Possibility to set the input specs for GLO QC dynamically, for K0s

[chiarazampolli]

@shahor02 @f3sch

It was tested in the FST enabling the sec vtx:

export PRINT_WORKFLOW=1
export NHBPERTF=32
export SHMSIZE=8000000000
export NJOBS="$JOBS"
export DPL_REPORT_PROCESSING=1
export ARGS_EXTRA_PROCESS_o2_secondary_vertexing_workflow='--disable-cascade-finder --disable-3body-finder --disable-strangeness-tracker'

export WORKFLOW_EXTRA_PROCESSING_STEPS=TPC_DEDX,MFT_RECO,MID_RECO,MCH_RECO,MATCH_MFTMCH,MATCH_MCHMID,MUON_SYNC_RECO,ZDC_RECO,MATCH_SECVTX,MATCH_TPCTOF 
export FST_SYNC_EXTRA_WORKFLOW_PARAMETERS=QC,CALIB_LOCAL_AGGREGATOR,CALIB_LOCAL_INTEGRATED_AGGREGATOR 
export QC_REDIRECT_MERGER_TO_LOCALHOST=1 
export GEN_TOPO_WORKDIR=`pwd` 
export ALICE_O2SIM_DUMPLOG=1 
export NEvents=5 
export NEventsQED=100 
export O2SIMSEED=12345 
export QC_CONFIG_PARAM="--local-batch=QC.root"
$O2_ROOT/prodtests/full_system_test.sh

During the FST, there were issues with MFTdigits (when saving the QC output to file via the --local-batch=QC.root option) and also the TOF raw QC (which was then disabled).
For the FST, to enable sec vtx, I set:

WORKFLOW_EXTRA_PROCESSING_STEPS=...MATCH_SECVTX,MATCH_TPCTOF

where MATCH_TPCTOF can also be omitted. This same WORKFLOW_EXTRA_PROCESSING_STEPS=...MATCH_SECVTX would then be needed in online.

Note that the reco_ASYNC in the FST also needs a change in the fst script, see AliceO2Group/AliceO2#13770, since we need strangeness tracking in ASYNC for the AODs (disabled above by export ARGS_EXTRA_PROCESS_o2_secondary_vertexing_workflow='--disable-cascade-finder --disable-3body-finder --disable-strangeness-tracker')

async reco was then tested on one run from zzo.

[github-actions]

REQUEST FOR PRODUCTION RELEASES:
To request your PR to be included in production software, please add the corresponding labels called "async-" to your PR. Add the labels directly (if you have the permissions) or add a comment of the form (note that labels are separated by a ",")

+async-label <label1>, <label2>, !<label3> ...

This will add <label1> and <label2> and removes <label3>.

The following labels are available
async-2023-pbpb-apass4
async-2023-pp-apass4
async-2024-pp-apass1
async-2022-pp-apass7
async-2024-pp-cpass0
async-2024-PbPb-cpass0
async-2024-PbPb-apass1
async-2024-ppRef-apass1

[shahor02]

@chiarazampolli why not putting the

export ARGS_EXTRA_PROCESS_o2_secondary_vertexing_workflow='--disable-cascade-finder --disable-3body-finder --disable-strangeness-tracker'
export WORKFLOW_EXTRA_PROCESSING_STEPS=TPC_DEDX,MFT_RECO,MID_RECO,MCH_RECO,MATCH_MFTMCH,MATCH_MCHMID,MUON_SYNC_RECO,ZDC_RECO,MATCH_SECVTX

to the full_system_test.sh (eventually, should go to dpl-workflow.sh). I would not add MATCH_TPCTOF since we don't want to run it online.

[chiarazampolli]

Hello @shahor02 ,

I guess I have to, or it will not work (the CI fails). Another way is to switch off the K0s QC in the JSON for sync.
Before that, I am actually trying to apply another modification: forcing the K0S QC to be off if there is no SECVTX. I need to test it. I will report here tomorrow.

For now, I am moving this PR to WIP.

Chiara

[chiarazampolli]

Hello @shahor02 , @f3sch ,
Now the K0sQC part will be forced to be off if there is no SECVTX. This should make the FST pass here, and I think it is the right way to do.

[shahor02]

OK, assuming that the jq manipulations are correct, I am not familiar with this.
BTW, EPNSYNCMODE implies SYNCMODE, it is enough to check the latter.

[chiarazampolli]

Hello @shahor02 ,

Comparing lines 61 and 147 (where there are an "if"... "elif") it seems they are different. That is why we did like this. Should I remove SYNCMODE?

Chiara

[shahor02]

SYNCMODE does not imply EPNSYNCMODE, but EPNSYNCMODE means SYNCMODE running on the EPN.
That said, checking both with OR does not hurt.

[chiarazampolli]

Checkes passed, merging.