Bug 1203172

Summary:	SMB: Iozone running for too long and hung after 13 hours on dis-rep volume accessed as samba share.
Product:	[Red Hat Storage] Red Hat Gluster Storage	Reporter:	surabhi <sbhaloth>
Component:	samba	Assignee:	rhs-smb <rhs-smb>
Status:	CLOSED ERRATA	QA Contact:	surabhi <sbhaloth>
Severity:	urgent	Docs Contact:
Priority:	urgent
Version:	rhgs-3.0	CC:	nlevinki, pgurusid, rcyriac, sbhaloth, vagarwal
Target Milestone:	---	Keywords:	ZStream
Target Release:	RHGS 3.0.4
Hardware:	Unspecified
OS:	Unspecified
Whiteboard:
Fixed In Version:		Doc Type:	Bug Fix
Doc Text:		Story Points:	---
Clone Of:		Environment:
Last Closed:	2015-03-26 06:37:12 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 surabhi 2015-03-18 10:27:36 UTC

Description of problem:
Iozone on samba share is too slow and on one instance getting hung.

Executed iozone on samba share with following settings for client and server threads:
client.event-threads: 4
server.event-threads: 4

On first instance where the server and clients are in different location , iozone ran for more than 13 hours and then got hung.

On second instance where the server and client are in same location iozone is running from 5 hours and still running.


Version-Release number of selected component (if applicable):
# rpm -qa | grep glusterfs
glusterfs-3.6.0.51-1.el6rhs.x86_64
samba-4.1.17-4.el6rhs.x86_64

How reproducible:
2/2


Steps to Reproduce:
1.Create dis-rep volume , set client.event-threads: 4
server.event-threads: 4 and access it as samba share on windows client
2.Run iozone on the share
3.Verify that iozone completed successfully without getting hung.

Actual results:
1. Iozone ran for 13 hours and get hung.
2. Iozone running from 5 hours and not yet completed.


Expected results:
Iozone should finish succesfully. Time taken should not be too long as compared to previous versions.

Additional info:
Machine details has been given to dev team for debugging the issue.

Comment 3 surabhi 2015-03-19 06:19:20 UTC

The iozone test has been run on previous 3.0.4 builds with samba4.1 as part of sanity and the execution didn't take this much of time.

Comment 5 Poornima G 2015-03-20 05:59:22 UTC

With the latest build glusterfs-3.6.0.53-1.el6rhs.x86_64, the issue is no more seen.
With this build, iozone test took around an hour to finish.

Tried with both cluster.consistent-metadata: on and off, there is no issue in either case.

Comment 6 surabhi 2015-03-24 12:44:01 UTC

With the latest build glusterfs-3.6.0.53-1.el6rhs.x86_64 the iozone finishes in expected time.there is no hung seen.
Marking the bug as verified.

Comment 8 errata-xmlrpc 2015-03-26 06:37:12 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://rhn.redhat.com/errata/RHBA-2015-0682.html