ViewVC Help
View File | Revision Log | Show Annotations | Root Listing
root/cvsroot/COMP/CRAB/python/cms_cmssw.py
(Generate patch)

Comparing COMP/CRAB/python/cms_cmssw.py (file contents):
Revision 1.73 by gutsche, Sun Apr 8 18:39:51 2007 UTC vs.
Revision 1.116.2.1 by fanzago, Tue Oct 9 10:46:56 2007 UTC

# Line 2 | Line 2 | from JobType import JobType
2   from crab_logger import Logger
3   from crab_exceptions import *
4   from crab_util import *
5 + from BlackWhiteListParser import BlackWhiteListParser
6   import common
6 – import PsetManipulator  
7 – import DataDiscovery
8 – import DataDiscovery_DBS2
9 – import DataLocation
7   import Scram
8  
9 < import os, string, re, shutil, glob
9 > import os, string, glob
10  
11   class Cmssw(JobType):
12      def __init__(self, cfg_params, ncjobs):
13          JobType.__init__(self, 'CMSSW')
14          common.logger.debug(3,'CMSSW::__init__')
15  
19 –        # Marco.
16          self._params = {}
17          self.cfg_params = cfg_params
18  
19 +        # init BlackWhiteListParser
20 +        self.blackWhiteListParser = BlackWhiteListParser(cfg_params)
21 +
22          try:
23              self.MaxTarBallSize = float(self.cfg_params['EDG.maxtarballsize'])
24          except KeyError:
25 <            self.MaxTarBallSize = 100.0
25 >            self.MaxTarBallSize = 9.5
26  
27          # number of jobs requested to be created, limit obj splitting
28          self.ncjobs = ncjobs
# Line 36 | Line 35 | class Cmssw(JobType):
35          self.executable = ''
36          self.executable_arch = self.scram.getArch()
37          self.tgz_name = 'default.tgz'
38 +        self.additional_tgz_name = 'additional.tgz'
39          self.scriptName = 'CMSSW.sh'
40          self.pset = ''      #scrip use case Da  
41          self.datasetPath = '' #scrip use case Da
# Line 44 | Line 44 | class Cmssw(JobType):
44          self.fjrFileName = 'crab_fjr.xml'
45  
46          self.version = self.scram.getSWVersion()
47 +        
48 +        #
49 +        # Try to block creation in case of arch/version mismatch
50 +        #
51 +
52 +        a = string.split(self.version, "_")
53 +
54 +        if int(a[1]) == 1 and (int(a[2]) < 5 and self.executable_arch.find('slc4') == 0):
55 +            msg = "Error: CMS does not support %s with %s architecture"%(self.version, self.executable_arch)
56 +            raise CrabException(msg)
57 +        if int(a[1]) == 1 and (int(a[2]) >= 5 and self.executable_arch.find('slc3') == 0):
58 +            msg = "Error: CMS does not support %s with %s architecture"%(self.version, self.executable_arch)
59 +            raise CrabException(msg)
60 +        
61          common.taskDB.setDict('codeVersion',self.version)
62          self.setParam_('application', self.version)
63  
# Line 51 | Line 65 | class Cmssw(JobType):
65  
66          ## get DBS mode
67          try:
68 <            self.use_dbs_2 = int(self.cfg_params['CMSSW.use_dbs_2'])
68 >            self.use_dbs_1 = int(self.cfg_params['CMSSW.use_dbs_1'])
69          except KeyError:
70 <            self.use_dbs_2 = 0
70 >            self.use_dbs_1 = 0
71              
72          try:
73              tmp =  cfg_params['CMSSW.datasetpath']
# Line 74 | Line 88 | class Cmssw(JobType):
88              self.setParam_('dataset', 'None')
89              self.setParam_('owner', 'None')
90          else:
91 <            datasetpath_split = self.datasetPath.split("/")
92 <            self.setParam_('dataset', datasetpath_split[1])
93 <            self.setParam_('owner', datasetpath_split[-1])
94 <
91 >            try:
92 >                datasetpath_split = self.datasetPath.split("/")
93 >                # standard style
94 >                self.setParam_('datasetFull', self.datasetPath)
95 >                if self.use_dbs_1 == 1 :
96 >                    self.setParam_('dataset', datasetpath_split[1])
97 >                    self.setParam_('owner', datasetpath_split[-1])
98 >                else:
99 >                    self.setParam_('dataset', datasetpath_split[1])
100 >                    self.setParam_('owner', datasetpath_split[2])
101 >            except:
102 >                self.setParam_('dataset', self.datasetPath)
103 >                self.setParam_('owner', self.datasetPath)
104 >                
105          self.setTaskid_()
106          self.setParam_('taskId', self.cfg_params['taskId'])
107  
# Line 127 | Line 151 | class Cmssw(JobType):
151                      self.output_file.append(tmp)
152                      pass
153              else:
154 <                log.message("No output file defined: only stdout/err and the CRAB Framework Job Report will be available")
154 >                log.message("No output file defined: only stdout/err and the CRAB Framework Job Report will be available\n")
155                  pass
156              pass
157          except KeyError:
158 <            log.message("No output file defined: only stdout/err and the CRAB Framework Job Report will be available")
158 >            log.message("No output file defined: only stdout/err and the CRAB Framework Job Report will be available\n")
159              pass
160  
161          # script_exe file as additional file in inputSandbox
# Line 157 | Line 181 | class Cmssw(JobType):
181                  tmp = string.strip(tmp)
182                  dirname = ''
183                  if not tmp[0]=="/": dirname = "."
184 <                files = glob.glob(os.path.join(dirname, tmp))
184 >                files = []
185 >                if string.find(tmp,"*")>-1:
186 >                    files = glob.glob(os.path.join(dirname, tmp))
187 >                    if len(files)==0:
188 >                        raise CrabException("No additional input file found with this pattern: "+tmp)
189 >                else:
190 >                    files.append(tmp)
191                  for file in files:
192                      if not os.path.exists(file):
193                          raise CrabException("Additional input file not found: "+file)
194                      pass
195 <                    storedFile = common.work_space.shareDir()+file
196 <                    shutil.copyfile(file, storedFile)
197 <                    self.additional_inbox_files.append(string.strip(storedFile))
195 >                    # fname = string.split(file, '/')[-1]
196 >                    # storedFile = common.work_space.pathForTgz()+'share/'+fname
197 >                    # shutil.copyfile(file, storedFile)
198 >                    self.additional_inbox_files.append(string.strip(file))
199                  pass
200              pass
201              common.logger.debug(5,"Additional input files: "+str(self.additional_inbox_files))
# Line 222 | Line 253 | class Cmssw(JobType):
253          except KeyError:
254              self.sourceSeedVtx = None
255              common.logger.debug(5,"No vertex seed given")
256 +
257 +        try:
258 +            self.sourceSeedG4 = int(cfg_params['CMSSW.g4_seed'])
259 +        except KeyError:
260 +            self.sourceSeedG4 = None
261 +            common.logger.debug(5,"No g4 sim hits seed given")
262 +
263 +        try:
264 +            self.sourceSeedMix = int(cfg_params['CMSSW.mix_seed'])
265 +        except KeyError:
266 +            self.sourceSeedMix = None
267 +            common.logger.debug(5,"No mix seed given")
268 +
269          try:
270              self.firstRun = int(cfg_params['CMSSW.first_run'])
271          except KeyError:
272              self.firstRun = None
273              common.logger.debug(5,"No first run given")
274          if self.pset != None: #CarlosDaniele
275 <            self.PsetEdit = PsetManipulator.PsetManipulator(self.pset) #Daniele Pset
275 >            ver = string.split(self.version,"_")
276 >            if (int(ver[1])>=1 and int(ver[2])>=5):
277 >                import PsetManipulator150 as pp
278 >            else:
279 >                import PsetManipulator as pp
280 >            PsetEdit = pp.PsetManipulator(self.pset) #Daniele Pset
281  
282          #DBSDLS-start
283          ## Initialize the variables that are extracted from DBS/DLS and needed in other places of the code
# Line 250 | Line 299 | class Cmssw(JobType):
299                  self.jobSplittingForScript()
300              else:
301                  self.jobSplittingNoInput()
302 <        else:
302 >        else:
303              self.jobSplittingByBlocks(blockSites)
304  
305          # modify Pset
# Line 258 | Line 307 | class Cmssw(JobType):
307              try:
308                  if (self.datasetPath): # standard job
309                      # allow to processa a fraction of events in a file
310 <                    self.PsetEdit.inputModule("INPUT")
311 <                    self.PsetEdit.maxEvent("INPUTMAXEVENTS")
312 <                    self.PsetEdit.skipEvent("INPUTSKIPEVENTS")
310 >                    PsetEdit.inputModule("INPUT")
311 >                    PsetEdit.maxEvent("INPUTMAXEVENTS")
312 >                    PsetEdit.skipEvent("INPUTSKIPEVENTS")
313                  else:  # pythia like job
314 <                    self.PsetEdit.maxEvent(self.eventsPerJob)
314 >                    PsetEdit.maxEvent(self.eventsPerJob)
315                      if (self.firstRun):
316 <                        self.PsetEdit.pythiaFirstRun("INPUTFIRSTRUN")  #First Run
316 >                        PsetEdit.pythiaFirstRun("INPUTFIRSTRUN")  #First Run
317                      if (self.sourceSeed) :
318 <                        self.PsetEdit.pythiaSeed("INPUT")
318 >                        PsetEdit.pythiaSeed("INPUT")
319                          if (self.sourceSeedVtx) :
320 <                            self.PsetEdit.pythiaSeedVtx("INPUTVTX")
320 >                            PsetEdit.vtxSeed("INPUTVTX")
321 >                        if (self.sourceSeedG4) :
322 >                            PsetEdit.g4Seed("INPUTG4")
323 >                        if (self.sourceSeedMix) :
324 >                            PsetEdit.mixSeed("INPUTMIX")
325                  # add FrameworkJobReport to parameter-set
326 <                self.PsetEdit.addCrabFJR(self.fjrFileName)
327 <                self.PsetEdit.psetWriter(self.configFilename())
326 >                PsetEdit.addCrabFJR(self.fjrFileName)
327 >                PsetEdit.psetWriter(self.configFilename())
328              except:
329                  msg='Error while manipuliating ParameterSet: exiting...'
330                  raise CrabException(msg)
331  
332      def DataDiscoveryAndLocation(self, cfg_params):
333  
334 +        import DataDiscovery
335 +        import DataDiscovery_DBS2
336 +        import DataLocation
337          common.logger.debug(10,"CMSSW::DataDiscoveryAndLocation()")
338  
339          datasetPath=self.datasetPath
340  
341          ## Contact the DBS
342 <        common.logger.message("Contacting DBS...")
342 >        common.logger.message("Contacting Data Discovery Services ...")
343          try:
344  
345 <            if self.use_dbs_2 == 1 :
290 <                self.pubdata=DataDiscovery_DBS2.DataDiscovery_DBS2(datasetPath, cfg_params)
291 <            else :
345 >            if self.use_dbs_1 == 1 :
346                  self.pubdata=DataDiscovery.DataDiscovery(datasetPath, cfg_params)
347 +            else :
348 +                self.pubdata=DataDiscovery_DBS2.DataDiscovery_DBS2(datasetPath, cfg_params)
349              self.pubdata.fetchDBSInfo()
350  
351          except DataDiscovery.NotExistingDatasetError, ex :
# Line 311 | Line 367 | class Cmssw(JobType):
367              msg = 'ERROR ***: failed Data Discovery in DBS :  %s'%ex.getErrorMessage()
368              raise CrabException(msg)
369  
314 –        ## get list of all required data in the form of dbs paths  (dbs path = /dataset/datatier/owner)
315 –        common.logger.message("Required data are :"+self.datasetPath)
316 –
370          self.filesbyblock=self.pubdata.getFiles()
371          self.eventsbyblock=self.pubdata.getEventsPerBlock()
372          self.eventsbyfile=self.pubdata.getEventsPerFile()
373  
374          ## get max number of events
375          self.maxEvents=self.pubdata.getMaxEvents() ##  self.maxEvents used in Creator.py
323 –        common.logger.message("The number of available events is %s\n"%self.maxEvents)
376  
325 –        common.logger.message("Contacting DLS...")
377          ## Contact the DLS and build a list of sites hosting the fileblocks
378          try:
379              dataloc=DataLocation.DataLocation(self.filesbyblock.keys(),cfg_params)
# Line 340 | Line 391 | class Cmssw(JobType):
391                  allSites.append(oneSite)
392          allSites = self.uniquelist(allSites)
393  
394 <        common.logger.message("Sites ("+str(len(allSites))+") hosting part/all of dataset: "+str(allSites))
395 <        common.logger.debug(6, "List of Sites: "+str(allSites))
394 >        # screen output
395 >        common.logger.message("Requested dataset: " + datasetPath + " has " + str(self.maxEvents) + " events in " + str(len(self.filesbyblock.keys())) + " blocks.\n")
396 >
397          return sites
398      
399      def jobSplittingByBlocks(self, blockSites):
# Line 403 | Line 455 | class Cmssw(JobType):
455          jobCount = 0
456          list_of_lists = []
457  
458 +        # list tracking which jobs are in which jobs belong to which block
459 +        jobsOfBlock = {}
460 +
461          # ---- Iterate over the blocks in the dataset until ---- #
462          # ---- we've met the requested total # of events    ---- #
463          while ( (eventsRemaining > 0) and (blockCount < numBlocksInDataset) and (jobCount < totalNumberOfJobs)):
464              block = blocks[blockCount]
465              blockCount += 1
466 +            if block not in jobsOfBlock.keys() :
467 +                jobsOfBlock[block] = []
468              
469              if self.eventsbyblock.has_key(block) :
470                  numEventsInBlock = self.eventsbyblock[block]
# Line 457 | Line 514 | class Cmssw(JobType):
514                              common.logger.debug(3,"Job "+str(jobCount+1)+" can run over "+str(filesEventCount - jobSkipEventCount)+" events (last file in block).")
515                              self.jobDestination.append(blockSites[block])
516                              common.logger.debug(5,"Job "+str(jobCount+1)+" Destination: "+str(self.jobDestination[jobCount]))
517 +                            # fill jobs of block dictionary
518 +                            jobsOfBlock[block].append(jobCount+1)
519                              # reset counter
520                              jobCount = jobCount + 1
521                              totalEventCount = totalEventCount + filesEventCount - jobSkipEventCount
# Line 480 | Line 539 | class Cmssw(JobType):
539                          common.logger.debug(3,"Job "+str(jobCount+1)+" can run over "+str(eventsPerJobRequested)+" events.")
540                          self.jobDestination.append(blockSites[block])
541                          common.logger.debug(5,"Job "+str(jobCount+1)+" Destination: "+str(self.jobDestination[jobCount]))
542 +                        jobsOfBlock[block].append(jobCount+1)
543                          # reset counter
544                          jobCount = jobCount + 1
545                          totalEventCount = totalEventCount + eventsPerJobRequested
# Line 500 | Line 560 | class Cmssw(JobType):
560                          common.logger.debug(3,"Job "+str(jobCount+1)+" can run over "+str(eventsPerJobRequested)+" events.")
561                          self.jobDestination.append(blockSites[block])
562                          common.logger.debug(5,"Job "+str(jobCount+1)+" Destination: "+str(self.jobDestination[jobCount]))
563 +                        jobsOfBlock[block].append(jobCount+1)
564                          # increase counter
565                          jobCount = jobCount + 1
566                          totalEventCount = totalEventCount + eventsPerJobRequested
# Line 517 | Line 578 | class Cmssw(JobType):
578          self.ncjobs = self.total_number_of_jobs = jobCount
579          if (eventsRemaining > 0 and jobCount < totalNumberOfJobs ):
580              common.logger.message("Could not run on all requested events because some blocks not hosted at allowed sites.")
581 <        common.logger.message("\n"+str(jobCount)+" job(s) can run on "+str(totalEventCount)+" events.\n")
581 >        common.logger.message(str(jobCount)+" job(s) can run on "+str(totalEventCount)+" events.\n")
582          
583 +        # screen output
584 +        screenOutput = "List of jobs and available destination sites:\n\n"
585 +
586 +        blockCounter = 0
587 +        for block in blocks:
588 +            if block in jobsOfBlock.keys() :
589 +                blockCounter += 1
590 +                screenOutput += "Block %5i: jobs %20s: sites: %s\n" % (blockCounter,spanRanges(jobsOfBlock[block]),','.join(self.blackWhiteListParser.checkWhiteList(self.blackWhiteListParser.checkBlackList(blockSites[block],block),block)))
591 +
592 +        common.logger.message(screenOutput)
593 +
594          self.list_of_args = list_of_lists
595          return
596  
# Line 563 | Line 635 | class Cmssw(JobType):
635              ## Since there is no input, any site is good
636             # self.jobDestination.append(["Any"])
637              self.jobDestination.append([""]) #must be empty to write correctly the xml
638 <            args=''
638 >            args=[]
639              if (self.firstRun):
640                      ## pythia first run
641                  #self.list_of_args.append([(str(self.firstRun)+str(i))])
642 <                args=args+(str(self.firstRun)+str(i))
642 >                args.append(str(self.firstRun)+str(i))
643              else:
644                  ## no first run
645                  #self.list_of_args.append([str(i)])
646 <                args=args+str(i)
646 >                args.append(str(i))
647              if (self.sourceSeed):
648 +                args.append(str(self.sourceSeed)+str(i))
649                  if (self.sourceSeedVtx):
650 <                    ## pythia + vtx random seed
651 <                    #self.list_of_args.append([
652 <                    #                          str(self.sourceSeed)+str(i),
653 <                    #                          str(self.sourceSeedVtx)+str(i)
654 <                    #                          ])
655 <                    args=args+str(',')+str(self.sourceSeed)+str(i)+str(',')+str(self.sourceSeedVtx)+str(i)
656 <                else:
657 <                    ## only pythia random seed
658 <                    #self.list_of_args.append([(str(self.sourceSeed)+str(i))])
659 <                    args=args +str(',')+str(self.sourceSeed)+str(i)
660 <            else:
661 <                ## no random seed
589 <                if str(args)=='': args=args+(str(self.firstRun)+str(i))
590 <            arguments=args.split(',')
591 <            if len(arguments)==3:self.list_of_args.append([str(arguments[0]),str(arguments[1]),str(arguments[2])])
592 <            elif len(arguments)==2:self.list_of_args.append([str(arguments[0]),str(arguments[1])])
593 <            else :self.list_of_args.append([str(arguments[0])])
650 >                    ## + vtx random seed
651 >                    args.append(str(self.sourceSeedVtx)+str(i))
652 >                if (self.sourceSeedG4):
653 >                    ## + G4 random seed
654 >                    args.append(str(self.sourceSeedG4)+str(i))
655 >                if (self.sourceSeedMix):    
656 >                    ## + Mix random seed
657 >                    args.append(str(self.sourceSeedMix)+str(i))
658 >                pass
659 >            pass
660 >            self.list_of_args.append(args)
661 >        pass
662              
663 <     #   print self.list_of_args
663 >        # print self.list_of_args
664  
665          return
666  
# Line 687 | Line 755 | class Cmssw(JobType):
755          try: # create tar ball
756              tar = tarfile.open(self.tgzNameWithPath, "w:gz")
757              ## First find the executable
758 <            if (executable != ''):
758 >            if (self.executable != ''):
759                  exeWithPath = self.scram.findFile_(executable)
760                  if ( not exeWithPath ):
761                      raise CrabException('User executable '+executable+' not found')
# Line 697 | Line 765 | class Cmssw(JobType):
765                      # the exe is private, so we must ship
766                      common.logger.debug(5,"Exe "+exeWithPath+" to be tarred")
767                      path = swArea+'/'
768 <                    exe = string.replace(exeWithPath, path,'')
769 <                    tar.add(path+exe,executable)
768 >                    # distinguish case when script is in user project area or given by full path somewhere else
769 >                    if exeWithPath.find(path) >= 0 :
770 >                        exe = string.replace(exeWithPath, path,'')
771 >                        tar.add(path+exe,os.path.basename(executable))
772 >                    else :
773 >                        tar.add(exeWithPath,os.path.basename(executable))
774                      pass
775                  else:
776                      # the exe is from release, we'll find it on WN
# Line 729 | Line 801 | class Cmssw(JobType):
801              pa = os.environ['CRABDIR'] + '/' + 'ProdAgentApi'
802              if os.path.isdir(pa):
803                  tar.add(pa,paDir)
804 +
805 +            ### FEDE FOR DBS PUBLICATION
806 +            ## Add PRODCOMMON dir to tar
807 +            prodcommonDir = 'ProdCommon'
808 +            prodcommonPath = os.environ['CRABDIR'] + '/' + 'ProdCommon'
809 +            if os.path.isdir(prodcommonPath):
810 +                tar.add(prodcommonPath,prodcommonDir)
811 +            #############################    
812          
813              common.logger.debug(5,"Files added to "+self.tgzNameWithPath+" : "+str(tar.getnames()))
814              tar.close()
# Line 754 | Line 834 | class Cmssw(JobType):
834          
835          return
836          
837 +    def additionalInputFileTgz(self):
838 +        """
839 +        Put all additional files into a tar ball and return its name
840 +        """
841 +        import tarfile
842 +        tarName=  common.work_space.pathForTgz()+'share/'+self.additional_tgz_name
843 +        tar = tarfile.open(tarName, "w:gz")
844 +        for file in self.additional_inbox_files:
845 +            tar.add(file,string.split(file,'/')[-1])
846 +        common.logger.debug(5,"Files added to "+self.additional_tgz_name+" : "+str(tar.getnames()))
847 +        tar.close()
848 +        return tarName
849 +
850      def wsSetupEnvironment(self, nj):
851          """
852          Returns part of a job script which prepares
# Line 765 | Line 858 | class Cmssw(JobType):
858          ## OLI_Daniele at this level  middleware already known
859  
860          txt += 'if [ $middleware == LCG ]; then \n'
861 +        txt += '    echo "### First set SCRAM ARCH and BUILD_ARCH to ' + self.executable_arch + ' ###"\n'
862 +        txt += '    export SCRAM_ARCH='+self.executable_arch+'\n'
863 +        txt += '    export BUILD_ARCH='+self.executable_arch+'\n'
864          txt += self.wsSetupCMSLCGEnvironment_()
865          txt += 'elif [ $middleware == OSG ]; then\n'
866          txt += '    WORKING_DIR=`/bin/mktemp  -d $OSG_WN_TMP/cms_XXXXXXXXXXXX`\n'
867          txt += '    echo "Created working directory: $WORKING_DIR"\n'
868          txt += '    if [ ! -d $WORKING_DIR ] ;then\n'
869          txt += '        echo "SET_CMS_ENV 10016 ==> OSG $WORKING_DIR could not be created on WN `hostname`"\n'
870 <        txt += '        echo "JOB_EXIT_STATUS = 10016"\n'
871 <        txt += '        echo "JobExitCode=10016" | tee -a $RUNTIME_AREA/$repo\n'
872 <        txt += '        dumpStatus $RUNTIME_AREA/$repo\n'
870 >        txt += '    echo "JOB_EXIT_STATUS = 10016"\n'
871 >        txt += '    echo "JobExitCode=10016" | tee -a $RUNTIME_AREA/$repo\n'
872 >        txt += '    dumpStatus $RUNTIME_AREA/$repo\n'
873          txt += '        rm -f $RUNTIME_AREA/$repo \n'
874          txt += '        echo "MonitorJobID=`echo $MonitorJobID`" | tee -a $RUNTIME_AREA/$repo \n'
875          txt += '        echo "MonitorID=`echo $MonitorID`" | tee -a $RUNTIME_AREA/$repo\n'
# Line 783 | Line 879 | class Cmssw(JobType):
879          txt += '    echo "Change to working directory: $WORKING_DIR"\n'
880          txt += '    cd $WORKING_DIR\n'
881          txt += self.wsSetupCMSOSGEnvironment_()
882 +        txt += '    echo "### Set SCRAM ARCH to ' + self.executable_arch + ' ###"\n'
883 +        txt += '    export SCRAM_ARCH='+self.executable_arch+'\n'
884          txt += 'fi\n'
885  
886          # Prepare JobType-specific part
# Line 805 | Line 903 | class Cmssw(JobType):
903          txt += '        cd $RUNTIME_AREA\n'
904          txt += '        /bin/rm -rf $WORKING_DIR\n'
905          txt += '        if [ -d $WORKING_DIR ] ;then\n'
906 <        txt += '            echo "SET_CMS_ENV 10018 ==> OSG $WORKING_DIR could not be deleted on WN `hostname` after CMSSW CMSSW_0_6_1 not found on `hostname`"\n'
907 <        txt += '            echo "JOB_EXIT_STATUS = 10018"\n'
908 <        txt += '            echo "JobExitCode=10018" | tee -a $RUNTIME_AREA/$repo\n'
909 <        txt += '            dumpStatus $RUNTIME_AREA/$repo\n'
906 >        txt += '            echo "SET_CMS_ENV 10018 ==> OSG $WORKING_DIR could not be deleted on WN `hostname` after CMSSW CMSSW_0_6_1 not found on `hostname`"\n'
907 >        txt += '            echo "JOB_EXIT_STATUS = 10018"\n'
908 >        txt += '            echo "JobExitCode=10018" | tee -a $RUNTIME_AREA/$repo\n'
909 >        txt += '            dumpStatus $RUNTIME_AREA/$repo\n'
910          txt += '            rm -f $RUNTIME_AREA/$repo \n'
911          txt += '            echo "MonitorJobID=`echo $MonitorJobID`" | tee -a $RUNTIME_AREA/$repo \n'
912          txt += '            echo "MonitorID=`echo $MonitorID`" | tee -a $RUNTIME_AREA/$repo\n'
# Line 817 | Line 915 | class Cmssw(JobType):
915          txt += '   exit 1 \n'
916          txt += 'fi \n'
917          txt += 'echo "CMSSW_VERSION =  '+self.version+'"\n'
820 –        txt += 'export SCRAM_ARCH='+self.executable_arch+'\n'
918          txt += 'cd '+self.version+'\n'
919 +        ########## FEDE FOR DBS2 ######################
920 +        txt += 'SOFTWARE_DIR=`pwd`\n'
921 +        txt += 'echo SOFTWARE_DIR=$SOFTWARE_DIR \n'
922 +        ###############################################
923          ### needed grep for bug in scramv1 ###
924          txt += scram+' runtime -sh\n'
925          txt += 'eval `'+scram+' runtime -sh | grep -v SCRAMRT_LSB_JOBNAME`\n'
# Line 844 | Line 945 | class Cmssw(JobType):
945          txt += '        cd $RUNTIME_AREA\n'
946          txt += '        /bin/rm -rf $WORKING_DIR\n'
947          txt += '        if [ -d $WORKING_DIR ] ;then\n'
948 <        txt += '            echo "SET_EXE_ENV 50114 ==> OSG $WORKING_DIR could not be deleted on WN `hostname` after Too few arguments for CRAB job wrapper"\n'
949 <        txt += '            echo "JOB_EXIT_STATUS = 50114"\n'
950 <        txt += '            echo "JobExitCode=50114" | tee -a $RUNTIME_AREA/$repo\n'
951 <        txt += '            dumpStatus $RUNTIME_AREA/$repo\n'
948 >        txt += '            echo "SET_EXE_ENV 50114 ==> OSG $WORKING_DIR could not be deleted on WN `hostname` after Too few arguments for CRAB job wrapper"\n'
949 >        txt += '            echo "JOB_EXIT_STATUS = 50114"\n'
950 >        txt += '            echo "JobExitCode=50114" | tee -a $RUNTIME_AREA/$repo\n'
951 >        txt += '            dumpStatus $RUNTIME_AREA/$repo\n'
952          txt += '            rm -f $RUNTIME_AREA/$repo \n'
953          txt += '            echo "MonitorJobID=`echo $MonitorJobID`" | tee -a $RUNTIME_AREA/$repo \n'
954          txt += '            echo "MonitorID=`echo $MonitorID`" | tee -a $RUNTIME_AREA/$repo\n'
# Line 859 | Line 960 | class Cmssw(JobType):
960  
961          # Prepare job-specific part
962          job = common.job_list[nj]
963 +        ### FEDE FOR DBS OUTPUT PUBLICATION
964 +        if (self.datasetPath):
965 +            txt += '\n'
966 +            txt += 'DatasetPath='+self.datasetPath+'\n'
967 +
968 +            datasetpath_split = self.datasetPath.split("/")
969 +            
970 +            txt += 'PrimaryDataset='+datasetpath_split[1]+'\n'
971 +            txt += 'DataTier='+datasetpath_split[2]+'\n'
972 +            #txt += 'ProcessedDataset='+datasetpath_split[3]+'\n'
973 +            txt += 'ApplicationFamily=cmsRun\n'
974 +
975 +        else:
976 +            txt += 'DatasetPath=MCDataTier\n'
977 +            txt += 'PrimaryDataset=null\n'
978 +            txt += 'DataTier=null\n'
979 +            #txt += 'ProcessedDataset=null\n'
980 +            txt += 'ApplicationFamily=MCDataTier\n'
981          if self.pset != None: #CarlosDaniele
982              pset = os.path.basename(job.configFilename())
983              txt += '\n'
984 +            txt += 'cp  $RUNTIME_AREA/'+pset+' .\n'
985              if (self.datasetPath): # standard job
986                  #txt += 'InputFiles=$2\n'
987                  txt += 'InputFiles=${args[1]}\n'
988                  txt += 'MaxEvents=${args[2]}\n'
989                  txt += 'SkipEvents=${args[3]}\n'
990                  txt += 'echo "Inputfiles:<$InputFiles>"\n'
991 <                txt += 'sed "s#{\'INPUT\'}#$InputFiles#" $RUNTIME_AREA/'+pset+' > pset_tmp_1.cfg\n'
991 >                txt += 'sed "s#{\'INPUT\'}#$InputFiles#" '+pset+' > tmp && mv -f tmp '+pset+'\n'
992                  txt += 'echo "MaxEvents:<$MaxEvents>"\n'
993 <                txt += 'sed "s#INPUTMAXEVENTS#$MaxEvents#" pset_tmp_1.cfg > pset_tmp_2.cfg\n'
993 >                txt += 'sed "s#INPUTMAXEVENTS#$MaxEvents#" '+pset+' > tmp && mv -f tmp '+pset+'\n'
994                  txt += 'echo "SkipEvents:<$SkipEvents>"\n'
995 <                txt += 'sed "s#INPUTSKIPEVENTS#$SkipEvents#" pset_tmp_2.cfg > pset.cfg\n'
995 >                txt += 'sed "s#INPUTSKIPEVENTS#$SkipEvents#" '+pset+' > tmp && mv -f tmp '+pset+'\n'
996              else:  # pythia like job
997 <                if (self.sourceSeed):
998 <                    txt += 'FirstRun=${args[1]}\n'
997 >                seedIndex=1
998 >                if (self.firstRun):
999 >                    txt += 'FirstRun=${args['+str(seedIndex)+']}\n'
1000                      txt += 'echo "FirstRun: <$FirstRun>"\n'
1001 <                    txt += 'sed "s#\<INPUTFIRSTRUN\>#$FirstRun#" $RUNTIME_AREA/'+pset+' > tmp_1.cfg\n'
1002 <                else:
1003 <                    txt += '# Copy untouched pset\n'
883 <                    txt += 'cp $RUNTIME_AREA/'+pset+' tmp_1.cfg\n'
1001 >                    txt += 'sed "s#\<INPUTFIRSTRUN\>#$FirstRun#" '+pset+' > tmp && mv -f tmp '+pset+'\n'
1002 >                    seedIndex=seedIndex+1
1003 >
1004                  if (self.sourceSeed):
1005 < #                    txt += 'Seed=$2\n'
1006 <                    txt += 'Seed=${args[2]}\n'
1007 <                    txt += 'echo "Seed: <$Seed>"\n'
1008 <                    txt += 'sed "s#\<INPUT\>#$Seed#" tmp_1.cfg > tmp_2.cfg\n'
1005 >                    txt += 'Seed=${args['+str(seedIndex)+']}\n'
1006 >                    txt += 'sed "s#\<INPUT\>#$Seed#" '+pset+' > tmp && mv -f tmp '+pset+'\n'
1007 >                    seedIndex=seedIndex+1
1008 >                    ## the following seeds are not always present
1009                      if (self.sourceSeedVtx):
1010 < #                        txt += 'VtxSeed=$3\n'
891 <                        txt += 'VtxSeed=${args[3]}\n'
1010 >                        txt += 'VtxSeed=${args['+str(seedIndex)+']}\n'
1011                          txt += 'echo "VtxSeed: <$VtxSeed>"\n'
1012 <                        txt += 'sed "s#INPUTVTX#$VtxSeed#" tmp_2.cfg > pset.cfg\n'
1013 <                    else:
1014 <                        txt += 'mv tmp_2.cfg pset.cfg\n'
1015 <                else:
1016 <                    txt += 'mv tmp_1.cfg pset.cfg\n'
1017 <                   # txt += '# Copy untouched pset\n'
1018 <                   # txt += 'cp $RUNTIME_AREA/'+pset+' pset.cfg\n'
1019 <
1012 >                        txt += 'sed "s#\<INPUTVTX\>#$VtxSeed#" '+pset+' > tmp && mv -f tmp '+pset+'\n'
1013 >                        seedIndex += 1
1014 >                    if (self.sourceSeedG4):
1015 >                        txt += 'G4Seed=${args['+str(seedIndex)+']}\n'
1016 >                        txt += 'echo "G4Seed: <$G4Seed>"\n'
1017 >                        txt += 'sed "s#\<INPUTG4\>#$G4Seed#" '+pset+' > tmp && mv -f tmp '+pset+'\n'
1018 >                        seedIndex += 1
1019 >                    if (self.sourceSeedMix):
1020 >                        txt += 'mixSeed=${args['+str(seedIndex)+']}\n'
1021 >                        txt += 'echo "MixSeed: <$mixSeed>"\n'
1022 >                        txt += 'sed "s#\<INPUTMIX\>#$mixSeed#" '+pset+' > tmp && mv -f tmp '+pset+'\n'
1023 >                        seedIndex += 1
1024 >                    pass
1025 >                pass
1026 >            txt += 'mv -f '+pset+' pset.cfg\n'
1027  
1028          if len(self.additional_inbox_files) > 0:
1029 <            for file in self.additional_inbox_files:
1030 <                relFile = file.split("/")[-1]
1031 <                txt += 'if [ -e $RUNTIME_AREA/'+relFile+' ] ; then\n'
906 <                txt += '   cp $RUNTIME_AREA/'+relFile+' .\n'
907 <                txt += '   chmod +x '+relFile+'\n'
908 <                txt += 'fi\n'
1029 >            txt += 'if [ -e $RUNTIME_AREA/'+self.additional_tgz_name+' ] ; then\n'
1030 >            txt += '  tar xzvf $RUNTIME_AREA/'+self.additional_tgz_name+'\n'
1031 >            txt += 'fi\n'
1032              pass
1033  
1034          if self.pset != None: #CarlosDaniele
# Line 916 | Line 1039 | class Cmssw(JobType):
1039              txt += 'cat pset.cfg\n'
1040              txt += 'echo "****** end pset.cfg ********"\n'
1041              txt += '\n'
1042 +            ### FEDE FOR DBS OUTPUT PUBLICATION
1043 +            txt += 'PSETHASH=`EdmConfigHash < pset.cfg` \n'
1044 +            txt += 'echo "PSETHASH = $PSETHASH" \n'
1045 +            ##############
1046 +            txt += '\n'
1047              # txt += 'echo "***** cat pset1.cfg *********"\n'
1048              # txt += 'cat pset1.cfg\n'
1049              # txt += 'echo "****** end pset1.cfg ********"\n'
# Line 957 | Line 1085 | class Cmssw(JobType):
1085              txt += '   echo "Successful untar" \n'
1086              txt += 'fi \n'
1087              txt += '\n'
1088 <            txt += 'echo "Include ProdAgentApi in PYTHONPATH"\n'
1088 >            txt += 'echo "Include ProdAgentApi and PRODCOMMON in PYTHONPATH"\n'
1089              txt += 'if [ -z "$PYTHONPATH" ]; then\n'
1090 <            txt += '   export PYTHONPATH=ProdAgentApi\n'
1090 >            #### FEDE FOR DBS OUTPUT PUBLICATION
1091 >            txt += '   export PYTHONPATH=$SOFTWARE_DIR/ProdAgentApi:$SOFTWARE_DIR/ProdCommon\n'
1092 >            #txt += '   export PYTHONPATH=`pwd`/ProdAgentApi:`pwd`/ProdCommon\n'
1093 >            #txt += '   export PYTHONPATH=ProdAgentApi\n'
1094              txt += 'else\n'
1095 <            txt += '   export PYTHONPATH=ProdAgentApi:${PYTHONPATH}\n'
1095 >            txt += '   export PYTHONPATH=$SOFTWARE_DIR/ProdAgentApi:$SOFTWARE_DIR/ProdCommon:${PYTHONPATH}\n'
1096 >            #txt += '   export PYTHONPATH=`pwd`/ProdAgentApi:`pwd`/ProdCommon:${PYTHONPATH}\n'
1097 >            #txt += '   export PYTHONPATH=ProdAgentApi:${PYTHONPATH}\n'
1098 >            txt += 'echo "PYTHONPATH=$PYTHONPATH"\n'
1099 >            ###################  
1100              txt += 'fi\n'
1101              txt += '\n'
1102  
# Line 984 | Line 1119 | class Cmssw(JobType):
1119      def executableArgs(self):
1120          if self.scriptExe:#CarlosDaniele
1121              return   self.scriptExe + " $NJob"
1122 <        else:
1123 <            return " -p pset.cfg"
1122 >        else:
1123 >            # if >= CMSSW_1_5_X, add -e
1124 >            version_array = self.scram.getSWVersion().split('_')
1125 >            major = 0
1126 >            minor = 0
1127 >            try:
1128 >                major = int(version_array[1])
1129 >                minor = int(version_array[2])
1130 >            except:
1131 >                msg = "Cannot parse CMSSW version string: " + "_".join(version_array) + " for major and minor release number!"  
1132 >                raise CrabException(msg)
1133 >            if major >= 1 and minor >= 5 :
1134 >                return " -e -p pset.cfg"
1135 >            else:
1136 >                return " -p pset.cfg"
1137  
1138      def inputSandbox(self, nj):
1139          """
# Line 1003 | Line 1151 | class Cmssw(JobType):
1151          if not self.pset is None:
1152              inp_box.append(common.work_space.pathForTgz() + 'job/' + self.configFilename())
1153          ## additional input files
1154 <        for file in self.additional_inbox_files:
1155 <            inp_box.append(file)
1154 >        tgz = self.additionalInputFileTgz()
1155 >        inp_box.append(tgz)
1156          return inp_box
1157  
1158      def outputSandbox(self, nj):
# Line 1038 | Line 1186 | class Cmssw(JobType):
1186              output_file_num = self.numberFile_(fileWithSuffix, '$NJob')
1187              txt += '\n'
1188              txt += '# check output file\n'
1189 <            txt += 'ls '+fileWithSuffix+'\n'
1190 <            txt += 'ls_result=$?\n'
1191 <            txt += 'if [ $ls_result -ne 0 ] ; then\n'
1192 <            txt += '   echo "ERROR: Problem with output file"\n'
1189 >            # txt += 'ls '+fileWithSuffix+'\n'
1190 >            # txt += 'ls_result=$?\n'
1191 >            txt += 'if [ -e ./'+fileWithSuffix+' ] ; then\n'
1192 >            ###### FEDE FOR OUTPUT DATA PUBLICATION ########
1193 >            txt += '    mv '+fileWithSuffix+' $RUNTIME_AREA\n'
1194 >            txt += '    cp $RUNTIME_AREA/'+fileWithSuffix+' $RUNTIME_AREA/'+output_file_num+'\n'
1195 >            ################################################
1196 >            txt += 'else\n'
1197 >            txt += '    exit_status=60302\n'
1198 >            txt += '    echo "ERROR: Problem with output file '+fileWithSuffix+'"\n'
1199 >            ############# FEDE ADDED CHECK FOR OUTPUT #############
1200 >            ## MATTY's FIX: the exit option was interrupting the execution
1201 >            if fileWithSuffix in self.output_file:
1202 >                txt += '    echo "JOB_EXIT_STATUS = $exit_status"\n'
1203 >                txt += '    # exit $exit_status\n'
1204 >            #######################################################    
1205              if common.scheduler.boss_scheduler_name == 'condor_g':
1206                  txt += '    if [ $middleware == OSG ]; then \n'
1207                  txt += '        echo "prepare dummy output file"\n'
1208                  txt += '        echo "Processing of job output failed" > $RUNTIME_AREA/'+output_file_num+'\n'
1209                  txt += '    fi \n'
1050 –            txt += 'else\n'
1051 –            txt += '   cp '+fileWithSuffix+' $RUNTIME_AREA/'+output_file_num+'\n'
1210              txt += 'fi\n'
1211 <      
1211 >        file_list = []
1212 >        for fileWithSuffix in (self.output_file):
1213 >             file_list.append(self.numberFile_(fileWithSuffix, '$NJob'))
1214 >            
1215 >        txt += 'file_list="'+string.join(file_list,' ')+'"\n'
1216          txt += 'cd $RUNTIME_AREA\n'
1055 –        txt += 'cd $RUNTIME_AREA\n'
1056 –        ### OLI_DANIELE
1057 –        txt += 'if [ $middleware == OSG ]; then\n'  
1058 –        txt += '    cd $RUNTIME_AREA\n'
1059 –        txt += '    echo "Remove working directory: $WORKING_DIR"\n'
1060 –        txt += '    /bin/rm -rf $WORKING_DIR\n'
1061 –        txt += '    if [ -d $WORKING_DIR ] ;then\n'
1062 –        txt += '        echo "SET_EXE 60999 ==> OSG $WORKING_DIR could not be deleted on WN `hostname` after cleanup of WN"\n'
1063 –        txt += '        echo "JOB_EXIT_STATUS = 60999"\n'
1064 –        txt += '        echo "JobExitCode=60999" | tee -a $RUNTIME_AREA/$repo\n'
1065 –        txt += '        dumpStatus $RUNTIME_AREA/$repo\n'
1066 –        txt += '        rm -f $RUNTIME_AREA/$repo \n'
1067 –        txt += '        echo "MonitorJobID=`echo $MonitorJobID`" | tee -a $RUNTIME_AREA/$repo \n'
1068 –        txt += '        echo "MonitorID=`echo $MonitorID`" | tee -a $RUNTIME_AREA/$repo\n'
1069 –        txt += '    fi\n'
1070 –        txt += 'fi\n'
1071 –        txt += '\n'
1072 –
1073 –        file_list = ''
1074 –        ## Add to filelist only files to be possibly copied to SE
1075 –        for fileWithSuffix in self.output_file:
1076 –            output_file_num = self.numberFile_(fileWithSuffix, '$NJob')
1077 –            file_list=file_list+output_file_num+' '
1078 –        file_list=file_list[:-1]
1079 –        txt += 'file_list="'+file_list+'"\n'
1080 –
1217          return txt
1218  
1219      def numberFile_(self, file, txt):
# Line 1088 | Line 1224 | class Cmssw(JobType):
1224          # take away last extension
1225          name = p[0]
1226          for x in p[1:-1]:
1227 <           name=name+"."+x
1227 >            name=name+"."+x
1228          # add "_txt"
1229          if len(p)>1:
1230 <          ext = p[len(p)-1]
1231 <          result = name + '_' + txt + "." + ext
1230 >            ext = p[len(p)-1]
1231 >            result = name + '_' + txt + "." + ext
1232          else:
1233 <          result = name + '_' + txt
1233 >            result = name + '_' + txt
1234          
1235          return result
1236  
# Line 1107 | Line 1243 | class Cmssw(JobType):
1243              req='Member("VO-cms-' + \
1244                   self.version + \
1245                   '", other.GlueHostApplicationSoftwareRunTimeEnvironment)'
1246 +        ## SL add requirement for OS version only if SL4
1247 +        #reSL4 = re.compile( r'slc4' )
1248 +        if self.executable_arch: # and reSL4.search(self.executable_arch):
1249 +            req+=' && Member("VO-cms-' + \
1250 +                 self.executable_arch + \
1251 +                 '", other.GlueHostApplicationSoftwareRunTimeEnvironment)'
1252  
1253          req = req + ' && (other.GlueHostNetworkAdapterOutboundIP)'
1254  
# Line 1126 | Line 1268 | class Cmssw(JobType):
1268          txt += '   echo "### SETUP CMS OSG  ENVIRONMENT ###"\n'
1269          txt += '   if [ -f $GRID3_APP_DIR/cmssoft/cmsset_default.sh ] ;then\n'
1270          txt += '      # Use $GRID3_APP_DIR/cmssoft/cmsset_default.sh to setup cms software\n'
1271 +        txt += '       export SCRAM_ARCH='+self.executable_arch+'\n'
1272          txt += '       source $GRID3_APP_DIR/cmssoft/cmsset_default.sh '+self.version+'\n'
1273          txt += '   elif [ -f $OSG_APP/cmssoft/cms/cmsset_default.sh ] ;then\n'
1274          txt += '      # Use $OSG_APP/cmssoft/cms/cmsset_default.sh to setup cms software\n'
1275 +        txt += '       export SCRAM_ARCH='+self.executable_arch+'\n'
1276          txt += '       source $OSG_APP/cmssoft/cms/cmsset_default.sh '+self.version+'\n'
1277          txt += '   else\n'
1278          txt += '       echo "SET_CMS_ENV 10020 ==> ERROR $GRID3_APP_DIR/cmssoft/cmsset_default.sh and $OSG_APP/cmssoft/cms/cmsset_default.sh file not found"\n'
# Line 1144 | Line 1288 | class Cmssw(JobType):
1288          txt += '       cd $RUNTIME_AREA\n'
1289          txt += '       /bin/rm -rf $WORKING_DIR\n'
1290          txt += '       if [ -d $WORKING_DIR ] ;then\n'
1291 <        txt += '            echo "SET_CMS_ENV 10017 ==> OSG $WORKING_DIR could not be deleted on WN `hostname` after $GRID3_APP_DIR/cmssoft/cmsset_default.sh and $OSG_APP/cmssoft/cms/cmsset_default.sh file not found"\n'
1292 <        txt += '            echo "JOB_EXIT_STATUS = 10017"\n'
1293 <        txt += '            echo "JobExitCode=10017" | tee -a $RUNTIME_AREA/$repo\n'
1294 <        txt += '            dumpStatus $RUNTIME_AREA/$repo\n'
1295 <        txt += '            rm -f $RUNTIME_AREA/$repo \n'
1296 <        txt += '            echo "MonitorJobID=`echo $MonitorJobID`" | tee -a $RUNTIME_AREA/$repo \n'
1297 <        txt += '            echo "MonitorID=`echo $MonitorID`" | tee -a $RUNTIME_AREA/$repo\n'
1291 >        txt += '           echo "SET_CMS_ENV 10017 ==> OSG $WORKING_DIR could not be deleted on WN `hostname` after $GRID3_APP_DIR/cmssoft/cmsset_default.sh and $OSG_APP/cmssoft/cms/cmsset_default.sh file not found"\n'
1292 >        txt += '           echo "JOB_EXIT_STATUS = 10017"\n'
1293 >        txt += '           echo "JobExitCode=10017" | tee -a $RUNTIME_AREA/$repo\n'
1294 >        txt += '           dumpStatus $RUNTIME_AREA/$repo\n'
1295 >        txt += '           rm -f $RUNTIME_AREA/$repo \n'
1296 >        txt += '           echo "MonitorJobID=`echo $MonitorJobID`" | tee -a $RUNTIME_AREA/$repo \n'
1297 >        txt += '           echo "MonitorID=`echo $MonitorID`" | tee -a $RUNTIME_AREA/$repo\n'
1298          txt += '       fi\n'
1299          txt += '\n'
1300          txt += '       exit 1\n'
# Line 1209 | Line 1353 | class Cmssw(JobType):
1353          txt += '   echo "### END SETUP CMS LCG ENVIRONMENT ###"\n'
1354          return txt
1355  
1356 +    ### FEDE FOR DBS OUTPUT PUBLICATION
1357 +    def modifyReport(self, nj):
1358 +        """
1359 +        insert the part of the script that modifies the FrameworkJob Report
1360 +        """
1361 +
1362 +        txt = ''
1363 +        try:
1364 +            publish_data = int(self.cfg_params['USER.publish_data'])          
1365 +        except KeyError:
1366 +            publish_data = 0
1367 +        if (publish_data == 1):  
1368 +            txt += 'echo "Modify Job Report" \n'
1369 +            #txt += 'chmod a+x $RUNTIME_AREA/'+self.version+'/ProdAgentApi/FwkJobRep/ModifyJobReport.py\n'
1370 +            ################ FEDE FOR DBS2 #############################################
1371 +            txt += 'chmod a+x $SOFTWARE_DIR/ProdAgentApi/FwkJobRep/ModifyJobReport.py\n'
1372 +            #############################################################################
1373 +            #try:
1374 +            #    publish_data = int(self.cfg_params['USER.publish_data'])          
1375 +            #except KeyError:
1376 +            #    publish_data = 0
1377 +
1378 +            txt += 'if [ -z "$SE" ]; then\n'
1379 +            txt += '    SE="" \n'
1380 +            txt += 'fi \n'
1381 +            txt += 'if [ -z "$SE_PATH" ]; then\n'
1382 +            txt += '    SE_PATH="" \n'
1383 +            txt += 'fi \n'
1384 +            txt += 'echo "SE = $SE"\n'
1385 +            txt += 'echo "SE_PATH = $SE_PATH"\n'
1386 +
1387 +        #if (publish_data == 1):  
1388 +            #processedDataset = self.cfg_params['USER.processed_datasetname']
1389 +            processedDataset = self.cfg_params['USER.publish_data_name']
1390 +            txt += 'ProcessedDataset='+processedDataset+'\n'
1391 +            #### LFN=/store/user/<user>/processedDataset_PSETHASH
1392 +            txt += 'if [ "$SE_PATH" == "" ]; then\n'
1393 +            #### FEDE: added slash in LFN ##############
1394 +            txt += '    FOR_LFN=/copy_problems/ \n'
1395 +            txt += 'else \n'
1396 +            txt += '    tmp=`echo $SE_PATH | awk -F \'store\' \'{print$2}\'` \n'
1397 +            #####  FEDE TO BE CHANGED, BECAUSE STORE IS HARDCODED!!!! ########
1398 +            txt += '    FOR_LFN=/store$tmp \n'
1399 +            txt += 'fi \n'
1400 +            txt += 'echo "ProcessedDataset = $ProcessedDataset"\n'
1401 +            txt += 'echo "FOR_LFN = $FOR_LFN" \n'
1402 +            txt += 'echo "CMSSW_VERSION = $CMSSW_VERSION"\n\n'
1403 +            #txt += 'echo "$RUNTIME_AREA/'+self.version+'/ProdAgentApi/FwkJobRep/ModifyJobReport.py crab_fjr_$NJob.xml $NJob $FOR_LFN $PrimaryDataset $DataTier $ProcessedDataset $ApplicationFamily $executable $CMSSW_VERSION $PSETHASH $SE $SE_PATH"\n'
1404 +            txt += 'echo "$SOFTWARE_DIR/ProdAgentApi/FwkJobRep/ModifyJobReport.py crab_fjr_$NJob.xml $NJob $FOR_LFN $PrimaryDataset $DataTier $ProcessedDataset $ApplicationFamily $executable $CMSSW_VERSION $PSETHASH $SE $SE_PATH"\n'
1405 +            txt += '$SOFTWARE_DIR/ProdAgentApi/FwkJobRep/ModifyJobReport.py crab_fjr_$NJob.xml $NJob $FOR_LFN $PrimaryDataset $DataTier $ProcessedDataset $ApplicationFamily $executable $CMSSW_VERSION $PSETHASH $SE $SE_PATH\n'
1406 +            #txt += '$RUNTIME_AREA/'+self.version+'/ProdAgentApi/FwkJobRep/ModifyJobReport.py crab_fjr_$NJob.xml $NJob $FOR_LFN $PrimaryDataset $DataTier $ProcessedDataset $ApplicationFamily $executable $CMSSW_VERSION $PSETHASH $SE $SE_PATH\n'
1407 +      
1408 +            txt += 'modifyReport_result=$?\n'
1409 +            txt += 'echo modifyReport_result = $modifyReport_result\n'
1410 +            txt += 'if [ $modifyReport_result -ne 0 ]; then\n'
1411 +            txt += '    exit_status=1\n'
1412 +            txt += '    echo "ERROR: Problem with ModifyJobReport"\n'
1413 +            txt += 'else\n'
1414 +            txt += '    mv NewFrameworkJobReport.xml crab_fjr_$NJob.xml\n'
1415 +            txt += 'fi\n'
1416 +        else:
1417 +            txt += 'echo "no data publication required"\n'
1418 +            #txt += 'ProcessedDataset=no_data_to_publish \n'
1419 +            #### FEDE: added slash in LFN ##############
1420 +            #txt += 'FOR_LFN=/local/ \n'
1421 +            #txt += 'echo "ProcessedDataset = $ProcessedDataset"\n'
1422 +            #txt += 'echo "FOR_LFN = $FOR_LFN" \n'
1423 +        return txt
1424 +
1425 +    def cleanEnv(self):
1426 +        ### OLI_DANIELE
1427 +        txt = ''
1428 +        txt += 'if [ $middleware == OSG ]; then\n'  
1429 +        txt += '    cd $RUNTIME_AREA\n'
1430 +        txt += '    echo "Remove working directory: $WORKING_DIR"\n'
1431 +        txt += '    /bin/rm -rf $WORKING_DIR\n'
1432 +        txt += '    if [ -d $WORKING_DIR ] ;then\n'
1433 +        txt += '              echo "SET_EXE 60999 ==> OSG $WORKING_DIR could not be deleted on WN `hostname` after cleanup of WN"\n'
1434 +        txt += '              echo "JOB_EXIT_STATUS = 60999"\n'
1435 +        txt += '              echo "JobExitCode=60999" | tee -a $RUNTIME_AREA/$repo\n'
1436 +        txt += '              dumpStatus $RUNTIME_AREA/$repo\n'
1437 +        txt += '        rm -f $RUNTIME_AREA/$repo \n'
1438 +        txt += '        echo "MonitorJobID=`echo $MonitorJobID`" | tee -a $RUNTIME_AREA/$repo \n'
1439 +        txt += '        echo "MonitorID=`echo $MonitorID`" | tee -a $RUNTIME_AREA/$repo\n'
1440 +        txt += '    fi\n'
1441 +        txt += 'fi\n'
1442 +        txt += '\n'
1443 +        return txt
1444 +
1445      def setParam_(self, param, value):
1446          self._params[param] = value
1447  
# Line 1221 | Line 1454 | class Cmssw(JobType):
1454      def getTaskid(self):
1455          return self._taskId
1456  
1224 – #######################################################################
1457      def uniquelist(self, old):
1458          """
1459          remove duplicates from a list
# Line 1230 | Line 1462 | class Cmssw(JobType):
1462          for e in old:
1463              nd[e]=0
1464          return nd.keys()
1465 +
1466 +
1467 +    def checkOut(self, limit):
1468 +        """
1469 +        check the dimension of the output files
1470 +        """
1471 +        txt = 'echo "*****************************************"\n'
1472 +        txt += 'echo "** Starting output sandbox limit check **"\n'
1473 +        txt += 'echo "*****************************************"\n'
1474 +        allOutFiles = ""
1475 +        listOutFiles = []
1476 +        for fileOut in (self.output_file+self.output_file_sandbox):
1477 +             if fileOut.find('crab_fjr') == -1:
1478 +                 allOutFiles = allOutFiles + " " + self.numberFile_(fileOut, '$NJob')
1479 +                 listOutFiles.append(self.numberFile_(fileOut, '$NJob'))
1480 +        txt += 'echo "OUTPUT files: '+str(allOutFiles)+'";\n'
1481 +        txt += 'ls -gGhrta;\n'
1482 +        txt += 'sum=0;\n'
1483 +        txt += 'for file in '+str(allOutFiles)+' ; do\n'
1484 +        txt += '    if [ -e $file ]; then\n'
1485 +        txt += '        tt=`ls -gGrta $file | awk \'{ print $3 }\'`\n'
1486 +        txt += '        sum=`expr $sum + $tt`\n'
1487 +        txt += '    else\n'
1488 +        txt += '        echo "WARNING: output file $file not found!"\n'
1489 +        txt += '    fi\n'
1490 +        txt += 'done\n'
1491 +        txt += 'echo "Total Output dimension: $sum";\n'
1492 +        txt += 'limit='+str(limit)+';\n'
1493 +        txt += 'echo "OUTPUT FILES LIMIT SET TO: $limit";\n'
1494 +        txt += 'if [ $limit -lt $sum ]; then\n'
1495 +        txt += '    echo "WARNING: output files have to big size - something will be lost;"\n'
1496 +        txt += '    echo "         checking the output file sizes..."\n'
1497 +        """
1498 +        txt += '    dim=0;\n'
1499 +        txt += '    exclude=0;\n'
1500 +        txt += '    for files in '+str(allOutFiles)+' ; do\n'
1501 +        txt += '        sumTemp=0;\n'
1502 +        txt += '        for file2 in '+str(allOutFiles)+' ; do\n'
1503 +        txt += '            if [ $file != $file2 ]; then\n'
1504 +        txt += '                tt=`ls -gGrta $file2 | awk \'{ print $3 }\';`\n'
1505 +        txt += '                sumTemp=`expr $sumTemp + $tt`;\n'
1506 +        txt += '            fi\n'
1507 +        txt += '        done\n'
1508 +        txt += '        if [ $sumTemp -lt $limit ]; then\n'
1509 +        txt += '            if [ $dim -lt $sumTemp ]; then\n'
1510 +        txt += '                dim=$sumTemp;\n'
1511 +        txt += '                exclude=$file;\n'
1512 +        txt += '            fi\n'
1513 +        txt += '        fi\n'
1514 +        txt += '    done\n'
1515 +        txt += '    echo "Dimension calculated: $dim"; echo "File to exclude: $exclude";\n'
1516 +        """
1517 +        txt += '    tot=0;\n'
1518 +        txt += '    for file2 in '+str(allOutFiles)+' ; do\n'
1519 +        txt += '        tt=`ls -gGrta $file2 | awk \'{ print $3 }\';`\n'
1520 +        txt += '        tot=`expr $tot + $tt`;\n'
1521 +        txt += '        if [ $limit -lt $tot ]; then\n'
1522 +        txt += '            tot=`expr $tot - $tt`;\n'
1523 +        txt += '            fileLast=$file;\n'
1524 +        txt += '            break;\n'
1525 +        txt += '        fi\n'
1526 +        txt += '    done\n'
1527 +        txt += '    echo "Dimension calculated: $tot"; echo "First file to exclude: $file";\n'
1528 +        txt += '    flag=0;\n'    
1529 +        txt += '    for filess in '+str(allOutFiles)+' ; do\n'
1530 +        txt += '        if [ $fileLast = $filess ]; then\n'
1531 +        txt += '            flag=1;\n'
1532 +        txt += '        fi\n'
1533 +        txt += '        if [ $flag -eq 1 ]; then\n'
1534 +        txt += '            rm -f $filess;\n'
1535 +        txt += '        fi\n'
1536 +        txt += '    done\n'
1537 +        txt += '    ls -agGhrt;\n'
1538 +        txt += '    echo "WARNING: output files are too big in dimension: can not put in the output_sandbox.";\n'
1539 +        txt += '    echo "JOB_EXIT_STATUS = 70000";\n'
1540 +        txt += '    exit_status=70000;\n'
1541 +        txt += 'else'
1542 +        txt += '    echo "Total Output dimension $sum is fine.";\n'
1543 +        txt += 'fi\n'
1544 +        txt += 'echo "*****************************************"\n'
1545 +        txt += 'echo "*** Ending output sandbox limit check ***"\n'
1546 +        txt += 'echo "*****************************************"\n'
1547 +        return txt

Diff Legend

– Removed lines
+ Added lines
< Changed lines
> Changed lines