Bug 1324465 - Can not know if the project is deleted in the delay period of deleting project after project controller delete the policy in the project
Summary: Can not know if the project is deleted in the delay period of deleting projec...
Keywords:
Status: CLOSED CURRENTRELEASE
Alias: None
Product: OKD
Classification: Red Hat
Component: Pod
Version: 3.x
Hardware: All
OS: Unspecified
medium
medium
Target Milestone: ---
: ---
Assignee: Cesar Wong
QA Contact: DeShuai Ma
URL:
Whiteboard:
Depends On:
Blocks:
TreeView+ depends on / blocked
 
Reported: 2016-04-06 11:31 UTC by Xiaolin
Modified: 2016-05-12 17:11 UTC (History)
5 users (show)

Fixed In Version:
Doc Type: Bug Fix
Doc Text:
Clone Of:
Environment:
Last Closed: 2016-05-12 17:11:20 UTC
Target Upstream Version:
Embargoed:


Attachments (Terms of Use)

Comment 1 Andy Goldstein 2016-04-06 13:41:43 UTC
Were pods actually stuck Terminating?

Do you have logs from the master and nodes?

Comment 6 Derek Carr 2016-04-07 14:27:07 UTC
The project name will not be released until the namespace is finally removed from Kubernetes, but we potentially can look to no longer charge the user for the project request limitation by not counting namespaces that are in terminating phase.

https://github.com/openshift/origin/blob/master/pkg/project/admission/requestlimit/admission.go#L114

This would mean that we would still be as latent as the watch notification propagating to that admission controller so the corresponding test scripts that drive the behavior may also have to change.

Comment 7 Xiaolin 2016-04-08 10:45:14 UTC
Add project request limit to config  and Verify on AMI rhel7_3934,it should be fixed and work well for our user case,
Thank you all.

$openshift version
openshift v1.1.6-38-g6ecbaaf
kubernetes v1.2.0-36-g4a3f9c5
etcd 2.2.5

But there is another issue,after do delete,create a project with the same name
immediately,it would fail,I wonder if we should think about this user case.

Comment 8 Cesar Wong 2016-04-08 13:19:06 UTC
Xiaolin, yes there will still be the limitation about reusing the name because the name has to be unique across the cluster.

Comment 9 Xiaolin 2016-04-11 05:05:30 UTC
Get it,have virified this fixed,thanks


Note You need to log in before you can comment on or make changes to this bug.