Bug 2009423

Summary:	fs: dlm: dlm_callback_resume is too noisy
Product:	Red Hat Enterprise Linux 9	Reporter:	Alexander Aring <aahringo>
Component:	kernel	Assignee:	Alexander Aring <aahringo>
kernel sub component:	DLM	QA Contact:	cluster-qe <cluster-qe>
Status:	CLOSED ERRATA	Docs Contact:
Severity:	low
Priority:	medium	CC:	aahringo, dhowells, gfs2-maint, mszeredi, swhiteho, teigland, xzhou
Version:	CentOS Stream	Keywords:	Triaged
Target Milestone:	rc
Target Release:	---
Hardware:	Unspecified
OS:	Unspecified
Whiteboard:
Fixed In Version:	kernel-5.14.0-148.el9	Doc Type:	If docs needed, set a value
Doc Text:		Story Points:	---
Clone Of:		Environment:
Last Closed:	2022-11-15 10:51:25 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:

Description Alexander Aring 2021-09-30 15:44:26 UTC

Description of problem:

The kernel log buffer gets filled up with a lot of messages like:

[241767.849302] dlm: bin: dlm_callback_resume 25
[241767.854846] dlm: bin: dlm_callback_resume 25
[241767.860373] dlm: bin: dlm_callback_resume 25
...
[241767.865920] dlm: bin: dlm_callback_resume 25
[241767.871352] dlm: bin: dlm_callback_resume 25
[241767.876733] dlm: bin: dlm_callback_resume 25

it is not an error. As DLM recovery stops callback handling by:

dlm_callback_suspend() every callback queue afterwards will be queued in a list until dlm_callback_resume() calls again. It seems lately in connection with GFS2 it becomes more load than usually for unknown reasons.

However we should make it quiet because we can never be sure about how many callbacks was be queued.

Version-Release number of selected component (if applicable):

How reproducible:

I can reproduce it by mount a gfs2 filesystem, run cityscape on all nodes and unmount all filesystems on all nodes at the same time.

However it depends on the load which is involved but it gets quite noisy with that procedure.

Actual results:

A lot of "[241767.871352] dlm: bin: dlm_callback_resume" messages in the kernel log. Which will fill the logs and confuses users.

Expected results:

Less logs with the same amount of information.

Additional info:

There is a patch prepared for upstream to accumulate all amounts of callback queues and print it out only once.

https://listman.redhat.com/archives/cluster-devel/2021-September/msg00107.html

Comment 15 Nate Straz 2022-08-24 14:54:45 UTC

Verified with multiple iterations of cityscape, umounting between iterations, on kernel-5.14.0-148.el9.x86_64

Comment 17 errata-xmlrpc 2022-11-15 10:51:25 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 (Moderate: kernel security, bug fix, and enhancement update), 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/RHSA-2022:8267