def monitorJobTracker(self):
"""This method is periodically called to monitor the JobTracker of the cluster."""
try:
if self.__isIdle():
if self.__idleJobTrackerHandler:
self.__log.info('Detected cluster as idle. Calling registered callback handler.')
self.__idleJobTrackerHandler.handleIdleJobTracker()
except:
self.__log.debug('Exception while monitoring job tracker. %s' % get_exception_string())
def monitorJobTracker(self):
"""This method is periodically called to monitor the JobTracker of the cluster."""
try:
if self.__isIdle():
if self.__idleJobTrackerHandler:
self.__log.info('Detected cluster as idle. Calling registered callback handler.')
self.__idleJobTrackerHandler.handleIdleJobTracker()
except:
self.__log.debug('Exception while monitoring job tracker. %s' % get_exception_string())
def getJobsStatus(self):
"""This method should return the status of all jobs that are run on the HOD allocated
hadoop cluster"""
jobStatusList = []
try:
hadoop16Version = { 'major' : '0', 'minor' : '16' }
if self.__isCompatibleHadoopVersion(hadoop16Version):
jtStatusCommand = self.__initStatusCommand(option='-list all')
jtStatusCommand.start()
jtStatusCommand.wait()
jtStatusCommand.join()
if jtStatusCommand.exit_code() == 0:
for line in jtStatusCommand.output():
jobStatus = self.__extractJobStatus(line)
if jobStatus is not None:
jobStatusList.append(jobStatus)
except:
self.__log.debug('Exception while getting job statuses. %s' % get_exception_string())
return jobStatusList
def getJobsStatus(self):
"""This method should return the status of all jobs that are run on the HOD allocated
hadoop cluster"""
jobStatusList = []
try:
hadoop16Version = { 'major' : '0', 'minor' : '16' }
if self.__isCompatibleHadoopVersion(hadoop16Version):
jtStatusCommand = self.__initStatusCommand(option='-list all')
jtStatusCommand.start()
jtStatusCommand.wait()
jtStatusCommand.join()
if jtStatusCommand.exit_code() == 0:
for line in jtStatusCommand.output():
jobStatus = self.__extractJobStatus(line)
if jobStatus is not None:
jobStatusList.append(jobStatus)
except:
self.__log.debug('Exception while getting job statuses. %s' % get_exception_string())
return jobStatusList
opList = self.__check_operation(operation)
if self.__opCode == 0:
if not self.__userState.checkStateFile(CLUSTER_DATA_FILE, (os.R_OK,)):
self.__log.critical(INVALID_STATE_FILE_MSGS[0] % \
self.__userState.get_state_file())
self.__opCode = 1
return self.__opCode
getattr(self, "_op_%s" % opList[0])(opList)
except HodInterruptException, h:
self.__log.critical("op: %s failed because of a process interrupt." \
% operation)
self.__opCode = HOD_INTERRUPTED_CODE
except:
self.__log.critical("op: %s failed: %s" % (operation,
get_exception_error_string()))
self.__log.debug(get_exception_string())
self.__cleanup()
self.__log.debug("return code: %s" % self.__opCode)
return self.__opCode
def script(self):
errorFlag = False
errorMsgs = []
scriptRet = 0 # return from the script, if run
script = self.__cfg['hod']['script']
nodes = self.__cfg['hod']['nodecount']
clusterDir = self.__cfg['hod']['clusterdir']
if self.__opCode == 0:
if not self.__userState.checkStateFile(CLUSTER_DATA_FILE,
(os.R_OK, )):
self.__log.critical(INVALID_STATE_FILE_MSGS[0] % \
self.__userState.get_state_file())
self.__opCode = 1
return self.__opCode
getattr(self, "_op_%s" % opList[0])(opList)
except HodInterruptException, h:
self.__log.critical("op: %s failed because of a process interrupt." \
% operation)
self.__opCode = HOD_INTERRUPTED_CODE
except:
self.__log.critical("op: %s failed: %s" %
(operation, get_exception_error_string()))
self.__log.debug(get_exception_string())
self.__cleanup()
self.__log.debug("return code: %s" % self.__opCode)
return self.__opCode
def script(self):
errorFlag = False
errorMsgs = []
scriptRet = 0 # return from the script, if run
script = self.__cfg['hod']['script']
nodes = self.__cfg['hod']['nodecount']
clusterDir = self.__cfg['hod']['clusterdir']
