Bug 1848633 - Snapshot e2es are leaving namespaces in terminating state
Summary: Snapshot e2es are leaving namespaces in terminating state
Keywords:
Status: CLOSED ERRATA
Alias: None
Product: OpenShift Container Platform
Classification: Red Hat
Component: Storage
Version: 4.6
Hardware: Unspecified
OS: Unspecified
unspecified
medium
Target Milestone: ---
: 4.6.0
Assignee: Christian Huffman
QA Contact: Wei Duan
URL:
Whiteboard:
Depends On:
Blocks:
TreeView+ depends on / blocked
 
Reported: 2020-06-18 16:21 UTC by Hemant Kumar
Modified: 2020-10-27 16:08 UTC (History)
5 users (show)

Fixed In Version:
Doc Type: If docs needed, set a value
Doc Text:
Clone Of:
Environment:
Last Closed: 2020-10-27 16:08:08 UTC
Target Upstream Version:
Embargoed:


Attachments (Terms of Use)


Links
System ID Private Priority Status Summary Last Updated
Red Hat Product Errata RHBA-2020:4196 0 None None None 2020-10-27 16:08:29 UTC

Description Hemant Kumar 2020-06-18 16:21:45 UTC
Snapshot tests are leaving namespaces behind as discovered in - https://bugzilla.redhat.com/show_bug.cgi?id=1814282

This is a follow up https://bugzilla.redhat.com/show_bug.cgi?id=1814282 because we want to fix snapshot e2es via a different PR.

Comment 13 Wei Duan 2020-08-13 02:15:43 UTC
Hi Huffman,

I performed two scenarios as below:

1. After creating volumesnapshot successfully --> Delete volumesnapshotclass --> Delete namespace, the namespace was deleted successfully
 
2. When this is a not READYTOUSE volumesnapshot exists --> Delete volumesnapshotclass --> Delete namespace, the namespace was is always in "Terminating" 
It should linkd to another issue like bug 1867167 or 1842747 (BTW, do you think they are the same root cause?)
$ oc get project wduan-test
NAME         DISPLAY NAME   STATUS
wduan-test                  Terminating

$ oc get volumesnapshot -n wduan-test
NAME            READYTOUSE   SOURCEPVC   SOURCESNAPSHOTCONTENT   RESTORESIZE   SNAPSHOTCLASS           SNAPSHOTCONTENT   CREATIONTIME   AGE
mysnapshot-01   false        mypvc01                                           csi-aws-ebs-snapclass                                    25m

Please let me know if this fix is only for scenario 1? Thanks.

Comment 14 Wei Duan 2020-08-13 03:44:56 UTC
Nightly used.
4.6.0-0.nightly-2020-08-10-200500

Comment 15 Christian Huffman 2020-08-13 13:14:08 UTC
You're correct that this bug is only for the first scenario.

The second scenario can have a couple of different causes, such as those tracked by the linked bugs. My understanding of the bugs is below:

- BZ1867167 - When the underlying PV is not created by a CSI driver, we never remove the source finalizer from the snapshot, as it never finishes creation. There's a WIP PR to address this that checks the volume type and will remove the finalizer.
- BZ1842747 - In this case the PV is from a correct driver, but not yet bound. This requires a different fix than BZ1867167, as there's been discussion upstream regarding how to handle this behavior.

Comment 16 Wei Duan 2020-08-13 14:11:29 UTC
Thanks for the explaination, it is more clear than reading all the comments for this bug and original bug.
I will mark "Verified" as we discussed above.

Comment 18 errata-xmlrpc 2020-10-27 16:08:08 UTC
Since the problem described in this bug report should be
resolved in a recent advisory, it has been closed with a
resolution of ERRATA.

For information on the advisory (OpenShift Container Platform 4.6 GA Images), and where to find the updated
files, follow the link below.

If the solution does not work for you, open a new bug report.

https://access.redhat.com/errata/RHBA-2020:4196


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