[ViewVC] Diff of: cvsroot/COMP/CRAB/python/cms

Comparing COMP/CRAB/python/cms_cmssw.py (file contents):
Revision 1.130 by fanzago, Wed Oct 17 13:24:41 2007 UTC vs.
Revision 1.205 by slacapra, Thu Jun 5 16:34:04 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
28
29		log = common.logger
30	<
30	>
31		self.scram = Scram.Scram(cfg_params)
32		self.additional_inbox_files = []
33		self.scriptExe = ''
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 = "Error: CMS does not support %s with %s architecture"%(self.version, self.executable_arch)
54	<	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)
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)
60	–
61	–	common.taskDB.setDict('codeVersion',self.version)
62	–	self.setParam_('application', self.version)
56
57		### collect Data cards
58
59	<	## get DBS mode
60	<	try:
68	<	self.use_dbs_1 = int(self.cfg_params['CMSSW.use_dbs_1'])
69	<	except KeyError:
70	<	self.use_dbs_1 = 0
71	<
72	<	try:
73	<	tmp = cfg_params['CMSSW.datasetpath']
74	<	log.debug(6, "CMSSW::CMSSW(): datasetPath = "+tmp)
75	<	if string.lower(tmp)=='none':
76	<	self.datasetPath = None
77	<	self.selectNoInput = 1
78	<	else:
79	<	self.datasetPath = tmp
80	<	self.selectNoInput = 0
81	<	except KeyError:
82	<	msg = "Error: datasetpath not defined "
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')
89	<	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("/")
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'])
68	>	self.datasetPath = tmp
69	>	self.selectNoInput = 0
70
71		self.dataTiers = []
72	<
72	>	self.debugWrap = ''
73	>	self.debug_wrapper = cfg_params.get('USER.debug_wrapper',False)
74	>	if self.debug_wrapper: self.debugWrap='--debug'
75		## now the application
76	<	try:
77	<	self.executable = cfg_params['CMSSW.executable']
113	<	self.setParam_('exe', self.executable)
114	<	log.debug(6, "CMSSW::CMSSW(): executable = "+self.executable)
115	<	msg = "Default executable cmsRun overridden. Switch to " + self.executable
116	<	log.debug(3,msg)
117	<	except KeyError:
118	<	self.executable = 'cmsRun'
119	<	self.setParam_('exe', self.executable)
120	<	msg = "User executable not defined. Use cmsRun"
121	<	log.debug(3,msg)
122	<	pass
76	>	self.executable = cfg_params.get('CMSSW.executable','cmsRun')
77	>	log.debug(6, "CMSSW::CMSSW(): executable = "+self.executable)
78
79	<	try:
125	<	self.pset = cfg_params['CMSSW.pset']
126	<	log.debug(6, "Cmssw::Cmssw(): PSet file = "+self.pset)
127	<	if self.pset.lower() != 'none' :
128	<	if (not os.path.exists(self.pset)):
129	<	raise CrabException("User defined PSet file "+self.pset+" does not exist")
130	<	else:
131	<	self.pset = None
132	<	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 140 \| 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)
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\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
157	<	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)
168	<	self.additional_inbox_files.append(string.strip(self.scriptExe))
169	<	except KeyError:
170	<	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
172	–	#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	>	# use parent files...
123	>	self.useParent = self.cfg_params.get('CMSSW.use_parent',False)
124
125		## additional input files
126	<	try:
126	>	if cfg_params.has_key('USER.additional_input_files'):
127		tmpAddFiles = string.split(cfg_params['USER.additional_input_files'],',')
128		for tmp in tmpAddFiles:
129		tmp = string.strip(tmp)
#	Line 192 \| Line 140 \| class Cmssw(JobType):
140		if not os.path.exists(file):
141		raise CrabException("Additional input file not found: "+file)
142		pass
195	–	# fname = string.split(file, '/')[-1]
196	–	# storedFile = common.work_space.pathForTgz()+'share/'+fname
197	–	# shutil.copyfile(file, storedFile)
143		self.additional_inbox_files.append(string.strip(file))
144		pass
145		pass
146		common.logger.debug(5,"Additional input files: "+str(self.additional_inbox_files))
147	<	except KeyError:
203	<	pass
204	<
205	<	# files per job
206	<	try:
207	<	if (cfg_params['CMSSW.files_per_jobs']):
208	<	raise CrabException("files_per_jobs no longer supported. Quitting.")
209	<	except KeyError:
210	<	pass
147	>	pass
148
149		## Events per job
150	<	try:
150	>	if cfg_params.has_key('CMSSW.events_per_job'):
151		self.eventsPerJob =int( cfg_params['CMSSW.events_per_job'])
152		self.selectEventsPerJob = 1
153	<	except KeyError:
153	>	else:
154		self.eventsPerJob = -1
155		self.selectEventsPerJob = 0
156	<
156	>
157		## number of jobs
158	<	try:
158	>	if cfg_params.has_key('CMSSW.number_of_jobs'):
159		self.theNumberOfJobs =int( cfg_params['CMSSW.number_of_jobs'])
160		self.selectNumberOfJobs = 1
161	<	except KeyError:
161	>	else:
162		self.theNumberOfJobs = 0
163		self.selectNumberOfJobs = 0
164
165	<	try:
165	>	if cfg_params.has_key('CMSSW.total_number_of_events'):
166		self.total_number_of_events = int(cfg_params['CMSSW.total_number_of_events'])
167		self.selectTotalNumberEvents = 1
168	<	except KeyError:
168	>	if self.selectNumberOfJobs == 1:
169	>	if (self.total_number_of_events != -1) and int(self.total_number_of_events) < int(self.theNumberOfJobs):
170	>	msg = 'Must specify at least one event per job. total_number_of_events > number_of_jobs '
171	>	raise CrabException(msg)
172	>	else:
173		self.total_number_of_events = 0
174		self.selectTotalNumberEvents = 0
175
176	<	if self.pset != None: #CarlosDaniele
176	>	if self.pset != None:
177		if ( (self.selectTotalNumberEvents + self.selectEventsPerJob + self.selectNumberOfJobs) != 2 ):
178		msg = 'Must define exactly two of total_number_of_events, events_per_job, or number_of_jobs.'
179		raise CrabException(msg)
#	Line 241 \| Line 182 \| class Cmssw(JobType):
182		msg = 'Must specify number_of_jobs.'
183		raise CrabException(msg)
184
185	<	## source seed for pythia
186	<	try:
187	<	self.sourceSeed = int(cfg_params['CMSSW.pythia_seed'])
188	<	except KeyError:
189	<	self.sourceSeed = None
190	<	common.logger.debug(5,"No seed given")
185	>	## New method of dealing with seeds
186	>	self.incrementSeeds = []
187	>	self.preserveSeeds = []
188	>	if cfg_params.has_key('CMSSW.preserve_seeds'):
189	>	tmpList = cfg_params['CMSSW.preserve_seeds'].split(',')
190	>	for tmp in tmpList:
191	>	tmp.strip()
192	>	self.preserveSeeds.append(tmp)
193	>	if cfg_params.has_key('CMSSW.increment_seeds'):
194	>	tmpList = cfg_params['CMSSW.increment_seeds'].split(',')
195	>	for tmp in tmpList:
196	>	tmp.strip()
197	>	self.incrementSeeds.append(tmp)
198	>
199	>	## Old method of dealing with seeds
200	>	## FUTURE: This is for old CMSSW and old CRAB. Can throw exceptions after a couple of CRAB releases and then
201	>	## remove
202	>	self.sourceSeed = cfg_params.get('CMSSW.pythia_seed',None)
203	>	if self.sourceSeed:
204	>	print "pythia_seed is a deprecated parameter. Use preserve_seeds or increment_seeds in the future.\n","Added to increment_seeds."
205	>	self.incrementSeeds.append('sourceSeed')
206	>	self.incrementSeeds.append('theSource')
207	>
208	>	self.sourceSeedVtx = cfg_params.get('CMSSW.vtx_seed',None)
209	>	if self.sourceSeedVtx:
210	>	print "vtx_seed is a deprecated parameter. Use preserve_seeds or increment_seeds in the future.\n","Added to increment_seeds."
211	>	self.incrementSeeds.append('VtxSmeared')
212	>
213	>	self.sourceSeedG4 = cfg_params.get('CMSSW.g4_seed',None)
214	>	if self.sourceSeedG4:
215	>	print "g4_seed is a deprecated parameter. Use preserve_seeds or increment_seeds in the future.\n","Added to increment_seeds."
216	>	self.incrementSeeds.append('g4SimHits')
217	>
218	>	self.sourceSeedMix = cfg_params.get('CMSSW.mix_seed',None)
219	>	if self.sourceSeedMix:
220	>	print "mix_seed is a deprecated parameter. Use preserve_seeds or increment_seeds in the future.\n","Added to increment_seeds."
221	>	self.incrementSeeds.append('mix')
222
223	<	try:
252	<	self.sourceSeedVtx = int(cfg_params['CMSSW.vtx_seed'])
253	<	except KeyError:
254	<	self.sourceSeedVtx = None
255	<	common.logger.debug(5,"No vertex seed given")
223	>	self.firstRun = cfg_params.get('CMSSW.first_run',None)
224
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")
225		if self.pset != None: #CarlosDaniele
226	<	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
226	>	import PsetManipulator as pp
227		PsetEdit = pp.PsetManipulator(self.pset) #Daniele Pset
228
229	+	# Copy/return
230	+
231	+	self.copy_data = int(cfg_params.get('USER.copy_data',0))
232	+	self.return_data = int(cfg_params.get('USER.return_data',0))
233	+
234		#DBSDLS-start
235	<	## Initialize the variables that are extracted from DBS/DLS and needed in other places of the code
235	>	## Initialize the variables that are extracted from DBS/DLS and needed in other places of the code
236		self.maxEvents=0 # max events available ( --> check the requested nb. of evts in Creator.py)
237		self.DBSPaths={} # all dbs paths requested ( --> input to the site local discovery script)
238		self.jobDestination=[] # Site destination(s) for each job (list of lists)
#	Line 289 \| Line 241 \| class Cmssw(JobType):
241		blockSites = {}
242		if self.datasetPath:
243		blockSites = self.DataDiscoveryAndLocation(cfg_params)
244	<	#DBSDLS-end
244	>	#DBSDLS-end
245	>
246
294	–	self.tgzNameWithPath = self.getTarBall(self.executable)
295	–
247		## Select Splitting
248	<	if self.selectNoInput:
249	<	if self.pset == None: #CarlosDaniele
248	>	if self.selectNoInput:
249	>	if self.pset == None:
250		self.jobSplittingForScript()
251		else:
252		self.jobSplittingNoInput()
#	Line 303 \| Line 254 \| class Cmssw(JobType):
254		self.jobSplittingByBlocks(blockSites)
255
256		# modify Pset
257	<	if self.pset != None: #CarlosDaniele
257	>	if self.pset != None:
258		try:
259	<	if (self.datasetPath): # standard job
260	<	# allow to processa a fraction of events in a file
261	<	PsetEdit.inputModule("INPUT")
262	<	PsetEdit.maxEvent("INPUTMAXEVENTS")
312	<	PsetEdit.skipEvent("INPUTSKIPEVENTS")
313	<	else: # pythia like job
314	<	PsetEdit.maxEvent(self.eventsPerJob)
315	<	if (self.firstRun):
316	<	PsetEdit.pythiaFirstRun("INPUTFIRSTRUN") #First Run
317	<	if (self.sourceSeed) :
318	<	PsetEdit.pythiaSeed("INPUT")
319	<	if (self.sourceSeedVtx) :
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	<	PsetEdit.addCrabFJR(self.fjrFileName)
259	>	# Add FrameworkJobReport to parameter-set, set max events.
260	>	# Reset later for data jobs by writeCFG which does all modifications
261	>	PsetEdit.addCrabFJR(self.fjrFileName) # FUTURE: Job report addition not needed by CMSSW>1.5
262	>	PsetEdit.maxEvent(self.eventsPerJob)
263		PsetEdit.psetWriter(self.configFilename())
264		except:
265	<	msg='Error while manipuliating ParameterSet: exiting...'
265	>	msg='Error while manipulating ParameterSet: exiting...'
266		raise CrabException(msg)
267	+	self.tgzNameWithPath = self.getTarBall(self.executable)
268
269		def DataDiscoveryAndLocation(self, cfg_params):
270
271		import DataDiscovery
335	–	import DataDiscovery_DBS2
272		import DataLocation
273		common.logger.debug(10,"CMSSW::DataDiscoveryAndLocation()")
274
#	Line 341 \| Line 277 \| class Cmssw(JobType):
277		## Contact the DBS
278		common.logger.message("Contacting Data Discovery Services ...")
279		try:
280	<
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)
280	>	self.pubdata=DataDiscovery.DataDiscovery(datasetPath, cfg_params)
281		self.pubdata.fetchDBSInfo()
282
283		except DataDiscovery.NotExistingDatasetError, ex :
#	Line 357 \| Line 289 \| class Cmssw(JobType):
289		except DataDiscovery.DataDiscoveryError, ex:
290		msg = 'ERROR ***: failed Data Discovery in DBS : %s'%ex.getErrorMessage()
291		raise CrabException(msg)
360	–	except DataDiscovery_DBS2.NotExistingDatasetError_DBS2, ex :
361	–	msg = 'ERROR ***: failed Data Discovery in DBS : %s'%ex.getErrorMessage()
362	–	raise CrabException(msg)
363	–	except DataDiscovery_DBS2.NoDataTierinProvenanceError_DBS2, ex :
364	–	msg = 'ERROR ***: failed Data Discovery in DBS : %s'%ex.getErrorMessage()
365	–	raise CrabException(msg)
366	–	except DataDiscovery_DBS2.DataDiscoveryError_DBS2, ex:
367	–	msg = 'ERROR ***: failed Data Discovery in DBS : %s'%ex.getErrorMessage()
368	–	raise CrabException(msg)
292
293		self.filesbyblock=self.pubdata.getFiles()
294		self.eventsbyblock=self.pubdata.getEventsPerBlock()
295		self.eventsbyfile=self.pubdata.getEventsPerFile()
296	+	self.parentFiles=self.pubdata.getParent()
297
298		## get max number of events
299	<	self.maxEvents=self.pubdata.getMaxEvents() ## self.maxEvents used in Creator.py
299	>	self.maxEvents=self.pubdata.getMaxEvents()
300
301		## Contact the DLS and build a list of sites hosting the fileblocks
302		try:
#	Line 381 \| Line 305 \| class Cmssw(JobType):
305		except DataLocation.DataLocationError , ex:
306		msg = 'ERROR ***: failed Data Location in DLS \n %s '%ex.getErrorMessage()
307		raise CrabException(msg)
308	<
308	>
309
310		sites = dataloc.getSites()
311		allSites = []
#	Line 395 \| Line 319 \| class Cmssw(JobType):
319		common.logger.message("Requested dataset: " + datasetPath + " has " + str(self.maxEvents) + " events in " + str(len(self.filesbyblock.keys())) + " blocks.\n")
320
321		return sites
322	<
322	>
323		def jobSplittingByBlocks(self, blockSites):
324		"""
325		Perform job splitting. Jobs run over an integer number of files
#	Line 445 \| Line 369 \| class Cmssw(JobType):
369		totalNumberOfJobs = 999999999
370		else :
371		totalNumberOfJobs = self.ncjobs
448	–
372
373		blocks = blockSites.keys()
374		blockCount = 0
#	Line 465 \| Line 388 \| class Cmssw(JobType):
388		blockCount += 1
389		if block not in jobsOfBlock.keys() :
390		jobsOfBlock[block] = []
391	<
391	>
392		if self.eventsbyblock.has_key(block) :
393		numEventsInBlock = self.eventsbyblock[block]
394		common.logger.debug(5,'Events in Block File '+str(numEventsInBlock))
395	<
395	>
396		files = self.filesbyblock[block]
397		numFilesInBlock = len(files)
398		if (numFilesInBlock <= 0):
#	Line 477 \| Line 400 \| class Cmssw(JobType):
400		fileCount = 0
401
402		# ---- New block => New job ---- #
403	<	parString = "\\{"
403	>	parString = ""
404		# counter for number of events in files currently worked on
405		filesEventCount = 0
406		# flag if next while loop should touch new file
407		newFile = 1
408		# job event counter
409		jobSkipEventCount = 0
410	<
410	>
411		# ---- Iterate over the files in the block until we've met the requested ---- #
412		# ---- total # of events or we've gone over all the files in this block ---- #
413	+	pString=''
414		while ( (eventsRemaining > 0) and (fileCount < numFilesInBlock) and (jobCount < totalNumberOfJobs) ):
415		file = files[fileCount]
416	+	if self.useParent:
417	+	parent = self.parentFiles[file]
418	+	for f in parent :
419	+	pString += '\\\"' + f + '\\\"\,'
420	+	common.logger.debug(6, "File "+str(file)+" has the following parents: "+str(parent))
421	+	common.logger.write("File "+str(file)+" has the following parents: "+str(parent))
422		if newFile :
423		try:
424		numEventsInFile = self.eventsbyfile[file]
#	Line 500 \| Line 430 \| class Cmssw(JobType):
430		newFile = 0
431		except KeyError:
432		common.logger.message("File "+str(file)+" has unknown number of events: skipping")
503	–
433
434	+	eventsPerJobRequested = min(eventsPerJobRequested, eventsRemaining)
435		# if less events in file remain than eventsPerJobRequested
436	<	if ( filesEventCount - jobSkipEventCount < eventsPerJobRequested ) :
436	>	if ( filesEventCount - jobSkipEventCount < eventsPerJobRequested):
437		# if last file in block
438		if ( fileCount == numFilesInBlock-1 ) :
439		# end job using last file, use remaining events in block
440		# close job and touch new file
441		fullString = parString[:-2]
442	<	fullString += '\\}'
443	<	list_of_lists.append([fullString,str(-1),str(jobSkipEventCount)])
442	>	if self.useParent:
443	>	fullParentString = pString[:-2]
444	>	list_of_lists.append([fullString,fullParentString,str(-1),str(jobSkipEventCount)])
445	>	else:
446	>	list_of_lists.append([fullString,str(-1),str(jobSkipEventCount)])
447		common.logger.debug(3,"Job "+str(jobCount+1)+" can run over "+str(filesEventCount - jobSkipEventCount)+" events (last file in block).")
448		self.jobDestination.append(blockSites[block])
449		common.logger.debug(5,"Job "+str(jobCount+1)+" Destination: "+str(self.jobDestination[jobCount]))
#	Line 522 \| Line 455 \| class Cmssw(JobType):
455		eventsRemaining = eventsRemaining - filesEventCount + jobSkipEventCount
456		jobSkipEventCount = 0
457		# reset file
458	<	parString = "\\{"
458	>	pString = ""
459	>	parString = ""
460		filesEventCount = 0
461		newFile = 1
462		fileCount += 1
#	Line 534 \| Line 468 \| class Cmssw(JobType):
468		elif ( filesEventCount - jobSkipEventCount == eventsPerJobRequested ) :
469		# close job and touch new file
470		fullString = parString[:-2]
471	<	fullString += '\\}'
472	<	list_of_lists.append([fullString,str(eventsPerJobRequested),str(jobSkipEventCount)])
471	>	if self.useParent:
472	>	fullParentString = pString[:-2]
473	>	list_of_lists.append([fullString,fullParentString,str(eventsPerJobRequested),str(jobSkipEventCount)])
474	>	else:
475	>	list_of_lists.append([fullString,str(eventsPerJobRequested),str(jobSkipEventCount)])
476		common.logger.debug(3,"Job "+str(jobCount+1)+" can run over "+str(eventsPerJobRequested)+" events.")
477		self.jobDestination.append(blockSites[block])
478		common.logger.debug(5,"Job "+str(jobCount+1)+" Destination: "+str(self.jobDestination[jobCount]))
#	Line 546 \| Line 483 \| class Cmssw(JobType):
483		eventsRemaining = eventsRemaining - eventsPerJobRequested
484		jobSkipEventCount = 0
485		# reset file
486	<	parString = "\\{"
486	>	pString = ""
487	>	parString = ""
488		filesEventCount = 0
489		newFile = 1
490		fileCount += 1
491	<
491	>
492		# if more events in file remain than eventsPerJobRequested
493		else :
494		# close job but don't touch new file
495		fullString = parString[:-2]
496	<	fullString += '\\}'
497	<	list_of_lists.append([fullString,str(eventsPerJobRequested),str(jobSkipEventCount)])
496	>	if self.useParent:
497	>	fullParentString = pString[:-2]
498	>	list_of_lists.append([fullString,fullParentString,str(eventsPerJobRequested),str(jobSkipEventCount)])
499	>	else:
500	>	list_of_lists.append([fullString,str(eventsPerJobRequested),str(jobSkipEventCount)])
501		common.logger.debug(3,"Job "+str(jobCount+1)+" can run over "+str(eventsPerJobRequested)+" events.")
502		self.jobDestination.append(blockSites[block])
503		common.logger.debug(5,"Job "+str(jobCount+1)+" Destination: "+str(self.jobDestination[jobCount]))
#	Line 570 \| Line 511 \| class Cmssw(JobType):
511		jobSkipEventCount = eventsPerJobRequested - (filesEventCount - jobSkipEventCount - self.eventsbyfile[file])
512		# remove all but the last file
513		filesEventCount = self.eventsbyfile[file]
514	<	parString = "\\{"
515	<	parString += '\\\"' + file + '\\\"\,'
514	>	if self.useParent:
515	>	for f in parent : pString += '\\\"' + f + '\\\"\,'
516	>	parString = '\\\"' + file + '\\\"\,'
517		pass # END if
518		pass # END while (iterate over files in the block)
519		pass # END while (iterate over blocks in the dataset)
#	Line 579 \| Line 521 \| class Cmssw(JobType):
521		if (eventsRemaining > 0 and jobCount < totalNumberOfJobs ):
522		common.logger.message("Could not run on all requested events because some blocks not hosted at allowed sites.")
523		common.logger.message(str(jobCount)+" job(s) can run on "+str(totalEventCount)+" events.\n")
524	<
524	>
525		# screen output
526		screenOutput = "List of jobs and available destination sites:\n\n"
527
#	Line 591 \| Line 533 \| class Cmssw(JobType):
533		for block in blocks:
534		if block in jobsOfBlock.keys() :
535		blockCounter += 1
536	<	screenOutput += "Block %5i: jobs %20s: sites: %s\n" % (blockCounter,spanRanges(jobsOfBlock[block]),','.join(self.blackWhiteListParser.checkWhiteList(self.blackWhiteListParser.checkBlackList(blockSites[block],block),block)))
536	>	screenOutput += "Block %5i: jobs %20s: sites: %s\n" % (blockCounter,spanRanges(jobsOfBlock[block]),
537	>	','.join(self.blackWhiteListParser.checkWhiteList(self.blackWhiteListParser.checkBlackList(blockSites[block],block),block)))
538		if len(self.blackWhiteListParser.checkWhiteList(self.blackWhiteListParser.checkBlackList(blockSites[block],block),block)) == 0:
539	<	noSiteBlock.append( spanRanges(jobsOfBlock[block]) )
539	>	noSiteBlock.append( spanRanges(jobsOfBlock[block]) )
540		bloskNoSite.append( blockCounter )
541	<
541	>
542		common.logger.message(screenOutput)
543		if len(noSiteBlock) > 0 and len(bloskNoSite) > 0:
544		msg = 'WARNING: No sites are hosting any part of data for block:\n '
#	Line 611 \| Line 554 \| class Cmssw(JobType):
554		for range_jobs in noSiteBlock:
555		msg += str(range_jobs) + virgola
556		msg += '\n will not be submitted and this block of data can not be analyzed!\n'
557	+	if self.cfg_params.has_key('EDG.se_white_list'):
558	+	msg += 'WARNING: SE White List: '+self.cfg_params['EDG.se_white_list']+'\n'
559	+	msg += '(Hint: By whitelisting you force the job to run at this particular site(s).\n'
560	+	msg += 'Please check if the dataset is available at this site!)\n'
561	+	if self.cfg_params.has_key('EDG.ce_white_list'):
562	+	msg += 'WARNING: CE White List: '+self.cfg_params['EDG.ce_white_list']+'\n'
563	+	msg += '(Hint: By whitelisting you force the job to run at this particular site(s).\n'
564	+	msg += 'Please check if the dataset is available at this site!)\n'
565	+
566		common.logger.message(msg)
567
568		self.list_of_args = list_of_lists
#	Line 622 \| Line 574 \| class Cmssw(JobType):
574		"""
575		common.logger.debug(5,'Splitting per events')
576
577	<	if (self.selectEventsPerJob):
577	>	if (self.selectEventsPerJob):
578		common.logger.message('Required '+str(self.eventsPerJob)+' events per job ')
579		if (self.selectNumberOfJobs):
580		common.logger.message('Required '+str(self.theNumberOfJobs)+' jobs in total ')
#	Line 636 \| Line 588 \| class Cmssw(JobType):
588		if (self.selectEventsPerJob):
589		if (self.selectTotalNumberEvents):
590		self.total_number_of_jobs = int(self.total_number_of_events/self.eventsPerJob)
591	<	elif(self.selectNumberOfJobs) :
591	>	elif(self.selectNumberOfJobs) :
592		self.total_number_of_jobs =self.theNumberOfJobs
593	<	self.total_number_of_events =int(self.theNumberOfJobs*self.eventsPerJob)
593	>	self.total_number_of_events =int(self.theNumberOfJobs*self.eventsPerJob)
594
595		elif (self.selectNumberOfJobs) :
596		self.total_number_of_jobs = self.theNumberOfJobs
597		self.eventsPerJob = int(self.total_number_of_events/self.total_number_of_jobs)
598	<
598	>
599		common.logger.debug(5,'N jobs '+str(self.total_number_of_jobs))
600
601		# is there any remainder?
#	Line 659 \| Line 611 \| class Cmssw(JobType):
611		self.list_of_args = []
612		for i in range(self.total_number_of_jobs):
613		## Since there is no input, any site is good
614	<	# self.jobDestination.append(["Any"])
663	<	self.jobDestination.append([""]) #must be empty to write correctly the xml
614	>	self.jobDestination.append([""]) #must be empty to write correctly the xml
615		args=[]
616		if (self.firstRun):
617	<	## pythia first run
667	<	#self.list_of_args.append([(str(self.firstRun)+str(i))])
617	>	## pythia first run
618		args.append(str(self.firstRun)+str(i))
669	–	else:
670	–	## no first run
671	–	#self.list_of_args.append([str(i)])
672	–	args.append(str(i))
673	–	if (self.sourceSeed):
674	–	args.append(str(self.sourceSeed)+str(i))
675	–	if (self.sourceSeedVtx):
676	–	## + vtx random seed
677	–	args.append(str(self.sourceSeedVtx)+str(i))
678	–	if (self.sourceSeedG4):
679	–	## + G4 random seed
680	–	args.append(str(self.sourceSeedG4)+str(i))
681	–	if (self.sourceSeedMix):
682	–	## + Mix random seed
683	–	args.append(str(self.sourceSeedMix)+str(i))
684	–	pass
685	–	pass
619		self.list_of_args.append(args)
687	–	pass
688	–
689	–	# print self.list_of_args
620
621		return
622
623
624	<	def jobSplittingForScript(self):#CarlosDaniele
624	>	def jobSplittingForScript(self):
625		"""
626		Perform job splitting based on number of job
627		"""
#	Line 707 \| Line 637 \| class Cmssw(JobType):
637		# argument is seed number.$i
638		self.list_of_args = []
639		for i in range(self.total_number_of_jobs):
710	–	## Since there is no input, any site is good
711	–	# self.jobDestination.append(["Any"])
640		self.jobDestination.append([""])
713	–	## no random seed
641		self.list_of_args.append([str(i)])
642		return
643
644		def split(self, jobParams):
645	<
719	<	common.jobDB.load()
720	<	#### Fabio
645	>
646		njobs = self.total_number_of_jobs
647		arglist = self.list_of_args
648		# create the empty structure
649		for i in range(njobs):
650		jobParams.append("")
651	<
651	>
652	>	listID=[]
653	>	listField=[]
654		for job in range(njobs):
655		jobParams[job] = arglist[job]
656	<	# print str(arglist[job])
657	<	# print jobParams[job]
658	<	common.jobDB.setArguments(job, jobParams[job])
659	<	common.logger.debug(5,"Job "+str(job)+" Destination: "+str(self.jobDestination[job]))
660	<	common.jobDB.setDestination(job, self.jobDestination[job])
656	>	listID.append(job+1)
657	>	job_ToSave ={}
658	>	concString = ' '
659	>	argu=''
660	>	if len(jobParams[job]):
661	>	argu += concString.join(jobParams[job] )
662	>	job_ToSave['arguments']= str(job+1)+' '+argu
663	>	job_ToSave['dlsDestination']= self.jobDestination[job]
664	>	listField.append(job_ToSave)
665	>	msg="Job "+str(job)+" Arguments: "+str(job+1)+" "+argu+"\n" \
666	>	+" Destination: "+str(self.jobDestination[job])
667	>	common.logger.debug(5,msg)
668	>	common._db.updateJob_(listID,listField)
669	>	self.argsList = (len(jobParams[0])+1)
670
735	–	common.jobDB.save()
671		return
672	<
738	<	def getJobTypeArguments(self, nj, sched):
739	<	result = ''
740	<	for i in common.jobDB.arguments(nj):
741	<	result=result+str(i)+" "
742	<	return result
743	<
672	>
673		def numberOfJobs(self):
745	–	# Fabio
674		return self.total_number_of_jobs
675
676		def getTarBall(self, exe):
677		"""
678		Return the TarBall with lib and exe
679		"""
752	–
753	–	# if it exist, just return it
754	–	#
755	–	# Marco. Let's start to use relative path for Boss XML files
756	–	#
680		self.tgzNameWithPath = common.work_space.pathForTgz()+'share/'+self.tgz_name
681		if os.path.exists(self.tgzNameWithPath):
682		return self.tgzNameWithPath
#	Line 767 \| Line 690 \| class Cmssw(JobType):
690
691		# First of all declare the user Scram area
692		swArea = self.scram.getSWArea_()
770	–	#print "swArea = ", swArea
771	–	# swVersion = self.scram.getSWVersion()
772	–	# print "swVersion = ", swVersion
693		swReleaseTop = self.scram.getReleaseTop_()
694	<	#print "swReleaseTop = ", swReleaseTop
775	<
694	>
695		## check if working area is release top
696		if swReleaseTop == '' or swArea == swReleaseTop:
697	+	common.logger.debug(3,"swArea = "+swArea+" swReleaseTop ="+swReleaseTop)
698		return
699
700		import tarfile
#	Line 785 \| Line 705 \| class Cmssw(JobType):
705		exeWithPath = self.scram.findFile_(executable)
706		if ( not exeWithPath ):
707		raise CrabException('User executable '+executable+' not found')
708	<
708	>
709		## then check if it's private or not
710		if exeWithPath.find(swReleaseTop) == -1:
711		# the exe is private, so we must ship
#	Line 801 \| Line 721 \| class Cmssw(JobType):
721		else:
722		# the exe is from release, we'll find it on WN
723		pass
724	<
724	>
725		## Now get the libraries: only those in local working area
726		libDir = 'lib'
727		lib = swArea+'/' +libDir
728		common.logger.debug(5,"lib "+lib+" to be tarred")
729		if os.path.exists(lib):
730		tar.add(lib,libDir)
731	<
731	>
732		## Now check if module dir is present
733		moduleDir = 'module'
734		module = swArea + '/' + moduleDir
#	Line 817 \| Line 737 \| class Cmssw(JobType):
737
738		## Now check if any data dir(s) is present
739		swAreaLen=len(swArea)
740	+	self.dataExist = False
741		for root, dirs, files in os.walk(swArea):
742		if "data" in dirs:
743	+	self.dataExist=True
744		common.logger.debug(5,"data "+root+"/data"+" to be tarred")
745		tar.add(root+"/data",root[swAreaLen:]+"/data")
746
747	<	## Add ProdAgent dir to tar
748	<	paDir = 'ProdAgentApi'
749	<	pa = os.environ['CRABDIR'] + '/' + 'ProdAgentApi'
750	<	if os.path.isdir(pa):
751	<	tar.add(pa,paDir)
747	>	### CMSSW ParameterSet
748	>	if not self.pset is None:
749	>	cfg_file = common.work_space.jobDir()+self.configFilename()
750	>	tar.add(cfg_file,self.configFilename())
751	>	common.logger.debug(5,"File added to "+self.tgzNameWithPath+" : "+str(tar.getnames()))
752	>
753
754	<	### FEDE FOR DBS PUBLICATION
832	<	## Add PRODCOMMON dir to tar
754	>	## Add ProdCommon dir to tar
755		prodcommonDir = 'ProdCommon'
756	<	prodcommonPath = os.environ['CRABDIR'] + '/' + 'ProdCommon'
756	>	prodcommonPath = os.environ['CRABDIR'] + '/' + 'external/'
757		if os.path.isdir(prodcommonPath):
758		tar.add(prodcommonPath,prodcommonDir)
837	–	#############################
838	–
759		common.logger.debug(5,"Files added to "+self.tgzNameWithPath+" : "+str(tar.getnames()))
760	+
761	+	##### ML stuff
762	+	ML_file_list=['report.py', 'DashboardAPI.py', 'Logger.py', 'ProcInfo.py', 'apmon.py']
763	+	path=os.environ['CRABDIR'] + '/python/'
764	+	for file in ML_file_list:
765	+	tar.add(path+file,file)
766	+	common.logger.debug(5,"Files added to "+self.tgzNameWithPath+" : "+str(tar.getnames()))
767	+
768	+	##### Utils
769	+	Utils_file_list=['parseCrabFjr.py','writeCfg.py', 'fillCrabFjr.py']
770	+	for file in Utils_file_list:
771	+	tar.add(path+file,file)
772	+	common.logger.debug(5,"Files added to "+self.tgzNameWithPath+" : "+str(tar.getnames()))
773	+
774	+	##### AdditionalFiles
775	+	for file in self.additional_inbox_files:
776	+	tar.add(file,string.split(file,'/')[-1])
777	+	common.logger.debug(5,"Files added to "+self.tgzNameWithPath+" : "+str(tar.getnames()))
778	+
779		tar.close()
780		except :
781		raise CrabException('Could not create tar-ball')
#	Line 847 \| Line 786 \| class Cmssw(JobType):
786		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.')
787
788		## create tar-ball with ML stuff
850	–	self.MLtgzfile = common.work_space.pathForTgz()+'share/MLfiles.tgz'
851	–	try:
852	–	tar = tarfile.open(self.MLtgzfile, "w:gz")
853	–	path=os.environ['CRABDIR'] + '/python/'
854	–	for file in ['report.py', 'DashboardAPI.py', 'Logger.py', 'ProcInfo.py', 'apmon.py', 'parseCrabFjr.py']:
855	–	tar.add(path+file,file)
856	–	common.logger.debug(5,"Files added to "+self.MLtgzfile+" : "+str(tar.getnames()))
857	–	tar.close()
858	–	except :
859	–	raise CrabException('Could not create ML files tar-ball')
860	–
861	–	return
862	–
863	–	def additionalInputFileTgz(self):
864	–	"""
865	–	Put all additional files into a tar ball and return its name
866	–	"""
867	–	import tarfile
868	–	tarName= common.work_space.pathForTgz()+'share/'+self.additional_tgz_name
869	–	tar = tarfile.open(tarName, "w:gz")
870	–	for file in self.additional_inbox_files:
871	–	tar.add(file,string.split(file,'/')[-1])
872	–	common.logger.debug(5,"Files added to "+self.additional_tgz_name+" : "+str(tar.getnames()))
873	–	tar.close()
874	–	return tarName
789
790	<	def wsSetupEnvironment(self, nj):
790	>	def wsSetupEnvironment(self, nj=0):
791		"""
792		Returns part of a job script which prepares
793		the execution environment for the job 'nj'.
794		"""
795	+	if (self.CMSSW_major >= 2 and self.CMSSW_minor >= 1) or (self.CMSSW_major >= 3):
796	+	psetName = 'pset.py'
797	+	else:
798	+	psetName = 'pset.cfg'
799		# Prepare JobType-independent part
800	<	txt = ''
801	<
802	<	## OLI_Daniele at this level middleware already known
885	<
886	<	txt += 'if [ $middleware == LCG ]; then \n'
887	<	txt += ' echo "### First set SCRAM ARCH and BUILD_ARCH to ' + self.executable_arch + ' ###"\n'
888	<	txt += ' export SCRAM_ARCH='+self.executable_arch+'\n'
889	<	txt += ' export BUILD_ARCH='+self.executable_arch+'\n'
800	>	txt = '\n#Written by cms_cmssw::wsSetupEnvironment\n'
801	>	txt += 'echo ">>> setup environment"\n'
802	>	txt += 'if [ $middleware == LCG ]; then \n'
803		txt += self.wsSetupCMSLCGEnvironment_()
804		txt += 'elif [ $middleware == OSG ]; then\n'
805		txt += ' WORKING_DIR=`/bin/mktemp -d $OSG_WN_TMP/cms_XXXXXXXXXXXX`\n'
806	<	txt += ' echo "Created working directory: $WORKING_DIR"\n'
807	<	txt += ' if [ ! -d $WORKING_DIR ] ;then\n'
808	<	txt += ' echo "SET_CMS_ENV 10016 ==> OSG $WORKING_DIR could not be created on WN `hostname`"\n'
809	<	txt += ' echo "JOB_EXIT_STATUS = 10016"\n'
897	<	txt += ' echo "JobExitCode=10016" \| tee -a $RUNTIME_AREA/$repo\n'
898	<	txt += ' dumpStatus $RUNTIME_AREA/$repo\n'
899	<	txt += ' rm -f $RUNTIME_AREA/$repo \n'
900	<	txt += ' echo "MonitorJobID=`echo $MonitorJobID`" \| tee -a $RUNTIME_AREA/$repo \n'
901	<	txt += ' echo "MonitorID=`echo $MonitorID`" \| tee -a $RUNTIME_AREA/$repo\n'
902	<	txt += ' exit 1\n'
806	>	txt += ' if [ ! $? == 0 ] ;then\n'
807	>	txt += ' echo "ERROR ==> OSG $WORKING_DIR could not be created on WN `hostname`"\n'
808	>	txt += ' job_exit_code=10016\n'
809	>	txt += ' func_exit\n'
810		txt += ' fi\n'
811	+	txt += ' echo ">>> Created working directory: $WORKING_DIR"\n'
812		txt += '\n'
813		txt += ' echo "Change to working directory: $WORKING_DIR"\n'
814		txt += ' cd $WORKING_DIR\n'
815	<	txt += self.wsSetupCMSOSGEnvironment_()
816	<	txt += ' echo "### Set SCRAM ARCH to ' + self.executable_arch + ' ###"\n'
909	<	txt += ' export SCRAM_ARCH='+self.executable_arch+'\n'
815	>	txt += ' echo ">>> current directory (WORKING_DIR): $WORKING_DIR"\n'
816	>	txt += self.wsSetupCMSOSGEnvironment_()
817		txt += 'fi\n'
818
819		# Prepare JobType-specific part
820		scram = self.scram.commandName()
821		txt += '\n\n'
822	<	txt += 'echo "### SPECIFIC JOB SETUP ENVIRONMENT ###"\n'
822	>	txt += 'echo ">>> specific cmssw setup environment:"\n'
823	>	txt += 'echo "CMSSW_VERSION = '+self.version+'"\n'
824		txt += scram+' project CMSSW '+self.version+'\n'
825		txt += 'status=$?\n'
826		txt += 'if [ $status != 0 ] ; then\n'
827	<	txt += ' echo "SET_EXE_ENV 10034 ==>ERROR CMSSW '+self.version+' not found on `hostname`" \n'
828	<	txt += ' echo "JOB_EXIT_STATUS = 10034"\n'
829	<	txt += ' echo "JobExitCode=10034" \| tee -a $RUNTIME_AREA/$repo\n'
922	<	txt += ' dumpStatus $RUNTIME_AREA/$repo\n'
923	<	txt += ' rm -f $RUNTIME_AREA/$repo \n'
924	<	txt += ' echo "MonitorJobID=`echo $MonitorJobID`" \| tee -a $RUNTIME_AREA/$repo \n'
925	<	txt += ' echo "MonitorID=`echo $MonitorID`" \| tee -a $RUNTIME_AREA/$repo\n'
926	<	## OLI_Daniele
927	<	txt += ' if [ $middleware == OSG ]; then \n'
928	<	txt += ' echo "Remove working directory: $WORKING_DIR"\n'
929	<	txt += ' cd $RUNTIME_AREA\n'
930	<	txt += ' /bin/rm -rf $WORKING_DIR\n'
931	<	txt += ' if [ -d $WORKING_DIR ] ;then\n'
932	<	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'
933	<	txt += ' echo "JOB_EXIT_STATUS = 10018"\n'
934	<	txt += ' echo "JobExitCode=10018" \| tee -a $RUNTIME_AREA/$repo\n'
935	<	txt += ' dumpStatus $RUNTIME_AREA/$repo\n'
936	<	txt += ' rm -f $RUNTIME_AREA/$repo \n'
937	<	txt += ' echo "MonitorJobID=`echo $MonitorJobID`" \| tee -a $RUNTIME_AREA/$repo \n'
938	<	txt += ' echo "MonitorID=`echo $MonitorID`" \| tee -a $RUNTIME_AREA/$repo\n'
939	<	txt += ' fi\n'
940	<	txt += ' fi \n'
941	<	txt += ' exit 1 \n'
827	>	txt += ' echo "ERROR ==> CMSSW '+self.version+' not found on `hostname`" \n'
828	>	txt += ' job_exit_code=10034\n'
829	>	txt += ' func_exit\n'
830		txt += 'fi \n'
943	–	txt += 'echo "CMSSW_VERSION = '+self.version+'"\n'
831		txt += 'cd '+self.version+'\n'
945	–	########## FEDE FOR DBS2 ######################
832		txt += 'SOFTWARE_DIR=`pwd`\n'
833	<	txt += 'echo SOFTWARE_DIR=$SOFTWARE_DIR \n'
948	<	###############################################
949	<	### needed grep for bug in scramv1 ###
950	<	txt += scram+' runtime -sh\n'
833	>	txt += 'echo ">>> current directory (SOFTWARE_DIR): $SOFTWARE_DIR" \n'
834		txt += 'eval `'+scram+' runtime -sh \| grep -v SCRAMRT_LSB_JOBNAME`\n'
835	<	txt += 'echo $PATH\n'
836	<
835	>	txt += 'if [ $? != 0 ] ; then\n'
836	>	txt += ' echo "ERROR ==> Problem with the command: "\n'
837	>	txt += ' echo "eval \`'+scram+' runtime -sh \| grep -v SCRAMRT_LSB_JOBNAME \` at `hostname`"\n'
838	>	txt += ' job_exit_code=10034\n'
839	>	txt += ' func_exit\n'
840	>	txt += 'fi \n'
841		# Handle the arguments:
842		txt += "\n"
843		txt += "## number of arguments (first argument always jobnumber)\n"
844		txt += "\n"
845	<	# txt += "narg=$#\n"
959	<	txt += "if [ $nargs -lt 2 ]\n"
845	>	txt += "if [ $nargs -lt "+str(self.argsList)+" ]\n"
846		txt += "then\n"
847	<	txt += " echo 'SET_EXE_ENV 1 ==> ERROR Too few arguments' +$nargs+ \n"
848	<	txt += ' echo "JOB_EXIT_STATUS = 50113"\n'
849	<	txt += ' echo "JobExitCode=50113" \| tee -a $RUNTIME_AREA/$repo\n'
964	<	txt += ' dumpStatus $RUNTIME_AREA/$repo\n'
965	<	txt += ' rm -f $RUNTIME_AREA/$repo \n'
966	<	txt += ' echo "MonitorJobID=`echo $MonitorJobID`" \| tee -a $RUNTIME_AREA/$repo \n'
967	<	txt += ' echo "MonitorID=`echo $MonitorID`" \| tee -a $RUNTIME_AREA/$repo\n'
968	<	## OLI_Daniele
969	<	txt += ' if [ $middleware == OSG ]; then \n'
970	<	txt += ' echo "Remove working directory: $WORKING_DIR"\n'
971	<	txt += ' cd $RUNTIME_AREA\n'
972	<	txt += ' /bin/rm -rf $WORKING_DIR\n'
973	<	txt += ' if [ -d $WORKING_DIR ] ;then\n'
974	<	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'
975	<	txt += ' echo "JOB_EXIT_STATUS = 50114"\n'
976	<	txt += ' echo "JobExitCode=50114" \| tee -a $RUNTIME_AREA/$repo\n'
977	<	txt += ' dumpStatus $RUNTIME_AREA/$repo\n'
978	<	txt += ' rm -f $RUNTIME_AREA/$repo \n'
979	<	txt += ' echo "MonitorJobID=`echo $MonitorJobID`" \| tee -a $RUNTIME_AREA/$repo \n'
980	<	txt += ' echo "MonitorID=`echo $MonitorID`" \| tee -a $RUNTIME_AREA/$repo\n'
981	<	txt += ' fi\n'
982	<	txt += ' fi \n'
983	<	txt += " exit 1\n"
847	>	txt += " echo 'ERROR ==> Too few arguments' +$nargs+ \n"
848	>	txt += ' job_exit_code=50113\n'
849	>	txt += " func_exit\n"
850		txt += "fi\n"
851		txt += "\n"
852
853		# Prepare job-specific part
854		job = common.job_list[nj]
855	<	### FEDE FOR DBS OUTPUT PUBLICATION
990	<	if (self.datasetPath):
855	>	if (self.datasetPath):
856		txt += '\n'
857		txt += 'DatasetPath='+self.datasetPath+'\n'
858
859		datasetpath_split = self.datasetPath.split("/")
860	<
860	>
861		txt += 'PrimaryDataset='+datasetpath_split[1]+'\n'
862		txt += 'DataTier='+datasetpath_split[2]+'\n'
998	–	#txt += 'ProcessedDataset='+datasetpath_split[3]+'\n'
863		txt += 'ApplicationFamily=cmsRun\n'
864
865		else:
866		txt += 'DatasetPath=MCDataTier\n'
867		txt += 'PrimaryDataset=null\n'
868		txt += 'DataTier=null\n'
1005	–	#txt += 'ProcessedDataset=null\n'
869		txt += 'ApplicationFamily=MCDataTier\n'
870	<	if self.pset != None: #CarlosDaniele
870	>	if self.pset != None:
871		pset = os.path.basename(job.configFilename())
872		txt += '\n'
873		txt += 'cp $RUNTIME_AREA/'+pset+' .\n'
874		if (self.datasetPath): # standard job
875	<	#txt += 'InputFiles=$2\n'
876	<	txt += 'InputFiles=${args[1]}\n'
877	<	txt += 'MaxEvents=${args[2]}\n'
878	<	txt += 'SkipEvents=${args[3]}\n'
875	>	txt += 'InputFiles=${args[1]}; export InputFiles\n'
876	>	if (self.useParent):
877	>	txt += 'ParentFiles=${args[2]}; export ParentFiles\n'
878	>	txt += 'MaxEvents=${args[3]}; export MaxEvents\n'
879	>	txt += 'SkipEvents=${args[4]}; export SkipEvents\n'
880	>	else:
881	>	txt += 'MaxEvents=${args[2]}; export MaxEvents\n'
882	>	txt += 'SkipEvents=${args[3]}; export SkipEvents\n'
883		txt += 'echo "Inputfiles:<$InputFiles>"\n'
884	<	txt += 'sed "s#{\'INPUT\'}#$InputFiles#" '+pset+' > tmp && mv -f tmp '+pset+'\n'
884	>	if (self.useParent): txt += 'echo "ParentFiles:<$ParentFiles>"\n'
885		txt += 'echo "MaxEvents:<$MaxEvents>"\n'
1019	–	txt += 'sed "s#INPUTMAXEVENTS#$MaxEvents#" '+pset+' > tmp && mv -f tmp '+pset+'\n'
886		txt += 'echo "SkipEvents:<$SkipEvents>"\n'
1021	–	txt += 'sed "s#INPUTSKIPEVENTS#$SkipEvents#" '+pset+' > tmp && mv -f tmp '+pset+'\n'
887		else: # pythia like job
888	<	seedIndex=1
888	>	txt += 'PreserveSeeds=' + ','.join(self.preserveSeeds) + '; export PreserveSeeds\n'
889	>	txt += 'IncrementSeeds=' + ','.join(self.incrementSeeds) + '; export IncrementSeeds\n'
890	>	txt += 'echo "PreserveSeeds: <$PreserveSeeds>"\n'
891	>	txt += 'echo "IncrementSeeds:<$IncrementSeeds>"\n'
892		if (self.firstRun):
893	<	txt += 'FirstRun=${args['+str(seedIndex)+']}\n'
893	>	txt += 'FirstRun=${args[1]}; export FirstRun\n'
894		txt += 'echo "FirstRun: <$FirstRun>"\n'
1027	–	txt += 'sed "s#\<INPUTFIRSTRUN\>#$FirstRun#" '+pset+' > tmp && mv -f tmp '+pset+'\n'
1028	–	seedIndex=seedIndex+1
895
896	<	if (self.sourceSeed):
1031	<	txt += 'Seed=${args['+str(seedIndex)+']}\n'
1032	<	txt += 'sed "s#\<INPUT\>#$Seed#" '+pset+' > tmp && mv -f tmp '+pset+'\n'
1033	<	seedIndex=seedIndex+1
1034	<	## the following seeds are not always present
1035	<	if (self.sourceSeedVtx):
1036	<	txt += 'VtxSeed=${args['+str(seedIndex)+']}\n'
1037	<	txt += 'echo "VtxSeed: <$VtxSeed>"\n'
1038	<	txt += 'sed "s#\<INPUTVTX\>#$VtxSeed#" '+pset+' > tmp && mv -f tmp '+pset+'\n'
1039	<	seedIndex += 1
1040	<	if (self.sourceSeedG4):
1041	<	txt += 'G4Seed=${args['+str(seedIndex)+']}\n'
1042	<	txt += 'echo "G4Seed: <$G4Seed>"\n'
1043	<	txt += 'sed "s#\<INPUTG4\>#$G4Seed#" '+pset+' > tmp && mv -f tmp '+pset+'\n'
1044	<	seedIndex += 1
1045	<	if (self.sourceSeedMix):
1046	<	txt += 'mixSeed=${args['+str(seedIndex)+']}\n'
1047	<	txt += 'echo "MixSeed: <$mixSeed>"\n'
1048	<	txt += 'sed "s#\<INPUTMIX\>#$mixSeed#" '+pset+' > tmp && mv -f tmp '+pset+'\n'
1049	<	seedIndex += 1
1050	<	pass
1051	<	pass
1052	<	txt += 'mv -f '+pset+' pset.cfg\n'
896	>	txt += 'mv -f ' + pset + ' ' + psetName + '\n'
897
1054	–	if len(self.additional_inbox_files) > 0:
1055	–	txt += 'if [ -e $RUNTIME_AREA/'+self.additional_tgz_name+' ] ; then\n'
1056	–	txt += ' tar xzvf $RUNTIME_AREA/'+self.additional_tgz_name+'\n'
1057	–	txt += 'fi\n'
1058	–	pass
898
899	<	if self.pset != None: #CarlosDaniele
900	<	txt += 'echo "### END JOB SETUP ENVIRONMENT ###"\n\n'
1062	<
1063	<	txt += '\n'
1064	<	txt += 'echo "*** cat pset.cfg *******"\n'
1065	<	txt += 'cat pset.cfg\n'
1066	<	txt += 'echo "**** end pset.cfg ******"\n'
899	>	if self.pset != None:
900	>	# FUTURE: Can simply for 2_1_x and higher
901		txt += '\n'
902	<	### FEDE FOR DBS OUTPUT PUBLICATION
903	<	txt += 'PSETHASH=`EdmConfigHash < pset.cfg` \n'
902	>	if self.debug_wrapper==True:
903	>	txt += 'echo "*** cat ' + psetName + ' *******"\n'
904	>	txt += 'cat ' + psetName + '\n'
905	>	txt += 'echo "**** end ' + psetName + ' ******"\n'
906	>	txt += '\n'
907	>	txt += 'PSETHASH=`edmConfigHash < ' + psetName + '` \n'
908		txt += 'echo "PSETHASH = $PSETHASH" \n'
1071	–	##############
909		txt += '\n'
1073	–	# txt += 'echo "*** cat pset1.cfg *******"\n'
1074	–	# txt += 'cat pset1.cfg\n'
1075	–	# txt += 'echo "**** end pset1.cfg ******"\n'
910		return txt
911
912	<	def wsBuildExe(self, nj=0):
912	>	def wsUntarSoftware(self, nj=0):
913		"""
914		Put in the script the commands to build an executable
915		or a library.
916		"""
917
918	<	txt = ""
918	>	txt = '\n#Written by cms_cmssw::wsUntarSoftware\n'
919
920		if os.path.isfile(self.tgzNameWithPath):
921	<	txt += 'echo "tar xzvf $RUNTIME_AREA/'+os.path.basename(self.tgzNameWithPath)+'"\n'
921	>	txt += 'echo ">>> tar xzvf $RUNTIME_AREA/'+os.path.basename(self.tgzNameWithPath)+' :" \n'
922		txt += 'tar xzvf $RUNTIME_AREA/'+os.path.basename(self.tgzNameWithPath)+'\n'
923	+	if self.debug_wrapper:
924	+	txt += 'ls -Al \n'
925		txt += 'untar_status=$? \n'
926		txt += 'if [ $untar_status -ne 0 ]; then \n'
927	<	txt += ' echo "SET_EXE 1 ==> ERROR Untarring .tgz file failed"\n'
928	<	txt += ' echo "JOB_EXIT_STATUS = $untar_status" \n'
929	<	txt += ' echo "JobExitCode=$untar_status" \| tee -a $RUNTIME_AREA/$repo\n'
1094	<	txt += ' if [ $middleware == OSG ]; then \n'
1095	<	txt += ' echo "Remove working directory: $WORKING_DIR"\n'
1096	<	txt += ' cd $RUNTIME_AREA\n'
1097	<	txt += ' /bin/rm -rf $WORKING_DIR\n'
1098	<	txt += ' if [ -d $WORKING_DIR ] ;then\n'
1099	<	txt += ' echo "SET_EXE 50999 ==> OSG $WORKING_DIR could not be deleted on WN `hostname` after Untarring .tgz file failed"\n'
1100	<	txt += ' echo "JOB_EXIT_STATUS = 50999"\n'
1101	<	txt += ' echo "JobExitCode=50999" \| tee -a $RUNTIME_AREA/$repo\n'
1102	<	txt += ' dumpStatus $RUNTIME_AREA/$repo\n'
1103	<	txt += ' rm -f $RUNTIME_AREA/$repo \n'
1104	<	txt += ' echo "MonitorJobID=`echo $MonitorJobID`" \| tee -a $RUNTIME_AREA/$repo \n'
1105	<	txt += ' echo "MonitorID=`echo $MonitorID`" \| tee -a $RUNTIME_AREA/$repo\n'
1106	<	txt += ' fi\n'
1107	<	txt += ' fi \n'
1108	<	txt += ' \n'
1109	<	txt += ' exit 1 \n'
927	>	txt += ' echo "ERROR ==> Untarring .tgz file failed"\n'
928	>	txt += ' job_exit_code=$untar_status\n'
929	>	txt += ' func_exit\n'
930		txt += 'else \n'
931		txt += ' echo "Successful untar" \n'
932		txt += 'fi \n'
933		txt += '\n'
934	<	txt += 'echo "Include ProdAgentApi and PRODCOMMON in PYTHONPATH"\n'
934	>	txt += 'echo ">>> Include ProdCommon in PYTHONPATH:"\n'
935		txt += 'if [ -z "$PYTHONPATH" ]; then\n'
936	<	#### FEDE FOR DBS OUTPUT PUBLICATION
1117	<	txt += ' export PYTHONPATH=$SOFTWARE_DIR/ProdAgentApi:$SOFTWARE_DIR/ProdCommon\n'
1118	<	#txt += ' export PYTHONPATH=`pwd`/ProdAgentApi:`pwd`/ProdCommon\n'
1119	<	#txt += ' export PYTHONPATH=ProdAgentApi\n'
936	>	txt += ' export PYTHONPATH=$RUNTIME_AREA/ProdCommon\n'
937		txt += 'else\n'
938	<	txt += ' export PYTHONPATH=$SOFTWARE_DIR/ProdAgentApi:$SOFTWARE_DIR/ProdCommon:${PYTHONPATH}\n'
1122	<	#txt += ' export PYTHONPATH=`pwd`/ProdAgentApi:`pwd`/ProdCommon:${PYTHONPATH}\n'
1123	<	#txt += ' export PYTHONPATH=ProdAgentApi:${PYTHONPATH}\n'
938	>	txt += ' export PYTHONPATH=$RUNTIME_AREA/ProdCommon:${PYTHONPATH}\n'
939		txt += 'echo "PYTHONPATH=$PYTHONPATH"\n'
1125	–	###################
940		txt += 'fi\n'
941		txt += '\n'
942
943		pass
944	<
944	>
945	>	return txt
946	>
947	>	def wsBuildExe(self, nj=0):
948	>	"""
949	>	Put in the script the commands to build an executable
950	>	or a library.
951	>	"""
952	>
953	>	txt = '\n#Written by cms_cmssw::wsBuildExe\n'
954	>	txt += 'echo ">>> moving CMSSW software directories in `pwd`" \n'
955	>
956	>	txt += 'rm -r lib/ module/ \n'
957	>	txt += 'mv $RUNTIME_AREA/lib/ . \n'
958	>	txt += 'mv $RUNTIME_AREA/module/ . \n'
959	>	if self.dataExist == True:
960	>	txt += 'rm -r src/ \n'
961	>	txt += 'mv $RUNTIME_AREA/src/ . \n'
962	>	if len(self.additional_inbox_files)>0:
963	>	for file in self.additional_inbox_files:
964	>	txt += 'mv $RUNTIME_AREA/'+os.path.basename(file)+' . \n'
965	>	txt += 'mv $RUNTIME_AREA/ProdCommon/ . \n'
966	>
967	>	txt += 'if [ -z "$PYTHONPATH" ]; then\n'
968	>	txt += ' export PYTHONPATH=$SOFTWARE_DIR/ProdCommon\n'
969	>	txt += 'else\n'
970	>	txt += ' export PYTHONPATH=$SOFTWARE_DIR/ProdCommon:${PYTHONPATH}\n'
971	>	txt += 'echo "PYTHONPATH=$PYTHONPATH"\n'
972	>	txt += 'fi\n'
973	>	txt += '\n'
974	>
975		return txt
976
977		def modifySteeringCards(self, nj):
978		"""
979	<	modify the card provided by the user,
979	>	modify the card provided by the user,
980		writing a new card into share dir
981		"""
982	<
982	>
983		def executableName(self):
984	<	if self.scriptExe: #CarlosDaniele
984	>	if self.scriptExe:
985		return "sh "
986		else:
987		return self.executable
988
989		def executableArgs(self):
990	+	# FUTURE: This function tests the CMSSW version. Can be simplified as we drop support for old versions
991		if self.scriptExe:#CarlosDaniele
992		return self.scriptExe + " $NJob"
993		else:
994	<	# if >= CMSSW_1_5_X, add -e
995	<	version_array = self.scram.getSWVersion().split('_')
996	<	major = 0
997	<	minor = 0
998	<	try:
999	<	major = int(version_array[1])
1000	<	minor = int(version_array[2])
1001	<	except:
1157	<	msg = "Cannot parse CMSSW version string: " + "_".join(version_array) + " for major and minor release number!"
1158	<	raise CrabException(msg)
1159	<	if major >= 1 and minor >= 5 :
1160	<	return " -e -p pset.cfg"
994	>	ex_args = ""
995	>	# FUTURE: This tests the CMSSW version. Can remove code as versions deprecated
996	>	# Framework job report
997	>	if (self.CMSSW_major >= 1 and self.CMSSW_minor >= 5) or (self.CMSSW_major >= 2):
998	>	ex_args += " -j $RUNTIME_AREA/crab_fjr_$NJob.xml"
999	>	# Type of config file
1000	>	if self.CMSSW_major >= 2 :
1001	>	ex_args += " -p pset.py"
1002		else:
1003	<	return " -p pset.cfg"
1003	>	ex_args += " -p pset.cfg"
1004	>	return ex_args
1005
1006		def inputSandbox(self, nj):
1007		"""
1008		Returns a list of filenames to be put in JDL input sandbox.
1009		"""
1010		inp_box = []
1169	–	# # dict added to delete duplicate from input sandbox file list
1170	–	# seen = {}
1171	–	## code
1011		if os.path.isfile(self.tgzNameWithPath):
1012		inp_box.append(self.tgzNameWithPath)
1013	<	if os.path.isfile(self.MLtgzfile):
1014	<	inp_box.append(self.MLtgzfile)
1176	<	## config
1177	<	if not self.pset is None:
1178	<	inp_box.append(common.work_space.pathForTgz() + 'job/' + self.configFilename())
1179	<	## additional input files
1180	<	tgz = self.additionalInputFileTgz()
1181	<	inp_box.append(tgz)
1013	>	wrapper = os.path.basename(str(common._db.queryTask('scriptName')))
1014	>	inp_box.append(common.work_space.pathForTgz() +'job/'+ wrapper)
1015		return inp_box
1016
1017		def outputSandbox(self, nj):
#	Line 1189 \| Line 1022 \| class Cmssw(JobType):
1022
1023		## User Declared output files
1024		for out in (self.output_file+self.output_file_sandbox):
1025	<	n_out = nj + 1
1025	>	n_out = nj + 1
1026		out_box.append(self.numberFile_(out,str(n_out)))
1027		return out_box
1028
#	Line 1204 \| Line 1037 \| class Cmssw(JobType):
1037		Returns part of a job script which renames the produced files.
1038		"""
1039
1040	<	txt = '\n'
1041	<	txt += '# directory content\n'
1042	<	txt += 'ls \n'
1043	<
1044	<	txt += 'output_exit_status=0\n'
1045	<
1046	<	for fileWithSuffix in (self.output_file_sandbox):
1214	<	output_file_num = self.numberFile_(fileWithSuffix, '$NJob')
1215	<	txt += '\n'
1216	<	txt += '# check output file\n'
1217	<	txt += 'if [ -e ./'+fileWithSuffix+' ] ; then\n'
1218	<	txt += ' mv '+fileWithSuffix+' $RUNTIME_AREA\n'
1219	<	txt += ' cp $RUNTIME_AREA/'+fileWithSuffix+' $RUNTIME_AREA/'+output_file_num+'\n'
1220	<	txt += 'else\n'
1221	<	txt += ' exit_status=60302\n'
1222	<	txt += ' echo "ERROR: Problem with output file '+fileWithSuffix+'"\n'
1223	<	if common.scheduler.boss_scheduler_name == 'condor_g':
1224	<	txt += ' if [ $middleware == OSG ]; then \n'
1225	<	txt += ' echo "prepare dummy output file"\n'
1226	<	txt += ' echo "Processing of job output failed" > $RUNTIME_AREA/'+output_file_num+'\n'
1227	<	txt += ' fi \n'
1228	<	txt += 'fi\n'
1229	<
1040	>	txt = '\n#Written by cms_cmssw::wsRenameOutput\n'
1041	>	txt += 'echo ">>> current directory (SOFTWARE_DIR): $SOFTWARE_DIR" \n'
1042	>	txt += 'echo ">>> current directory content:"\n'
1043	>	if self.debug_wrapper:
1044	>	txt += 'ls -Al\n'
1045	>	txt += '\n'
1046	>
1047		for fileWithSuffix in (self.output_file):
1048		output_file_num = self.numberFile_(fileWithSuffix, '$NJob')
1049		txt += '\n'
1050		txt += '# check output file\n'
1051		txt += 'if [ -e ./'+fileWithSuffix+' ] ; then\n'
1052	<	txt += ' mv '+fileWithSuffix+' $RUNTIME_AREA\n'
1053	<	txt += ' cp $RUNTIME_AREA/'+fileWithSuffix+' $RUNTIME_AREA/'+output_file_num+'\n'
1052	>	if (self.copy_data == 1): # For OSG nodes, file is in $WORKING_DIR, should not be moved to $RUNTIME_AREA
1053	>	txt += ' mv '+fileWithSuffix+' '+output_file_num+'\n'
1054	>	#txt += ' ln -s `pwd`/'+output_file_num+' $RUNTIME_AREA/'+fileWithSuffix+'\n'
1055	>	else:
1056	>	txt += ' mv '+fileWithSuffix+' $RUNTIME_AREA/'+output_file_num+'\n'
1057	>	txt += ' ln -s $RUNTIME_AREA/'+output_file_num+' $RUNTIME_AREA/'+fileWithSuffix+'\n'
1058		txt += 'else\n'
1059	<	txt += ' exit_status=60302\n'
1060	<	txt += ' echo "ERROR: Problem with output file '+fileWithSuffix+'"\n'
1061	<	txt += ' echo "JOB_EXIT_STATUS = $exit_status"\n'
1241	<	txt += ' output_exit_status=$exit_status\n'
1242	<	if common.scheduler.boss_scheduler_name == 'condor_g':
1059	>	txt += ' job_exit_code=60302\n'
1060	>	txt += ' echo "WARNING: Output file '+fileWithSuffix+' not found"\n'
1061	>	if common.scheduler.name().upper() == 'CONDOR_G':
1062		txt += ' if [ $middleware == OSG ]; then \n'
1063		txt += ' echo "prepare dummy output file"\n'
1064		txt += ' echo "Processing of job output failed" > $RUNTIME_AREA/'+output_file_num+'\n'
#	Line 1248 \| Line 1067 \| class Cmssw(JobType):
1067		file_list = []
1068		for fileWithSuffix in (self.output_file):
1069		file_list.append(self.numberFile_(fileWithSuffix, '$NJob'))
1070	<
1070	>
1071		txt += 'file_list="'+string.join(file_list,' ')+'"\n'
1072	+	txt += '\n'
1073	+	txt += 'echo ">>> current directory (SOFTWARE_DIR): $SOFTWARE_DIR" \n'
1074	+	txt += 'echo ">>> current directory content:"\n'
1075	+	if self.debug_wrapper:
1076	+	txt += 'ls -Al\n'
1077	+	txt += '\n'
1078		txt += 'cd $RUNTIME_AREA\n'
1079	+	txt += 'echo ">>> current directory (RUNTIME_AREA): $RUNTIME_AREA"\n'
1080		return txt
1081
1082		def numberFile_(self, file, txt):
#	Line 1268 \| Line 1094 \| class Cmssw(JobType):
1094		result = name + '_' + txt + "." + ext
1095		else:
1096		result = name + '_' + txt
1097	<
1097	>
1098		return result
1099
1100		def getRequirements(self, nj=[]):
1101		"""
1102	<	return job requirements to add to jdl files
1102	>	return job requirements to add to jdl files
1103		"""
1104		req = ''
1105		if self.version:
1106		req='Member("VO-cms-' + \
1107		self.version + \
1108		'", other.GlueHostApplicationSoftwareRunTimeEnvironment)'
1109	<	## SL add requirement for OS version only if SL4
1284	<	#reSL4 = re.compile( r'slc4' )
1285	<	if self.executable_arch: # and reSL4.search(self.executable_arch):
1109	>	if self.executable_arch:
1110		req+=' && Member("VO-cms-' + \
1111		self.executable_arch + \
1112		'", other.GlueHostApplicationSoftwareRunTimeEnvironment)'
1113
1114		req = req + ' && (other.GlueHostNetworkAdapterOutboundIP)'
1115	+	if common.scheduler.name() == "glitecoll":
1116	+	req += ' && other.GlueCEStateStatus == "Production" '
1117
1118		return req
1119
1120		def configFilename(self):
1121		""" return the config filename """
1122	<	return self.name()+'.cfg'
1122	>	# FUTURE: Can remove cfg mode for CMSSW >= 2_1_x
1123	>	if (self.CMSSW_major >= 2 and self.CMSSW_minor >= 1) or (self.CMSSW_major >= 3):
1124	>	return self.name()+'.py'
1125	>	else:
1126	>	return self.name()+'.cfg'
1127
1298	–	### OLI_DANIELE
1128		def wsSetupCMSOSGEnvironment_(self):
1129		"""
1130		Returns part of a job script which is prepares
1131		the execution environment and which is common for all CMS jobs.
1132		"""
1133	<	txt = '\n'
1134	<	txt += ' echo "### SETUP CMS OSG ENVIRONMENT ###"\n'
1135	<	txt += ' if [ -f $GRID3_APP_DIR/cmssoft/cmsset_default.sh ] ;then\n'
1136	<	txt += ' # Use $GRID3_APP_DIR/cmssoft/cmsset_default.sh to setup cms software\n'
1137	<	txt += ' export SCRAM_ARCH='+self.executable_arch+'\n'
1138	<	txt += ' source $GRID3_APP_DIR/cmssoft/cmsset_default.sh '+self.version+'\n'
1310	<	txt += ' elif [ -f $OSG_APP/cmssoft/cms/cmsset_default.sh ] ;then\n'
1133	>	txt = '\n#Written by cms_cmssw::wsSetupCMSOSGEnvironment_\n'
1134	>	txt += ' echo ">>> setup CMS OSG environment:"\n'
1135	>	txt += ' echo "set SCRAM ARCH to ' + self.executable_arch + '"\n'
1136	>	txt += ' export SCRAM_ARCH='+self.executable_arch+'\n'
1137	>	txt += ' echo "SCRAM_ARCH = $SCRAM_ARCH"\n'
1138	>	txt += ' if [ -f $OSG_APP/cmssoft/cms/cmsset_default.sh ] ;then\n'
1139		txt += ' # Use $OSG_APP/cmssoft/cms/cmsset_default.sh to setup cms software\n'
1140	<	txt += ' export SCRAM_ARCH='+self.executable_arch+'\n'
1141	<	txt += ' source $OSG_APP/cmssoft/cms/cmsset_default.sh '+self.version+'\n'
1142	<	txt += ' else\n'
1143	<	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'
1144	<	txt += ' echo "JOB_EXIT_STATUS = 10020"\n'
1145	<	txt += ' echo "JobExitCode=10020" \| tee -a $RUNTIME_AREA/$repo\n'
1318	<	txt += ' dumpStatus $RUNTIME_AREA/$repo\n'
1319	<	txt += ' rm -f $RUNTIME_AREA/$repo \n'
1320	<	txt += ' echo "MonitorJobID=`echo $MonitorJobID`" \| tee -a $RUNTIME_AREA/$repo \n'
1321	<	txt += ' echo "MonitorID=`echo $MonitorID`" \| tee -a $RUNTIME_AREA/$repo\n'
1322	<	txt += ' exit 1\n'
1323	<	txt += '\n'
1324	<	txt += ' echo "Remove working directory: $WORKING_DIR"\n'
1325	<	txt += ' cd $RUNTIME_AREA\n'
1326	<	txt += ' /bin/rm -rf $WORKING_DIR\n'
1327	<	txt += ' if [ -d $WORKING_DIR ] ;then\n'
1328	<	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'
1329	<	txt += ' echo "JOB_EXIT_STATUS = 10017"\n'
1330	<	txt += ' echo "JobExitCode=10017" \| tee -a $RUNTIME_AREA/$repo\n'
1331	<	txt += ' dumpStatus $RUNTIME_AREA/$repo\n'
1332	<	txt += ' rm -f $RUNTIME_AREA/$repo \n'
1333	<	txt += ' echo "MonitorJobID=`echo $MonitorJobID`" \| tee -a $RUNTIME_AREA/$repo \n'
1334	<	txt += ' echo "MonitorID=`echo $MonitorID`" \| tee -a $RUNTIME_AREA/$repo\n'
1335	<	txt += ' fi\n'
1336	<	txt += '\n'
1337	<	txt += ' exit 1\n'
1338	<	txt += ' fi\n'
1140	>	txt += ' source $OSG_APP/cmssoft/cms/cmsset_default.sh '+self.version+'\n'
1141	>	txt += ' else\n'
1142	>	txt += ' echo "ERROR ==> $OSG_APP/cmssoft/cms/cmsset_default.sh file not found"\n'
1143	>	txt += ' job_exit_code=10020\n'
1144	>	txt += ' func_exit\n'
1145	>	txt += ' fi\n'
1146		txt += '\n'
1147	<	txt += ' echo "SET_CMS_ENV 0 ==> setup cms environment ok"\n'
1148	<	txt += ' echo " END SETUP CMS OSG ENVIRONMENT "\n'
1147	>	txt += ' echo "==> setup cms environment ok"\n'
1148	>	txt += ' echo "SCRAM_ARCH = $SCRAM_ARCH"\n'
1149
1150		return txt
1151	<
1345	<	### OLI_DANIELE
1151	>
1152		def wsSetupCMSLCGEnvironment_(self):
1153		"""
1154		Returns part of a job script which is prepares
1155		the execution environment and which is common for all CMS jobs.
1156		"""
1157	<	txt = ' \n'
1158	<	txt += ' echo " ### SETUP CMS LCG ENVIRONMENT ### "\n'
1159	<	txt += ' if [ ! $VO_CMS_SW_DIR ] ;then\n'
1160	<	txt += ' echo "SET_CMS_ENV 10031 ==> ERROR CMS software dir not found on WN `hostname`"\n'
1161	<	txt += ' echo "JOB_EXIT_STATUS = 10031" \n'
1162	<	txt += ' echo "JobExitCode=10031" \| tee -a $RUNTIME_AREA/$repo\n'
1163	<	txt += ' dumpStatus $RUNTIME_AREA/$repo\n'
1164	<	txt += ' rm -f $RUNTIME_AREA/$repo \n'
1165	<	txt += ' echo "MonitorJobID=`echo $MonitorJobID`" \| tee -a $RUNTIME_AREA/$repo \n'
1166	<	txt += ' echo "MonitorID=`echo $MonitorID`" \| tee -a $RUNTIME_AREA/$repo\n'
1167	<	txt += ' exit 1\n'
1168	<	txt += ' else\n'
1169	<	txt += ' echo "Sourcing environment... "\n'
1170	<	txt += ' if [ ! -s $VO_CMS_SW_DIR/cmsset_default.sh ] ;then\n'
1171	<	txt += ' echo "SET_CMS_ENV 10020 ==> ERROR cmsset_default.sh file not found into dir $VO_CMS_SW_DIR"\n'
1172	<	txt += ' echo "JOB_EXIT_STATUS = 10020"\n'
1173	<	txt += ' echo "JobExitCode=10020" \| tee -a $RUNTIME_AREA/$repo\n'
1174	<	txt += ' dumpStatus $RUNTIME_AREA/$repo\n'
1175	<	txt += ' rm -f $RUNTIME_AREA/$repo \n'
1176	<	txt += ' echo "MonitorJobID=`echo $MonitorJobID`" \| tee -a $RUNTIME_AREA/$repo \n'
1177	<	txt += ' echo "MonitorID=`echo $MonitorID`" \| tee -a $RUNTIME_AREA/$repo\n'
1178	<	txt += ' exit 1\n'
1179	<	txt += ' fi\n'
1180	<	txt += ' echo "sourcing $VO_CMS_SW_DIR/cmsset_default.sh"\n'
1181	<	txt += ' source $VO_CMS_SW_DIR/cmsset_default.sh\n'
1182	<	txt += ' result=$?\n'
1183	<	txt += ' if [ $result -ne 0 ]; then\n'
1378	<	txt += ' echo "SET_CMS_ENV 10032 ==> ERROR problem sourcing $VO_CMS_SW_DIR/cmsset_default.sh"\n'
1379	<	txt += ' echo "JOB_EXIT_STATUS = 10032"\n'
1380	<	txt += ' echo "JobExitCode=10032" \| tee -a $RUNTIME_AREA/$repo\n'
1381	<	txt += ' dumpStatus $RUNTIME_AREA/$repo\n'
1382	<	txt += ' rm -f $RUNTIME_AREA/$repo \n'
1383	<	txt += ' echo "MonitorJobID=`echo $MonitorJobID`" \| tee -a $RUNTIME_AREA/$repo \n'
1384	<	txt += ' echo "MonitorID=`echo $MonitorID`" \| tee -a $RUNTIME_AREA/$repo\n'
1385	<	txt += ' exit 1\n'
1386	<	txt += ' fi\n'
1387	<	txt += ' fi\n'
1388	<	txt += ' \n'
1389	<	txt += ' echo "SET_CMS_ENV 0 ==> setup cms environment ok"\n'
1390	<	txt += ' echo "### END SETUP CMS LCG ENVIRONMENT ###"\n'
1157	>	txt = '\n#Written by cms_cmssw::wsSetupCMSLCGEnvironment_\n'
1158	>	txt += ' echo ">>> setup CMS LCG environment:"\n'
1159	>	txt += ' echo "set SCRAM ARCH and BUILD_ARCH to ' + self.executable_arch + ' ###"\n'
1160	>	txt += ' export SCRAM_ARCH='+self.executable_arch+'\n'
1161	>	txt += ' export BUILD_ARCH='+self.executable_arch+'\n'
1162	>	txt += ' if [ ! $VO_CMS_SW_DIR ] ;then\n'
1163	>	txt += ' echo "ERROR ==> CMS software dir not found on WN `hostname`"\n'
1164	>	txt += ' job_exit_code=10031\n'
1165	>	txt += ' func_exit\n'
1166	>	txt += ' else\n'
1167	>	txt += ' echo "Sourcing environment... "\n'
1168	>	txt += ' if [ ! -s $VO_CMS_SW_DIR/cmsset_default.sh ] ;then\n'
1169	>	txt += ' echo "ERROR ==> cmsset_default.sh file not found into dir $VO_CMS_SW_DIR"\n'
1170	>	txt += ' job_exit_code=10020\n'
1171	>	txt += ' func_exit\n'
1172	>	txt += ' fi\n'
1173	>	txt += ' echo "sourcing $VO_CMS_SW_DIR/cmsset_default.sh"\n'
1174	>	txt += ' source $VO_CMS_SW_DIR/cmsset_default.sh\n'
1175	>	txt += ' result=$?\n'
1176	>	txt += ' if [ $result -ne 0 ]; then\n'
1177	>	txt += ' echo "ERROR ==> problem sourcing $VO_CMS_SW_DIR/cmsset_default.sh"\n'
1178	>	txt += ' job_exit_code=10032\n'
1179	>	txt += ' func_exit\n'
1180	>	txt += ' fi\n'
1181	>	txt += ' fi\n'
1182	>	txt += ' \n'
1183	>	txt += ' echo "==> setup cms environment ok"\n'
1184		return txt
1185
1393	–	### FEDE FOR DBS OUTPUT PUBLICATION
1186		def modifyReport(self, nj):
1187		"""
1188	<	insert the part of the script that modifies the FrameworkJob Report
1188	>	insert the part of the script that modifies the FrameworkJob Report
1189		"""
1190	+	txt = '\n#Written by cms_cmssw::modifyReport\n'
1191	+	publish_data = int(self.cfg_params.get('USER.publish_data',0))
1192	+	if (publish_data == 1):
1193	+	processedDataset = self.cfg_params['USER.publish_data_name']
1194	+	LFNBaseName = LFNBase(processedDataset)
1195
1196	<	txt = ''
1197	<	try:
1198	<	publish_data = int(self.cfg_params['USER.publish_data'])
1199	<	except KeyError:
1200	<	publish_data = 0
1201	<	if (publish_data == 1):
1202	<	txt += 'echo "Modify Job Report" \n'
1406	<	#txt += 'chmod a+x $RUNTIME_AREA/'+self.version+'/ProdAgentApi/FwkJobRep/ModifyJobReport.py\n'
1407	<	################ FEDE FOR DBS2 #############################################
1408	<	txt += 'chmod a+x $SOFTWARE_DIR/ProdAgentApi/FwkJobRep/ModifyJobReport.py\n'
1409	<	#############################################################################
1410	<	#try:
1411	<	# publish_data = int(self.cfg_params['USER.publish_data'])
1412	<	#except KeyError:
1413	<	# publish_data = 0
1414	<
1415	<	txt += 'if [ -z "$SE" ]; then\n'
1416	<	txt += ' SE="" \n'
1417	<	txt += 'fi \n'
1418	<	txt += 'if [ -z "$SE_PATH" ]; then\n'
1419	<	txt += ' SE_PATH="" \n'
1420	<	txt += 'fi \n'
1421	<	txt += 'echo "SE = $SE"\n'
1422	<	txt += 'echo "SE_PATH = $SE_PATH"\n'
1196	>	txt += 'if [ $copy_exit_status -eq 0 ]; then\n'
1197	>	txt += ' FOR_LFN=%s_${PSETHASH}/\n'%(LFNBaseName)
1198	>	txt += 'else\n'
1199	>	txt += ' FOR_LFN=/copy_problems/ \n'
1200	>	txt += ' SE=""\n'
1201	>	txt += ' SE_PATH=""\n'
1202	>	txt += 'fi\n'
1203
1204	<	#if (publish_data == 1):
1205	<	#processedDataset = self.cfg_params['USER.processed_datasetname']
1426	<	processedDataset = self.cfg_params['USER.publish_data_name']
1204	>	txt += 'echo ">>> Modify Job Report:" \n'
1205	>	txt += 'chmod a+x $SOFTWARE_DIR/ProdCommon/ProdCommon/FwkJobRep/ModifyJobReport.py\n'
1206		txt += 'ProcessedDataset='+processedDataset+'\n'
1428	–	#### LFN=/store/user/<user>/processedDataset_PSETHASH
1429	–	txt += 'if [ "$SE_PATH" == "" ]; then\n'
1430	–	#### FEDE: added slash in LFN ##############
1431	–	txt += ' FOR_LFN=/copy_problems/ \n'
1432	–	txt += 'else \n'
1433	–	txt += ' tmp=`echo $SE_PATH \| awk -F \'store\' \'{print$2}\'` \n'
1434	–	##### FEDE TO BE CHANGED, BECAUSE STORE IS HARDCODED!!!! ########
1435	–	txt += ' FOR_LFN=/store$tmp \n'
1436	–	txt += 'fi \n'
1207		txt += 'echo "ProcessedDataset = $ProcessedDataset"\n'
1208	+	txt += 'echo "SE = $SE"\n'
1209	+	txt += 'echo "SE_PATH = $SE_PATH"\n'
1210		txt += 'echo "FOR_LFN = $FOR_LFN" \n'
1211		txt += 'echo "CMSSW_VERSION = $CMSSW_VERSION"\n\n'
1212	<	#txt += 'echo "$RUNTIME_AREA/'+self.version+'/ProdAgentApi/FwkJobRep/ModifyJobReport.py crab_fjr_$NJob.xml $NJob $FOR_LFN $PrimaryDataset $DataTier $ProcessedDataset $ApplicationFamily $executable $CMSSW_VERSION $PSETHASH $SE $SE_PATH"\n'
1213	<	txt += 'echo "$SOFTWARE_DIR/ProdAgentApi/FwkJobRep/ModifyJobReport.py crab_fjr_$NJob.xml $NJob $FOR_LFN $PrimaryDataset $DataTier $ProcessedDataset $ApplicationFamily $executable $CMSSW_VERSION $PSETHASH $SE $SE_PATH"\n'
1442	<	txt += '$SOFTWARE_DIR/ProdAgentApi/FwkJobRep/ModifyJobReport.py crab_fjr_$NJob.xml $NJob $FOR_LFN $PrimaryDataset $DataTier $ProcessedDataset $ApplicationFamily $executable $CMSSW_VERSION $PSETHASH $SE $SE_PATH\n'
1443	<	#txt += '$RUNTIME_AREA/'+self.version+'/ProdAgentApi/FwkJobRep/ModifyJobReport.py crab_fjr_$NJob.xml $NJob $FOR_LFN $PrimaryDataset $DataTier $ProcessedDataset $ApplicationFamily $executable $CMSSW_VERSION $PSETHASH $SE $SE_PATH\n'
1444	<
1212	>	txt += 'echo "$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 $SE_PATH"\n'
1213	>	txt += '$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 $SE_PATH\n'
1214		txt += 'modifyReport_result=$?\n'
1446	–	txt += 'echo modifyReport_result = $modifyReport_result\n'
1215		txt += 'if [ $modifyReport_result -ne 0 ]; then\n'
1216	<	txt += ' exit_status=1\n'
1217	<	txt += ' echo "ERROR: Problem with ModifyJobReport"\n'
1216	>	txt += ' modifyReport_result=70500\n'
1217	>	txt += ' job_exit_code=$modifyReport_result\n'
1218	>	txt += ' echo "ModifyReportResult=$modifyReport_result" \| tee -a $RUNTIME_AREA/$repo\n'
1219	>	txt += ' echo "WARNING: Problem with ModifyJobReport"\n'
1220		txt += 'else\n'
1221	<	txt += ' mv NewFrameworkJobReport.xml crab_fjr_$NJob.xml\n'
1221	>	txt += ' mv NewFrameworkJobReport.xml $RUNTIME_AREA/crab_fjr_$NJob.xml\n'
1222		txt += 'fi\n'
1453	–	else:
1454	–	txt += 'echo "no data publication required"\n'
1455	–	#txt += 'ProcessedDataset=no_data_to_publish \n'
1456	–	#### FEDE: added slash in LFN ##############
1457	–	#txt += 'FOR_LFN=/local/ \n'
1458	–	#txt += 'echo "ProcessedDataset = $ProcessedDataset"\n'
1459	–	#txt += 'echo "FOR_LFN = $FOR_LFN" \n'
1223		return txt
1224
1225	<	def cleanEnv(self):
1226	<	### OLI_DANIELE
1227	<	txt = ''
1228	<	txt += 'if [ $middleware == OSG ]; then\n'
1229	<	txt += ' cd $RUNTIME_AREA\n'
1230	<	txt += ' echo "Remove working directory: $WORKING_DIR"\n'
1231	<	txt += ' /bin/rm -rf $WORKING_DIR\n'
1232	<	txt += ' if [ -d $WORKING_DIR ] ;then\n'
1233	<	txt += ' echo "SET_EXE 60999 ==> OSG $WORKING_DIR could not be deleted on WN `hostname` after cleanup of WN"\n'
1234	<	txt += ' echo "JOB_EXIT_STATUS = 60999"\n'
1235	<	txt += ' echo "JobExitCode=60999" \| tee -a $RUNTIME_AREA/$repo\n'
1236	<	txt += ' dumpStatus $RUNTIME_AREA/$repo\n'
1237	<	txt += ' rm -f $RUNTIME_AREA/$repo \n'
1238	<	txt += ' echo "MonitorJobID=`echo $MonitorJobID`" \| tee -a $RUNTIME_AREA/$repo \n'
1239	<	txt += ' echo "MonitorID=`echo $MonitorID`" \| tee -a $RUNTIME_AREA/$repo\n'
1225	>	def wsParseFJR(self):
1226	>	"""
1227	>	Parse the FrameworkJobReport to obtain useful infos
1228	>	"""
1229	>	txt = '\n#Written by cms_cmssw::wsParseFJR\n'
1230	>	txt += 'echo ">>> Parse FrameworkJobReport crab_fjr.xml"\n'
1231	>	txt += 'if [ -s $RUNTIME_AREA/crab_fjr_$NJob.xml ]; then\n'
1232	>	txt += ' if [ -s $RUNTIME_AREA/parseCrabFjr.py ]; then\n'
1233	>	txt += ' cmd_out=`python $RUNTIME_AREA/parseCrabFjr.py --input $RUNTIME_AREA/crab_fjr_$NJob.xml --dashboard $MonitorID,$MonitorJobID '+self.debugWrap+'`\n'
1234	>	if self.debug_wrapper :
1235	>	txt += ' echo "Result of parsing the FrameworkJobReport crab_fjr.xml: $cmd_out"\n'
1236	>	txt += ' executable_exit_status=`python $RUNTIME_AREA/parseCrabFjr.py --input $RUNTIME_AREA/crab_fjr_$NJob.xml --exitcode`\n'
1237	>	txt += ' if [ $executable_exit_status -eq 50115 ];then\n'
1238	>	txt += ' echo ">>> crab_fjr.xml contents: "\n'
1239	>	txt += ' cat $RUNTIME_AREA/crab_fjr_NJob.xml\n'
1240	>	txt += ' echo "Wrong FrameworkJobReport --> does not contain useful info. ExitStatus: $executable_exit_status"\n'
1241	>	txt += ' elif [ $executable_exit_status -eq -999 ];then\n'
1242	>	txt += ' echo "ExitStatus from FrameworkJobReport not available. not available. Using exit code of executable from command line."\n'
1243	>	txt += ' else\n'
1244	>	txt += ' echo "Extracted ExitStatus from FrameworkJobReport parsing output: $executable_exit_status"\n'
1245	>	txt += ' fi\n'
1246	>	txt += ' else\n'
1247	>	txt += ' echo "CRAB python script to parse CRAB FrameworkJobReport crab_fjr.xml is not available, using exit code of executable from command line."\n'
1248		txt += ' fi\n'
1249	+	#### Patch to check input data reading for CMSSW16x Hopefully we-ll remove it asap
1250	+
1251	+	if self.datasetPath:
1252	+	# VERIFY PROCESSED DATA
1253	+	txt += ' if [ $executable_exit_status -eq 0 ];then\n'
1254	+	txt += ' echo ">>> Verify list of processed files:"\n'
1255	+	txt += ' echo $InputFiles \|tr -d \'\\\\\' \|tr \',\' \'\\n\'\|tr -d \'"\' > input-files.txt\n'
1256	+	txt += ' python $RUNTIME_AREA/parseCrabFjr.py --input $RUNTIME_AREA/crab_fjr_$NJob.xml --lfn > processed-files.txt\n'
1257	+	txt += ' cat input-files.txt \| sort \| uniq > tmp.txt\n'
1258	+	txt += ' mv tmp.txt input-files.txt\n'
1259	+	txt += ' echo "cat input-files.txt"\n'
1260	+	txt += ' echo "----------------------"\n'
1261	+	txt += ' cat input-files.txt\n'
1262	+	txt += ' cat processed-files.txt \| sort \| uniq > tmp.txt\n'
1263	+	txt += ' mv tmp.txt processed-files.txt\n'
1264	+	txt += ' echo "----------------------"\n'
1265	+	txt += ' echo "cat processed-files.txt"\n'
1266	+	txt += ' echo "----------------------"\n'
1267	+	txt += ' cat processed-files.txt\n'
1268	+	txt += ' echo "----------------------"\n'
1269	+	txt += ' diff -q input-files.txt processed-files.txt\n'
1270	+	txt += ' fileverify_status=$?\n'
1271	+	txt += ' if [ $fileverify_status -ne 0 ]; then\n'
1272	+	txt += ' executable_exit_status=30001\n'
1273	+	txt += ' echo "ERROR ==> not all input files processed"\n'
1274	+	txt += ' echo " ==> list of processed files from crab_fjr.xml differs from list in pset.cfg"\n'
1275	+	txt += ' echo " ==> diff input-files.txt processed-files.txt"\n'
1276	+	txt += ' fi\n'
1277	+	txt += ' fi\n'
1278	+	txt += '\n'
1279	+	txt += 'else\n'
1280	+	txt += ' echo "CRAB FrameworkJobReport crab_fjr.xml is not available, using exit code of executable from command line."\n'
1281		txt += 'fi\n'
1282		txt += '\n'
1283	+	txt += 'echo "ExeExitCode=$executable_exit_status" \| tee -a $RUNTIME_AREA/$repo\n'
1284	+	txt += 'echo "EXECUTABLE_EXIT_STATUS = $executable_exit_status"\n'
1285	+	txt += 'job_exit_code=$executable_exit_status\n'
1286	+
1287		return txt
1288
1289		def setParam_(self, param, value):
#	Line 1485 \| Line 1292 \| class Cmssw(JobType):
1292		def getParams(self):
1293		return self._params
1294
1488	–	def setTaskid_(self):
1489	–	self._taskId = self.cfg_params['taskId']
1490	–
1491	–	def getTaskid(self):
1492	–	return self._taskId
1493	–
1295		def uniquelist(self, old):
1296		"""
1297		remove duplicates from a list
#	Line 1500 \| Line 1301 \| class Cmssw(JobType):
1301		nd[e]=0
1302		return nd.keys()
1303
1304	<
1504	<	def checkOut(self, limit):
1304	>	def outList(self):
1305		"""
1306		check the dimension of the output files
1307		"""
1308	<	txt = 'echo "*****************************************"\n'
1309	<	txt += 'echo " Starting output sandbox limit check "\n'
1510	<	txt += 'echo "*****************************************"\n'
1511	<	allOutFiles = ""
1308	>	txt = ''
1309	>	txt += 'echo ">>> list of expected files on output sandbox"\n'
1310		listOutFiles = []
1311	<	for fileOut in (self.output_file+self.output_file_sandbox):
1312	<	if fileOut.find('crab_fjr') == -1:
1313	<	allOutFiles = allOutFiles + " " + self.numberFile_(fileOut, '$NJob')
1314	<	listOutFiles.append(self.numberFile_(fileOut, '$NJob'))
1315	<	txt += 'echo "OUTPUT files: '+str(allOutFiles)+'";\n'
1316	<	txt += 'ls -gGhrta;\n'
1317	<	txt += 'sum=0;\n'
1318	<	txt += 'for file in '+str(allOutFiles)+' ; do\n'
1319	<	txt += ' if [ -e $file ]; then\n'
1320	<	txt += ' tt=`ls -gGrta $file \| awk \'{ print $3 }\'`\n'
1321	<	txt += ' sum=`expr $sum + $tt`\n'
1322	<	txt += ' else\n'
1323	<	txt += ' echo "WARNING: output file $file not found!"\n'
1324	<	txt += ' fi\n'
1325	<	txt += 'done\n'
1528	<	txt += 'echo "Total Output dimension: $sum";\n'
1529	<	txt += 'limit='+str(limit)+';\n'
1530	<	txt += 'echo "OUTPUT FILES LIMIT SET TO: $limit";\n'
1531	<	txt += 'if [ $limit -lt $sum ]; then\n'
1532	<	txt += ' echo "WARNING: output files have to big size - something will be lost;"\n'
1533	<	txt += ' echo " checking the output file sizes..."\n'
1534	<	"""
1535	<	txt += ' dim=0;\n'
1536	<	txt += ' exclude=0;\n'
1537	<	txt += ' for files in '+str(allOutFiles)+' ; do\n'
1538	<	txt += ' sumTemp=0;\n'
1539	<	txt += ' for file2 in '+str(allOutFiles)+' ; do\n'
1540	<	txt += ' if [ $file != $file2 ]; then\n'
1541	<	txt += ' tt=`ls -gGrta $file2 \| awk \'{ print $3 }\';`\n'
1542	<	txt += ' sumTemp=`expr $sumTemp + $tt`;\n'
1543	<	txt += ' fi\n'
1544	<	txt += ' done\n'
1545	<	txt += ' if [ $sumTemp -lt $limit ]; then\n'
1546	<	txt += ' if [ $dim -lt $sumTemp ]; then\n'
1547	<	txt += ' dim=$sumTemp;\n'
1548	<	txt += ' exclude=$file;\n'
1549	<	txt += ' fi\n'
1550	<	txt += ' fi\n'
1551	<	txt += ' done\n'
1552	<	txt += ' echo "Dimension calculated: $dim"; echo "File to exclude: $exclude";\n'
1553	<	"""
1554	<	txt += ' tot=0;\n'
1555	<	txt += ' for file2 in '+str(allOutFiles)+' ; do\n'
1556	<	txt += ' tt=`ls -gGrta $file2 \| awk \'{ print $3 }\';`\n'
1557	<	txt += ' tot=`expr $tot + $tt`;\n'
1558	<	txt += ' if [ $limit -lt $tot ]; then\n'
1559	<	txt += ' tot=`expr $tot - $tt`;\n'
1560	<	txt += ' fileLast=$file;\n'
1561	<	txt += ' break;\n'
1562	<	txt += ' fi\n'
1563	<	txt += ' done\n'
1564	<	txt += ' echo "Dimension calculated: $tot"; echo "First file to exclude: $file";\n'
1565	<	txt += ' flag=0;\n'
1566	<	txt += ' for filess in '+str(allOutFiles)+' ; do\n'
1567	<	txt += ' if [ $fileLast = $filess ]; then\n'
1568	<	txt += ' flag=1;\n'
1569	<	txt += ' fi\n'
1570	<	txt += ' if [ $flag -eq 1 ]; then\n'
1571	<	txt += ' rm -f $filess;\n'
1572	<	txt += ' fi\n'
1573	<	txt += ' done\n'
1574	<	txt += ' ls -agGhrt;\n'
1575	<	txt += ' echo "WARNING: output files are too big in dimension: can not put in the output_sandbox.";\n'
1576	<	txt += ' echo "JOB_EXIT_STATUS = 70000";\n'
1577	<	txt += ' exit_status=70000;\n'
1578	<	txt += 'else'
1579	<	txt += ' echo "Total Output dimension $sum is fine.";\n'
1580	<	txt += 'fi\n'
1581	<	txt += 'echo "*****************************************"\n'
1582	<	txt += 'echo "* Ending output sandbox limit check *"\n'
1583	<	txt += 'echo "*****************************************"\n'
1311	>	stdout = 'CMSSW_$NJob.stdout'
1312	>	stderr = 'CMSSW_$NJob.stderr'
1313	>	if (self.return_data == 1):
1314	>	for file in (self.output_file+self.output_file_sandbox):
1315	>	listOutFiles.append(self.numberFile_(file, '$NJob'))
1316	>	listOutFiles.append(stdout)
1317	>	listOutFiles.append(stderr)
1318	>	else:
1319	>	for file in (self.output_file_sandbox):
1320	>	listOutFiles.append(self.numberFile_(file, '$NJob'))
1321	>	listOutFiles.append(stdout)
1322	>	listOutFiles.append(stderr)
1323	>	txt += 'echo "output files: '+string.join(listOutFiles,' ')+'"\n'
1324	>	txt += 'filesToCheck="'+string.join(listOutFiles,' ')+'"\n'
1325	>	txt += 'export filesToCheck\n'
1326		return txt

Diff Legend

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

Comparing COMP/CRAB/python/cms_cmssw.py (file contents): Revision 1.130 by fanzago, Wed Oct 17 13:24:41 2007 UTC vs. Revision 1.205 by slacapra, Thu Jun 5 16:34:04 2008 UTC

Diff Legend

Comparing COMP/CRAB/python/cms_cmssw.py (file contents):
Revision 1.130 by fanzago, Wed Oct 17 13:24:41 2007 UTC vs.
Revision 1.205 by slacapra, Thu Jun 5 16:34:04 2008 UTC