k8s 下有两个版本的nginx ingress  分别是 

ingress nginx   以及 nginx ingress 

Ingress-nginx 与 Nginx-ingress - LeoZhanggg - 博客园

这里我讨论的是 nginx ingress

Nginx Ingress 使用Prometheus 导出数据

nginx ingress 本身支持通过支持这个提供prometheus 格式的 metrics 供prometheus 抓取， 需要配置  -enable-prometheus-metrics。 如果需要响应时间则需要单独配置 -enable-latency-metrics. 

Prometheus | NGINX Ingress Controller

遇到的bug

一开始看上面的教程 在deployment 中设置好参数之后，发现没有latency 相关的内容。 一看 nginx 的log 就发现了 

2024/11/23 10:33:36 [alert] 44#44: connect() failed (13: Permission denied) while logging to syslog, server: unix:/var/lib/nginx/nginx-syslog.sock
2024/11/23 10:33:36 [alert] 46#46: connect() failed (13: Permission denied) while logging to syslog, server: unix:/var/lib/nginx/nginx-syslog.sock
2024/11/23 10:33:36 [warn] 46#46: *20 send() to syslog failed while logging request, client: 192.168.65.6, server: example.api.com, request: "OPTIONS /Auth/GetUserDetail HTTP/1.1", upstream: "http://10.1.1.143:5036/Auth/GetUserDetail", host: "example.api.com:30080", referrer: "http://example.itpp.com:30080/"
2024/11/23 10:33:36 [warn] 44#44: *16 send() to syslog failed while logging request, client: 192.168.65.6, server: example.api.com, request: "OPTIONS /Auth/GetUserDetail HTTP/1.1", upstream: "http://10.1.1.143:5036/Auth/GetUserDetail", host: "example.api.com:30080", referrer: "http://example.itpp.com:30080/"

因为 nginx 读取响应时间延迟是通过首先记录API到 syslog 再通过读取syslog 产生 metrics。 如果不做特殊配置的话，虽然nginx 能够运行起来但是实际上是没有配置好 syslog 的 需要配置相关 configmap 来指定好 nginx.conf  里面 syslog 相关

apiVersion: v1
kind: ConfigMap
metadata:name: nginx-config
data:nginx.conf: |worker_processes  1;events {worker_connections  1024;}http {include       mime.types;default_type  application/octet-stream;error_log syslog:server=unix:/var/lib/nginx/nginx-syslog.sock debug;access_log syslog:server=unix:/var/lib/nginx/nginx-syslog.sock combined;server {listen 8080;location /stub_status {stub_status;allow 127.0.0.1;allow 10.0.0.0/8;deny all;}location / {root /usr/share/nginx/html;index index.html;}}}

这样卸载 再重新安装就没有问题了。