21.2 muon selection updates

This merge request provides updates to the MuonSelectorTools for upcoming recommendations updates. In particular:

• Veto of BMG chambers for high-pT working point due to possible misalignments (only in data17 and data18).
• Reoptimized bad muon veto in association with the high-pT working point for some efficiency gain at very high pT.
• A particular selection of good 2-station muons for the high-pT working point, provided for now as an optional addition configurable via the property "Use2stationMuonsHighPt".
• A new low-pT working point based on multivariate analysis, provided for now primarily for testing, enabled via the property "UseMVALowPt".

Tagging @szambito

NB: Selector updates are not going to affect derivations

This merge request affects 1 package:

• PhysicsAnalysis/MuonID/MuonSelectorTools

added Analysis review-pending-level-1 labels

added urgent label

Hi everyone, adding the "urgent" label - it'd be nice to have this merged in parallel to !20869 (merged) as part of the same release. The changes are small and hopefully uncontroversial. This version of the tool was already tested by few analyses, and approved by MCP.

Cheers, Stefano (cc: @fsforza)

I had a quick look and I think it looks mostly good. I had one concrete suggestion for an improvement.

Since CMake files are touched I'm afraid I have to escalate to L2 though.

Best, Christian

added review-pending-level-2 label and removed review-pending-level-1 label

Few additional comments, otherwise looks fine.

Cheers,
         Pavol [L2]

added review-user-action-required label

Hi all,

thanks for the useful feedback! In a burst of proactivity, I added few scattered replies here and there; leaving the rest to @mkbugge (probably tomorrow morning).

Cheers, Stefano

changed the description

CI Result SUCCESS

	AnalysisBase	AnalysisTop	AthAnalysis	AthDerivation
externals
cmake
make
required tests
optional tests

Full details available at NICOS MR-21067-2019-02-12-19-17
AnalysisBase: number of compilation errors 0, warnings 9
AnalysisTop: number of compilation errors 0, warnings 10
AthAnalysis: number of compilation errors 0, warnings 10
AthDerivation: number of compilation errors 1, warnings 54
For experts only: Jenkins output [CI-MERGE-REQUEST 33694]

Hi all, I take that the AthDerivation compilation error has nothing to do with this?

Cheers, Stefano

Actually, can somebody confirm that this doesn't affect/break the derivations?

added 1 commit

• 700ddf07 - Updates following MR comments

Compare with previous version

This merge request affects 1 package:

• PhysicsAnalysis/MuonID/MuonSelectorTools

added review-pending-level-1 label and removed review-pending-level-2 review-user-action-required labels

CI Result SUCCESS

	AnalysisBase	AnalysisTop	AthAnalysis	AthDerivation
externals
cmake
make
required tests
optional tests

Full details available at NICOS MR-21067-2019-02-13-12-35
AnalysisBase: number of compilation errors 0, warnings 9
AnalysisTop: number of compilation errors 0, warnings 10
AthAnalysis: number of compilation errors 0, warnings 10
AthDerivation: number of compilation errors 0, warnings 54
For experts only: Jenkins output [CI-MERGE-REQUEST 33726]

I think the new commit looks fine.

@pavol can the open discussions be resolved?

Eric (L1)

All issues have been addressed and CI passed, so approving the review. I'll leave Nils or derivation coordination to resolve the final discussion before merging.

AR Shifter

added review-approved label and removed review-pending-level-1 label

added alsoTargeting:21.0 alsoTargeting:21.3 labels

resolved all discussions

@egramsta, @guirriec, @boeriu: Just tagging you as an FYI, as this may affect derivations.

mentioned in commit 27a5dd7b

merged

mentioned in merge request !21110 (merged)

added sweep:done label

added sweep:failed label

Sweep summary
failed:

• 21.0
• 21.3

removed alsoTargeting:21.0 alsoTargeting:21.3 labels

hi all,

Yesterday I tried to add the labels for the sweep into 21.0 and 21.3 but it failed...

This update is not stricktly needed for the two mentioned releases, so no problem. But I tried to look for the log file of the failure to debug it and I didn't find it. Could you point it to me?

Thanks a lot for the help!

The automatic "sweep_MR.py [ERRORS]" email says:

03:03:00 DEBUG MR title for target branch 21.0: 'Sweeping !21067 (merged) from 21.2 to 21.0. 21.2 muon selection updates' 03:03:00 DEBUG MR title for target branch 21.3: 'Sweeping !21067 (merged) from 21.2 to 21.3. 21.2 muon selection updates' 03:03:01 DEBUG https://gitlab.cern.ch:443 "POST /api/v4/projects/53790/repository/branches HTTP/1.1" 201 905 03:03:02 DEBUG https://gitlab.cern.ch:443 "POST /api/v4/projects/53790/repository/commits/27a5dd7b/cherry_pick HTTP/1.1" 400 194

03:03:02 CRITICAL failed to cherry pick merge commit '27a5dd7b' with Sorry, we cannot cherry-pick this merge request automatically. This merge request may already have been cherry-picked, or a more recent commit may have updated some of its content. 03:03:02 CRITICAL failed to cherry-pick '27a5dd7b' into '21.0' ***** Hint: check merge conflicts on a local copy of atlas/athena.git ********** git checkout 21.0 git cherry-pick -m 1 27a5dd7b git status

---

if that helps. Not sure if there is a log somewhere else.

do I understand it correctly that sweeping failure means this MR doesn't appear in upstream/21.2 even if I do fetch upstream? But since this MR was merged, this one is considered in CI?

I am asking because I failed to grab this MR locally but I saw this update in the CI of my MR (!21110 (merged))

-Keisuke

No it means it won't appear in upstream/21.0 and upstream/21.3.

ah, OK, thanks for clarifying @krumnack !

-Keisuke

All sweeps failed, but in !29763 (merged), the MuonSelectorTools package was copied to master, thus, these changes went into master. They did not go into 21.0, 21.3 and 21.9, but since the reprocessing and also Run3 will be done with master, we do not need to update the muon working point definitions in 21.0, 21.3 and 21.9 I think. Thus, removing sweep:failed label.

removed sweep:failed label