https://grafana.com/docs/k6/latest/get-started

install and run

1. install

# mac
brew install k6

2. 當前目錄下生成壓測腳本

# create file script.js
k6 new [filename]     # create file ‘script.js’ in the current directory

3 run

k6 run script.js

4. reports
   默認情況下，k6 將總結結果打印到 stdout 。

script.js

import http from 'k6/http';
import { check, sleep } from 'k6';export const options = {vus: 10,duration: '30s',
};export default function () {const res = http.get('http://test.k6.io');check(res, { 'status was 200': (r) => r.status == 200 });sleep(1);
}

options

https://grafana.com/docs/k6/latest/using-k6/k6-options/reference/

• vus
  虛擬用戶數
• duration
  持續時間
• rps
  每秒請求數量

請求總數 = vus * rps * durance（s）

import http from 'k6/http';
import { check, sleep } from 'k6';export const options = {maxRedirects: 4,duration: '300s',vus: 10,rps: 300
};// stages : 逐步提升/降低 
export const options = {stages: [{ target: 200, duration: '30s' },{ target: 0, duration: '30s' },],
};export const options = {stages: [{ duration: '10s', target: 100 },{ duration: '5m', target: 100 },{ duration: '10s', target: 0 },],rps: 100,
};

最佳實踐

VUs太大的情況下, 直接啟動所有 VUs 會幾乎同時發起請求，導致請求的瞬間激增。這樣的突發性負載可能會導致系統未能及時響應，進而出現錯誤。
改為 分階段增加 VUs , 系統逐步適應增加的負載

report 解析

默認report

         /\      Grafana   /￣￣/  /\  /  \     |\  __   /  /   /  \/    \    | |/ /  /   ￣￣\ /          \   |   (  |  (￣)  |/ __________ \  |_|\_\  \_____/ execution: localscript: script.jsoutput: -scenarios: (100.00%) 1 scenario, 2000 max VUs, 1m15s max duration (incl. graceful stop):* default: Up to 2000 looping VUs for 45s over 6 stages (gracefulRampDown: 30s, gracefulStop: 30s)? status is 200checks.........................: 100.00% 191878 out of 191878data_received..................: 160 MB  3.5 MB/sdata_sent......................: 20 MB   443 kB/shttp_req_blocked...............: avg=5.84μs   min=0s    med=1μs      max=9.49ms  p(90)=4μs      p(95)=7μs     http_req_connecting............: avg=2.71μs   min=0s    med=0s       max=9.43ms  p(90)=0s       p(95)=0s      http_req_duration..............: avg=246.16ms min=383μs med=103.16ms max=6.91s   p(90)=617.17ms p(95)=979.14ms{ expected_response:true }...: avg=246.16ms min=383μs med=103.16ms max=6.91s   p(90)=617.17ms p(95)=979.14mshttp_req_failed................: 0.00%   0 out of 191878http_req_receiving.............: avg=40.53μs  min=5μs   med=18μs     max=65.69ms p(90)=74μs     p(95)=102.14μshttp_req_sending...............: avg=10.06μs  min=1μs   med=3μs      max=53.75ms p(90)=15μs     p(95)=23μs    http_req_tls_handshaking.......: avg=0s       min=0s    med=0s       max=0s      p(90)=0s       p(95)=0s      http_req_waiting...............: avg=246.11ms min=373μs med=103.09ms max=6.91s   p(90)=617.11ms p(95)=979.11mshttp_reqs......................: 191878  4263.707029/siteration_duration.............: avg=246.24ms min=406μs med=103.26ms max=6.91s   p(90)=617.2ms  p(95)=979.24msiterations.....................: 191878  4263.707029/svus............................: 7       min=7                max=1990vus_max........................: 2000    min=2000             max=2000running (0m45.0s), 0000/2000 VUs, 191878 complete and 0 interrupted iterations
default ? [======================================] 0000/2000 VUs  45s

總請求量 191878, 持續時間 45.0s
p(95)=979.14ms : 95% 的請求響應時間在 979.14ms 以內