Note: This bug is displayed in read-only format because the product is no longer active in Red Hat Bugzilla.

Bug 1727430

Summary:	CPU Spike casue files unavailable
Product:	[Community] GlusterFS	Reporter:	pandiarajan.alwargandhi
Component:	arbiter	Assignee:	Ravishankar N <ravishankar>
Status:	CLOSED INSUFFICIENT_DATA	QA Contact:
Severity:	urgent	Docs Contact:
Priority:	unspecified
Version:	4.1	CC:	bugs, pandiarajan.alwargandhi, ravishankar
Target Milestone:	---	Keywords:	Triaged
Target Release:	---
Hardware:	Unspecified
OS:	Linux
Whiteboard:
Fixed In Version:		Doc Type:	If docs needed, set a value
Doc Text:		Story Points:	---
Clone Of:		Environment:
Last Closed:	2019-08-05 10:04:43 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 pandiarajan.alwargandhi 2019-07-06 08:03:04 UTC

Description of problem:

CPU spike on any node causing files/volume unavailable

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

We have 3 node gluster, out of which we are using node 3 as arbiter
Node1 and Node2 version: 4.1.5
Node3 version:4.1.7

How reproducible:

Production envirnoment, so we need to wait until the issue occurs

Steps to Reproduce:
1.
2.
3.

Actual results:


Expected results:


Additional info:

Comment 1 Ravishankar N 2019-07-15 04:13:21 UTC

Could you provide some more details?
1. gluster volume info
2. what is the access mechanism (fuse?) and its gluster version
3. On which node is the CPU spike observed?
4. By files being unavailable, do you mean the file is missing when you try to access from the client or is it that the client appears to be hung?
5. Were there any self-heals going on at the time of the spike?
6. The next time this occurs, please capture the a) volume profile info and  b)`top -H <pid of the process that consumes cpu (I'm guessing it is the brick)`. You cab attach the results to this bug.

Note: It is advisable to have all servers and clients at the same gluster version.

Comment 2 Ravishankar N 2019-07-15 04:34:07 UTC

(In reply to Ravishankar N from comment #1)
> b)`top -H <pid of the process that consumes cpu (I'm guessing it is the
> brick)`. 
Sorry made a typo, this is just `top -H`

Comment 3 Ravishankar N 2019-08-05 10:04:43 UTC

Hi, I'm  closing this bug since I haven't heard from you. Please feel free to re-open with the information I requested/ steps to reproduce.

Comment 4 Red Hat Bugzilla 2023-09-14 05:31:25 UTC

The needinfo request[s] on this closed bug have been removed as they have been unresolved for 1000 days