Skip to:

  1. Skip to Jira Navigation
  2. Skip to Side Navigation
  3. Skip to Main Content

cannot run physical restores with non-voting/arbiter/delayed replica members

Description

There's this ticket for PBM: https://jira.percona.com/browse/PBM-979
After it gets fixed we need to do our fixes for these types of nodes and then re-test.

If we have at least one non-voting member in replica set it will be impossible to run restore - it will be stuck in requested state forever.

$ k get psmdb NAME ENDPOINT STATUS AGE my-cluster-name my-cluster-name-rs0.test.svc.cluster.local initializing 26m $ k get pods NAME READY STATUS RESTARTS AGE my-cluster-name-rs0-0 1/1 Running 0 12m my-cluster-name-rs0-1 1/1 Running 0 12m my-cluster-name-rs0-2 1/1 Running 0 13m my-cluster-name-rs0-nv-0 2/2 Running 0 26m percona-server-mongodb-operator-84675fc76f-fb6jb 1/1 Running 0 26m $ k get psmdb-restore NAME CLUSTER STATUS AGE restore1 my-cluster-name requested 14m

Normal replica set nodes were restarted with new sts, but non voting one was not (as you can see the "nv" pod still has two containers).

Environment

None

Activity

Slava Sarzhan September 21, 2023 at 6:34 PM

we need to remove these limitations from our doc:
https://docs.percona.com/percona-operator-for-mongodb/backups.html#physical

  • physical backups can’t be used in the cluster with arbiter nodes,

Done

Details

Assignee

Reporter

Labels

Needs QA

Yes

Needs Doc

Yes

Fix versions

Affects versions

Priority

Smart Checklist

Created February 28, 2023 at 11:40 AM
Updated March 5, 2024 at 4:32 PM
Resolved October 3, 2023 at 3:56 PM

Flag notifications