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

Re: [condor-users] more jobs running than computers?



On Tue, May 11, 2004 at 09:28:45AM +0930, Carl Howard wrote:
> This might seem a naive question but how is it possible to have more jobs
> running than there are computers in the pool?
> 

Were they recently running? There's no global source of information
in Condor, so it's possible that different things come up with different
views and answers about what is happening in the pool - in particular,
condor_status can have stale information.)

> I run plain vanilla jobs and the condor_q output says:
> 
> 339 jobs; 156 idle, 183 running, 0 held
> 

What does 'condor_q -run' have to say? It should tell you where the
schedd *thinks* those 183 jobs are running. 

> and the condor_status says:
> 
> 
>                      Machines Owner Claimed Unclaimed Matched Preempting
> 
>        INTEL/WINNT50        1     0       1         0       0          0
>        INTEL/WINNT51      183    67     115         1       0          0
>      SUN4u/SOLARIS29        2     0       0         2       0          0
> 
>                Total      186    67     116         3       0          0
> 
> and of those that are claimed, there are only 7 that say Busy, the rest seem
> stuck at idle.
> 

What does the ShadowLog on the submit machine have to say?

-Erik
Condor Support Information:
http://www.cs.wisc.edu/condor/condor-support/
To Unsubscribe, send mail to majordomo@xxxxxxxxxxx with
unsubscribe condor-users <your_email_address>