Considering all surfaces (not only material ones) in TimedExtrapolator.cxx

Considering all surfaces (not only material ones) as 'compatible' in TimedExtrapolator.cxx to restore SCT hits on both sides. ATLASSIM-3745

This merge request affects 1 package:

• Tracking/TrkExtrapolation/TrkExTools

added Tracking review-pending-level-1 labels

added Simulation bugfix labels

Jenkins please retry a build

This merge request affects 1 package:

• Tracking/TrkExtrapolation/TrkExTools

CI Result SUCCESS

	Athena
externals
cmake
make
required tests
optional tests

Full details available at NICOS MR-20694-2019-01-29-02-03
Athena: number of compilation errors 0, warnings 166
CI Jenkins server is switched to https://atlas-sit-ci.cern.ch. It is accessible world-wide (behind CERN SSO). In old links to Jenkins server aibuild080.cern.ch:8080 should be replaced with atlas-sit-ci.cern.ch For experts only: Jenkins output [CI-MERGE-REQUEST 32897]

Hi, @arnaez.

While the output ROOT files look sane, I'm getting segmentation faults in Athena,21.3,latest with both ATLFASTIIF_IDOnly and Fatras_newExtrapolation_IDOnly:

PyJobTransforms.trfExe.execute 2019-01-29 06:09:22,968 INFO EVNTtoHITS executor returns 139
PyJobTransforms.trfExe.validate 2019-01-29 06:09:23,896 ERROR Validation of return code failed: EVNTtoHITS got a SIGSEGV signal (exit code 139) (Error code 65)
PyJobTransforms.trfExe.validate 2019-01-29 06:09:23,913 INFO Scanning logfile log.EVNTtoHITS for errors in substep sim
PyJobTransforms.transform.execute 2019-01-29 06:09:24,010 CRITICAL Transform executor raised TransformValidationException: EVNTtoHITS got a SIGSEGV signal (exit code 139)
PyJobTransforms.transform.execute 2019-01-29 06:09:27,379 WARNING Transform now exiting early with exit code 65 (EVNTtoHITS got a SIGSEGV signal (exit code 139))

On the bright side, using the ATLFASTII simulator EVNTtoHITS is successful. You said you ran successfully over 2 events -- I guess you ran with ATLFASTII?

P.S. Maybe make this a WIP? (I guess it doesn't matter since it's aimed for 21.3 and won't be swept into master until first merged into 21.3.)

Hi,

Yes, I got 'Py:Athena INFO leaving with code 0: "successful run"' after 2 events indeed, nothing bad apparently when trying on 'asetup Athena,21.3,latest' All other changes I had locally are the ones in arnaez/athena@e730c126, arnaez/athena@a9235075, arnaez/athena@6608512c and arnaez/athena@bd3b06e4. I thought they should not make any real difference but now I'm seeing the very bottom of arnaez/athena@bd3b06e4 so I'm not sure any more. @jchapman might want to comment.

Best, Olivier.

Marking review-user-action-required until the discussion is resolved.

Tadej (L1)

added review-user-action-required label and removed review-pending-level-1 label

Could the stack trace of the seg fault (if available) but posted here?

@arnaez has anyone else been able to reproduce the crash that @vpascuzz saw?

FYI @nstyles the seg fault is right at the end of the job when all the descructors are being called and doesn't appear to be related to these changes as Vince sees it in 21.3 builds without these changes too.

added review-approved label and removed review-user-action-required label

Approving this one as we can't reproduce the seg faults, so I think it's safe to try out in the nightly.

merged

mentioned in commit b95e0dcc

Hi, @jchapman. Thanks for the info. Just to add, the crash only occurs in Athena,>=21.3.9. Things are fine in 21.3.8, for example. @tlari confirms he sees the same crash as well with Sim_tf.py but FastChain_tf.py runs without problems.

Sorry to @nstyles for not having posted the stack trace yet — I meant to but forgot. Will attach it when I get to the office.

Hi @vpascuzz,

OK - most likely this is related to the externals update in 21.3.9 then. Let's open a separate JIRA ticket and discuss it there.

Cheers,

John

added sweep:done label

mentioned in commit c2c57cf9