cta-observatory / ctapipe

Low-level data processing pipeline software for CTAO or similar arrays of Imaging Atmospheric Cherenkov Telescopes
https://ctapipe.readthedocs.org
BSD 3-Clause "New" or "Revised" License
63 stars 267 forks source link

ctapipe-process all calibration fields skipped #2620

Closed BastienLacave closed 2 hours ago

BastienLacave commented 4 hours ago

Describe the bug When using ctapipe-process on real LST data, all calibration, pedestal and flatfields fields are skipped and the calibration therefore crashes. (see .txt attached).

To Reproduce Steps to reproduce the behavior:

  1. On run 2929, date 20201118 (see config file attached):

pedestal : /fefs/aswg/data/real/monitoring/PixelCalibration/Cat-A/drs4_baseline/20201118/v0.9.2/drs4_pedestal.Run02920.0000.h5

calibration : /fefs/aswg/data/real/monitoring/PixelCalibration/Cat-A/calibration/20201118/v0.9.2/calibration_filters_52.Run02921.0000.h5

time calibration : /fefs/aswg/data/real/monitoring/PixelCalibration/Cat-A/drs4_time_sampling_from_FF/20191124/v0.9.2/time_calibration.Run01625.0000.h5

pointing : /fefs/onsite/monitoring/driveLST1/DrivePositioning/DrivePosition_log_20201118.txt

run summary : /fefs/aswg/data/real/monitoring/RunSummary/RunSummary_20201118.ecsv

R0 file : /fefs/aswg/data/real/R0/20201118/LST-1.1.Run02929.0000.fits.fz

Command: ctapipe-process -t 1 --overwrite --input={file} --output={output_file} --config={args.config} --write-parameters --EventTypeFilter.allowed_types=SUBARRAY --progress

This happens on some runs and not others, I have checked that the calib files are not empty and the fields skipped are in the files. I find the calibration files using the onsite package from lstchain. The same run works but with calib files found in the "pro" directory, but I am interested by the files in the "v0.9.2" directory, is there any reason this works for one and not the other ?

ctapipe_standard_pmt_config_02929.json sba0.sh.39890718.txt

morcuended commented 4 hours ago

The same run works but with calib files found in the "pro" directory, but I am interested by the files in the "v0.9.2" directory, is there any reason this works for one and not the other ?

Just one note, pro directory points to the latest version of calibration produced with lstchain v0.10

maxnoe commented 2 hours ago

You cannot use hdf5 files produced with ctapipe 0.12 (lstchain 0.9) with ctapipe versions newer than 0.16 (lstchain 0.10 uses ctapipe 0.19).

This is not a bug, it was a conscious breaking change.

BastienLacave commented 1 hour ago

Ah very well then, thank you