Linkerd2 helm chart installation and Rancher2 monitoring integration.

[hetii]

Hello.

I use terraform+rancher2 to provision my cluster.

Rancher2 has own monitoring chart that provision grafana and prometheus instances.

My linkerd and linkerd-cni is installed via helm charts and I disable in its value file grafana and prometheus installation.

Now to reuse monitoring from rancher2 in my other deployments usually I enable serviceMonitor in
cattle-prometheus namespace. Then the metrics was automatically scrapped by Prometheus instance.

For linkerd helm chart I don't see such possibilities, so any clue how this integration should look like?

Also do I need then install some special plugins/charts to this existing grafana?

[grampelberg]

You'll want to copy the scrape config and import the grafana dashboards. That should be it though!

[hetii]

Ok Thank you for this tip, I will check it.

[hetii]

Btw witch chart are linkerd specific that should be imported?

[hetii]

Small update from my side.
I get secret in rancher called prometheus-cluster-monitoring-additional-scrape-configs where rancher2 monitoring store scrape configs for istio and I add there also scrape for linkerd-proxy.
After applying new secret in prometheus target i see linkerd-proxy (8/8 up).
In service discovery tab I see also: linkerd-proxy (8/79 active targets)

Then when I try search some metrics based on https://linkerd.io/2/reference/proxy-metrics I get some statistic but
but linkerd stat deployments -A show stats like that:

NAMESPACE                 NAME                      MESHED   SUCCESS   RPS   LATENCY_P50   LATENCY_P95   LATENCY_P99   TCP_CONN
linkerd                   linkerd-controller           1/1         -     -             -             -             -          -
linkerd                   linkerd-destination          1/1         -     -             -             -             -          -
linkerd                   linkerd-identity             1/1         -     -             -             -             -          -
linkerd                   linkerd-proxy-injector       1/1         -     -             -             -             -          -
linkerd                   linkerd-sp-validator         1/1         -     -             -             -             -          -
linkerd                   linkerd-tap                  1/1         -     -             -             -             -          -
linkerd                   linkerd-web                  1/1         -     -             -             -             -          -

So what I missing there and how linkerd cli know with instance of prometheus it should use to display properly it's stats?