#!/bin/bash # # This script checks whether data can be deleted # source `dirname $0`/../Sourcefile.sh printprocesslog "INFO starting $0" numchecktransfer=`/usr/sbin/lsof $0 | grep -o -c $0` if [ $numchecktransfer -gt 1 ] then printprocesslog "INFO "$0" already running -> exit. " echo "INFO "$0" already running -> exit. " finish fi logfile2=$logpath"/transfer/CheckTransfer.log" date > $logfile2 2>&1 diskusage=( `ssh fact@161.72.93.131 "df -P /daq" | grep daq ` ) # check if more than X GB are left on /daq if [ ${diskusage[3]} -lt 700000 ] then printprocesslog "DISK less than 700 GB left on /daq ("${diskusage[3]}")" echo "WARN less than 700 GB left on /daq ("${diskusage[3]}")" echo "WARN less than 700 GB left on /daq ("${diskusage[3]}")" >> $logfile2 2>&1 sendemail="yes" fi diskusage2=( `df -P /scratch | grep scratch ` ) # check if more than X GB are left on /scratch if [ ${diskusage2[3]} -lt 500000 ] then printprocesslog "DISK less than 500 GB left on /scratch ("${diskusage2[3]}")" echo "WARN less than 500 GB left on /scratch "${diskusage2[3]}")" echo "WARN less than 500 GB left on /scratch "${diskusage2[3]}")" >> $logfile2 2>&1 sendemail="yes" fi # needed for transfer to phido #source /home_nfs/isdc/fact_opr/myagent.sh # check first the disk in LP and on dl00 ssh fact@161.72.93.131 "df -h /*da*" df -h /scratch echo "" echo "" >> $logfile2 2>&1 # check next the DB to know if some transfer processes failed or crashed function check_runs_in_db() { query="SELECT "$toquery" FROM "$1" "$where #echo $query runs=( `sendquery $query` ) if [ ${#runs[@]} -gt 0 ] then #for run in ${runs[@]} #do # echo $run #done sendemail="yes" echo -e "\e[1;31m\x1b[5m ==>\e[00m "$1": "${runs[@]}"\e[1;31m\x1b[5m <==\e[00m " echo "SELECT fNight, fRunId, fStartTime, fStopTime, fReturnCode FROM "$1" "$where";" echo "UPDATE "$1" SET fStartTime=NULL, fStopTime=NULL, fAvailable=NULL, fReturnCode=NULL, fProcessingSiteKey=NULL "$where";" echo -e " ==> "$1": "${runs[@]}" <== " >> $logfile2 2>&1 echo "---> Please check the DB and reset the processes if needed. " >> $logfile2 2>&1 echo "to check: SELECT fNight, fRunId, fStartTime, fStopTime, fReturnCode FROM "$1" "$where";" >> $logfile2 2>&1 echo "to reset: UPDATE "$1" SET fStartTime=NULL, fStopTime=NULL, fAvailable=NULL, fReturnCode=NULL, fProcessingSiteKey=NULL "$where";" >> $logfile2 2>&1 fi } # get information of runs where transfer had a problem toquery="fNight, fRunID, fStartTime, fStopTime, fAvailable, fProcessingSiteKey, fReturnCode " toquery="CONCAT(fNight, '_', fRunID, '(', fStartTime, '-', fStopTime, ':', fReturnCode, ')') " toquery="CONCAT(fNight, '_', fRunID, ':', fReturnCode) " toquery="IF (ISNULL(fReturnCode), CONCAT(fNight, '_', fRunID, 'crashed'), CONCAT(fNight, '_', fRunID, 'failed', fReturnCode)) " where="WHERE NOT ISNULL(fReturnCode) OR (NOT ISNULL(fStartTime) AND ISNULL(fStopTime) AND fStartTime < DATE_ADD(Now(), INTERVAL -5 HOUR)) OR (ISNULL(fStartTime) AND NOT ISNULL(fStopTime)) " check_runs_in_db "RawFileRsyncedISDCStatus" check_runs_in_db "RawFileAvailWueStatus" # RawFileAvailISDC needs a different treatment # as return code 0 means that file is in fails folder in archive where="WHERE fReturnCode>0 OR (NOT ISNULL(fStartTime) AND ISNULL(fStopTime) AND fStartTime < DATE_ADD(Now(), INTERVAL -1 HOUR)) OR (ISNULL(fStartTime) AND NOT ISNULL(fStopTime)) " check_runs_in_db "RawFileAvailISDCStatus" # get nights from directory in LP dates=( `ssh fact@161.72.93.131 "find /loc_data/zipraw -mindepth 3 -type d | sort | sed -e 's/\/loc_data\/zipraw\///g' "` ) numdaysok=0 numdaysoklimit=10 sumdata=0 checklimit=5000 for date in ${dates[@]} do logfile=$logpath"/transfer/CheckTransfer_"`echo $date | sed -e 's/\//-/g'`".log" echo "" echo "" >> $logfile2 2>&1 echo "Processing "$date" ..." echo "Processing "$date" ..." >> $logfile2 2>&1 # avoid that already checked days are checked again if grep "EVERYTHING" $logfile >/dev/null 2>&1 then echo " "$date" has been checked already and is fine. Please check logfile "$logfile echo " "$date" has been checked already and is fine. Please check logfile "$logfile >> $logfile2 2>&1 sendemail="yes" continue fi if [ "$certaindate" != "" ] then checkstring=`echo $certaindate | grep -E -o '^20[0-9][0-9]\/[01][0-9]\/[0-3][0-9]$'` if [ "$checkstring" = "" ] then echo "Please give the variable certaindate in the correct format (YYYY/MM/DD)" finish fi if [ "$certaindate" != "$date" ] then printprocesslog "INFO continue, as certaindate has been set to "$certaindate echo " continue, as certaindate has been set to "$certaindate continue fi fi # some counters numdiff=0 numok=0 numpb=0 # check always only $numdaysoklimit days # and require at least $checklimit GB that have been checked # remark: bc: expr1 < expr2: the result is 1 if expr1 is strictly less than expr2 if [ $numdaysok -ge $numdaysoklimit ] && [ $(echo " $sumdata > $checklimit " | bc -l) -eq 1 ] then printprocesslog "INFO more than "$numdaysoklimit" ok and more than "$checklimit" GB checked. " continue fi # get paths date2=`echo $date | sed -e 's/\///g'` # la palma lprawpath="/newdaq/raw/"$date lprawpath2="/daq/raw/"$date # lprawpath2="/newdaq/raw/"$date #if daq is down lpziprawpath="/loc_data/zipraw/"$date qlapath="/daq/analysis/callisto/"$date # qlapath="/newdaq/analysis/callisto/"$date #if daq is down # isdc localrawpath="/scratch/from_lapalma/raw/"$date localrawpath3="/fact/raw/"$date # localfailpath="/archive/fact/fails/raw/"$date localfailpath="/gpfs/fact/fact-archive/fails/raw/"$date # wuerzburg wuerawpath="/fact/raw/"$date # dortmund #phidorawpath="/fhgfs/groups/app/fact-construction/raw/"$date # get disk usage and number of files for directory # on newdaq files with more than 3 digits as run number are excluded newdaq=( `ssh fact@161.72.93.131 "if [ -d $lprawpath ]; then ls $lprawpath/${date2}_[0-9][0-9][0-9][.]* | wc -l; du -s -b --apparent-size $lprawpath; else echo '-1 -1 -1'; fi"` ) daq=( `ssh fact@161.72.93.131 "if [ -d $lprawpath2 ]; then ls $lprawpath2/${date2}_[0-9][0-9][0-9][.]* | wc -l; du -s -b --apparent-size $lprawpath2; else echo '-1 -1 -1'; fi"` ) if [ ${newdaq[0]} -eq -1 ] && [ ${daq[0]} -eq -1 ] then printprocesslog "INFO no data available on newdaq for "$date echo "INFO no data available on newdaq for "$date >> $logfile 2>&1 continue fi zip=( `ssh fact@161.72.93.131 "if [ -d $lpziprawpath ]; then ls $lpziprawpath/${date2}_[0-9][0-9][0-9][.]* 2>/dev/null | wc -l; du -s -b --apparent-size $lpziprawpath; else echo '-1 -1 -1'; fi"` ) dl00=( `if [ -d $localrawpath ]; then ls $localrawpath/${date2}_[0-9][0-9][0-9][.]* | wc -l; du -s -b --apparent-size $localrawpath; else echo '-1 -1 -1'; fi` ) archive=( `if [ -d $localrawpath3 ]; then ls $localrawpath3/${date2}_[0-9][0-9][0-9][.]* 2>/dev/null | wc -l; du -L -s -b --apparent-size $localrawpath3; else echo '-1 -1 -1'; fi` ) fails=( `if [ -d $localfailpath ]; then ls $localfailpath/${date2}_[0-9][0-9][0-9][.]* | wc -l; du -L -s -b --apparent-size $localfailpath; else echo '-1 -1 -1'; fi` ) wue=( `ssh operator@coma.astro.uni-wuerzburg.de "if [ -d $wuerawpath ]; then ls $wuerawpath/${date2}_[0-9][0-9][0-9][.]* | wc -l; du -s -b --apparent-size $wuerawpath; else echo '-1 -1 -1'; fi"` ) #phido=( `ssh -i /home_nfs/isdc/fact_opr/.ssh/id_rsa.fact_opr.phido 129.217.160.201 "if [ -d $phidorawpath ]; then ls $phidorawpath/* | wc -l; du -s -b --apparent-size $phidorawpath; else echo '-1 -1 -1'; fi"` ) qla=( `ssh fact@161.72.93.131 "ls $qlapath/20*_C.root 2>/dev/null | wc -l"` ) query="SELECT Sum(if(fHasDrsFile=1,2,1)) FROM RunInfo WHERE fNight="$date2 querystart="SELECT Sum(if(fHasDrsFile=1,2,1)) FROM " queryjoin="LEFT JOIN RunInfo USING(fNight,fRunID) " querywhere="WHERE fNight="$date2" AND NOT ISNULL(fStartTime) AND NOT ISNULL(fStopTime) AND ISNULL(fReturnCode)" numruns=`sendquery` if [ "$numruns" == "" ] then numruns=0 fi query=$querystart"RawFileRsyncedISDCStatus "$queryjoin" "$querywhere numrsynced=`sendquery` if [ "$numrsynced" == "" ] then numrsynced=0 fi query=$querystart"RawFileAvailISDCStatus "$queryjoin" "$querywhere numisdc=`sendquery` if [ "$numisdc" == "" ] then numisdc=0 fi query=$querystart"RawFileAvailWueStatus "$queryjoin" "$querywhere numwue=`sendquery` if [ "$numwue" == "" ] then numwue=0 fi #query=$querystart"RawFileAvailPhidoStatus "$queryjoin" "$querywhere #numphido=`sendquery` #if [ "$numphido" == "" ] #then # numphido=0 #fi # select number of data runs query="SELECT COUNT(*) FROM RunInfo WHERE fNight="$date2 query=$query" AND fRunTypeKey=1" numdatruns=`sendquery` printprocesslog "disk: " printprocesslog " newdaq "${newdaq[@]} printprocesslog " daq "${daq[@]} printprocesslog " data "${zip[@]} printprocesslog " dl00 "${dl00[@]} printprocesslog " wue "${wue[@]} printprocesslog " arch "${archive[@]} printprocesslog " fail "${fails[@]} #printprocesslog " phido "${phido[@]} printprocesslog " qla "${qla[@]} printprocesslog "db: " printprocesslog " runinfo "$numruns printprocesslog " rsynced "$numrsynced printprocesslog " isdc "$numisdc printprocesslog " wue "$numwue #printprocesslog " phido "$numphido printprocesslog " data "$numdatruns echo "" >> $logfile 2>&1 echo "" >> $logfile 2>&1 echo "" >> $logfile 2>&1 echo `date`": checking the transfer..." >> $logfile 2>&1 echo "" >> $logfile 2>&1 echo "disk: " >> $logfile 2>&1 echo " newdaq "${newdaq[@]} >> $logfile 2>&1 echo " daq "${daq[@]} >> $logfile 2>&1 echo " data "${zip[@]} >> $logfile 2>&1 echo " dl00 "${dl00[@]} >> $logfile 2>&1 echo " wue "${wue[@]} >> $logfile 2>&1 echo " arch "${archive[@]} >> $logfile 2>&1 echo " fail "${fails[@]} >> $logfile 2>&1 #echo " phido "${phido[@]} >> $logfile 2>&1 echo " qla "${qla[@]} >> $logfile 2>&1 echo "db: " >> $logfile 2>&1 echo " runinfo "$numruns >> $logfile 2>&1 echo " rsynced "$numrsynced >> $logfile 2>&1 echo " isdc "$numisdc >> $logfile 2>&1 echo " wue "$numwue >> $logfile 2>&1 #echo " phido "$numphido >> $logfile 2>&1 echo " datruns "$numdatruns >> $logfile 2>&1 if ! [ $numdatruns -eq $qla ] then printprocesslog "WARN not all data runs are processed yet by the QLA for "$date"." echo "WARN not all data runs are processed yet by the QLA for "$date >> $logfile 2>&1 result1="1-" else result1="0-" fi # check if file are available in the different places if [ ${dl00[0]} -eq -1 ] && [ $date2 -lt 20120308 ] then printprocesslog "INFO data not available on /scratch on dl00 for "$date echo "INFO data not available on /scratch on dl00 for "$date >> $logfile 2>&1 fi if [ ${archive[0]} -eq -1 ] then printprocesslog "INFO data not in archive for "$date echo "INFO data not in archive for "$date >> $logfile 2>&1 fi # check if number of files agree in the different places # lp if ! [ ${daq[0]} -eq -1 ] && ! [ ${daq[0]} -eq ${newdaq[0]} ] then printprocesslog "WARN number of files on daq (" ${daq[0]}") does not agree with number of files on newdaq (" ${newdaq[0]}") for "$date echo "WARN number of files on daq (" ${daq[0]}") does not agree with number of files on newdaq (" ${newdaq[0]}") for "$date >> $logfile 2>&1 numpb=`echo " $numpb + 1 " | bc -l ` result1=$result1"1" else numok=`echo " $numok + 1 " | bc -l ` result1=$result1"0" fi # dl00 #if ! [ ${dl00[0]} -eq -1 ] && ! [ ${dl00[0]} -eq ${newdaq[0]} ] if ! [ ${dl00[0]} -eq ${newdaq[0]} ] then printprocesslog "WARN number of files on dl00 (" ${dl00[0]}") does not agree with number of files in LP (" ${newdaq[0]}") for "$date echo "WARN number of files on dl00 (" ${dl00[0]}") does not agree with number of files in LP (" ${newdaq[0]}") for "$date >> $logfile 2>&1 numpb=`echo " $numpb + 1 " | bc -l ` result1=$result1"1" else numok=`echo " $numok + 1 " | bc -l ` result1=$result1"0" fi # archive if ! [ ${archive[0]} -eq -1 ] && ! [ ${archive[0]} -eq ${newdaq[0]} ] then printprocesslog "WARN number of files in archive (" ${archive[0]}") does not agree with number of files in LP (" ${newdaq[0]}") for "$date echo "WARN number of files in archive (" ${archive[0]}") does not agree with number of files in LP (" ${newdaq[0]}") for "$date >> $logfile 2>&1 #check /archive/rev_1/failed if ! [ ${fails[0]} -eq -1 ] then sum=`echo " ${fails[0]} + ${archive[0]} " | bc -l ` if ! [ $sum -eq ${newdaq[0]} ] then printprocesslog "ERROR number of files in whole archive ("$sum") is different from number of files in La Palma ("${newdaq[0]}")." echo "ERROR number of files in whole archive ("$sum") is different from number of files in La Palma ("${newdaq[0]}")." >> $logfile 2>&1 numpb=`echo " $numpb + 1 " | bc -l ` result1=$result1"1" else numok=`echo " $numok + 1 " | bc -l ` result1=$result1"0" fi else result1=$result1"1" fi else if [ ${archive[0]} -eq -1 ] then numpb=`echo " $numpb + 1 " | bc -l ` result1=$result1"1" else numok=`echo " $numok + 1 " | bc -l ` result1=$result1"0" fi fi # wue #if ! [ ${wue[0]} -eq -1 ] && ! [ ${wue[0]} -eq ${newdaq[0]} ] if ! [ ${wue[0]} -eq ${newdaq[0]} ] then printprocesslog "WARN number of files in Wue (" ${wue[0]}") does not agree with number of files in LP (" ${newdaq[0]}") for "$date echo "WARN number of files in Wue (" ${wue[0]}") does not agree with number of files in LP (" ${newdaq[0]}") for "$date >> $logfile 2>&1 numpb=`echo " $numpb + 1 " | bc -l ` result1=$result1"1" else numok=`echo " $numok + 1 " | bc -l ` result1=$result1"0" fi ## phido #if ! [ ${phido[0]} -eq -1 ] && ! [ ${phido[0]} -eq ${newdaq[0]} ] #then # printprocesslog "WARN number of files on Phido (" ${phido[0]}") does not agree with number of files in LP (" ${newdaq[0]}") for "$date # echo "WARN number of files on Phido (" ${phido[0]}") does not agree with number of files in LP (" ${newdaq[0]}") for "$date >> $logfile 2>&1 # numpb=`echo " $numpb + 1 " | bc -l ` # result1=$result1"1" #else # numok=`echo " $numok + 1 " | bc -l ` # result1=$result1"0" #fi short2= # num files if [ "$result1" != "0-0000" ] && [ "$short" != "no" ] then short2="yes" fi if [ "$short2" = "yes" ] then printprocesslog "number of files does not yet agree in all sites ("$result1") -> do no further checking." echo "" >> $logfile 2>&1 echo "number of files does not yet agree in all sites ("$result1") -> do no further checking." >> $logfile 2>&1 echo "" >> $logfile 2>&1 # print to console echo "SUMMARY for "$date echo "-----------------------" echo " number of files does not yet agree in all sites: "${newdaq[0]}" (newdaq) "${daq[0]}" (daq) "${zip[0]}" (data) "${dl00[0]}" (dl) "${wue[0]}" (wue) "${archive[0]}" (arch) "${fails[0]}" (fails) "${qla[@]}" (qla) "$numdatruns" (datruns) " echo " "$date" is not yet transfered completely. Please check the logfile "$logfile # print to single logfile echo "SUMMARY for "$date >> $logfile 2>&1 echo "-----------------------" >> $logfile 2>&1 echo " number of files does not yet agree in all sites: "${newdaq[0]}" (newdaq) "${daq[0]}" (daq) "${zip[0]}" (data) "${dl00[0]}" (dl) "${wue[0]}" (wue) "${archive[0]}" (arch) "${fails[0]}" (fails) "${qla[@]}" (qla) "$numdatruns" (datruns) " >> $logfile 2>&1 echo " "$date" is not yet transfered completely. Please check the logfile "$logfile >> $logfile 2>&1 # print to global logfile echo "SUMMARY for "$date >> $logfile2 2>&1 echo "-----------------------" >> $logfile2 2>&1 echo " number of files does not yet agree in all sites: "${newdaq[0]}" (newdaq) "${daq[0]}" (daq) "${zip[0]}" (data) "${dl00[0]}" (dl) "${wue[0]}" (wue) "${archive[0]}" (arch) "${fails[0]}" (fails) "${qla[@]}" (qla) "$numdatruns" (datruns) " >> $logfile2 2>&1 echo " "$date" is not yet transfered completely. Please check the logfile "$logfile >> $logfile2 2>&1 continue fi # check du for raw files # la palma if ! [ ${newdaq[1]} -eq ${daq[1]} ] then printprocesslog "WARN size of data doesn't agree on newdaq ("${newdaq[1]}") and daq ("${daq[1]}") for "$date echo "WARN size of data doesn't agree on newdaq ("${newdaq[1]}") and daq ("${daq[1]}") for "$date >> $logfile 2>&1 numdiff=`echo " $numdiff + 1 " | bc -l ` numpb=`echo " $numpb + 1 " | bc -l ` result2="1" else numok=`echo " $numok + 1 " | bc -l ` result2="0" fi # check du for zipped raw files # dl00 if ! [ ${zip[1]} -eq ${dl00[1]} ] && [ $date2 -gt 20120307 ] then printprocesslog "WARN size of data doesn't agree on data ("${zip[1]}") and dl00 ("${dl00[1]}") for "$date echo "WARN size of data doesn't agree on data ("${zip[1]}") and dl00 ("${dl00[1]}") for "$date >> $logfile 2>&1 if ! [ ${dl00[1]} -eq -1 ] then numdiff=`echo " $numdiff + 1 " | bc -l ` numpb=`echo " $numpb + 1 " | bc -l ` result2=$result2"1" else numok=`echo " $numok + 1 " | bc -l ` result2=$result2"0" fi else numok=`echo " $numok + 1 " | bc -l ` result2=$result2"0" fi # archive if ! [ ${zip[1]} -eq ${archive[1]} ] then printprocesslog "WARN size of data doesn't agree on data ("${zip[1]}") and in archive ("${archive[1]}") for "$date echo "WARN size of data doesn't agree on data ("${zip[1]}") and in archive ("${archive[1]}") for "$date >> $logfile 2>&1 if ! [ ${archive[1]} -eq -1 ] then numdiff=`echo " $numdiff + 1 " | bc -l ` numpb=`echo " $numpb + 1 " | bc -l ` result2=$result2"1" else numok=`echo " $numok + 1 " | bc -l ` result2=$result2"0" fi else numok=`echo " $numok + 1 " | bc -l ` result2=$result2"0" fi # wue if ! [ ${zip[1]} -eq ${wue[1]} ] then printprocesslog "WARN size of data doesn't agree on data ("${zip[1]}") and in Wue ("${wue[1]}") for "$date echo "WARN size of data doesn't agree on data ("${zip[1]}") and in Wue ("${wue[1]}") for "$date >> $logfile 2>&1 if ! [ ${wue[1]} -eq -1 ] then numdiff=`echo " $numdiff + 1 " | bc -l ` numpb=`echo " $numpb + 1 " | bc -l ` result2=$result2"1" else numok=`echo " $numok + 1 " | bc -l ` result2=$result2"0" fi else numok=`echo " $numok + 1 " | bc -l ` result2=$result2"0" fi ## phido #if ! [ ${zip[1]} -eq ${phido[1]} ] #then # printprocesslog "WARN size of data doesn't agree on data ("${zip[1]}") and on Phido ("${phido[1]}") for "$date # echo "WARN size of data doesn't agree on data ("${zip[1]}") and on Phido ("${phido[1]}") for "$date >> $logfile 2>&1 # if ! [ ${phido[1]} -eq -1 ] # then # numdiff=`echo " $numdiff + 1 " | bc -l ` # numpb=`echo " $numpb + 1 " | bc -l ` # result2=$result2"1" # else # numok=`echo " $numok + 1 " | bc -l ` # result2=$result2"0" # fi #else # numok=`echo " $numok + 1 " | bc -l ` # result2=$result2"0" #fi #result=$result"-" # check DB (only starting from 8.3.2012) (if-clause to be removed later) if [ $date2 -gt 20120307 ] then # lp if ! [ $numruns -eq ${newdaq[0]} ] then printprocesslog "WARN number of runs on newdaq ("${newdaq[0]}") not equal to number of runs ("$numruns")" echo "WARN number of runs on newdaq ("${newdaq[0]}") not equal to number of runs ("$numruns")" >> $logfile 2>&1 numpb=`echo " $numpb + 1 " | bc -l ` result3="1" else numok=`echo " $numok + 1 " | bc -l ` result3="0" fi # dl00 if ! [ $numruns -eq $numrsynced ] then printprocesslog "WARN number of rsynced runs ("$numrsynced") not equal to number of runs ("$numruns")" echo "WARN number of rsynced runs ("$numrsynced") not equal to number of runs ("$numruns")" >> $logfile 2>&1 numpb=`echo " $numpb + 1 " | bc -l ` result3=$result3"1" else numok=`echo " $numok + 1 " | bc -l ` result3=$result3"0" fi # archive if ! [ $numruns -eq $numisdc ] then printprocesslog "WARN number of ingested files in archive ("$numisdc") not equal to number of runs ("$numruns")" echo "WARN number of ingested files in archive ("$numisdc") not equal to number of runs ("$numruns")" >> $logfile 2>&1 numpb=`echo " $numpb + 1 " | bc -l ` result3=$result3"1" else numok=`echo " $numok + 1 " | bc -l ` result3=$result3"0" fi # wue if ! [ $numruns -eq $numwue ] then printprocesslog "WARN number of backuped in Wue ("$numrsynced") not equal to number of runs ("$numruns")" echo "WARN number of backuped in Wue ("$numrsynced") not equal to number of runs ("$numruns")" >> $logfile 2>&1 numpb=`echo " $numpb + 1 " | bc -l ` result3=$result3"1" else numok=`echo " $numok + 1 " | bc -l ` result3=$result3"0" fi ## phido #if ! [ $numruns -eq $numphido ] #then # printprocesslog "WARN number of backuped on Phido ("$numrsynced") not equal to number of runs ("$numruns")" # echo "WARN number of backuped on Phido ("$numrsynced") not equal to number of runs ("$numruns")" >> $logfile 2>&1 # numpb=`echo " $numpb + 1 " | bc -l ` # result3=$result3"1" #else # numok=`echo " $numok + 1 " | bc -l ` # result3=$result3"0" #fi fi # db if [ "$result3" != "0000" ] && [ "$short" != "no" ] then short2="yes" fi if [ "$short2" = "yes" ] then printprocesslog "number of files does not yet agree in db ("$result3") -> do no further checking." echo "" >> $logfile 2>&1 echo "number of files does not yet agree in all sites ("$result1") -> do no further checking." >> $logfile 2>&1 echo "" >> $logfile 2>&1 # print to console echo "SUMMARY for "$date echo "-----------------------" echo " number of files does not yet agree in all sites: "${newdaq[0]}" (newdaq) "${daq[0]}" (daq) "${zip[0]}" (data) "${dl00[0]}" (dl) "${wue[0]}" (wue) "${archive[0]}" (arch) "${fails[0]}" (fails) "${qla[@]}" (qla) "$numdatruns" (datruns) " echo " "$date" is not yet transfered completely. Please check the logfile "$logfile # print to single logfile echo "SUMMARY for "$date >> $logfile 2>&1 echo "-----------------------" >> $logfile 2>&1 echo " number of files does not yet agree in all sites: "${newdaq[0]}" (newdaq) "${daq[0]}" (daq) "${zip[0]}" (data) "${dl00[0]}" (dl) "${wue[0]}" (wue) "${archive[0]}" (arch) "${fails[0]}" (fails) "${qla[@]}" (qla) "$numdatruns" (datruns) " >> $logfile 2>&1 echo " "$date" is not yet transfered completely. Please check the logfile "$logfile >> $logfile 2>&1 # print to global logfile echo "SUMMARY for "$date >> $logfile2 2>&1 echo "-----------------------" >> $logfile2 2>&1 echo " number of files does not yet agree in all sites: "${newdaq[0]}" (newdaq) "${daq[0]}" (daq) "${zip[0]}" (data) "${dl00[0]}" (dl) "${wue[0]}" (wue) "${archive[0]}" (arch) "${fails[0]}" (fails) "${qla[@]}" (qla) "$numdatruns" (datruns) " >> $logfile2 2>&1 echo " "$date" is not yet transfered completely. Please check the logfile "$logfile >> $logfile2 2>&1 continue fi #numdiff=0 # add for debugging so that single file sizes are not checked printprocesslog "numdiff: "$numdiff printprocesslog "INFO numok: "$numok printprocesslog "INFO numpb: "$numpb echo "numdiff: "$numdiff >> $logfile 2>&1 echo "INFO numok: "$numok >> $logfile 2>&1 echo "INFO numpb: "$numpb >> $logfile 2>&1 #if [ $numdiff -gt 0 ] if [ $numdiff -ge 0 ] then query="SELECT fRunID FROM RunInfo WHERE fNight="$date2 runs=( `sendquery` ) archivediffcounter=0 archiveokcounter=0 wuediffcounter=0 wueokcounter=0 #phidodiffcounter=0 #phidookcounter=0 dl00diffcounter=0 dl00okcounter=0 daqdiffcounter=0 daqokcounter=0 printprocesslog "INFO found "${#runs[@]}" rawfiles in DB." echo "INFO found "${#runs[@]}" rawfiles in DB." >> $logfile 2>&1 for run in ${runs[@]} do rawfile=$date2"_"`printf %03d $run`".fits" #rawfile2=$rawfile".gz" rawfile2=$rawfile".*z" printprocesslog "INFO checking "$rawfile # get file sizes for run #sizenewdaq=( `ssh fact@161.72.93.131 "ls -l ${lprawpath}/${rawfile} 2>/dev/null | awk '{ print \\\$5 }'"` ) sizenewdaq=( `ssh fact@161.72.93.131 "ls -lH ${lprawpath}/${rawfile} 2>/dev/null | awk '{ print \\\$5 }'"` ) sizedaq=( `ssh fact@161.72.93.131 "ls -l $lprawpath2/$rawfile 2>/dev/null | awk '{ print \\\$5 }'"` ) sizezip=( `ssh fact@161.72.93.131 "ls -l $lpziprawpath/$rawfile2 2>/dev/null | awk '{ print \\\$5 }'"` ) if ! [ ${dl00[1]} -eq -1 ] then sizedl00=( `ls -l $localrawpath/$rawfile2 2>/dev/null | awk '{ print \$5 }'` ) fi if ! [ ${archive[1]} -eq -1 ] then sizearchive=( `ls -lH $localrawpath3/$rawfile2 2>/dev/null | awk '{ print \$5 }'` ) fi if ! [ ${fails[1]} -eq -1 ] then sizefails=( `ls -lH $localfailpath/$rawfile2 2>/dev/null | awk '{ print \$5 }'` ) fi if ! [ ${wue[1]} -eq -1 ] then sizewue=( `ssh operator@coma.astro.uni-wuerzburg.de "ls -l $wuerawpath/$rawfile2 2>/dev/null | awk '{ print \\\$5 }'"` ) fi #if ! [ ${phido[1]} -eq -1 ] #then # sizephido=( `ssh -i /home_nfs/isdc/fact_opr/.ssh/id_rsa.fact_opr.phido 129.217.160.201 "ls -l $phidorawpath/$rawfile2 2>/dev/null | awk '{ print \\\$5 }'"` ) #fi # check file sizes for run # lp if ! [ "$sizenewdaq" = "$sizedaq" ] then printprocesslog "WARN "$rawfile" newdaq("$sizenewdaq") daq("$sizedaq")" echo " "$rawfile" newdaq("$sizenewdaq") daq("$sizedaq")" >> $logfile 2>&1 daqdiffcounter=`echo " $daqdiffcounter + 1 " | bc -l ` else daqokcounter=`echo " $daqokcounter + 1 " | bc -l ` fi # dl00 if ! [ "$sizezip" = "$sizedl00" ] && ! [ ${dl00[1]} -eq -1 ] then printprocesslog "WARN "$rawfile2" data("$sizezip") dl00("$sizedl00")" echo " "$rawfile2" data("$sizezip") dl00("$sizedl00")" >> $logfile 2>&1 dl00diffcounter=`echo " $dl00diffcounter + 1 " | bc -l ` else dl00okcounter=`echo " $dl00okcounter + 1 " | bc -l ` fi # archive #if [ "$sizezip" != "$sizearchive" -a ${archive[1]} -ne -1 -a "$sizearchive" != "" ] || [ "$sizezip" != "$sizefails" -a ${fails[1]} -ne -1 -a "$sizefails" != "" ] #not yet ingested files are treated wrongly #if [ ${archive[1]} -ne -1 ] && [ "$sizezip" != "$sizearchive" -o "$sizezip" != "$sizefails" ] if [ ${archive[1]} -ne -1 -a "$sizezip" != "$sizearchive" -a "$sizezip" != "$sizefails" ] then printprocesslog "WARN "$rawfile2" data("$sizezip") archive("$sizearchive"/"$sizefails")" echo " "$rawfile2" data("$sizezip") archive("$sizearchive"/"$sizefails")" >> $logfile 2>&1 #echo " "$sizezip"-"$sizearchive"-"${archive[1]}"-"$sizezip"-"$sizefails"-"${fails[1]} archivediffcounter=`echo " $archivediffcounter + 1 " | bc -l ` else archiveokcounter=`echo " $archiveokcounter + 1 " | bc -l ` fi # wue if ! [ "$sizezip" = "$sizewue" ] && ! [ ${wue[1]} -eq -1 ] then printprocesslog "WARN "$rawfile2" data("$sizezip") wue("$sizewue")" echo " "$rawfile2" data("$sizezip") wue("$sizewue")" >> $logfile 2>&1 wuediffcounter=`echo " $wuediffcounter + 1 " | bc -l ` else wueokcounter=`echo " $wueokcounter + 1 " | bc -l ` fi ## phido #if ! [ "$sizezip" = "$sizephido" ] && ! [ ${phido[1]} -eq -1 ] #then # printprocesslog " "$rawfile2" data("$sizezip") phido("$sizephido")" # echo " "$rawfile2" data("$sizezip") phido("$sizephido")" >> $logfile 2>&1 # phidodiffcounter=`echo " $phidodiffcounter + 1 " | bc -l ` #else # phidookcounter=`echo " $phidookcounter + 1 " | bc -l ` #fi done query="SELECT fRunID FROM RunInfo WHERE fNight="$date2" AND fHasDrsFile=1" drsruns=( `sendquery` ) printprocesslog "INFO found "${#drsruns[@]}" drsfiles in DB." echo "INFO found "${#drsruns[@]}" drsfiles in DB." >> $logfile 2>&1 for drsrun in ${drsruns[@]} do rawfile=$date2"_"`printf %03d $drsrun`".drs.fits" #rawfile2=$rawfile".gz" rawfile2=$rawfile".*z" # get file sizes for run #sizenewdaq=( `ssh fact@161.72.93.131 "ls -l ${lprawpath}/${rawfile} 2>/dev/null | awk '{ print \\\$5 }'"` ) sizenewdaq=( `ssh fact@161.72.93.131 "ls -lH ${lprawpath}/${rawfile} 2>/dev/null | awk '{ print \\\$5 }'"` ) sizedaq=( `ssh fact@161.72.93.131 "ls -l $lprawpath2/$rawfile 2>/dev/null | awk '{ print \\\$5 }'"` ) sizezip=( `ssh fact@161.72.93.131 "ls -l $lpziprawpath/$rawfile2 2>/dev/null | awk '{ print \\\$5 }'"` ) if ! [ ${dl00[1]} -eq -1 ] then sizedl00=( `ls -l $localrawpath/$rawfile2 2>/dev/null | awk '{ print \$5 }'` ) fi if ! [ ${archive[1]} -eq -1 ] then sizearchive=( `ls -lH $localrawpath3/$rawfile2 2>/dev/null | awk '{ print \$5 }'` ) fi if ! [ ${fails[1]} -eq -1 ] then sizefails=( `ls -lH $localfailpath/$rawfile2 2>/dev/null | awk '{ print \$5 }'` ) fi if ! [ ${wue[1]} -eq -1 ] then sizewue=( `ssh operator@coma.astro.uni-wuerzburg.de "ls -l $wuerawpath/$rawfile2 2>/dev/null | awk '{ print \\\$5 }'"` ) fi #if ! [ ${phido[1]} -eq -1 ] #then # sizephido=( `ssh -i /home_nfs/isdc/fact_opr/.ssh/id_rsa.fact_opr.phido 129.217.160.201 "ls -l $phidorawpath/$rawfile2 2>/dev/null | awk '{ print \\\$5 }'"` ) #fi # check file sizes for run # lp if ! [ "$sizenewdaq" = "$sizedaq" ] then printprocesslog "WARN "$rawfile" newdaq("$sizenewdaq") daq("$sizedaq")" echo " "$rawfile" newdaq("$sizenewdaq") daq("$sizedaq")" >> $logfile 2>&1 daqdiffcounter=`echo " $daqdiffcounter + 1 " | bc -l ` else daqokcounter=`echo " $daqokcounter + 1 " | bc -l ` fi # dl00 if ! [ "$sizezip" = "$sizedl00" ] && ! [ ${dl00[1]} -eq -1 ] then printprocesslog "WARN "$rawfile2" data("$sizezip") dl00("$sizedl00")" echo " "$rawfile2" data("$sizezip") dl00("$sizedl00")" >> $logfile 2>&1 dl00diffcounter=`echo " $dl00diffcounter + 1 " | bc -l ` else dl00okcounter=`echo " $dl00okcounter + 1 " | bc -l ` fi #if [ "$sizezip" != "$sizearchive" -a ${archive[1]} -ne -1 ] || [ "$sizezip" != "$sizefails" -a ${fails[1]} -ne -1 ] if [ "$sizezip" != "$sizearchive" -a ${archive[1]} -ne -1 -a "$sizearchive" != "" ] || [ "$sizezip" != "$sizefails" -a ${fails[1]} -ne -1 -a "$sizefails" != "" ] then printprocesslog "WARN "$rawfile2" data("$sizezip") archive("$sizearchive"/"$sizefails")" echo " "$rawfile2" data("$sizezip") archive("$sizearchive"/"$sizefails")" >> $logfile 2>&1 archivediffcounter=`echo " $archivediffcounter + 1 " | bc -l ` else archiveokcounter=`echo " $archiveokcounter + 1 " | bc -l ` fi # wue if ! [ "$sizezip" = "$sizewue" ] && ! [ ${wue[1]} -eq -1 ] then printprocesslog "WARN "$rawfile2" data("$sizezip") wue("$sizewue")" echo " "$rawfile2" data("$sizezip") wue("$sizewue")" >> $logfile 2>&1 wuediffcounter=`echo " $wuediffcounter + 1 " | bc -l ` else wueokcounter=`echo " $wueokcounter + 1 " | bc -l ` fi ## phido #if ! [ "$sizezip" = "$sizephido" ] && ! [ ${phido[1]} -eq -1 ] #then # printprocesslog "WARN "$rawfile2" data("$sizezip") phido("$sizephido")" # echo " "$rawfile2" data("$sizezip") phido("$sizephido")" >> $logfile 2>&1 # phidodiffcounter=`echo " $phidodiffcounter + 1 " | bc -l ` #else # phidookcounter=`echo " $phidookcounter + 1 " | bc -l ` #fi done #result=$result"-" # raw files if [ $daqokcounter -eq ${daq[0]} ] then numok=`echo " $numok + 1 " | bc -l ` result4="0" else result4="1" numpb=`echo " $numpb + 1 " | bc -l ` fi # zipped files # dl00 if [ $dl00okcounter -eq ${dl00[0]} ] then result4=$result4"0" numok=`echo " $numok + 1 " | bc -l ` else result4=$result4"1" numpb=`echo " $numpb + 1 " | bc -l ` fi # archive # daq had been used, because archive[0] doesn't include fails[0] #if [ $archiveokcounter -eq ${newdaq[0]} ] # archive[0] had been used, because newdaq[0] might be empty # in case the data was taken on data #if [ $archiveokcounter -eq ${archive[0]} ] # compare with daq[0] as there should be always data on data if [ $archiveokcounter -eq ${daq[0]} ] then result4=$result4"0" numok=`echo " $numok + 1 " | bc -l ` else result4=$result4"1" numpb=`echo " $numpb + 1 " | bc -l ` fi # wue if [ $wueokcounter -eq ${wue[0]} ] then result4=$result4"0" numok=`echo " $numok + 1 " | bc -l ` else result4=$result4"1" numpb=`echo " $numpb + 1 " | bc -l ` fi ## phido #if [ $phidookcounter -eq ${phido[0]} ] #then # result4=$result4"0" # numok=`echo " $numok + 1 " | bc -l ` #else # result4=$result4"1" # numpb=`echo " $numpb + 1 " | bc -l ` #fi printprocesslog "INFO "$daqokcounter" files are ok on daq (raw)." printprocesslog "INFO "$dl00okcounter" files are ok on dl00." printprocesslog "INFO "$wueokcounter" files are ok in Wue." printprocesslog "INFO "$archiveokcounter" files are ok in the archive." #printprocesslog "INFO "$phidookcounter" files are ok on Phido." echo "INFO "$daqokcounter" files are ok on daq (raw)." >> $logfile 2>&1 echo "INFO "$dl00okcounter" files are ok on dl00." >> $logfile 2>&1 echo "INFO "$wueokcounter" files are ok in Wue." >> $logfile 2>&1 echo "INFO "$archiveokcounter" files are ok in the archive." >> $logfile 2>&1 #echo "INFO "$phidookcounter" files are ok on Phido." >> $logfile 2>&1 if [ $daqdiffcounter -gt 0 ] then printprocesslog "WARN "$daqdiffcounter" files have a different size on daq (raw)." echo "WARN "$daqdiffcounter" files have a different size on daq (raw)." >> $logfile 2>&1 else printprocesslog "INFO "$daqdiffcounter" files have a different size on daq (raw)." echo "INFO "$daqdiffcounter" files have a different size on daq (raw)." >> $logfile 2>&1 fi if [ $dl00diffcounter -gt 0 ] then printprocesslog "WARN "$dl00diffcounter" files have a different size on dl00." echo "WARN "$dl00diffcounter" files have a different size on dl00." >> $logfile 2>&1 else printprocesslog "INFO "$dl00diffcounter" files have a different size on dl00." echo "INFO "$dl00diffcounter" files have a different size on dl00." >> $logfile 2>&1 fi if [ $wuediffcounter -gt 0 ] then printprocesslog "WARN "$wuediffcounter" files have a different size in Wue." echo "WARN "$wuediffcounter" files have a different size in Wue." >> $logfile 2>&1 else printprocesslog "INFO "$wuediffcounter" files have a different size in Wue." echo "INFO "$wuediffcounter" files have a different size in Wue." >> $logfile 2>&1 fi if [ $archivediffcounter -gt 0 ] then printprocesslog "WARN "$archivediffcounter" files have a different size in the archive." echo "WARN "$archivediffcounter" files have a different size in the archive." >> $logfile 2>&1 else printprocesslog "INFO "$archivediffcounter" files have a different size in the archive." echo "INFO "$archivediffcounter" files have a different size in the archive." >> $logfile 2>&1 fi #printprocesslog "WARN "$phidodiffcounter" files have a different size on Phido." #echo "WARN "$phidodiffcounter" files have a different size on Phido." >> $logfile 2>&1 fi # print summary: printprocesslog "INFO day ok: "$numdaysok printprocesslog "INFO numok: "$numok printprocesslog "INFO numpb: "$numpb printprocesslog "result:" printprocesslog "(qla-#files-dudir-db-filesize)" #printprocesslog " ldawp-ldawp-ldawp-ldawp" printprocesslog " q-ldaw-ldaw-ldaw-ldaw" printprocesslog " "$result1"-"$result2"-"$result3"-"$result4 echo "INFO day ok: "$numdaysok >> $logfile 2>&1 echo "INFO numok: "$numok >> $logfile 2>&1 echo "INFO numpb: "$numpb >> $logfile 2>&1 echo "result:" >> $logfile 2>&1 echo "(qla-#files-dudir-db-filesize)" >> $logfile 2>&1 #echo " ldawp-ldawp-ldawp-ldawp" >> $logfile 2>&1 echo " q-ldaw-ldaw-ldaw-ldaw" >> $logfile 2>&1 echo " "$result1"-"$result2"-"$result3"-"$result4 >> $logfile 2>&1 sumdatanew=`echo " ( ${daq[1]} + ${zip[1]} ) / 1024 / 1024 / 1024 " | bc -l | cut -d. -f1` sumdata=`echo " $sumdata + $sumdatanew " | bc -l | cut -d. -f1` printprocesslog "checked alread "$sumdata" GB. "$sumdatanew" "${daq[1]}" "${zip[1]} if [ $numpb -lt 4 ] then numdaysok=`echo " $numdaysok + 1 " | bc -l ` fi echo "SUMMARY for "$date echo "-----------------------" echo "SUMMARY for "$date >> $logfile 2>&1 echo "-----------------------" >> $logfile 2>&1 echo "SUMMARY for "$date >> $logfile2 2>&1 echo "-----------------------" >> $logfile2 2>&1 #echo "res1:"$result1 #echo "res3:"$result3 #echo "res4:"$result4 #echo "arch:"${archive[0]} #echo "isdc:"$numisdc if [ "$result1" = "0-0000" ] && [ "$result3" = "0000" ] && [ "$result4" = "0000" ] then echo " EVERYTHING is ok. "$date" can be deleted. " echo "" >> $logfile 2>&1 echo " EVERYTHING is ok. "$date" can be deleted. " >> $logfile 2>&1 echo " EVERYTHING is ok. "$date" can be deleted. " >> $logfile2 2>&1 echo " Details in the logfile "$logfile echo "" >> $logfile 2>&1 echo " Details in the logfile "$logfile >> $logfile2 2>&1 sendemail="yes" else if [ "$result1" = "0-0000" ] && [ "$result3" = "0010" ] && [ "$result4" = "0000" ] && [ ${archive[0]} -eq $numisdc ] then echo " "${fails[0]}" file(s) corrupt (fails folder), but files are transfered correctly. " echo " "${fails[0]}" file(s) corrupt (fails folder), but files are transfered correctly. " >> $logfile 2>&1 echo " "${fails[0]}" file(s) corrupt (fails folder), but files are transfered correctly. " >> $logfile2 2>&1 echo " TRANSFER is ok. "$date" can be deleted. " echo "" >> $logfile 2>&1 echo " TRANSFER is ok. "$date" can be deleted. " >> $logfile 2>&1 echo " TRANSFER is ok. "$date" can be deleted. " >> $logfile2 2>&1 echo " Details in the logfile "$logfile echo "" >> $logfile 2>&1 echo " Details in the logfile "$logfile >> $logfile2 2>&1 sendemail="yes" else echo " "$date" is not yet transfered completely. Please check the logfile "$logfile echo "" >> $logfile 2>&1 echo " "$date" is not yet transfered completely. Please check the logfile "$logfile >> $logfile 2>&1 echo " "$date" is not yet transfered completely. Please check the logfile "$logfile >> $logfile2 2>&1 echo " resetting jobs in the DB might be needed." echo " resetting jobs in the DB might be needed." >> $logfile 2>&1 echo " resetting jobs in the DB might be needed." >> $logfile2 2>&1 fi fi if [ $daqdiffcounter -gt 0 ] then echo " WARN "$daqdiffcounter" files have a different size on daq (raw)." echo " WARN "$daqdiffcounter" files have a different size on daq (raw)." >> $logfile 2>&1 echo " WARN "$daqdiffcounter" files have a different size on daq (raw)." >> $logfile2 2>&1 fi if [ $dl00diffcounter -gt 0 ] then echo " WARN "$dl00diffcounter" files have a different size on dl00." echo " WARN "$dl00diffcounter" files have a different size on dl00." >> $logfile 2>&1 echo " WARN "$dl00diffcounter" files have a different size on dl00." >> $logfile2 2>&1 fi if [ $wuediffcounter -gt 0 ] then echo " WARN "$wuediffcounter" files have a different size in Wue." echo " WARN "$wuediffcounter" files have a different size in Wue." >> $logfile 2>&1 echo " WARN "$wuediffcounter" files have a different size in Wue." >> $logfile2 2>&1 fi if [ $archivediffcounter -gt 0 ] then echo " WARN "$archivediffcounter" files have a different size in the archive." echo " WARN "$archivediffcounter" files have a different size in the archive." >> $logfile 2>&1 echo " WARN "$archivediffcounter" files have a different size in the archive." >> $logfile2 2>&1 fi if ! [ $numdatruns -eq $qla ] then echo "WARN not all data runs are processed yet by the QLA. " echo "WARN not all data runs are processed yet by the QLA. " >> $logfile 2>&1 echo "WARN not all data runs are processed yet by the QLA. " >> $logfile2 2>&1 fi echo "" echo "" >> $logfile 2>&1 echo "" >> $logfile2 2>&1 done #sendemail="yes" if [ "$sendemail" = "yes" ] then echo "" echo "INFO send email with "$logfile2"to shift@fact-project.org " printprocesslog "INFO send email with "$logfile2"to shift@fact-project.org " cat $logfile2 | mail -s "testmail for info on deleting data" shift@fact-project.org fi printprocesslog "INFO finished $0"