Fix bug for allowed truncation collections without Aux
There was a bug included to allow the truncation of collections, which prevented the serialisation of the Aux collections allowed for truncation.
Seen in various repro, flagged in ATR-29387 and ATR-29384.
Tested with:
- CI test: Trigger_athenaHLT_v1PhysP1, checking that the flagged collections are still present with
trigbs_dumpHLTContentInBS_run3.py --hltres --sizeSummary data....
-
test_trigP1_v1PhysP1_T0Mon_build.py
locally fails:Test trigP1_v1PhysP1_T0Mon_build finished with code 65 because the following required steps failed: ['Reco_tf.Tier0Reco', 'CheckLog’]
, and in particular failing in DQHistogramMerge for reasons not correlated with this MR - run on 7 truncated events in the debug stream, and checked that errors on the flagged collections disappeared:
athenaHLT.py -o output --file=/eos/atlas/atlastier0/rucio/data24_13p6TeV/debug_TruncatedHLTResult/00475192/data24_13p6TeV.00475192.debug_TruncatedHLTResult.daq.RAW/data24_13p6TeV.00475192.debug_TruncatedHLTResult.daq.RAW._lb0388._SFO-12._0001.data --file=/eos/atlas/atlastier0/rucio/data24_13p6TeV/debug_TruncatedHLTResult/00475192/data24_13p6TeV.00475192.debug_TruncatedHLTResult.daq.RAW/data24_13p6TeV.00475192.debug_TruncatedHLTResult.daq.RAW._lb0428._SFO-13._0001.data --file=/eos/atlas/atlastier0/rucio/data24_13p6TeV/debug_TruncatedHLTResult/00475192/data24_13p6TeV.00475192.debug_TruncatedHLTResult.daq.RAW/data24_13p6TeV.00475192.debug_TruncatedHLTResult.daq.RAW._lb0435._SFO-12._0001.data --file=/eos/atlas/atlastier0/rucio/data24_13p6TeV/debug_TruncatedHLTResult/00475192/data24_13p6TeV.00475192.debug_TruncatedHLTResult.daq.RAW/data24_13p6TeV.00475192.debug_TruncatedHLTResult.daq.RAW._lb0777._SFO-15._0001.data --file=/eos/atlas/atlastier0/rucio/data24_13p6TeV/debug_TruncatedHLTResult/00475192/data24_13p6TeV.00475192.debug_TruncatedHLTResult.daq.RAW/data24_13p6TeV.00475192.debug_TruncatedHLTResult.daq.RAW._lb0777._SFO-20._0001.data --file=/eos/atlas/atlastier0/rucio/data24_13p6TeV/debug_TruncatedHLTResult/00475192/data24_13p6TeV.00475192.debug_TruncatedHLTResult.daq.RAW/data24_13p6TeV.00475192.debug_TruncatedHLTResult.daq.RAW._lb0780._SFO-16._0001.data --file=/eos/atlas/atlastier0/rucio/data24_13p6TeV/debug_TruncatedHLTResult/00475192/data24_13p6TeV.00475192.debug_TruncatedHLTResult.daq.RAW/data24_13p6TeV.00475192.debug_TruncatedHLTResult.daq.RAW._lb0898._SFO-14._0001.data --file=/eos/atlas/atlastier0/rucio/data24_13p6TeV/debug_TruncatedHLTResult/00475192/data24_13p6TeV.00475192.debug_TruncatedHLTResult.daq.RAW/data24_13p6TeV.00475192.debug_TruncatedHLTResult.daq.RAW._lb0898._SFO-19._0001.data TriggerJobOpts.runHLT Trigger.triggerMenuSetup="PhysicsP1_pp_run3_v1_HLTReprocessing_prescale" Trigger.doLVL1=True >athenaHLT.log 2>&1
I added more info in the ERROR to identify the first collection that got oversized. Indeed even with the TrucantionAllowed on 2 collections, some errors are raised for other collections going oversize before them in the list of the EDM file, but the message was indicating only the largest collections. The current message tells which is the collections that caused the oversize (the first one dropped) for each moduleId.
Merge request reports
Activity
added 24.0 label
added Trigger review-pending-level-1 labels
added bugfix label
added urgent label
CI Result FAILURE (hash 3d214e13)Athena externals cmake make tests Full details available on this CI monitor view. Check the JIRA CI status board for known problems
Athena: number of compilation errors 0, warnings 0
For experts only: Jenkins output [CI-MERGE-REQUEST-EL9 9877] (remote access info)Please check the failing tests
~L1
removed review-pending-level-1 label
added review-user-action-required label
removed urgent label
added 69 commits
-
3d214e13...1079387c - 62 commits from branch
atlas:24.0
- 1de8d43c - added ZBTopo to L1exclued thresholds
- 86948c71 - add LArSaturation as not cheked TOB
- 2d1f7e10 - fix bug in allow-trauncation collections
- 17b34da2 - added debug, to be removed
- 91d49985 - more comments
- c3a022f4 - fix deferred buffer per module ID
- c6ea5a2f - fix conflicts
Toggle commit list-
3d214e13...1079387c - 62 commits from branch
added review-pending-level-1 label and removed review-user-action-required label
I made some more tests, and found that my change was not enough. The deferredInterfaceBuffer needs to target the corresponding module ID to be serialised together with the Aux. The latter change should fix the crash here. I’m also trying to run other tests that failed with release 44, but they take long
added TriggerEDM label
CI Result SUCCESS (hash c6ea5a2f)Athena externals cmake make tests Full details available on this CI monitor view. Check the JIRA CI status board for known problems
Athena: number of compilation errors 0, warnings 0
For experts only: Jenkins output [CI-MERGE-REQUEST-EL9 10068] (remote access info)