new GT

[jpata]

Spring16_25nsV10, https://hypernews.cern.ch/HyperNews/CMS/get/jes/639.html

[degrutto]

Hi,

how do we make sure when we run on data we use the correct GT? Is there some code already from some repository we can steal or for example we work at crab submission level?

[jpata]

Ciao, sorry I'm not precisely sure what your question is, but the GT is configured in the usual way in vhbb.py. Also someone needs to dump the JEC files into the data dir.

[degrutto]

Hi @jpata I am referring to the fact that I read IOV BCD: [1,276811] corresponds to Spring16_25nsV10BCD_DATA.tar.gz IOV E: [276831,277420] corresponds to Spring16_25nsV10E_DATA.tar.gz IOV F: [277772,278801] corresponds to Spring16_25nsV10F_DATA.tar.gz IOV G: [278802,Infinity] corresponds to Spring16_25nsV10p2_DATA.tar.gz

I'm not sure, actually I don't think that in heppy we have an IOV (i.e. a way to associate the run number to the payload) as we have in the full framework

[jpata]

ah, I see. Good question indeed... For crab submission we can pass the GT as an extra argument, I think.

[jpata]

to confirm, likely we'll need run-dependent GT/JEC configuration for data.

[arizzi]

I'll take care of this in the crab script, can someone meanwhile dump all payloads in the relevant vhbb heppy directories and make a PR?

[jpata]

OK, doing it!

[jpata]

https://github.com/vhbb/cmssw/pull/593

[jpata]

looks like the newest GT for Data is Spring16_23Sep2016V2, the one for MC is identical to 25nsV10: https://twiki.cern.ch/twiki/bin/view/CMS/JECDataMC

Waiting for #592 to be merged in order to test coherently in CMSSW 8_0_21.

[arizzi]

but we were using Spring16_25nsV6_MC... so we still need to update, right?

[jpata]

yes, we skipped one.

[arizzi]

crab mechanism implemented too #599 (to be tested after some issue with running on data are solved). We can close this issue