Geometry D110: failure in creating Tracker misalignment scenarios

[mmusich]

This issue stems from PR #45764.
After #45370 a bunch of unit tests relying on deprecated geometries have started to fail, which was fixed at #45764.
The default geometry (D110 - i.e. Tracker T35) cannot be used for createRandomlyMisalignedGeometry_Phase2_cfg.py as doing so as in:

diff --git a/Alignment/TrackerAlignment/test/Misalignments/createRandomlyMisalignedGeometry_Phase2_cfg.py b/Alignment/TrackerAlignment/test/Misalignments/createRandomlyMisalignedGeometry_Phase2_cfg.py
index e73a48ca19a..9ab4f03e6dc 100644
--- a/Alignment/TrackerAlignment/test/Misalignments/createRandomlyMisalignedGeometry_Phase2_cfg.py
+++ b/Alignment/TrackerAlignment/test/Misalignments/createRandomlyMisalignedGeometry_Phase2_cfg.py
@@ -4,7 +4,8 @@ import FWCore.ParameterSet.VarParsing as VarParsing
 import copy, sys, os
 
 import Configuration.Geometry.defaultPhase2ConditionsEra_cff as _settings
-GLOBAL_TAG, ERA = _settings.get_era_and_conditions("2026D107")  # this should really be _settings.DEFAULT_VERSION :(
+#GLOBAL_TAG, ERA = _settings.get_era_and_conditions("2026D107")  # this should really be _settings.DEFAULT_VERSION :(
+GLOBAL_TAG, ERA = _settings.get_era_and_conditions(_settings.DEFAULT_VERSION)
 process = cms.Process("Misaligner", ERA)
 
 ###################################################################
@@ -48,7 +49,7 @@ process.MessageLogger.cout = cms.untracked.PSet(
 ###################################################################
 # Ideal geometry producer and standard includes
 ###################################################################
-process.load('Configuration.Geometry.GeometryExtended2026D107Reco_cff')
+process.load('Configuration.Geometry.GeometryExtended2026DefaultReco_cff')
 process.trackerGeometry.applyAlignment = True
 
 ###################################################################

results in

----- Begin Fatal Exception 24-Aug-2024 17:54:17 CEST-----------------------
An exception of category 'BadHierarchy' occurred while
   [0] Calling EventProcessor::runToCompletion (which does almost everything after beginJob and before endJob)
Exception Message:
[AlignableTrackerBuilder] Pixel GeomDet (subdetector 1) is not a GeomDetUnit.
----- End Fatal Exception -------------------------------------------------

see also #45764 (comment).

Possibly related to #43097

[cmsbuild]

cms-bot internal usage

[cmsbuild]

A new Issue was created by @mmusich.

@Dr15Jones, @antoniovilela, @makortel, @mandrenguyen, @rappoccio, @sextonkennedy, @smuzaffar can you please review it and eventually sign/assign? Thanks.

cms-bot commands are listed here

[makortel]

assign geometry, upgrade, alca

[cmsbuild]

New categories assigned: geometry,upgrade,alca

@Dr15Jones,@civanch,@bsunanda,@makortel,@mdhildreth,@srimanob,@subirsarkar,@saumyaphor4252,@perrotta,@consuegs you have been requested to review this Pull request/Issue and eventually sign? Thanks

[makortel]

FYI @cms-sw/trk-dpg-l2

[mmusich]

Another symptom.
By adding this change:

diff --git a/Geometry/TrackerGeometryBuilder/test/python/testTrackerHierarchy_cfg.py b/Geometry/TrackerGeometryBuilder/test/python/testTrackerHierarchy_cfg.py
index a10920d3b09..be6f958f2f1 100644
--- a/Geometry/TrackerGeometryBuilder/test/python/testTrackerHierarchy_cfg.py
+++ b/Geometry/TrackerGeometryBuilder/test/python/testTrackerHierarchy_cfg.py
@@ -1,10 +1,14 @@
 import FWCore.ParameterSet.Config as cms
 
-process = cms.Process("GeometryTest")
+from Configuration.Eras.Era_Phase2C17I13M9_cff import Phase2C17I13M9
+
+process = cms.Process('GeometryTest',Phase2C17I13M9)
+
 process.load("FWCore.MessageLogger.MessageLogger_cfi")
 
 # Choose Tracker Geometry
-process.load("Configuration.Geometry.GeometryExtended2021Reco_cff")
+process.load('Configuration.Geometry.GeometryExtended2026D110Reco_cff')
 
 process.source = cms.Source("EmptySource")
 

and then running the test, it results in:

$ cmsRun testTrackerHierarchy_cfg.py                                                                                                                                              Begin processing the 1st record. Run 1, Event 1, LumiSection 1 on stream 0 at 03-Sep-2024 16:03:31.938 CEST

GeometricDet Hierarchy

In Tracker Geom there are 30400 modules
in filling An exception of category 'InvalidReference' occurred.
Exception Message:
Trie::insert called with a key already in collection;
key value: 

303042566 1
1/1/1/1/ 303042565


GDet Hierarchy

In Tracker Geom there are 43708 modules
in filling An exception of category 'InvalidReference' occurred.
Exception Message:
Trie::insert called with a key already in collection;
key value: 

303042564 1
1/1/1/1/ 303042565

[mmusich]

This will be solved by #45993

[mmusich]

solutions proposed:

• bugfix: remove DoubleSensGeomDet (ph-2 InnerTracker) #45993 (master, CMSSW_14_2_X)
• backport in 14_1_X of PR#45993: bug-fix: remove DoubleSensGeomDet (ph-2 InnerTracker) #45996 (CMSSW_14_1_X)

[mmusich]

@cms-sw/alca-l2 @cms-sw/geometry-l2 in order to be sure to process correctly ph2 samples produced in CMSSW_14_0_X, how open would you be to a backport of #45993 to that cycle?

[perrotta]

@cms-sw/alca-l2 @cms-sw/geometry-l2 in order to be sure to process correctly ph2 samples produced in CMSSW_14_0_X, how open would you be to a backport of #45993 to that cycle?

In the PR description of #45993 is written "removing the DoubleSensGeomDet introduced for phase-2 Inner Tracker modules (T33 tracker geometry and following T35, T36, T37, T38 and T39) but actually not used in the code (and unlikely to be used in future)." Given that such PR removes some deprecated geometries, and it can help processing correctly Phase2 samples, for AlCaDB I don't see counterindications to the backport.

[mmusich]

thanks @perrotta, I opened #46030

[srimanob]

Can this issue be closed?

[mmusich]

Can this issue be closed?

As far as I am concerned yes, though people are still checking if there are issues when using a release with the fix in conjunction with the samples already produced in 14.0.x. I let the L2s to sign if they are satisfied.

[perrotta]

+alca

• This issue was fixed
• Possible additional problems as mentioned in Geometry D110: failure in creating Tracker misalignment scenarios #45792 (comment) can be dealt with by some other dedicated github issue if and when they show up

[srimanob]

+Upgrade

[civanch]

+geometry

[cmsbuild]

This issue is fully signed and ready to be closed.

[makortel]

@cmsbuild, please close