+++ This bug was initially created as a clone of Bug #1886600 +++
panic: assignment to entry in nil map
goroutine 80 [running]:
k8s.io/apimachinery/pkg/util/runtime.HandleCrash(0x0, 0x0, 0x0)
k8s.io/client-go/tools/leaderelection/resourcelock.(*ConfigMapLock).Update(0xc0000f6000, 0x0, 0x0, 0x0, 0x0, 0x0, 0x0, 0x0, 0x0, 0x0, ...)
k8s.io/client-go/tools/leaderelection.(*LeaderElector).renew(0xc000f56b40, 0x1abc1a0, 0xc0004a4180)
k8s.io/client-go/tools/leaderelection.(*LeaderElector).Run(0xc000f56b40, 0x1abc1a0, 0xc0004a4140)
k8s.io/client-go/tools/leaderelection.RunOrDie(0x1abc1a0, 0xc00030de00, 0x1acbcc0, 0xc0000f6000, 0x14f46b0400, 0xa7a358200, 0x6fc23ac00, 0xc001060f30, 0xc0010baa10, 0x0, ...)
github.com/openshift/cluster-version-operator/pkg/start.(*Options).run.func4(0x1abc1a0, 0xc00030de00, 0xc0000f6000, 0xc000f3ab40, 0xc000099d80, 0x1abc1a0, 0xc00030dd80, 0xc00094b500, 0xc0010ba9d0)
created by github.com/openshift/cluster-version-operator/pkg/start.(*Options).run
--- Additional comment from Scott Dodson on 2020-10-09 00:09:02 UTC ---
Both of these are in 4.5 where we recently merged https://github.com/openshift/cluster-version-operator/pull/446 which is relevant to leader election, this is probably not a 4.6 blocker but we'll evaluate deeper tomorrow, leaving target release unassigned for now.
--- Additional comment from W. Trevor King on 2020-10-09 04:19:58 UTC ---
Looks like we need an upstream client-go fix backported. I'll try to figure out what our process is for that...
--- Additional comment from W. Trevor King on 2020-10-09 05:51:19 UTC ---
I've filed an upstream PR requesting a backport to 1.17.
--- Additional comment from Scott Dodson on 2020-10-09 13:51:13 UTC ---
This is 4.5 specific but the bot will revert my changes if i set target release to 4.5 directly, so setting it 4.7 now to get off the blocker list.
--- Additional comment from W. Trevor King on 2020-10-09 18:51:56 UTC ---
4.6+ use 1.19 clients which have this patched, so setting this modified to unblock a new-in-4.5 patch.
Here's the patch landing in 4.6+: https://github.com/openshift/cluster-version-operator/commit/fb84815ff1eb5ff21378965acc497b526b839f69#diff-c554e6c97e8d61bb9449784b8630b2abR91
I'm verifying it on 4.6.0-0.nightly-2020-10-10-041109.
Steps of verification are as below:
1. Fresh install an Azure cluster -- passed
2. Set up an OVN cluster on GCP with 4.5.14, and then do an upgrade to 4.6.0-0.nightly-2020-10-10-041109 - passed
There's no panic found in CVO, hence moving it to verified state.
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.