Data management
Split book-keeping files
The .dst
description files downloaded directly from Dirac
contains a list with
a large number of files (700 files or more). For example, this file contains 1690 lines:
/lhcb/LHCb/Collision12/SEMILEPTONIC.DST/00067251/0000/00067251_00001536_1.Semileptonic.dst
/lhcb/LHCb/Collision12/SEMILEPTONIC.DST/00067251/0000/00067251_00001039_1.Semileptonic.dst
/lhcb/LHCb/Collision12/SEMILEPTONIC.DST/00067251/0000/00067251_00000344_1.Semileptonic.dst
/lhcb/LHCb/Collision12/SEMILEPTONIC.DST/00067251/0000/00067251_00000301_1.Semileptonic.dst
/lhcb/LHCb/Collision12/SEMILEPTONIC.DST/00067251/0000/00067251_00000850_1.Semileptonic.dst
/lhcb/LHCb/Collision12/SEMILEPTONIC.DST/00067251/0000/00067251_00000988_1.Semileptonic.dst
/lhcb/LHCb/Collision12/SEMILEPTONIC.DST/00067251/0000/00067251_00001423_1.Semileptonic.dst
/lhcb/LHCb/Collision12/SEMILEPTONIC.DST/00067251/0000/00067251_00000841_1.Semileptonic.dst
...
/lhcb/LHCb/Collision12/SEMILEPTONIC.DST/00067251/0000/00067251_00001096_1.Semileptonic.dst
It is impractical to download every single one of them, as we don't have enough
storage space in lxplus
, even with EOS
.
To split, we use split
, a linux tool that should have been installed by default:
split --lines=50 <filename.txt>
<filename.txt>
will be sliced for every 50 lines, and the
output filenames will be x??
.
Download data files with lxplus
To download all dst files listed in a bookkeeping file:
lhcb-proxy-init
lb-dirac dirac-dms-get-file -D <target_dir> --File <file_with_list_of_lfns>
If you know a LFN
, instead of a file that contains a list of LFN
s, you can:
lb-dirac dirac-dms-get-file -D <target_dir> -l <lfn>
Query MC DDDB and CONDDB tags
Warning
MagUp and MagDown have different tags. Typically you can replace mu100
with md100
.
Find the production number for a given .dst
LFN, say:
/lhcb/LHCb/Collision12/SEMILEPTONIC.DST/00067251/0000/00067251_00000841_1.Semileptonic.dst
It's production number is 67251
.
On lxplus, do the following:
lhcb-proxy-init
lb-dirac dirac-bookkeeping-production-information 67251
The output contains both DDDB and CONDDB tags.