Bug #1044: TOF data structure 'charge' - MAUS - MICEmine

Bug #1044

TOF data structure 'charge'

Added by Rogers, Chris over 11 years ago. Updated almost 11 years ago.

Status:

Closed

Priority:

High

Assignee:

Rajaram, Durga

Category:

TOF

Target version:

MAUS-v0.4.2

Start date:

10 July 2012

Due date:

% Done:

100%

Estimated time:

Workflow:

New Issue

Description

I am getting an occasional exception in TOF reconstruction

Traceback (most recent call last):
  File "/home/cr67/MAUS/maus_release/src/common_py/ErrorHandler.py", line 159, in HandleCppException
    raise(CppError(error_message))
ErrorHandler.CppError: In branch recon_events
In branch tof_event
In branch tof_slab_hits
In branch tof2
In branch charge
Missing required branch charge converting json->cpp at PointerItem::SetCppChild

For some reason the charge branch is sometimes missing. Why is this? Anyway, every time this happens it causes the reconstruction to crash for that spill... a bit unpleasant. FYI I attach a log file - note I was doing some load testing here and hit ctrl-c at the end (MAUS was still running). Probably worth checking with a smaller data file...

Files

recon.log (76.3 KB) recon.log

Rogers, Chris, 10 July 2012 11:35

Updated by Rajaram, Durga over 11 years ago

The only thing I see so far [ running on the same run# ] is that sometimes there are no slab_hits for TOF2
and as a result there is no "charge".
In the JSON output, the slab_hits for tof2 show up as null.
Will look more closely.

From the log it's not clear what the event # is on the error -- when it says "processed 38 spills" is it spills?
or actual particle events?
it would be helpful if there's a way to skip events and get to the offending event for purposes of debugging.
Is there such an option?

Updated by Rogers, Chris over 11 years ago

when it says "processed 38 spills" is it spills? or actual particle events?

No, good point, it is "Events" (which in this case means DAQ events), including start of burst and end of burst

it would be helpful if there's a way to skip events and get to the offending event for purposes of debugging. Is there such an option?

No, the place to implement such a thing would probably be line 76-89 of src/common_py/framework/single_thread.py which is the main single-threaded for loop

Updated by Rogers, Chris over 11 years ago

The only thing I see so far [ running on the same run# ] is that sometimes there are no slab_hits for TOF2 and as a result there is no "charge". In the JSON output, the slab_hits for tof2 show up as null.

Yeah I tried quite hard to get the output to be reasonably consistent, removing things like hits showing up as null, when I set up the Json <-> C++ conversion. It was quite a painful process. Probably indicates that our output data in json was flaky and inconsistent, hopefully we can start to resolve that...