Context

In order for plugins to inform better the kubelet for scheduling and scaling, we need to include a readiness and liveness probes for the plugins backend. The plugins already include a readiness probe, we are missing the liveness probe

Outcome

• All the observability plugins deployments in COO have a liveness probe checking for that the backend process is running
• All the observability plugins deployments use the /health endpoint as a readiness probe

Steps

• Change in COO the deployment configuration to add the liveness and readiness probes
• The liveness probe is a command that checks that the backend process is running. Is not an HTTP request
• Check timeouts for the probes