Trigger simulation fails for some runs #844
Description
I noticed that the trigger simulation fails for some runs in spring 2017. I am running mcwrapper in the CentOS7 container with the settings to reproduce REST production 2017-01 ver04 and analysis ver70:
Environment file: /group/halld/www/halldweb/html/halld_versions/recon-2019_11-ver01_13.xml
Analysis Environment file: /group/halld/www/halldweb/html/halld_versions/analysis-2017_01-ver70.xml
For run 30788, the reconstruction step prints this message:
---------DL1MCTrigger (Brun): JANA_CALIB_CONTEXT =variation=mc
JANA >>Creating DTranslationTable for run 30788
JANA >>Reading translation table from calib DB: Translation/DAQ2detector ...
JANA >>23166 channels defined in translation table
File with name: /home/hdops/CDAQ/daq_dev_v0.31/daq/config/hd_all/LIST doesn't exist (not associated) with run: 30788
------------ Trigger Settings ---------------
----------- FCAL -----------
FCAL_CELL_THR = 65
FCAL_NSA = 10
FCAL_NSB = 3
FCAL_WINDOW = 10
----------- BCAL -----------
BCAL_CELL_THR = 20
BCAL_NSA = 19
BCAL_NSB = 3
BCAL_WINDOW = 20
Do not use RCDB for the trigger simulation. Default (spring 2017) trigger settings are used
As a consequence, no event passes the trigger selection and all monitoring histograms are empty.
For a other runs of the spring period (e.g. 30300), very different settings are loaded and my events pass the trigger:
---------DL1MCTrigger (Brun): JANA_CALIB_CONTEXT =variation=mc
JANA >>Creating DTranslationTable for run 30300
JANA >>Reading translation table from calib DB: Translation/DAQ2detector ...
JANA >>23166 channels defined in translation table
------------ Trigger Settings ---------------
----------- FCAL -----------
FCAL_CELL_THR = 165
FCAL_NSA = 10
FCAL_NSB = 3
FCAL_WINDOW = 10
----------- BCAL -----------
BCAL_CELL_THR = 120
BCAL_NSA = 19
BCAL_NSB = 3
BCAL_WINDOW = 20
TYPE FCAL_E BCAL_E EN_THR NHIT LANE FCAL_EMIN FCAL_EMAX BCAL_EMIN BCAL_EMAX PATTERN
BFCAL 25 1 45000 0 0 200 65535 0 65535 0x0
BFCAL 0 1 54000 0 2 0 0 0 0 0x0
PS 0 0 0 0 3 0 0 0 0 0x0
JANA >>In DL1MCTrigger_factory, loading constants...
I don't know yet for how many runs this happens, but this silent effect could potentially bias the trigger efficiency and by that the determination of the cross section.