<br><font size=2 face="sans-serif">The data used on the file system is
pretty transient. Files are created and then moved off to other locations
not on the lustre system. </font>
<br>
<br><font size=2 face="sans-serif">I did look into top and do not find
1 specific process that is hogging the cpu. Pretty much the same
across each OSS. Quiet a few of these is about it. ll_ost_io_01</font>
<br>
<br><font size=2 face="sans-serif">The stripping seems to be going across
all of the ost's correctly. </font>
<br>
<br><font size=2 face="Courier New">UUID
bytes Used
Available Use% Mounted on</font>
<br><font size=2 face="Courier New">lustre-MDT0000_UUID
726.2G 1.7G 683.0G
0% /san[MDT:0]</font>
<br><font size=2 face="Courier New">lustre-OST0000_UUID
2.4T 742.2G 1.5T
30% /san[OST:0]</font>
<br><font size=2 face="Courier New">lustre-OST0001_UUID
2.4T 696.6G 1.6T
28% /san[OST:1]</font>
<br><font size=2 face="Courier New">lustre-OST0002_UUID
2.4T 729.9G 1.5T
30% /san[OST:2]</font>
<br><font size=2 face="Courier New">lustre-OST0003_UUID
2.4T 736.1G 1.5T
30% /san[OST:3]</font>
<br><font size=2 face="Courier New">lustre-OST0004_UUID
2.4T 757.1G 1.5T
31% /san[OST:4]</font>
<br><font size=2 face="Courier New">lustre-OST0005_UUID
2.4T 784.7G 1.5T
32% /san[OST:5]</font>
<br><font size=2 face="Courier New">lustre-OST0006_UUID
2.4T 898.8G 1.4T
37% /san[OST:6]</font>
<br><font size=2 face="Courier New">lustre-OST0007_UUID
2.4T 762.2G 1.5T
31% /san[OST:7]</font>
<br>
<br><font size=2 face="Courier New">filesystem summary:
18.9T 6.0T 12.0T
31% /san</font>
<br>
<br><font size=2 face="Courier New">Thanks Again.</font>
<br>
<br>
<br><font size=2 face="sans-serif"><br>
Rocky <br>
<br>
</font>
<br>
<br>
<br>
<table width=100%>
<tr valign=top>
<td><font size=1 color=#5f5f5f face="sans-serif">From:</font>
<td><font size=1 face="sans-serif">Wang Yibin <wang.yibin@oracle.com></font>
<tr valign=top>
<td><font size=1 color=#5f5f5f face="sans-serif">To:</font>
<td><font size=1 face="sans-serif">Ronald K Long <rklong@usgs.gov></font>
<tr>
<td valign=top><font size=1 color=#5f5f5f face="sans-serif">Cc:</font>
<td><font size=1 face="sans-serif">lustre-discuss@lists.lustre.org</font>
<tr valign=top>
<td><font size=1 color=#5f5f5f face="sans-serif">Date:</font>
<td><font size=1 face="sans-serif">11/16/2010 09:54 AM</font>
<tr valign=top>
<td><font size=1 color=#5f5f5f face="sans-serif">Subject:</font>
<td><font size=1 face="sans-serif">Re: [Lustre-discuss] High CPU load,
only on 1 OSS</font></table>
<br>
<hr noshade>
<br>
<br>
<br><font size=3 face="sans-serif">Hello,</font>
<br>
<br><font size=3 face="sans-serif">Normally when stripe_offset is set to
-1, MDS will do load/space balancing automatically.</font>
<br><font size=3 face="sans-serif">What is your use pattern of the filesystem?</font>
<br><font size=3 face="sans-serif">It sounds like that your applications
are doing extensive I/O on that particular OSS.</font>
<br>
<br><font size=3 face="sans-serif">To find out why the load on the OSS
is so high, please </font>
<br><font size=3 face="sans-serif">- find what processes are hogging the
CPUs using top(1).</font>
<br><font size=3 face="sans-serif">- get the stripe info of your in-use
files to see whether most of them reside on the same OSS.</font>
<br>
<br><font size=3 face="sans-serif">If the files in use are not distributed
among the OSS servers, or your file usage pattern is one-OSS bound, you
may want to consider tuning the stripe_count/stripe_size.</font>
<br>
<br><font size=3 face="sans-serif">ÔÚ 2010-11-16£¬ÏÂÎç10:38£¬ Ronald
K Long дµÀ£º</font>
<br>
<br><font size=2 face="sans-serif"><br>
We recently setup a lustre config. 1 MDS 4 OSS's. Everything
is running fine except on the first OSS we are experiencing very high cpu
load. The first OSS is running a CPU load in the high 50's. The
other 3 OSS's are steady at around 8. Everything is the same between
all of the OSS's.</font><font size=3 face="sans-serif"> <br>
</font><font size=2 face="sans-serif"><br>
The stripe is setup </font><font size=3 face="sans-serif"><br>
</font><font size=2 face="sans-serif"><br>
stripe_count: 1 stripe_offset: -1</font><font size=3 face="sans-serif">
<br>
</font><font size=2 face="sans-serif"><br>
Red Hat 5 64bit</font><font size=3 face="sans-serif"> <br>
</font><font size=2 face="sans-serif"><br>
kernel-2.6.18-194.3.1.el5_lustre.1.8.4</font><font size=3 face="sans-serif">
</font><font size=2 face="sans-serif"><br>
kernel-devel-2.6.18-194.3.1.el5_lustre.1.8.4</font><font size=3 face="sans-serif">
</font><font size=2 face="sans-serif"><br>
lustre-ldiskfs-3.1.3-2.6.18_194.3.1.el5_lustre.1.8.4</font><font size=3 face="sans-serif">
</font><font size=2 face="sans-serif"><br>
lustre-1.8.4-2.6.18_194.3.1.el5_lustre.1.8.4</font><font size=3 face="sans-serif">
</font><font size=2 face="sans-serif"><br>
lustre-modules-1.8.4-2.6.18_194.3.1.el5_lustre.1.8.4</font><font size=3 face="sans-serif">
<br>
<br>
</font><font size=2 face="sans-serif"><br>
Any thing I can check on the problem OSS to rectify this issue.</font><font size=3 face="sans-serif">
<br>
</font><font size=2 face="sans-serif"><br>
Thank you in advance</font><font size=3 face="sans-serif"><br>
</font><font size=2 face="sans-serif"><br>
Rocky </font><font size=3 face="sans-serif"><br>
_______________________________________________<br>
Lustre-discuss mailing list</font><font size=3 color=blue face="sans-serif"><u><br>
</u></font><a href="mailto:Lustre-discuss@lists.lustre.org"><font size=3 color=blue face="sans-serif"><u>Lustre-discuss@lists.lustre.org</u></font></a><font size=3 face="sans-serif"><br>
</font><a href="http://lists.lustre.org/mailman/listinfo/lustre-discuss"><font size=3 face="sans-serif">http://lists.lustre.org/mailman/listinfo/lustre-discuss</font></a>
<br>
<br>
<br>