Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Add option to set PodDisruptionBudget #1764

Closed
xoxys opened this issue Sep 9, 2023 · 7 comments · Fixed by #1772
Closed

Add option to set PodDisruptionBudget #1764

xoxys opened this issue Sep 9, 2023 · 7 comments · Fixed by #1772
Labels

Comments

@xoxys
Copy link

xoxys commented Sep 9, 2023

Is your feature request related to a problem? Please describe.

It's not possible to override the default PodDisruptionBudget while the default might not fit for every use case.

Describe the solution you'd like

Add an option to set a custom PodDisruptionBudget or disable it.

@jiuker
Copy link
Contributor

jiuker commented Sep 10, 2023

Under which case you need to disable this? @xoxys

@xoxys
Copy link
Author

xoxys commented Sep 10, 2023

I'm using a single server multi disk tenant on a small setup. During node drains, I would still like to get the tenant started on another node even if this results in a short downtime, which is acceptable in my case. The current behavior is that during node drains, the hard-coded PodDisruptionBudget blocks the node drain and the pod is not automatically transferred to another node. You can still keep the default, but is it required to enforce it for everyone?

@jiuker
Copy link
Contributor

jiuker commented Sep 11, 2023

I think this is acceptable, you can now manually delete this resource because this resource is mandatory for cluster health to create.

@jiuker jiuker closed this as completed Sep 11, 2023
@xoxys
Copy link
Author

xoxys commented Sep 11, 2023

I don't understand your answer. Manually deleting is pointless if the operator will ensure it again? So basically the operator does only support multi-server deployments and is not intended to be used for single-server deployments?

@jiuker
Copy link
Contributor

jiuker commented Sep 11, 2023

I don't understand your answer. Manually deleting is pointless if the operator will ensure it again? So basically the operator does only support multi-server deployments and is not intended to be used for single-server deployments?

Yes, a single-node minio cluster, which we consider to be a cluster that does not have production properties. @xoxys . BTW, @harshavardhana @pjuarezd cc

@D1StrX
Copy link

D1StrX commented Sep 16, 2023

MinIO in a k8s cluster obstructs the draining in 1.27.x as well
(will retry after 5s): Cannot evict pod as it would violate the pod's disruption budget.
This is a deployment of 2 pods spread on 2/3 workers. It doesn't delete the pod on the draining node, and draining command logs the above.

Will the PR also fix the issue for the environment outlined above?

@jiuker
Copy link
Contributor

jiuker commented Sep 18, 2023

@D1StrX Yes. It will not create for servers are 1 or 2.

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
Projects
None yet
Development

Successfully merging a pull request may close this issue.

4 participants