Bug 1840820
| Summary: | Additional PVCs created in openshift-monitoring namespace after upgrading from 4.3 to 4.4.4 | ||
|---|---|---|---|
| Product: | OpenShift Container Platform | Reporter: | Aditya Deshpande <adeshpan> |
| Component: | Monitoring | Assignee: | Pawel Krupa <pkrupa> |
| Status: | CLOSED DUPLICATE | QA Contact: | Junqi Zhao <juzhao> |
| Severity: | high | Docs Contact: | |
| Priority: | unspecified | ||
| Version: | 4.4 | CC: | aghadge, alegrand, anpicker, erooth, kakkoyun, lcosic, mloibl, pkrupa, surbania |
| Target Milestone: | --- | ||
| Target Release: | --- | ||
| Hardware: | Unspecified | ||
| OS: | Unspecified | ||
| Whiteboard: | |||
| Fixed In Version: | Doc Type: | If docs needed, set a value | |
| Doc Text: | Story Points: | --- | |
| Clone Of: | Environment: | ||
| Last Closed: | 2020-05-28 09:52:26 UTC | Type: | Bug |
| Regression: | --- | Mount Type: | --- |
| Documentation: | --- | CRM: | |
| Verified Versions: | Category: | --- | |
| oVirt Team: | --- | RHEL 7.3 requirements from Atomic Host: | |
| Cloudforms Team: | --- | Target Upstream Version: | |
| Embargoed: | |||
*** This bug has been marked as a duplicate of bug 1833427 *** |
Description of problem: After upgrading from OCP 4.3.21 to 4.4.4, there are additional PVCs getting created for prometheus and alertmanager though there were PVCs present before upgrade. Before upgrade: ~~~ # oc get pvc NAME STATUS VOLUME CAPACITY ACCESS MODES STORAGECLASS AGE prom-prometheus-k8s-0 Bound pvc-af243bba-8281-4617-965b-a9cf43adcf41 1Gi RWO gp2 4h59m prom-prometheus-k8s-1 Bound pvc-679cf5a9-83d7-4927-b15f-fbfecf824720 1Gi RWO gp2 4h59m # oc get cm cluster-monitoring-config -o yaml apiVersion: v1 data: config.yaml: | prometheusK8s: volumeClaimTemplate: metadata: name: prom <==== PVC-prefix-name spec: storageClassName: gp2 resources: requests: storage: 1Gi kind: ConfigMap ~~~ Here, it was already PVCs were present with PVC-prefix-name: "prom" But after upgrade to 4.4.4, it is observed that additional PVC with default PVC-prefix-name got created and new PVCs are getting used by the respective pods. After upgrade: ~~~ # oc get pvc NAME STATUS VOLUME CAPACITY ACCESS MODES STORAGECLASS AGE prom-prometheus-k8s-0 Bound pvc-af243bba-8281-4617-965b-a9cf43adcf41 1Gi RWO gp2 4h59m prom-prometheus-k8s-1 Bound pvc-679cf5a9-83d7-4927-b15f-fbfecf824720 1Gi RWO gp2 4h59m prometheus-k8s-db-prometheus-k8s-0 Bound pvc-ef25921b-1a59-4145-ad9f-14904f0feb97 1Gi RWO gp2 109m prometheus-k8s-db-prometheus-k8s-1 Bound pvc-6e8ea9fd-eb97-4a3a-a1ef-32319b615fa1 1Gi RWO gp2 109m # oc describe pod prometheus-k8s-0 Volumes: prometheus-k8s-db: Type: PersistentVolumeClaim (a reference to a PersistentVolumeClaim in the same namespace) ClaimName: prometheus-k8s-db-prometheus-k8s-0 ReadOnly: false ~~~ Version-Release number of selected component (if applicable): # oc get clusterversion NAME VERSION AVAILABLE PROGRESSING SINCE STATUS version 4.4.4 True False 69m Cluster version is 4.4.4 How reproducible: Always Steps to Reproduce: 1. deploy OCP v4.3.21 with customized PVC-prefix-name for prometheus and alertmanager pods with persistent storage. 2. upgrade to 4.4.4 using stable channel 3. check the pvc inside openshift-monitoring namespace after upgrade Actual results: Additional PVCs with default pvc-prefix-name are getting created though there are already present PVCs which are created before upgrade. Existing PVCs are not being used by the new prometheus and alertmanager pods. Expected results: Prometheus and alertmanager pods should use the existing PVCs which are created before upgrade. Also, there should not be any new creation of PVCs with default pvc-prefix-name as there are PVCs present already. Additional info: The cluster-monitoring-config configmpa inside the openshift-monitoring namespace is not changed during upgrade process.