<div dir="ltr">Are you mounting -o writeback?<br></div><div class="gmail_extra"><br><br><div class="gmail_quote">On Fri, Mar 29, 2013 at 12:28 PM, Andy <span dir="ltr"><<a href="mailto:aryan1@allantgroup.com" target="_blank">aryan1@allantgroup.com</a>></span> wrote:<br>
<blockquote class="gmail_quote" style="margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex">I have been having performance issues from time to time on our<br>
production ocfs2 volumes, so I set up a test system to try to reproduce<br>
what I was seeing on the production systems. This is what I found out:<br>
<br>
I have a 2 node test system sharing a 2TB volume with a journal size of<br>
256MB. I can easily trigger the slowdown by starting to processes to<br>
write a 10GB file each, then I delete a different large file (7GB+)<br>
while the other processes are writing. The slowdown is significant and<br>
very disruptive. Not only did it take over 3 minutes to delete the<br>
file, every else with pause when entering that directory too. A du<br>
command with stop and nfs access to that file system will think the<br>
server is not responding. Under heavier amounts of writes, I have had a<br>
delete takes 13mins for a 8GB file, and NFS mounts return I/O errors.<br>
We often deal with large files, so this situation above is fairly common.<br>
<br>
I would like any ideas that would provide smoother performance of the<br>
OCFS2 volume and somehow eliminate the long pauses during deletes.<br>
<br>
Thanks,<br>
<br>
Andy<br>
<br>
_______________________________________________<br>
Ocfs2-users mailing list<br>
<a href="mailto:Ocfs2-users@oss.oracle.com">Ocfs2-users@oss.oracle.com</a><br>
<a href="https://oss.oracle.com/mailman/listinfo/ocfs2-users" target="_blank">https://oss.oracle.com/mailman/listinfo/ocfs2-users</a><br>
</blockquote></div><br></div>