[Gluster-users] glusterfs and glusterfsd process utilization

Krutika Dhananjay kdhananj at redhat.com
Fri Dec 26 05:40:59 UTC 2014


----- Original Message -----

> From: "Kyle Harris" <kyle.harris98 at gmail.com>
> To: "Krutika Dhananjay" <kdhananj at redhat.com>
> Cc: gluster-users at gluster.org
> Sent: Wednesday, December 24, 2014 9:11:14 PM
> Subject: Re: [Gluster-users] glusterfs and glusterfsd process utilization

> Thanks Krutika for taking time to troubleshoot this for me. Would this be
> considered a bug that needs to be fixed? It seems as though the heal process
> should not be able to use all the available resources of a system. Might I
> suggest some type of option that limits the number of heals that could take
> place at a single time? Perhaps some type of rate limit for healing?
That's right. We are in fact planning to introduce such a throttling ability to keep self-heal from affecting functioning of normal clients in this manner. 
-Krutika 

> Interestingly enough I attempted to rate limit the gluster processes by using
> cgroups and although it worked, it kept the processes from working properly.

> Thanks again.

> Kyle
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://www.gluster.org/pipermail/gluster-users/attachments/20141226/3cc8be00/attachment.html>


More information about the Gluster-users mailing list