<html>
<head>
<meta content="text/html; charset=ISO-8859-1"
http-equiv="Content-Type">
</head>
<body bgcolor="#FFFFFF" text="#000000">
The kernel developers introduced a bug into ext4 that has yet to be
fixed. If you use xfs you won't have those hangs.<br>
<br>
<div class="moz-cite-prefix">On 03/01/2013 01:30 AM, Nikita A
Kardashin wrote:<br>
</div>
<blockquote
cite="mid:CALA_TQhmwhBeBgGEyDuivZXJm-zDW-7W4vXTJf7g3nGEHwRQuQ@mail.gmail.com"
type="cite">
<div dir="ltr">Hello again!
<div><br>
</div>
<div style="">I am complete rebuild my storage. </div>
<div style="">As base: ext4 over mdadm-raid1 </div>
<div style="">Gluster volume in distributed-replicated mode with
settings:</div>
<div style="">
<br>
</div>
<div style="">
<div>Options Reconfigured:</div>
<div>performance.cache-size: 1024MB</div>
<div>nfs.disable: on</div>
<div>performance.write-behind-window-size: 4MB</div>
<div>performance.io-thread-count: 64</div>
<div>features.quota: off</div>
<div>features.quota-timeout: 1800</div>
<div>performance.io-cache: on</div>
<div>performance.write-behind: on</div>
<div>performance.flush-behind: on</div>
<div>performance.read-ahead: on</div>
<div><br>
</div>
<div style="">As result, I got write performance about 80MB/s
on dd if=/dev/zero of=testfile.bin bs=100M count=10, but I
also got some strange things.</div>
<div style="">If I try to execute above command inside virtual
machine (KVM), first time all going right - about 900MB/s
(cache effect, I think), but if I run this test again on
existing file - task (dd) hungs up and can be stopped only
by Ctrl+C.</div>
<div style=""><br>
</div>
<div style="">Overall virtual system latency is poor too. For
example, apt-get upgrade upgrading system very, very slow,
freezing on "Unpacking replacement" and other io-related
steps. </div>
<div style="">
<br>
</div>
<div style="">Does glusterfs have any tuning options, that can
help me?</div>
</div>
</div>
<div class="gmail_extra"><br>
<br>
<div class="gmail_quote">2013/2/27 Brian Candler <span
dir="ltr"><<a moz-do-not-send="true"
href="mailto:B.Candler@pobox.com" target="_blank">B.Candler@pobox.com</a>></span><br>
<blockquote class="gmail_quote" style="margin:0 0 0
.8ex;border-left:1px #ccc solid;padding-left:1ex">
<div class="im">On Wed, Feb 27, 2013 at 02:46:28PM +0000,
Robert van Leeuwen wrote:<br>
> You could try to trunk the network on your client but
10Gbit ethernet ( i suggest to use fiber, because of
latency issues with copper 10Gbit)<br>
<br>
</div>
Aside: 10G with SFP+ direct-attach cables also works well,
even though it's<br>
copper. The latency problem is with 10Gbase-T (RJ45 /
CAT6(a))<br>
<div class="HOEnZb">
<div class="h5">_______________________________________________<br>
Gluster-users mailing list<br>
<a moz-do-not-send="true"
href="mailto:Gluster-users@gluster.org">Gluster-users@gluster.org</a><br>
<a moz-do-not-send="true"
href="http://supercolony.gluster.org/mailman/listinfo/gluster-users"
target="_blank">http://supercolony.gluster.org/mailman/listinfo/gluster-users</a><br>
</div>
</div>
</blockquote>
</div>
<br>
<br clear="all">
<div><br>
</div>
-- <br>
With best regards,<br>
differentlocal (<a moz-do-not-send="true"
href="http://www.differentlocal.ru">www.differentlocal.ru</a>
| <a moz-do-not-send="true"
href="mailto:differentlocal@gmail.com">differentlocal@gmail.com</a>),<br>
System administrator.
</div>
<br>
<fieldset class="mimeAttachmentHeader"></fieldset>
<br>
<pre wrap="">_______________________________________________
Gluster-users mailing list
<a class="moz-txt-link-abbreviated" href="mailto:Gluster-users@gluster.org">Gluster-users@gluster.org</a>
<a class="moz-txt-link-freetext" href="http://supercolony.gluster.org/mailman/listinfo/gluster-users">http://supercolony.gluster.org/mailman/listinfo/gluster-users</a></pre>
</blockquote>
<br>
</body>
</html>