Bug 1636037

Summary:	Observed MDS failover and events in logs while dumping cache
Product:	[Red Hat Storage] Red Hat Ceph Storage	Reporter:	Ramakrishnan Periyasamy <rperiyas>
Component:	CephFS	Assignee:	Venky Shankar <vshankar>
Status:	CLOSED ERRATA	QA Contact:	Persona non grata <nobody+410372>
Severity:	medium	Docs Contact:	Erin Donnelly <edonnell>
Priority:	high
Version:	3.1	CC:	anharris, ceph-eng-bugs, edonnell, hnallurv, jbrier, pasik, pdonnell, rperiyas, tchandra, tserlin, vshankar
Target Milestone:	z1
Target Release:	3.2
Hardware:	Unspecified
OS:	Unspecified
Whiteboard:
Fixed In Version:	RHEL: ceph-12.2.8-64.el7cp Ubuntu: ceph_12.2.8-49redhat1	Doc Type:	Bug Fix
Doc Text:	.The Ceph MDS no longer allows dumping of cache larger than 1 GB Previously, if you attempted to dump a Ceph Metadata Server (MDS) cache with a size of around 1 GB or larger, the MDS could terminate unexpectedly. With this update, MDS no longer allows dumping of cache that size so the MDS no longer terminates in the described situation.	Story Points:	---
Clone Of:		Environment:
Last Closed:	2019-03-07 15:50:55 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:
Bug Blocks:	1629656

Description Ramakrishnan Periyasamy 2018-10-04 10:16:17 UTC

Description of problem:
Observed MDS failover when dumping cache using admin-daemon command.

command used: "ceph --admin-daemon ceph-mds.magna083.asok dump cache"

Command was hung for sometime and observed failover in cluster status. Command not exited.

Log: faildover mds log available in magna002 path "/home/rperiyas/fs_4thoct"

Version-Release number of selected component (if applicable):
ceph version 12.2.5-42.el7cp

How reproducible:
1/1

Steps to Reproduce:
1. Configured FS with 2 active and 1 standby
2. Mounted fuse and kernel and started smalfile and crefi IO on different dir
3. During IO's try to dump the cache using admin-daemon command.

Actual results:
MDS failover and dump cache command hung

Expected results:
There should not be a failover

Additional info:
When tried to perform the same active in another MDS, there was no failover of MDS and observed dump cache command output.

Comment 16 Patrick Donnelly 2019-01-04 20:30:58 UTC

*** Bug 1622243 has been marked as a duplicate of this bug. ***

Comment 27 errata-xmlrpc 2019-03-07 15:50:55 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, 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-2019:0475