Allow specifying volume for workspace storage of the task-runner deployment
Summary
When trying to run a backup, since registry can't be skipped at the moment, the process requires about 50G of space to complete. However my nodes are only configured with 60G of space total, and the pod gets killed by the kubelet.
I do have ample space on remote storage though.
Current behavior
Events:
Type Reason Age From Message
---- ------ ---- ---- -------
Warning Evicted 10m kubelet, k002 The node was low on resource: ephemeral-storage. Container task-runner was using 24778672Ki, which exceeds its request of 0.
Normal Killing 10m kubelet, k002 Killing container with id docker://task-runner:Need to kill Pod
Expected behavior
Able to complete backups
Versions
- Chart: latest
- Platform:
- Self-hosted: Bootkube
- Kubernetes: (
kubectl version
)- Client: 1.12.2
- Server: 1.12.2
- Helm: (
helm version
)- Client: 2.11.0
- Server: 2.11.0
Relevant logs
.
.
.
done
Dumping registry ...
done
Dumping uploads ...
done
Dumping artifacts ...
done
Dumping lfs ...
WARNING: This version of GitLab depends on gitlab-shell 8.3.3, but you're running Unknown. Please update gitlab-shell.
Packing up backup tar
command terminated with exit code 137
rpc error: code = Unknown desc = Error: No such container: 66afb4df2850bb11182291023dc34500df9e6a1f6629a9603e744b9732e41a9d
Edited by Wei K Huang