<!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 text="#000000" bgcolor="#ffffff">
Fragmentation has been atop our dev priority list for sometime<br>
now. That is, both, reducing it and handling it better when it does<br>
get fragmented.<br>
<br>
Just last week we pushed patches for the same into the newly<br>
created 2.6.35.<br>
<a class="moz-txt-link-freetext" href="http://oss.oracle.com/pipermail/ocfs2-devel/2010-May/006511.html">http://oss.oracle.com/pipermail/ocfs2-devel/2010-May/006511.html</a><br>
<br>
As always, these features will eventually trickle down to the prod<br>
releases.<br>
<br>
Sunil<br>
<br>
On 05/24/2010 12:41 PM, Brian Kroth wrote:
<blockquote cite="mid:20100524194131.GH20060@gmail.com" type="cite">
<pre wrap="">Michael Austin <a class="moz-txt-link-rfc2396E" href="mailto:onedbguru@gmail.com"><onedbguru@gmail.com></a> 2010-05-24 13:32:
</pre>
<blockquote type="cite">
<pre wrap=""> I would like to get some feedback on the overall perception on the support
and stability of OCFS2 (latest). This tool looks like a perfect fit for
a production system I am planning, but, due to it's open source roots,
there are some concerns about s&s. The app will be deemed mission
critical with very little tolerance for any downtime (24x365).
Thanks.
M. Austin
Consultant
</pre>
</blockquote>
<pre wrap="">
It pains me to, but I can't say I'd recommend it for something like a
mail setup that has heavy write of tiny files. There's a fragmentation
issue that burned us bad recently and before that a locking issue
(search the archives). Even then I have to say that the Oracle devs
were responsive to us even without a service contract, for which I'm
very grateful. You might have better luck with a "supported" distro.
I've always used mainline kernels with Debian.
That said, I had been using an earlier version for a web server backend
(couple of TB, mostly read) and a video streaming library (_many_ TB and
_lots_ of read traffic) for a long time without any reports of problems.
I don't work there anymore, but from what I hear everything's still
humming along without interruption (that should be read overall cluster
interruption) for almost 3 years now. That even with crummy server rooms
that try bake their inhabitants from time to time :)
I will also say just off hand that OCFS2 is still the best OSS shared
disk cluster fs I've tried. I've tested GFS2 off and on for a couple of
years and it still has a rather trivial deadlock case:
# cssh node1 node2 node3
# mkdir /cluster/$HOSTNAME
# touch /cluster/$HOSTNAME/test
# rm -rf /cluster/*
Cheers,
Brian
</pre>
<pre wrap="">
<fieldset class="mimeAttachmentHeader"></fieldset>
_______________________________________________
Ocfs2-users mailing list
<a class="moz-txt-link-abbreviated" href="mailto:Ocfs2-users@oss.oracle.com">Ocfs2-users@oss.oracle.com</a>
<a class="moz-txt-link-freetext" href="http://oss.oracle.com/mailman/listinfo/ocfs2-users">http://oss.oracle.com/mailman/listinfo/ocfs2-users</a></pre>
</blockquote>
<br>
</body>
</html>