2025 Wfs Skeleton and Updates to Data Wfs (2024,2023,2022)#46445
2025 Wfs Skeleton and Updates to Data Wfs (2024,2023,2022)#46445cmsbuild merged 4 commits intocms-sw:masterfrom
Conversation
|
@AdrianoDee, CMSSW_14_2_X branch is closed for direct updates. cms-bot is going to move this PR to master branch. |
|
cms-bot internal usage |
|
please test |
|
+code-checks Logs: https://cmssdt.cern.ch/SDT/code-checks/cms-sw-PR-46445/42304
|
|
A new Pull Request was created by @AdrianoDee for master. It involves the following packages:
@AdrianoDee, @Martin-Grunewald, @Moanwar, @atpathak, @consuegs, @kskovpen, @miquork, @mmusich, @perrotta, @srimanob, @subirsarkar, @sunilUIET can you please review it and eventually sign? Thanks. cms-bot commands are listed here |
| @@ -13,6 +13,7 @@ | |||
| 'relval2022' : 'Fake2', | |||
| 'relval2023' : 'Fake2', | |||
| 'relval2024' : 'GRun', | |||
There was a problem hiding this comment.
are we going to get 2025 relvals to run in IBs and PRs?
There was a problem hiding this comment.
IBs no, PRs I've added a single ttbar noPu.
There was a problem hiding this comment.
(explaining myself, if we overhaul completely 2024 in favor of 2025, we can demote relval2024 to Fake2 in this cycle as the purpose of having GRun here is to keep having workflows running it somewhere)
There was a problem hiding this comment.
Meaning: I didn't add the in this PR but it would be doable.
There was a problem hiding this comment.
(explaining myself, if we overhaul completely 2024 in favor of 2025, we can demote
relval2024toFake2in this cycle as the purpose of havingGRunhere is to keep having workflows running it somewhere)
Then yes we could do that.
There was a problem hiding this comment.
we're not going to re-run the 2024 HLT in any other release than 14.0.X in production. I don't see the reason to validate the 2024 HLT in this cycle. Of course "offline" validators can continue to validate offline reconstruction.
There was a problem hiding this comment.
This is clear, I wanted to understand if it's anyway useful to keep an eye on the 2024 GT (for the "offline" validations).
There was a problem hiding this comment.
If it's not, I'll move everything to 2025 and that's it.
There was a problem hiding this comment.
I've moved @relval2024 to Fake and added a couple of 2025 wfs to IB tests.
There was a problem hiding this comment.
This is clear, I wanted to understand if it's anyway useful to keep an eye on the 2024 GT (for the "offline" validations).
2024 pp GTs are supposed to be "completed" by now.
There is still some work in progress for the 2024 HI GTs, but they are validated via different workflows than the ones addressed here.
|
-1 Failed Tests: RelVals RelValsValueError: Undefined workflows: 16834.0 |
|
please test |
|
Pull request #46445 was updated. @AdrianoDee, @Martin-Grunewald, @Moanwar, @atpathak, @cmsbuild, @consuegs, @kskovpen, @miquork, @mmusich, @perrotta, @srimanob, @subirsarkar, @sunilUIET can you please check and sign again. |
|
please test |
|
-1 Failed Tests: RelVals RelVals-INPUT RelValsRelVals-INPUT
|
| 'relval2024' : 'GRun', | ||
| 'relval2024' : 'Fake2', | ||
| 'relval2025' : 'GRun', | ||
| 'relval2026' : '75e33', |
There was a problem hiding this comment.
Hmm, 2026 is rather confusing by now as it is by now still phase1, rather than phase2 which should now be 2030?
When is that update redefining 2026 foreseen?
There was a problem hiding this comment.
Indeed, it is the next item on my todo list.
| steps['HLTDR3_2023B']=merge( [ {'-s':'L1REPACK:Full,HLT:@%s'%hltKey2023,},{'--conditions':'auto:run3_hlt_relval'},{'--era' : 'Run3'},steps['HLTD'] ] ) | ||
|
|
||
| steps['HLTDR3_2024']=merge( [ {'-s':'L1REPACK:Full,HLT:@%s'%hltKey2024,},{'--conditions':'auto:run3_hlt_relval'},{'--era' : 'Run3_2024'},steps['HLTD'] ] ) | ||
| steps['HLTDR3_2024']=merge( [ {'-s':'L1REPACK:Full,HLT:@%s'%hltKey2025,},{'--conditions':'auto:run3_hlt_relval'},{'--era' : 'Run3_2024'},steps['HLTD'] ] ) |
There was a problem hiding this comment.
I understand you use hltKey2025 here to exercise the GRun menu as hltKey2024 is now the Fake2 menu, but it is a bit confusing - when do you foresee to change it back here to hltKey2024? - or could one simply swap in 2025 wfs alltogether rather than making 2024 hybrid?
There was a problem hiding this comment.
Being this data wfs I think we could roll it back only when 2025 data wfs come. To have 2025 before that (to have 2024 wfs running with @relval2024, the only possibility I see is to have some "fake" 2025 data workflows running actually on 2024 data. But I think this could be even more confusing. Note that for MC instead the key used for 2024 workflows is @relval2024 and 2025 wfs will exercise the GRun.
|
(files are being copied to T2_CH_CERN) |
|
please test |
|
-1 Failed Tests: RelVals RelVals-INPUT
RelVals
Expand to see more relval errors ...
RelVals-INPUT
Expand to see more relval errors ...
|
|
please test with #46559 |
|
+1 Size: This PR adds an extra 12KB to repository Comparison SummarySummary:
|
|
+pdmv |
|
+hlt
|
| wf_number = wf_number + offset_era * e_n | ||
| wf_number = wf_number + offset_pd * p_n | ||
| wf_number = wf_number + offset_events * evs | ||
| wf_number = wf_number + 0.0001 * 0.01 |
There was a problem hiding this comment.
Just curious: why did you move to the hardcoded numbers rather than sticking on the parameterized ones? Why do you do so for all eras but the eras_2022_2 one?
There was a problem hiding this comment.
It was a stupid copy paste. Thanks for spotting it. Just reverted.
|
please test with #46559 |
|
+code-checks Logs: https://cmssdt.cern.ch/SDT/code-checks/cms-sw-PR-46445/42447
|
|
Pull request #46445 was updated. @AdrianoDee, @Moanwar, @atpathak, @consuegs, @DickyChant, @miquork, @perrotta, @srimanob, @subirsarkar can you please check and sign again. |
|
please abort |
|
please test with #46559 |
|
+1 Size: This PR adds an extra 16KB to repository Comparison SummarySummary:
|
|
+pdmv |
|
+alca |
|
+Upgrade |
|
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. @mandrenguyen, @sextonkennedy, @rappoccio, @antoniovilela (and backports should be raised in the release meeting by the corresponding L2) |
|
+1 |
PR description:
This PR proposes the addition of 2025 wfs to the matrix. For the moment they are a copy of what we run currently for 2024 with the idea that once 2025 parameters (conditions, HLT) come we will simply need to fill the dummy ones (keeping them in sync one with the other). Here I add also:
@relval2025pointing toGRun;auto:phase1_2025_*symbolic GTs;The era is
Era_2025that for now is simply a copy ofEra_2024. A TTbar 2025 noPU wf is added to the limited matrix to be tested with PRs.The rationale is that I would like to avoid (with some anticipation) that we fall again in a situation such as #44028 or #41410.
Additional changes:
auto:run3_data_relval);ZeroBias2024 wfs to have the RECO datatier produced for them (as it was in 2023 and 2022) and to run the proper DQM sequence.relval2025to keepGRuntested;relval_2024moved toFake2;relval_2023(Fake2);PR validation:
New wfs run.