Node Leaves the cluster after POD restart

Description

Mongo Node leaves the cluster once pod restarts - 

```sh
2020-10-01T15:33:09.866+0000 I REPL [replexec-0] This node is not a member of the config
2020-10-01T15:33:09.866+0000 I REPL [replexec-0] transition to REMOVED from STARTUP
2020-10-01T15:33:09.866+0000 I NETWORK [replexec-0] Skip closing connection for connection # 2
```

Environment

None

Smart Checklist

Activity

Lalit Choudhary November 12, 2020 at 1:17 PM

I also tested the same on GEK with a 1.5 version operator. Don't see issue after pod restart.

 

 

Slava Sarzhan October 20, 2020 at 1:32 PM

Hi  , 

Which version of psmdb do you use?  Also could you please provide your CR.

Thank you.

Devesh Kumar October 2, 2020 at 9:48 AM

Same issue with 1.3.0

Devesh Kumar October 2, 2020 at 9:33 AM

The same bug is there in 1.4.0 also, when a pod gets restarted we get following error - 

Heartbeat to 172.26.2.204:27017 failed after 2 retries, response status: InvalidReplicaSetConfig: Our replica set configuration is invalid or does not include us

 

When you connect to the cluster and remove that node manually from the rs config, the it again added by the operator to the cluster and works fine. Same scenario  is there for 1.5.0

Cannot Reproduce

Details

Assignee

Reporter

Labels

Time tracking

40m logged

Affects versions

Priority

Smart Checklist

Created October 1, 2020 at 3:37 PM
Updated March 5, 2024 at 5:04 PM
Resolved November 23, 2020 at 10:57 AM