Bug #13949

Basic reco failure from Track3DKalman

Added by Thomas Brooks over 3 years ago. Updated over 3 years ago.

Target version:
Start date:
Due date:
% Done:


Estimated time:
First Occurred:
Occurs In:


Working on sbndcode (develop branch)

Some cry generated events cause the FHiCL file standard_reco_sbnd_basic.fcl to give the error

%MSG-s ArtException:  PostPathEndRun end_path 27-Sep-2016 06:04:29 CDT  PostEndRun
cet::exception caught in art
---- EventProcessorFailure BEGIN
  An exception occurred during current event processing
  ---- ScheduleExecutionFailure BEGIN

    ---- FatalRootError BEGIN
      Fatal Root Error: @SUB=TRotation::Rotate(angle,axis)
       zero axis
      cet::exception going through module Track3DKalman/trackkalspscc run: 1 subRun: 0 event: 3
    ---- FatalRootError END
    Exception going through path reco
  ---- ScheduleExecutionFailure END
  cet::exception caught in EventProcessor and rethrown
  Another exception was caught while trying to clean up files after
  the primary exception.  We give up trying to clean up files at
  this point.  The description of this additional exception follows:
  ---- FatalRootError BEGIN
    Fatal Root Error: @SUB=TTree::SetEntries
    Tree branches have different numbers of entries, with 10 maximum.
  ---- FatalRootError END
---- EventProcessorFailure END

And then art exits with status 65

This error does not occur for every event and the event number which it occurs at changes.

The commands I am using are as follows:

lar -c prodcosmics_cry_sbnd.fcl -n 10
lar -c standard_g4_sbnd.fcl -s prodcosmics...
lar -c standard_detsim_sbnd.fcl -s prodcosmics...
lar -c standard_reco_sbnd_basic.fcl -s prodcosmics...

The outputs of all the commands can be found at /sbnd/data/users/tbrooks/cry_reco_bug/
Specifically, the detsim output is at prodcosmics_cry_sbnd_CosmicsGen-20160927T105732_G4-20160927T110023_DetSim-20160927T110249.root and the full reco stage output is cry_reco_bug.log.


#1 Updated by Thomas Brooks over 3 years ago

I've just Noticed that the same thing happens with corsika generated events as well.

#2 Updated by Gianluca Petrillo over 3 years ago

  • Subject changed from Basic reco failure on some cry events to Basic reco failure from Track3DKalman
  • Status changed from New to Accepted
  • Priority changed from Normal to High

The module Track3DKalman is not widely used any more. It might even be unsupported, as far as I know. The author, Eric Church, might know something about that.
In general, I think at this stage it would benefit SBND to stick to an algorithm/module that is used widely enough that we can get bug fixes and maintenance for free,
My reading of "this stage" is: we need something working out of the box for simple studies.

I am raising the priority of the issue, since it is a potential stopper for a basic SBND resource (that is, standard_reco_sbnd_basic.fcl).

#3 Updated by Gianluca Petrillo over 3 years ago

  • Status changed from Accepted to Resolved
  • Assignee set to Gianluca Petrillo
  • % Done changed from 0 to 80

An attempt to fix the problem has been pushed as ed0b2d3f8efa43dd4f3241ba141193459f0016b9 .
In there, we are switching to PMAlgTrackMaker for tracking, and to LineCluster for clustering (since it's the only cluster maker that also produces hits, which is a feature required by the old PMAlgTrackMaker from LArSoft 5.12.01 which we are using).
I was able to produce some tracks in some of the events from a single particle sample (prodsingle_sbnd.fcl). I did not performe any physics verification.

Feedback is strongly appreciated.

Thanks to Robert Sulej, Michelle Stancari and Hector Mendez who in different way contributed to this commit.

#4 Updated by Gianluca Petrillo over 3 years ago

  • Status changed from Resolved to Closed

Closing as resolved. Reopen if there are still pending problems.

Also available in: Atom PDF