Scavenge incorrectly marked as failed

[rmfaheem]

Describe the bug
When the auto-scavenge plugin is scavenging a node and that node is restarted, the scavenge will be marked as failed on the restart, and yet it will resume and continue until completion.

To Reproduce
Steps to reproduce the behavior:

1. Enable auto-scavenge
2. Kill the node with the scavenge in-progress
3. Observe the logs/scavenge history

Expected behavior
Not mark the scavenge as failed, and resume from where it left off.

Actual behavior
Scavenge is marked as failed.

Config/Logs/Screenshots
If applicable, please attach your node configuration, logs or any screenshots.

KurrentDB details

• KurrentDB server version: 25.0.1

[linear]

DB-804 Scavenge incorrectly marked as failed

[timothycoleman]

Yeah this seems to be a bug. On startup it looks for incomplete scavenges and marks them as failed. However in this case the autoscavenge has managed to start a scavenge already (much more likely than a human doing the same thing) which has resulted in that scavenge being marked as failed even though it is still running.

It won't affect the running of the scavenge fortunately