Bug 639073

Summary: GFS1 vs GFS2 performance issue [rhel-5.5.z]
Product: Red Hat Enterprise Linux 5 Reporter: RHEL Program Management <pm-rhel>
Component: kernelAssignee: Jiri Pirko <jpirko>
Status: CLOSED ERRATA QA Contact: Red Hat Kernel QE team <kernel-qe>
Severity: urgent Docs Contact:
Priority: urgent    
Version: 5.3CC: adas, al, bmarson, bmarzins, brsmith, casmith, cww, dhoward, djansa, eguan, iannis, jkortus, jpirko, jwilleford, kai, pm-eus, rkhan, rpeterso, rwheeler, scooter, swhiteho, tao, teigland, ttracy, yugzhang
Target Milestone: rcKeywords: ZStream
Target Release: ---   
Hardware: All   
OS: Linux   
Whiteboard:
Fixed In Version: Doc Type: Bug Fix
Doc Text:
In a two node cluster, the lock master of two folders can move 100 files from one folder to the other in less than 1 second. If a server is not the lock master for that folder, it would take that server 3-5 seconds to perform the same task on GFS1 (Global File System 1), and 30-50 seconds on GFS2 (Global File System 2). With this update, the aforementioned task takes less than 1 second on GFS1 and about 3 seconds on GFS2.
Story Points: ---
Clone Of: Environment:
Last Closed: 2010-11-09 18:08:03 UTC Type: ---
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: 504188    
Bug Blocks:    

Description RHEL Program Management 2010-09-30 18:37:27 UTC
This bug has been copied from bug #504188 and has been proposed
to be backported to 5.5 z-stream (EUS).

Comment 4 Jiri Pirko 2010-10-11 08:55:31 UTC
in kernel 2.6.18-194.20.1.el5

linux-2.6-fs-dlm-fix-try-1cb-failure-part-2.patch
linux-2.6-fs-dlm-no-node-callback-when-try-1cb-lock-req-fails.patch

Comment 6 Eryu Guan 2010-11-09 02:18:09 UTC
Confirmed patches listed in comment 4 are applied in kernel 2.6.18-194.26.1.el5 correctly. Set SanityOnly verified.

Comment 9 errata-xmlrpc 2010-11-09 18:08:03 UTC
An advisory has been issued which should help the problem
described in this bug report. This report is therefore being
closed with a resolution of ERRATA. For more information
on therefore solution and/or where to find the updated files,
please follow the link below. You may reopen this bug report
if the solution does not work for you.

http://rhn.redhat.com/errata/RHSA-2010-0839.html

Comment 10 Martin Prpič 2010-11-11 13:56:17 UTC
    Technical note added. If any revisions are required, please edit the "Technical Notes" field
    accordingly. All revisions will be proofread by the Engineering Content Services team.
    
    New Contents:
In a two node cluster, the lock master of two folders can move 100 files from one folder to the other in less than 1 second. If a server is not the lock master for that folder, it would take that server 3-5 seconds to perform the same task on GFS1 (Global File System 1), and 30-50 seconds on GFS2 (Global File System 2). With this update, the aforementioned task takes less than 1 second on GFS1 and about 3 seconds on GFS2.