[Ocfs2-users] Unmounting ocfs2 at shutdown

Kurt Hackel Kurt.Hackel at oracle.com
Thu Jun 30 10:33:11 CDT 2005


Hi Ivan,

Sounds like something is spinning while trying to migrate its locks to
the other node on umount.  To debug this, if the boxes are responsive
enough, try to do an ALT-SYSRQ-T on both nodes (maybe 2 or 3 times) and
send me the output from /var/log/messages.  In some cases, ALT-SYSRQ-P
may work to get one process stack.

Another thing worth trying is to run this command on both nodes
immediately before you unmount:
  debugfs.ocfs2 -l dlm dlm_domain dlm_thread dlm_master dlm_recovery dlm_glue allow

This sets up our debugging to dump all lockmanager-related logging in
/proc/fs/ocfs2_nodemanager/log_mask.  This may result in a lot of
output in /var/log/messages.  Just reboot the machines after about a 
minute or two of logging.

Once I get your info, I'll log a bugzilla bug for you and attach any
output you may have.

Thanks!
Kurt Hackel
  

On Thu, Jun 30, 2005 at 08:42:58AM -0400, Ivan Vasquez wrote:
> We are installing a 2 node cluster and are seeing this behavior: When
> shared storage is manually unmounted by umount /dev/sd*#, it works well
> and the other node is not affected. However, when the system tries to do
> the same at shutdown, the machine going down gets stuck in it, while the
> remaining machine gets a very high cpu utilization. We end up having to
> turn off the servers as the system stops being responsive.
> 
> Our question:
> What is the suggested way to unmount an ocfs2 filesystem at shutdown?
> 
> Thanks,
> Ivan
> 
> -----Original Message-----
> From: ocfs2-users-bounces at oss.oracle.com
> [mailto:ocfs2-users-bounces at oss.oracle.com] On Behalf Of Joel Becker
> Sent: Tuesday, June 28, 2005 3:19 PM
> To: Zafar_Mahmood at Dell.com
> Cc: ocfs2-users at oss.oracle.com
> Subject: Re: [Ocfs2-users] OCFS2 volumes do no mount automatically on
> RHEL4also
> 
> On Tue, Jun 28, 2005 at 01:55:02PM -0500, Zafar_Mahmood at Dell.com wrote:
> > Here it how it is configured:
> > 
> > [root at z6850n1 rc3.d]# ls -l *netfs
> > lrwxrwxrwx  1 root root 15 May 22 08:14 S25netfs -> ../init.d/netfs
> > [root at z6850n1 rc3.d]# ls -l *o2cb
> > lrwxrwxrwx  1 root root 14 Jun 19 12:44 S50o2cb -> ../init.d/o2cb
> 
> 	Did you install months ago and then always upgrade?  Try
> 'chkconfig o2cb off; chkconfig o2cb on'
> 
> Joel
> -- 
> 
> Life's Little Instruction Book #267
> 
> 	"Lie on your back and look at the stars."
> 
> Joel Becker
> Senior Member of Technical Staff
> Oracle
> E-mail: joel.becker at oracle.com
> Phone: (650) 506-8127
> _______________________________________________
> Ocfs2-users mailing list
> Ocfs2-users at oss.oracle.com
> http://oss.oracle.com/mailman/listinfo/ocfs2-users
> _______________________________________________
> Ocfs2-users mailing list
> Ocfs2-users at oss.oracle.com
> http://oss.oracle.com/mailman/listinfo/ocfs2-users



More information about the Ocfs2-users mailing list