Fix and add some more metrics for Grafana and Thanos
- Add k8s metrics for
thanos-query
- Fix the apdex metric for
thanos-query-frontend
- Breakdown by pod instead or in addition of FQDN where applicable
- Update some Thanos selectors now that it's all migrated to k8s (also the
thanos-store
SLI was picking up metrics from the Thanos sidecar because of this until now) - Remove some outdated/broken links to Elastic APM
- Add 3 SLIs for Grafana:
- HTTP requests to Grafana
- HTTP requests from Grafana to Grafana Image Renderer
- HTTP requests from Grafana to its datasources
Part of https://gitlab.com/gitlab-com/gl-infra/reliability/-/issues/15257
Edited by Pierre Guinoiseau