Workflow 4204, Stage 1

Priority	50
Processors	1
Wall seconds	80000
RSS bytes	4194304000 (4000 MiB)
Max distance for inputs	30.0
Enabled input RSEs	CERN_PDUNE_EOS, DUNE_CA_SFU, DUNE_CERN_EOS, DUNE_ES_PIC, DUNE_FR_CCIN2P3_DISK, DUNE_IN_TIFR, DUNE_IT_INFN_CNAF, DUNE_UK_GLASGOW, DUNE_UK_LANCASTER_CEPH, DUNE_UK_MANCHESTER_CEPH, DUNE_US_BNL_SDCC, DUNE_US_FNAL_DISK_STAGE, FNAL_DCACHE, FNAL_DCACHE_STAGING, FNAL_DCACHE_TEST, MONTECARLO, NIKHEF, PRAGUE, QMUL, RAL-PP, RAL_ECHO, SURFSARA, T3_US_NERSC
Enabled output RSEs	CERN_PDUNE_EOS, DUNE_CA_SFU, DUNE_CERN_EOS, DUNE_ES_PIC, DUNE_FR_CCIN2P3_DISK, DUNE_IN_TIFR, DUNE_IT_INFN_CNAF, DUNE_UK_GLASGOW, DUNE_UK_LANCASTER_CEPH, DUNE_UK_MANCHESTER_CEPH, DUNE_US_BNL_SDCC, DUNE_US_FNAL_DISK_STAGE, FNAL_DCACHE, FNAL_DCACHE_STAGING, FNAL_DCACHE_TEST, NIKHEF, PRAGUE, QMUL, RAL-PP, RAL_ECHO, SURFSARA, T3_US_NERSC
Enabled sites	BR_CBPF, CA_SFU, CA_Victoria, CERN, CH_UNIBE-LHEP, CZ_FZU, ES_CIEMAT, ES_PIC, FR_CCIN2P3, IN_TIFR, IT_CNAF, NL_NIKHEF, NL_SURFsara, UK_Bristol, UK_Brunel, UK_Durham, UK_Edinburgh, UK_Imperial, UK_Lancaster, UK_Liverpool, UK_Manchester, UK_Oxford, UK_QMUL, UK_RAL-PPD, UK_RAL-Tier1, UK_Sheffield, US_BNL, US_Caltech, US_Colorado, US_FNAL-FermiGrid, US_FNAL-T1, US_Michigan, US_MIT, US_Nebraska, US_NotreDame, US_PuertoRico, US_SU-ITS, US_Swan, US_UChicago, US_UConn-HPC, US_UCSD, US_Wisconsin
Scope	usertests
Events for this stage

Output patterns

	Destination	Pattern	Lifetime	For next stage
1	https://fndcadoor.fnal.gov:2880/dune/scratch/users/lavaut/04204/1	singleHit.*

Environment variables

Name	Value
INPUT_TAR_DIR_LOCAL	/cvmfs/fifeuser1.opensciencegrid.org/sw/dune/473df62631a815dba35936260654a011b3a47478

File states

Total files	Finding	Unallocated	Allocated	Outputting	Processed	Not found	Failed
1000	0	1	640	0	0	0	359

Job states

Total	Submitted	Started	Processing	Outputting	Finished	Notused	Aborted	Stalled	Jobscript error	Outputting failed	None processed
3028	2	0	64	0	2362	0	0	70	161	0	369

RSEs used

Name	Inputs	Outputs
DUNE_US_FNAL_DISK_STAGE	4214	0

Stats of processed input files as CSV or JSON, and of uploaded output files as CSV or JSON (up to 10000 files included)

File reset events, by site

Site	Allocated	Outputting
US_NotreDame	224	0
US_FNAL-FermiGrid	160	0
US_Wisconsin	100	0
CA_SFU	50	0
US_Colorado	40	0

Jobscript

#!/bin/bash
:<<'EOF'

To use this jobscript to process 10 files from the dc4-vd-coldbox-bottom
data and put the output in the usertests namespace (MetaCat) and 
scope (Rucio), and in the usertests:output-test-01 dataset in MetaCat and
Rucio, use this command to create the workflow:

justin simple-workflow \
--mql \
"files from dune:all where core.run_type='dc4-vd-coldbox-bottom' and dune.campaign='dc4' limit 10" \
--jobscript dc4-vd-coldbox-bottom.jobscript --max-distance 30 --rss-mb 4000 \
--scope usertests --output-pattern '*_reco_data_*.root:output-test-01' 

The following optional environment variables can be set when creating the
workflow/stage: FCL_FILE, NUM_EVENTS, DUNE_VERSION, DUNE_QUALIFIER 

EOF

# fcl file and DUNE software version/qualifier to be used
FCL_FILE=MyrunSingleHit_PDHD.fcl
DUNE_VERSION=${DUNE_VERSION:-v10_00_02d00}
DUNE_QUALIFIER=${DUNE_QUALIFIER:-e26:prof}

# number of events to process from the input file
if [ "$NUM_EVENTS" != "" ] ; then
 events_option="-n $NUM_EVENTS"
fi


if [ -z ${JUSTIN_PROCESSORS} ]; then
  JUSTIN_PROCESSORS=1
fi

echo "Justin processors: ${JUSTIN_PROCESSORS}"

export TF_NUM_THREADS=${JUSTIN_PROCESSORS}
export OPENBLAS_NUM_THREADS=${JUSTIN_PROCESSORS}
export JULIA_NUM_THREADS=${JUSTIN_PROCESSORS}
export MKL_NUM_THREADS=${JUSTIN_PROCESSORS}
export NUMEXPR_NUM_THREADS=${JUSTIN_PROCESSORS}
export OMP_NUM_THREADS=${JUSTIN_PROCESSORS}


# First get an unprocessed file from this stage
for nf in {1..10}
do
   DID_PFN_RSE=`$JUSTIN_PATH/justin-get-file`
   ##Check that any file was returned
   if [ "${DID_PFN_RSE}" == "" ] ; then
     echo "Could not get file"
     exit 0
     #continue
   fi
   FILE=`echo ${DID_PFN_RSE} | cut -f2 -d' '`
   DID=`echo ${DID_PFN_RSE} | cut -f1 -d' '`
   echo ${DID}  >> did.list
   echo ${FILE} >> file.list

done

#did_pfn_rse=`$JUSTIN_PATH/justin-get-file`

#if [ "$did_pfn_rse" = "" ] ; then
 # echo "Nothing to process - exit jobscript"
 # exit 0
#fi

# Keep a record of all input DIDs, for pdjson2meta file -> DID mapping
#echo "$did_pfn_rse" | cut -f1 -d' ' >>all-input-dids.txt

# pfn is also needed when creating justin-processed-pfns.txt
#pfn=`echo $did_pfn_rse | cut -f2 -d' '`
#echo "Input PFN = $pfn"

# Setup DUNE environment
source /cvmfs/dune.opensciencegrid.org/products/dune/setup_dune.sh

export PRODUCTS=$INPUT_TAR_DIR_LOCAL:${PRODUCTS}
echo "PRODUCTS: ${PRODUCTS}"
echo "DUNESW_DIR: ${DUNESW_DIR}"


# Properly setup custom code with INPUT_TAR_DIR_LOCAL

export DUNESW_DIR=${INPUT_TAR_DIR_LOCAL}/dunesw/${DUNE_VERSION}
export DUNESW_FQ_DIR=${DUNESW_DIR}/slf7.x86_64.e26.prof
export DUNESW_LIB=${DUNESW_FQ_DIR}/lib

export PROTODUNEANA_DIR=${INPUT_TAR_DIR_LOCAL}/protoduneana/${DUNE_VERSION}
export PROTODUNEANA_LIB=${PROTODUNEANA_DIR}/include
export PROTODUNEANA_FQ_DIR=${PROTODUNEANA_DIR}/slf7.x86_64.e26.prof
export PROTODUNEANA_LIB=${PROTODUNEANA_FQ_DIR}/lib

export DUNEPROTOTYPES_DIR=${INPUT_TAR_DIR_LOCAL}/duneprototypes/${DUNE_VERSION}
export DUNEPROTOTYPES_INC=${DUNEPROTOTYPES_DIR}/include
export DUNEPROTOTYPES_FQ_DIR=${DUNEPROTOTYPES_DIR}/slf7.x86_64.e26.prof
export DUNEPROTOTYPES_LIB=${DUNEPROTOTYPES_FQ_DIR}/lib

export FHICL_FILE_PATH=${DUNESW_DIR}/fcl:${FHICL_FILE_PATH}
export FHICL_FILE_PATH=${DUNEPROTOTYPES_DIR}/fcl:${FHICL_FILE_PATH}
export FHICL_FILE_PATH=${PROTODUNEANA_DIR}/fcl:${FHICL_FILE_PATH}
export FHICL_FILE_PATH=${INPUT_TAR_DIR_LOCAL}/myfcl:${FHICL_FILE_PATH}

echo "DUNESW_DIR: ${DUNESW_DIR}"
echo "PROTODUNEANA_DIR: ${PROTODUNEANA_DIR}"
echo "DUNEPROTOTYPES_DIR: ${DUNEPROTOTYPES_DIR}"

source "${INPUT_TAR_DIR_LOCAL}/localProducts_larsoft_v10_00_02d00_prof_e26/setup2"
setup dunesw "$DUNE_VERSION" -q "$DUNE_QUALIFIER"

echo " VERSION ACTIVE =  VERSION ACTIVE =  VERSION ACTIVE = "
ups active | grep "dunesw\|duneprototypes"
mrbslp

echo "now run lar on these files"

cat file.list
cat did.list

# Construct outFile from input $pfn 
now=$(date -u +"%Y-%m-%dT_%H%M%SZ")
#Ffname=`echo $pfn | awk -F/ '{print $NF}'`
#fname=`echo $Ffname | awk -F. '{print $1}'`
outFile=singleHit_output_${now}.root

campaign="justIN.r${JUSTIN_WORKFLOW_ID}s${JUSTIN_STAGE_ID}"

# Do the scary preload stuff in a subshell!
lar -c $events_option ${INPUT_TAR_DIR_LOCAL}/myfcl/$FCL_FILE  -T $outFile -S file.list  > log_singleHit_${now}.log 2>&1

# Subshell exits with exit code of last command
larExit=$?
echo "lar exit code $larExit"

# echo "$FILE" > justin-processed-pfns.txt
cat file.list > justin-processed-pfns.txt

echo "processed files"
cat justin-processed-pfns.txt

ls -lRS

# Create compressed tar file with all log files 
#tar zcf `echo "$JUSTIN_JOBSUB_ID.logs.tgz" | sed 's/@/_/g'` *.log
#exit $jobscriptExit
exit $larExit