GitLab

Frank Siegert authored Feb 07, 2020 and Spyros Argyropoulos committed Feb 07, 2020

In particular the "find ." JO discovery was super slow in large trees on networked file systems.

Zach Marshall authored Feb 06, 2020 and Christian Gutschow committed Feb 06, 2020

Pointed out on the egroup, the first 'launch' was missing from the
reweighting card.

Zach Marshall authored Feb 04, 2020 and Christian Gutschow committed Feb 04, 2020

The event multiplier is just a hair too low in these jobs, so single
event failures cause the job to crash. See e.g.:

https://bigpanda.cern.ch/job?pandaid=4634345509

https://bigpanda.cern.ch/job?pandaid=4634364264

Frank Siegert authored Feb 04, 2020 and Christian Gutschow committed Feb 04, 2020

[skip modfiles]

Zach Marshall authored Jan 30, 2020 and Spyros Argyropoulos committed Jan 30, 2020

This adds events to the LHE runs for one MGC test job to ensure that we
make it to the end of the Pythia8 job.

The other meatier piece of the MR adds the possibility to logParser.py
to run stand-alone on a local log file, using the job options and input
files that live on cvmfs. This should allow people who have not locally
checked out mcjoboptions but do want to run the log parser to check the
output.

Frank Siegert authored Jan 22, 2020 and Spyros Argyropoulos committed Jan 22, 2020

The minimum number of events to run in central production should not be 10 but 1.

[skip all]