Adding PU in 2019 scenario #26463
Adding PU in 2019 scenario #26463
Conversation
|
The code-checks are being triggered in jenkins. |
|
Adding @luciasilvestris as a watcher |
|
The code-checks are being triggered in jenkins. |
|
The code-checks are being triggered in jenkins. |
|
+code-checks Logs: https://cmssdt.cern.ch/SDT/code-checks/cms-sw-PR-26463/9289
|
|
A new Pull Request was created by @boudoul (boudoul) for master. It involves the following packages: Configuration/PyReleaseValidation @pgunnell, @zhenhu, @prebello, @cmsbuild, @kpedro88 can you please review it and eventually sign? Thanks. cms-bot commands are listed here |
|
The code-checks are being triggered in jenkins. |
|
+code-checks Logs: https://cmssdt.cern.ch/SDT/code-checks/cms-sw-PR-26463/9291
|
| @@ -14,9 +14,9 @@ | |||
| '2018Design', | |||
| '2018DesignPU', | |||
| '2019', | |||
| # '2019PU', | |||
| '2019PU', | |||
There was a problem hiding this comment.
@boudoul I guess this line with the corresponding changed order is the responsible of the visible difference in the standard matrix:
< 12024.0 TTbar_13TeV_TuneCUETP8M1_2019Design_GenSimFull+DigiFull_2019Design+RecoFull_2019Design+HARVESTFull_2019Design
---
> 12024.0 TTbar_13TeV_TuneCUETP8M1_2019PU_GenSimFull+DigiFullPU_2019PU+RecoFullPU_2019PU+HARVESTFullPU_2019PU+NanoFull_2019PU
Not a problem in itself, just to be aware of it
There was a problem hiding this comment.
Hi @boudoul @fabiocos this number is related to TTbar Design in IB as you can see at
https://github.com/cms-sw/cmssw/blob/master/Configuration/PyReleaseValidation/python/relval_2017.py
When you change https://github.com/cms-sw/cmssw/blob/master/Configuration/PyReleaseValidation/python/upgradeWorkflowComponents.py, moving the comment of line 17, you will change the order of the wfs numbers.
We had a similar problem in the past (when we started 2017 relvals) and to keep all under control @kpedro88 has changed something at
numWFConflict = [[11800,12000],[12200,12400],[20800,21200],[21600,22400],[25000,26000],[50000,51000]]
Maybe increasing the range [11800,12000] would work.
There was a problem hiding this comment.
ok yes I will try to keep the WF numbers unchanged in a next commit later today
There was a problem hiding this comment.
If you are activating the 2019 PU workflows, you should remove the ranges [11800,12000],[12200,12400] from the conflict list. These were reserved for that purpose.
There was a problem hiding this comment.
ah you're right- I missed your comment I just re-oreder to preserve but actually what you wrote makes more sense - Let me fix in a next commit
|
Comparison is ready @slava77 comparisons for the following workflows were not done due to missing matrix map:
Comparison Summary:
|
|
+upgrade |
|
+1 |
cmsbuild
added
fully-signed
pdmv-approved
and removed
pdmv-pending
pending-signatures
labels
|
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. @davidlange6, @slava77, @smuzaffar, @fabiocos (and backports should be raised in the release meeting by the corresponding L2) |
|
+1 |
|
@boudoul @prebello @zhenhu @pgunnell a number of 2019 workflows have failed in the latest IB CMSSW_10_6_X_2019-04-26-1100. The failure reports a DAS error, but I think the underlying cause is that a default dataset is now provided for 2019, so it wants to reuse GEN-SIM RelVals, but the needed samples are not hosted at T2_CH_CERN. Can we transfer them so the tests will run? |
|
@kpedro88 as cross checked with @smuzaffar in principle the bot should take care of caching the needed files at CERN using the utilities in https://github.com/cms-sw/cms-bot/tree/master/das-utils but there were additional problems on top. For the remaining issue (11626.0) the sample apparently still needs to be produced: is giving an empty result. |
@kpedro88 the samples are in FNAL disk AFAIK. Is it not possible to read them from there? |
|
@prebello the IB tests require reading files from CERN. Please see #26463 (comment) for more details. |
hi @zhenhu , as you have submitted other 2019 relvals, and machinery is set up in your local area, please take care of /RelValQCD_Pt_600_800_13/CMSSW_10_6_0_pre3-105X_postLS2_realistic_v6-v1/GEN-SIM too. |
ok @kpedro88 , I see but the issue is related to only RelValQCD_Pt_600_800_13 that needs still to be produced, or the ready ones from my list above? |
|
According to @fabiocos it is just RelValQCD_Pt_600_800_13 that is needed, and the other samples were automatically transferred after fixing some bug in the bot. The last few IB test results (https://cmssdt.cern.ch/SDT/html/showIB.html) confirm this. |
|
great. So @zhenhu will provide the missing one. Thank you. |
|
We will produce this sample soon. |
|
The missing relval has been produced: |
|
@zhenhu thank you, I now see the files at FNAL. In principle the bot should copy and cache them, let's see next IBs... |
PR description:
In order to help the pdmV team towards a 2019 validation, I'm adding the PU workflows for 2019 , so that 2019 relvals with or without PU can be submitted with a fair comparison with 2018 (to start the process ...)
@franzoni @srimanob FYI
PR validation:
runTheMatrix.py --what upgrade -l 12000 -ne seems to give the correct answer (attention , usual process : it does miss a 2019 MinBias sample to run (and also to pick the correct PU scenario , without the 2019 monbais , it picks PU35, once the Minbias will be added, it will automatically pick up PU50, like 2018)