Crash in PyCBC Live #5171
Description
I do not remember seeing this one before. X's added by myself.
2025-08-21T17:31:05.895-07:00 pycbc-live 0 Analyzing from X
2025-08-21T17:31:05.946-07:00 pycbc-live 0 Skipping recalculation of L1 PSD, X
2025-08-21T17:31:05.995-07:00 pycbc-live 0 Skipping recalculation of H1 PSD, X
2025-08-21T17:31:12.061-07:00 pycbc-live 0 V1 frame is late, giving up
2025-08-21T17:31:12.062-07:00 pycbc-live 0 Insufficient data for V1 analysis
2025-08-21T17:31:12.062-07:00 pycbc-live 0 Gathering triggers
2025-08-21T17:31:12.076-07:00 pycbc-live 0 Checking H1's DQ vector
2025-08-21T17:31:12.077-07:00 pycbc-live 0 Keeping X H1 triggers after DQ flags
2025-08-21T17:31:12.077-07:00 pycbc-live 0 Reading H1's iDQ information
2025-08-21T17:31:12.077-07:00 pycbc-live 0 iDQ flagged X H1 triggers
2025-08-21T17:31:12.077-07:00 pycbc-live 0 Checking L1's DQ vector
2025-08-21T17:31:12.078-07:00 pycbc-live 0 Keeping X L1 triggers after DQ flags
2025-08-21T17:31:12.078-07:00 pycbc-live 0 Reading L1's iDQ information
2025-08-21T17:31:12.079-07:00 pycbc-live 0 iDQ flagged X L1 triggers
2025-08-21T17:31:12.082-07:00 pycbc-live 0 H1-L1: X coincs, X bytes
2025-08-21T17:31:12.082-07:00 pycbc-live 0 adding singles to the background estimate...
2025-08-21T17:31:13.754-07:00 pycbc-live 0 H1-L1: X background and zerolag coincs
2025-08-21T17:31:13.772-07:00 pycbc-live 0 Clustering H1-L1 coincs
2025-08-21T17:31:13.773-07:00 pycbc-live 0 Clustering events over 8.024025692304535 s window
2025-08-21T17:31:13.779-07:00 pycbc-live 0 X triggers remaining
2025-08-21T17:31:13.871-07:00 pycbc-live 0 Coincs 0: L1-H1: X in cbuffer
2025-08-21T17:31:13.873-07:00 pycbc-live 0 Found L1-H1 coinc with ifar X
2025-08-21T17:31:13.874-07:00 pycbc-live 0 computing followup data for coinc
2025-08-21T17:31:13.874-07:00 pycbc-live 0 Generating SPAtmplt, duration X s, index X, starting from X Hz
2025-08-21T17:31:13.984-07:00 pycbc-live 0 Generating SPAtmplt, duration X s, index X, starting from X Hz
2025-08-21T17:31:14.195-07:00 pycbc-live 0 Adding V1 to candidate, pvalue X, X samples
2025-08-21T17:31:14.763-07:00 pycbc-live 0 Computing p_astro
Traceback (most recent call last):
File "…/lib/python3.9/runpy.py", line 197, in _run_module_as_main
return _run_code(code, main_globals, None,
File "…/lib/python3.9/runpy.py", line 87, in _run_code
exec(code, run_globals)
File "…/lib/python3.9/site-packages/mpi4py/__main__.py", line 7, in <module>
main()
File "…/lib/python3.9/site-packages/mpi4py/run.py", line 198, in main
run_command_line(args)
File "…lib/python3.9/site-packages/mpi4py/run.py", line 47, in run_command_line
run_path(sys.argv[0], run_name='__main__')
File "…/lib/python3.9/runpy.py", line 288, in run_path
return _run_module_code(code, init_globals, run_name,
File "…/lib/python3.9/runpy.py", line 97, in _run_module_code
_run_code(code, mod_globals, init_globals,
File "…/lib/python3.9/runpy.py", line 87, in _run_code
exec(code, run_globals)
File "…/bin/pycbc_live", line 1379, in <module>
evnt.check_coincs(list(results.keys()), best_coinc, psds)
File "…/bin/pycbc_live", line 527, in check_coincs
event = CandidateForGraceDB(
File "…/lib/python3.9/site-packages/pycbc/io/gracedb.py", line 272, in __init__
min(eff_distances),
TypeError: '>' not supported between instances of 'float' and 'NoneType'
V1 frame files were not being delivered at that time, so the message "Adding V1 to candidate, pvalue X, X samples" points out an incorrect decision on which detectors to use. This then probably caused the source probabilities code to crash.