Fwd: [Gluster-devel] rc8

Gordan Bobic gordan at bobich.net
Tue Apr 21 09:58:52 UTC 2009


On Tue, 21 Apr 2009 00:46:20 -0700, Liam Slusser <lslusser at gmail.com>
wrote:
> First off, i apologize to the both of you for emailing you directly.  I
> don't have access to post to the gluster-devel mailing list but i am
still
> having memory leak issues with the rc8 release.  I've tried emailing the
> gluster-users list but i haven't gotten a reply from anybody.
>
> Can either of you forward my email below to the devel list - i think the
> developers should be aware there is some memory issues.

You really should subscribe to the devel list, if you're not already. The
forwarding isn't exactly a sustainable solution.

> There is still a memory leak with rc8 on my setup.  The first server in a
> cluster or two servers starts out using 18M and just slowly increases.
>  After 30mins it has doubled in size to over 30M and just keeps growing -
> the more memory it uses the worst the performance.  Funny that the 2nd
> server in my cluster using the same configuration file has no such memory
> problem.

Interesting - I only have one server working in the cluster (the 2nd
server isn't built yet), but I was getting a leak before with this setup
(just one server being up), so something has clearly been fixed.

Are you using any performance translators? IIRC, io-threads and/or
io-cache were known to have leakyness issues.

Gordan





More information about the Gluster-devel mailing list