Aller au contenu

GarbageCollectorInactive#

Alert is triggered when harbor_jobservice_task_total metric is missing from Prometheus.

This alert either indicates that :

  • the pod harbor or jobservice are not in a running state in the kubernetes cluster
  • or they can't purge unused data.

Scope#

The scope of the alert is for Product Factory internal tooling.

Troubleshooting Hints#

Identify the cluster and check the status/logs of the pods.

Tip

   kubectl get pods -n harbor-svc-oci-registry-stg | grep harbor
   kubectl get pods -n harbor-svc-oci-registry-stg | grep jobservice
   kubectl logs -n harbor-svc-oci-registry-stg -l app=harbor,component=core | grep -i "garbage"
   kubectl logs -n harbor-svc-oci-registry-stg -l app=harbor,component=jobservice | grep -i "garbage"

   kubectl get pods -n harbor-svc-oci-registry-prd | grep harbor
   kubectl get pods -n harbor-svc-oci-registry-prd | grep jobservice
   kubectl logs -n harbor-svc-oci-registry-prd -l app=harbor,component=trivy | grep -i "garbage"
   kubectl logs -n harbor-svc-oci-registry-prd -l app=harbor,component=jobservice | grep -i "garbage"

Solutions Hints#

Restart the trivy and the jobservice container and check again.