[GE users] Re: Startup times and other issues with 6.0u3

Ron Chen ron_chen_123 at yahoo.com
Sat Mar 19 02:39:38 GMT 2005


They can be related, since the difference between
tight and non-tight PE jobs is that the tight one is
started by the execd/shepherd.

When a job is started by execd/shepherd, SGE adds a
new group ID to the job, and thus all the processes
are not allowed to escape. So if CentOS also has
problems with the additional gid, then you will
experience performance problem.

I will wait for Reuti to wake up (he's in Europe), and
see if there's anything wrong with your PE - he knows
more about PEs than I.

 -Ron


--- Brian R Smith <brian at cypher.acomp.usf.edu> wrote:
> That problem was on RHEL 3.0 with the 2.4.x kernel
> series and is 
> completely different than mine.  He is running
> simultaneous compile jobs 
> while i am running communications-laden (well, not
> too badly) numerical 
> computations.  CentOS 4 is build of RHEL 4 and is
> based on a recent 
> 2.6.x kernel.  I can't see the correlation.
> 
> Either way, if I can get a fix for my current setup,
> great.  Otherwise, 
> I will revert back to SGE 6.0u2 and wait for u4 to
> come out with the fixes.



		
__________________________________ 
Do you Yahoo!? 
Yahoo! Small Business - Try our new resources site!
http://smallbusiness.yahoo.com/resources/ 

---------------------------------------------------------------------
To unsubscribe, e-mail: users-unsubscribe at gridengine.sunsource.net
For additional commands, e-mail: users-help at gridengine.sunsource.net




More information about the gridengine-users mailing list