def testLoad(self):
"""
_testLoad_
Test the loading of file meta data using the ID of a file and the
LFN of a file.
"""
testFileA = File(lfn="/this/is/a/lfn", size=1024, events=10,
checksums={'cksum': 101}, first_event=2, merged=True)
testFileA.create()
testFileB = File(lfn=testFileA["lfn"])
testFileB.load()
testFileC = File(id=testFileA["id"])
testFileC.load()
self.assertEqual(testFileA, testFileB, "File load by LFN didn't work")
self.assertEqual(testFileA, testFileC, "File load by ID didn't work")
self.assertTrue(isinstance(testFileB["id"], int), "File id is not an integer type.")
self.assertTrue(isinstance(testFileB["size"], int), "File size is not an integer type.")
self.assertTrue(isinstance(testFileB["events"], int), "File events is not an integer type.")
self.assertTrue(isinstance(testFileB["checksums"], dict), "File cksum is not a string type.")
self.assertTrue(isinstance(testFileB["first_event"], int), "File first_event is not an integer type.")
self.assertTrue(isinstance(testFileC["id"], int), "File id is not an integer type.")
self.assertTrue(isinstance(testFileC["size"], int), "File size is not an integer type.")
self.assertTrue(isinstance(testFileC["events"], int), "File events is not an integer type.")
self.assertTrue(isinstance(testFileC["checksums"], dict), "File cksum is not an string type.")
self.assertTrue(isinstance(testFileC["first_event"], int), "File first_event is not an integer type.")
self.assertEqual(testFileC['checksums'], {'cksum': '101'})
testFileA.delete()
return
def testSetLocationOrder(self):
"""
_testSetLocationOrder_
This tests that you can specify a location before creating the file,
instead of having to do it afterwards.
"""
myThread = threading.currentThread()
testFileA = File(lfn="/this/is/a/lfn", size=1024, events=10)
testFileA.setLocation("se1.cern.ch")
testFileA.create()
testFileB = File(lfn=testFileA["lfn"])
testFileB.load()
daoFactory = DAOFactory(package="WMCore.WMBS",
logger=logging,
dbinterface=myThread.dbi)
locationFac = daoFactory(classname="Files.GetLocation")
location = locationFac.execute(testFileB['lfn']).pop()
self.assertEqual(location, 'se1.cern.ch')
return
def testLoad(self):
"""
_testLoad_
Test the loading of file meta data using the ID of a file and the
LFN of a file.
"""
testFileA = File(lfn="/this/is/a/lfn",
size=1024,
events=10,
checksums={'cksum': 101},
first_event=2,
merged=True)
testFileA.create()
testFileB = File(lfn=testFileA["lfn"])
testFileB.load()
testFileC = File(id=testFileA["id"])
testFileC.load()
assert testFileA == testFileB, \
"ERROR: File load by LFN didn't work"
assert testFileA == testFileC, \
"ERROR: File load by ID didn't work"
assert type(testFileB["id"]) == int, \
"ERROR: File id is not an integer type."
assert type(testFileB["size"]) == int, \
"ERROR: File size is not an integer type."
assert type(testFileB["events"]) == int, \
"ERROR: File events is not an integer type."
assert type(testFileB["checksums"]) == dict, \
"ERROR: File cksum is not a string type."
assert type(testFileB["first_event"]) == int, \
"ERROR: File first_event is not an integer type."
assert type(testFileC["id"]) == int, \
"ERROR: File id is not an integer type."
assert type(testFileC["size"]) == int, \
"ERROR: File size is not an integer type."
assert type(testFileC["events"]) == int, \
"ERROR: File events is not an integer type."
assert type(testFileC["checksums"]) == dict, \
"ERROR: File cksum is not an string type."
assert type(testFileC["first_event"]) == int, \
"ERROR: File first_event is not an integer type."
self.assertEqual(testFileC['checksums'], {'cksum': '101'})
testFileA.delete()
return
def testLoad(self):
"""
_testLoad_
Test the loading of file meta data using the ID of a file and the
LFN of a file.
"""
testFileA = File(
lfn="/this/is/a/lfn",
size=1024,
events=10,
checksums={"cksum": 101},
first_event=2,
last_event=3,
merged=True,
)
testFileA.create()
testFileB = File(lfn=testFileA["lfn"])
testFileB.load()
testFileC = File(id=testFileA["id"])
testFileC.load()
assert testFileA == testFileB, "ERROR: File load by LFN didn't work"
assert testFileA == testFileC, "ERROR: File load by ID didn't work"
assert type(testFileB["id"]) == int, "ERROR: File id is not an integer type."
assert type(testFileB["size"]) == int, "ERROR: File size is not an integer type."
assert type(testFileB["events"]) == int, "ERROR: File events is not an integer type."
assert type(testFileB["checksums"]) == dict, "ERROR: File cksum is not a string type."
assert type(testFileB["first_event"]) == int, "ERROR: File first_event is not an integer type."
assert type(testFileB["last_event"]) == int, "ERROR: File last_event is not an integer type."
assert type(testFileC["id"]) == int, "ERROR: File id is not an integer type."
assert type(testFileC["size"]) == int, "ERROR: File size is not an integer type."
assert type(testFileC["events"]) == int, "ERROR: File events is not an integer type."
assert type(testFileC["checksums"]) == dict, "ERROR: File cksum is not an string type."
assert type(testFileC["first_event"]) == int, "ERROR: File first_event is not an integer type."
assert type(testFileC["last_event"]) == int, "ERROR: File last_event is not an integer type."
self.assertEqual(testFileC["checksums"], {"cksum": "101"})
testFileA.delete()
return
def filesOfStatusByRun(self, status, runID):
"""
_filesOfStatusByRun_
Return all the files in the given subscription and the given run which
have the given status.
"""
existingTransaction = self.beginTransaction()
files = []
action = self.daofactory(classname = "Subscriptions.Get%sFilesByRun" % status)
for f in action.execute(self["id"], runID, conn = self.getDBConn(),
transaction = self.existingTransaction()):
fl = File(id = f["file"])
fl.load()
files.append(fl)
self.commitTransaction(existingTransaction)
return files
def filesOfStatusByRun(self, status, runID):
"""
_filesOfStatusByRun_
Return all the files in the given subscription and the given run which
have the given status.
"""
existingTransaction = self.beginTransaction()
files = []
action = self.daofactory(classname="Subscriptions.Get%sFilesByRun" % status)
for f in action.execute(self["id"], runID, conn=self.getDBConn(),
transaction=self.existingTransaction()):
fl = File(id=f["file"])
fl.load()
files.append(fl)
self.commitTransaction(existingTransaction)
return files
def testSetLocationOrder(self):
"""
_testSetLocationOrder_
This tests that you can specify a location before creating the file,
instead of having to do it afterwards.
"""
myThread = threading.currentThread()
testFileA = File(lfn = "/this/is/a/lfn", size = 1024, events = 10)
testFileA.setLocation("se1.cern.ch")
testFileA.create()
testFileB = File(lfn = testFileA["lfn"])
testFileB.load()
daoFactory = DAOFactory(package = "WMCore.WMBS", logger = logging, dbinterface = myThread.dbi)
locationFac = daoFactory(classname = "Files.GetLocation")
location = locationFac.execute(testFileB['lfn']).pop()
self.assertEqual(location, 'se1.cern.ch')
return
def testReportHandling(self):
"""
_testReportHandling_
Verify that we're able to parse a CMSSW report, convert it to a Report()
style report, pickle it and then have the accountant process it.
"""
self.procPath = os.path.join(WMCore.WMBase.getTestBase(),
"WMCore_t/FwkJobReport_t/CMSSWProcessingReport.xml")
myReport = Report("cmsRun1")
myReport.parse(self.procPath)
# Fake some metadata that should be added by the stageout scripts.
for fileRef in myReport.getAllFileRefsFromStep("cmsRun1"):
fileRef.size = 1024
fileRef.location = "cmssrm.fnal.gov"
fwjrPath = os.path.join(self.tempDir, "ProcReport.pkl")
cmsRunStep = myReport.retrieveStep("cmsRun1")
cmsRunStep.status = 0
myReport.setTaskName('/TestWF/None')
myReport.persist(fwjrPath)
self.setFWJRAction.execute(jobID = self.testJob["id"], fwjrPath = fwjrPath)
pFile = DBSBufferFile(lfn = "/path/to/some/lfn", size = 600000, events = 60000)
pFile.setAlgorithm(appName = "cmsRun", appVer = "UNKNOWN",
appFam = "RECO", psetHash = "GIBBERISH",
configContent = "MOREGIBBERISH")
pFile.setDatasetPath("/bogus/dataset/path")
#pFile.addRun(Run(1, *[45]))
pFile.create()
config = self.createConfig(workerThreads = 1)
accountant = JobAccountantPoller(config)
accountant.setup()
accountant.algorithm()
self.verifyJobSuccess(self.testJob["id"])
self.verifyFileMetaData(self.testJob["id"], myReport.getAllFilesFromStep("cmsRun1"))
inputFile = File(lfn = "/store/backfill/2/unmerged/WMAgentCommissioining10/MinimumBias/RECO/rereco_GR09_R_34X_V5_All_v1/0000/outputRECORECO.root")
inputFile.load()
self.testMergeJob = Job(name = "testMergeJob", files = [inputFile])
self.testMergeJob.create(group = self.mergeJobGroup)
self.testMergeJob["state"] = "complete"
self.stateChangeAction.execute(jobs = [self.testMergeJob])
self.mergePath = os.path.join(WMCore.WMBase.getTestBase(),
"WMCore_t/FwkJobReport_t/CMSSWMergeReport.xml")
myReport = Report("mergeReco")
myReport.parse(self.mergePath)
# Fake some metadata that should be added by the stageout scripts.
for fileRef in myReport.getAllFileRefsFromStep("mergeReco"):
fileRef.size = 1024
fileRef.location = "cmssrm.fnal.gov"
fileRef.dataset = {"applicationName": "cmsRun", "applicationVersion": "CMSSW_3_4_2_patch1",
"primaryDataset": "MinimumBias", "processedDataset": "Rereco-v1",
"dataTier": "RECO"}
fwjrPath = os.path.join(self.tempDir, "MergeReport.pkl")
myReport.setTaskName('/MergeWF/None')
cmsRunStep = myReport.retrieveStep("mergeReco")
cmsRunStep.status = 0
myReport.persist(fwjrPath)
self.setFWJRAction.execute(jobID = self.testMergeJob["id"], fwjrPath = fwjrPath)
accountant.algorithm()
self.verifyJobSuccess(self.testMergeJob["id"])
self.verifyFileMetaData(self.testMergeJob["id"], myReport.getAllFilesFromStep("mergeReco"))
return
def testReportHandling(self):
"""
_testReportHandling_
Verify that we're able to parse a CMSSW report, convert it to a Report()
style report, pickle it and then have the accountant process it.
"""
self.procPath = os.path.join(WMCore.WMBase.getTestBase(),
"WMCore_t/FwkJobReport_t/CMSSWProcessingReport.xml")
myReport = Report("cmsRun1")
myReport.parse(self.procPath)
# Fake some metadata that should be added by the stageout scripts.
for fileRef in myReport.getAllFileRefsFromStep("cmsRun1"):
fileRef.size = 1024
fileRef.location = "cmssrm.fnal.gov"
fwjrPath = os.path.join(self.tempDir, "ProcReport.pkl")
cmsRunStep = myReport.retrieveStep("cmsRun1")
cmsRunStep.status = 0
myReport.setTaskName('/TestWF/None')
myReport.persist(fwjrPath)
self.setFWJRAction.execute(jobID = self.testJob["id"], fwjrPath = fwjrPath)
pFile = DBSBufferFile(lfn = "/path/to/some/lfn", size = 600000, events = 60000)
pFile.setAlgorithm(appName = "cmsRun", appVer = "UNKNOWN",
appFam = "RECO", psetHash = "GIBBERISH",
configContent = "MOREGIBBERISH")
pFile.setDatasetPath("/bogus/dataset/path")
#pFile.addRun(Run(1, *[45]))
pFile.create()
config = self.createConfig(workerThreads = 1)
accountant = JobAccountantPoller(config)
accountant.setup()
accountant.algorithm()
self.verifyJobSuccess(self.testJob["id"])
self.verifyFileMetaData(self.testJob["id"], myReport.getAllFilesFromStep("cmsRun1"))
inputFile = File(lfn = "/store/backfill/2/unmerged/WMAgentCommissioining10/MinimumBias/RECO/rereco_GR09_R_34X_V5_All_v1/0000/outputRECORECO.root")
inputFile.load()
self.testMergeJob = Job(name = "testMergeJob", files = [inputFile])
self.testMergeJob.create(group = self.mergeJobGroup)
self.testMergeJob["state"] = "complete"
self.stateChangeAction.execute(jobs = [self.testMergeJob])
self.mergePath = os.path.join(WMCore.WMBase.getTestBase(),
"WMCore_t/FwkJobReport_t/CMSSWMergeReport.xml")
myReport = Report("mergeReco")
myReport.parse(self.mergePath)
# Fake some metadata that should be added by the stageout scripts.
for fileRef in myReport.getAllFileRefsFromStep("mergeReco"):
fileRef.size = 1024
fileRef.location = "cmssrm.fnal.gov"
fileRef.dataset = {"applicationName": "cmsRun", "applicationVersion": "CMSSW_3_4_2_patch1",
"primaryDataset": "MinimumBias", "processedDataset": "Rereco-v1",
"dataTier": "RECO"}
fwjrPath = os.path.join(self.tempDir, "MergeReport.pkl")
myReport.setTaskName('/MergeWF/None')
cmsRunStep = myReport.retrieveStep("mergeReco")
cmsRunStep.status = 0
myReport.persist(fwjrPath)
self.setFWJRAction.execute(jobID = self.testMergeJob["id"], fwjrPath = fwjrPath)
accountant.algorithm()
self.verifyJobSuccess(self.testMergeJob["id"])
self.verifyFileMetaData(self.testMergeJob["id"], myReport.getAllFilesFromStep("mergeReco"))
return
def printBlocksAndFiles(self, location, blockList, type = "missing", verbose = 1):
"""
print blocks and files: verbose = 1 only blocks, verbose > 1 blocks and files
"""
for block in blockList:
print "\n"
print "#################################################################"
print "Block ID: %s Name: %s" % (block["BLOCK_ID"], block["BLOCK_NAME"])
if verbose < 2:
continue
if location == "global":
print "\n"
print "List all files from %s block from Global DBS: Get info from local DBS " % type
for file in self.localDbsApi.listFiles(blockName=block["BLOCK_NAME"]):
print "parent list: %s" % file["ParentList"]
print "file LFN: %s" % file["LogicalFileName"]
print "\n"
fileIDs = ListFiles.listFileIDsByBlockID(self.t0astDBConn, block["BLOCK_ID"])
print "List all files from %s block from Global DBS: Get info from T0AST " % type
print "====================================================="
for fileID in fileIDs:
wmbsFile = WMBSFile(id = fileID)
wmbsFile.load()
print "--------------------------------------"
print "Info from: T0AST"
print "file LFN: %s" % wmbsFile["lfn"]
print ""
file = T0ASTFile(wmbsFile)
file.datasetPathID = \
ListDatasets.listDatasetIDForWMBSFile(self.t0astDBConn, wmbsFile["id"])
datasetNames = \
ListDatasets.listDatasetNamesForWMBSFile(self.t0astDBConn, wmbsFile["id"])
file["PRIMARY_DATASET"] = datasetNames["PRIMARY"]
file["PROCESSED_DATASET"] = datasetNames["PROCESSED"]
file["DATA_TIER"] = datasetNames["TIER"]
if file["DATA_TIER"] == "RECO":
t0ParentFileList = file.getParentList(type="file")
for wmbsFile in t0ParentFileList:
t0File = T0ASTFile(wmbsFile)
t0File["BLOCK_ID"] = ListBlock.getBlockIDByFileID(self.t0astDBConn, wmbsFile["id"])
print "Block ID: %s : Parent File: %s" % (t0File["BLOCK_ID"], t0File["LFN"])
if t0File["BLOCK_ID"] != None:
blockInfo = ListBlock.getBlockInfoByID(self.t0astDBConn, t0File["BLOCK_ID"])
print "Block Name: %s \nStatus: %s" % (blockInfo["BLOCK_NAME"], blockInfo["STATUS"])
if blockInfo["STATUS"] == "InFlight" or blockInfo["MIGRATE_STATUS"] == "Migrated":
dbsBlock = self.localDbsApi.listBlocks(block_name=blockInfo["BLOCK_NAME"])
if dbsBlock == []:
print "It doesn't exist in Local dbs: Something wrong"
else:
print "Block: %s exist in Local DBS" % blockInfo["BLOCK_NAME"]
try:
for file in self.localDbsApi.listFiles(patternLFN=t0File["LFN"]):
print "File: %s exist in Local DBS" % file["LogicalFileName"]
except:
print "File doesn't exist in Local DBS"
if blockInfo["MIGRATE_STATUS"] == "Migrated":
dbsBlock = self.globalDbsApi.listBlocks(block_name=blockInfo["BLOCK_NAME"])
if dbsBlock == []:
print "It doesn't exist in Global dbs: Something wrong"
else:
print "Block: %s exist in Global DBS" % blockInfo["BLOCK_NAME"]
try:
for file in self.globalDbsApi.listFiles(patternLFN=t0File["LFN"]):
print "File: %s exist in Global DBS" % file["LogicalFileName"]
except:
print "File doesn't exist in Global DBS"
print "\n"
print "Info from Local DBS: List all parent files from %s block:" % type
try:
for pfile in self.localDbsApi.listFileParents(file["LFN"]):
print "Parent Block: %s" % pfile["Block"]["NAME"]
print "Parent File: %s" % pfile["LogicalFileName"]
print "Info from Global DBS: parent block for %s block:" % type
blockList = self.globalDbsApi.listBlocks(block_name=block["BLOCK_NAME"])
if blockList == []:
print "Global DBS doen't have block %s "% pfile["Block"]["NAME"]
else:
for dbsBlock in blockList:
print "Global DBS Parent block %s exsist" % dbsBlock["NAME"]
except Exception, ex:
print "No parents file found in Local DBS "
print "====================================================="
