Delphes 3.0.9 crash

[jstupak]

I ordinarily run jobs with ~5k input events and have no problems. But I am trying to run some larger jobs (50k events) now and the majority of my jobs crash. I get two different errors:

INFO: initializing module FastJetFinder
INFO: initializing module CAJetFinder
INFO: initializing module ConstituentFilter
INFO: initializing module BTagging
INFO: initializing module BTaggingLoose
INFO: initializing module TauTagging
INFO: initializing module UniqueObjectFinder
INFO: initializing module ScalarHT
INFO: initializing module TreeWriter
Reading pythia_events.hep
ERROR: Unsupported block type.-----------------------------------] (0.00%)
Command exited with non-zero status 1

and

INFO: initializing module FastJetFinder
INFO: initializing module CAJetFinder
INFO: initializing module ConstituentFilter
INFO: initializing module BTagging
INFO: initializing module BTaggingLoose
INFO: initializing module TauTagging
INFO: initializing module UniqueObjectFinder
INFO: initializing module ScalarHT
INFO: initializing module TreeWriter
Reading pythia_events.hep
ERROR: Inconsistent size of arrays. File is probably corrupted.--] (0.00%)

The card I used is available here: ​http://cmssw.cvs.cern.ch/cgi-bin/cmssw.cgi/UserCode/spadhi/Snowmass/Cards/delphes_card_Snowmass_NoPileUp.tcl?revision=1.3&view=markup

Do you understand why I would see these errors when running on 50k events, but not ~5k, and can something be done about this?

Thanks,
John Stupak

[Alexandre Mertens]

Hello,

Thanks for reporting the errors.

I just tried to run your card with a 50k events file I have generated with madgraph+pythia6 and it did work.

Can u provide me one of your files, I'll try to find what's happening?

Cheers,

Alexandre

[Saptaparna Bhattacharya]

Hi Delphes authors,

I see the same errors as well. My pythia_events.hep files have 100K events each and I see both types of error messages and subsequent crashes when I run Delphes on them.

Is there a way to circumvent this situation?

Thanks a lot!
Sapta

[Pavel Demin]

I'm not sure that it's a Delphes problem.

There was a problem with STDHEP files generated with MadGraph+Pythia. If I remember well, files greater than 2GB were corrupted.

We should check with the MadGraph developers if this problem is fixed.

Is there any way to check if the files that can be read by any other program? Maybe PGS?

[Pavel Demin]

I've just tested several STDHEP files produced by MadGraph.

I have 50MB, 3.3GB and 4.0GB files.

The 50MB and 3.3GB files can be read without any problem.

The 4.0GB file can't be read.

In the 4.0GB file, already the 2nd block has a block ID number equal 2. Normally, it should be 4. The following blocks have completely wrong block ID numbers. So, I'd say that it's corrupted from the very beginning.

I've just sent a bug report to the MadGraph developers:

​https://bugs.launchpad.net/madgraph5/+bug/1218842

I'm afraid that I can't do more than this for this problem.

[Pavel Demin]

I've just proposed a fix for this problem to the MadGraph developers.

Hopefully, it'll be available in on of the future MadGraph releases.

For more details, please see

​https://bugs.launchpad.net/madgraph5/+bug/1218842