Bug 1880354

Summary:	cluster-image-registry-operator: Fix bug in reflector not recovering from "Too large resource version"
Product:	OpenShift Container Platform	Reporter:	Oleg Bulatov <obulatov>
Component:	Image Registry	Assignee:	Oleg Bulatov <obulatov>
Status:	CLOSED ERRATA	QA Contact:	Wenjing Zheng <wzheng>
Severity:	high	Docs Contact:
Priority:	high
Version:	4.5	CC:	aos-bugs, lszaszki, wzheng
Target Milestone:	---
Target Release:	4.6.0
Hardware:	Unspecified
OS:	Unspecified
Whiteboard:
Fixed In Version:		Doc Type:	Bug Fix
Doc Text:	Cause: the operator does not handle "Too large resource version" properly Consequence: when this error happens, the operator stops getting events from the cluster Fix: bump client-go Result: the operator can recover from this error	Story Points:	---
Clone Of:	1880314	Environment:
Last Closed:	2020-10-27 16:42:20 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:	1880369
Bug Blocks:	1880314

Description Oleg Bulatov 2020-09-18 10:42:45 UTC

+++ This bug was initially created as a clone of Bug #1880314 +++

A recent fix in the reflector/informer https://github.com/kubernetes/kubernetes/pull/92688 prevents components/operators from entering a hotloop and stuck.

There are already reported cases that have run into that issue and were stuck for hours or even days. For example https://bugzilla.redhat.com/show_bug.cgi?id=1877346.

The root cause of the issue is the fact that a watch cache is initialized from the global revision (etcd) and might stay on it for an undefined period (if no changes were (add, modify) made). 
That means that the watch cache across server instances may be out of sync. 
That might lead to a situation in which a client gets a resource version from a server that has observed a newer rv, disconnect (due to a network error) from it, and reconnect to a server that is behind, resulting in “Too large resource version“ errors.

More details in https://github.com/kubernetes/kubernetes/issues/91073 and https://github.com/kubernetes/enhancements/tree/master/keps/sig-api-machinery/1904-efficient-watch-resumption


It looks like the issue only affects 1.18. According to https://github.com/kubernetes/kubernetes/issues/91073#issuecomment-652251669 the issue was first introduced in that version by changes done to the reflector. 
The fix is already present in 1.19.


Please make sure that cluster-image-registry-operator is using a client-go that includes https://github.com/kubernetes/kubernetes/pull/92688 if not please use this BZ and file a PR.
In case you are using a framework to build your operator make sure it uses the right version of the client-go library.

Comment 1 Oleg Bulatov 2020-09-18 10:44:18 UTC

4.6.0 already has changes from https://github.com/kubernetes/kubernetes/pull/92688:

https://github.com/kubernetes/kubernetes/pull/92688/files#diff-df2b34adba86da0485290415620d737aR86

https://github.com/openshift/cluster-image-registry-operator/blob/4a8f3c1d5d6a25b77b7ed28af1fa06a9afe64019/vendor/k8s.io/client-go/tools/cache/reflector.go#L86

Comment 3 Wenjing Zheng 2020-09-21 09:28:24 UTC

$ oc logs pods/cluster-image-registry-operator-fc454db88-4pbw2 | grep "Timeout: Too large resource version"

Verified on 4.6.0-0.nightly-2020-09-20-184226.

Comment 7 errata-xmlrpc 2020-10-27 16:42:20 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