In order to assess if some buspatches systematically have bad pedestal values,
I’ve scanned the 2017 OCDB with the AliMUONCDB::ShowFaultyPedestalsBusPatches.
In order to do so, first task is to copy OCDB files locally (to speed up subsequent scans).
The simplest for that task is probably to make a list of all files :
aliensh
> find /alice/data/2017/OCDB/MUON/Calib/Pedestals *.root > ped.txt
Edit the ped.txt to prepend alien:// in front of all files.
Then use the VAF::CopyFromRemote method to copy the files locally.
Once done the scan can proceed on some runlist.
AliMUONCDB::ShowFaultyPedestalsBusPatches("runlist.ped",0.9,4000,4000,0.33,"buspatch.badped.33percent-of-the-time","local:///alice/data/2017/OCDB")
Leading, in this example, to the list of buspatches that have more than 90% of bad pedestal channels (where bad means either mean larger than 4000 or sigma larger than 4000) more than 33% of the time.
❯ cat buspatch.badped.33percent-of-the-time.txt
BP 0420 is bad for 4 runs in 4 ( 100.00%)
BP 0427 is bad for 4 runs in 4 ( 100.00%)
BP 0705 is bad for 3 runs in 8 ( 37.50%)
BP 0811 is bad for 627 runs in 808 ( 77.60%)
BP 0826 is bad for 672 runs in 800 ( 84.00%)
BP 0827 is bad for 647 runs in 808 ( 80.07%)
BP 1002 is bad for 654 runs in 814 ( 80.34%)
BP 1120 is bad for 654 runs in 821 ( 79.66%)
BP 1121 is bad for 10 runs in 30 ( 33.33%)
BP 1336 is bad for 760 runs in 798 ( 95.24%)
BP 1722 is bad for 6 runs in 10 ( 60.00%)