Error when running gridpack

Hello again,

ater a closer look I would like to rectify some statements.

Indeed the creation of the gridpack is not as successful.
I am running it in cluster mode and jobs finishes extremely quickly.

at the end I see the error:

INFO: All jobs finished ESC[0m
End surveyESC[0m
combine_eventsESC[0m
Combining EventsESC[0m
ESC[1;34mFail to read the number of unweighted events in the combine.log fileESC[0mESC[0m
Creating root filesESC[0m
** Calculating number of events to process. Please wait...
** Input file contains 0 events
** Exiting...

yet there is nothing concerning which type of error was causing this.

As additional information I can say that I installed LHApdf and I do manage to run with (for instance) CT10,
but when I try to revert back to the default pdf in the run card, the lines above is what I see.

thanks a lot,

Valerio

Hi Valerio,

Which cluster did you use?
Is the jobs submitted on the cluster?
Do you set the queue correctly?
Is your cluster accepting the jobs?
(Some cluster requires some additional information in addition of the queue, and you need then to modify the submission script by hand)

Cheers,

Olivier

On Nov 18, 2013, at 2:01 PM, Valerio Dao <email address hidden> wrote:

> Question #239401 on MadGraph5 changed:
> https://answers.launchpad.net/madgraph5/+question/239401
>
> Valerio Dao gave more information on the question:
> Hello again,
>
> ater a closer look I would like to rectify some statements.
>
> Indeed the creation of the gridpack is not as successful.
> I am running it in cluster mode and jobs finishes extremely quickly.
>
> at the end I see the error:
>
> INFO: All jobs finished ESC[0m
> End surveyESC[0m
> combine_eventsESC[0m
> Combining EventsESC[0m
> ESC[1;34mFail to read the number of unweighted events in the combine.log fileESC[0mESC[0m
> Creating root filesESC[0m
> ** Calculating number of events to process. Please wait...
> ** Input file contains 0 events
> ** Exiting...
>
> yet there is nothing concerning which type of error was causing this.
>
> As additional information I can say that I installed LHApdf and I do manage to run with (for instance) CT10,
> but when I try to revert back to the default pdf in the run card, the lines above is what I see.
>
> thanks a lot,
>
> Valerio
>
> --
> You received this question notification because you are a member of
> MadTeam, which is an answer contact for MadGraph5.

Hi Olivier,

thanks for the prompt answer.
sorry for not providing necessary information (I am currently travelling).

I don't think this problem is related to the usage of a cluster since:
- when setting CT10 PDFs the jobs run without problems
- with the cteq6.1l (default in MG) the jobs fails also without the cluster option.
  In particular when running interactively I get a lot of:
STOP
STOP
STOP
STOP
STOP
STOP
... for each subprocess.

My only suspicion is that after I follow the procedure to install the LHA pdf somehow
I am not able anymore to run with the default pdfs within MG.
I know that this sounds weird but so far is a possible explanation that describes the symthoms
..... (maybe I am totally wrong)

I will try to provide additional information as soon as I will be connected to a terminal.

thanks,

Valerio

Hi Valerio,

of course you are right this doesn't seems to be linked to the cluster then.
I'll try to reproduce your error tommorow.

Cheers,

Olivier

Hello Olivier,

any news on this topic?

Even when I run a new process I keep getting a lot of:

STOP
STOP
STOP
STOP

... do you have any idea of what can cause this?

thanks a lot,

Valerio

Hi,

I can't reproduce your problem.
Could you look at some of the files:
madevent/SubProcesses/P?????/G??/GridRun_1_log.txt
This might indicates why it crashes.

Cheers,

Olivier