OpenWAF
第一个全方位开源的Web应用防护系统（WAF），更全面的防护功能，更多样的防护策略
- Name
- Version
- Synopsis
- Description
- Installation
- Community
- Bugs and Patches
- TODO
- Changes
- Copyright and License
- Modules Configuration Directives
- Nginx Variables
- SecRules
- Donation
This document describes OpenWAF v0.0.6 released on 26 Dec 2017.
Docker Version
- titansec/openwaf:0.0.1.161130_beta
SHA: 596dee9d2b9ce44d59dc445141f72b3607f9fbe6
https://github.com/titansec/OpenWAF/tree/596dee9d2b9ce44d59dc445141f72b3607f9fbe6 - titansec/openwaf:0.0.3.170103_beta
SHA: 28ce1556250301f26f31b46d9cd9dde5a3b3f03f
https://github.com/titansec/OpenWAF/tree/28ce1556250301f26f31b46d9cd9dde5a3b3f03f - titansec/openwaf:v0.0.4
SHA: cdef1e0eac4f12c374feb6ac3c3c465610daaf6c
https://github.com/titansec/OpenWAF/tree/cdef1e0eac4f12c374feb6ac3c3c465610daaf6c - titansec/openwaf:v0.0.5(titansec/openwaf:latest)
https://github.com/titansec/OpenWAF/tree/f9a3a6cfb8b75b1f4af9f44477a5b64337a7da47
#nginx.conf
lua_package_path '/twaf/?.lua;;';
init_by_lua_file /twaf/app/twaf_init.lua;
lua_shared_dict twaf_shm 50m;
upstream test {
server 0.0.0.1; #just an invalid address as a place holder
balancer_by_lua_file twaf_balancer.lua;
}
server {
listen 443 ssl;
server_name _;
ssl_certificate_by_lua_file twaf_ssl_cert.lua;
rewrite_by_lua_file /twaf/app/twaf_rewrite.lua;
access_by_lua_file /twaf/app/twaf_access.lua;
header_filter_by_lua_file /twaf/app/twaf_header_filter.lua;
body_filter_by_lua_file /twaf/app/twaf_body_filter.lua
log_by_lua_file /twaf/app/twaf_log.lua;
set $twaf_https 1;
set $twaf_upstream_server "";
ssl_certificate nginx.crt;
ssl_certificate_key nginx.key;
location / {
lua_need_request_body on;
proxy_pass $twaf_upstream_server;
}
}
server {
listen 80;
server_name _;
rewrite_by_lua_file /twaf/app/twaf_rewrite.lua;
access_by_lua_file /twaf/app/twaf_access.lua;
header_filter_by_lua_file /twaf/app/twaf_header_filter.lua;
body_filter_by_lua_file /twaf/app/twaf_body_filter.lua
log_by_lua_file /twaf/app/twaf_log.lua;
set $twaf_upstream_server "";
location / {
lua_need_request_body on;
proxy_pass $twaf_upstream_server;
}
} #default_config-json
#main_safe_policy-jsonOpenWAF是第一个全方位开源的Web应用防护系统（WAF），他基于nginx_lua API分析HTTP请求信息。OpenWAF由行为分析引擎和规则引擎两大功能引擎构成。其中规则引擎主要对单个请求进行分析，行为分析引擎主要负责跨请求信息追踪。
规则引擎的启发来自modsecurity及freewaf(lua-resty-waf)，将ModSecurity的规则机制用lua实现。基于规则引擎可以进行协议规范，自动工具，注入攻击，跨站攻击，信息泄露，异常请求等安全防护，支持动态添加规则，及时修补漏洞。
行为分析引擎包含基于频率的模糊识别，防恶意爬虫，人机识别等防探测模块，防CSRF，防CC，防提权，文件上传防护等防攻击模块，cookie防篡改，防盗链，自定义响应头，攻击响应页面等防信息泄露模块。
除了两大引擎之外，还包含统计，日志，攻击响应页面，接入规则等基础模块。除了已有的功能模块，OpenWAF还支持动态修改配置， 动态添加第三方模块，使得在不重启引擎中断业务的条件下，升级防护。
OpenWAF支持将上述功能封装为策略，不同的web application应用不同的策略来防护。将来还会打造云平台，策略还可分享供他人参考。
基础模块如下:
功能模块如下:
- 规则引擎 openwaf_rule_engine
- 攻击响应页面 openwaf_attack_response
- API openwaf_api
- 防恶意爬虫 openwaf_anti_mal_crawler
- 防CC openwaf_anti_cc
详细配置文档及示例请看上述各模块文档
The OpenWAF-en mailing list is for English speakers.
The OpenWAF-cn mailing list is for Chinese speakers.
579790127
Please submit bug reports, wishlists, or patches by
- creating a ticket on the GitHub Issue Tracker,
- or posting to the OpenWAF community.
-
- 上传人机识别模块
-
- 上传防盗链模块
-
- 上传cookie防篡改模块
-
- 上传基于频率的模糊识别防探测模块
-
- 上传WebShell上传防护模块
-
- 上传防CSRF模块
-
- 放开动态配置规则引擎API
-
- 放开动态配置行为分析引擎API
Time: 2017/03/20
Version: v0.0.4
1. New Module - twaf_anti_cc
Anti http flood
Time: 2017/01/03
Version: v0.0.3.170103_beta
1. New Module - twaf_anti_mal_crawler
Distinguish malicious crawler and some scan tools
Time: 2016/12/05
Version: v0.0.2.161205_beta
1. New Module - twaf_attack_response
Return custom response page When the request is rejected by OpenWAF
2. Api - api/stat[/policy_uuid]
Show statistical infomation
Time: 2016/12/05
Version: v0.0.1.161130_beta
1. Docker
build OpenWAF with docker
Time: 2016/12/05
Version: v0.0.1.161012_beta
1. log module
Send tcp/udp log
2. reqstat module
Statistics of request infomation
3. access rule
Publish applications
4. rule engine
Access Control
This module is licensed under the BSD license.
Copyright (C) 2016-2016, by Jian "Miracle" Qi (齐健) [email protected], Titan Co.Ltd.
All rights reserved.
Redistribution and use in source and binary forms, with or without modification, are permitted provided that the following conditions are met:
-
Redistributions of source code must retain the above copyright notice, this list of conditions and the following disclaimer.
-
Redistributions in binary form must reproduce the above copyright notice, this list of conditions and the following disclaimer in the documentation and/or other materials provided with the distribution.
THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS" AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT HOLDER OR CONTRIBUTORS BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.
- twaf_access_rule
- twaf_anti_hotlink
- twaf_anti_mal_crawler
- twaf_reqstat
- twaf_log
- twaf_secrules
- twaf_anti_cc
{
"twaf_access_rule": {
"rules": [ -- 注意先后顺序
{
"client_ssl": false, -- 客户端认证的开关，与ngx_ssl组成双向认证
"client_ssl_cert": "path", -- 客户端认证所需PEM证书地址
"ngx_ssl": false, -- nginx认证的开关
"ngx_ssl_cert": "path", -- nginx认证所需PEM证书地址
"ngx_ssl_key": "path", -- nginx认证所需PEM私钥地址
"host": "^1\\.1\\.1\\.1$", -- 域名，正则匹配
"port": 80, -- 端口号（缺省80）
"path": "\/", -- 路径，正则匹配
"server_ssl": false, -- 后端服务器ssl开关
"forward": "server_5", -- 后端服务器upstream名称
"forward_addr": "1.1.1.2", -- 后端服务器ip地址
"forward_port": "8080", -- 后端服务器端口号（缺省80）
"uuid": "access_567b067ff2060", -- 用来标记此规则的uuid
"policy": "policy_uuid" -- 安全策略ID
}
]
}
}syntax: "rules": table
default: none
context: twaf_access_rule
table类型，接入规则，顺序匹配
syntax: "client_ssl": true|false
default: false
context: twaf_access_rule
客户端认证开关，与ngx_ssl组成双向认证，默认false
PS: 当前客户端认证不生效
syntax: "client_ssl_cert": "path"
default: none
context: twaf_access_rule
string类型，客户端认证所需PEM证书地址，目前仅支持绝对地址
syntax: "ngx_ssl": true|false
default: false
context: twaf_access_rule
boolean类型，服务器端(nginx)认证开关，与client_ssl组成双向认证，默认关闭
syntax: "ngx_ssl_cert": "path"
default: none
context: twaf_access_rule
string类型，服务器端(nginx)认证所需PEM证书地址，目前仅支持绝对地址
syntax: "ngx_ssl_key": "path"
default: none
context: twaf_access_rule
string类型，服务器端(nginx)认证所需PEM私钥地址，目前仅支持绝对地址
syntax: "host": "ip|domain name regex"
default: none
context: twaf_access_rule
string类型，域名，正则匹配(匹配时，忽略大小写)
例如:
"host": "^1\\.1\\.1\\.1$"
"host": "test\\.com"
"host": "^.*\\.com$"
"host": "www.baidu.com"
syntax: "port": number
default: 80
context: twaf_access_rule
number类型，端口号
syntax: "path": "regex"
default: none
context: twaf_access_rule
string类型，路径，正则匹配
例如:
"path": "/"
"path": "/images"
"path": "/[a|b]test"
syntax: "server_ssl": true|false
default: false
context: twaf_access_rule
boolean类型，OpenWAF向后端服务器连接的ssl开关
例如:
upstream test {
server 1.1.1.1;
}
http {
server {
listen 80;
server_name _;
location / {
#server_ssl为true，相当于proxy_pass后为https
proxy_pass https://test;
#server_ssl为false，相当于proxy_pass后为http
#proxy_pass http://test;
}
}
}
syntax: "forward": "upstream_uuid"
default: none
context: twaf_access_rule
string类型，forward表示后端服务器的uuid，即upstream的名称
#如：forward值为test
upstream test {
server 1.1.1.1;
}
syntax: "forward_addr": "ip"
default: none
context: twaf_access_rule
string类型，forward_addr表示后端服务器的ip地址（TODO：支持域名）
upstream test {
#如：forward_addr值为1.1.1.1
server 1.1.1.1;
}
syntax: "forward_port": port
default: 80
context: twaf_access_rule
number类型，forward_port表示后端服务器端口号，默认80
upstream test {
#如：forward_port值为50001
server 1.1.1.1:50001;
}
syntax: "uuid": "string"
default: none
context: twaf_access_rule
string类型，接入规则的唯一标识
syntax: "policy": "policy_uuid"
default: none
context: twaf_access_rule
string类型，满足此接入规则的请求，所使用安全策略的uuid
{
"twaf_anti_hotlink":{
"state":false,
"log_state":true,
"event_id":"110001",
"event_severity":"medium",
"ct_state":false,
"action_meta":403,
"action":"DENY",
"mode":"referer",
"allow_noreferer":true,
"cookie_name":"TWAF_AH",
"uri_ext":["javascript", "css", "html", ""]
}
}syntax: "state": true|false|"$dynamic_state"
default: false
context: twaf_anti_hotlink
当前模块暂未开源
syntax: "log_state": true|false|"$dynamic_state"
default: true
context: twaf_anti_hotlink
syntax: "ct_state": true|false|"$dynamic_state"
default: false
context: twaf_anti_hotlink
syntax: "event_id": "string"
default: "110001"
context: twaf_anti_hotlink
syntax: "event_severity": "string"
default: "medium"
context: twaf_anti_hotlink
syntax: "action": "string"
default: "DENY"
context: twaf_anti_hotlink
syntax: "action_meta": "string"|number
default: 403
context: twaf_anti_hotlink
syntax: "mode": "string"
default: "referer"
context: twaf_anti_hotlink
syntax: "allow_noreferer": true|false
default: true
context: twaf_anti_hotlink
syntax: "cookie_name": "string"
default: TWAF_AH
context: twaf_anti_hotlink
cookie_name表示盗链模块发送COOKIE的名称，默认"TWAF_AH"
此配置只有mode为cookie模式下生效
syntax: "uri_ext": array|exten|"all"
default: none
context: twaf_anti_hotlink
uri_ext表示对哪些资源进行盗链防护
#对html类型资源进行盗链防护
"uri_ext": "html"
#对未知类型资源进行盗链防护，nginx无法解析出资源类型时为空字符串
"uri_ext": ""
#对html、css及未知类型资源进行盗链防护
"uri_ext": ["html", "css", ""]
#对所有资源进行盗链防护
"uri_ext": "all"
{
"state": false, -- 模块开关，支持 true，false
"log_state":true, -- 日志开关
"dict_state": false, -- shared_dict 开关
"shared_dict_name":"twaf_anti_mal_crawler", -- shared_dict 名称,若为空，则值为 "twaf_global" 下的 "dict_name"
"shared_dict_key": "remote_addr", -- shared_dict 键值
"timeout":300, -- shared_dict 保存状态有效时长（单位秒）
"timer_flush_expired":200, -- shared_dict 清除过期信息的间隔时间（单位秒）,若为空，则值为 "twaf_global" 下的 "timer_flush_expired"
"cookie_state":true, -- cookie机制开关
"crawler_cookie_name":"TWAF_crawler", -- 爬虫cookie名称
"mal_cookie_name":"TWAF_mcrawler", -- 恶意爬虫cookie名称
"force_scan_robots_state":true, -- 页面注入诱捕路径的开关
"force_scan_times": 3, -- 注入诱捕路径的页面个数
"trap_uri":"/abc/abc.html", -- 诱捕路径
"trap_args":"id=1", -- 诱捕参数
"action":"DENY", -- 执行动作，支持 "ALLOW", "DENY", "REDIRECT", "ROBOT", "RESET_CONNECTION", "PASS" 等
"action_meta": 403 -- 执行动作的附属信息，若 action 为 DENY，action_meta为响应码，若 action 为 REDIRECT，action_meta 为重定向 url
}syntax: state true|false|$dynamic_state
default: false
context: twaf_anti_mal_crawler
模块开关，默认false（关闭），支持动态开关
syntax: log_state true|false|$dynamic_state
default: true
context: twaf_anti_mal_crawler
安全日志开关， 默认true（记录），支持动态开关
syntax: dict_state true|false
default: false
context: twaf_anti_mal_crawler
shared_dict 开关。当 dict_state 为 true，某 IP 被此模块拦截，会被记录在内存中，在 timeout 时间内访问会被拦截(且重置timeout)
syntax: shared_dict_name
default: nil
context: twaf_anti_mal_crawler
shared_dict 名称。对应 nginx 中的配置项，不可轻易修改
若为空，则值为 "twaf_global" 下的 "dict_name"
syntax: shared_dict_key |
default: remote_addr
context: twaf_anti_mal_crawler
shared_dict 键值。支持数组
syntax: timeout
default: 300
context: twaf_anti_mal_crawler
shared_dict 保存状态有效时长（单位秒）
syntax: timeout
default: 200
context: twaf_anti_mal_crawler
shared_dict 清除过期信息的间隔时间（单位秒）,若为空，则值为 "twaf_global" 下的 "timer_flush_expired"
syntax: cookie_state true|false|$dynamic_state
default: true
context: twaf_anti_mal_crawler
是否发送cookie,默认true（发送），支持动态开关
syntax: crawler_cookie_name
default: "TWAF_crawler"
context: twaf_anti_mal_crawler
爬虫 cookie 名称
syntax: mal_cookie_name
default: TWAF_mcrawler
context: twaf_anti_mal_crawler
恶意爬虫cookie名称
syntax: force_scan_robots_state true|false
default: true
context: twaf_anti_mal_crawler
页面注入诱捕路径的开关
某些扫描工具不会去访问 /robots.txt，因此在他访问的页面中插入禁爬目录的暗链
syntax: force_scan_times
default: 3
context: twaf_anti_mal_crawler
注入诱捕路径的页面数
syntax: trap_uri
default: /abc/abc.html
context: twaf_anti_mal_crawler
诱捕路径，访问此路径，被标识为恶意爬虫
syntax: trap_args
default: id=1
context: twaf_anti_mal_crawler
诱捕参数。携带此参数访问诱捕路径，不会标识为攻击
"twaf_reqstat": {
"state":true,
"safe_state":true,
"access_state":true,
"upstream_state":true,
"shared_dict_name":"twaf_reqshm"
}syntax: state true|false|$dynamic_state
default: true
context: twaf_reqstat
统计模块开关，支持动态开关，默认开启
syntax: access_state true|false|$dynamic_state
default: true
context: twaf_reqstat
访问信息统计开关，支持动态开关，默认开启
syntax: safe_state true|false|$dynamic_state
default: true
context: twaf_reqstat
安全信息统计开关，支持动态开关，默认开启
syntax: upstream_state true|false|$dynamic_state
default: true
context: twaf_reqstat
转发信息统计开关，支持动态开关，默认开启
syntax: shared_dict_name string
default: openwaf_reqshm
context: twaf_reqstat
指定shared_dict名称，在这之前需在nginx配置文件中配置lua_shared_dict
默认shared_dict名称为openwaf_reqshm
"twaf_log": {
"access_log_state":false, -- 访问日志开关
"security_log_state":true, -- 安全日志开关
"sock_type":"udp", -- 支持tcp和udp两种协议
"content_type":"JSON", -- 支持JSON和INFLUXDB两种日志格式
"host":"127.0.0.1", -- 日志服务器地址
"port":60055, -- 日志服务器端口号
"flush_limit":0, -- 缓冲，当存储的日志大于阈值才发送
"size_limit": 200, -- 控制日志中每一项的字符上限，如'raw_header'或请求体响应体，可能会使udp日志报错
"drop_limit":65507, -- 缓冲上限，达到此值，丢弃当前日志，发送缓存并清空缓存，当sock_type为udp时，drop_limit值最大为65507（65508会报错message too long）
"max_retry_times":5, -- 最大容错次数
"ssl":false, -- 是否开启ssl协议
"access_log":{} -- 访问日志格式
"security_log":{} -- 安全日志格式
}syntax: "access_log_state": true|false
default: false
context: twaf_log
访问日志开关，默认关闭
syntax: "security_log_state": true|false
default: true
context: twaf_log
安全事件日志开关，默认开启
syntax: "sock_type": tcp|udp
default: udp
context: twaf_log
日志传输协议，默认udp
syntax: "content_type": JSON|INFLUXDB
default: JSON
context: twaf_log
日志格式，默认JSON
syntax: "host": string
default: "127.0.0.1"
context: twaf_log
日志接收服务器的ip地址
syntax: "port": number
default: 60055
context: twaf_log
日志接收服务器的端口号
syntax: "flush_limit": number
default: 0
context: twaf_log
缓冲区大小，当存储的日志大于阈值才发送，默认值为0，即立即发送日志
syntax: "size_limit": number
default: 200
context: twaf_log
控制日志中每一项的字符上限，如'raw_header'或请求体响应体，可能会使udp日志报错
syntax: "drop_limit": number
default: 65507
context: twaf_log
缓冲上限，达到此值，丢弃当前日志，发送缓存并清空缓存，当sock_type为udp时，drop_limit值最大为65507（65508会报错message too long）
syntax: "max_retry_times": number
default: 5
context: twaf_log
最大容错次数
syntax: "ssl": true|false
default: false
context: twaf_log
是否开启ssl协议，默认false
syntax: "access_log": table
default: false
context: twaf_log
访问日志格式
syntax: "security_log": table
default: false
context: twaf_log
安全事件日志格式
若content_type为JSON，则日志格式为
[
variable_key_1,
variable_key_2,
...
]
若content_type为INFLUXDB，则日志格式为
{
"db":MEASUREMENT名称,
"tags":[variable_key_1, variable_key_2, ...],
"fileds"[variable_key_1, variable_key_2, ...],
"time":true|false
}
变量名称详见规则引擎模块twaf_secrules
#日志格式举例
#JSON格式
"security_log": [
"remote_addr",
"remote_port",
"userid",
"dev_uuid",
"original_dst_addr",
"original_dst_port",
"remote_user",
"time_local",
"msec",
"request_method",
"request_uri",
"request_protocol",
"status",
"bytes_sent",
"http_referer",
"http_user_agent",
"gzip_ratio",
"http_host",
"raw_header"
]
#INFLUXDB格式
"security_log": {
"db":"test", -- MEASUREMENT名称
"tags":[], -- tags keys
"fileds":[ -- fileds keys
"remote_addr",
"remote_port",
"userid",
"dev_uuid",
"original_dst_addr",
"original_dst_port",
"remote_user",
"time_local",
"msec",
"request_method",
"request_uri",
"request_protocol",
"status",
"bytes_sent",
"http_referer",
"http_user_agent",
"gzip_ratio",
"http_host",
"raw_header"
],
"time":true -- 日志是否携带时间戳
}
"twaf_secrules":{
"state": true, -- 总开关
"reqbody_state": true, -- 请求体检测开关
"header_filter_state": true, -- 响应头检测开关
"body_filter_state": true, -- 响应体检测开关
"system_rules_state": true, -- 系统规则集检测开关
"reqbody_limit":134217728, -- 请求体检测阈值，大于阈值不检测
"respbody_limit":524288, -- 响应体检测阈值，大于阈值不检测
"pre_path": "/opt/OpenWAF/", -- OpenWAF安装路径
"path": "lib/twaf/inc/knowledge_db/twrules", -- 特征规则库在OpenWAF中的路径
"user_defined_rules":[ -- 用户自定义规则，数组
],
"rules_id":{ -- 特征排除
"111112": [{"REMOTE_HOST":"a.com", "URI":"^/ab"}], -- 匹配中数组中信息则对应规则失效，数组中key为变量名称，值支持正则
"111113": {}, -- 特征未被排除
"111114": [{}] -- 特征被无条件排除
}
}syntax: state true|false
default: true
context: twaf_secrules
规则引擎总开关
syntax: reqbody_state true|false
default: true
context: twaf_secrules
请求体检测开关
syntax: header_filter_state true|false
default: true
context: twaf_secrules
响应头检测开关
syntax: body_filter_state true|false
default: false
context: twaf_secrules
响应体检测开关，默认关闭，若开启需添加第三方模块[ngx_http_twaf_header_sent_filter_module暂未开源]
syntax: system_rules_state true|false
default: true
context: twaf_secrules
系统规则集检测开关
lib/twaf/inc/knowledge_db/twrules 目录下的规则，都是系统规则
除了系统规则外，还有 twaf_secrules 模块下 user_defined_rules 的用户自定义规则
系统规则一般很少改动，而用户自定义规则却随着业务而增减，如动态配置缓存、压缩、时域控制和黑白名单等。
syntax: reqbody_limit number
default: 134217728
context: twaf_secrules
请求体检测大小上限，默认134217728B(128MB)，若请求体超过设置上限，则不检测
PS：reqbody_limit值要小于nginx中client_body_buffer_size的值才会生效
syntax: respbody_limit number
default: 134217728
context: twaf_secrules
响应体检测大小上限，默认134217728B(128MB)，若响应体大小超过设置上限，则不检测
syntax: pre_path string
default: /opt/OpenWAF/
context: twaf_secrules
OpenWAF的安装路径
syntax: path string
default: lib/twaf/inc/knowledge_db/twrules
context: twaf_secrules
特征规则库在OpenWAF中的路径
syntax: user_defined_rules
default: []
context: twaf_secrules
用户自定义规则
先执行用户自定义规则，再执行系统规则
syntax: rules_id table
default: none
context: twaf_secrules
用于排除特征
{
"twaf_limit_conn": {
"state":false, -- CC防护模块开关
"log_state":true, -- CC日志开关
"trigger_state":true, -- 触发开关
"clean_state":true, -- 清洗开关
"trigger_thr":{ -- 触发阈值（关系为“或”）
"req_flow_max":1073741824, -- 每秒请求流量，单位B
"req_count_max":10000 -- 每秒请求数
},
"clean_thr":{ -- 清洗阈值
"new_conn_max":40, -- 单一源ip每秒新建连接数
"conn_max":100, -- 单一源ip防护期间内连接总数
"req_max":50, -- 单一源ip每秒请求总数
"uri_frequency_max":3000 -- 单一路径每秒请求总数
},
"attacks": 1, -- 在一次CC攻击过程中，某ip触发清洗值的次数大于attacks，则此ip会一直被拦截，直到CC攻击结束
"timer_flush_expired":10, -- 清理shared_dict过期数据的时间间隔
"interval":10, -- 进入CC防护后发送日志的时间间隔，单位秒
"shared_dict_name":"twaf_limit_conn", -- 存放其他信息的shared_dict
"shared_dict_key": "remote_addr", -- shared_dict的键值
"action":"DENY", -- 触发CC防护执行的动作
"action_meta":403,
"timeout":30 -- 清洗时长（当再次触发清洗值时，重置）
}
}syntax: "state": true|false
default: false
context: twaf_limit_conn
boolean类型，CC防护模块总开关，默认关闭
syntax: "log_state": true|false
default: true
context: twaf_limit_conn
boolean类型，CC防护模块日志开关，默认开启
syntax: "trigger_state": true|false
default: true
context: twaf_limit_conn
boolean类型，CC防护模块的触发开关，默认开启
若关闭，则触发机制关闭，时刻进入CC清洗状态
syntax: "clean_state": true|false
default: true
context: twaf_limit_conn
boolean类型，CC防护模块总开关，默认开启
若关闭（仅用于测试），则清洗机制关闭，CC模块将无法拦截请求
syntax: "trigger_thr": table
default: {"req_flow_max":1073741824,"req_count_max":10000}
context: twaf_limit_conn
table类型，触发阈值
当达到其中一个触发阈值，进入CC清洗状态
当前有两个触发阈值
"trigger_thr":{ -- 触发阈值（关系为“或”）
"req_flow_max":1073741824, -- 每秒请求流量，单位B，默认1GB/s
"req_count_max":10000 -- 每秒请求数，默认10000个/秒
}syntax: "clean_thr": table
default: {"new_conn_max":40,"conn_max":100,"req_max":50,"uri_frequency_max":3000}
context: twaf_limit_conn
table类型，清洗阈值
当进入CC清洗状态，达到其中一个清洗阈值，则执行相应动作
当前有四个清洗阈值
"clean_thr":{ -- 清洗阈值（关系为“或”）
"new_conn_max":40, -- 单一源ip每秒新建连接数，默认40个/秒
"conn_max":100, -- 单一源ip防护期间内连接总数，默认100个
"req_max":50, -- 单一源ip每秒请求总数，默认50个/秒
"uri_frequency_max":3000 -- 单一路径每秒请求总数，默认3000个/秒
}syntax: "attacks": number
default: 1
context: twaf_limit_conn
在一次 CC 攻击过程中，某ip触发清洗阈值的次数大于 attacks ，则此 ip 会一直被拦截，直到 CC 攻击结束
此前，在一次 CC 攻击过程中，当达到清洗阈值时，才会进行拦截。若未达到清洗阈值，即使之前被拦截过，也可正常访问后端服务器
正确设置此参数，可以大大提升 CC 防护性能
若想恢复以前的 CC 防护机制，只需 attacks 设为 0 即可
此参数出现在 OpenWAF-0.0.6 版本， twaf_anti_cc 的 0.0.3 版本
syntax: "timer_flush_expired": number
default: 10
context: twaf_limit_conn
number类型，清理shared_dict过期数据的时间间隔，默认10秒
syntax: "interval": number
default: 10
context: twaf_limit_conn
number类型，进入CC防护后发送日志的时间间隔，默认10秒
syntax: "shared_dict_name": string
default: "twaf_limit_conn"
context: twaf_limit_conn
string类型，存放当前CC防护信息的shared_dict名称
syntax: "shared_dict_key": string|table
default: "remote_addr"
context: twaf_limit_conn
string或table类型，shared_dict的键值，支持nginx变量
支持字符串类型和数组类型
"shared_dict_key": "remote_addr"
"shared_dict_key": ["remote_addr", "http_user_agent"]
syntax: "action": string
default: "DENY"
context: twaf_limit_conn
string类型，触发CC防护执行的动作，默认"DENY"
syntax: "action_meta": number|string
default: 403
context: twaf_limit_conn
string或number类型，执行动作的附属信息，默认403
syntax: "timeout": number
default: 30
context: twaf_limit_conn
number类型，清洗时长，N秒内不再达到触发阈值，则退出CC清洗状态
在清洗过程中，再次达到触发阈值，则时间重置为30秒
syntax: set $twaf_https 0|1
default: 0
context: server
用于标记请求是否通过ssl加密
"set $twaf_https 1"，则表示请求通过ssl加密
"set $twaf_https 0"，则表示请求未通过ssl加密
server {
listen 443 ssl;
set $twaf_https 1;
...
}
server {
listen 80;
set $twaf_https 0;
...
}
syntax: set $twaf_upstream_server ""
default: none
context: server
用于指定后端服务器地址，只需初始化为空字符串即可，其值由"server_ssl"和"forward"确定
upstream server_1 {
...
}
upstream server_2 {
...
}
server {
...
set $twaf_upstream_server "";
location / {
...
proxy_pass $twaf_upstream_server;
}
}
若"server_ssl"值为true, "forward"值为"server_1"
等价于proxy_pass https://server_1;
若"server_ssl"值为false, "forward"值为"server_2"
等价于proxy_pass http://server_2;
- ARGS
- ARGS_COMBINED_SIZE
- ARGS_GET
- ARGS_GET_NAMES
- ARGS_NAMES
- ARGS_POST
- ARGS_POST_NAMES
- BYTES_IN
- CONNECTION_REQUESTS
- DURATION
- FILES
- FILES_NAMES
- GEO
- GEO_CODE3
- GEO_CODE3
- GEO_ID
- GEO_CONTINENT
- GEO_NAME
- GZIP_RATIO
- HTTP_COOKIE
- HTTP_HOST
- HTTP_REFERER
- HTTP_USER_AGENT
- IP_VERSION
- MATCHED_VAR
- MATCHED_VARS
- MATCHED_VAR_NAME
- MATCHED_VARS_NAMES
- ORIGINAL_DST_ADDR
- ORIGINAL_DST_PORT
- POLICYID
- QUERY_STRING
- RAW_HEADER
- RAW_HEADER_TRUE
- REMOTE_ADDR
- REMOTE_HOST
- REMOTE_PORT
- REMOTE_USER
- REQUEST_BASENAME
- REQUEST_BODY
- REQUEST_COOKIES
- REQUEST_COOKIES_NAMES
- REQUEST_FILENAME
- REQUEST_HEADERS
- REQUEST_HEADERS_NAMES
- REQUEST_LINE
- REQUEST_METHOD
- REQUEST_PROTOCOL
- HTTP_VERSION
- URI
- URL
- REQUEST_URI
- RESPONSE_BODY
- RESPONSE_HEADERS
- RESPONSE_STATUS
- SCHEME
- SERVER_ADDR
- SERVER_NAME
- SERVER_PORT
- SESSION
- SESSION_DATA
- TIME
- TIME_DAY
- TIME_EPOCH
- TIME_HOUR
- TIME_MIN
- TIME_MON
- TIME_SEC
- TIME_WDAY
- TIME_YEAR
- TIME_LOCAL
- TX
- UNIQUE_ID
- UPSTREAM_CACHE_STATUS
- USERID
table类型，所有的请求参数，包含ARGS_GET和ARGS_POST
例如：POST http://www.baidu.com?name=miracle&age=5
请求体为：time=123456&day=365
ARGS变量值为{"name": "miracle", "age": "5", "time": "123456", "day": "365"}
number类型，请求参数总长度，只包含key和value的长度，不包含'&'或'='等符号
例如：GET http://www.baidu.com?name=miracle&age=5
ARGS_COMBINED_SIZE变量值为15，而不是18
table类型，querystring参数
例如：GET http://www.baidu.com?name=miracle&age=5
ARGS_GET变量值为{"name": "miracle", "age": "5"}
table类型，querystring参数key值
例如：GET http://www.baidu.com?name=miracle&age=5
ARGS_GET_NAMES变量值为["name", "age"]
table类型，querystring参数key值及post参数key值
例如：POST http://www.baidu.com?name=miracle&age=5
请求体为：time=123456&day=365
ARGS_NAMES变量值为["name", "age", "time", "day"]
table类型，POST参数
例如：
POST http://www.baidu.com/login.html
请求体为：time=123456&day=365
ARGS_POST变量值为{"time": "123456", "day": "365"}
table类型，POST参数key值
例如：
POST http://www.baidu.com/login.html
请求体为：time=123456&day=365
ARGS_POST_NAMES变量值为["time", "day"]
number类型，接收信息字节数
number类型，当前连接中的请求数
string类型，处理事务用时时间，单位:微秒(μs)
table类型，从请求体中得到的原始文件名(带有文件后缀名)
table类型，上传文件名称（不带有后缀名）
table类型，包含code3,code,id,continent,name等字段信息
string类型，3个字母长度的国家缩写
string类型，2个字母长度的国家缩写
number类型，国家ID
string类型，国家所在大洲
string类型，国家全称
string类型，压缩比率
string类型，请求头中的cookie字段
string类型，请求头中的host字段值，既域名:端口(80缺省)
string类型，请求头中的referer字段
string类型，请求头中的user-agent字段
string类型，IPv4 or IPv6
类型不定，当前匹配中的变量
table类型，单条规则匹配中的所有变量
string类型，当前匹配中的变量名称
table类型，单条规则匹配中的所有变量名称
string类型，服务器地址，应用代理模式为WAF地址，透明模式为后端服务器地址
string类型，服务器端口号，应用代理模式为WAF端口号，透明模式为后端服务器端口号
string类型，策略ID
string类型，未解码的请求参数
string类型，请求头信息，带请求行
string类型，请求头信息，不带请求行
string类型，客户端地址
string类型，域名
number类型，端口号
string类型，用于身份验证的用户名
string类型，请求的文件名
例如: GET http://www.baidu.com/test/login.php
REQUEST_BASENAME值为/login.php
类型不定，请求体
table类型，请求携带的cookie
table类型，请求携带cookie的名称
string类型，relative request URL(相对请求路径)
例如: GET http://www.baidu.com/test/login.php
REQUEST_FILENAME值为/test/login.php
table类型，请求头信息
table类型，请求头key值
string类型，请求行
string类型，请求方法
string类型，http请求协议，如: HTTP/1.1
string类型，http请求协议版本，如: 1.1
string类型，请求路径，既不带域名，也不带参数
例如: GET http://www.baid.com/test/login.php?name=miracle
URI变量值为/test/login.php
string类型，统一资源定位符，SCHEME与HTTP_HOST与URI的拼接
例如: GET http://www.baid.com/test/login.php?name=miracle
URL变量值为http://www.baid.com/test/login.php
string类型，请求路径，带参数，但不带有域名
例如: GET http://www.baid.com/test/login.php?name=miracle
REQUEST_URI变量值为/test/login.php?name=miracle
string类型，响应体
table类型，响应头信息
function类型，响应状态码
string类型，http or https
例如：GET http://www.baidu.com/
SCHEME变量值为http
string类型，服务器地址
string类型，服务器名称
number类型，服务器端口号
table类型，第三方模块lua-resty-session提供的变量
table类型，session信息，第三方模块lua-resty-session提供的变量
string类型，hour:minute:second
number类型，天(1-31)
number类型，时间戳，seconds since 1970
number类型，小时(0-23)
number类型，分钟(0-59)
number类型，月份(1-12)
number类型，秒(0-59)
number类型，周(0-6)
number类型，年份，four-digit，例如: 1997
string类型，当前时间，例如: 26/Aug/2016:01:32:16 -0400
table类型，用于存储当前请求信息的变量，作用域仅仅是当前请求
string类型，ID标识，随机生成的字符串，可通过配置来控制随机字符串的长度
keeps the status of accessing a response cache (0.8.3). The status can be either “MISS”, “BYPASS”, “EXPIRED”, “STALE”, “UPDATING”, “REVALIDATED”, or “HIT”.
string类型，从接入规则配置得到的用于ID标识
- base64_decode
- sql_hex_decode
- base64_encode
- counter
- compress_whitespace
- hex_decode
- hex_encode
- html_decode
- length
- lowercase
- md5
- normalise_path
- remove_nulls
- remove_whitespace
- replace_comments
- remove_comments_char
- remove_comments
- uri_decode
- uri_encode
- sha1
- trim_left
- trim_right
- trim
Decodes a Base64-encoded string.
Note: 注意transform的执行顺序
例如：
{
"id": "xxxx",
...
"transform": ["base64_decode", "lowercase"],
...
}
先执行base64解码，然后字符串最小化，若顺序调换，会影响结果
Decode sql hex data.
Encodes input string using Base64 encoding.
计数，相当于modsecurity中的'&'符号
Converts any of the whitespace characters (0x20, \f, \t, \n, \r, \v, 0xa0) to spaces (ASCII 0x20), compressing multiple consecutive space characters into one.
Decodes a string that has been encoded using the same algorithm as the one used in hexEncode
Encodes string (possibly containing binary characters) by replacing each input byte with two hexadecimal characters.
Decodes the characters encoded as HTML entities.
Looks up the length of the input string in bytes
Converts all characters to lowercase
Calculates an MD5 hash from the data in input. The computed hash is in a raw binary form and may need encoded into text to be printed (or logged). Hash functions are commonly used in combination with hex_encode (for example: "transform": ["md5", "hex_encode").
Removes multiple slashes, directory self-references, and directory back-references (except when at the beginning of the input) from input string.
Removes all NUL bytes from input
Removes all whitespace characters from input.
移除空白字符\s，包含水平定位字符 ('\t')、归位键('\r')、换行('\n')、垂直定位字符('\v')或翻页('\f')等
用一个空格代替/.../注释内容
Removes common comments chars (/*, */, --, #).
去掉/.../注释内容
Unescape str as an escaped URI component.
例如:
"b%20r56+7" 使用uri_decode转换后为 b r56 7
Escape str as a URI component.
Calculates a SHA1 hash from the input string. The computed hash is in a raw binary form and may need encoded into text to be printed (or logged). Hash functions are commonly used in combination with hex_encode (for example, "transform": ["sha1", "hex_encode"]).
Removes whitespace from the left side of the input string.
Removes whitespace from the right side of the input string.
Removes whitespace from both the left and right sides of the input string.
- begins_with
- contains
- contains_word
- detect_sqli
- detect_xss
- ends_with
- equal
- greater_eq
- greater
- ip_utils
- less_eq
- less
- pf
- regex
- str_match
- validate_url_encoding
- num_range
- str_range
Returns true if the parameter string is found at the beginning of the input.
Returns true if the parameter string is found anywhere in the input.
operator为contains且pattern为数组，相当于modsecurity的pm
PS: modsecurity的pm忽略大小写，OpenWAF中contains不忽略大小写
例如:
{
"id": "xxx",
...
"operator": "contains",
"pattern": ["abc", "def"],
...
}
Returns true if the parameter string (with word boundaries) is found anywhere in the input.
This operator uses LibInjection to detect SQLi attacks.
This operator uses LibInjection to detect XSS attacks.
Returns true if the parameter string is found at the end of the input.
Performs a string comparison and returns true if the parameter string is identical to the input string.
相当于modsecurity的eq和streq
例如:
{
"id": "xxx",
...
"operator": "equal",
"pattern": [12345, "html", "23456"]
...
}
Performs numerical comparison and returns true if the input value is greater than or equal to the provided parameter.
return false, if a value is provided that cannot be converted to a number.
Performs numerical comparison and returns true if the input value is greater than the operator parameter.
return false, if a value is provided that cannot be converted to a number.
Performs a fast ipv4 or ipv6 match of REMOTE_ADDR variable data. Can handle the following formats:
Full IPv4 Address: 192.168.1.100 Network Block/CIDR Address: 192.168.1.0/24 IPv4 Address Region: 1.1.1.1-2.2.2.2
ip_utils与pf的组合相当于modsecurity中的ipMatchF和ipMatchFromFile
例如:
规则如下：
{
"id": "xxxx",
...
"operator": "ip_utils",
"pf": "/tmp/ip_blacklist.txt",
...
}
"/tmp/ip_blacklist.txt"文件内容如下：
192.168.1.100
192.168.1.0/24
1.1.1.1-2.2.2.2
Performs numerical comparison and returns true if the input value is less than or equal to the operator parameter.
return false, if a value is provided that cannot be converted to a number.
Performs numerical comparison and returns true if the input value is less than to the operator parameter.
return false, if a value is provided that cannot be converted to a number.
pattern是operator操作的参数
pf是指pattern from file，与pattern互斥（二者不可同时出现），目前仅支持绝对路径
pf与contains组合，相当于modsecurity的pmf或pmFromFile
pf与ip_utils组合，相当于modsecurity的ipMatchF或ipMatchFromFile
Performs a regular expression match of the pattern provided as parameter.
regex还有modecurity的capture捕获功能
modsecurity有关capture的描述如下： When used together with the regular expression operator (@rx), the capture action will create copies of the regular expression captures and place them into the transaction variable collection.
OpenWAF中无capture指令，但使用regex默认开启capture功能
例如:
{
"id": "000031",
"release_version": "858",
"charactor_version": "001",
"opts": {
"nolog": false
},
"phase": "access",
"action": "deny",
"meta": 403,
"severity": "low",
"rule_name": "protocol.reqHeader.c",
"desc": "协议规范性约束，检测含有不合规Range或Request-Range值的HTTP请求",
"match": [
{
"vars": [
{
"var": "REQUEST_HEADERS",
"parse": {
"specific": "Range"
}
},
{
"var": "REQUEST_HEADERS",
"parse": {
"specific": "Request-Range"
}
}
],
"operator": "regex",
"pattern": "(\\d+)\\-(\\d+)\\,"
},
{
"vars": [{
"var": "TX",
"parse": {
"specific": "2"
}
}],
"operator": "greater_eq",
"pattern": "%{TX.1}",
"parse_pattern": true,
"op_negated": true
}
]
}
等同于contains
Validates the URL-encoded characters in the provided input string.
判断是否在数字范围内
它与transform的length组合，相当于modsecurity的validateByteRange
{
"id": "xxx",
...
"operator": "num_range",
"pattern": [10, "13", "32-126"],
"transform": "length",
...
}
判断是否在字符串范围内
例如时间区间判断:
{
"id": "xxx",
...
"operator": "str_range",
"pattern": ["01:42:00-04:32:00"],
...
}
- allow
- allow_phase
- deny
- id
- nolog
- op_negated
- parse
- pass
- warn
- audit
- phase
- proxy_cache
- redirect
- charactor_version
- severity
- setvar
- meta
- transform
- tag
- release_version
- robot
- add_resp_headers
Stops processing of the current phase but also skipping over all other phases.
"action": "allow"
一旦执行此动作，则后面的防护规则及其他安全模块均不进行安全检测，此动作一般用于白名单
Stops processing of the current phase.
"action": "allow_phase"
Stops rule processing and intercepts transaction.
"action": "deny",
"meta": 403
Stops rule processing and intercepts transaction.
"id": "xxxxxxx"
不记录日志
"opts": {
"nolog": true
}
对operator结果的取反
"match": [{
"vars": [{
"var": "HTTP_USER_AGENT"
}],
"transform": "length",
"operator": "less_eq",
"pattern": 50,
"op_negated": true
}]
等价于
"match": [{
"vars": [{
"var": "HTTP_USER_AGENT"
}],
"transform": "length",
"operator": "greater",
"pattern": 50
}]
若请求头中user_agent字段长度大于50，则匹配中此条规则
对变量进一步解析
若请求GET http://www.baidu.com?name=miracle&age=5
"match": [{
"vars": [{
"var": "ARGS_GET"
}]，
...
}]
得到的值为{"name": "miracle", "age": "5"}
"match": [{
"vars": [{
"var": "ARGS_GET",
"parse": {
"specific": "name"
}
}]
}]
得到的值为["miracle"]
"match": [{
"vars": [{
"var": "ARGS_GET",
"parse": {
"specific": ["name", "age"]
}
}]
}]
得到的值为["miracle", "5"]
"match": [{
"vars": [{
"var": "ARGS_GET",
"parse": {
"ignore": "name"
}
}]
}]
得到的值为{"age": "5"}
"match": [{
"vars": [{
"var": "ARGS_GET",
"parse": {
"ignore": ["name", "age"]
}
}]
}]
得到的值为[]
"match": [{
"vars": [{
"var": "ARGS_GET",
"parse": {
"keys": true
}
}]
}]
得到的值为["name", "age"]
"match": [{
"vars": [{
"var": "ARGS_GET",
"parse": {
"values": true
}
}]
}]
得到的值为["miracle", "5"]
"match": [{
"vars": [{
"var": "ARGS_GET",
"parse": {
"all": true
}
}]
}]
得到的值为["name", "age", "miracle", "5"]
Continues processing with the next rule in spite of a successful match.
"action": "pass"
like 'pass'
"action": "warn"
like 'pass'
"action": "audit"
规则执行的阶段，取值可为"access","header_filter","body_filter"的组合
{
"id": "xxx_01",
"phase": "access",
...
}
"xxx_01"规则在access阶段执行
{
"id": "xxx_02",
"phase": ["access", "header_filter"],
...
}
"xxx_02规则在access阶段和"header_filter"阶段各执行一次
{
...
phase = "header_filter", -- 缓存开关需在header_filter阶段配置
action = "pass", -- 无需拦截请求
opts = {
nolog = true, -- 不需记录日志
proxy_cache = {
state = true|false, -- 缓存开关
expired = 600 -- 缓存时长（单位秒）,默认600秒
}
}
...
}
若state为true，且得到的缓存状态为"MISS"或"EXPIRED"，则对响应内容进行缓存，同时设置缓存时长
若state为false，则清除对应缓存键的缓存（包含其缓存文件）
举例如下：
# nginx.conf 有关proxy cache 配置如下
http {
proxy_cache_path /opt/cache/OpenWAF-proxy levels=2:2 keys_zone=twaf_cache:101m max_size=100m use_temp_path=off;
proxy_cache_key $host$uri;
proxy_cache twaf_cache;
proxy_ignore_headers X-Accel-Expires Cache-Control Set-Cookie;
proxy_no_cache $twaf_cache_flag;
server {
set $twaf_cache_flag 1; #默认不缓存
}
}
# lua 格式 配置
{
id = "test_x01", -- id 全局唯一
opts = {
nolog = true,
proxy_cache = {
state = true,
expired = 300
}
},
phase = "header_filter",
action = "pass",
match = {{
vars = {{
var = "URI"
},{
var = "REQUEST_HEADERS",
parse = {
specific = "Referer"
}
}},
operator = "equal",
pattern = {"/xampp/", "%{SCHEME}://%{HTTP_HOST}/xampp/"},
parse_pattern = true
}}
}
此规则将缓存URI为'/xampp/'的页面，更新时间为300秒
若match中过滤条件为响应码，则相当于Nginx的proxy_cache_valid指令
若match中过滤条件为请求方法，则相当于Nginx的proxy_cache_methods指令
若macth中过滤条件为资源类型，则相当于Nginx的proxy_cache_content_type指令
PS: proxy_cache_content_type指令为官方指令，是miracle Qi修改Nginx源码扩展的功能
"action": "redirect",
"meta": "/index.html"
指定此条规则的版本，同modsecurity中Action的rev功能
"charactor_version": "001"
Assigns severity to the rule in which it is used.
The data below is used by the OWASP ModSecurity Core Rule Set (CRS):
EMERGENCY: is generated from correlation of anomaly scoring data where there is an inbound attack and an outbound leakage. ALERT: is generated from correlation where there is an inbound attack and an outbound application level error. CRITICAL: Anomaly Score of 5. Is the highest severity level possible without correlation. It is normally generated by the web attack rules (40 level files). ERROR: Error - Anomaly Score of 4. Is generated mostly from outbound leakage rules (50 level files). WARNING: Anomaly Score of 3. Is generated by malicious client rules (35 level files). NOTICE: Anomaly Score of 2. Is generated by the Protocol policy and anomaly files. INFO DEBUG
也可自定义严重等级，如:low，medium，high，critical等
"severity": "high"
Creates, removes, or updates a variable.
{
"id": "xxx_01",
"opts":{
"nolog": false,
"setvar": [{
"column": "TX",
"key": "score",
"value": 5,
"incr": true
}]
},
...
}
"xxx_01"规则中，给变量TX中score成员的值加5，若TX中无score成员，则初始化为0，再加5
{
"id": "xxx_02",
"opts":{
"nolog": false,
"setvar": [{
"column": "TX",
"key": "score",
"value": 5
}]
},
...
}
"xxx_02"规则中，给变量TX中score成员的值赋为5
"action"的附属信息
若"action"为"deny"，则"meta"为响应码
"action": "deny",
"meta": 403
若"action"为"redirect"，则"meta"为重定向地址
"action": "redirect",
"meta": "/index.html"
This action is used to specify the transformation pipeline to use to transform the value of each variable used in the rule before matching.
Assigns a tag (category) to a rule.
支持数组 "tag": ["xxx_1", "xxx_2"]
支持字符串 "tag": "xxx_3"
规则集版本，等同于modsecurity中Action的ver功能
"release_version": "858"
人机识别
需提前配置人机识别模块配置，此功能暂未放开
"action": "robot"
增删改响应头
例如隐藏server字段:
"opts": {
"add_resp_headers": {
"server": ""
}
}
