<html>
<head>
<meta content="text/html; charset=ISO-8859-1"
http-equiv="Content-Type">
</head>
<body bgcolor="#FFFFFF" text="#000000">
<div class="moz-cite-prefix">On 04/15/2013 09:02 AM, Daryl Herzmann
wrote:<br>
</div>
<blockquote
cite="mid:CALOVXPeCVP=DPVkn9AsoqjWZxgZFmWErqBBN4Ns4=gngZ-N4vg@mail.gmail.com"
type="cite">
<meta http-equiv="Content-Type" content="text/html;
charset=ISO-8859-1">
<div dir="ltr">
<div style="">Thanks for your help.</div>
<div><br>
</div>
On Mon, Apr 15, 2013 at 8:39 AM, Pat Riehecky <span dir="ltr"><<a
moz-do-not-send="true" href="mailto:riehecky@fnal.gov"
target="_blank">riehecky@fnal.gov</a>></span> wrote:<br>
<div class="gmail_extra">
<div class="gmail_quote">
<blockquote class="gmail_quote" style="margin:0px 0px 0px
0.8ex;border-left-width:1px;border-left-color:rgb(204,204,204);border-left-style:solid;padding-left:1ex">
<div bgcolor="#FFFFFF" text="#000000">
<div>I've run into some terrible performance when I've
had a lot of add/remove actions on the filesystem in
parallel. They were mostly due to fragmentation.
Alas, XFS can get some horrid fragmentation.<br>
<br>
xfs_db -c frag -r /dev/<node><br>
<br>
should give you the stats on its fragmentation.<br>
</div>
</div>
</blockquote>
<div><br>
</div>
<div>
<div># xfs_db -c frag -r /dev/md127</div>
<div>actual 140539575, ideal 139998847, fragmentation
factor 0.38%</div>
</div>
<div><br>
</div>
<div style="">Here's an iostat snapshot while I was running
xfs_db, the tps numbers for md127 seem strange. sd[b-f]
are a part of the raid5....</div>
<div style=""><br>
</div>
<div style="">
<div>avg-cpu: %user %nice %system %iowait %steal
%idle</div>
<div> 5.12 0.00 7.12 8.81 0.00
78.94</div>
<div><br>
</div>
<div>Device: tps Blk_read/s Blk_wrtn/s
Blk_read Blk_wrtn</div>
<div>sda 35.00 1032.00 656.00
1032 656</div>
<div>sdc 458.00 32451.00 680.00
32451 680</div>
<div>sde 451.00 30936.00 544.00
30936 544</div>
<div>sdb 573.00 32448.00 784.00
32448 784</div>
<div>sdd 527.00 30728.00 696.00
30728 696</div>
<div>sdf 593.00 31736.00 624.00
31736 624</div>
<div>md127 157986.00 157983.00 1592.00
157983 1592</div>
<div><br>
</div>
</div>
<div> </div>
<blockquote class="gmail_quote" style="margin:0px 0px 0px
0.8ex;border-left-width:1px;border-left-color:rgb(204,204,204);border-left-style:solid;padding-left:1ex">
<div bgcolor="#FFFFFF" text="#000000">
<div>I can't speak for others, but I've got 'xfs_fsr'
linked into /etc/cron.weekly/ on my personal systems
with large XFS filesystems.<br>
</div>
</div>
</blockquote>
<div><br>
</div>
<div style="">Seems like I shouldn't have to do that given
the numbers above?</div>
<div style=""><br>
</div>
<div style="">daryl</div>
<br>
</div>
</div>
</div>
</blockquote>
<br>
I would agree, your drive looks nice and optimized..... this thread
already has you checking for lazy counts.... those are my two
"tricks" for getting XFS to run beautifully. But I've never had a
volume over 8TB before.....<br>
<br>
Pat<br>
<br>
<pre class="moz-signature" cols="78">--
Pat Riehecky
</pre>
</body>
</html>