Re: Thread 0 hasn't responded in 8 seconds.

["Tarbert, Claire" <ctarbert@indiana.edu>]

Hello,

I've been seeing the same thing running analysis code on the Indiana 
cluster.  Any suggestions how to avoid this would be much appreciated.

Cheers
Claire


----- Message from leverinb@uregina.ca ---------

> Hi  all,
>
> While trying to run analysis code on the cluster, which has been busy
> lately between Claire and I, sometimes a job gets a little slow and
> if nothing happens for 8 seconds it kills the job. Isn't this a
> little impatient, I suppose, for running many jobs? Is there a reason
> for 8 seconds and not 30 seconds? Just curious and a little
> frustrated.  (I've included some of the log file for my job below.)
>
>
> -Blake
>
>
>
>
>
>
>
> 145.0 events   12.0Hz  (average rate: 3.2Hz)
>               174.0 events   32.0Hz  (average rate: 3.4Hz)
>                             234.0 events   0.0Hz  (average rate:
> 4.8Hz)                                        347.0 events   0.0Hz
> (average rate: 6.8Hz)                                         347.0
> events   0.0Hz  (average rate: 6.5Hz)
>         347.0 events   0.0Hz  (average rate: 6.2Hz)
>                      Thread 0 hasn't responded in 8 seconds.
> (run:event=1:337) Delisting ...             Caught HUP signal for
> thread 0x423c0bb0 thread exiting...
>
> *** Break *** segmentation violation
> Generating stack trace...
>


----- End message from leverinb@uregina.ca -----