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

Re: [Condor-users] Condor acts weird



On 05/07/2012 14:13, "Marios Kleanthous" <mklean@xxxxxxxxxxxx> wrote:

>Hello,
>
>We have setup a condor cluster with 13 machines, 8 nodes each machine and
>on an attemp to run we have the following problem:
>
>When a 1 user submits from one 1 machine we get max 48 nodes Claimed out
>of 104 nodes in total. Most of the nodes get Matched and then timeout and
>return to unclaimed. This continues on and on no matter how big is the
>jobs or match we increase the Match timeout
> variable.
>
>When multiple users submit from multiple machines then all nodes get
>Claimed without any problems.
>
>We have been banging our heads for few days and couldn¹t figure it out.
>Is it possible that it¹s a network problem, saturation, and we need to
>talk with the administrator to look it up?
>
>Thank you,
>Marios
>  

Are you using static or dynamic slots?
What version of Condor?
-Ian

-- 
Ian Cottam
IT Services -- supporting research
Faculty of Engineering and Physical Sciences
The University of Manchester
"The only strategy that is guaranteed to fail is not taking risks." Mark
Zuckerberg