search

hpe-storage / python-hpedockerplugin

HPE Native Docker Plugin
Apache License 2.0
36 stars 64 forks source link

Multimaster 1.15 Statefulset : After reboot 2 PODs are in Completed State and 3 PODs are in Running state. #758

Closed sandesh-desai closed 4 years ago

sandesh-desai commented 4 years ago

Testbed Details: Host: -- K8s multimaster Host OS: CentOS K8s version: GitVersion:"v1.15.3

Steps performed: 1.Created SC,PVC,ConfigMap. 2.Created Statefulset with replicas=5 and all are in Running state on worker-1(cssosbe01-196150). 3.Scaled up to 12 PODs successfully and all PODs are in Running state. 4.Scaled down to 5 PODs successfully and all PODs are in Running state. 5.Rebooted worker-1. Expected Result: All PODs should be in Running State after the node came up.

Observation: Actual Result: 2 PODs are in Completed State and 3 PODs are in Running state.

Attaching the logs and output.

dcvlog_statefulset.txt dorylog_statefulset.txt output.txt

imran-ansari commented 4 years ago

PR https://github.com/hpe-storage/python-hpedockerplugin/pull/761 raised that describes the steps to recover PODs after rebooting a node

sandesh-desai commented 4 years ago

Verified and working fine.closing the bug. Attaching the logs:

Statefulset testing.txt