<div dir="ltr">Hi,<br><br>Even I ran into 100% cpu utilization with patch-796. But it was due to badly configured aggregate-size. An aggregate-size of 1KB kept the cpu usage well around 54%.<br><br>regards,<br><div class="gmail_quote">
On Thu, Aug 28, 2008 at 5:12 PM, Dai, Manhong <span dir="ltr"><<a href="mailto:daimh@umich.edu">daimh@umich.edu</a>></span> wrote:<br><blockquote class="gmail_quote" style="border-left: 1px solid rgb(204, 204, 204); margin: 0pt 0pt 0pt 0.8ex; padding-left: 1ex;">
<div>
<p><font size="2">Hi,<br>
<br>
<br>
Repository revision: glusterfs--mainline--2.5--patch-795 is the version<br>
I am using.<br>
<br>
<br>
Our cluster is in use, so I cannot confirm this problem on other<br>
releases. Since removing write-behind can get rid of this problem, I can<br>
live with that.<br>
<br>
<br>
<br>
Best,<br><font color="#888888">
Manhong</font><div><div></div><div class="Wj3C7c"><br>
<br>
On Thu, 2008-08-28 at 07:48 +0400, Raghavendra G wrote:<br>
> Hi,<br>
><br>
> what patch are you using? with glusterfs--mainline--3.0--patch-329 and<br>
> a basics setup of write-behind over protocol/client, glusterfs cpu<br>
> usage never went above 65% in my tests. Can you please confirm whether<br>
> the problem persists in patch-329?<br>
><br>
> regards,<br>
><br>
> On Thu, Aug 28, 2008 at 6:27 AM, Dai, Manhong <<a href="mailto:daimh@umich.edu" target="_blank">daimh@umich.edu</a>> wrote:<br>
> Hi,<br>
> <br>
> client.vol is<br>
> <br>
> volume unify-brick<br>
> type cluster/unify<br>
> option scheduler rr # round robin<br>
> option namespace muskie-ns<br>
> # subvolumes muskie-brick pike1-brick pike2-brick pike3-brick<br>
> subvolumes muskie-brick pike1-brick pike3-brick<br>
> end-volume<br>
> <br>
> volume wb<br>
> type performance/write-behind<br>
> option aggregate-size 1MB<br>
> option flush-behind on<br>
> subvolume<br>
> client.vol is<br>
> <br>
> volume unify-brick<br>
> type cluster/unify<br>
> option scheduler rr # round robin<br>
> option namespace muskie-ns<br>
> # subvolumes muskie-brick pike1-brick pike2-brick pike3-brick<br>
> subvolumes muskie-brick pike1-brick pike3-brick<br>
> end-volume<br>
> <br>
> volume wb<br>
> type performance/write-behind<br>
> option aggregate-size 1MB<br>
> option flush-behind on<br>
> subvolumes unify-brick<br>
> end-volume<br>
> <br>
> <br>
> yes abcdefghijklmn | while read l; do echo $l; done > a<br>
> <br>
> would cause glusterfs process 100% busy and files system hang<br>
> when the output size is around the aggregate-size.<br>
> <br>
> <br>
> removing write-behind translator would get rid of this<br>
> problem.<br>
> <br>
> <br>
> s unify-brick<br>
> end-volume<br>
> <br>
> <br>
> command "yes abcdefghijklmn | while read l; do echo $l; done ><br>
> a" would cause glusterfs process 100% busy and files system<br>
> hang when the output size is around the aggregate-size.<br>
> removing write-behind translator would get rid of this<br>
> problem.<br>
> <br>
> <br>
> <br>
> <br>
> Best,<br>
> Manhong<br>
> <br>
> <br>
> <br>
> _______________________________________________<br>
> Gluster-users mailing list<br>
> <a href="mailto:Gluster-users@gluster.org" target="_blank">Gluster-users@gluster.org</a><br>
> <a href="http://zresearch.com/cgi-bin/mailman/listinfo/gluster-users" target="_blank">http://zresearch.com/cgi-bin/mailman/listinfo/gluster-users</a><br>
> <br>
><br>
><br>
><br>
> --<br>
> Raghavendra G<br>
><br>
> A centipede was happy quite, until a toad in fun,<br>
> Said, "Prey, which leg comes after which?",<br>
> This raised his doubts to such a pitch,<br>
> He fell flat into the ditch,<br>
> Not knowing how to run.<br>
> -Anonymous<br>
><br>
</div></div></font>
</p>
</div>
</blockquote></div><br><br clear="all"><br>-- <br>Raghavendra G<br><br>A centipede was happy quite, until a toad in fun,<br>Said, "Prey, which leg comes after which?",<br>This raised his doubts to such a pitch,<br>
He fell flat into the ditch,<br>Not knowing how to run.<br>-Anonymous<br>
</div>