MadGraph5Error : Histogram with run_id '0' was not found in the specified XML source.

[José David Ruiz]

Dear MG authors,

I'm experiencing a problem when running pythia 8 in a server. The error message is the following:

$ ./bin/madevent
launch in debug mode
************************************************************
* *
* W E L C O M E to *
* M A D G R A P H 5 _ a M C @ N L O *
* M A D E V E N T *
* *
* * * *
* * * * * *
* * * * * 5 * * * * *
* * * * * *
* * * *
* *
* VERSION 2.5.1 20xx-xx-xx *
* *
* The MadGraph5_aMC@NLO Development Team - Find us at *
* https://server06.fynu.ucl.ac.be/projects/madgraph *
* *
* Type 'help' for in-line help. *
* *
************************************************************
INFO: load configuration from /afs/cern.ch/work/j/jruizalv/public/MC_tools/MG5_aMC_v2_5_1/BBbar_13TeV/Cards/me5_configuration.txt
INFO: load configuration from /afs/cern.ch/work/j/jruizalv/public/MC_tools/MG5_aMC_v2_5_1/input/mg5_configuration.txt
INFO: load configuration from /afs/cern.ch/work/j/jruizalv/public/MC_tools/MG5_aMC_v2_5_1/BBbar_13TeV/Cards/me5_configuration.txt
Using default eps viewer "evince". Set another one in ./input/mg5_configuration.txt
Using default web browser "firefox". Set another one in ./input/mg5_configuration.txt
BBbar_13TeV>pythia8 run_01
INFO: compile directory
compile Source Directory
Which programs do you want to run?
    0 / auto : running existing cards
    1. pythia8 : Pythia8
    3. delphes : Pythia8 + Delphes.
 [0, 1, 3, auto, pythia8, delphes][60s to answer]
>1
Will run in mode pythia8
Do you want to edit a card (press enter to bypass editing)?
/------------------------------------------------------------\
| 1. pythia8 : pythia8_card.dat |
\------------------------------------------------------------/
 you can also
   - enter the path to a valid card.
 [0, done, 1, pythia8, enter path][90s to answer]
>0
INFO: Update the dependent parameter of the param_card.dat
INFO: Running Pythia8 [arXiv:1410.3012]
Splitting .lhe event file for PY8 parallelization...
Submitting Pythia8 jobs...
Pythia8 shower jobs: 1 Idle, 7 Running, 0 Done [1 seconds]
Pythia8 shower jobs: 0 Idle, 3 Running, 5 Done [3 seconds]
Pythia8 shower jobs: 0 Idle, 2 Running, 6 Done [3 seconds]
Pythia8 shower jobs: 0 Idle, 1 Running, 7 Done [3 seconds]
Pythia8 shower jobs: 0 Idle, 0 Running, 8 Done [3 seconds]
Merging results from the split PY8 runs...
Error detected in "pythia8 run_01 "
write debug file /afs/cern.ch/work/j/jruizalv/public/MC_tools/MG5_aMC_v2_5_1/BBbar_13TeV/run_01_tag_1_debug.log
If you need help with this issue please contact us on https://answers.launchpad.net/mg5amcnlo
MadGraph5Error : Histogram with run_id '0' was not found in the specified XML source.

I attach the log file and the config of the system where I'm running. I'm also experiencing the same problem in lxplus.

Thanks,
Jose

[Valentin Hirschi (valentin-hirschi)]

Could you check if the problem is also present when turning off PY8 parallelization?
You can do so by setting the MG5aMC option 'nb_core' to 1.
Also, could you attach the xml file that contains the DJR histograms generated by Pythia8?
It should be in:

/afs/cern.ch/work/j/jruizalv/public/MC_tools/MG5_aMC_v2_5_1/BBbar_13TeV/Events/run_01/djrs.dat

if it was not already erased when the error occurred.

Also, given how quickly Pythia8 ran, it seems unlikely that there was not a problem there.
Could you look at the log of the Shower runs and report if you seen any mention of a problem at the end, i.e. in:

/afs/cern.ch/work/j/jruizalv/public/MC_tools/MG5_aMC_v2_5_1/BBbar_13TeV/Events/run_01/tag_1_pythia8.log

(All the paths above are for a single run of Pythia8. The corresponding path for a parallel run of Pythia8 will be in the split_<x> directories).

Let me know

[José David Ruiz (jotadram6)]

Answers between lines:

>Could you check if the problem is also present when turning off PY8 parallelization?
>You can do so by setting the MG5aMC option 'nb_core' to 1.

The issue disappears when turning off the parallelization.

>Also, could you attach the xml file that contains the DJR histograms generated by Pythia8?
>It should be in:
>
>/afs/cern.ch/work/j/jruizalv/public/MC_tools/MG5_aMC_v2_5_1/BBbar_13TeV/Events/run_01/djrs.dat
>
>if it was not already erased when the error occurred.
>

It is not there, so I guess it was erased before the error occured.

>Also, given how quickly Pythia8 ran, it seems unlikely that there was not a problem there.
>Could you look at the log of the Shower runs and report if you seen any mention of a problem at >the end, i.e. in:
>
>/afs/cern.ch/work/j/jruizalv/public/MC_tools/MG5_aMC_v2_5_1/BBbar_13TeV/Events/run_01/tag_1_pyth>ia8.log
>

I'm attaching the log.

>(All the paths above are for a single run of Pythia8. The corresponding path for a parallel run >of Pythia8 will be in the split_<x> directories).

I'm also attaching the directories.

[Valentin Hirschi (valentin-hirschi)]

I don't think you attached the shower log, but only the MG5aMC crash log,
which is not useful to check the behavior of Pythia8.
Ideally, I would like to look at the Pythia8 log at (when turning on
parallelization):

/afs/cern.ch/work/j/jruizalv/public/MC_tools/MG5_aMC_v2_5_
1/BBbar_13TeV/Events/run_01/split_<x>/[...]_pyth
<http://cern.ch/work/j/jruizalv/public/MC_tools/MG5_aMC_v2_5_1/BBbar_13TeV/Events/run_01/tag_1_pyth>
ia8.log

In any case, It's likely not a problem of PY8 since it appears only with
parallelization, so could you tell me exactly the process and the run
parameters (including number of events and cores) that you are considering
so that I can try to reproduce this issue?

On Tue, Dec 13, 2016 at 7:48 PM, José David Ruiz <email address hidden>
wrote:

> Answers between lines:
>
> >Could you check if the problem is also present when turning off PY8
> parallelization?
> >You can do so by setting the MG5aMC option 'nb_core' to 1.
>
> The issue disappears when turning off the parallelization.
>
> >Also, could you attach the xml file that contains the DJR histograms
> generated by Pythia8?
> >It should be in:
> >
> >/afs/cern.ch/work/j/jruizalv/public/MC_tools/MG5_aMC_v2_5_
> 1/BBbar_13TeV/Events/run_01/djrs.dat
> >
> >if it was not already erased when the error occurred.
> >
>
> It is not there, so I guess it was erased before the error occured.
>
> >Also, given how quickly Pythia8 ran, it seems unlikely that there was not
> a problem there.
> >Could you look at the log of the Shower runs and report if you seen any
> mention of a problem at >the end, i.e. in:
> >
> >/afs/cern.ch/work/j/jruizalv/public/MC_tools/MG5_aMC_v2_5_
> 1/BBbar_13TeV/Events/run_01/tag_1_pyth>ia8.log
> >
>
> I'm attaching the log.
>
> >(All the paths above are for a single run of Pythia8. The corresponding
> path for a parallel run >of Pythia8 will be in the split_<x>
> directories).
>
> I'm also attaching the directories.
>
> ** Attachment added: "PY8_parallelization.tar.gz"
> https://bugs.launchpad.net/mg5amcnlo/+bug/1649725/+
> attachment/4791418/+files/PY8_parallelization.tar.gz
>
> --
> You received this bug notification because you are a bug assignee.
> https://bugs.launchpad.net/bugs/1649725
>
> Title:
> MadGraph5Error : Histogram with run_id '0' was not found in the
> specified XML source.
>
> Status in MadGraph5_aMC@NLO:
> New
>
> Bug description:
> Dear MG authors,
>
> I'm experiencing a problem when running pythia 8 in a server. The
> error message is the following:
>
> $ ./bin/madevent
> launch in debug mode
> ************************************************************
> * *
> * W E L C O M E to *
> * M A D G R A P H 5 _ a M C @ N L O *
> * M A D E V E N T *
> * *
> * * * *
> * * * * * *
> * * * * * 5 * * * * *
> * * * * * ...

[José David Ruiz (jotadram6)]

Please look inside the tar.gz file I have attached. Inside I have put the file tag_1_pythia8.log

[José David Ruiz (jotadram6)]

I attach the Cards directory in a tar.gz file.

[Valentin Hirschi (valentin-hirschi)]

Looking into the PY8 log files that you sent in your PY8_parallelization tarball, I see the following:

"""
 PYTHIA Abort from Pythia::Pythia: unmatched version numbers : in code 8.219 but in XML 8.205

Start generating events
 PYTHIA Error in Settings::flag: unknown key JetMatching:doVeto

Inclusive cross section: 0.00000000e+00 +- 0.00000000e+00 mb
"""

So there seems to be some mangling between different version of Pythia8 installed on your system.

If you are not running in multi-core but on a cluster instead, then the origin of this problem is that the XML read by Pythia8 at run time will be read on the worker node which has an older version of PY8 installed.
There is not workaround implemented yet, and we have ideas on how to solve that in partnership with PY8 authors, and we will hopefully have this issue fixed in the next releases.
Therefore please confirm that you are attempting to run PY8 in parallel on a cluster and not in multicore.
I assign Stefan P. to this issue, since we already discussed it with him and I want him to be aware of this thread.

[deborah (pinna)]

Dear MG authors,

I am trying to run Pythia8 (installed via Madgraph MG5_aMC_v2_5_5) and I am experiencing the same problem. I have set in the mg5_configurations file run_mode = 0 or 2 but still having the same issue.

Is there now a solution to this problem?

Thank you in advance,
Deborah

[Valentin Hirschi (valentin-hirschi)]

Deborah,

Can you confirm that the error reported in the tag_1_pythia8.log file is also in your case something along the lines:

"""
 PYTHIA Abort from Pythia::Pythia: unmatched version numbers : in code 8.219 but in XML 8.205

Start generating events
 PYTHIA Error in Settings::flag: unknown key JetMatching:doVeto

Inclusive cross section: 0.00000000e+00 +- 0.00000000e+00 mb
"""

If so, it is probably a mangling of Pythia8 version if you are running on a cluster.
So can you confirm that you are running on a cluster and not in multicore?

[deborah (pinna)]

Dear Valentin,

yes the error I have is along the lines of:

"""
 PYTHIA Abort from Pythia::Pythia: unmatched version numbers : in code 8.219 but in XML 8.205

Start generating events
 PYTHIA Error in Settings::flag: unknown key JetMatching:doVeto

Inclusive cross section: 0.00000000e+00 +- 0.00000000e+00 mb
"""

I have set in the mg5_configuration run_mode = 0 or 2 and I am working on a lxplus machine.
Probably it was due to an incopatibility with some settings I had on lxplus because setting explicitely:

export PYTHIA8DATA=/afs/cern.ch/work/d/dpinna/couplingStudies/MG5_aMC_v2_5_5/HEPTools/pythia8/share/Pythia8/xmldoc/

did solve the problem.

Thanks,
Deborah