Skip to content

Releases: JeffersonLab/gluex_MCwrapper

Beginner release (with fixed backwards compatibility)

14 Oct 09:03
a252c94
Compare
Choose a tag to compare

Beginner release

26 Sep 12:52
dcae152
Compare
Choose a tag to compare

With this release MCWrapper is fully Alma9 compatible.

What's Changed

New Contributors

Full Changelog: v2.9.0...v2.10.0

Whodini Release

07 Jun 14:08
7fd6e1b
Compare
Choose a tag to compare

What's Changed

Full Changelog: v2.8.1...v2.9.0

Sir Mix-a-Lot Release (bugfixes)

30 Apr 13:23
7722368
Compare
Choose a tag to compare

What's Changed

New Contributors

Full Changelog: v2.8.0...v2.8.1

Sir Mix-a-Lot Release

03 Apr 15:24
a915453
Compare
Choose a tag to compare

This fully incorporates more robust bundling and xrootd writing as well as a few generator enhancements

What's Changed

New Contributors

Full Changelog: v2.7.0...v2.8.0

Bun B release

03 Apr 12:10
9ec51ca
Compare
Choose a tag to compare

The main thrust of this update is the inclusion of bundling for OSG jobs. Also updates to bggen_jpsi and gen_ee and more customization to how you run MCwrapper

What's Changed

New Contributors

Full Changelog: v2.6.2...v2.7.0

himalayan salt-N-pink peppercorn

31 May 15:37
160440a
Compare
Choose a tag to compare

mainly some hotfixes for pythia
and workarounds for OSG running

What's Changed

Full Changelog: v2.6.1...v2.6.2

Salt-N-Pepa Release

13 Apr 13:57
Compare
Choose a tag to compare

What's Changed

Full Changelog: v2.5.2...v2.6.1

Afrika Bambaataa Release

12 Nov 15:54
9ed8a41
Compare
Choose a tag to compare

This release is FULL of changes, some of them should go unnoticed by users and others were specifically requested...besides the smattering of bug fixes for specific use cases we have:

The MCwrapper-bot FE has been split from the gluex_MCwrapper repo. This should make maintenance of both easier.

Big changes to the submission and recording of new jobs. Now jobs that only differ by file_number are submitted all at once in a single condor_submit. All checks for duplicated batch_job_id have been removed (have never seen this ever happen again and it was needlessly slowing things down). DB inserts are also done in one transaction per "bundle". This means that high volume MC production, especially those of a single run number get submitted in seconds as opposed to hours (pending number of jobs).

scosg20 production: scosg20 is quite a bit different from its predecessor scosg16. From file system mountings to network topology running on scosg20 took quite a few tweaks to MCWrapper. These tweaks should enable future moves to be far less painful.

geantBEAM is now runnable via MCwrapper (thanks @s6pepaul)

gen_vec_ps generator now added to MCwrapper (thanks @jrstevenjlab)

and probably the biggest addition, evtgen. That's right! decay_evtgen is now fully supported in MCwrapper and should be available via the submit sim page!

What's Changed

New Contributors

Full Changelog: v2.5.2...v2.6.0

Too $hort Release

12 Apr 12:20
1c53489
Compare
Choose a tag to compare

besides some touchups (e.g. better error handling, a few new specific options) there are 2 main changes in this release:

  1. When running overt a range of run numbers MCwrapper now uses uses the luminosity in the energy range specified by the user to normalize the number of events to produce for each run number which satisfies the appropriate RCDB_QUERY. Thanks @jrstevenjlab !

  2. Future osg submit nodes will not have lustre mounted. It was discovered that CLAS was abusing lustre via the submit node thus to prevent future probles lustre mounts were removed from the current submit node. As such quite a few changes were required in MCwrapper bot to remove the need for lustre mounts. Currently, the output sizes are no longer updated while a better solution is found. This does not impact running at all