concourse: Clean up noisy 'failed-to-transition' log during volume GC
Bug Report
Nov 03 20:09:38 somewhere.io docker[3779]: {"timestamp":"1509739778.442883492","source":"atc","message":"atc.volume-collector.run.orphaned-volumes.mark-created-as-destroying.failed-to-transition","log_level":2,"data":{"error":"volume cannot be destroyed as children are present","session":"26.14766.1.3","volume":"480df3bf-97f8-4b3b-6b15-464eea436eb5","worker":"primaryworker"}}
The work dir is btrfs based.
The following can also be handy:
- Concourse version: 3.6.0
- Deployment type (BOSH/Docker/binary): binary
- Infrastructure/IaaS:
- Browser (if applicable):
- Did this used to work? yes
About this issue
- Original URL
- State: closed
- Created 7 years ago
- Reactions: 5
- Comments: 16 (8 by maintainers)
Also seeing this on our 3.8.0 k8s deployment
Any update on this issue? I’m trying to diagnose a volume inflation on the 8 workers we have in our concourse instance. I’m constantly having to retire workers and nuke the folder the workers use on the file system. The volume count never goes down and always increases on all workers. I’m trying to investigate the database rows to see if I can find any correlation between anything.
Any relation to this: https://github.com/concourse/concourse/issues/1419
@xtremerui we’re on 3.9 and get log messages like https://github.com/concourse/concourse/issues/1780#issuecomment-350853575 every 30 seconds. Let you know once we upgrade.