[tempo-distributed] Container is missing readiness or liveness probe

[alita1991]

The Tempo service is currently lacking proper probes, leading to potential issues with monitoring, health checks, and graceful handling of service disruptions. Probes play a critical role in ensuring the stability and reliability of services within our Kubernetes cluster.

This issue was discovered while scanning the helm chart via kube-score: https://github.com/zegl/kube-score/blob/master/README_PROBES.md

The components with missing readiness probes are:

• apps/v1/StatefulSet tempo-memcached
• apps/v1/Deployment tempo-query-frontend
• apps/v1/Deployment tempo-compactor

The components with missing liveness probes are:

• apps/v1/StatefulSet tempo-ingester
• apps/v1/Deployment tempo-distributor
• apps/v1/Deployment tempo-querier

Is there any reason for not having probes for the given components? If not, what would be the recommendation?

[hkailantzis]

@alita1991 thanks for opening this. Is this something that could be also done for the monolithic version (https://github.com/grafana/helm-charts/tree/main/charts/tempo) ? Or make sense to open another issue for this? Thanks in advance!

[Mistral-valaise]

Hi,

I'm experiencing the same issue with the Grafana Tempo Helm chart (https://github.com/grafana/helm-charts/tree/main/charts/tempo). I've also created an issue to address this problem. https://github.com/grafana/helm-charts/issues/3185