Thanks to visit codestin.com
Credit goes to github.com

Skip to content

Conversation

@francescobrivio
Copy link
Contributor

@francescobrivio francescobrivio commented Nov 9, 2022

Replay Request

Requestor
Francesco Brivio for AlCaDB/HI

Describe the configuration

  • Release: CMSSW_12_5_2_patch1 (still to be built)
  • Run: 361697
  • GTs:
    • expressGlobalTag: 124X_dataRun3_Express_HIreplay2022_v1
    • promptrecoGlobalTag: 124X_dataRun3_Prompt_HIreplay2022_v1
    • alcap0GlobalTag: 124X_dataRun3_Prompt_HIreplay2022_v1
  • Additional changes:
    • AlCaRecoMatrix for 2022 HI as documented in AlCaRecoMatrix#HeavyIon_Replay_on_2022_dry_run
    • Added new Run3 HI scenarios:
      • ppEra_Run3_pp_on_PbPb
      • hcalnzsEra_Run3_pp_on_PbPb
      • trackingOnlyEra_Run3_pp_on_PbPb
      • ppEra_Run3_pp_on_PbPb_approxSiStripClusters
    • The HI Datasets should be more or less the ones that will be produced in the actual HI data taking next week:
      • HIExpress
      • HIExpressAlignment
      • HIHLTMonitor
      • HIHcalNZS
      • HIHLTPhysics
      • HITestRaw[0-23]
      • HITestRawPrime[0-19]

Purpose of the test
This replay is to test the HI configuration for 2022 including the new scenarios, the GTs, the HI alcarecomatrix and a first attempt at the RawPrime repacking and prompt-reprocessing.
The input run is 361697 which is the longest run with the most recent menu (/cdaq/special/2022/HIonTest/HLT/V8), unfortunately some of the streams (HIExpressAlignment and HIHcalNZS) are empty as can be seen in this OMS link, but there is nothing that we can do at the moment to test it more.

T0 Operations cmsTalk thread
https://cms-talk.web.cern.ch/t/replay-for-2022-hi-dry-run-data/17329

FYI @mandrenguyen (for HI) @mpresill (as ORM)

@francescobrivio
Copy link
Contributor Author

test syntax please

@francescobrivio
Copy link
Contributor Author

test syntax please

@francescobrivio
Copy link
Contributor Author

test syntax please

@germanfgv
Copy link
Contributor

@germanfgv
Copy link
Contributor

We have 5 paused jobs, al of them with PD HIEmptyBX. The error message is:

cmsRun1
Fatal Exception (Exit Code: 8001)
An exception of category 'NoRecord' occurred while
   [0] Processing  Event run: 361697 lumi: 14 event: 22054465 stream: 2
   [1] Running path 'write_MINIAOD_step'
   [2] Prefetching for module PoolOutputModule/'write_MINIAOD'
   [3] Calling method for module HiEvtPlaneFlatProducer/'hiEvtPlaneFlat'
Exception Message:
No "HeavyIonRPRcd" record found in the EventSetup.

 Please add an ESSource or ESProducer that delivers such a record.

This looks like a GT error to me, is that right @francescobrivio ?

You can find the tarball for one of the jobs here:

/eos/cms/store/logs/prod/recent/PromptReco/PromptReco_Run361697_HIEmptyBX_Tier0_REPLAY_2022_ID221110152242_v10152242/Reco/vocms001.cern.ch-413-3-log.tar.gz

@malbouis
Copy link
Contributor

Hi @germanfgv , thanks for the tarball!

I have created a new GT with the missing record: 124X_dataRun3_Prompt_HIreplay2022_v2.

With this new GT, I could run on the tarball you provided without crashes (for as many events as I could :-) ), while I could reproduce the crash right away with the previous GT.

Would you please try this new GT and let us know of problems?

Thanks!

@germanfgv
Copy link
Contributor

Thanks @malbouis. I'll rerun these jobs with the new GT, and I'll also add another run to be processed from scratch with it.

I'll let you know if we find any problems

@germanfgv
Copy link
Contributor

I resumed the jobs with the new GTs and they were successful. I added run 361696 to the replay to process from scratch and it has been successful so far.

@francescobrivio
Copy link
Contributor Author

I resumed the jobs with the new GTs and they were successful. I added run 361696 to the replay to process from scratch and it has been successful so far.

Very good! Thanks German for confirming!
So we (AlCa) will have to update the production Prompt GT to include this fix from Helena!

@germanfgv
Copy link
Contributor

The replay finished without issues. We processed runs 361696, 361696, and 361779.

@francescobrivio
Copy link
Contributor Author

test syntax please

@francescobrivio
Copy link
Contributor Author

At the request of DQM (@rvenditti) I have added a few more dqm_sequences to the HITestRaw[0-23] datasets so the DPGs can check if monitoring their subsystems is possible even in HI conditions and prepare for the actual HI data-taking next year.

Changes:

dqm_sequences=["@commonSiStripZeroBias"],

-->

dqm_sequences=["@commonSiStripZeroBias", "@ecal", "@L1TMon", "@hcal", "@muon", "@jetmet"],

@germanfgv could you please start again the replay with this configuration (I think 1 run is enough) so we can check nothing is crashing with the new dqm_sequences added.

@rvenditti
Copy link

Hi @germanfgv did you have the possibility to launch the reply with updated DQM configuration?

@germanfgv
Copy link
Contributor

@rvenditti yes. I launched it yesterday evening. It should have finished by now, but I see there are still several idling jobs. I'm going to check what the problem is.

@germanfgv
Copy link
Contributor

The jobs were idle because, they were issued with low priority. Now they are running and should finish in the next few hours

@germanfgv
Copy link
Contributor

germanfgv commented Nov 16, 2022

The replay finished succesfully. I'll add the DQM sequences to #4780

@germanfgv germanfgv closed this Nov 16, 2022
@francescobrivio francescobrivio deleted the HI_replay2022 branch March 14, 2024 07:54
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

None yet

Projects

None yet

Development

Successfully merging this pull request may close these issues.

5 participants