[Ocfs2-users] Slow write performance

Volker.Gruenewald at deutschepost.de Volker.Gruenewald at deutschepost.de
Wed Jul 27 23:45:52 PDT 2011


Hi,

 

we have several ocfs2 Clusters.

 

All are based on 

RHEL 5.5 (2.6.18-194.el5 #1 SMP Tue Mar 16 21:52:39 EDT 2010 x86_64 x86_64 x86_64 GNU/Linux)

Version of OCFS2 is

 

ocfs2-tools-1.4.4-1.el5

ocfs2-tools-devel-1.4.4-1.el5

ocfs2console-1.4.4-1.el5

ocfs2-2.6.18-194.el5-1.4.7-1.el5

 

Every Cluster have on Node, which is exporting the (OCFS") share via NFS (right now NFS v2).

The entry in /etc/exports looks as 

 

/share server (rw,sync,no_root_squash,no_subtree_check,nohide)

 

There are a lot of File Operations on the share.

 

We are right now facing a lot of problems, mostly with poor write performance or Cluster Hangs.

 

The system with poor performance (since a few days) is a 4 Node System

 

I will attach the output of ./stat_sysdir.sh (run on the nfs node)

 

Next steps from me would be (after what I have learned here right now)

-          umount the file system

-          reduce the node count to 6

-          fsck.ocfs2 on the filesystem

 

please prove further help, regards

 

Volker Grünewald

 

-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://oss.oracle.com/pipermail/ocfs2-users/attachments/20110728/cc76edd9/attachment-0001.html 
-------------- next part --------------
A non-text attachment was scrubbed...
Name: out.log
Type: application/octet-stream
Size: 924771 bytes
Desc: out.log
Url : http://oss.oracle.com/pipermail/ocfs2-users/attachments/20110728/cc76edd9/attachment-0001.obj 


More information about the Ocfs2-users mailing list