Closed katilp closed 6 years ago
Test as in #1877
Start 17:49:
[cms-opendata@localhost ~]$ df
Filesystem 1K-blocks Used Available Use% Mounted on
aufs 40182924 901688 37216968 3% /
tmpfs 1022272 0 1022272 0% /dev/shm
cvmfs2 20480000 539998 19940002 3% /cvmfs/cvmfs-config.cern.ch
cvmfs2 20480000 539998 19940002 3% /cvmfs/cms.cern.ch
cvmfs2 20480000 539998 19940002 3% /cvmfs/cms-ib.cern.ch
cvmfs2 20480000 539998 19940002 3% /cvmfs/cms-opendata-conddb.cern.ch
end 18:36:
[cms-opendata@localhost ~]$ df
Filesystem 1K-blocks Used Available Use% Mounted on
aufs 40182924 11269640 26849016 30% /
tmpfs 1022272 0 1022272 0% /dev/shm
cvmfs2 20480000 10869733 9610268 54% /cvmfs/cvmfs-config.cern.ch
cvmfs2 20480000 10869733 9610268 54% /cvmfs/cms.cern.ch
cvmfs2 20480000 10869733 9610268 54% /cvmfs/cms-ib.cern.ch
cvmfs2 20480000 10869733 9610268 54% /cvmfs/cms-opendata-conddb.cern.ch
Testing the RAW reco with this VM, see http://opendata-dev.web.cern.ch/record/463 or direct in https://github.com/cms-opendata-validation/2011-minimumbias-raw-recotest/blob/master/README.md Interested to see whether this solves the problem of killed jobs for rereco UPDATE: does not help, killed after close to 20000 evts
@tiborsimko Can we add another version of 2011/2012 VM to the 2011 record? I would prefer to have it there for the sake of simplicity for users (instructions why to choose the new VM can reside in the VM record and not be spread all over in different instructions). It will have its own contextualization file record, though,
Tested by @zenaiev, no problem with the new VM
Closing as superseded by #2044
We can update the 2011 VM with a newer version with an increased cache volume for #1877 The VM is available from https://cernbox.cern.ch/index.php/s/SkLCf42lo6DqdAE with the following changes with respect to the old one (from Jakob)
Testing