Addition of histograms to the Muon DQM #22456

parbol · 2018-03-05T11:47:34Z

In this pull request I'm adding histograms to the Muon DQM in order to correct/improve three aspects:

Try to improve the monitoring of the pixel problem from the muon perspective:

Two Monitor Elements per Muon ID have been added in EfficiencyAnalyzer.cc and EfficiencyPlotter.cc showing the efficiency as a function of the phi and eta of the innermost hit of the innertrack of the muon.
One Monitor Element per Muon ID (Loose, Medium, Tight) has been added to DiMuonAnalyzer.cc to monitor the ratio between the number of bad hits and good hits in the inner track for events where the two leptons have a mass compatible with the Z mass.

Monitor variables used by the Soft Muon MVA ID

About 15 new monitor elements have been added to MuonRecoAnalyzer.cc to monitor the variables used by the Soft Muon MVA as requested by the POG.

Occupancy plot for monitoring of bad lumis

One Occupancy plot per muon ID has been added to MuonRecoAnalyzer.cc to be filled only with events where the DCS flags are not OK.

… variables used by the soft MVA muon and occupancy plots for events with DCS flags set to BAD

cmsbuild · 2018-03-05T11:47:49Z

@parbol, CMSSW_10_1_X branch is closed for direct updates. cms-bot is going to move this PR to master branch.
In future, please use cmssw master branch to submit your changes.

cmsbuild · 2018-03-05T11:48:05Z

The code-checks are being triggered in jenkins.

cmsbuild · 2018-03-05T11:50:44Z

+code-checks

Logs: https://cmssdt.cern.ch/SDT/code-checks/cms-sw-PR-22456/3730

cmsbuild · 2018-03-05T11:51:03Z

A new Pull Request was created by @parbol for master.

It involves the following packages:

DQMOffline/Muon

@vazzolini, @kmaeshima, @dmitrijus, @cmsbuild, @jfernan2, @vanbesien can you please review it and eventually sign? Thanks.
@barvic, @bellan, @abbiendi, @jhgoh, @calderona, @HuguesBrun, @drkovalskyi, @ptcox, @battibass, @trocino, @folguera, @rociovilar this is something you requested to watch as well.
@davidlange6, @slava77, @fabiocos you are the release manager for this.

cms-bot commands are listed here

jfernan2 · 2018-03-05T15:41:43Z

please test

cmsbuild · 2018-03-05T15:42:02Z

The tests are being triggered in jenkins.
https://cmssdt.cern.ch/jenkins/job/ib-any-integration/26517/console Started: 2018/03/05 17:19

cmsbuild · 2018-03-05T18:14:17Z

-1

Tested at: c48201b

You can see the results of the tests here:
https://cmssdt.cern.ch/SDT/jenkins-artifacts/pull-request-integration/PR-22456/26517/summary.html

I found follow errors while testing this PR

Failed tests: RelVals

RelVals:

When I ran the RelVals I found an error in the following worklfows:
140.53 step2

runTheMatrix-results/140.53_RunHI2011+RunHI2011+RECOHID11+HARVESTDHI/step2_RunHI2011+RunHI2011+RECOHID11+HARVESTDHI.log

150.0 step3

runTheMatrix-results/150.0_HydjetQ_B12_5020GeV_2018+HydjetQ_B12_5020GeV_2018+DIGIHI2018+RECOHI2018+HARVESTHI2018/step3_HydjetQ_B12_5020GeV_2018+HydjetQ_B12_5020GeV_2018+DIGIHI2018+RECOHI2018+HARVESTHI2018.log

cmsbuild · 2018-03-05T18:14:19Z

Comparison not run due to runTheMatrix errors (RelVals and Igprof tests were also skipped)

cmsbuild · 2018-03-16T17:05:49Z

Comparison job queued.

cmsbuild · 2018-03-16T19:09:10Z

Comparison is ready
https://cmssdt.cern.ch/SDT/jenkins-artifacts/pull-request-integration/PR-22456/26922/summary.html

Comparison Summary:

No significant changes to the logs found
Reco comparison results: 0 differences found in the comparisons
DQMHistoTests: Total files compared: 29
DQMHistoTests: Total histograms compared: 2477528
DQMHistoTests: Total failures: 1
DQMHistoTests: Total nulls: 0
DQMHistoTests: Total successes: 2477351
DQMHistoTests: Total skipped: 176
DQMHistoTests: Total Missing objects: 0
DQMHistoSizes: Histogram memory added: 1599.39 KiB( 23 files compared)
Checked 118 log files, 9 edm output root files, 29 DQM output files

jfernan2 · 2018-03-17T20:31:02Z

+1

cmsbuild · 2018-03-17T20:31:19Z

This pull request is fully signed and it will be integrated in one of the next master IBs (tests are also fine). This pull request will now be reviewed by the release team before it's merged. @davidlange6, @slava77, @smuzaffar, @fabiocos (and backports should be raised in the release meeting by the corresponding L2)

fabiocos · 2018-03-26T12:55:44Z

+1

adding ~ 78 kB of histograms, not negligible but still manageable IMO

diff on dqmMemoryStats.py output for wf 10824.0

< 2564.04 KiB Muons_miniAOD/MuonRecoAnalyzer
< 2561.48 KiB Muons/MuonRecoAnalyzer

2603.06 KiB Muons_miniAOD/MuonRecoAnalyzer
2600.19 KiB Muons/MuonRecoAnalyzer

slava77 · 2018-04-13T21:45:56Z

DQMOffline/Muon/src/EfficiencyAnalyzer.cc

@@ -264,6 +270,9 @@ void EfficiencyAnalyzer::analyze(const edm::Event & iEvent,const edm::EventSetup
      h_allProbes_pt->Fill(muon2->pt());
      h_allProbes_eta->Fill(muon2->eta());
      h_allProbes_phi->Fill(muon2->phi());
+      h_allProbes_inner_pt->Fill(muon2->innerTrack()->innerMomentum().Rho());
+      h_allProbes_inner_eta->Fill(muon2->innerTrack()->innerPosition().Eta());
+      h_allProbes_inner_phi->Fill(muon2->innerTrack()->innerPosition().Phi());


These changes have broken DQM for miniAOD jobs running separately from AOD inputs.
E.g. in workflow 136.7611 running on 1000 events

[0] Processing Event run: 277069 lumi: 81 event: 35893405 stream: 0 [1] Calling method for module EfficiencyAnalyzer/'LooseMuonEfficiencyAnalyzer_miniAOD' Exception Message: RefCore: A request to resolve a reference to a product of type 'std::vector<reco::TrackExtra>' with ProductID '2:3025' can not be satisfied because the product cannot be found. Probably the branch containing the product is not stored in the input file.

@fabozzi
did we notice the problems already in relvals? Or are we not running these regularly?
jenkins tests will likely miss this problem because it takes 2 muons somewhat consistent with Z->mumu to make it to this problematic part of the code

@slava77
there workflows are only tested in IB to maintain the sequence, we do not produce them in the relval set for regular validation campaigns

I see. So, there are no relvals which run only miniAOD with DQM included?
How are you planning to test the reminiAOD of 2016?

Sorry for the late reply. I've been absent the whole weekend. So, if I understand correctly, these jobs are failing when asking for the innerTrack to the Muons in miniAOD jobs. I guess this is because the innerTrack is not there anymore and the reference it's not valid. I have added a "IsminiAOD" boolean in such a way that this operation is not done when we are running a miniAOD workflow. However I am unsure how to test this new code is working. Is it enough to run it over a miniAOD file? Any suggestion? Thanks a lot.

Don't we store the inner track inside pat::Muon? Just use pat::Muon::track() call instead. It will check if the track is embedded and if not, it will call innerTrack() method.

Don't we store the inner track inside pat::Muon? Just use pat::Muon::track() call instead. It will check if the track is embedded and if not, it will call innerTrack() method.

Only the track is saved, not its TrackExtra.
The TrackExtra is not available in AOD

I have added a "IsminiAOD" boolean in such a way that this operation is not done when we are running a miniAOD workflow.

good

However I am unsure how to test this new code is working. Is it enough to run it over a miniAOD file? Any suggestion?

In case 2017 AOD is at hand, I suggest to start from the matrix workflow 1325.51 and substitute the ttbar inputs with a Z->mumu AOD file.
The cmsDriver command with 1K events in the test is
cmsDriver.py step2 --conditions auto:phase1_2017_realistic -s PAT,VALIDATION:@miniAODValidation,DQM:@miniAODDQM --process PAT --era Run2_2017,run2_miniAOD_94XFall17 --eventcontent MINIAODSIM,DQM --runUnscheduled --scenario pp --datatier MINIAODSIM,DQMIO --mc -n 1000 --filein filelist:step1_dasquery.log --fileout file:step2.root

I don't see why DQM would care about parameters at the inner point vs at point of closest approach, which is stored in Track. Also, it gets eta and phi of the inner point. Is it really what is needed? @parbol can you please clarify?

This was an attempt to have in the DQM variables that might be sensitive to the pixels problem. Probably the same thing can be seen using the point of closest approach (what do you think Dima?), since everything we want to see is some degradation. I'm fine with both solutions: the IsminiAOD switch (already implemented), or changing to the point of closest approach.

If it's to understand how deeply the track is going, don't you want then rho of the inner point, instead of momentum? I would expect that direction is good enough from POCA. Only location of the first hit can be interesting. Guys, what do you think?

slava77 · 2018-04-13T21:51:47Z

@fabiocos
very soon we need to validate the re-miniAOD workflows in the context of the 2016 reminiAOD.
Changes from this PR are blocking validation of these workflows.
Somewhat expedient resolution would be nice (revert is easy but perhaps not so nice for the proponents of the changes).

fabiocos · 2018-04-17T07:14:36Z

@slava77 let me follow-up in a separate issue

kpedro88 · 2018-04-18T13:19:51Z

DQMOffline/Muon/src/MuonRecoAnalyzer.cc

+	   if (&(*beamSpot)==NULL) {
+	     refPoint = beamSpot->position();
+	   } else {
+	     cout << "ERROR: No beam sport found!" << endl;


This message is appearing e.g. in 20034.0 step3 logs (phase2 workflow). cout should not be used.

Also, I'm not sure I understand the logic that you try to assign the refPoint only if the beamSpot address is null...

parbol · 2018-04-19T16:33:04Z

Hello,

I have made a new branch correcting the issues before:

https://github.com/parbol/cmssw/tree/DQMMuonDCSFlag_v2

In particular it corrects three things:

The innerTrack is tested to be "isnonNull()" before being used, otherwise the code is not executed.
The logic of the beamspot spotted before is now corrected to uses beamSpot().isValid()
One remaining cout has been replaced by a edm::Log

Please, could you let me know to which CMSSW should I send the PR against? CMSSW_10_1_X, 2_X?

Thanks

kpedro88 · 2018-04-19T16:47:58Z

Please make the PR to master branch, which currently corresponds to 10_2_X. You will need to rebase your branch (which is currently based on 10_1_X). It would also be nice to fix the typo "sport" -> "spot" in the printout. I leave it up to the DQM reviewers to decide if it should be backported.

parbol · 2018-04-19T21:17:55Z

Ok, it's done in:

#23015

Thanks

parbol added 2 commits March 4, 2018 09:23

Added plots to monitor possible problems in the pixels, monitoring of…

d161410

… variables used by the soft MVA muon and occupancy plots for events with DCS flags set to BAD

Fix adding the modules that I switheched to allow the testing

c48201b

cmsbuild changed the base branch from CMSSW_10_1_X to master March 5, 2018 11:47

cmsbuild added this to the CMSSW_10_1_X milestone Mar 5, 2018

cmsbuild added code-checks-pending comparison-pending dqm-pending orp-pending pending-signatures tests-pending labels Mar 5, 2018

cmsbuild added code-checks-approved and removed code-checks-pending labels Mar 5, 2018

cmsbuild added tests-started and removed tests-pending labels Mar 5, 2018

cmsbuild added comparison-notrun tests-rejected and removed comparison-pending tests-started labels Mar 5, 2018

Fixed the problem associated to the reading of vertex and beamspot

f400de2

cmsbuild removed code-checks-approved comparison-notrun tests-rejected labels Mar 11, 2018

cmsbuild added tests-approved and removed tests-started labels Mar 16, 2018

cmsbuild added comparison-available and removed comparison-pending labels Mar 16, 2018

cmsbuild added dqm-approved fully-signed and removed dqm-pending pending-signatures labels Mar 17, 2018

cmsbuild added orp-approved and removed orp-pending labels Mar 26, 2018

cmsbuild merged commit af82bc6 into cms-sw:master Mar 26, 2018

slava77 reviewed Apr 13, 2018

View reviewed changes

cmsbuild modified the milestones: CMSSW_10_1_X, CMSSW_10_2_X Apr 13, 2018

fabiocos mentioned this pull request Apr 17, 2018

Validation of re-miniAOD and muon DQM #22981

Closed

slava77 mentioned this pull request Apr 17, 2018

features needed for re-miniAOD of 2016 data/MC in 94X #22978

Closed

4 tasks

kpedro88 reviewed Apr 18, 2018

View reviewed changes

parbol mentioned this pull request Apr 19, 2018

The use of the innerTrack is now protected with an if, also a cout ha… #23015

Merged

smuzaffar modified the milestones: CMSSW_10_2_X, CMSSW_10_1_X Nov 26, 2018

Provide feedback

Saved searches

Use saved searches to filter your results more quickly

Addition of histograms to the Muon DQM #22456

Addition of histograms to the Muon DQM #22456

parbol commented Mar 5, 2018

cmsbuild commented Mar 5, 2018

cmsbuild commented Mar 5, 2018

cmsbuild commented Mar 5, 2018

cmsbuild commented Mar 5, 2018

jfernan2 commented Mar 5, 2018

cmsbuild commented Mar 5, 2018 •

edited

cmsbuild commented Mar 5, 2018

cmsbuild commented Mar 5, 2018

cmsbuild commented Mar 16, 2018

cmsbuild commented Mar 16, 2018

jfernan2 commented Mar 17, 2018

cmsbuild commented Mar 17, 2018

fabiocos commented Mar 26, 2018

slava77 Apr 13, 2018

fabozzi Apr 15, 2018

slava77 Apr 15, 2018

parbol Apr 16, 2018

drkovalskyi Apr 16, 2018 •

edited

slava77 Apr 17, 2018

slava77 Apr 17, 2018

drkovalskyi Apr 17, 2018

parbol Apr 17, 2018

drkovalskyi Apr 17, 2018

slava77 commented Apr 13, 2018

fabiocos commented Apr 17, 2018

kpedro88 Apr 18, 2018

parbol commented Apr 19, 2018

kpedro88 commented Apr 19, 2018

parbol commented Apr 19, 2018

Addition of histograms to the Muon DQM #22456

Addition of histograms to the Muon DQM #22456

Conversation

parbol commented Mar 5, 2018

cmsbuild commented Mar 5, 2018

cmsbuild commented Mar 5, 2018

cmsbuild commented Mar 5, 2018

cmsbuild commented Mar 5, 2018

jfernan2 commented Mar 5, 2018

cmsbuild commented Mar 5, 2018 • edited

cmsbuild commented Mar 5, 2018

cmsbuild commented Mar 5, 2018

cmsbuild commented Mar 16, 2018

cmsbuild commented Mar 16, 2018

jfernan2 commented Mar 17, 2018

cmsbuild commented Mar 17, 2018

fabiocos commented Mar 26, 2018

< 2564.04 KiB Muons_miniAOD/MuonRecoAnalyzer < 2561.48 KiB Muons/MuonRecoAnalyzer

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

drkovalskyi Apr 16, 2018 • edited

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

slava77 commented Apr 13, 2018

fabiocos commented Apr 17, 2018

Choose a reason for hiding this comment

parbol commented Apr 19, 2018

kpedro88 commented Apr 19, 2018

parbol commented Apr 19, 2018

cmsbuild commented Mar 5, 2018 •

edited

< 2564.04 KiB Muons_miniAOD/MuonRecoAnalyzer
< 2561.48 KiB Muons/MuonRecoAnalyzer

drkovalskyi Apr 16, 2018 •

edited