[IPython-dev] Problems with ipcluster and LSF protocol (v.0.13.2)

[Florian M. Wagner]

Hey Min,

thanks for the reply. I think c.HubFactory.ip = '*' did it for me. The 
engines are found by the Client now. Although, I am experiencing memory 
issues. I think, since ipcluster starts the engines with bsub only, I 
just have the standard resources available. Is it possible to pass 
arguments like the requested walltime etc. to ipcluster or would I have 
to do that over the config files prior to job execution?

Great work with IPython.parallel. I love it!

Am 30.07.2013 20:05, schrieb MinRK:
> Do you see any jobs that are running or queued? Does the batch system 
> log output of jobs?
>
>
> On Tue, Jul 30, 2013 at 3:07 AM, Florian M. Wagner 
> <wagnerfl at student.ethz.ch <mailto:wagnerfl at student.ethz.ch>> wrote:
>
>     Dear IPython Users,
>
>     I am trying to start engines on a big cluster, which uses the Load
>     Sharing Facility protocol. The engines seem to start correctly:
>
>         *[wagnerfl at brutus1 ~]$ ipcluster start --profile=cluster
>         --engines=LSF --n=4 --IPClusterApp.log_level=DEBUG*
>         2013-07-30 11:55:31,893.893 [IPClusterStart] Using existing
>         profile dir:
>         u'/cluster/home02/erdw/wagnerfl/.ipython/profile_cluster'
>         2013-07-30 11:55:31.914 [IPClusterStart] Starting ipcluster
>         with [daemon=False]
>         2013-07-30 11:55:31.915 [IPClusterStart] Creating pid file:
>         /cluster/home02/erdw/wagnerfl/.ipython/profile_cluster/pid/ipcluster.pid
>         2013-07-30 11:55:31.916 [IPClusterStart] Starting Controller
>         with LocalControllerLauncher
>         2013-07-30 11:55:32.915 [IPClusterStart] Starting 4 Engines
>         with LSFJob array.
>         2013-07-30 11:55:33.199 [IPClusterStart] Job submitted with
>         job id: '27633338'
>         2013-07-30 11:56:03.200 [IPClusterStart] Engines appear to
>         have started successfully
>
>
>     However, I cannot access them from the Client.
>
>         In [1]: from IPython.parallel import Client
>         In [2]: c = Client(profile='cluster')
>         In [3]: c.ids
>         Out[3]: []
>
>     Could you give me a hint how to debug this, or is it not worth
>     trying and should I better wait for 1.0?
>
>     Thanks a lot in advance
>     Florian
>
>
>
>
>
>
>     _______________________________________________
>     IPython-dev mailing list
>     IPython-dev at scipy.org <mailto:IPython-dev at scipy.org>
>     http://mail.scipy.org/mailman/listinfo/ipython-dev
>
>
>
>
> _______________________________________________
> IPython-dev mailing list
> IPython-dev at scipy.org
> http://mail.scipy.org/mailman/listinfo/ipython-dev

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mail.python.org/pipermail/ipython-dev/attachments/20130731/e5ca81f0/attachment.html>