[torqueusers] ulimit -l

Brian D. Haymore brian at chpc.utah.edu
Tue Mar 27 14:54:32 MDT 2007


You will need to do a couple things still.

First is to go to your /etc/pam.d directory and edit one or more of the 
following files (this is based on RedHat EL4 btw so a different dist may 
be different):

rsh (if your using rsh for your remote job dispatch mechanism)
sshd (if your using ssh for your remote job dispatch mechanism)
login

and add in this line into each of those files:

session    required     pam_limits.so

Some of those files on my RedHat system said certain rules already in 
the file had to be first or last so take note of that if they are there 
for you as well.


Then the next thing is you have to restart the torque daemons on each 
system.  The reason for this is if they were started before you made 
these changes they hold the old limits and all processes they spawn will 
inherit those limits.  So restarting will get Torque going with the 
"unlimited" mode so child processes inherit that.


Brock Palen wrote:
> Hello,  We are making our first trip into the world of OpenIB and in 
> this environment they require us to up the lockable memory.  While we 
> have upped the amount in  /etc/security/limits.conf and shows up if i 
> ssh to a node.  But in a pbs job it falls back to the default 32kb.  Is 
> there a method to getting around this?
> 
> Brock Palen
> Center for Advanced Computing
> brockp at umich.edu
> (734)936-1985
> 
> 
> _______________________________________________
> torqueusers mailing list
> torqueusers at supercluster.org
> http://www.supercluster.org/mailman/listinfo/torqueusers

-- 
Brian D. Haymore
University of Utah
Center for High Performance Computing
155 South 1452 East RM 405
Salt Lake City, Ut 84112-0190
Phone: (801) 558-1150, Fax: (801) 585-5366
http://www.map.utah.edu/umaplink/0019.html


More information about the torqueusers mailing list