Bug 1880366

Summary:	[release 4.5] cluster-storage-operator: Fix bug in reflector not recovering from "Too large resource version"
Product:	OpenShift Container Platform	Reporter:	Lukasz Szaszkiewicz <lszaszki>
Component:	Storage	Assignee:	Jan Safranek <jsafrane>
Status:	CLOSED NOTABUG	QA Contact:	Qin Ping <piqin>
Severity:	high	Docs Contact:
Priority:	high
Version:	4.5	CC:	aos-bugs, jsafrane
Target Milestone:	---
Target Release:	4.5.z
Hardware:	Unspecified
OS:	Unspecified
Whiteboard:
Fixed In Version:		Doc Type:	If docs needed, set a value
Doc Text:		Story Points:	---
Clone Of:		Environment:
Last Closed:	2020-09-22 11:19:10 UTC	Type:	Bug
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:	1879901

Description Lukasz Szaszkiewicz 2020-09-18 11:08:42 UTC

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-storage-operator and its operands are 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 2 Jan Safranek 2020-09-22 11:19:10 UTC

cluster-storage-operator in 4.5.z uses client-go from kubernetes-1.12.5 (yes, that old. We rebased it in 4.6)

Comment 3 Jan Safranek 2020-09-22 11:24:18 UTC

Reading https://bugzilla.redhat.com/show_bug.cgi?id=1880309, there is no way we're updating k8s.io/client-go to 1.18.* in 4.5.z. The operator is really old, it uses "go dep", old operator-sdk and whatnot. On the positive side, it just creates one default storage class.