[Gluster-users] maxing out cpu

[Dave Hall]

Hi Shehjar,

On Wed, 2010-01-06 at 14:38 +0530, Shehjar Tikoo wrote:
> Dave Hall wrote:
> > Sorry, I should have been clearer, it is the glusterfs client. 
> > glusterfsd never seems to go above 10% even under load.  it is only
> > 1 glusterfs process and it max's out a whole core.
> 
> Is it always the same client that is showing this problem
> or do the clients on all machines have CPU usage shoot-ups?
> 
> If it is always the same machine, will it be possible to run the
> "glusterfs" client process under valgrind on this system? It'll slow
> it down but should help us figure out where the CPU usage is going.

It is 2 of the 3 nodes which I recall doing it.  I have tried comparing
the environments between the 3 to see what is different and can't find
anything.  

> You can do this by using the following command to start up the
> glusterfs client process:
> 
>   $ valgrind --tool=callgrind <path-to-glusterfs> -f
> <path-to-client-volfile> -L NONE --no-daemon <client-mount-point>

I'll try to get it to trigger while running under valgrind and send you
the output.  It is frustrating that there doesn't seem to be an easy way
of triggering it, but I suppose if there was it would have already been
fixed :)

Cheers

Dave