一般用法
wrk -t1 -c400 —timeout 5s -d 100s -H “Authorization: ABCDEFGH123OPQ” —latency 'http://localhost:8080/index.html'
参数说明
$ wrk
wrk 4.1.0 [kqueue] Copyright (C) 2012 Will Glozer
Usage: wrk <options> <url>
Options:
# HTTP 连接数
-c, --connections <N> Connections to keep open
# 测试持续时间，如 2s 2m 2h
-d, --duration <T> Duration of test
# 开启的线程数
-t, --threads <N> Number of threads to use
# 进阶功能，使用 lua 脚本
-s, --script <S> Load Lua script file
# 添加请求头，如 "User-Agent: wrk"
-H, --header <H> Add header to request
# 打印详细延迟统计
--latency Print latency statistics
# 设置请求超时时间，大于该时间的请求将被记录
--timeout <T> Socket/request timeout
post
- 准备一个lua脚本文件
wrk.method = "POST"
wrk.headers["Content-Type"] = "application/json"
wrk.body = '{"valuationPrice":100000}'
通过wrk -t10 -c2000 -d60s -T5s --script=post.lua --latency http://192.168.31.107/user/login指定post.lua进行压测
自定义函数
wrk.method = "POST"
wrk.headers["Content-Type"] = "application/json"
wrk.body = '{"valuationPrice":100000}'
function response(status, headers, body)
-- print(tostring(status))
writeFile("wrk.log","status:"..tostring(status)..",body:"..body.."\n")
end
function writeFile(fileName,content)
local f = assert(io.open(fileName,'a+'))
f:write(content)
f:close()
end
通过wrk -t10 -c2000 -d60s -T5s --script=post.lua --latency http://192.168.31.107/user/login进行压测
响应内容分析
Thread Stats Avg Stdev Max +/- Stdev
Latency 105.65ms 1.92ms 130.03ms 97.87%
Req/Sec 8.87 0.98 10.00 95.24%
189 requests in 20.08s, 21.43KB read
Socket errors: connect 0, read 5, write 0, timeout 1
Requests/sec: 9.41
Transfer/sec: 1.07KB
- Latency: 响应时间 Avg平均响应时间 Max最大响应时间
- Req/Sec: 每个线程每秒的请求数，Avg平均每秒的请求数 Max最大每秒请求数
- Requests/sec: 线程总共平均每秒钟完成的请求个数
- Transfer/sec: 每秒钟读取3.26MB数据量
- Non-2xx or 3xx responses: 状态码非2xx或3xx的请求数量
mac安装
brew install wrk
