Fix Prometheus router's metrics

[tomMoulard]

What does this PR do?

This PR intends to solve 2 issues in the metrics exports on routers:

The first is when you try to export more than once the metrics of a router. The first time, it exports the correct value, and the second time, metrics are not exported.

Metrics are not exported the second time

$ curl -S localhost/
Hostname: whoami-78447d957f-n75xz
IP: 127.0.0.1
IP: ::1
IP: 10.42.0.7
IP: fe80::dc7f:11ff:feb4:5088
RemoteAddr: 10.42.0.6:33288
GET / HTTP/1.1
Host: localhost
User-Agent: curl/7.68.0
Accept: application/json, application/xml, text/plain, */*
Accept-Encoding: gzip
X-Forwarded-For: 10.42.0.1
X-Forwarded-Host: localhost
X-Forwarded-Port: 80
X-Forwarded-Proto: http
X-Forwarded-Server: traefik-788d95b68f-tn7h7
X-Real-Ip: 10.42.0.1

$ curl -S localhost:8080/metrics | grep simpleingressroute | grep router
traefik_router_open_connections{method="GET",protocol="http",router="default-simpleingressroute-5a0ff67b89637da8ca52@kubernetescrd",service="default-simpleingressroute-5a0ff67b89637da8ca52"} 0
traefik_router_request_duration_seconds_bucket{code="200",method="GET",protocol="http",router="default-simpleingressroute-5a0ff67b89637da8ca52@kubernetescrd",service="default-simpleingressroute-5a0ff67b89637da8ca52",le="0.1"} 4
traefik_router_request_duration_seconds_bucket{code="200",method="GET",protocol="http",router="default-simpleingressroute-5a0ff67b89637da8ca52@kubernetescrd",service="default-simpleingressroute-5a0ff67b89637da8ca52",le="0.3"} 4
traefik_router_request_duration_seconds_bucket{code="200",method="GET",protocol="http",router="default-simpleingressroute-5a0ff67b89637da8ca52@kubernetescrd",service="default-simpleingressroute-5a0ff67b89637da8ca52",le="1.2"} 4
traefik_router_request_duration_seconds_bucket{code="200",method="GET",protocol="http",router="default-simpleingressroute-5a0ff67b89637da8ca52@kubernetescrd",service="default-simpleingressroute-5a0ff67b89637da8ca52",le="5"} 4
traefik_router_request_duration_seconds_bucket{code="200",method="GET",protocol="http",router="default-simpleingressroute-5a0ff67b89637da8ca52@kubernetescrd",service="default-simpleingressroute-5a0ff67b89637da8ca52",le="+Inf"} 4
traefik_router_request_duration_seconds_sum{code="200",method="GET",protocol="http",router="default-simpleingressroute-5a0ff67b89637da8ca52@kubernetescrd",service="default-simpleingressroute-5a0ff67b89637da8ca52"} 0.002946457
traefik_router_request_duration_seconds_count{code="200",method="GET",protocol="http",router="default-simpleingressroute-5a0ff67b89637da8ca52@kubernetescrd",service="default-simpleingressroute-5a0ff67b89637da8ca52"} 4
traefik_router_requests_total{code="200",method="GET",protocol="http",router="default-simpleingressroute-5a0ff67b89637da8ca52@kubernetescrd",service="default-simpleingressroute-5a0ff67b89637da8ca52"} 4
$ curl -S localhost:8080/metrics | grep simpleingressroute | grep router
$

The second issue is related to the isOutdated method on the Prometheus exporter:
The exporter did not check if routers were still up-to-date, and if old routers needed to be removed from Prometheus metrics.

Motivation

Solve Prometheus metrics issues:

• Have correct router metrics
• Don't export metrics of removed routers

More

• Added/updated tests
• Added/updated documentation

Co-authored-by: Michael michael.matur@gmail.com
Co-authored-by: Romain rtribotte@users.noreply.github.com

[rtribotte]

LGTM

[jbdoumenjou]

LGTM