Draft: Introduce trigger error for CITest validation

Make the TriggerMC CITest (!50455 (merged)) fail on purpose to validate error reporting.

cc @rbielski

Jenkins please retry a build

After this will also try trigger count changes.

This merge request affects 1 package:

• Trigger/TriggerCommon/TriggerMenuMT

This merge request affects 1 file:

• Trigger/TriggerCommon/TriggerMenuMT/python/HLT/Menu/GenerateMenuMT.py

Adding @sutt ,@ademaria ,@iriu ,@malconad ,@carquin ,@ggonella ,@bernius ,@hrussell ,@gipezzul ,@dzanzi as watchers

added Trigger TriggerMenu master labels

CI Result FAILURE (hash 5c922e44)

	Athena	AthSimulation	AthGeneration	AnalysisBase	AthAnalysis	DetCommon
externals
cmake
make
required tests
optional 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
AthSimulation: number of compilation errors 0, warnings 0
AthGeneration: number of compilation errors 0, warnings 0
AnalysisBase: number of compilation errors 0, warnings 0
AthAnalysis: number of compilation errors 0, warnings 0
DetCommon: number of compilation errors 0, warnings 0
For experts only: Jenkins output [CI-MERGE-REQUEST-CC7 47033]

Looks good - I'm just not sure about running noerror.sh which duplicates a subset of the CheckLog functionality. Especially since this extra output moves the key summary line:

2022-02-16T133332 CET TrigValSteering INFO     Test trigAna_RDOtoRDOTrig_v1Dev_build finished with code 65 because the following required steps failed: ['Reco_tf', 'CheckLog', 'CountRefComp']

further from the end, hiding it a bit in the flood of messages.

Yeah, could add an option to disable noerror.sh. Will think about that. On the other side, there is some merit of having a common error reporting across all tests that cannot be disabled. Another easy fix would be to change the log level of that INFO messsage to ERROR in TrigValSteering

Sure, the changing of the level for this message is a good idea in any case, but the workaround then will be by even more duplication. Probably OK, just a bit confusing.

One more comment on noerror.sh - I noticed only now, but this message:

 noerror.sh> Please check /build/ci-builds/master/Athena/build/Athena/AtlasTest/CITest/CMakeFiles/ciTestRun/TriggerMC/TriggerMC.log

is only confusing and not helpful. No one has access to this file and even if they had, it's deleted shortly after the pipeline finishes (or as soon as a new one starts on the node). Also the content of that file is just printed above, so no need to check the file.

It's the same message post.sh would print. We use that noerror.sh script also for regular unit tests. And when running locally, the message is relevant. We could certainly tweak it to discover if it runs in the CI or not.

added 1 commit

• 083840d9 - Draft: Introduce trigger count diffs for CITest validation

Compare with previous version

Jenkins please retry a build

WARNING: big files (>100K) are found in the changeset

276K in file Trigger/TrigValidation/TrigAnalysisTest/share/ref_RDOtoRDOTrig_v1Dev_build.ref

This merge request affects 1 package:

• Trigger/TrigValidation/TrigAnalysisTest

This merge request affects 1 file:

• Trigger/TrigValidation/TrigAnalysisTest/share/ref_RDOtoRDOTrig_v1Dev_build.ref

Adding @vmartin ,@okumura as watchers

added changes-trigger-counts label

CI Result FAILURE (hash 083840d9)

	Athena	AthSimulation	AthGeneration	AnalysisBase	AthAnalysis	DetCommon
externals
cmake
make
required tests
optional 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
AthSimulation: number of compilation errors 0, warnings 0
AthGeneration: number of compilation errors 0, warnings 0
AnalysisBase: number of compilation errors 0, warnings 0
AthAnalysis: number of compilation errors 0, warnings 0
DetCommon: number of compilation errors 0, warnings 0
For experts only: Jenkins output [CI-MERGE-REQUEST-CC7 47046]

This one seems all right :)

All looks as expected, so we can close this experiment.

closed