Bug 787307

Summary: dlm: dev_write no op messages
Product: Red Hat Enterprise Linux 6 Reporter: vmiszczak
Component: clusterAssignee: David Teigland <teigland>
Status: CLOSED WONTFIX QA Contact: Cluster QE <mspqa-list>
Severity: medium Docs Contact:
Priority: unspecified    
Version: 6.2CC: ccaulfie, cluster-maint, lhh, rpeterso, teigland
Target Milestone: rc   
Target Release: ---   
Hardware: Unspecified   
OS: Linux   
Whiteboard:
Fixed In Version: Doc Type: Bug Fix
Doc Text:
Story Points: ---
Clone Of: Environment:
Last Closed: 2012-02-10 18:28:13 UTC Type: ---
Regression: --- Mount Type: ---
Documentation: --- CRM:
Verified Versions: Category: ---
oVirt Team: --- RHEL 7.3 requirements from Atomic Host:
Cloudforms Team: --- Target Upstream Version:

Description vmiszczak 2012-02-03 21:27:23 UTC 
Description of problem:
I'm using a 2 nodes ctdb cluster. When ctdb is running on both nodes, all consoles of one node are spammed with "dlm: dev_write no op xxxxx xxxxx" messages and this is logged to /var/log/messages (only on one node at a time, but this can be the first or the second across ctdb restarts). Logs stop when stopping ctdb. My setup should be correct as ping_pong read or read/write tests are successful.

Version-Release number of selected component (if applicable):
6.2

How reproducible:
Every time.
Configure a ctdb cluster with 2 nodes over GFS2 (all latest software via yum upgrade at the time of today) and configure CTDB_RECOVERY_LOCK on the shared storage. If I remove the CTDB_RECOVERY_LOCK option, the messages disappear but it's really required. 

Actual results:
Feb 3 19:36:23 gfs-01 kernel: dlm: dev_write no op 40001 102b5
Feb 3 19:36:24 gfs-01 kernel: dlm: dev_write no op 40001 102b5
Feb 3 19:36:25 gfs-01 kernel: dlm: dev_write no op 40001 102b5
Feb 3 19:36:26 gfs-01 kernel: dlm: dev_write no op 40001 102b5
Feb 3 19:36:27 gfs-01 kernel: dlm: dev_write no op 40001 102b5
Feb 3 19:36:28 gfs-01 kernel: dlm: dev_write no op 40001 102b5
Feb 3 19:36:29 gfs-01 kernel: dlm: dev_write no op 40001 102b5
Feb 3 19:36:30 gfs-01 kernel: dlm: dev_write no op 40001 102b5
Feb 3 19:36:31 gfs-01 kernel: dlm: dev_write no op 40001 102b5
....

Expected results:
No message if everything is OK, or a fix if it's not.
Comment 2 Sumit Bose 2012-02-03 23:04:16 UTC 
Hi Cluster Team, I think this is a duplicate of 
https://bugzilla.redhat.com/show_bug.cgi?id=731775 . Can you confirm this? If yes, please close the ticket.
Comment 3 vmiszczak 2012-02-04 10:27:28 UTC 
Hello,
I have searched and found the bug 731775 you mention before posting.
This bug is marked solved in cluster-3.0.12.1-13.el6 

Currently the bug is triggered with thoses packages :
uname -r;rpm -q cman;rpm -q ctdb
2.6.32-220.4.1.el6.x86_64
cman-3.0.12.1-23.el6.x86_64
ctdb-1.0.114.3-3.el6.x86_64
Comment 4 vmiszczak 2012-02-06 17:46:34 UTC 
I have some news about this issue. It only appears if you run the cluster in compat mode (<group groupd_compat="1" />). As I do not need compat, I just turned it off and all is fine.
Comment 5 Lon Hohberger 2012-02-10 18:28:13 UTC 
Ok, thanks for the bug report and the workaround.