[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

[Condor-users] random negotiation?


this post is somewhat related to my previous one -
One of our Condor users has been evicted from a few machines due to 
overheating (summer is back!), and since the memory footprint had grown
beyond the physical memory size, these jobs have been stuck.
Unfortunately, none of his other - still virgin - jobs have been scheduled
as long as the "showstopper" jobs were in queue.
I'd expect Condor to skip jobs it cannot schedule properly and continue
with others (this way leaving the problematic stuff for the end). Is there
a config file setting which controls this kind of behaviour, may I have
overlooked something (again)?
Any suggestion is welcome.


Steffen Grunewald * MPI fuer Gravitationsphysik (Albert-Einstein-Institut)
SciencePark Golm, Am Mühlenberg 1, D-14476 Potsdam * http://www.aei.mpg.de
* e-mail: steffen.grunewald(*)aei.mpg.de * +49-331-567-{fon:7233,fax:7298}
No Word/PPT mails - http://www.gnu.org/philosophy/no-word-attachments.html