE3YWNoNC51cy5hdXRoMC5jb20vIn0..WP_WRVM-y3fM1sN4.fAQqtKoKZNG9Wj0OhtrMgtsjTJ2J72M2klDRd9SvUKGbiYsZNPmIl_qJUf81D3VIjD59o9xrOOJIzXTgsfFVA2x15g-jBlNh68N7dyhXu9237Tbplweu1jA25IZDSnjitQ3pbf7xJVIfPnWcrzl6uT8G1EP-omFcl6AQprV2FoKFMCGFCgeafuttppKe1a8mpJDj7AFLPs-344tT9mvCWmI4DuoLFh0PiqMMJBByoijRSxcSdXLPxZng84j8JVF7H6mFa-dj-icP-KLy6yvzEaRKz_uwBzQCzgYK434LIpqw_PRuN3ClEsenwRgIsNdVjvKcoAysfoZhmRy9BQaE0I7qTohSBFNX6A.mgGGeeWgugfXcUcsX4T5dQ",
"refresh_token": "GrZ1f2JvzjAZQzSXmyr1ScWbv8aMFBvzAXHBUSiILcDEG",
"id_token": "eyJhbGciOiJSUzI1NiIsInR5cCI6IkpXVCIsImtpZCI6Imc1Z1ExSF9ZbTY0WUlvVkQwSVpXTCJ9.eyJlbWFpbCI6IjE2MDExNTYyNjhAcXEuY29tIiwiZW1haWxfdmVyaWZpZWQiOmZhbHNlLCJpc3MiOiJodHRwczovL2Rldi1vNDN4YjFtejd5YTdhY2g0LnVzLmF1dGgwLmNvbS8iLCJhdWQiOiJZYWdGcVJEOXRmTklhYWM1QmFtamhzU2F0anJBbnNuWiIsImlhdCI6MTcxOTE5ODYzOCwiZXhwIjoxNzE5MjM0NjM4LCJzdWIiOiJhdXRoMHw2NjVkNzFlNzRjMTMxMTc3YmU2NmU2MDciLCJzaWQiOiJjdDJVOF9ZUS16VDdFOGkwRTNNeUstejc5ZGlWUWhhVSJ9.gfzXKJ0FeqzYqOUDLQHWcUG19IOLqkpLN09xTmIat0umrlGV5VNSumgWH3XJmmwnhdb8AThH3Jf-7kbRJzu4rM-BbGbFTRBTzNHeUajFOFrIgld5VENQ_M_sXHkTp0psWKSr9vF24kmilCfSbvC5lBKjt878ljZ7-xteWuaUYOMUdcJb4DSv0-zjX01sonJxYamTlhji3M4TAW7VwhwqyZt8dBhVSNaRw1wUKj-M1JrBDLyx65sroZtSqVA0udIrqMHEbWYb2de7JjzlqG003HRMzwOm7OXgEd5ZVFqgmBLosgixOU5DJ4A26nlqK92Sp6VqDMRvA-3ym8W_m-wJ_A",
"scope": "openid email offline_access",
"expires_in": 86400,
"token_type": "Bearer"
}
```
8. The obtained id_token, access_token, and refresh_token are encrypted and stored in the `_oauth2_proxy` cookie.
9. Redirect to the backend service accessed by the user and set cookies for subsequent user login state verification while clearing the `_oauth2_proxy_csrf` cookie.
```json
"Set-Cookie": [
"_oauth2_proxy_csrf=; Path=/; Expires=Mon, 24 Jun 2024 02:17:39 GMT; HttpOnly",
"_oauth2_proxy=8zM_Pcfpp_gesKFe4SMg08o5Iv0A8WAOQOmG1-vZBbQ56UggYVC0Cu-gFMEoxJZU5q1O5vqRlVBizlLetgVjRCksGVbttwl8tQ7h5YiyIubbbtvF1T4JzLh3QfzUUrwbB-VznOkh8qLbjAhddocecjBt4rMiDyceKXqMr4eO5TUEMx4vHtJYnTYalMeTYhGXk5MNSyrdZX9NnQnkdrCjiOQM13ggwob2nYwhGWaAlgzFSWkgkdtBy2Cl_YMWZ8_gKk9rDX289-JrJyGpr5k9O9RzRhZoY2iE3Mcr8-Q37RTji1Ga22QO-XkAcSaGqY1Qo7jLdmgZTYKC5JvtdLc4rj3vcbveYxU7R3Pt2vEribQjKTh4Sqb0aA03p4cxXyZN4SUfBW1NAOm4JLPUhKJy8frqC9_E0nVqPvpvnacaoQs8WkX2zp75xHoMa3SD6KZhQ5JUiPEiNkOaUsyafLvht6lLkNDhgzW3BP2czoe0DCDBLnsot0jH-qQpMZYkaGr-ZnRKI1OPl1vHls3mao5juOAW1VB2A9aughgc8SJ55IFZpMfFMdHdTDdMqPODkItX2PK44GX-pHeLxkOqrzp3GHtMInpL5QIQlTuux3erm3CG-ntlUE7JBtN2T9LEb8XfIFu58X9_vzMun4JQlje2Thi9_taI_z1DSaTtvNNb54wJfSPwYCCl4OsH-BacVmPQhH6TTZ6gP2Qsm5TR2o1U2D9fuVkSM-OPCG9l3tILambIQwC3vofMW6X8SIFSmhJUDvN7NbwxowBiZ6Y7GJRZlAk_GKDkpsdrdIvC67QqczZFphRVnm6qi-gPO41APCbcO6fgTwyOhbP3RrZZKWSIqWJYhNE3_Sfkf0565H7sC7Hc8XUUjJvP3WnjKS9x7KwzWa-dsUjV3-Q-VNl-rXTguVNAIirYK-qrMNMZGCRcJqcLnUF0V_J2lVmFyVsSlE3t0sDw2xmbkOwDptXFOjQL5Rb4esUMYdCBWFajBfvUtcZEFtYhD0kb6VcbjXO3NCVW5qKh_l9C9SRCc7TG1vcRAqUQlRXHacTGWfcWsuQkCJ3Mp_oWaDxs1GRDykQYxAn5sTICovThWEU2C6o75grWaNrkj5NU-0eHh3ryvxLmGLBOXZV9OQhtKShWmUgywSWMxOHOuZAqdAPULc8KheuGFjXYp-RnCbFYWePJmwzfQw89kSkj1KUZgMYwKEjSz62z2qc9KLczomv76ortQzvo4Hv9kaW6xVuQj5R5Oq6_WMBOqsmUMzcXpxCIOGjcdcZRBc0Fm09Uy9oV1PRqvAE4PGtfyrCaoqILBix8UIww63B07YGwzQ-hAXDysBK-Vca2x7GmGdXsNXXcTgu00bdsjtHZPDBBWGfL3g_rMAXr2vWyvK4CwNjcaPAmrlF3geHPwbIePT0hskBboX1v1bsuhzsai7rGM4r53pnb1ZEoTQDa1B-HyokFgo14XiwME0zE1ifpNzefjpkz1YY2krJlqfCydNwoKaTit4tD2yHlnxAeFF9iIrxzSKErNUFpmyLa7ge7V33vhEH-6k5oBTLE2Q2BrC6aAkLCcPwU9xv_SzBDQPRY0MEYv3kGF03Swo1crRbGh-aifYX9NiHDsmG6r1vAnx0MAOw2Jzuz2x6SSdfBrzlcoWBlrwiZzd9kAKq75n1Uy9uzZ8SRnkBrEZySHBwEbu196VklkRE0jqwC-e3wWNNuviSOfwkVeX-7QdOoO10yw9VK2sW52lFvIEf4chv_ta7bGfAZOWBjpktG6ZLD81SE6A88zpqG2SysSyNMp9hl-umG-5sFsjCn_c9E8bDvwkUOUVb9bNqhBDsZgR0BNPawiOZjmyfhzmwmWf-zgFzfFSV6BvOwNRi3sCOHTsWcuk9NBQ_YK8CpNkVl3WeIBSDfidimuC_QV9UWKs1GPk35ZRkM4zKtLY2JsBFWKaDy_P80TcOzcMBoP8gIBClXZ-WUqfE8s1yyc4jrq-qL1_wJ24ef1O9FktsbyZiDKXw2vnqsT8-g_hCeG-unrT1ZFscf8oNdqczARHX-K4vKH2k3uIqEx1M=|1719199056|2rsgdUIClHNEpxBLlHOVRYup6e4oKensQfljtmn4B80=; Path=/; Expires=Mon, 01 Jul 2024 03:17:36 GMT; HttpOnly"
]
```
10. Verify the existence of the cookie storing user's token information and check if it has expired.
11. Use the access token with the Authorization header to access the corresponding API.
```shell
curl --url "foo.bar.com/headers"
--header "Authorization: Bearer eyJhbGciOiJkaXIiLCJlbmMiOiJBMjU2R0NNIiwiaXNzIjoiaHR0cHM6Ly9kZXYtbzQzeGIxbXo3eWE3YWNoNC51cy5hdXRoMC5jb20vIn0..WP_WRVM-y3fM1sN4.fAQqtKoKZNG9Wj0OhtrMgtsjTJ2J72M2klDRd9SvUKGbiYsZNPmIl_qJUf81D3VIjD59o9xrOOJIzXTgsfFVA2x15g-jBlNh68N7dyhXu9237Tbplweu1jA25IZDSnjitQ3pbf7xJVIfPnWcrzl6uT8G1EP-omFcl6AQprV2FoKFMCGFCgeafuttppKe1a8mpJDj7AFLPs-344tT9mvCWmI4DuoLFh0PiqMMJBByoijRSxcSdXLPxZng84j8JVF7H6mFa-dj-icP-KLy6yvzEaRKz_uwBzQCzgYK434LIpqw_PRuN3ClEsenwRgIsNdVjvKcoAysfoZhmRy9BQaE0I7qTohSBFNX6A.mgGGeeWgugfXcUcsX4T5dQ"
```
12. The backend service obtains user authorization information based on the access token and returns the corresponding HTTP response.
```json
{
"email": "******",
"email_verified": false,
"iss": "https://dev-o43xb1mz7ya7ach4.us.auth0.com/",
"aud": "YagFqRD9tfNIaac5BamjhsSatjrAnsnZ",
"iat": 1719198638,
"exp": 1719234638,
"sub": "auth0|665d71e74c131177be66e607",
"sid": "ct2U8_YQ-zT7E8i0E3MyK-z79diVQhaU"
}
```
#### User Token Refresh
1. Simulate user accessing the corresponding service API.
```shell
curl --url "foo.bar.com/headers"
```
2. Verify the expiration time of the token.
3. If a refresh_token is detected in the cookie, access the corresponding interface to exchange a new id_token and access_token.
```shell
curl -X POST \
--url "https://dev-o43xb1mz7ya7ach4.us.auth0.com/oauth/token" \
--data "grant_type=refresh_token" \
--data "client_id=YagFqRD9tfNIaac5BamjhsSatjrAnsnZ" \
--data "client_secret=ekqv5XoZuMFtYms1NszEqRx03qct6BPvGeJUeptNG4y09PrY16BKT9IWezTrrhJJ" \
--data "refresh_token=GrZ1f2JvzjAZQzSXmyr1ScWbv8aMFBvzAXHBUSiILcDEG"
```
4. Access the corresponding API with the access token using the Authorization header.
5. The backend service obtains user authorization information based on the access token and returns the corresponding HTTP response.

59
plugins/wasm-go/extensions/opa/README.md
View File

@@ -1,10 +1,17 @@
# 功能说明
---
title: OPA
keywords: [higress,opa]
description: OPA 策略控制插件配置参考
---
## 功能说明
该插件实现了 `OPA` 策略控制
# 该教程使用k8s，[k8s配置文件](../../../../test/e2e/conformance/tests/go-wasm-opa.yaml)
## 运行属性
支持client `k8s,nacos,ip,route` 策略去访问
插件执行阶段：`认证阶段`
插件执行优先级：`225`
## 配置字段
@@ -18,8 +25,6 @@
| servicePort | string | 非必填 | - | 服务端口（serviceSource为`k8s,nacos,ip`必填） |
| namespace | string | 非必填 | - | 服务端口（serviceSource为`k8s,nacos`必填） |
这是一个用于OPA认证配置的表格，确保在提供所有必要的信息时遵循上述指导。
## 配置示例
```yaml
@@ -31,15 +36,15 @@ policy: example1
timeout: 5s
```
# 在宿主机上执行OPA的流程
## OPA 服务安装参考
## 启动opa服务
### 启动 OPA 服务
```shell
docker run -d --name opa -p 8181:8181 openpolicyagent/opa:0.35.0 run -s
```
## 创建opa策略
### 创建 OPA 策略
```shell
curl -X PUT '127.0.0.1:8181/v1/policies/example1' \
@@ -56,46 +61,10 @@ allow {
}'
```
## 查询策略
### 查询策略
```shell
curl -X POST '127.0.0.1:8181/v1/data/example1/allow' \
-H 'Content-Type: application/json' \
-d '{"input":{"request":{"method":"GET"}}}'
```
# 测试插件
## 打包 WASM 插件
> 在 `wasm-go` 目录下把Dockerfile文件改成`PLUGIN_NAME=opa`，然后执行以下命令
```shell
docker build -t build-wasm-opa --build-arg GOPROXY=https://goproxy.cn,direct --platform=linux/amd64 .
```
## 拷贝插件
> 在当前的目录执行以下命令，将插件拷贝当前的目录
```shell
docker cp wasm-opa:/plugin.wasm .
```
## 运行插件
> 运行前修改envoy.yaml 这两个字段 `OPA_SERVER` `OPA_PORT` 替换宿主机上的IP和端口
```shell
docker compose up
```
## 使用curl测试插件
```shell
curl http://127.0.0.1:10000/get -X GET -v
```
```shell
curl http://127.0.0.1:10000/get -X POST -v
```

58
plugins/wasm-go/extensions/opa/README_EN.md Normal file
View File

@@ -0,0 +1,58 @@
---
title: OPA
keywords: [higress,opa]
description: OPA policy control plugin configuration reference
---
## Function Description
This plugin implements `OPA` policy control.
## Running Attributes
Plugin Execution Phase: `Authentication Phase`
Plugin Execution Priority: `225`
## Configuration Fields
| Field | Data Type | Required | Default Value | Description |
|------------------|-------------|----------|---------------|-----------------------------------------------|
| policy | string | Required | - | OPA Policy |
| timeout | string | Required | - | Timeout setting for access |
| serviceSource | string | Required | - | k8s, nacos, ip, route |
| host | string | Optional | - | Service host (required if serviceSource is `ip`) |
| serviceName | string | Optional | - | Service name (required if serviceSource is `k8s,nacos,ip`) |
| servicePort | string | Optional | - | Service port (required if serviceSource is `k8s,nacos,ip`) |
| namespace | string | Optional | - | Namespace (required if serviceSource is `k8s,nacos`) |
## Configuration Example
```yaml
serviceSource: k8s
serviceName: opa
servicePort: 8181
namespace: higress-backend
policy: example1
timeout: 5s
```
## OPA Service Installation Reference
### Start OPA Service
```shell
docker run -d --name opa -p 8181:8181 openpolicyagent/opa:0.35.0 run -s
```
### Create OPA Policy
```shell
curl -X PUT '127.0.0.1:8181/v1/policies/example1' \
-H 'Content-Type: text/plain' \
-d 'package example1
import input.request
default allow = false
allow {
# HTTP method must GET
request.method == "GET"
}'
```
### Query Policy
```shell
curl -X POST '127.0.0.1:8181/v1/data/example1/allow' \
-H 'Content-Type: application/json' \
-d '{"input":{"request":{"method":"GET"}}}'
```

36
plugins/wasm-go/extensions/request-validation/README.md
View File

@@ -1,7 +1,18 @@
# 功能说明
---
title: 请求协议校验
keywords: [higress,request validation]
description: 请求协议校验插件配置参考
---
## 功能说明
`request-validation`插件用于提前验证向上游服务转发的请求。该插件使用`JSON Schema`机制进行数据验证，可以验证请求的body及header数据。
# 配置字段
## 运行属性
插件执行阶段：`认证阶段`
插件执行优先级：`220`
## 配置字段
| 名称 | 数据类型 | 填写要求 | 默认值 | 描述 |
| -------- | -------- | -------- |-----| -------- |
@@ -14,9 +25,9 @@
**校验规则对header和body是一样的，下面以body为例说明**
# 配置示例
## 配置示例
## 枚举（Enum）验证
### 枚举（Enum）验证
```yaml
body_schema:
type: object
@@ -31,7 +42,7 @@ body_schema:
default: "enum_string_1"
```
## 布尔（Boolean）验证
### 布尔（Boolean）验证
```yaml
body_schema:
type: object
@@ -43,7 +54,7 @@ body_schema:
default: true
```
## 数字范围（Number or Integer）验证
### 数字范围（Number or Integer）验证
```yaml
body_schema:
type: object
@@ -56,7 +67,7 @@ body_schema:
maximum: 10
```
## 字符串长度（String）验证
### 字符串长度（String）验证
```yaml
body_schema:
type: object
@@ -69,7 +80,7 @@ body_schema:
maxLength: 10
```
## 正则表达式（Regex）验证
### 正则表达式（Regex）验证
```yaml
body_schema:
type: object
@@ -83,7 +94,7 @@ body_schema:
pattern: "^[a-zA-Z0-9_]+$"
```
## 数组（Array）验证
### 数组（Array）验证
```yaml
body_schema:
type: object
@@ -101,7 +112,7 @@ body_schema:
default: [1, 2, 3]
```
## 多字段组合（Combined）验证
### 多字段组合（Combined）验证
```yaml
body_schema:
type: object
@@ -128,7 +139,7 @@ body_schema:
pattern: "^[a-zA-Z0-9_]+$"
```
## 自定义拒绝信息
### 自定义拒绝信息
```yaml
body_schema:
type: object
@@ -141,6 +152,3 @@ rejected_code: 403
rejected_msg: "请求被拒绝"
```
# 本地调试
参考[使用 GO 语言开发 WASM 插件](https://higress.io/zh-cn/docs/user/wasm-go#%E4%B8%89%E6%9C%AC%E5%9C%B0%E8%B0%83%E8%AF%95)

149
plugins/wasm-go/extensions/request-validation/README_EN.md Normal file
View File

@@ -0,0 +1,149 @@
---
title: Request Protocol Validation
keywords: [higress,request validation]
description: Configuration reference for request protocol validation plugin
---
## Function Description
The `request-validation` plugin is used to validate requests forwarded to upstream services in advance. This plugin utilizes the `JSON Schema` mechanism for data validation, capable of validating both the body and header data of requests.
## Execution Attributes
Plugin Execution Phase: `Authentication Phase`
Plugin Execution Priority: `220`
## Configuration Fields
| Name | Data Type | Requirements | Default Value | Description |
|-----------------|-----------|--------------|---------------|---------------------------------------------------|
| header_schema | object | Optional | - | Configuration for JSON Schema to validate request headers |
| body_schema | object | Optional | - | Configuration for JSON Schema to validate request body |
| rejected_code | number | Optional | 403 | HTTP status code returned when the request is rejected |
| rejected_msg | string | Optional | - | HTTP response body returned when the request is rejected |
| enable_swagger | bool | Optional | false | Configuration to enable Swagger documentation validation |
| enable_oas3 | bool | Optional | false | Configuration to enable OAS3 documentation validation |
**Validation rules for header and body are the same, below is an example using body.**
## Configuration Examples
### Enumeration (Enum) Validation
```yaml
body_schema:
type: object
required:
- enum_payload
properties:
enum_payload:
type: string
enum:
- "enum_string_1"
- "enum_string_2"
default: "enum_string_1"
```
### Boolean Validation
```yaml
body_schema:
type: object
required:
- boolean_payload
properties:
boolean_payload:
type: boolean
default: true
```
### Number Range (Number or Integer) Validation
```yaml
body_schema:
type: object
required:
- integer_payload
properties:
integer_payload:
type: integer
minimum: 1
maximum: 10
```
### String Length Validation
```yaml
body_schema:
type: object
required:
- string_payload
properties:
string_payload:
type: string
minLength: 1
maxLength: 10
```
### Regular Expression (Regex) Validation
```yaml
body_schema:
type: object
required:
- regex_payload
properties:
regex_payload:
type: string
minLength: 1
maxLength: 10
pattern: "^[a-zA-Z0-9_]+$"
```
### Array Validation
```yaml
body_schema:
type: object
required:
- array_payload
properties:
array_payload:
type: array
minItems: 1
items:
type: integer
minimum: 1
maximum: 10
uniqueItems: true
default: [1, 2, 3]
```
### Combined Validation
```yaml
body_schema:
type: object
required:
- boolean_payload
- array_payload
- regex_payload
properties:
boolean_payload:
type: boolean
array_payload:
type: array
minItems: 1
items:
type: integer
minimum: 1
maximum: 10
uniqueItems: true
default: [1, 2, 3]
regex_payload:
type: string
minLength: 1
maxLength: 10
pattern: "^[a-zA-Z0-9_]+$"
```
### Custom Rejection Message
```yaml
body_schema:
type: object
required:
- boolean_payload
properties:
boolean_payload:
type: boolean
rejected_code: 403
rejected_msg: "Request rejected"
```

15
plugins/wasm-go/extensions/traffic-tag/README.md
View File

@@ -1,10 +1,19 @@
# traffic-tag插件说明文档
---
title: 流量染色
keywords: [higress,traffic tag]
description: 流量染色插件配置参考
---
本文档提供了 `traffic-tag` 插件的配置选项的详细信息。此配置用于管理基于灰度发布和测试目的定义的特定规则的流量标记。
## 功能说明
`traffic-tag` 插件允许根据权重或特定请求内容通过添加特定请求头的方式对请求流量进行标记。它支持复杂的逻辑来确定如何根据用户定义的标准标记流量。
`traffic-tag` 插件允许根据权重或特定请求内容通过添加特定请求头的方式对请求流量进行染色。它支持复杂的逻辑来确定如何根据用户定义的标准染色流量。
## 运行属性
插件执行阶段：`默认阶段`
插件执行优先级：`400`
## 配置字段

118
plugins/wasm-go/extensions/traffic-tag/README_EN.md
View File

@@ -1,79 +1,76 @@
# Traffic-Tag Plugin Documentation
---
title: Traffic Tagging
keywords: [higress, traffic tag]
description: Traffic tagging plugin configuration reference
---
## Function Description
The `traffic-tag` plugin allows for tagging request traffic by adding specific request headers based on weight or specific request content. It supports complex logic to determine how to tag traffic according to user-defined standards.
This document provides detailed information about the configuration options for the `traffic-tag` plugin. This configuration is used to manage traffic tags based on rules defined for canary releases and testing purposes.
## Feature Description
The `traffic-tag` plugin allows for tagging request traffic by adding specific request headers based on weight or specific request content. It supports complex logic to determine how to tag traffic based on user-defined criteria.
## Running Attributes
Plugin execution phase: `Default Phase`
Plugin execution priority: `400`
## Configuration Fields
This section provides a detailed description of the configuration fields.
| Field Name | Type | Default | Required | Description |
|-------------------|------------------|---------|----------|-------------------------------------------------------------------|
| `conditionGroups` | array of objects | - | No | Defines groups of content-based tagging conditions, see **Condition Group Configuration**. |
| `weightGroups` | array of objects | - | No | Defines groups of weight-based tagging conditions, see **Weight Group Configuration**. |
| `defaultTagKey` | string | - | No | The default tag key name used when no conditions are matched. It only takes effect when **defaultTagValue** is also configured. |
| `defaultTagValue` | string | - | No | The default tag value used when no conditions are matched. It only takes effect when **defaultTagKey** is also configured. |
| Field Name | Type | Default Value | Required | Description |
|------------------|-------------------|---------------|----------|-------------------------------------------------------------------|
| `conditionGroups` | array of object | - | No | Defines content-based tagging condition groups, detailed structure in **Condition Group Configuration**. |
| `weightGroups` | array of object | - | No | Defines weight-based tagging condition groups, detailed structure in **Weight Group Configuration**. |
| `defaultTagKey` | string | - | No | Default tagging key name used when no conditions are matched. Only effective when **defaultTagVal** is also configured. |
| `defaultTagVal` | string | - | No | Default tagging value used when no conditions are matched. Only effective when **defaultTagKey** is also configured. |
### Condition Group Configuration
The configuration fields for each item in `conditionGroups` are described as follows:
Each item in `conditionGroups` has the following configuration fields:
| Field Name | Type | Default | Required | Description |
|--------------|--------|---------|----------|------------------------------------------------------------------|
| `headerName` | string | - | Yes | The name of the HTTP header to add or modify. |
| `headerValue`| string | - | Yes | The value of the HTTP header. |
| `logic` | string | - | Yes | The logical relation within the condition group, supports `and`, `or`, all in lowercase. |
| `conditions` | array of objects | - | Yes | Describes specific tagging conditions, detailed structure below. |
| Field Name | Type | Default Value | Required | Description |
|------------------|--------|---------------|----------|-------------------------------------------------------------------|
| `headerName` | string | - | Yes | The HTTP header name to be added or modified. |
| `headerValue` | string | - | Yes | The value of the HTTP header. |
| `logic` | string | - | Yes | Logical relationship in the condition group, supports `and`, `or`, must be in lowercase. |
| `conditions` | array of object | - | Yes | Describes specific tagging conditions, detailed structure below. |
---
The configuration fields for each item in `conditions` are described as follows:
The configuration fields for each item in `conditions` are as follows:
| Field Name | Type | Default Value | Required | Description |
|-------------------|-------------------|---------------|----------|-------------------------------------------------------------------|
| `conditionType` | string | - | Yes | Condition type, supports `header`, `parameter`, `cookie`. |
| `key` | string | - | Yes | The key of the condition. |
| `operator` | string | - | Yes | Operator, supports `equal`, `not_equal`, `prefix`, `in`, `not_in`, `regex`, `percentage`. |
| `value` | array of string | - | Yes | The value of the condition. **Only when** the operator is `in` and `not_in` multiple values are supported. |
| Field Name | Type | Default | Required | Description |
|----------------|--------|---------|----------|---------------------------------------------------------------------|
| `conditionType`| string | - | Yes | The type of condition, supports `header`, `parameter`, `cookie`. |
| `key` | string | - | Yes | The keyword for the condition. |
| `operator` | string | - | Yes | The operator, supports `equal`, `not_equal`, `prefix`, `in`, `not_in`, `regex`, `percentage`. |
| `value` | array of strings | - | Yes | The value for the condition, **only** supports multiple values for `in` and `not_in`. |
> **Note: When `operator` is `regex`, the regex engine used is [RE2](https://github.com/google/re2). For more details, see the [RE2 official documentation](https://github.com/google/re2/wiki/Syntax).**
> **Note: When the `operator` is `regex`, the regular expression engine used is [RE2](https://github.com/google/re2). For details, please refer to the [RE2 Official Documentation](https://github.com/google/re2/wiki/Syntax).**
### Weight Group Configuration
The configuration fields for each item in `weightGroups` are described as follows:
Each item in `weightGroups` has the following configuration fields:
| Field Name | Type | Default Value | Required | Description |
|------------------|-------------------|---------------|----------|-------------------------------------------------------------------|
| `headerName` | string | - | Yes | The HTTP header name to be added or modified. |
| `headerValue` | string | - | Yes | The value of the HTTP header. |
| `weight` | integer | - | Yes | Traffic weight percentage. |
| Field Name | Type | Default | Required | Description |
|--------------|---------|---------|----------|--------------------------------------------------------|
| `headerName` | string | - | Yes | The name of the HTTP header to add or modify. |
| `headerValue`| string | - | Yes | The value of the HTTP header. |
| `weight` | integer | - | Yes | The percentage of traffic weight. |
### Operator Description
| Operator | Description |
|---------------|----------------------------------------------------|
| `equal` | Exact match, values must be identical. |
| `not_equal` | Not equal match, condition met when values are different. |
| `prefix` | Prefix match, condition met when the specified value is a prefix of the actual value. |
| `in` | Inclusion match, actual value must be in the specified list. |
| `not_in` | Exclusion match, condition met when actual value is not in the specified list. |
| `regex` | Regular expression match, matched according to regex rules. |
| `percentage` | Percentage match, principle: `hash(get(key)) % 100 < value`, condition met when true. |
### Operator Descriptions
| Operator | Description |
|--------------|---------------------------------------------------------|
| `equal` | Exact match, the value must be completely equal. |
| `not_equal` | Not equal match, condition is met when values are not equal. |
| `prefix` | Prefix match, condition is met when the specified value is a prefix of the actual value. |
| `in` | Inclusion match, actual value needs to be in the specified list. |
| `not_in` | Exclusion match, condition is met when the actual value is not in the specified list. |
| `regex` | Regex match, matches according to regex rules. |
| `percentage` | Percentage match, the principle: `hash(get(key)) % 100 < value` satisfies the condition. |
> **Tip: Differences between `percentage` and `weight`**
> **Tip: About the difference between `percentage` and `weight`**
>
> - **`percentage` operator**: Used in condition expressions, based on a specified percentage and specified key-value pair to determine whether to perform a certain action. For the same key-value pair, the result of multiple matches is idempotent, i.e., if it hits the condition this time, it will also hit next time.
> - **`weight` field**: Used to define the traffic weight of different treatment paths. In weight-based traffic marking, `weight` determines the proportion of traffic that a path should receive. Unlike `percentage`, which is based on a fixed comparison basis, `weight` involves a random distribution of traffic weight, so the same request may match multiple outcomes in multiple matches.
> - **`percentage` operator**: Used in conditional expressions, it determines whether to perform a certain operation based on specified percentages and key-value pairs. For the same key-value pair, the result of multiple matches is idempotent, meaning if a condition is hit this time, it will hit it next time as well.
> - **`weight` field**: Used to define traffic weights for different processing paths. In weight-based traffic tagging, `weight` determines the proportion of traffic a particular path should receive. Unlike `percentage`, since there is no fixed comparison basis and it is based on random weight distribution, multiple matches for the same request may yield multiple results.
>
> When using `percentage` for condition matching, each request is evaluated to see if it meets a specific percentage condition; whereas `weight` statically randomly allocates the proportion of overall traffic.
> When using `percentage` for conditional matching, it assesses whether each request meets specific percentage conditions; while `weight` is a static random allocation of overall traffic distribution.
## Configuration Example
**Example 1: Content-Based Matching**
According to the following configuration, requests that meet both the condition of the request header `role` being one of `user`, `viewer`, or `editor` and the query parameter `foo=bar` will have the request header `x-mse-tag: gray` added. Since `defaultTagKey` and `defaultTagVal` are configured, when no conditions are matched, the request header `x-mse-tag: base` will be added to the request.
**Example 1: Content-based Matching**
According to the configuration below, requests where the request header `role` has a value of `user`, `viewer`, or `editor` and contain query parameter `foo=bar` will have the request header `x-mse-tag: gray` added. Since `defaultTagKey` and `defaultTagVal` are configured, when no conditions are matched, the request will have the request header `x-mse-tag: base` added.
```yaml
defaultTagKey: x-mse-tag
@@ -94,15 +91,14 @@ conditionGroups:
key: foo
operator: equal
value:
- bar
- bar
```
**Example 2: Weight-Based Matching**
According to the following configuration, requests have a 30% chance of having the request header `x-mse-tag: gray` added, a 30% chance of having `x-mse-tag: blue` added, and a 40% chance of not having any header added.
**Example 2: Weight-based Matching**
According to the configuration below, there is a 30% chance that the request will have the request header `x-mse-tag: gray` added, a 30% chance it will have `x-mse-tag: blue` added, and a 40% chance that no header will be added.
```yaml
# The total weight sums to 100, the unconfigured 40 weight will not add a header
# The total weight is 100; the 40 weight not configured in the example will not add a header.
weightGroups:
- headerName: x-mse-tag
headerValue: gray

213
plugins/wasm-go/extensions/transformer/README.md
View File

@@ -1,8 +1,19 @@
# 功能说明
---
title: 请求响应转换
keywords: [higress,transformer]
description: 请求响应转换插件配置参考
---
## 功能说明
`transformer` 插件可以对请求/响应头、请求查询参数、请求/响应体参数进行转换，支持的转换操作类型包括删除、重命名、更新、添加、追加、映射、去重。
## 运行属性
# 配置字段
插件执行阶段：`认证阶段`
插件执行优先级：`410`
## 配置字段
| 名称 | 数据类型 | 填写要求 | 默认值 | 描述 |
| :----: | :----: | :----: | :----: | -------- |
@@ -52,7 +63,7 @@
# 转换操作类型
## 转换操作类型
| 操作类型 | key 字段含义 | value 字段含义 | 描述 |
| :----: | :----: | :----: | ------------------------------------------------------------ |
@@ -67,11 +78,97 @@
# 配置示例
## 配置示例
## Request Transformer
### 实现基于Body参数路由
### 转换请求头部
配置示例：
```yaml
reqRules:
- operate: map
headers:
- fromKey: userId
toKey: x-user-id
mapSource: body
```
此规则将请求body中的`userId`解析出后，设置到请求Header`x-user-id`中，这样就可以基于Higress请求Header匹配路由的能力来实现基于Body参数的路由了。
此配置同时支持`application/json`和`application/x-www-form-urlencoded`两种类型的请求Body。
举例来说：
**对于application/json类型的body**
```bash
curl localhost -d '{"userId":12, "userName":"johnlanni"}' -H 'content-type:application/json'
```
将从json中提取出`userId`字段的值，设置到`x-user-id`中，后端服务收到的请求头将增加:`x-usr-id: 12`。
因为在插件新增这个Header后，网关将重新计算路由，所以可以实现网关路由配置根据这个请求头来匹配路由到特定的目标服务。
**对于application/x-www-form-urlencoded类型的body**
```bash
curl localhost -d 'userId=12&userName=johnlanni'
```
将从`k1=v1&k2=v2`这样的表单格式中提取出`userId`字段的值，设置到`x-user-id`中，后端服务收到的请求头将增加:`x-usr-id: 12`。
因为在插件新增这个Header后，网关将重新计算路由，所以可以实现网关路由配置根据这个请求头来匹配路由到特定的目标服务。
#### json path 支持
可以根据 [GJSON Path 语法](https://github.com/tidwall/gjson/blob/master/SYNTAX.md)，从复杂的 json 中提取出字段。
比较常用的操作举例，对于以下 json:
```json
{
"name": {"first": "Tom", "last": "Anderson"},
"age":37,
"children": ["Sara","Alex","Jack"],
"fav.movie": "Deer Hunter",
"friends": [
{"first": "Dale", "last": "Murphy", "age": 44, "nets": ["ig", "fb", "tw"]},
{"first": "Roger", "last": "Craig", "age": 68, "nets": ["fb", "tw"]},
{"first": "Jane", "last": "Murphy", "age": 47, "nets": ["ig", "tw"]}
]
}
```
可以实现这样的提取:
```text
name.last "Anderson"
name.first "Tom"
age 37
children ["Sara","Alex","Jack"]
children.0 "Sara"
children.1 "Alex"
friends.1 {"first": "Roger", "last": "Craig", "age": 68}
friends.1.first "Roger"
```
现在如果想从上面这个 json 格式的 body 中提取出 friends 中第二项的 first 字段，来设置到 Header `x-first-name` 中，同时抽取 last 字段，来设置到 Header `x-last-name` 中，则可以使用这份插件配置:
```yaml
reqRules:
- operate: map
headers:
- fromKey: friends.1.first
toKey: x-first-name
- fromKey: friends.1.last
toKey: x-last-name
mapSource: body
```
### Request Transformer
#### 转换请求头部
```yaml
reqRules:
@@ -139,7 +236,7 @@ $ curl -v console.higress.io/get -H 'host: foo.bar.com' \
}
```
### 转换请求查询参数
#### 转换请求查询参数
```yaml
reqRules:
@@ -193,7 +290,7 @@ $ curl -v "console.higress.io/get?k1=v11&k1=v12&k2=v2"
}
```
### 转换请求体
#### 转换请求体
```yaml
reqRules:
@@ -313,11 +410,11 @@ $ curl -v -X POST console.higress.io/post -H 'host: foo.bar.com' \
}
```
## Response Transformer
### Response Transformer
与 Request Transformer 类似，在此仅说明转换 JSON 形式的请求/响应体时的注意事项：
### key 嵌套 `.`
#### key 嵌套 `.`
1.通常情况下，指定的 key 中含有 `.` 表示嵌套含义，如下：
@@ -365,7 +462,7 @@ $ curl -v console.higress.io/get
}
```
### 访问数组元素 `.index`
#### 访问数组元素 `.index`
可以通过数组下标 `array.index 访问数组元素，下标从 0 开始：
@@ -449,7 +546,7 @@ $ curl -v -X POST console.higress.io/post \
}
```
### 遍历数组元素 `.#`
#### 遍历数组元素 `.#`
可以使用 `array.#` 对数组进行遍历操作：
@@ -502,95 +599,3 @@ $ curl -v -X POST console.higress.io/post \
...
}
```
# 特殊用法：实现基于Body参数路由
**Note**
> 需要数据面的proxy wasm版本大于等于0.2.100
> 编译时，需要带上版本的tag，例如：`tinygo build -o main.wasm -scheduler=none -target=wasi -gc=custom -tags="custommalloc nottinygc_finalizer proxy_wasm_version_0_2_100" ./`
配置示例：
```yaml
reqRules:
- operate: map
headers:
- fromKey: userId
toKey: x-user-id
mapSource: body
```
此规则将请求body中的`userId`解析出后，设置到请求Header`x-user-id`中，这样就可以基于Higress请求Header匹配路由的能力来实现基于Body参数的路由了。
此配置同时支持`application/json`和`application/x-www-form-urlencoded`两种类型的请求Body。
举例来说：
**对于application/json类型的body**
```bash
curl localhost -d '{"userId":12, "userName":"johnlanni"}' -H 'content-type:application/json'
```
将从json中提取出`userId`字段的值，设置到`x-user-id`中，后端服务收到的请求头将增加:`x-usr-id: 12`。
因为在插件新增这个Header后，网关将重新计算路由，所以可以实现网关路由配置根据这个请求头来匹配路由到特定的目标服务。
**对于application/x-www-form-urlencoded类型的body**
```bash
curl localhost -d 'userId=12&userName=johnlanni'
```
将从`k1=v1&k2=v2`这样的表单格式中提取出`userId`字段的值，设置到`x-user-id`中，后端服务收到的请求头将增加:`x-usr-id: 12`。
因为在插件新增这个Header后，网关将重新计算路由，所以可以实现网关路由配置根据这个请求头来匹配路由到特定的目标服务。
## json path 支持
可以根据 [GJSON Path 语法](https://github.com/tidwall/gjson/blob/master/SYNTAX.md)，从复杂的 json 中提取出字段。
比较常用的操作举例，对于以下 json:
```json
{
"name": {"first": "Tom", "last": "Anderson"},
"age":37,
"children": ["Sara","Alex","Jack"],
"fav.movie": "Deer Hunter",
"friends": [
{"first": "Dale", "last": "Murphy", "age": 44, "nets": ["ig", "fb", "tw"]},
{"first": "Roger", "last": "Craig", "age": 68, "nets": ["fb", "tw"]},
{"first": "Jane", "last": "Murphy", "age": 47, "nets": ["ig", "tw"]}
]
}
```
可以实现这样的提取:
```text
name.last "Anderson"
name.first "Tom"
age 37
children ["Sara","Alex","Jack"]
children.0 "Sara"
children.1 "Alex"
friends.1 {"first": "Roger", "last": "Craig", "age": 68}
friends.1.first "Roger"
```
现在如果想从上面这个 json 格式的 body 中提取出 friends 中第二项的 first 字段，来设置到 Header `x-first-name` 中，同时抽取 last 字段，来设置到 Header `x-last-name` 中，则可以使用这份插件配置:
```yaml
reqRules:
- operate: map
headers:
- fromKey: friends.1.first
toKey: x-first-name
- fromKey: friends.1.last
toKey: x-last-name
mapSource: body
```

522
plugins/wasm-go/extensions/transformer/README_EN.md Normal file
View File

@@ -0,0 +1,522 @@
---
title: Request Response Transformation
keywords: [higress,transformer]
description: Request response transformation plugin configuration reference
---
## Function Description
The `transformer` plugin can transform request/response headers, request query parameters, and request/response body parameters. Supported transformation operation types include deletion, renaming, updating, adding, appending, mapping, and deduplication.
## Execution Attributes
Plugin execution phase: `authentication phase`
Plugin execution priority: `410`
## Configuration Fields
| Name | Data Type | Fill Requirement | Default Value | Description |
| :----: | :----: | :----: | :----: | -------- |
| reqRules | string | Optional, at least one of reqRules or respRules must be filled | - | Request transformer configuration, specifying the transformation operation type and rules for transforming request headers, request query parameters, and request body |
| respRules | string | Optional, at least one of reqRules or respRules must be filled | - | Response transformer configuration, specifying the transformation operation type and rules for transforming response headers and response body |
The configuration fields for each item in `reqRules` and `respRules` are as follows:
| Name | Data Type | Fill Requirement | Default Value | Description |
| :----: | :----: | :----: | :----: | -------- |
| operate | string | Required, optional values are `remove`, `rename`, `replace`, `add`, `append`, `map`, `dedupe` | - | Specifies the transformation operation type. Supported operation types include remove (remove), rename (rename), replace (replace), add (add), append (append), map (map), dedupe (dedupe). When there are multiple transformation rules of different types, they are executed in the order of the above operation types. |
| mapSource | string | Optional, optional values are `headers`, `querys`, `body` | - | Valid only when operate is `map`. Specifies the mapping source. If this field is not filled, the default mapping source is itself. |
| headers | array of object | Optional | - | Specifies transformation rules for request/response headers. |
| querys | array of object | Optional | - | Specifies transformation rules for request query parameters. |
| body | array of object | Optional | - | Specifies transformation rules for request/response body parameters. Request body transformations allow content-types of `application/json`, `application/x-www-form-urlencoded`, and `multipart/form-data` while response body transformations only allow content-type of `application/json`. |
The configuration fields for each item in `headers`, `querys`, `body` are as follows:
| Name | Data Type | Fill Requirement | Default Value | Description |
| :----: | :----: | :----: | -------- | --------------------------------------------------- |
| key | string | Optional | - | Used when operate is `remove`, see [Transformation Operation Types](#转换操作类型) for details. |
| oldKey | string | Optional | - | Used when operate is `rename`, see [Transformation Operation Types](#转换操作类型) for details. |
| newKey | string | Optional | - | Used when operate is `rename`, see [Transformation Operation Types](#转换操作类型) for details. |
| key | string | Optional | - | Used when operate is `replace`, see [Transformation Operation Types](#转换操作类型) for details. |
| newValue | string | Optional | - | Used when operate is `replace`, see [Transformation Operation Types](#转换操作类型) for details. |
| key | string | Optional | - | Used when operate is `add`, see [Transformation Operation Types](#转换操作类型) for details. |
| value | string | Optional | - | Used when operate is `add`, see [Transformation Operation Types](#转换操作类型) for details. |
| key | string | Optional | - | Used when operate is `append`, see [Transformation Operation Types](#转换操作类型) for details. |
| appendValue | string | Optional | - | Used when operate is `append`, see [Transformation Operation Types](#转换操作类型) for details. |
| fromKey | string | Optional | - | Used when operate is `map`, see [Transformation Operation Types](#转换操作类型) for details. |
| toKey | string | Optional | - | Used when operate is `map`, see [Transformation Operation Types](#转换操作类型) for details. |
| key | string | Optional | - | Used when operate is `dedupe`, see [Transformation Operation Types](#转换操作类型) for details. |
| strategy | string | Optional | - | Used when operate is `dedupe`, see [Transformation Operation Types](#转换操作类型) for details. |
| value_type | string | Optional, optional values are `object`, `boolean`, `number`, `string` | string | When `content-type: application/json`, this field specifies the value type of request/response body parameters. |
| host_pattern | string | Optional | - | Specifies the request hostname matching rule. Valid when transformation operation type is `replace`, `add`, `append`. |
| path_pattern | string | Optional | - | Specifies the request path matching rule. Valid when transformation operation type is `replace`, `add`, `append`. |
Note:
* `request transformer` supports the following transformation objects: request headers, request query parameters, request body (application/json, application/x-www-form-urlencoded, multipart/form-data).
* `response transformer` supports the following transformation objects: response headers, response body (application/json).
* The plugin supports bidirectional conversion capability, meaning that a single plugin can perform transformations on both requests and responses.
* The execution order of transformation operation types is the order written in the configuration file, e.g., remove → rename → replace → add → append → map → dedupe or dedupe → map → append → add → replace → rename → remove.
* When the transformation object is headers, `key` is case-insensitive. When headers are operated and are `rename` or `map`, `value` is also case-insensitive (as this field has a key meaning). However, `key` and `value` fields in querys and body are case-sensitive.
* `value_type` is only effective for content type application/json for request/response bodies.
* `host_pattern` and `path_pattern` support [RE2 syntax](https://pkg.go.dev/regexp/syntax), valid only for `replace`, `add`, `append` operations. In a transformation rule, only one of the two can be optionally filled. If both are filled, then `host_pattern` takes effect while `path_pattern` becomes ineffective.
## Transformation Operation Types
| Operation Type | Key Field Meaning | Value Field Meaning | Description |
| :----: | :----: | :----: | ------------------------------------------------------------ |
| Remove remove | Target key | Not required | If the specified `key` exists, delete it; otherwise, no operation |
| Rename rename | Target oldKey | New key name newKey | If the specified `oldKey:value` exists, rename its key to `newKey`, resulting in `newKey:value`; otherwise, no operation |
| Replace replace | Target key | New value newValue | If the specified `key:value` exists, update its value to `newValue`, resulting in `key:newValue`; otherwise, no operation |
| Add add | Added key | Added value | If the specified `key:value` does not exist, add it; otherwise, no operation |
| Append append | Target key | Appending value appendValue | If the specified `key:value` exists, append appendValue to get `key:[value..., appendValue]`; otherwise, it is equivalent to performing add operation, resulting in `key:appendValue`. |
| Map map | Mapping source fromKey | Mapping target toKey | If the specified `fromKey:fromValue` exists, map its value fromValue to the value of toKey, resulting in `toKey:fromValue`, while retaining `fromKey:fromValue` (note: if toKey already exists, its value will be overwritten); otherwise, no operation. |
| Deduplicate dedupe | Target key | Specified deduplication strategy strategy | `strategy` optional values include: <br>`RETAIN_UNIQUE`: Retain all unique values in order, e.g., `k1:[v1,v2,v3,v3,v2,v1]`, deduplication results in `k1:[v1,v2,v3]`. <br>`RETAIN_LAST`: Retain the last value, e.g., `k1:[v1,v2,v3]`, deduplication results in `k1:v3`. <br>`RETAIN_FIRST` (default): Retain the first value, e.g., `k1:[v1,v2,v3]`, deduplication results in `k1:v1`. <br>(Note: When deduplication results in only one element v1, the key-value pair becomes `k1:v1`, not `k1:[v1]`.) |
## Configuration Example
### Implement Routing Based on Body Parameters
Configuration example:
```yaml
reqRules:
- operate: map
headers:
- fromKey: userId
toKey: x-user-id
mapSource: body
```
This rule extracts the `userId` from the request body and sets it in the request header `x-user-id`. This allows routing based on body parameters using Higress's ability to match on request headers.
This configuration supports both `application/json` and `application/x-www-form-urlencoded` types of request bodies.
For example:
**For application/json type body**
```bash
curl localhost -d '{"userId":12, "userName":"johnlanni"}' -H 'content-type:application/json'
```
The value of the `userId` field will be extracted from the JSON and set to `x-user-id`. The backend service will receive a request header with: `x-user-id: 12`.
After the plugin adds this header, the gateway will recalculate the routes, allowing the routing configuration to match the specific target service based on this request header.
**For application/x-www-form-urlencoded type body**
```bash
curl localhost -d 'userId=12&userName=johnlanni'
```
The value of the `userId` field will be extracted from the form format `k1=v1&k2=v2` and set to `x-user-id`. The backend service will receive a request header with: `x-user-id: 12`.
After the plugin adds this header, the gateway will recalculate the routes, allowing the routing configuration to match the specific target service based on this request header.
#### JSON Path Support
You can extract fields from complex JSON according to [GJSON Path syntax](https://github.com/tidwall/gjson/blob/master/SYNTAX.md).
Common operations include, for the following JSON:
```json
{
"name": {"first": "Tom", "last": "Anderson"},
"age": 37,
"children": ["Sara","Alex","Jack"],
"fav.movie": "Deer Hunter",
"friends": [
{"first": "Dale", "last": "Murphy", "age": 44, "nets": ["ig", "fb", "tw"]},
{"first": "Roger", "last": "Craig", "age": 68, "nets": ["fb", "tw"]},
{"first": "Jane", "last": "Murphy", "age": 47, "nets": ["ig", "tw"]}
]
}
```
You can achieve such extractions:
```text
name.last "Anderson"
name.first "Tom"
age 37
children ["Sara","Alex","Jack"]
children.0 "Sara"
children.1 "Alex"
friends.1 {"first": "Roger", "last": "Craig", "age": 68}
friends.1.first "Roger"
```
Now, if you want to extract the `first` field from the second item in `friends` from the above JSON formatted body and set it to the header `x-first-name`, while also extracting the `last` field to set it to the header `x-last-name`, you can use this plugin configuration:
```yaml
reqRules:
- operate: map
headers:
- fromKey: friends.1.first
toKey: x-first-name
- fromKey: friends.1.last
toKey: x-last-name
mapSource: body
```
### Request Transformer
#### Transforming Request Headers
```yaml
reqRules:
- operate: remove
headers:
- key: X-remove
- operate: rename
headers:
- oldKey: X-not-renamed
newKey: X-renamed
- operate: replace
headers:
- key: X-replace
newValue: replaced
- operate: add
headers:
- key: X-add-append
value: host-\$1
host_pattern: ^(.*)\.com$
- operate: append
headers:
- key: X-add-append
appendValue: path-\$1
path_pattern: ^.*?\/(\w+)[\?]{0,1}.*$
- operate: map
headers:
- fromKey: X-add-append
toKey: X-map
- operate: dedupe
headers:
- key: X-dedupe-first
strategy: RETAIN_FIRST
- key: X-dedupe-last
strategy: RETAIN_LAST
- key: X-dedupe-unique
strategy: RETAIN_UNIQUE
```
Send Request
```bash
$ curl -v console.higress.io/get -H 'host: foo.bar.com' \
-H 'X-remove: exist' -H 'X-not-renamed:test' -H 'X-replace:not-replaced' \
-H 'X-dedupe-first:1' -H 'X-dedupe-first:2' -H 'X-dedupe-first:3' \
-H 'X-dedupe-last:a' -H 'X-dedupe-last:b' -H 'X-dedupe-last:c' \
-H 'X-dedupe-unique:1' -H 'X-dedupe-unique:2' -H 'X-dedupe-unique:3' \
-H 'X-dedupe-unique:3' -H 'X-dedupe-unique:2' -H 'X-dedupe-unique:1'
# httpbin response result
{
"args": {},
"headers": {
...
"X-Add-Append": "host-foo.bar,path-get",
...
"X-Dedupe-First": "1",
"X-Dedupe-Last": "c",
"X-Dedupe-Unique": "1,2,3",
...
"X-Map": "host-foo.bar,path-get",
"X-Renamed": "test",
"X-Replace": "replaced"
},
...
}
```
#### Transforming Request Query Parameters
```yaml
reqRules:
- operate: remove
querys:
- key: k1
- operate: rename
querys:
- oldKey: k2
newKey: k2-new
- operate: replace
querys:
- key: k2-new
newValue: v2-new
- operate: add
querys:
- key: k3
value: v31-\$1
path_pattern: ^.*?\/(\w+)[\?]{0,1}.*$
- operate: append
querys:
- key: k3
appendValue: v32
- operate: map
querys:
- fromKey: k3
toKey: k4
- operate: dedupe
querys:
- key: k4
strategy: RETAIN_FIRST
```
Send Request
```bash
$ curl -v "console.higress.io/get?k1=v11&k1=v12&k2=v2"
# httpbin response result
{
"args": {
"k2-new": "v2-new",
"k3": [
"v31-get",
"v32"
],
"k4": "v31-get"
},
...
"url": "http://foo.bar.com/get?k2-new=v2-new&k3=v31-get&k3=v32&k4=v31-get"
}
```
#### Transforming Request Body
```yaml
reqRules:
- operate: remove
body:
- key: a1
- operate: rename
body:
- oldKey: a2
newKey: a2-new
- operate: replace
body:
- key: a3
newValue: t3-new
value_type: string
- operate: add
body:
- key: a1-new
value: t1-new
value_type: string
- operate: append
body:
- key: a1-new
appendValue: t1-\$1-append
value_type: string
host_pattern: ^(.*)\.com$
- operate: map
body:
- fromKey: a1-new
toKey: a4
- operate: dedupe
body:
- key: a4
strategy: RETAIN_FIRST
```
Send Requests:
**1. Content-Type: application/json**
```bash
$ curl -v -X POST console.higress.io/post -H 'host: foo.bar.com' \
-H 'Content-Type: application/json' -d '{"a1":"t1","a2":"t2","a3":"t3"}'
# httpbin response result
{
...
"headers": {
...
"Content-Type": "application/json",
...
},
"json": {
"a1-new": [
"t1-new",
"t1-foo.bar-append"
],
"a2-new": "t2",
"a3": "t3-new",
"a4": "t1-new"
},
...
}
```
**2. Content-Type: application/x-www-form-urlencoded**
```bash
$ curl -v -X POST console.higress.io/post -H 'host: foo.bar.com' \
-d 'a1=t1&a2=t2&a3=t3'
# httpbin response result
{
...
"form": {
"a1-new": [
"t1-new",
"t1-foo.bar-append"
],
"a2-new": "t2",
"a3": "t3-new",
"a4": "t1-new"
},
"headers": {
...
"Content-Type": "application/x-www-form-urlencoded",
...
},
...
}
```
**3. Content-Type: multipart/form-data**
```bash
$ curl -v -X POST console.higress.io/post -H 'host: foo.bar.com' \
-F a1=t1 -F a2=t2 -F a3=t3
# httpbin response result
{
...
"form": {
"a1-new": [
"t1-new",
"t1-foo.bar-append"
],
"a2-new": "t2",
"a3": "t3-new",
"a4": "t1-new"
},
"headers": {
...
"Content-Type": "multipart/form-data; boundary=------------------------1118b3fab5afbc4e",
...
},
...
}
```
### Response Transformer
Similar to Request Transformer, this only describes the precautions for transforming JSON-formatted request/response bodies:
#### Key Nesting `.`
1. In general, a key containing `.` indicates a nested meaning, as follows:
```yaml
respRules:
- operate: add
body:
- key: foo.bar
value: value
```
```bash
$ curl -v console.higress.io/get
# httpbin response result
{
...
"foo": {
"bar": "value"
},
...
}
```
2. When using `\.` to escape `.` in the key, it indicates a non-nested meaning, as follows:
> When enclosing a string with double quotes, use `\\.` for escaping
```yaml
respRules:
- operate: add
body:
- key: foo\.bar
value: value
```
```bash
$ curl -v console.higress.io/get
# httpbin response result
{
...
"foo.bar": "value",
...
}
```
#### Accessing Array Elements `.index`
You can access array elements by their index `array.index`, where the index starts from 0:
```json
{
"users": [
{
"123": { "name": "zhangsan", "age": 18 }
},
{
"456": { "name": "lisi", "age": 19 }
}
]
}
```
1. Remove the first element of `user`:
```yaml
reqRules:
- operate: remove
body:
- key: users.0
```
```bash
$ curl -v -X POST console.higress.io/post \
-H 'Content-Type: application/json' \
-d '{"users":[{"123":{"name":"zhangsan"}},{"456":{"name":"lisi"}}]}'
# httpbin response result
{
...
"json": {
"users": [
{
"456": {
"name": "lisi"
}
}
]
},
...
}
```
2. Rename the key `123` of the first element of `users` to `msg`:
```yaml
reqRules:
- operate: rename
body:
- oldKey: users.0.123
newKey: users.0.first
```
```bash
$ curl -v -X POST console.higress.io/post \
-H 'Content-Type: application/json' \
-d '{"users":[{"123":{"name":"zhangsan"}},{"456":{"name":"lisi"}}]}'
# httpbin response result
{
...
"json": {
"users": [
{
"msg": {
"name": "zhangsan"
}
},
{
"456": {
"name": "lisi"
}
}
]
},
...
}
```
#### Iterating Array Elements `.#`
You can use `array.#` to iterate over an array:
> ❗️This operation can only be used in replace, do not attempt this operation in other transformations to avoid unpredictable results
```json
{
"users": [
{
"name": "zhangsan",
"age": 18
},
{
"name": "lisi",
"age": 19
}
]
}
```
```yaml
reqRules:
- operate: replace
body:
- key: users.#.age
newValue: 20
```
```bash
$ curl -v -X POST console.higress.io/post \
-H 'Content-Type: application/json' \
-d '{"users":[{"name":"zhangsan","age":18},{"name":"lisi","age":19}]}'
# httpbin response result
{
...
"json": {
"users": [
{
"age": "20",
"name": "zhangsan"
},
{
"age": "20",
"name": "lisi"
}
]
},
...
}
```

45
plugins/wasm-go/extensions/waf/README.md
View File

@@ -1,13 +1,25 @@
# 功能说明
---
title: WAF
keywords: [higress,waf]
description: WAF 插件配置参考
---
## 功能说明
waf插件实现了基于ModSecurity的规则防护引擎，可以根据用户配置的规则屏蔽可疑请求，并支持OWASP CRS，为站点提供基础的防护功能。
# 配置字段
## 运行属性
插件执行阶段：`授权阶段`
插件执行优先级：`330`
## 配置字段
| 名称 | 数据类型 | 填写要求 | 默认值 | 描述 |
| -------- | -------- | -------- | -------- | -------- |
| useCRS | bool | 选填 | false | 是否开启OWASP CRS，详情可参考[coreruleset](https://github.com/coreruleset/coreruleset/tree/v3.3.2) |
| secRules | array of string | 选填 | - | 用户自定义的waf防护规则，语法规则可参考[ModSecurity中文手册](http://www.modsecurity.cn/chm/) |
# 配置示例
## 配置示例
```yaml
useCRS: true
secRules:
@@ -23,30 +35,3 @@ secRules:
curl http://example.com/admin
curl http://example.com -d "maliciouspayload"
```
# 对特定路由或域名开启
```yaml
useCRS: true
secRules:
- "SecDebugLogLevel 3"
- "SecRuleEngine On"
- "SecAction \"id:100,phase:1,pass\""
- "SecRule REQUEST_URI \"@streq /admin\" \"id:101,phase:1,t:lowercase,deny\""
- "SecRule REQUEST_BODY \"@rx maliciouspayload\" \"id:102,phase:2,t:lowercase,deny\""
_rules_:
- _match_route_:
- "route-1"
secRules:
- "SecDebugLogLevel 3"
- "SecRuleEngine On"
- "SecAction \"id:102,phase:1,deny\""
- _match_domain_:
- "*.example.com"
- test.com
secRules:
- "SecDebugLogLevel 3"
- "SecRuleEngine On"
- "SecAction \"id:102,phase:1,pass\""
```
此例 `_match_route_` 中指定的 `route-1` 即在创建网关路由时填写的路由名称，当匹配到这两个路由时，将使用此段配置； 此例 `_match_domain_` 中指定的 `*.example.com` 和 `test.com` 用于匹配请求的域名，当发现域名匹配时，将使用此段配置； 配置的匹配生效顺序，将按照 `_rules_` 下规则的排列顺序，匹配第一个规则后生效对应配置，后续规则将被忽略。

39
plugins/wasm-go/extensions/waf/README_EN.md Normal file
View File

@@ -0,0 +1,39 @@
---
title: WAF
keywords: [higress,waf]
description: WAF plugin configuration reference
---
## Function Description
The waf plugin implements a ModSecurity-based rule protection engine, which can block suspicious requests based on user-defined rules, and supports OWASP CRS, providing basic protection features for the site.
## Running Attributes
Plugin execution phase: `authorization phase`
Plugin execution priority: `330`
## Configuration Fields
| Name | Data Type | Filling Requirements | Default Value | Description |
|----------|--------------------|----------------------|---------------|-----------------------------------------------------------------------------|
| useCRS | bool | Optional | false | Whether to enable OWASP CRS, for details refer to [coreruleset](https://github.com/coreruleset/coreruleset/tree/v3.3.2) |
| secRules | array of string | Optional | - | User-defined WAF protection rules, syntax rules can refer to [ModSecurity Chinese Manual](http://www.modsecurity.cn/chm/) |
## Configuration Example
```yaml
useCRS: true
secRules:
- "SecDebugLogLevel 3"
- "SecRuleEngine On"
- "SecAction \"id:100,phase:1,pass\""
- "SecRule REQUEST_URI \"@streq /admin\" \"id:101,phase:1,t:lowercase,deny\""
- "SecRule REQUEST_BODY \"@rx maliciouspayload\" \"id:102,phase:2,t:lowercase,deny\""
```
Based on this configuration, the following requests will be prohibited from access:
```bash
curl http://example.com/admin
curl http://example.com -d "maliciouspayload"
```