cmssw
cmssw copied to clipboard
cms-sw
L1 Unprefirable event information is missing in MINIAOD for Run 3
Hello,
The information related to L1 unprefirable events is not available in Run 3 MINIAOD.
Indeed the relevant producer (simGtExtUnprefireable) is only enabled for run2_miniAOD_UL eras:
https://github.com/cms-sw/cmssw/blob/CMSSW_12_4_X/PhysicsTools/PatAlgos/python/slimming/miniAOD_tools.py#L622-L623
Obviously this impacts also current NANOAOD for which the current unprefirable event bit is always false: https://github.com/cms-sw/cmssw/blob/master/PhysicsTools/NanoAOD/python/triggerObjects_cff.py#L281-L286
Having this information in Prompt Reco for the next data to be taken this year would be extremely useful.
I believe the fix is rather trivial (I guess we want this information available by default for data now, not behind an era modifier) but I'm leaving for two weeks tomorrow so won't be able to follow this myself. I would really appreciate if someone could follow this up.
Thanks a lot !
Laurent
FYI @cecilecaillol @epalencia @mariadalfonso @gouskos @swertz @vlimant @bundocka @eyigitba @jpata @clacaputo
A new Issue was created by @lathomas .
@Dr15Jones, @perrotta, @dpiparo, @rappoccio, @makortel, @smuzaffar, @qliphy can you please review it and eventually sign/assign? Thanks.
cms-bot commands are listed here
New categories assigned: xpog,reconstruction,l1
@jpata,@mariadalfonso,@epalencia,@gouskos,@swertz,@clacaputo,@rekovic,@cecilecaillol,@vlimant you have been requested to review this Pull request/Issue and eventually sign? Thanks
@jpata @clacaputo how do we handle the Run3 era in mini ? are many things that have been fixed for the run2_miniAOD_UL, and are not pipelined for Run3 ? or this is an isolated case ?
It looks like this was pointed out in the review of the PR that introduced this feature: https://github.com/cms-sw/cmssw/pull/31985#discussion_r526325765
From reading the thread, it seems that the agreement was that it will go in as-is for the UL production, and to follow up with a fix: https://github.com/cms-sw/cmssw/issues/32304.
However, this never happened (I closed the issue by mistake after 2 years of no activity and an L1 signature). @rekovic are you able to take a look?
how do we handle the Run3 era in mini? are many things that have been fixed for the run2_miniAOD_UL, and are not pipelined for Run3 ? or this is an isolated case?
In general, so far in reco we tried to avoid switching things on ONLY for a particular era unless it's a special case. Rather, our approach has been that new stuff should be enabled by default, and switched off for old eras if necessary. It's probably still worth to do a git grep run2_miniAOD_UL and check that it's used as intended.
+reconstruction This is solved by #39169 (and the 14X backport #39253) and can be closed
