def downloadDataSets(InputDatasets, Destination, RSE="", use_singularity=False):
### Apparently rucio does no longer work in combination with AthAnalysis. So let's
### execute it from a singulartity container
Cmds = []
image_to_choose = setupBatchSubmitArgParser().get_default("SingularityImage")
home_dir = setupBatchSubmitArgParser().get_default("BaseFolder") + "/TMP/.singularity/"
CreateDirectory(Destination, False)
if use_singularity:
CreateDirectory(home_dir, False)
to_clean = []
for DS in InputDatasets:
if not use_singularity:
Cmds += ["rucio download %s --ndownloader 32 %s --dir %s" % (DS, "" if len(RSE) == 0 else "--rse %s" % (RSE), Destination)]
else:
singularity_dir = home_dir + "/" + id_generator(21)
to_clean += [singularity_dir]
singularity_script = WriteList([
"#!/bin/bash",
"export ATLAS_LOCAL_ROOT_BASE=/cvmfs/atlas.cern.ch/repo/ATLASLocalRootBase",
"source ${ATLAS_LOCAL_ROOT_BASE}/user/atlasLocalSetup.sh",
"lsetup rucio",
"echo 'rucio download %s --ndownloader 32 %s --dir %s'" % (DS, "" if len(RSE) == 0 else "--rse %s" % (RSE), Destination),
"rucio download %s --ndownloader 32 %s --dir %s" % (DS, "" if len(RSE) == 0 else "--rse %s" % (RSE), Destination),
], "%s/to_exec.sh" % (singularity_dir))
os.system("chmod 0777 " + singularity_script)
Cmds += [
"singularity exec --cleanenv -H %s:/alrb -B %s:/srv %s/%s %s" %
(singularity_dir, Destination, SINGULARITY_DIR, image_to_choose, singularity_script)
]
ExecuteCommands(ListOfCmds=Cmds, MaxCurrent=8)
for c in to_clean:
os.system("rm -rf %s" % (c))
def exclusiveBatchOpt():
parser = setupBatchSubmitArgParser()
parser.add_argument(
'--BaseProject',
help=
'choose project containing file lists to adjust the filepath to the DS list (default is \"XAMPPbase\")',
default='XAMPPbase')
parser.set_defaults(Merge_vmem=600)
parser.add_argument('--inputDS',
help='Input datasets',
default=[],
nargs="+")
parser.add_argument(
"--RSE",
help='RSE storage element for files located via dcache.',
default='MPPMU_LOCALGROUPDISK')
parser.add_argument(
'--RunTime',
help='Changes the RunTime of the Jobs: default 19:59:59 ',
default='19:59:59')
parser.add_argument(
'--EventsPerJob',
help='Changes the Events per Batch job. Default: 10000 ',
default=10000,
type=int)
parser.add_argument('--FilesPerMergeJob',
help='Number of files per merge',
default=8,
type=int)
parser.add_argument("--FinalSplit",
help="How many files should be left after merge",
default=1,
type=int)
parser.add_argument('--vmem',
help='Virtual memory reserved for each analysis jobs',
default=3500,
type=int)
parser.add_argument(
'--HoldJob',
default=[],
nargs="+",
help=
'Specfiy job names which should be finished before your job is starting. '
)
parser.add_argument(
"--SpareWhatsProcessedIn",
help=
"If the cluster decided to die during production, you can skip the processed files in the directories",
default=[],
nargs="+")
return parser
def getArgumentParser():
"""Get arguments from command line."""
parser = setupBatchSubmitArgParser()
parser.add_argument("--fileLists",
help="Specify the file lists to submit",
default=[],
nargs="+")
parser.add_argument("--fileListsFolder",
help="Specify a folder containing file lists",
type=str,
default="")
parser.add_argument("--remainingSplit",
help="Specify a remaining split of the files",
default=1,
type=int)
### Arguments to read a list of dataset files
parser.add_argument(
"--RucioDSList",
help="Specify a file containing a list of rucio datasets to be merged",
default="")
parser.add_argument(
"--RucioRSE",
help=
"Specify the RSE on which the datasets are stored. Required if download is not activated",
default="")
parser.add_argument("--download",
help="Download the files in advance",
default=False,
action='store_true')
parser.add_argument("--batch_size",
help="What is the final size of each file in the end",
default=-1,
type=int)
parser.add_argument("--nFilesPerJob",
help="Specify number of files per merge job",
default=10,
type=int)
parser.add_argument("--HoldJob",
help="Specify a list of jobs to hold on",
default=[])
parser.set_defaults(Merge_vmem=4000)
return parser
def setupScriptSubmitParser():
parser = setupBatchSubmitArgParser()
parser.add_argument("--ListOfCmds",
help="Text file where all commands are listed",
required=True)
parser.add_argument(
'--HoldJob',
default=[],
nargs="+",
help=
'Specfiy job names which should be finished before your job is starting. '
)
parser.add_argument('--RunTime',
help='Changes the RunTime of the analysis Jobs',
default='07:59:59')
parser.add_argument('--vmem',
help='Changes the virtual memory needed by each jobs',
type=int,
default=2000)
return parser
def setupCIparser():
parser = argparse.ArgumentParser(
prog='Update reference cutflows',
conflict_handler='resolve',
formatter_class=argparse.ArgumentDefaultsHelpFormatter)
parser.add_argument("--ciDir",
help="Directory where the CI files are located.",
default=ResolvePath("XAMPPbase/test"))
parser.add_argument("--joboptions",
help="Which job options shall be run",
default="XAMPPbase/runXAMPPbase.py")
parser.add_argument("--regions",
help="what are the regions to consider",
default=[],
nargs="+",
required=True)
parser.add_argument("--serviceAccount",
help="what is the name of the service account to use",
default="xampp")
parser.add_argument(
"--EOSpath",
help="Where are all the reference samples located on EOS",
default="root://eoshome.cern.ch//eos/user/x/xampp/ci/base/")
parser.add_argument(
"--TEMPdir",
help="Where to store all of the temporary files",
default="%s/CI_temp" %
(setupBatchSubmitArgParser().get_default("BaseFolder")))
parser.add_argument("--athenaArgParser",
help="Which athena argument parser shall be used",
default=ResolvePath("XAMPPbase/python/runAthena.py"))
parser.add_argument("--evtMax",
help="Limit the number of events to run on",
default=-1,
type=int)
parser.add_argument('--noSyst',
help='run without systematic uncertainties',
action='store_true',
default=False)
return parser
def getArguments():
USERNAME = os.getenv("USER")
parser = setupBatchSubmitArgParser()
parser.set_defaults(BaseFolder="/nfs/dust/atlas/user/{username}/MC".format(
username=USERNAME))
parser.set_defaults(maxCurrentJobs=100)
parser.set_defaults(noContainerShipping=False)
parser.set_defaults(jobOptionsDir="./..")
parser.add_argument("--modelsDir",
default="",
help="Path to directory containing additional models.")
parser.add_argument("-r", "--runNumbers", nargs="+", default=[], type=int)
parser.add_argument("-R", "--runRange", nargs=2, action="store", type=int)
parser.add_argument(
"--nJobs",
help="Number of jobs to be submitted [default: 5]",
type=int,
default=5,
)
parser.add_argument(
"--eventsPerJob",
help="Events generated per job [default: 10000]",
type=int,
default=10000,
)
parser.add_argument(
"--keepOutput",
help="Keep RunDir after task",
action="store_true",
default=False,
)
parser.add_argument("--AthGeneration",
help="Event generation release version",
default="21.6.61")
parser.add_argument(
"--noDerivationJob",
help="Disable launching derivation jobs.",
action="store_true",
)
parser.add_argument(
"--derivations",
help="Format of TRUTH derivations [default: TRUTH1]",
nargs="+",
default=["TRUTH1"],
)
parser.add_argument("--AthDerivation",
help="Derivation release version",
default="21.2.63.0")
parser.add_argument("--nCores",
help="Number of cores per node.",
type=int,
default=1)
parser.add_argument(
"--evgen_runtime",
help="Event generation job time limit [default: 8 hours]",
default="08:00:00",
)
parser.add_argument(
"--evgen_memory",
help="Event generation job memory limit [default: 4 GB]",
default=4000,
type=int,
)
parser.add_argument(
"--deriv_runtime",
help="Derivation job time limit [default: 8 hours]",
default="08:00:00",
)
parser.add_argument(
"--deriv_memory",
help="Derivation job memory limit [default: 2.5 GB]",
default=2500,
type=int,
)
# Pre includes and execs for the event generation
parser.add_argument("--evgen_postExec",
default="",
help="Reco_tf --postExec",
type=str)
parser.add_argument("--evgen_preExec",
default="",
help="Reco_tf --preExec",
type=str)
parser.add_argument("--evgen_postInclude",
default="",
help="Reco_tf --postInclude",
type=str)
parser.add_argument("--evgen_preInclude",
default="",
help="Reco_tf --preInclude",
type=str)
# Pre includes and execs for the derivation
parser.add_argument(
"--deriv_postExec",
default="",
help="Reco_tf --postExec in derivation",
type=str,
)
parser.add_argument("--deriv_preExec",
default="",
help="Reco_tf --preExec in derivation",
type=str)
parser.add_argument(
"--deriv_postInclude",
default="",
help="Reco_tf --postInclude in derivation",
type=str,
)
parser.add_argument(
"--deriv_preInclude",
default="",
help="Reco_tf --preInclude in derivation",
type=str,
)
return parser