<!DOCTYPE HTML PUBLIC "-//W3C//DTD HTML 4.01 Transitional//EN">
<html>
<head>
<meta content="text/html; charset=ISO-8859-1"
http-equiv="Content-Type">
</head>
<body bgcolor="#ffffff" text="#000000">
It shouldn't. But I have been known to be wrong before.<br>
And testing with different commit values is easy enough.<br>
<br>
On 02/02/2011 06:03 PM, Henrique Fernandes wrote:
<blockquote
cite="mid:AANLkTimbM1M=jcC5qkg6yJh3ouCq1MHfCK4ZLXa6V5ut@mail.gmail.com"
type="cite">Thanks for the tips.<br>
<br>
So you are saying that if both nodes have same load, inscrease
comit time will not make any diferences ?<br>
<br>
<br>
<br clear="all">
[]'sf.rique <br>
<br>
<br>
<div class="gmail_quote">On Wed, Feb 2, 2011 at 11:54 PM, Sunil
Mushran <span dir="ltr"><<a moz-do-not-send="true"
href="mailto:sunil.mushran@oracle.com">sunil.mushran@oracle.com</a>></span>
wrote:<br>
<blockquote class="gmail_quote" style="margin: 0pt 0pt 0pt
0.8ex; border-left: 1px solid rgb(204, 204, 204);
padding-left: 1ex;">
<div bgcolor="#ffffff" text="#000000"> # grep "Resources"
/sys/kernel/debug/o2dlm/UUID/dlm_state<br>
Lock Resources: 20713 (1139459)<br>
<br>
The first number is number of live resources. Second the
number<br>
of resources created for the life of the domain.<br>
<br>
If you see the the number of resources levelling off but the
second<br>
number incrementing, that _could_ mean that you could
benefit with<br>
a larger cache. But you have to supplement that number with
the<br>
workload. Say you are deleting and creating files. In that
case, you<br>
will see the same number movement but not because of the
cache limit.<br>
<br>
We are in the process of adding more counters to provide
more visibility<br>
to the users.<br>
<br>
Well, using any software without support has its risks.<br>
<br>
Using noatime will help. Increasing commit will help if
workload if largely<br>
from one node only.
<div>
<div class="h5"><br>
<br>
On 02/02/2011 05:10 PM, Henrique Fernandes wrote:
<blockquote type="cite">centos 5.5 <br>
ocfs2 1.4<br>
The mail solution, dovecot+postifx+mailscaner uses
about 1 gb the machine has 1.5 in my tests i am rising
it to 3gb so it have 2gb of cached<br>
<br>
How many memory do you thing i should separeted for
cache ?<br>
<br>
When we made some tests ocfs2 1.4 had better
performance than 1.6 but or tests were very simple, an
script that writes lots of files and anothe rone that
reads it.<br>
<br>
And it is a problem to have ocfs2 1.6 in ths centos!!<br>
<br>
Should i use ocfs2 in production ?<br>
<br>
how about the commit and noatime configs ??<br>
<br>
<br>
thanks!!<br>
<br clear="all">
[]'sf.rique <br>
<br>
<br>
<div class="gmail_quote">On Wed, Feb 2, 2011 at 9:27
PM, Sunil Mushran <span dir="ltr"><<a
moz-do-not-send="true"
href="mailto:sunil.mushran@oracle.com"
target="_blank">sunil.mushran@oracle.com</a>></span>
wrote:<br>
<blockquote class="gmail_quote" style="margin: 0pt
0pt 0pt 0.8ex; border-left: 1px solid rgb(204,
204, 204); padding-left: 1ex;">
<div bgcolor="#ffffff" text="#000000"> version?
distro?<br>
<br>
This workload will benefit a lot with the
indexed directories available in<br>
ocfs2 1.6 (and mainline and sles11).<br>
<br>
The other thing to check is the amount memory in
the virtual machines.<br>
File systems need memory to cache the inodes. If
memory is lacking,<br>
the inodes are freed and have to be re-read from
disk time and again.<br>
While this is a problem even in a local fs, it
is a bigger problem in a cfs<br>
as a cfs needs to do lock mastery for the same
inode time and again.
<div>
<div><br>
<br>
On 02/02/2011 01:09 PM, Henrique Fernandes
wrote: </div>
</div>
<blockquote type="cite">
<div>
<div>Hello,<br>
<br>
First of all, i am new at the list and i
have several questions about ocfs2
performance.<br>
<br>
Where i am working i am having huge
performance problens with ocfs2.<br>
<br>
Let me tell my envoriment.<br>
<br>
3 Xen VirtualMachines withs ocfs2 mounting
an LUN exported over iSCSI. ( acctualy 3
LUNS, 3 ocfs2 clusters )<br>
<br>
I am not the one who configured the
envoriment, but it is making the
performance of my MAIL system to bad.<br>
<br>
Have about 9k accounts but only 4k are
active. It is a maildir system. ( postfix
+ dovecot )<br>
<br>
Now that this performance problens are
afecting my system i am gonna try help to
tunning the ocfs2.<br>
<br>
Pretty much all default settings.<br>
<br>
OCFS2 is configured to write with ordered
mode. We know that changing to writeback
will make performance much better, but we
are not considering lose anydata, so it i
snot an option.<br>
<br>
Now we are going to implemente noatime
options in mount. This make better
performace ?<br>
<br>
Other one, how about the commit mount
options ? The default is set to 5s if i
increse it how is the potential data loss
in case we lost lose power?<br>
<br>
Does anyone have any other paramenter that
should help us ?<br>
<br>
Another info, the inscremental backup is
taking 10 to 12 hours. <br>
<br>
All nodes have VERY high I/O wait.<br>
<br>
Thanks to all!!<br>
<br>
If you could tell me any doc that i sould
read would be nice to!<br>
<br>
<br>
<br>
<br>
<br>
<br clear="all">
[]'sf.rique <br>
</div>
</div>
<pre><fieldset></fieldset>
_______________________________________________
Ocfs2-users mailing list
<a moz-do-not-send="true" href="mailto:Ocfs2-users@oss.oracle.com" target="_blank">Ocfs2-users@oss.oracle.com</a>
<a moz-do-not-send="true" href="http://oss.oracle.com/mailman/listinfo/ocfs2-users" target="_blank">http://oss.oracle.com/mailman/listinfo/ocfs2-users</a></pre>
</blockquote>
<br>
</div>
</blockquote>
</div>
<br>
</blockquote>
<br>
</div>
</div>
</div>
</blockquote>
</div>
<br>
</blockquote>
<br>
</body>
</html>