Bug 2001796

Summary:	[4.8.2] NNCP creation failures after nmstate-handler pod deletion
Product:	Container Native Virtualization (CNV)	Reporter:	Petr Horáček <phoracek>
Component:	Documentation	Assignee:	Shikha Jhala <sjhala>
Status:	CLOSED NOTABUG	QA Contact:	Meni Yakove <myakove>
Severity:	high	Docs Contact:
Priority:	urgent
Version:	4.8.1	CC:	cnv-qe-bugs, gkapoor, myakove, oramraz, rhrazdil, ycui
Target Milestone:	---	Keywords:	Regression
Target Release:	4.8.2
Hardware:	Unspecified
OS:	Unspecified
Whiteboard:
Fixed In Version:		Doc Type:	Known Issue
Doc Text:	Cause: kubernetes-nmstate configures vlan trunks linux-bridge ports without using nmstatectl desiredState API. Consequence: This causes nmstatectl to fail when policy with linux-bridge re-reconciles (ie. on handler pod restart or policy update) Result: When this happens, the policy goes to Degraded state and the policy desiredState is rolled back. Workaround (if any): When policy is in degraded state due to the mismatching VLAN IDs, the issue may be resolved by re-triggerring reconciliation (for example by changing linux-bridge description).	Story Points:	---
Clone Of:	2000052
Clones:	2001901 2004527 (view as bug list)		Environment:
Last Closed:	2021-09-16 14:33:22 UTC	Type:	---
Regression:	---	Mount Type:	---
Documentation:	---	CRM:
Verified Versions:		Category:	---
oVirt Team:	---	RHEL 7.3 requirements from Atomic Host:
Cloudforms Team:	---	Target Upstream Version:
Embargoed:
Bug Depends On:	2000052
Bug Blocks:	2001901, 2004527

Comment 1 Petr Horáček 2021-09-07 08:23:27 UTC

We have a reason to believe that this knmstate bug severely affects 4.8.2, which is also built with 1.0.2-14. We are currently backporting a fix, it should be available by tomorrow.

I'm proposing this as a blocker.

Comment 2 Petr Horáček 2021-09-07 09:20:19 UTC

This bug already affects 4.8.1. The fix https://github.com/nmstate/kubernetes-nmstate/pull/793 is quite large and has not been tested on 4.9 yet, thus back-porting it to 4.8.2 is a gamble. On one hand we have this regression described in the BZ which is already present in 4.8.1. On other hand we may introduce a new regression with the backport (the fix was not tested on 4.9) and that would delay 4.8.2 even further.

Comment 3 Petr Horáček 2021-09-07 12:17:07 UTC

Turning this into a documentation BZ, so we cover the workaround in a release note. The resolution will be then tackled in 4.8.3 via https://bugzilla.redhat.com/show_bug.cgi?id=2001901.

Comment 4 ctomasko 2021-09-08 05:04:57 UTC

@sjhala Please add release note for maintenance release 4.8.2

Comment 5 Shikha Jhala 2021-09-15 13:39:37 UTC

@phoracek Docs PR is ready for review: https://github.com/openshift/openshift-docs/pull/36377. Thanks!

Comment 6 Shikha Jhala 2021-09-15 15:46:22 UTC

SME review is complete. @oramraz Can you or someone from your team please review the PR from QE perspective: https://github.com/openshift/openshift-docs/pull/36377. Thank you.

Comment 7 Radim Hrazdil 2021-09-16 14:33:22 UTC

We're unable to reproduce this issue on 4.8.2, closing.

Comment 8 Red Hat Bugzilla 2023-09-15 01:14:44 UTC

The needinfo request[s] on this closed bug have been removed as they have been unresolved for 500 days