| 91 |
|
try: |
| 92 |
|
datasetpath_split = self.datasetPath.split("/") |
| 93 |
|
# standard style |
| 94 |
+ |
self.setParam_('datasetFull', self.datasetPath) |
| 95 |
|
if self.use_dbs_1 == 1 : |
| 96 |
|
self.setParam_('dataset', datasetpath_split[1]) |
| 97 |
|
self.setParam_('owner', datasetpath_split[-1]) |
| 319 |
|
if (self.sourceSeedVtx) : |
| 320 |
|
PsetEdit.vtxSeed("INPUTVTX") |
| 321 |
|
if (self.sourceSeedG4) : |
| 322 |
< |
self.PsetEdit.g4Seed("INPUTG4") |
| 322 |
> |
PsetEdit.g4Seed("INPUTG4") |
| 323 |
|
if (self.sourceSeedMix) : |
| 324 |
< |
self.PsetEdit.mixSeed("INPUTMIX") |
| 324 |
> |
PsetEdit.mixSeed("INPUTMIX") |
| 325 |
|
# add FrameworkJobReport to parameter-set |
| 326 |
|
PsetEdit.addCrabFJR(self.fjrFileName) |
| 327 |
|
PsetEdit.psetWriter(self.configFilename()) |
| 583 |
|
# screen output |
| 584 |
|
screenOutput = "List of jobs and available destination sites:\n\n" |
| 585 |
|
|
| 586 |
+ |
# keep trace of block with no sites to print a warning at the end |
| 587 |
+ |
noSiteBlock = [] |
| 588 |
+ |
bloskNoSite = [] |
| 589 |
+ |
|
| 590 |
|
blockCounter = 0 |
| 591 |
|
for block in blocks: |
| 592 |
|
if block in jobsOfBlock.keys() : |
| 593 |
|
blockCounter += 1 |
| 594 |
|
screenOutput += "Block %5i: jobs %20s: sites: %s\n" % (blockCounter,spanRanges(jobsOfBlock[block]),','.join(self.blackWhiteListParser.checkWhiteList(self.blackWhiteListParser.checkBlackList(blockSites[block],block),block))) |
| 595 |
< |
|
| 595 |
> |
if len(self.blackWhiteListParser.checkWhiteList(self.blackWhiteListParser.checkBlackList(blockSites[block],block),block)) == 0: |
| 596 |
> |
noSiteBlock.append( spanRanges(jobsOfBlock[block]) ) |
| 597 |
> |
bloskNoSite.append( blockCounter ) |
| 598 |
> |
|
| 599 |
|
common.logger.message(screenOutput) |
| 600 |
+ |
if len(noSiteBlock) > 0 and len(bloskNoSite) > 0: |
| 601 |
+ |
msg = 'WARNING: No sites are hosting any part of data for block:\n ' |
| 602 |
+ |
virgola = "" |
| 603 |
+ |
if len(bloskNoSite) > 1: |
| 604 |
+ |
virgola = "," |
| 605 |
+ |
for block in bloskNoSite: |
| 606 |
+ |
msg += ' ' + str(block) + virgola |
| 607 |
+ |
msg += '\n Related jobs:\n ' |
| 608 |
+ |
virgola = "" |
| 609 |
+ |
if len(noSiteBlock) > 1: |
| 610 |
+ |
virgola = "," |
| 611 |
+ |
for range_jobs in noSiteBlock: |
| 612 |
+ |
msg += str(range_jobs) + virgola |
| 613 |
+ |
msg += '\n will not be submitted and this block of data can not be analyzed!\n' |
| 614 |
+ |
common.logger.message(msg) |
| 615 |
|
|
| 616 |
|
self.list_of_args = list_of_lists |
| 617 |
|
return |
| 621 |
|
Perform job splitting based on number of event per job |
| 622 |
|
""" |
| 623 |
|
common.logger.debug(5,'Splitting per events') |
| 624 |
< |
common.logger.message('Required '+str(self.eventsPerJob)+' events per job ') |
| 625 |
< |
common.logger.message('Required '+str(self.theNumberOfJobs)+' jobs in total ') |
| 626 |
< |
common.logger.message('Required '+str(self.total_number_of_events)+' events in total ') |
| 624 |
> |
|
| 625 |
> |
if (self.selectEventsPerJob): |
| 626 |
> |
common.logger.message('Required '+str(self.eventsPerJob)+' events per job ') |
| 627 |
> |
if (self.selectNumberOfJobs): |
| 628 |
> |
common.logger.message('Required '+str(self.theNumberOfJobs)+' jobs in total ') |
| 629 |
> |
if (self.selectTotalNumberEvents): |
| 630 |
> |
common.logger.message('Required '+str(self.total_number_of_events)+' events in total ') |
| 631 |
|
|
| 632 |
|
if (self.total_number_of_events < 0): |
| 633 |
|
msg='Cannot split jobs per Events with "-1" as total number of events' |
| 794 |
|
# distinguish case when script is in user project area or given by full path somewhere else |
| 795 |
|
if exeWithPath.find(path) >= 0 : |
| 796 |
|
exe = string.replace(exeWithPath, path,'') |
| 797 |
< |
tar.add(path+exe,os.path.basename(executable)) |
| 797 |
> |
tar.add(path+exe,exe) |
| 798 |
|
else : |
| 799 |
|
tar.add(exeWithPath,os.path.basename(executable)) |
| 800 |
|
pass |
| 1208 |
|
txt += '# directory content\n' |
| 1209 |
|
txt += 'ls \n' |
| 1210 |
|
|
| 1211 |
< |
for fileWithSuffix in (self.output_file+self.output_file_sandbox): |
| 1211 |
> |
txt += 'output_exit_status=0\n' |
| 1212 |
> |
|
| 1213 |
> |
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" > $RUNTIME_AREA/'+output_file_num+'\n' |
| 1227 |
> |
txt += ' fi \n' |
| 1228 |
> |
txt += 'fi\n' |
| 1229 |
> |
|
| 1230 |
> |
for fileWithSuffix in (self.output_file): |
| 1231 |
|
output_file_num = self.numberFile_(fileWithSuffix, '$NJob') |
| 1232 |
|
txt += '\n' |
| 1233 |
|
txt += '# check output file\n' |
| 1188 |
– |
# txt += 'ls '+fileWithSuffix+'\n' |
| 1189 |
– |
# txt += 'ls_result=$?\n' |
| 1234 |
|
txt += 'if [ -e ./'+fileWithSuffix+' ] ; then\n' |
| 1235 |
< |
###### FEDE 14444 - 08 - 2007 ######## |
| 1236 |
< |
txt += ' mv '+fileWithSuffix+' $RUNTIME_AREA\n' |
| 1193 |
< |
txt += ' cp $RUNTIME_AREA/'+fileWithSuffix+' $RUNTIME_AREA/'+output_file_num+'\n' |
| 1194 |
< |
################################### |
| 1235 |
> |
txt += ' mv '+fileWithSuffix+' $RUNTIME_AREA\n' |
| 1236 |
> |
txt += ' cp $RUNTIME_AREA/'+fileWithSuffix+' $RUNTIME_AREA/'+output_file_num+'\n' |
| 1237 |
|
txt += 'else\n' |
| 1238 |
< |
txt += ' exit_status=60302\n' |
| 1239 |
< |
txt += ' echo "ERROR: Problem with output file '+fileWithSuffix+'"\n' |
| 1238 |
> |
txt += ' exit_status=60302\n' |
| 1239 |
> |
txt += ' echo "ERROR: Problem with output file '+fileWithSuffix+'"\n' |
| 1240 |
> |
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': |
| 1243 |
|
txt += ' if [ $middleware == OSG ]; then \n' |
| 1244 |
|
txt += ' echo "prepare dummy output file"\n' |
| 1248 |
|
file_list = [] |
| 1249 |
|
for fileWithSuffix in (self.output_file): |
| 1250 |
|
file_list.append(self.numberFile_(fileWithSuffix, '$NJob')) |
| 1251 |
+ |
|
| 1252 |
|
txt += 'file_list="'+string.join(file_list,' ')+'"\n' |
| 1208 |
– |
|
| 1253 |
|
txt += 'cd $RUNTIME_AREA\n' |
| 1210 |
– |
#### FEDE this is the cleanEnv function |
| 1211 |
– |
### OLI_DANIELE |
| 1212 |
– |
#txt += 'if [ $middleware == OSG ]; then\n' |
| 1213 |
– |
#txt += ' cd $RUNTIME_AREA\n' |
| 1214 |
– |
#txt += ' echo "Remove working directory: $WORKING_DIR"\n' |
| 1215 |
– |
#txt += ' /bin/rm -rf $WORKING_DIR\n' |
| 1216 |
– |
#txt += ' if [ -d $WORKING_DIR ] ;then\n' |
| 1217 |
– |
#txt += ' echo "SET_EXE 60999 ==> OSG $WORKING_DIR could not be deleted on WN `hostname` after cleanup of WN"\n' |
| 1218 |
– |
#txt += ' echo "JOB_EXIT_STATUS = 60999"\n' |
| 1219 |
– |
#txt += ' echo "JobExitCode=60999" | tee -a $RUNTIME_AREA/$repo\n' |
| 1220 |
– |
#txt += ' dumpStatus $RUNTIME_AREA/$repo\n' |
| 1221 |
– |
#txt += ' rm -f $RUNTIME_AREA/$repo \n' |
| 1222 |
– |
#txt += ' echo "MonitorJobID=`echo $MonitorJobID`" | tee -a $RUNTIME_AREA/$repo \n' |
| 1223 |
– |
#txt += ' echo "MonitorID=`echo $MonitorID`" | tee -a $RUNTIME_AREA/$repo\n' |
| 1224 |
– |
#txt += ' fi\n' |
| 1225 |
– |
#txt += 'fi\n' |
| 1226 |
– |
#txt += '\n' |
| 1227 |
– |
|
| 1228 |
– |
|
| 1254 |
|
return txt |
| 1255 |
|
|
| 1256 |
|
def numberFile_(self, file, txt): |
| 1397 |
|
""" |
| 1398 |
|
|
| 1399 |
|
txt = '' |
| 1375 |
– |
txt += 'echo "Modify Job Report" \n' |
| 1376 |
– |
#txt += 'chmod a+x $RUNTIME_AREA/'+self.version+'/ProdAgentApi/FwkJobRep/ModifyJobReport.py\n' |
| 1377 |
– |
################ FEDE FOR DBS2 ############################################# |
| 1378 |
– |
txt += 'chmod a+x $SOFTWARE_DIR/ProdAgentApi/FwkJobRep/ModifyJobReport.py\n' |
| 1379 |
– |
############################################################################# |
| 1400 |
|
try: |
| 1401 |
|
publish_data = int(self.cfg_params['USER.publish_data']) |
| 1402 |
|
except KeyError: |
| 1403 |
|
publish_data = 0 |
| 1404 |
+ |
if (publish_data == 1): |
| 1405 |
+ |
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 = $SE_PATH"\n' |
| 1415 |
> |
txt += 'if [ -z "$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 = $SE_PATH"\n' |
| 1423 |
|
|
| 1424 |
< |
if (publish_data == 1): |
| 1424 |
> |
#if (publish_data == 1): |
| 1425 |
|
#processedDataset = self.cfg_params['USER.processed_datasetname'] |
| 1426 |
|
processedDataset = self.cfg_params['USER.publish_data_name'] |
| 1427 |
|
txt += 'ProcessedDataset='+processedDataset+'\n' |
| 1451 |
|
txt += ' mv NewFrameworkJobReport.xml crab_fjr_$NJob.xml\n' |
| 1452 |
|
txt += 'fi\n' |
| 1453 |
|
else: |
| 1454 |
< |
txt += 'ProcessedDataset=no_data_to_publish \n' |
| 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' |
| 1457 |
> |
#txt += 'FOR_LFN=/local/ \n' |
| 1458 |
> |
#txt += 'echo "ProcessedDataset = $ProcessedDataset"\n' |
| 1459 |
> |
#txt += 'echo "FOR_LFN = $FOR_LFN" \n' |
| 1460 |
|
return txt |
| 1461 |
|
|
| 1462 |
|
def cleanEnv(self): |
| 1499 |
|
for e in old: |
| 1500 |
|
nd[e]=0 |
| 1501 |
|
return nd.keys() |
| 1502 |
+ |
|
| 1503 |
+ |
|
| 1504 |
+ |
def checkOut(self, limit): |
| 1505 |
+ |
""" |
| 1506 |
+ |
check the dimension of the output files |
| 1507 |
+ |
""" |
| 1508 |
+ |
txt = 'echo "*****************************************"\n' |
| 1509 |
+ |
txt += 'echo "** Starting output sandbox limit check **"\n' |
| 1510 |
+ |
txt += 'echo "*****************************************"\n' |
| 1511 |
+ |
allOutFiles = "" |
| 1512 |
+ |
listOutFiles = [] |
| 1513 |
+ |
for fileOut in (self.output_file+self.output_file_sandbox): |
| 1514 |
+ |
if fileOut.find('crab_fjr') == -1: |
| 1515 |
+ |
allOutFiles = allOutFiles + " " + self.numberFile_(fileOut, '$NJob') |
| 1516 |
+ |
listOutFiles.append(self.numberFile_(fileOut, '$NJob')) |
| 1517 |
+ |
txt += 'echo "OUTPUT files: '+str(allOutFiles)+'";\n' |
| 1518 |
+ |
txt += 'ls -gGhrta;\n' |
| 1519 |
+ |
txt += 'sum=0;\n' |
| 1520 |
+ |
txt += 'for file in '+str(allOutFiles)+' ; do\n' |
| 1521 |
+ |
txt += ' if [ -e $file ]; then\n' |
| 1522 |
+ |
txt += ' tt=`ls -gGrta $file | awk \'{ print $3 }\'`\n' |
| 1523 |
+ |
txt += ' sum=`expr $sum + $tt`\n' |
| 1524 |
+ |
txt += ' else\n' |
| 1525 |
+ |
txt += ' echo "WARNING: output file $file not found!"\n' |
| 1526 |
+ |
txt += ' fi\n' |
| 1527 |
+ |
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' |
| 1584 |
+ |
return txt |
