Jack Lin

The questions here are For `e2e-test-volume-0` (single replica) 1. When the s3 got deleted due to reboot, the backup should failed with error message, why was it stuck 2. The...

I tried to reproduce the issue but failed to make the backup stuck in `inProgress` state ### delete the minio and the longhorn-manager when doing back up 1. Upload resumed...

I will keep investigating the issue. From my perspective, If we want to test negative effect on backup, we don't need recurring job here because it doesn't make any different...

So it doesn't related to single replicas or multiple replicas We reboot one node On that node, there could have 1. minio 2. replica which was requested by the engine...

@innobead We haven't found the root cause of this issue The PR @derekbit created is not the fix for this issue, but just refactor another error handling minor issue found...

I can reproduce the issue now, need to spend some time investigating 1. Create a 3Gi Volume (I use 1 replica, but it doesn't matter) and write 3Gi data 2....

The reproduce steps are not correct After a while(~17mins, after the s3 put retry backoff) the backup resumed and failed due to ``` cannot find backupstore/volumes/5b/d5/nginx/volume.cfg in backupstore ``` Still...

This issue has been present since Day 1 and can occur in any version of Longhorn. The symptom is that **the backup fails in the replica process (check the logs),...

Hi @chriscchien Could it be possible that the image was not updated because in the PR I even remove the logging message https://github.com/longhorn/backing-image-manager/pull/166/files So it is strange that it still...

PR: https://github.com/longhorn/backing-image-manager/pull/163