add X3872 and Psi2S to Jpsi pi pi

[ffionda]

No description provided.

[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

[sawenzel]

The CI has problems (out of disc space). I'd like to wait until this is fixed then run the CI again. @jackal1-66 : Can you trigger the CI once ready?

[jackal1-66]

Restarted after fix merged in ali-bot alisw/ali-bot#1423

[jackal1-66]

The problem was not fixed because it seems not related to the actual disk getting full, but the intermediate files generated by gcc that are getting too large. The tests will likely fail, as they did for other PRs. They will be relaunched when the issue is fixed.

[jackal1-66]

@ffionda the build/O2DPG/sim/o2 failure is real now. Could you please check why the test of /O2DPG/MC/config/PWGDQ/ini/Generator_InjectedPromptX3782Midy_TriggerGap.ini is failing? Do you have a similar behaviour locally (or on lxplus)?

[ffionda]

Ciao @jackal1-66, thanks for checking. I think the problem is that the tag R02-02-00-alice2 is needed for EvtGen
https://github.com/alisw/EVTGEN (as the PR alisw/EVTGEN@a979196 is needed for the generator in this PR). I installed this EvtGen version locally and it works for me, however I see that on lxplus only R02-02-00-alice1-XX versions are available, so the simulation of the X3782 doesn't work. Would it be possible to make R02-02-00-alice2 EvtGen version available on grid / lxplus for simulations ? Thank you

[jackal1-66]

Ciao @ffionda I checked the differences of the tags and I see no reason why the version shouldn't be bumped.
Let's wait for alisw/alidist#5693 to be merged (which should be done independently from this PR), then I will restart the CI.

[jackal1-66]

evtgen version was bumped @ffionda and the CI was restarted. The tests are still failing and this time due to a segmentation violation.

[ffionda]

Thanks @jackal1-66 for merging the PR. We would need to include it in the production software for anchored pp simulations (I tried the option here https://its.cern.ch/jira/browse/O2-5366 but it seems it doesn't work in this case since if I understood correctly it's really a new generator which is introduced and not just a different configuration of an existing generator).
So, we would need then to add the following labels:
+async-2023-pp-apass4
+async-2024-pp-apass1
+async-2022-pp-apass7
Thank you.

[sawenzel]

@ffionda @jackal1-66 : In principle this approach should work. It doesn't matter if something is updated or not as long as we use the new variable O2DPG_MC_CONFIG_ROOT (and the relevant commit for this change is itself backported).

@jackal1-66 : Could you do a test with some anchored MC job and see what's missing? I suspect that we also need the alidist change alisw/alidist#5681 to be part of older software tags. @alcaliva @chiarazampolli

[ffionda]

Ciao @sawenzel, ok thanks a lot! I don't know if this could help, but I had a deeper look (and make this grid test here https://alimonitor.cern.ch/jobs/jdl.jsp?pid=3212370991), and I think one of the issues is that we would need to make the change O2DPG_ROOT -> O2DPG_MC_CONFIG_ROOT also in other places inside the DQ folder,
in particular here https://github.com/AliceO2Group/O2DPG/blob/master/MC/config/PWGDQ/external/generator/GeneratorPromptCharmonia.C (for this specific generator)

However, there is probably another thing which I'm not sure whether it could be a problem. For this production, we would need to have Evtgen R02-02-00-alice2 (while the version pointed by existing pp production software from O2sim package, e.g. pp2022 apass7 VO_ALICE@O2PDPSuite::async-async-v1-01-02a-slc9-alidist-async-v1-01-02-1, is R02-02-00-alice1-XX). So, I don't know if this would also be fixed. Hope this helps, thanks a lot!

[jackal1-66]

Ciao @ffionda, I just opened #1817 to fix the wrong environment variables inside the scripts. It was my bad, I committed the .ini configurations change before without actually changing the macros.

[ffionda]

Thanks @jackal1-66, this PR (#1817) actually helps to pick up the correct decay tables. However, I ran some grid tests (anchoring to pp apass7), and while everything works fine for psi2s -> jpsi pi pi (see output here https://alimonitor.cern.ch/catalogue/#/alice/cern.ch/user/f/ffionda/anchorMC_PSI2_test2/output/526641), for the X3782 it doesn't work (see test on aliensh here /alice/cern.ch/user/f/ffionda/anchorMC_X3782_test2/) because of the EvtGen issue I mentioned above (i.e. EvtGen R02-02-00-alice1-XX is picked up by the production software, while we would need R02-02-00-alice2).
So, do you (@sawenzel, @jackal1-66 ) think it is possible to make the porting of this EvtGen PR alisw/EVTGEN#3 in the production software ? Or do you have any other suggestion on how we could solve this ?
Thanks a lot.

[jackal1-66]

Hello @ffionda, thanks for the tests. The modifications to EvtGen are not extreme, so I don't think it would be a problem porting them in production @chiarazampolli @alcaliva. @ffionda did you try changing the O2DPG_MC_CONFIG_ROOT variable or did you simply run the tests on the 30 of Novermber?

[ffionda]

Ciao @jackal1-66, thanks for the answer. Yes, I confirm that I changed the O2DPG_MC_CONFIG_ROOT variable setting it to the version of 30 November (here you can see the corresponding script https://alimonitor.cern.ch/users/download.jsp?view=true&path=/alice/cern.ch/user/f/ffionda/anchorMC_X3782_test2/runme.sh I used in the test to set it), but it doesn't help (as I guess EvtGen is picked up from O2sim package). Thanks a lot.

[chiarazampolli]

Hello @ffionda, thanks for the tests. The modifications to EvtGen are not extreme, so I don't think it would be a problem porting them in production @chiarazampolli @alcaliva. @ffionda did you try changing the O2DPG_MC_CONFIG_ROOT variable or did you simply run the tests on the 30 of Novermber?

Hello @jackal1-66 ,
Do I understand that you'd like to update alidist? For which build? What is the target production for this PR? Do I understand from @ffionda that she needs:
+async-2023-pp-apass4
+async-2024-pp-apass1
+async-2022-pp-apass7
The right way is that she tries to compile the sw related to those labels with updated EvGen (so updated alidist, that might need to be patched since taking master might not work). If it works, we can do the same centrally.

Can you follow this with @alcaliva and @catalinristea ? I am away these days, with very slow reaction time...

Chiara

[ffionda]

Ciao @chiarazampolli
thanks a lot for your answer and for the details. Yes, I confirm that we would need the anchoring to pp 2022 / 23 / 24 (i.e. +async-2023-pp-apass4, +async-2024-pp-apass1, +async-2022-pp-apass7).
Ok, so if I understood correctly, I have to try to compile the software production version (with the EvtGen fix in alidist) locally and see if it works (I'm not 100% sure if I should do it for all the three software production versions for the different pp years, maybe I'll start from the one of 2022...). I can try in the next days (if needed I will probably contact Alberto and Catalin to ask for more details).
Thanks a lot for your help,
Cheers,

Fiorella

[ffionda]

Ciao @chiarazampolli, @alcaliva, @catalinristea, @jackal1-66
I managed to compile locally the O2Physics@async-v1-01 / O2@async-v1-01 packages, as well as EvtGen for which I considered the updated alidist/evtgen.sh (where the correct EvtGen version R02-02-00-alice2 is picked up). The compilation of O2 / O2Physics and EvtGen (as well as the loading of the corresponding packages) worked well for me (for completeness I'm attaching the screenshot). Do you think it's possible to try it centrally ? Just let me know if you need something else from our side.
Thanks, Fiorella