ViewVC Help

View File | Revision Log | Show Annotations | Root Listing

root/cvsroot/COMP/CRAB/python/cms_cmssw.py

Comparing COMP/CRAB/python/cms_cmssw.py (file contents):
Revision 1.137 by slacapra, Fri Nov 16 11:09:31 2007 UTC vs.
Revision 1.196 by ewv, Wed May 28 16:46:17 2008 UTC

#	Line 5 | Line 5 | from crab_util import *
5		from BlackWhiteListParser import BlackWhiteListParser
6		import common
7		import Scram
8	+	from LFNBaseName import *
9
10		import os, string, glob
11
#	Line 13 | Line 14 | class Cmssw(JobType):
14		JobType.__init__(self, 'CMSSW')
15		common.logger.debug(3,'CMSSW::__init__')
16
17	+	self.argsList = []
18	+
19		self._params = {}
20		self.cfg_params = cfg_params
18	–
21		# init BlackWhiteListParser
22		self.blackWhiteListParser = BlackWhiteListParser(cfg_params)
23
24	<	try:
23	<	self.MaxTarBallSize = float(self.cfg_params['EDG.maxtarballsize'])
24	<	except KeyError:
25	<	self.MaxTarBallSize = 9.5
24	>	self.MaxTarBallSize = float(self.cfg_params.get('EDG.maxtarballsize',9.5))
25
26		# number of jobs requested to be created, limit obj splitting
27		self.ncjobs = ncjobs
#	Line 35 | Line 34 | class Cmssw(JobType):
34		self.executable = ''
35		self.executable_arch = self.scram.getArch()
36		self.tgz_name = 'default.tgz'
38	–	self.additional_tgz_name = 'additional.tgz'
37		self.scriptName = 'CMSSW.sh'
38	<	self.pset = ''      #scrip use case Da
39	<	self.datasetPath = '' #scrip use case Da
38	>	self.pset = ''
39	>	self.datasetPath = ''
40
41		# set FJR file name
42		self.fjrFileName = 'crab_fjr.xml'
43
44		self.version = self.scram.getSWVersion()
45	<
46	<	#
47	<	# Try to block creation in case of arch/version mismatch
48	<	#
49	<
50	<	a = string.split(self.version, "_")
51	<
52	<	if int(a[1]) == 1 and (int(a[2]) < 5 and self.executable_arch.find('slc4') == 0):
53	<	msg = "Warning: You are using %s version of CMSSW  with %s architecture. \n--> Did you compile your libraries with SLC3? Otherwise you can find some problems running on SLC4 Grid nodes.\n"%(self.version, self.executable_arch)
54	<	common.logger.message(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)
45	>	version_array = self.version.split('_')
46	>	self.CMSSW_major = 0
47	>	self.CMSSW_minor = 0
48	>	self.CMSSW_patch = 0
49	>	try:
50	>	self.CMSSW_major = int(version_array[1])
51	>	self.CMSSW_minor = int(version_array[2])
52	>	self.CMSSW_patch = int(version_array[3])
53	>	except:
54	>	msg = "Cannot parse CMSSW version string: " + self.version + " for major and minor release number!"
55		raise CrabException(msg)
56
61	–	common.taskDB.setDict('codeVersion',self.version)
62	–	self.setParam_('application', self.version)
63	–
57		### collect Data cards
58
59	<	try:
67	<	tmp =  cfg_params['CMSSW.datasetpath']
68	<	log.debug(6, "CMSSW::CMSSW(): datasetPath = "+tmp)
69	<	if string.lower(tmp)=='none':
70	<	self.datasetPath = None
71	<	self.selectNoInput = 1
72	<	else:
73	<	self.datasetPath = tmp
74	<	self.selectNoInput = 0
75	<	except KeyError:
59	>	if not cfg_params.has_key('CMSSW.datasetpath'):
60		msg = "Error: datasetpath not defined "
61		raise CrabException(msg)
62	<
63	<	# ML monitoring
64	<	# split dataset path style: /PreProdR3Minbias/SIM/GEN-SIM
65	<	if not self.datasetPath:
66	<	self.setParam_('dataset', 'None')
83	<	self.setParam_('owner', 'None')
62	>	tmp =  cfg_params['CMSSW.datasetpath']
63	>	log.debug(6, "CMSSW::CMSSW(): datasetPath = "+tmp)
64	>	if string.lower(tmp)=='none':
65	>	self.datasetPath = None
66	>	self.selectNoInput = 1
67		else:
68	<	try:
69	<	datasetpath_split = self.datasetPath.split("/")
87	<	# standard style
88	<	self.setParam_('datasetFull', self.datasetPath)
89	<	self.setParam_('dataset', datasetpath_split[1])
90	<	self.setParam_('owner', datasetpath_split[2])
91	<	except:
92	<	self.setParam_('dataset', self.datasetPath)
93	<	self.setParam_('owner', self.datasetPath)
94	<
95	<	self.setTaskid_()
96	<	self.setParam_('taskId', self.cfg_params['taskId'])
68	>	self.datasetPath = tmp
69	>	self.selectNoInput = 0
70
71		self.dataTiers = []
72
73	+	self.debug_pset = cfg_params.get('USER.debug_pset',False)
74	+
75		## now the application
76	<	try:
77	<	self.executable = cfg_params['CMSSW.executable']
103	<	self.setParam_('exe', self.executable)
104	<	log.debug(6, "CMSSW::CMSSW(): executable = "+self.executable)
105	<	msg = "Default executable cmsRun overridden. Switch to " + self.executable
106	<	log.debug(3,msg)
107	<	except KeyError:
108	<	self.executable = 'cmsRun'
109	<	self.setParam_('exe', self.executable)
110	<	msg = "User executable not defined. Use cmsRun"
111	<	log.debug(3,msg)
112	<	pass
76	>	self.executable = cfg_params.get('CMSSW.executable','cmsRun')
77	>	log.debug(6, "CMSSW::CMSSW(): executable = "+self.executable)
78
79	<	try:
115	<	self.pset = cfg_params['CMSSW.pset']
116	<	log.debug(6, "Cmssw::Cmssw(): PSet file = "+self.pset)
117	<	if self.pset.lower() != 'none' :
118	<	if (not os.path.exists(self.pset)):
119	<	raise CrabException("User defined PSet file "+self.pset+" does not exist")
120	<	else:
121	<	self.pset = None
122	<	except KeyError:
79	>	if not cfg_params.has_key('CMSSW.pset'):
80		raise CrabException("PSet file missing. Cannot run cmsRun ")
81	+	self.pset = cfg_params['CMSSW.pset']
82	+	log.debug(6, "Cmssw::Cmssw(): PSet file = "+self.pset)
83	+	if self.pset.lower() != 'none' :
84	+	if (not os.path.exists(self.pset)):
85	+	raise CrabException("User defined PSet file "+self.pset+" does not exist")
86	+	else:
87	+	self.pset = None
88
89		# output files
90		## stuff which must be returned always via sandbox
#	Line 130 | Line 94 | class Cmssw(JobType):
94		self.output_file_sandbox.append(self.fjrFileName)
95
96		# other output files to be returned via sandbox or copied to SE
97	<	try:
98	<	self.output_file = []
99	<	tmp = cfg_params['CMSSW.output_file']
100	<	if tmp != '':
101	<	tmpOutFiles = string.split(cfg_params['CMSSW.output_file'],',')
102	<	log.debug(7, 'cmssw::cmssw(): output files '+str(tmpOutFiles))
103	<	for tmp in tmpOutFiles:
104	<	tmp=string.strip(tmp)
141	<	self.output_file.append(tmp)
142	<	pass
143	<	else:
144	<	log.message("No output file defined: only stdout/err and the CRAB Framework Job Report will be available\n")
97	>	self.output_file = []
98	>	tmp = cfg_params.get('CMSSW.output_file',None)
99	>	if tmp :
100	>	tmpOutFiles = string.split(tmp,',')
101	>	log.debug(7, 'cmssw::cmssw(): output files '+str(tmpOutFiles))
102	>	for tmp in tmpOutFiles:
103	>	tmp=string.strip(tmp)
104	>	self.output_file.append(tmp)
105		pass
106	<	pass
147	<	except KeyError:
106	>	else:
107		log.message("No output file defined: only stdout/err and the CRAB Framework Job Report will be available\n")
108	<	pass
108	>	pass
109
110		# script_exe file as additional file in inputSandbox
111	<	try:
112	<	self.scriptExe = cfg_params['USER.script_exe']
113	<	if self.scriptExe != '':
114	<	if not os.path.isfile(self.scriptExe):
115	<	msg ="ERROR. file "+self.scriptExe+" not found"
116	<	raise CrabException(msg)
158	<	self.additional_inbox_files.append(string.strip(self.scriptExe))
159	<	except KeyError:
160	<	self.scriptExe = ''
111	>	self.scriptExe = cfg_params.get('USER.script_exe',None)
112	>	if self.scriptExe :
113	>	if not os.path.isfile(self.scriptExe):
114	>	msg ="ERROR. file "+self.scriptExe+" not found"
115	>	raise CrabException(msg)
116	>	self.additional_inbox_files.append(string.strip(self.scriptExe))
117
162	–	#CarlosDaniele
118		if self.datasetPath == None and self.pset == None and self.scriptExe == '' :
119	<	msg ="Error. script_exe  not defined"
120	<	raise CrabException(msg)
119	>	msg ="Error. script_exe  not defined"
120	>	raise CrabException(msg)
121
122		## additional input files
123	<	try:
123	>	if cfg_params.has_key('USER.additional_input_files'):
124		tmpAddFiles = string.split(cfg_params['USER.additional_input_files'],',')
125		for tmp in tmpAddFiles:
126		tmp = string.strip(tmp)
#	Line 182 | Line 137 | class Cmssw(JobType):
137		if not os.path.exists(file):
138		raise CrabException("Additional input file not found: "+file)
139		pass
185	–	# fname = string.split(file, '/')[-1]
186	–	# storedFile = common.work_space.pathForTgz()+'share/'+fname
187	–	# shutil.copyfile(file, storedFile)
140		self.additional_inbox_files.append(string.strip(file))
141		pass
142		pass
143		common.logger.debug(5,"Additional input files: "+str(self.additional_inbox_files))
144	<	except KeyError:
193	<	pass
194	<
195	<	# files per job
196	<	try:
197	<	if (cfg_params['CMSSW.files_per_jobs']):
198	<	raise CrabException("files_per_jobs no longer supported.  Quitting.")
199	<	except KeyError:
200	<	pass
144	>	pass
145
146		## Events per job
147	<	try:
147	>	if cfg_params.has_key('CMSSW.events_per_job'):
148		self.eventsPerJob =int( cfg_params['CMSSW.events_per_job'])
149		self.selectEventsPerJob = 1
150	<	except KeyError:
150	>	else:
151		self.eventsPerJob = -1
152		self.selectEventsPerJob = 0
153
154		## number of jobs
155	<	try:
155	>	if cfg_params.has_key('CMSSW.number_of_jobs'):
156		self.theNumberOfJobs =int( cfg_params['CMSSW.number_of_jobs'])
157		self.selectNumberOfJobs = 1
158	<	except KeyError:
158	>	else:
159		self.theNumberOfJobs = 0
160		self.selectNumberOfJobs = 0
161
162	<	try:
162	>	if cfg_params.has_key('CMSSW.total_number_of_events'):
163		self.total_number_of_events = int(cfg_params['CMSSW.total_number_of_events'])
164		self.selectTotalNumberEvents = 1
165	<	except KeyError:
165	>	if self.selectNumberOfJobs  == 1:
166	>	if int(self.total_number_of_events) < int(self.theNumberOfJobs):
167	>	msg = 'Must specify at least one event per job. total_number_of_events > number_of_jobs '
168	>	raise CrabException(msg)
169	>	else:
170		self.total_number_of_events = 0
171		self.selectTotalNumberEvents = 0
172
173	<	if self.pset != None: #CarlosDaniele
173	>	if self.pset != None:
174		if ( (self.selectTotalNumberEvents + self.selectEventsPerJob + self.selectNumberOfJobs) != 2 ):
175		msg = 'Must define exactly two of total_number_of_events, events_per_job, or number_of_jobs.'
176		raise CrabException(msg)
#	Line 231 | Line 179 | class Cmssw(JobType):
179		msg = 'Must specify  number_of_jobs.'
180		raise CrabException(msg)
181
182	<	## source seed for pythia
183	<	try:
184	<	self.sourceSeed = int(cfg_params['CMSSW.pythia_seed'])
185	<	except KeyError:
186	<	self.sourceSeed = None
187	<	common.logger.debug(5,"No seed given")
188	<
189	<	try:
190	<	self.sourceSeedVtx = int(cfg_params['CMSSW.vtx_seed'])
191	<	except KeyError:
192	<	self.sourceSeedVtx = None
193	<	common.logger.debug(5,"No vertex seed given")
194	<
195	<	try:
196	<	self.sourceSeedG4 = int(cfg_params['CMSSW.g4_seed'])
197	<	except KeyError:
198	<	self.sourceSeedG4 = None
199	<	common.logger.debug(5,"No g4 sim hits seed given")
182	>	## New method of dealing with seeds
183	>	self.incrementSeeds = []
184	>	self.preserveSeeds = []
185	>	if cfg_params.has_key('CMSSW.preserve_seeds'):
186	>	tmpList = cfg_params['CMSSW.preserve_seeds'].split(',')
187	>	for tmp in tmpList:
188	>	tmp.strip()
189	>	self.preserveSeeds.append(tmp)
190	>	if cfg_params.has_key('CMSSW.increment_seeds'):
191	>	tmpList = cfg_params['CMSSW.increment_seeds'].split(',')
192	>	for tmp in tmpList:
193	>	tmp.strip()
194	>	self.incrementSeeds.append(tmp)
195	>
196	>	## Old method of dealing with seeds
197	>	## FUTURE: This is for old CMSSW and old CRAB. Can throw exceptions after a couple of CRAB releases and then
198	>	## remove
199	>	self.sourceSeed = cfg_params.get('CMSSW.pythia_seed',None)
200	>	if self.sourceSeed:
201	>	print "pythia_seed is a deprecated parameter. Use preserve_seeds or increment_seeds in the future.\n","Added to increment_seeds."
202	>	self.incrementSeeds.append('sourceSeed')
203	>	self.incrementSeeds.append('theSource')
204	>
205	>	self.sourceSeedVtx = cfg_params.get('CMSSW.vtx_seed',None)
206	>	if self.sourceSeedVtx:
207	>	print "vtx_seed is a deprecated parameter. Use preserve_seeds or increment_seeds in the future.\n","Added to increment_seeds."
208	>	self.incrementSeeds.append('VtxSmeared')
209	>
210	>	self.sourceSeedG4 = cfg_params.get('CMSSW.g4_seed',None)
211	>	if self.sourceSeedG4:
212	>	print "g4_seed is a deprecated parameter. Use preserve_seeds or increment_seeds in the future.\n","Added to increment_seeds."
213	>	self.incrementSeeds.append('g4SimHits')
214	>
215	>	self.sourceSeedMix = cfg_params.get('CMSSW.mix_seed',None)
216	>	if self.sourceSeedMix:
217	>	print "mix_seed is a deprecated parameter. Use preserve_seeds or increment_seeds in the future.\n","Added to increment_seeds."
218	>	self.incrementSeeds.append('mix')
219
220	<	try:
254	<	self.sourceSeedMix = int(cfg_params['CMSSW.mix_seed'])
255	<	except KeyError:
256	<	self.sourceSeedMix = None
257	<	common.logger.debug(5,"No mix seed given")
220	>	self.firstRun = cfg_params.get('CMSSW.first_run',None)
221
259	–	try:
260	–	self.firstRun = int(cfg_params['CMSSW.first_run'])
261	–	except KeyError:
262	–	self.firstRun = None
263	–	common.logger.debug(5,"No first run given")
222		if self.pset != None: #CarlosDaniele
223		import PsetManipulator as pp
224		PsetEdit = pp.PsetManipulator(self.pset) #Daniele Pset
225
226	+	# Copy/return
227	+
228	+	self.copy_data = int(cfg_params.get('USER.copy_data',0))
229	+	self.return_data = int(cfg_params.get('USER.return_data',0))
230	+
231		#DBSDLS-start
232		## Initialize the variables that are extracted from DBS/DLS and needed in other places of the code
233		self.maxEvents=0  # max events available   ( --> check the requested nb. of evts in Creator.py)
#	Line 277 | Line 240 | class Cmssw(JobType):
240		blockSites = self.DataDiscoveryAndLocation(cfg_params)
241		#DBSDLS-end
242
280	–	self.tgzNameWithPath = self.getTarBall(self.executable)
243
244		## Select Splitting
245		if self.selectNoInput:
246	<	if self.pset == None: #CarlosDaniele
246	>	if self.pset == None:
247		self.jobSplittingForScript()
248		else:
249		self.jobSplittingNoInput()
#	Line 289 | Line 251 | class Cmssw(JobType):
251		self.jobSplittingByBlocks(blockSites)
252
253		# modify Pset
254	<	if self.pset != None: #CarlosDaniele
254	>	if self.pset != None:
255		try:
256	<	if (self.datasetPath): # standard job
257	<	# allow to processa a fraction of events in a file
258	<	PsetEdit.inputModule("INPUTFILE")
259	<	PsetEdit.maxEvent(0)
298	<	PsetEdit.skipEvent(0)
299	<	else:  # pythia like job
300	<	PsetEdit.maxEvent(self.eventsPerJob)
301	<	if (self.firstRun):
302	<	PsetEdit.pythiaFirstRun(0)  #First Run
303	<	if (self.sourceSeed) :
304	<	PsetEdit.pythiaSeed(0)
305	<	if (self.sourceSeedVtx) :
306	<	PsetEdit.vtxSeed(0)
307	<	if (self.sourceSeedG4) :
308	<	PsetEdit.g4Seed(0)
309	<	if (self.sourceSeedMix) :
310	<	PsetEdit.mixSeed(0)
311	<	# add FrameworkJobReport to parameter-set
312	<	PsetEdit.addCrabFJR(self.fjrFileName)
256	>	# Add FrameworkJobReport to parameter-set, set max events.
257	>	# Reset later for data jobs by writeCFG which does all modifications
258	>	PsetEdit.addCrabFJR(self.fjrFileName) # FUTURE: Job report addition not needed by CMSSW>1.5
259	>	PsetEdit.maxEvent(self.eventsPerJob)
260		PsetEdit.psetWriter(self.configFilename())
261		except:
262	<	msg='Error while manipuliating ParameterSet: exiting...'
262	>	msg='Error while manipulating ParameterSet: exiting...'
263		raise CrabException(msg)
264	+	self.tgzNameWithPath = self.getTarBall(self.executable)
265
266		def DataDiscoveryAndLocation(self, cfg_params):
267
#	Line 326 | Line 274 | class Cmssw(JobType):
274		## Contact the DBS
275		common.logger.message("Contacting Data Discovery Services ...")
276		try:
329	–
277		self.pubdata=DataDiscovery.DataDiscovery(datasetPath, cfg_params)
278		self.pubdata.fetchDBSInfo()
279
#	Line 345 | Line 292 | class Cmssw(JobType):
292		self.eventsbyfile=self.pubdata.getEventsPerFile()
293
294		## get max number of events
295	<	self.maxEvents=self.pubdata.getMaxEvents() ##  self.maxEvents used in Creator.py
295	>	self.maxEvents=self.pubdata.getMaxEvents()
296
297		## Contact the DLS and build a list of sites hosting the fileblocks
298		try:
#	Line 419 | Line 366 | class Cmssw(JobType):
366		else :
367		totalNumberOfJobs = self.ncjobs
368
422	–
369		blocks = blockSites.keys()
370		blockCount = 0
371		# Backup variable in case self.maxEvents counted events in a non-included block
#	Line 474 | Line 420 | class Cmssw(JobType):
420		except KeyError:
421		common.logger.message("File "+str(file)+" has unknown number of events: skipping")
422
423	<
423	>	eventsPerJobRequested = min(eventsPerJobRequested, eventsRemaining)
424		# if less events in file remain than eventsPerJobRequested
425	<	if ( filesEventCount - jobSkipEventCount < eventsPerJobRequested ) :
425	>	if ( filesEventCount - jobSkipEventCount < eventsPerJobRequested):
426		# if last file in block
427		if ( fileCount == numFilesInBlock-1 ) :
428		# end job using last file, use remaining events in block
#	Line 540 | Line 486 | class Cmssw(JobType):
486		jobSkipEventCount = eventsPerJobRequested - (filesEventCount - jobSkipEventCount - self.eventsbyfile[file])
487		# remove all but the last file
488		filesEventCount = self.eventsbyfile[file]
489	<	parString = ""
544	<	parString += '\\\"' + file + '\\\"\,'
489	>	parString = '\\\"' + file + '\\\"\,'
490		pass # END if
491		pass # END while (iterate over files in the block)
492		pass # END while (iterate over blocks in the dataset)
#	Line 561 | Line 506 | class Cmssw(JobType):
506		for block in blocks:
507		if block in jobsOfBlock.keys() :
508		blockCounter += 1
509	<	screenOutput += "Block %5i: jobs %20s: sites: %s\n" % (blockCounter,spanRanges(jobsOfBlock[block]),','.join(self.blackWhiteListParser.checkWhiteList(self.blackWhiteListParser.checkBlackList(blockSites[block],block),block)))
509	>	screenOutput += "Block %5i: jobs %20s: sites: %s\n" % (blockCounter,spanRanges(jobsOfBlock[block]),
510	>	','.join(self.blackWhiteListParser.checkWhiteList(self.blackWhiteListParser.checkBlackList(blockSites[block],block),block)))
511		if len(self.blackWhiteListParser.checkWhiteList(self.blackWhiteListParser.checkBlackList(blockSites[block],block),block)) == 0:
512		noSiteBlock.append( spanRanges(jobsOfBlock[block]) )
513		bloskNoSite.append( blockCounter )
#	Line 581 | Line 527 | class Cmssw(JobType):
527		for range_jobs in noSiteBlock:
528		msg += str(range_jobs) + virgola
529		msg += '\n               will not be submitted and this block of data can not be analyzed!\n'
530	+	if self.cfg_params.has_key('EDG.se_white_list'):
531	+	msg += 'WARNING: SE White List: '+self.cfg_params['EDG.se_white_list']+'\n'
532	+	msg += '(Hint: By whitelisting you force the job to run at this particular site(s).\n'
533	+	msg += 'Please check if the dataset is available at this site!)\n'
534	+	if self.cfg_params.has_key('EDG.ce_white_list'):
535	+	msg += 'WARNING: CE White List: '+self.cfg_params['EDG.ce_white_list']+'\n'
536	+	msg += '(Hint: By whitelisting you force the job to run at this particular site(s).\n'
537	+	msg += 'Please check if the dataset is available at this site!)\n'
538	+
539		common.logger.message(msg)
540
541		self.list_of_args = list_of_lists
#	Line 629 | Line 584 | class Cmssw(JobType):
584		self.list_of_args = []
585		for i in range(self.total_number_of_jobs):
586		## Since there is no input, any site is good
632	–	# self.jobDestination.append(["Any"])
587		self.jobDestination.append([""]) #must be empty to write correctly the xml
588		args=[]
589		if (self.firstRun):
590	<	## pythia first run
637	<	#self.list_of_args.append([(str(self.firstRun)+str(i))])
590	>	## pythia first run
591		args.append(str(self.firstRun)+str(i))
639	–	else:
640	–	## no first run
641	–	#self.list_of_args.append([str(i)])
642	–	args.append(str(i))
643	–	if (self.sourceSeed):
644	–	args.append(str(self.sourceSeed)+str(i))
645	–	if (self.sourceSeedVtx):
646	–	## + vtx random seed
647	–	args.append(str(self.sourceSeedVtx)+str(i))
648	–	if (self.sourceSeedG4):
649	–	## + G4 random seed
650	–	args.append(str(self.sourceSeedG4)+str(i))
651	–	if (self.sourceSeedMix):
652	–	## + Mix random seed
653	–	args.append(str(self.sourceSeedMix)+str(i))
654	–	pass
655	–	pass
592		self.list_of_args.append(args)
657	–	pass
658	–
659	–	# print self.list_of_args
593
594		return
595
596
597	<	def jobSplittingForScript(self):#CarlosDaniele
597	>	def jobSplittingForScript(self):
598		"""
599		Perform job splitting based on number of job
600		"""
#	Line 677 | Line 610 | class Cmssw(JobType):
610		# argument is seed number.$i
611		self.list_of_args = []
612		for i in range(self.total_number_of_jobs):
680	–	## Since there is no input, any site is good
681	–	# self.jobDestination.append(["Any"])
613		self.jobDestination.append([""])
683	–	## no random seed
614		self.list_of_args.append([str(i)])
615		return
616
617		def split(self, jobParams):
618
689	–	common.jobDB.load()
690	–	#### Fabio
619		njobs = self.total_number_of_jobs
620		arglist = self.list_of_args
621		# create the empty structure
622		for i in range(njobs):
623		jobParams.append("")
624
625	+	listID=[]
626	+	listField=[]
627		for job in range(njobs):
628		jobParams[job] = arglist[job]
629	<	# print str(arglist[job])
630	<	# print jobParams[job]
631	<	common.jobDB.setArguments(job, jobParams[job])
632	<	common.logger.debug(5,"Job "+str(job)+" Destination: "+str(self.jobDestination[job]))
633	<	common.jobDB.setDestination(job, self.jobDestination[job])
629	>	listID.append(job+1)
630	>	job_ToSave ={}
631	>	concString = ' '
632	>	argu=''
633	>	if len(jobParams[job]):
634	>	argu +=   concString.join(jobParams[job] )
635	>	job_ToSave['arguments']= str(job+1)+' '+argu
636	>	job_ToSave['dlsDestination']= self.jobDestination[job]
637	>	listField.append(job_ToSave)
638	>	msg="Job "+str(job)+" Arguments:   "+str(job+1)+" "+argu+"\n"  \
639	>	+"                     Destination: "+str(self.jobDestination[job])
640	>	common.logger.debug(5,msg)
641	>	common._db.updateJob_(listID,listField)
642	>	self.argsList = (len(jobParams[0])+1)
643
705	–	common.jobDB.save()
644		return
645
708	–	def getJobTypeArguments(self, nj, sched):
709	–	result = ''
710	–	for i in common.jobDB.arguments(nj):
711	–	result=result+str(i)+" "
712	–	return result
713	–
646		def numberOfJobs(self):
715	–	# Fabio
647		return self.total_number_of_jobs
648
649		def getTarBall(self, exe):
650		"""
651		Return the TarBall with lib and exe
652		"""
722	–
723	–	# if it exist, just return it
724	–	#
725	–	# Marco. Let's start to use relative path for Boss XML files
726	–	#
653		self.tgzNameWithPath = common.work_space.pathForTgz()+'share/'+self.tgz_name
654		if os.path.exists(self.tgzNameWithPath):
655		return self.tgzNameWithPath
#	Line 737 | Line 663 | class Cmssw(JobType):
663
664		# First of all declare the user Scram area
665		swArea = self.scram.getSWArea_()
740	–	#print "swArea = ", swArea
741	–	# swVersion = self.scram.getSWVersion()
742	–	# print "swVersion = ", swVersion
666		swReleaseTop = self.scram.getReleaseTop_()
744	–	#print "swReleaseTop = ", swReleaseTop
667
668		## check if working area is release top
669		if swReleaseTop == '' or swArea == swReleaseTop:
670	+	common.logger.debug(3,"swArea = "+swArea+" swReleaseTop ="+swReleaseTop)
671		return
672
673		import tarfile
#	Line 787 | Line 710 | class Cmssw(JobType):
710
711		## Now check if any data dir(s) is present
712		swAreaLen=len(swArea)
713	+	self.dataExist = False
714		for root, dirs, files in os.walk(swArea):
715		if "data" in dirs:
716	+	self.dataExist=True
717		common.logger.debug(5,"data "+root+"/data"+" to be tarred")
718		tar.add(root+"/data",root[swAreaLen:]+"/data")
719
720	<	## Add ProdAgent dir to tar
721	<	paDir = 'ProdAgentApi'
722	<	pa = os.environ['CRABDIR'] + '/' + 'ProdAgentApi'
723	<	if os.path.isdir(pa):
724	<	tar.add(pa,paDir)
720	>	### CMSSW ParameterSet
721	>	if not self.pset is None:
722	>	cfg_file = common.work_space.jobDir()+self.configFilename()
723	>	tar.add(cfg_file,self.configFilename())
724	>	common.logger.debug(5,"File added to "+self.tgzNameWithPath+" : "+str(tar.getnames()))
725	>
726
727	<	### FEDE FOR DBS PUBLICATION
802	<	## Add PRODCOMMON dir to tar
727	>	## Add ProdCommon dir to tar
728		prodcommonDir = 'ProdCommon'
729		prodcommonPath = os.environ['CRABDIR'] + '/' + 'ProdCommon'
730		if os.path.isdir(prodcommonPath):
731		tar.add(prodcommonPath,prodcommonDir)
732	<	#############################
732	>	common.logger.debug(5,"Files added to "+self.tgzNameWithPath+" : "+str(tar.getnames()))
733	>
734	>	##### ML stuff
735	>	ML_file_list=['report.py', 'DashboardAPI.py', 'Logger.py', 'ProcInfo.py', 'apmon.py']
736	>	path=os.environ['CRABDIR'] + '/python/'
737	>	for file in ML_file_list:
738	>	tar.add(path+file,file)
739	>	common.logger.debug(5,"Files added to "+self.tgzNameWithPath+" : "+str(tar.getnames()))
740	>
741	>	##### Utils
742	>	Utils_file_list=['parseCrabFjr.py','writeCfg.py', 'JobReportErrorCode.py']
743	>	for file in Utils_file_list:
744	>	tar.add(path+file,file)
745	>	common.logger.debug(5,"Files added to "+self.tgzNameWithPath+" : "+str(tar.getnames()))
746
747	+	##### AdditionalFiles
748	+	for file in self.additional_inbox_files:
749	+	tar.add(file,string.split(file,'/')[-1])
750		common.logger.debug(5,"Files added to "+self.tgzNameWithPath+" : "+str(tar.getnames()))
751	+
752		tar.close()
753		except :
754		raise CrabException('Could not create tar-ball')
#	Line 817 | Line 759 | class Cmssw(JobType):
759		raise CrabException('Input sandbox size of ' + str(float(tarballinfo.st_size)/1024.0/1024.0) + ' MB is larger than the allowed ' + str(self.MaxTarBallSize) + ' MB input sandbox limit and not supported by the used GRID submission system. Please make sure that no unnecessary files are in all data directories in your local CMSSW project area as they are automatically packed into the input sandbox.')
760
761		## create tar-ball with ML stuff
820	–	self.MLtgzfile =  common.work_space.pathForTgz()+'share/MLfiles.tgz'
821	–	try:
822	–	tar = tarfile.open(self.MLtgzfile, "w:gz")
823	–	path=os.environ['CRABDIR'] + '/python/'
824	–	for file in ['report.py', 'DashboardAPI.py', 'Logger.py', 'ProcInfo.py', 'apmon.py', 'parseCrabFjr.py']:
825	–	tar.add(path+file,file)
826	–	common.logger.debug(5,"Files added to "+self.MLtgzfile+" : "+str(tar.getnames()))
827	–	tar.close()
828	–	except :
829	–	raise CrabException('Could not create ML files tar-ball')
830	–
831	–	return
762
763	<	def additionalInputFileTgz(self):
834	<	"""
835	<	Put all additional files into a tar ball and return its name
836	<	"""
837	<	import tarfile
838	<	tarName=  common.work_space.pathForTgz()+'share/'+self.additional_tgz_name
839	<	tar = tarfile.open(tarName, "w:gz")
840	<	for file in self.additional_inbox_files:
841	<	tar.add(file,string.split(file,'/')[-1])
842	<	common.logger.debug(5,"Files added to "+self.additional_tgz_name+" : "+str(tar.getnames()))
843	<	tar.close()
844	<	return tarName
845	<
846	<	def wsSetupEnvironment(self, nj):
763	>	def wsSetupEnvironment(self, nj=0):
764		"""
765		Returns part of a job script which prepares
766		the execution environment for the job 'nj'.
767		"""
768	+	if (self.CMSSW_major >= 2 and self.CMSSW_minor >= 1) or (self.CMSSW_major >= 3):
769	+	psetName = 'pset.py'
770	+	else:
771	+	psetName = 'pset.cfg'
772		# Prepare JobType-independent part
773	<	txt = ''
773	>	txt = '\n#Written by cms_cmssw::wsSetupEnvironment\n'
774		txt += 'echo ">>> setup environment"\n'
775		txt += 'if [ $middleware == LCG ]; then \n'
776		txt += self.wsSetupCMSLCGEnvironment_()
777		txt += 'elif [ $middleware == OSG ]; then\n'
778		txt += '    WORKING_DIR=`/bin/mktemp  -d $OSG_WN_TMP/cms_XXXXXXXXXXXX`\n'
779		txt += '    if [ ! $? == 0 ] ;then\n'
780	<	txt += '        echo "SET_CMS_ENV 10016 ==> OSG $WORKING_DIR could not be created on WN `hostname`"\n'
781	<	txt += '        echo "JOB_EXIT_STATUS = 10016"\n'
782	<	txt += '        echo "JobExitCode=10016" | tee -a $RUNTIME_AREA/$repo\n'
862	<	txt += '        dumpStatus $RUNTIME_AREA/$repo\n'
863	<	txt += '        exit 1\n'
780	>	txt += '        echo "ERROR ==> OSG $WORKING_DIR could not be created on WN `hostname`"\n'
781	>	txt += '        job_exit_code=10016\n'
782	>	txt += '        func_exit\n'
783		txt += '    fi\n'
784		txt += '    echo ">>> Created working directory: $WORKING_DIR"\n'
785		txt += '\n'
#	Line 868 | Line 787 | class Cmssw(JobType):
787		txt += '    cd $WORKING_DIR\n'
788		txt += '    echo ">>> current directory (WORKING_DIR): $WORKING_DIR"\n'
789		txt += self.wsSetupCMSOSGEnvironment_()
871	–	#txt += '    echo "### Set SCRAM ARCH to ' + self.executable_arch + ' ###"\n'
872	–	#txt += '    export SCRAM_ARCH='+self.executable_arch+'\n'
790		txt += 'fi\n'
791
792		# Prepare JobType-specific part
#	Line 880 | Line 797 | class Cmssw(JobType):
797		txt += scram+' project CMSSW '+self.version+'\n'
798		txt += 'status=$?\n'
799		txt += 'if [ $status != 0 ] ; then\n'
800	<	txt += '    echo "SET_EXE_ENV 10034 ==>ERROR CMSSW '+self.version+' not found on `hostname`" \n'
801	<	txt += '    echo "JOB_EXIT_STATUS = 10034"\n'
802	<	txt += '    echo "JobExitCode=10034" | tee -a $RUNTIME_AREA/$repo\n'
886	<	txt += '    dumpStatus $RUNTIME_AREA/$repo\n'
887	<	txt += '    if [ $middleware == OSG ]; then \n'
888	<	txt += '        cd $RUNTIME_AREA\n'
889	<	txt += '        echo ">>> current directory (RUNTIME_AREA): $RUNTIME_AREA"\n'
890	<	txt += '        echo ">>> Remove working directory: $WORKING_DIR"\n'
891	<	txt += '        /bin/rm -rf $WORKING_DIR\n'
892	<	txt += '        if [ -d $WORKING_DIR ] ;then\n'
893	<	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'
894	<	txt += '            echo "JOB_EXIT_STATUS = 10018"\n'
895	<	txt += '            echo "JobExitCode=10018" | tee -a $RUNTIME_AREA/$repo\n'
896	<	txt += '            dumpStatus $RUNTIME_AREA/$repo\n'
897	<	txt += '        fi\n'
898	<	txt += '    fi \n'
899	<	txt += '    exit 1 \n'
800	>	txt += '    echo "ERROR ==> CMSSW '+self.version+' not found on `hostname`" \n'
801	>	txt += '    job_exit_code=10034\n'
802	>	txt += '    func_exit\n'
803		txt += 'fi \n'
804		txt += 'cd '+self.version+'\n'
902	–	########## FEDE FOR DBS2 ######################
805		txt += 'SOFTWARE_DIR=`pwd`\n'
806		txt += 'echo ">>> current directory (SOFTWARE_DIR): $SOFTWARE_DIR" \n'
905	–	###############################################
906	–	### needed grep for bug in scramv1 ###
907	–	txt += scram+' runtime -sh\n'
807		txt += 'eval `'+scram+' runtime -sh | grep -v SCRAMRT_LSB_JOBNAME`\n'
808	<	txt += 'echo $PATH\n'
809	<
808	>	txt += 'if [ $? != 0 ] ; then\n'
809	>	txt += '    echo "ERROR ==> Problem with the command: "\n'
810	>	txt += '    echo "eval \`'+scram+' runtime -sh | grep -v SCRAMRT_LSB_JOBNAME \` at `hostname`"\n'
811	>	txt += '    job_exit_code=10034\n'
812	>	txt += '    func_exit\n'
813	>	txt += 'fi \n'
814		# Handle the arguments:
815		txt += "\n"
816		txt += "## number of arguments (first argument always jobnumber)\n"
817		txt += "\n"
818	<	txt += "if [ $nargs -lt 2 ]\n"
818	>	txt += "if [ $nargs -lt "+str(self.argsList)+" ]\n"
819		txt += "then\n"
820	<	txt += "    echo 'SET_EXE_ENV 1 ==> ERROR Too few arguments' +$nargs+ \n"
821	<	txt += '    echo "JOB_EXIT_STATUS = 50113"\n'
822	<	txt += '    echo "JobExitCode=50113" | tee -a $RUNTIME_AREA/$repo\n'
920	<	txt += '    dumpStatus $RUNTIME_AREA/$repo\n'
921	<	txt += '    if [ $middleware == OSG ]; then \n'
922	<	txt += '        cd $RUNTIME_AREA\n'
923	<	txt += '        echo ">>> current directory (RUNTIME_AREA): $RUNTIME_AREA"\n'
924	<	txt += '        echo ">>> Remove working directory: $WORKING_DIR"\n'
925	<	txt += '        /bin/rm -rf $WORKING_DIR\n'
926	<	txt += '        if [ -d $WORKING_DIR ] ;then\n'
927	<	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'
928	<	txt += '            echo "JOB_EXIT_STATUS = 50114"\n'
929	<	txt += '            echo "JobExitCode=50114" | tee -a $RUNTIME_AREA/$repo\n'
930	<	txt += '            dumpStatus $RUNTIME_AREA/$repo\n'
931	<	txt += '        fi\n'
932	<	txt += '    fi \n'
933	<	txt += "    exit 1\n"
820	>	txt += "    echo 'ERROR ==> Too few arguments' +$nargs+ \n"
821	>	txt += '    job_exit_code=50113\n'
822	>	txt += "    func_exit\n"
823		txt += "fi\n"
824		txt += "\n"
825
826		# Prepare job-specific part
827		job = common.job_list[nj]
939	–	### FEDE FOR DBS OUTPUT PUBLICATION
828		if (self.datasetPath):
829		txt += '\n'
830		txt += 'DatasetPath='+self.datasetPath+'\n'
#	Line 952 | Line 840 | class Cmssw(JobType):
840		txt += 'PrimaryDataset=null\n'
841		txt += 'DataTier=null\n'
842		txt += 'ApplicationFamily=MCDataTier\n'
843	<	if self.pset != None: #CarlosDaniele
843	>	if self.pset != None:
844		pset = os.path.basename(job.configFilename())
845		txt += '\n'
846		txt += 'cp  $RUNTIME_AREA/'+pset+' .\n'
847		if (self.datasetPath): # standard job
848	<	txt += 'InputFiles=${args[1]}\n'
849	<	txt += 'MaxEvents=${args[2]}\n'
850	<	txt += 'SkipEvents=${args[3]}\n'
848	>	txt += 'InputFiles=${args[1]}; export InputFiles\n'
849	>	txt += 'MaxEvents=${args[2]}; export MaxEvents\n'
850	>	txt += 'SkipEvents=${args[3]}; export SkipEvents\n'
851		txt += 'echo "Inputfiles:<$InputFiles>"\n'
964	–	txt += 'sed "s#\'INPUTFILE\'#$InputFiles#" '+pset+' > tmp && mv -f tmp '+pset+'\n'
852		txt += 'echo "MaxEvents:<$MaxEvents>"\n'
966	–	txt += 'sed "s#int32 input = 0#int32 input = $MaxEvents#" '+pset+' > tmp && mv -f tmp '+pset+'\n'
853		txt += 'echo "SkipEvents:<$SkipEvents>"\n'
968	–	txt += 'sed "s#uint32 skipEvents = 0#uint32 skipEvents = $SkipEvents#" '+pset+' > tmp && mv -f tmp '+pset+'\n'
854		else:  # pythia like job
855	<	seedIndex=1
855	>	txt += 'PreserveSeeds='  + ','.join(self.preserveSeeds)  + '; export PreserveSeeds\n'
856	>	txt += 'IncrementSeeds=' + ','.join(self.incrementSeeds) + '; export IncrementSeeds\n'
857	>	txt += 'echo "PreserveSeeds: <$PreserveSeeds>"\n'
858	>	txt += 'echo "IncrementSeeds:<$IncrementSeeds>"\n'
859		if (self.firstRun):
860	<	txt += 'FirstRun=${args['+str(seedIndex)+']}\n'
860	>	txt += 'FirstRun=${args[1]}; export FirstRun\n'
861		txt += 'echo "FirstRun: <$FirstRun>"\n'
974	–	txt += 'sed "s#uint32 firstRun = 0#uint32 firstRun = $FirstRun#" '+pset+' > tmp && mv -f tmp '+pset+'\n'
975	–	seedIndex=seedIndex+1
862
863	<	if (self.sourceSeed):
978	<	txt += 'Seed=${args['+str(seedIndex)+']}\n'
979	<	txt += 'sed "s#uint32 sourceSeed = 0#uint32 sourceSeed = $Seed#" '+pset+' > tmp && mv -f tmp '+pset+'\n'
980	<	seedIndex=seedIndex+1
981	<	## the following seeds are not always present
982	<	if (self.sourceSeedVtx):
983	<	txt += 'VtxSeed=${args['+str(seedIndex)+']}\n'
984	<	txt += 'echo "VtxSeed: <$VtxSeed>"\n'
985	<	txt += 'sed "s#uint32 VtxSmeared = 0#uint32 VtxSmeared = $VtxSeed#" '+pset+' > tmp && mv -f tmp '+pset+'\n'
986	<	seedIndex += 1
987	<	if (self.sourceSeedG4):
988	<	txt += 'G4Seed=${args['+str(seedIndex)+']}\n'
989	<	txt += 'echo "G4Seed: <$G4Seed>"\n'
990	<	txt += 'sed "s#uint32 g4SimHits = 0#uint32 g4SimHits = $G4Seed#" '+pset+' > tmp && mv -f tmp '+pset+'\n'
991	<	seedIndex += 1
992	<	if (self.sourceSeedMix):
993	<	txt += 'mixSeed=${args['+str(seedIndex)+']}\n'
994	<	txt += 'echo "MixSeed: <$mixSeed>"\n'
995	<	txt += 'sed "s#uint32 mix = 0#uint32 mix = $mixSeed#" '+pset+' > tmp && mv -f tmp '+pset+'\n'
996	<	seedIndex += 1
997	<	pass
998	<	pass
999	<	txt += 'mv -f '+pset+' pset.cfg\n'
863	>	txt += 'mv -f ' + pset + ' ' + psetName + '\n'
864
1001	–	if len(self.additional_inbox_files) > 0:
1002	–	txt += 'if [ -e $RUNTIME_AREA/'+self.additional_tgz_name+' ] ; then\n'
1003	–	txt += '  tar xzvf $RUNTIME_AREA/'+self.additional_tgz_name+'\n'
1004	–	txt += 'fi\n'
1005	–	pass
865
866	<	if self.pset != None: #CarlosDaniele
866	>	if self.pset != None:
867	>	# FUTURE: Can simply for 2_1_x and higher
868		txt += '\n'
869	<	txt += 'echo "***** cat pset.cfg *********"\n'
870	<	txt += 'cat pset.cfg\n'
871	<	txt += 'echo "****** end pset.cfg ********"\n'
872	<	txt += '\n'
873	<	### FEDE FOR DBS OUTPUT PUBLICATION
874	<	txt += 'PSETHASH=`EdmConfigHash < pset.cfg` \n'
869	>	if self.debug_pset==True:
870	>	txt += 'echo "***** cat ' + psetName + ' *********"\n'
871	>	txt += 'cat ' + psetName + '\n'
872	>	txt += 'echo "****** end ' + psetName + ' ********"\n'
873	>	txt += '\n'
874	>	txt += 'PSETHASH=`edmConfigHash < ' + psetName + '` \n'
875		txt += 'echo "PSETHASH = $PSETHASH" \n'
1016	–	##############
876		txt += '\n'
877		return txt
878
879	<	def wsBuildExe(self, nj=0):
879	>	def wsUntarSoftware(self, nj=0):
880		"""
881		Put in the script the commands to build an executable
882		or a library.
883		"""
884
885	<	txt = ""
885	>	txt = '\n#Written by cms_cmssw::wsUntarSoftware\n'
886
887		if os.path.isfile(self.tgzNameWithPath):
888		txt += 'echo ">>> tar xzvf $RUNTIME_AREA/'+os.path.basename(self.tgzNameWithPath)+' :" \n'
889		txt += 'tar xzvf $RUNTIME_AREA/'+os.path.basename(self.tgzNameWithPath)+'\n'
890	+	txt += 'ls -Al \n'
891		txt += 'untar_status=$? \n'
892		txt += 'if [ $untar_status -ne 0 ]; then \n'
893	<	txt += '   echo "SET_EXE 1 ==> ERROR Untarring .tgz file failed"\n'
894	<	txt += '   echo "JOB_EXIT_STATUS = $untar_status" \n'
895	<	txt += '   echo "JobExitCode=$untar_status" | tee -a $RUNTIME_AREA/$repo\n'
1036	<	txt += '   if [ $middleware == OSG ]; then \n'
1037	<	txt += '       cd $RUNTIME_AREA\n'
1038	<	txt += '        echo ">>> current directory (RUNTIME_AREA): $RUNTIME_AREA"\n'
1039	<	txt += '        echo ">>> Remove working directory: $WORKING_DIR"\n'
1040	<	txt += '       /bin/rm -rf $WORKING_DIR\n'
1041	<	txt += '       if [ -d $WORKING_DIR ] ;then\n'
1042	<	txt += '           echo "SET_EXE 50999 ==> OSG $WORKING_DIR could not be deleted on WN `hostname` after Untarring .tgz file failed"\n'
1043	<	txt += '           echo "JOB_EXIT_STATUS = 50999"\n'
1044	<	txt += '           echo "JobExitCode=50999" | tee -a $RUNTIME_AREA/$repo\n'
1045	<	txt += '           dumpStatus $RUNTIME_AREA/$repo\n'
1046	<	txt += '       fi\n'
1047	<	txt += '   fi \n'
1048	<	txt += '   \n'
1049	<	txt += '   exit 1 \n'
893	>	txt += '   echo "ERROR ==> Untarring .tgz file failed"\n'
894	>	txt += '   job_exit_code=$untar_status\n'
895	>	txt += '   func_exit\n'
896		txt += 'else \n'
897		txt += '   echo "Successful untar" \n'
898		txt += 'fi \n'
899		txt += '\n'
900	<	txt += 'echo ">>> Include ProdAgentApi and PRODCOMMON in PYTHONPATH:"\n'
900	>	txt += 'echo ">>> Include ProdCommon in PYTHONPATH:"\n'
901		txt += 'if [ -z "$PYTHONPATH" ]; then\n'
902	<	#### FEDE FOR DBS OUTPUT PUBLICATION
1057	<	txt += '   export PYTHONPATH=$SOFTWARE_DIR/ProdAgentApi:$SOFTWARE_DIR/ProdCommon\n'
902	>	txt += '   export PYTHONPATH=$RUNTIME_AREA/ProdCommon\n'
903		txt += 'else\n'
904	<	txt += '   export PYTHONPATH=$SOFTWARE_DIR/ProdAgentApi:$SOFTWARE_DIR/ProdCommon:${PYTHONPATH}\n'
904	>	txt += '   export PYTHONPATH=$RUNTIME_AREA/ProdCommon:${PYTHONPATH}\n'
905		txt += 'echo "PYTHONPATH=$PYTHONPATH"\n'
1061	–	###################
906		txt += 'fi\n'
907		txt += '\n'
908
#	Line 1066 | Line 910 | class Cmssw(JobType):
910
911		return txt
912
913	+	def wsBuildExe(self, nj=0):
914	+	"""
915	+	Put in the script the commands to build an executable
916	+	or a library.
917	+	"""
918	+
919	+	txt = '\n#Written by cms_cmssw::wsBuildExe\n'
920	+	txt += 'echo ">>> moving CMSSW software directories in `pwd`" \n'
921	+
922	+	txt += 'rm -r lib/ module/ \n'
923	+	txt += 'mv $RUNTIME_AREA/lib/ . \n'
924	+	txt += 'mv $RUNTIME_AREA/module/ . \n'
925	+	if self.dataExist == True:
926	+	txt += 'rm -r src/ \n'
927	+	txt += 'mv $RUNTIME_AREA/src/ . \n'
928	+	if len(self.additional_inbox_files)>0:
929	+	for file in self.additional_inbox_files:
930	+	txt += 'mv $RUNTIME_AREA/'+os.path.basename(file)+' . \n'
931	+	txt += 'mv $RUNTIME_AREA/ProdCommon/ . \n'
932	+
933	+	txt += 'if [ -z "$PYTHONPATH" ]; then\n'
934	+	txt += '   export PYTHONPATH=$SOFTWARE_DIR/ProdCommon\n'
935	+	txt += 'else\n'
936	+	txt += '   export PYTHONPATH=$SOFTWARE_DIR/ProdCommon:${PYTHONPATH}\n'
937	+	txt += 'echo "PYTHONPATH=$PYTHONPATH"\n'
938	+	txt += 'fi\n'
939	+	txt += '\n'
940	+
941	+	return txt
942	+
943		def modifySteeringCards(self, nj):
944		"""
945		modify the card provided by the user,
#	Line 1073 | Line 947 | class Cmssw(JobType):
947		"""
948
949		def executableName(self):
950	<	if self.scriptExe: #CarlosDaniele
950	>	if self.scriptExe:
951		return "sh "
952		else:
953		return self.executable
954
955		def executableArgs(self):
956	+	# FUTURE: This function tests the CMSSW version. Can be simplified as we drop support for old versions
957		if self.scriptExe:#CarlosDaniele
958		return   self.scriptExe + " $NJob"
959		else:
960	<	# if >= CMSSW_1_5_X, add -e
961	<	version_array = self.scram.getSWVersion().split('_')
962	<	major = 0
963	<	minor = 0
964	<	try:
965	<	major = int(version_array[1])
966	<	minor = int(version_array[2])
967	<	except:
1093	<	msg = "Cannot parse CMSSW version string: " + "_".join(version_array) + " for major and minor release number!"
1094	<	raise CrabException(msg)
1095	<	if major >= 1 and minor >= 5 :
1096	<	return " -e -p pset.cfg"
960	>	ex_args = ""
961	>	# FUTURE: This tests the CMSSW version. Can remove code as versions deprecated
962	>	# Framework job report
963	>	if (self.CMSSW_major >= 1 and self.CMSSW_minor >= 5) or (self.CMSSW_major >= 2):
964	>	ex_args += " -j $RUNTIME_AREA/crab_fjr_$NJob.xml"
965	>	# Type of config file
966	>	if self.CMSSW_major >= 2 :
967	>	ex_args += " -p pset.py"
968		else:
969	<	return " -p pset.cfg"
969	>	ex_args += " -p pset.cfg"
970	>	return ex_args
971
972		def inputSandbox(self, nj):
973		"""
974		Returns a list of filenames to be put in JDL input sandbox.
975		"""
976		inp_box = []
1105	–	# # dict added to delete duplicate from input sandbox file list
1106	–	# seen = {}
1107	–	## code
977		if os.path.isfile(self.tgzNameWithPath):
978		inp_box.append(self.tgzNameWithPath)
979	<	if os.path.isfile(self.MLtgzfile):
980	<	inp_box.append(self.MLtgzfile)
1112	<	## config
1113	<	if not self.pset is None:
1114	<	inp_box.append(common.work_space.pathForTgz() + 'job/' + self.configFilename())
1115	<	## additional input files
1116	<	tgz = self.additionalInputFileTgz()
1117	<	inp_box.append(tgz)
979	>	wrapper = os.path.basename(str(common._db.queryTask('scriptName')))
980	>	inp_box.append(common.work_space.pathForTgz() +'job/'+ wrapper)
981		return inp_box
982
983		def outputSandbox(self, nj):
#	Line 1140 | Line 1003 | class Cmssw(JobType):
1003		Returns part of a job script which renames the produced files.
1004		"""
1005
1006	<	txt = '\n'
1007	<	txt += 'echo" >>> directory content:"\n'
1006	>	txt = '\n#Written by cms_cmssw::wsRenameOutput\n'
1007	>	txt += 'echo ">>> current directory (SOFTWARE_DIR): $SOFTWARE_DIR" \n'
1008	>	txt += 'echo ">>> current directory content:"\n'
1009		txt += 'ls \n'
1010	<	txt = '\n'
1147	<
1148	<	txt += 'output_exit_status=0\n'
1149	<
1150	<	for fileWithSuffix in (self.output_file_sandbox):
1151	<	output_file_num = self.numberFile_(fileWithSuffix, '$NJob')
1152	<	txt += '\n'
1153	<	txt += '# check output file\n'
1154	<	txt += 'if [ -e ./'+fileWithSuffix+' ] ; then\n'
1155	<	txt += '    mv '+fileWithSuffix+' $RUNTIME_AREA\n'
1156	<	txt += '    cp $RUNTIME_AREA/'+fileWithSuffix+' $RUNTIME_AREA/'+output_file_num+'\n'
1157	<	txt += 'else\n'
1158	<	txt += '    exit_status=60302\n'
1159	<	txt += '    echo "ERROR: Problem with output file '+fileWithSuffix+'"\n'
1160	<	if common.scheduler.boss_scheduler_name == 'condor_g':
1161	<	txt += '    if [ $middleware == OSG ]; then \n'
1162	<	txt += '        echo "prepare dummy output file"\n'
1163	<	txt += '        echo "Processing of job output failed" > $RUNTIME_AREA/'+output_file_num+'\n'
1164	<	txt += '    fi \n'
1165	<	txt += 'fi\n'
1010	>	txt += '\n'
1011
1012		for fileWithSuffix in (self.output_file):
1013		output_file_num = self.numberFile_(fileWithSuffix, '$NJob')
1014		txt += '\n'
1015		txt += '# check output file\n'
1016		txt += 'if [ -e ./'+fileWithSuffix+' ] ; then\n'
1017	<	txt += '    mv '+fileWithSuffix+' $RUNTIME_AREA\n'
1018	<	txt += '    cp $RUNTIME_AREA/'+fileWithSuffix+' $RUNTIME_AREA/'+output_file_num+'\n'
1017	>	if (self.copy_data == 1):  # For OSG nodes, file is in $WORKING_DIR, should not be moved to $RUNTIME_AREA
1018	>	txt += '    mv '+fileWithSuffix+' '+output_file_num+'\n'
1019	>	txt += '    ln -s `pwd`/'+output_file_num+' $RUNTIME_AREA/'+fileWithSuffix+'\n'
1020	>	else:
1021	>	txt += '    mv '+fileWithSuffix+' $RUNTIME_AREA/'+output_file_num+'\n'
1022	>	txt += '    ln -s $RUNTIME_AREA/'+output_file_num+' $RUNTIME_AREA/'+fileWithSuffix+'\n'
1023		txt += 'else\n'
1024	<	txt += '    exit_status=60302\n'
1025	<	txt += '    echo "ERROR: Problem with output file '+fileWithSuffix+'"\n'
1026	<	txt += '    echo "JOB_EXIT_STATUS = $exit_status"\n'
1178	<	txt += '    output_exit_status=$exit_status\n'
1179	<	if common.scheduler.boss_scheduler_name == 'condor_g':
1024	>	txt += '    job_exit_code=60302\n'
1025	>	txt += '    echo "WARNING: Output file '+fileWithSuffix+' not found"\n'
1026	>	if common.scheduler.name().upper() == 'CONDOR_G':
1027		txt += '    if [ $middleware == OSG ]; then \n'
1028		txt += '        echo "prepare dummy output file"\n'
1029		txt += '        echo "Processing of job output failed" > $RUNTIME_AREA/'+output_file_num+'\n'
#	Line 1187 | Line 1034 | class Cmssw(JobType):
1034		file_list.append(self.numberFile_(fileWithSuffix, '$NJob'))
1035
1036		txt += 'file_list="'+string.join(file_list,' ')+'"\n'
1037	+	txt += '\n'
1038	+	txt += 'echo ">>> current directory (SOFTWARE_DIR): $SOFTWARE_DIR" \n'
1039	+	txt += 'echo ">>> current directory content:"\n'
1040	+	txt += 'ls \n'
1041	+	txt += '\n'
1042		txt += 'cd $RUNTIME_AREA\n'
1043		txt += 'echo ">>> current directory (RUNTIME_AREA):  $RUNTIME_AREA"\n'
1044		return txt
#	Line 1218 | Line 1070 | class Cmssw(JobType):
1070		req='Member("VO-cms-' + \
1071		self.version + \
1072		'", other.GlueHostApplicationSoftwareRunTimeEnvironment)'
1073	<	## SL add requirement for OS version only if SL4
1222	<	#reSL4 = re.compile( r'slc4' )
1223	<	if self.executable_arch: # and reSL4.search(self.executable_arch):
1073	>	if self.executable_arch:
1074		req+=' && Member("VO-cms-' + \
1075		self.executable_arch + \
1076		'", other.GlueHostApplicationSoftwareRunTimeEnvironment)'
1077
1078		req = req + ' && (other.GlueHostNetworkAdapterOutboundIP)'
1079	+	if common.scheduler.name() == "glitecoll":
1080	+	req += ' && other.GlueCEStateStatus == "Production" '
1081
1082		return req
1083
1084		def configFilename(self):
1085		""" return the config filename """
1086	<	return self.name()+'.cfg'
1086	>	# FUTURE: Can remove cfg mode for CMSSW >= 2_1_x
1087	>	if (self.CMSSW_major >= 2 and self.CMSSW_minor >= 1) or (self.CMSSW_major >= 3):
1088	>	return self.name()+'.py'
1089	>	else:
1090	>	return self.name()+'.cfg'
1091
1092		def wsSetupCMSOSGEnvironment_(self):
1093		"""
1094		Returns part of a job script which is prepares
1095		the execution environment and which is common for all CMS jobs.
1096		"""
1097	<	txt = '    echo ">>> setup CMS OSG environment:"\n'
1097	>	txt = '\n#Written by cms_cmssw::wsSetupCMSOSGEnvironment_\n'
1098	>	txt += '    echo ">>> setup CMS OSG environment:"\n'
1099		txt += '    echo "set SCRAM ARCH to ' + self.executable_arch + '"\n'
1100		txt += '    export SCRAM_ARCH='+self.executable_arch+'\n'
1101		txt += '    echo "SCRAM_ARCH = $SCRAM_ARCH"\n'
#	Line 1246 | Line 1103 | class Cmssw(JobType):
1103		txt += '      # Use $OSG_APP/cmssoft/cms/cmsset_default.sh to setup cms software\n'
1104		txt += '        source $OSG_APP/cmssoft/cms/cmsset_default.sh '+self.version+'\n'
1105		txt += '    else\n'
1106	<	txt += '        echo "SET_CMS_ENV 10020 ==> ERROR $OSG_APP/cmssoft/cms/cmsset_default.sh file not found"\n'
1107	<	txt += '        echo "JOB_EXIT_STATUS = 10020"\n'
1108	<	txt += '        echo "JobExitCode=10020" | tee -a $RUNTIME_AREA/$repo\n'
1252	<	txt += '        dumpStatus $RUNTIME_AREA/$repo\n'
1253	<	txt += '\n'
1254	<	txt += '        cd $RUNTIME_AREA\n'
1255	<	txt += '        echo ">>> current directory (RUNTIME_AREA): $RUNTIME_AREA"\n'
1256	<	txt += '        echo ">>> Remove working directory: $WORKING_DIR"\n'
1257	<	txt += '        /bin/rm -rf $WORKING_DIR\n'
1258	<	txt += '        if [ -d $WORKING_DIR ] ;then\n'
1259	<	txt += '            echo "SET_CMS_ENV 10017 ==> OSG $WORKING_DIR could not be deleted on WN `hostname` after $OSG_APP/cmssoft/cms/cmsset_default.sh file not found"\n'
1260	<	txt += '            echo "JOB_EXIT_STATUS = 10017"\n'
1261	<	txt += '            echo "JobExitCode=10017" | tee -a $RUNTIME_AREA/$repo\n'
1262	<	txt += '            dumpStatus $RUNTIME_AREA/$repo\n'
1263	<	txt += '        fi\n'
1264	<	txt += '\n'
1265	<	txt += '        exit 1\n'
1106	>	txt += '        echo "ERROR ==> $OSG_APP/cmssoft/cms/cmsset_default.sh file not found"\n'
1107	>	txt += '        job_exit_code=10020\n'
1108	>	txt += '        func_exit\n'
1109		txt += '    fi\n'
1110		txt += '\n'
1111	<	txt += '    echo "SET_CMS_ENV 0 ==> setup cms environment ok"\n'
1111	>	txt += '    echo "==> setup cms environment ok"\n'
1112		txt += '    echo "SCRAM_ARCH = $SCRAM_ARCH"\n'
1113
1114		return txt
1115
1273	–	### OLI_DANIELE
1116		def wsSetupCMSLCGEnvironment_(self):
1117		"""
1118		Returns part of a job script which is prepares
1119		the execution environment and which is common for all CMS jobs.
1120		"""
1121	<	txt = '    echo ">>> setup CMS LCG environment:"\n'
1121	>	txt = '\n#Written by cms_cmssw::wsSetupCMSLCGEnvironment_\n'
1122	>	txt += '    echo ">>> setup CMS LCG environment:"\n'
1123		txt += '    echo "set SCRAM ARCH and BUILD_ARCH to ' + self.executable_arch + ' ###"\n'
1124		txt += '    export SCRAM_ARCH='+self.executable_arch+'\n'
1125		txt += '    export BUILD_ARCH='+self.executable_arch+'\n'
1126		txt += '    if [ ! $VO_CMS_SW_DIR ] ;then\n'
1127	<	txt += '        echo "SET_CMS_ENV 10031 ==> ERROR CMS software dir not found on WN `hostname`"\n'
1128	<	txt += '        echo "JOB_EXIT_STATUS = 10031" \n'
1129	<	txt += '        echo "JobExitCode=10031" | tee -a $RUNTIME_AREA/$repo\n'
1287	<	txt += '        dumpStatus $RUNTIME_AREA/$repo\n'
1288	<	txt += '        exit 1\n'
1127	>	txt += '        echo "ERROR ==> CMS software dir not found on WN `hostname`"\n'
1128	>	txt += '        job_exit_code=10031\n'
1129	>	txt += '        func_exit\n'
1130		txt += '    else\n'
1131		txt += '        echo "Sourcing environment... "\n'
1132		txt += '        if [ ! -s $VO_CMS_SW_DIR/cmsset_default.sh ] ;then\n'
1133	<	txt += '            echo "SET_CMS_ENV 10020 ==> ERROR cmsset_default.sh file not found into dir $VO_CMS_SW_DIR"\n'
1134	<	txt += '            echo "JOB_EXIT_STATUS = 10020"\n'
1135	<	txt += '            echo "JobExitCode=10020" | tee -a $RUNTIME_AREA/$repo\n'
1295	<	txt += '            dumpStatus $RUNTIME_AREA/$repo\n'
1296	<	txt += '            exit 1\n'
1133	>	txt += '            echo "ERROR ==> cmsset_default.sh file not found into dir $VO_CMS_SW_DIR"\n'
1134	>	txt += '            job_exit_code=10020\n'
1135	>	txt += '            func_exit\n'
1136		txt += '        fi\n'
1137		txt += '        echo "sourcing $VO_CMS_SW_DIR/cmsset_default.sh"\n'
1138		txt += '        source $VO_CMS_SW_DIR/cmsset_default.sh\n'
1139		txt += '        result=$?\n'
1140		txt += '        if [ $result -ne 0 ]; then\n'
1141	<	txt += '            echo "SET_CMS_ENV 10032 ==> ERROR problem sourcing $VO_CMS_SW_DIR/cmsset_default.sh"\n'
1142	<	txt += '            echo "JOB_EXIT_STATUS = 10032"\n'
1143	<	txt += '            echo "JobExitCode=10032" | tee -a $RUNTIME_AREA/$repo\n'
1305	<	txt += '            dumpStatus $RUNTIME_AREA/$repo\n'
1306	<	txt += '            exit 1\n'
1141	>	txt += '            echo "ERROR ==> problem sourcing $VO_CMS_SW_DIR/cmsset_default.sh"\n'
1142	>	txt += '            job_exit_code=10032\n'
1143	>	txt += '            func_exit\n'
1144		txt += '        fi\n'
1145		txt += '    fi\n'
1146		txt += '    \n'
1147	<	txt += '    echo "SET_CMS_ENV 0 ==> setup cms environment ok"\n'
1147	>	txt += '    echo "==> setup cms environment ok"\n'
1148		return txt
1149
1313	–	### FEDE FOR DBS OUTPUT PUBLICATION
1150		def modifyReport(self, nj):
1151		"""
1152		insert the part of the script that modifies the FrameworkJob Report
1153		"""
1154	<
1155	<	txt = ''
1320	<	try:
1321	<	publish_data = int(self.cfg_params['USER.publish_data'])
1322	<	except KeyError:
1323	<	publish_data = 0
1154	>	txt = '\n#Written by cms_cmssw::modifyReport\n'
1155	>	publish_data = int(self.cfg_params.get('USER.publish_data',0))
1156		if (publish_data == 1):
1157	<	txt += 'echo ">>> Modify Job Report:" \n'
1158	<	################ FEDE FOR DBS2 #############################################
1327	<	txt += 'chmod a+x $SOFTWARE_DIR/ProdAgentApi/FwkJobRep/ModifyJobReport.py\n'
1328	<	#############################################################################
1157	>	processedDataset = self.cfg_params['USER.publish_data_name']
1158	>	LFNBaseName = LFNBase(processedDataset)
1159
1160	<	txt += 'if [ -z "$SE" ]; then\n'
1161	<	txt += '    SE="" \n'
1162	<	txt += 'fi \n'
1163	<	txt += 'if [ -z "$SE_PATH" ]; then\n'
1164	<	txt += '    SE_PATH="" \n'
1165	<	txt += 'fi \n'
1166	<	txt += 'echo "SE = $SE"\n'
1337	<	txt += 'echo "SE_PATH = $SE_PATH"\n'
1160	>	txt += 'if [ $copy_exit_status -eq 0 ]; then\n'
1161	>	txt += '    FOR_LFN=%s_${PSETHASH}/\n'%(LFNBaseName)
1162	>	txt += 'else\n'
1163	>	txt += '    FOR_LFN=/copy_problems/ \n'
1164	>	txt += '    SE=""\n'
1165	>	txt += '    SE_PATH=""\n'
1166	>	txt += 'fi\n'
1167
1168	<	processedDataset = self.cfg_params['USER.publish_data_name']
1168	>	txt += 'echo ">>> Modify Job Report:" \n'
1169	>	txt += 'chmod a+x $SOFTWARE_DIR/ProdCommon/ProdCommon/FwkJobRep/ModifyJobReport.py\n'
1170		txt += 'ProcessedDataset='+processedDataset+'\n'
1341	–	#### LFN=/store/user/<user>/processedDataset_PSETHASH
1342	–	txt += 'if [ "$SE_PATH" == "" ]; then\n'
1343	–	#### FEDE: added slash in LFN ##############
1344	–	txt += '    FOR_LFN=/copy_problems/ \n'
1345	–	txt += 'else \n'
1346	–	txt += '    tmp=`echo $SE_PATH | awk -F \'store\' \'{print$2}\'` \n'
1347	–	#####  FEDE TO BE CHANGED, BECAUSE STORE IS HARDCODED!!!! ########
1348	–	txt += '    FOR_LFN=/store$tmp \n'
1349	–	txt += 'fi \n'
1171		txt += 'echo "ProcessedDataset = $ProcessedDataset"\n'
1172	+	txt += 'echo "SE = $SE"\n'
1173	+	txt += 'echo "SE_PATH = $SE_PATH"\n'
1174		txt += 'echo "FOR_LFN = $FOR_LFN" \n'
1175		txt += 'echo "CMSSW_VERSION = $CMSSW_VERSION"\n\n'
1176	<	#txt += 'echo "$RUNTIME_AREA/'+self.version+'/ProdAgentApi/FwkJobRep/ModifyJobReport.py crab_fjr_$NJob.xml $NJob $FOR_LFN $PrimaryDataset $DataTier $ProcessedDataset $ApplicationFamily $executable $CMSSW_VERSION $PSETHASH $SE $SE_PATH"\n'
1177	<	txt += 'echo "$SOFTWARE_DIR/ProdAgentApi/FwkJobRep/ModifyJobReport.py crab_fjr_$NJob.xml $NJob $FOR_LFN $PrimaryDataset $DataTier $ProcessedDataset $ApplicationFamily $executable $CMSSW_VERSION $PSETHASH $SE $SE_PATH"\n'
1355	<	txt += '$SOFTWARE_DIR/ProdAgentApi/FwkJobRep/ModifyJobReport.py crab_fjr_$NJob.xml $NJob $FOR_LFN $PrimaryDataset $DataTier $ProcessedDataset $ApplicationFamily $executable $CMSSW_VERSION $PSETHASH $SE $SE_PATH\n'
1356	<	#txt += '$RUNTIME_AREA/'+self.version+'/ProdAgentApi/FwkJobRep/ModifyJobReport.py crab_fjr_$NJob.xml $NJob $FOR_LFN $PrimaryDataset $DataTier $ProcessedDataset $ApplicationFamily $executable $CMSSW_VERSION $PSETHASH $SE $SE_PATH\n'
1357	<
1176	>	txt += 'echo "$SOFTWARE_DIR/ProdCommon/ProdCommon/FwkJobRep/ModifyJobReport.py $RUNTIME_AREA/crab_fjr_$NJob.xml $NJob $FOR_LFN $PrimaryDataset $DataTier $ProcessedDataset $ApplicationFamily $executable $CMSSW_VERSION $PSETHASH $SE $SE_PATH"\n'
1177	>	txt += '$SOFTWARE_DIR/ProdCommon/ProdCommon/FwkJobRep/ModifyJobReport.py $RUNTIME_AREA/crab_fjr_$NJob.xml $NJob $FOR_LFN $PrimaryDataset $DataTier $ProcessedDataset $ApplicationFamily $executable $CMSSW_VERSION $PSETHASH $SE $SE_PATH\n'
1178		txt += 'modifyReport_result=$?\n'
1359	–	txt += 'echo modifyReport_result = $modifyReport_result\n'
1179		txt += 'if [ $modifyReport_result -ne 0 ]; then\n'
1180	<	txt += '    exit_status=1\n'
1181	<	txt += '    echo "ERROR: Problem with ModifyJobReport"\n'
1180	>	txt += '    modifyReport_result=70500\n'
1181	>	txt += '    job_exit_code=$modifyReport_result\n'
1182	>	txt += '    echo "ModifyReportResult=$modifyReport_result" | tee -a $RUNTIME_AREA/$repo\n'
1183	>	txt += '    echo "WARNING: Problem with ModifyJobReport"\n'
1184		txt += 'else\n'
1185	<	txt += '    mv NewFrameworkJobReport.xml crab_fjr_$NJob.xml\n'
1185	>	txt += '    mv NewFrameworkJobReport.xml $RUNTIME_AREA/crab_fjr_$NJob.xml\n'
1186		txt += 'fi\n'
1366	–	else:
1367	–	txt += 'echo "no data publication required"\n'
1187		return txt
1188
1189	<	def cleanEnv(self):
1190	<	txt = ''
1191	<	txt += 'if [ $middleware == OSG ]; then\n'
1192	<	txt += '    cd $RUNTIME_AREA\n'
1193	<	txt += '    echo ">>> current directory (RUNTIME_AREA): $RUNTIME_AREA"\n'
1194	<	txt += '    echo ">>> Remove working directory: $WORKING_DIR"\n'
1195	<	txt += '    /bin/rm -rf $WORKING_DIR\n'
1196	<	txt += '    if [ -d $WORKING_DIR ] ;then\n'
1197	<	txt += '        echo "SET_EXE 60999 ==> OSG $WORKING_DIR could not be deleted on WN `hostname` after cleanup of WN"\n'
1198	<	txt += '        echo "JOB_EXIT_STATUS = 60999"\n'
1199	<	txt += '        echo "JobExitCode=60999" | tee -a $RUNTIME_AREA/$repo\n'
1200	<	txt += '        dumpStatus $RUNTIME_AREA/$repo\n'
1189	>	def wsParseFJR(self):
1190	>	"""
1191	>	Parse the FrameworkJobReport to obtain useful infos
1192	>	"""
1193	>	txt = '\n#Written by cms_cmssw::wsParseFJR\n'
1194	>	txt += 'echo ">>> Parse FrameworkJobReport crab_fjr.xml"\n'
1195	>	txt += 'if [ -s $RUNTIME_AREA/crab_fjr_$NJob.xml ]; then\n'
1196	>	txt += '    if [ -s $RUNTIME_AREA/parseCrabFjr.py ]; then\n'
1197	>	txt += '        cmd_out=`python $RUNTIME_AREA/parseCrabFjr.py --input $RUNTIME_AREA/crab_fjr_$NJob.xml --MonitorID $MonitorID --MonitorJobID $MonitorJobID`\n'
1198	>	txt += '        echo "Result of parsing the FrameworkJobReport crab_fjr.xml: $cmd_out"\n'
1199	>	txt += '        tmp_executable_exit_status=`echo $cmd_out | awk -F\; \'{print $1}\' | awk -F \' \' \'{print $NF}\'`\n'
1200	>	txt += '        if [ -n $tmp_executable_exit_status ];then\n'
1201	>	txt += '            executable_exit_status=$tmp_executable_exit_status\n'
1202	>	txt += '        fi\n'
1203	>	txt += '        if [ $executable_exit_status -eq 50115 ];then\n'
1204	>	txt += '            echo ">>> crab_fjr.xml contents: "\n'
1205	>	txt += '            cat $RUNTIME_AREA/crab_fjr_NJob.xml\n'
1206	>	txt += '            echo "Wrong FrameworkJobReport --> does not contain useful info. ExitStatus: $executable_exit_status"\n'
1207	>	txt += '        else\n'
1208	>	txt += '            echo "Extracted ExitStatus from FrameworkJobReport parsing output: $executable_exit_status"\n'
1209	>	txt += '        fi\n'
1210	>	txt += '    else\n'
1211	>	txt += '        echo "CRAB python script to parse CRAB FrameworkJobReport crab_fjr.xml is not available, using exit code of executable from command line."\n'
1212		txt += '    fi\n'
1213	+	#### Patch to check input data reading for CMSSW16x Hopefully we-ll remove it asap
1214	+
1215	+	if self.datasetPath:
1216	+	# VERIFY PROCESSED DATA
1217	+	txt += '    if [ $executable_exit_status -eq 0 ];then\n'
1218	+	txt += '      echo ">>> Verify list of processed files:"\n'
1219	+	txt += '      echo $InputFiles |tr -d \'\\\\\' |tr \',\' \'\\n\'|tr -d \'"\' > input-files.txt\n'
1220	+	txt += '      grep LFN $RUNTIME_AREA/crab_fjr_$NJob.xml |cut -d">" -f2|cut -d"<" -f1|grep "/" > processed-files.txt\n'
1221	+	txt += '      cat input-files.txt  | sort | uniq > tmp.txt\n'
1222	+	txt += '      mv tmp.txt input-files.txt\n'
1223	+	txt += '      echo "cat input-files.txt"\n'
1224	+	txt += '      echo "----------------------"\n'
1225	+	txt += '      cat input-files.txt\n'
1226	+	txt += '      cat processed-files.txt | sort | uniq > tmp.txt\n'
1227	+	txt += '      mv tmp.txt processed-files.txt\n'
1228	+	txt += '      echo "----------------------"\n'
1229	+	txt += '      echo "cat processed-files.txt"\n'
1230	+	txt += '      echo "----------------------"\n'
1231	+	txt += '      cat processed-files.txt\n'
1232	+	txt += '      echo "----------------------"\n'
1233	+	txt += '      diff -q input-files.txt processed-files.txt\n'
1234	+	txt += '      fileverify_status=$?\n'
1235	+	txt += '      if [ $fileverify_status -ne 0 ]; then\n'
1236	+	txt += '         executable_exit_status=30001\n'
1237	+	txt += '         echo "ERROR ==> not all input files processed"\n'
1238	+	txt += '         echo "      ==> list of processed files from crab_fjr.xml differs from list in pset.cfg"\n'
1239	+	txt += '         echo "      ==> diff input-files.txt processed-files.txt"\n'
1240	+	txt += '      fi\n'
1241	+	txt += '    fi\n'
1242	+	txt += '\n'
1243	+	txt += 'else\n'
1244	+	txt += '    echo "CRAB FrameworkJobReport crab_fjr.xml is not available, using exit code of executable from command line."\n'
1245		txt += 'fi\n'
1246		txt += '\n'
1247	+	txt += 'echo "ExeExitCode=$executable_exit_status" | tee -a $RUNTIME_AREA/$repo\n'
1248	+	txt += 'echo "EXECUTABLE_EXIT_STATUS = $executable_exit_status"\n'
1249	+	txt += 'job_exit_code=$executable_exit_status\n'
1250	+
1251		return txt
1252
1253		def setParam_(self, param, value):
#	Line 1390 | Line 1256 | class Cmssw(JobType):
1256		def getParams(self):
1257		return self._params
1258
1393	–	def setTaskid_(self):
1394	–	self._taskId = self.cfg_params['taskId']
1395	–
1396	–	def getTaskid(self):
1397	–	return self._taskId
1398	–
1259		def uniquelist(self, old):
1260		"""
1261		remove duplicates from a list
#	Line 1405 | Line 1265 | class Cmssw(JobType):
1265		nd[e]=0
1266		return nd.keys()
1267
1268	<
1409	<	def checkOut(self, limit):
1268	>	def outList(self):
1269		"""
1270		check the dimension of the output files
1271		"""
1272	<	txt += 'echo ">>> Starting output sandbox limit check :"\n'
1273	<	allOutFiles = ""
1272	>	txt = ''
1273	>	txt += 'echo ">>> list of expected files on output sandbox"\n'
1274		listOutFiles = []
1275	<	for fileOut in (self.output_file+self.output_file_sandbox):
1276	<	if fileOut.find('crab_fjr') == -1:
1277	<	allOutFiles = allOutFiles + " " + self.numberFile_(fileOut, '$NJob')
1278	<	listOutFiles.append(self.numberFile_(fileOut, '$NJob'))
1279	<	txt += 'echo "OUTPUT files: '+str(allOutFiles)+'";\n'
1280	<	txt += 'ls -gGhrta;\n'
1281	<	txt += 'sum=0;\n'
1282	<	txt += 'for file in '+str(allOutFiles)+' ; do\n'
1283	<	txt += '    if [ -e $file ]; then\n'
1284	<	txt += '        tt=`ls -gGrta $file | awk \'{ print $3 }\'`\n'
1285	<	txt += '        sum=`expr $sum + $tt`\n'
1286	<	txt += '    else\n'
1287	<	txt += '        echo "WARNING: output file $file not found!"\n'
1288	<	txt += '    fi\n'
1289	<	txt += 'done\n'
1431	<	txt += 'echo "Total Output dimension: $sum";\n'
1432	<	txt += 'limit='+str(limit)+';\n'
1433	<	txt += 'echo "OUTPUT FILES LIMIT SET TO: $limit";\n'
1434	<	txt += 'if [ $limit -lt $sum ]; then\n'
1435	<	txt += '    echo "WARNING: output files have to big size - something will be lost;"\n'
1436	<	txt += '    echo "         checking the output file sizes..."\n'
1437	<	"""
1438	<	txt += '    dim=0;\n'
1439	<	txt += '    exclude=0;\n'
1440	<	txt += '    for files in '+str(allOutFiles)+' ; do\n'
1441	<	txt += '        sumTemp=0;\n'
1442	<	txt += '        for file2 in '+str(allOutFiles)+' ; do\n'
1443	<	txt += '            if [ $file != $file2 ]; then\n'
1444	<	txt += '                tt=`ls -gGrta $file2 | awk \'{ print $3 }\';`\n'
1445	<	txt += '                sumTemp=`expr $sumTemp + $tt`;\n'
1446	<	txt += '            fi\n'
1447	<	txt += '        done\n'
1448	<	txt += '        if [ $sumTemp -lt $limit ]; then\n'
1449	<	txt += '            if [ $dim -lt $sumTemp ]; then\n'
1450	<	txt += '                dim=$sumTemp;\n'
1451	<	txt += '                exclude=$file;\n'
1452	<	txt += '            fi\n'
1453	<	txt += '        fi\n'
1454	<	txt += '    done\n'
1455	<	txt += '    echo "Dimension calculated: $dim"; echo "File to exclude: $exclude";\n'
1456	<	"""
1457	<	txt += '    tot=0;\n'
1458	<	txt += '    for file2 in '+str(allOutFiles)+' ; do\n'
1459	<	txt += '        tt=`ls -gGrta $file2 | awk \'{ print $3 }\';`\n'
1460	<	txt += '        tot=`expr $tot + $tt`;\n'
1461	<	txt += '        if [ $limit -lt $tot ]; then\n'
1462	<	txt += '            tot=`expr $tot - $tt`;\n'
1463	<	txt += '            fileLast=$file;\n'
1464	<	txt += '            break;\n'
1465	<	txt += '        fi\n'
1466	<	txt += '    done\n'
1467	<	txt += '    echo "Dimension calculated: $tot"; echo "First file to exclude: $file";\n'
1468	<	txt += '    flag=0;\n'
1469	<	txt += '    for filess in '+str(allOutFiles)+' ; do\n'
1470	<	txt += '        if [ $fileLast = $filess ]; then\n'
1471	<	txt += '            flag=1;\n'
1472	<	txt += '        fi\n'
1473	<	txt += '        if [ $flag -eq 1 ]; then\n'
1474	<	txt += '            rm -f $filess;\n'
1475	<	txt += '        fi\n'
1476	<	txt += '    done\n'
1477	<	txt += '    ls -agGhrt;\n'
1478	<	txt += '    echo "WARNING: output files are too big in dimension: can not put in the output_sandbox.";\n'
1479	<	txt += '    echo "JOB_EXIT_STATUS = 70000";\n'
1480	<	txt += '    exit_status=70000;\n'
1481	<	txt += 'else'
1482	<	txt += '    echo "Total Output dimension $sum is fine.";\n'
1483	<	txt += 'fi\n'
1484	<	txt += 'echo "Ending output sandbox limit check"\n'
1275	>	stdout = 'CMSSW_$NJob.stdout'
1276	>	stderr = 'CMSSW_$NJob.stderr'
1277	>	if (self.return_data == 1):
1278	>	for file in (self.output_file+self.output_file_sandbox):
1279	>	listOutFiles.append(self.numberFile_(file, '$NJob'))
1280	>	listOutFiles.append(stdout)
1281	>	listOutFiles.append(stderr)
1282	>	else:
1283	>	for file in (self.output_file_sandbox):
1284	>	listOutFiles.append(self.numberFile_(file, '$NJob'))
1285	>	listOutFiles.append(stdout)
1286	>	listOutFiles.append(stderr)
1287	>	txt += 'echo "output files: '+string.join(listOutFiles,' ')+'"\n'
1288	>	txt += 'filesToCheck="'+string.join(listOutFiles,' ')+'"\n'
1289	>	txt += 'export filesToCheck\n'
1290		return txt

Diff Legend

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