def check(self, datum):
log.debug("Got check data '%s'." % datum)
status = self.factory.cfg.app.state_definitions.unknown
if self.isIn(datum, self.getOkThreshold()):
log.debug("Status data is in 'ok' threshold.")
status = self.factory.cfg.app.state_definitions.ok
# The 'current status' index hasn't been updated yet, so
# 'current status' is really 'last status', and 'last status'
# is really the run prior to last.
if self.factory.state.get('current status') not in (
self.factory.cfg.app.state_definitions.ok,
self.factory.cfg.app.state_definitions.recovering):
status = self.factory.cfg.app.state_definitions.recovering
elif self.isIn(datum, self.getWarnThreshold()):
log.debug("Status data is in 'warn' threshold.")
status = self.factory.cfg.app.state_definitions.warn
elif self.isIn(datum, self.getErrorThreshold()):
log.debug("Status data is in 'error' threshold.")
status = self.factory.cfg.app.state_definitions.error
elif self.isIn(datum, self.getFailedThreshold()):
log.debug("Status data is in 'failed' threshold.")
status = self.factory.cfg.app.state_definitions.failed
elif datum == self.factory.cfg.app.state_definitions.failed:
status = self.factory.cfg.app.state_definitions.failed
return status
def setFilename(self, filename=None):
backupDir = self.monitor.cfg.app.admin.backups.state_dir
if not filename:
filename = self.monitor.uid
self.filename = os.path.join(backupDir,
re.sub('[^0-9A-Za-z_]', '_', filename))
self.data.filename = self.filename
log.debug("Backup file named '%s'." % self.filename)
def setFilename(self, filename=None):
backupDir = self.monitor.cfg.app.admin.backups.state_dir
if not filename:
filename = self.monitor.uid
self.filename = os.path.join(backupDir,
re.sub('[^0-9A-Za-z_]', '_', filename))
self.data.filename = self.filename
log.debug("Backup file named '%s'." % self.filename)
def getPingReturn(self, results, port):
try:
gain = self.data['gain'][port]
except KeyError:
gain = 0
try:
loss = self.data['loss'][port]
except KeyError:
loss = 0
results = gain / (float(gain + loss) or 1)
log.debug('Ping results: %s' % results)
return results
def isIn(self, datum, threshold):
'''
Generic method for checking data against thresholds
'''
if self.threshold_type == 'ranged':
isInFunc = isInRange
elif self.threshold_type == 'listed':
isInFunc = isInList
elif self.threshold_type == 'exact':
isInFunc = isExactly
log.debug("Using method '%s'..." % isInFunc.func_name)
log.debug("datum: %s" % datum)
log.debug("datum type: %s" % type(datum))
log.debug("threshold: %s" % threshold)
log.debug("threshold type: %s" % type(threshold))
return isInFunc(datum, threshold)
def setInterval(self, seconds=None):
#import pdb;pdb.set_trace()
def useDef():
interval = self.cfg.defaults.interval
log.debug('Set interval from service check defaults')
return interval
if seconds:
interval = seconds
log.debug('Manually set interval')
else:
interval = self.cfg.check.interval
if interval:
log.debug('Set interval from service check config')
else:
interval = useDef()
if not interval:
interval = useDef()
self._interval = int(interval)
def connectionLost(self, reason):
# parse returned data
log.debug(self.factory.data)
parse = OutputParser(self.factory.data)
loss = parse.getPingLoss()
gain = parse.getPingGain()
host = self.getHost()
# threshold checks, messaging, and workflow
self.processRules(gain, host=host, loss=loss, gain=gain)
# update state information
self.updateState()
# dump info to log file
log.info('State Data: ' + str(self.factory.state.data) + '\n')
# final cleanup
LocalAgentClient.connectionLost(self, reason)
ClientMixin.teardown(self)
def connectionLost(self, reason):
# parse returned data
log.debug(self.factory.data)
parse = OutputParser(self.factory.data)
loss = parse.getPingLoss()
gain = parse.getPingGain()
host = self.getHost()
# threshold checks, messaging, and workflow
self.processRules(gain, host=host, loss=loss, gain=gain)
# update state information
self.updateState()
# dump info to log file
log.info('State Data: '+str(self.factory.state.data)+'\n')
# final cleanup
LocalAgentClient.connectionLost(self, reason)
ClientMixin.teardown(self)
def _setupNullClient(self):
'''
This needs to be called anytime we didn't get a connection.
Rules processing occurs
'''
self.status = '0'
log.debug('Entered null client setup...')
self.original_protocol = self.protocol
log.debug('Original Protocol: %s' % self.original_protocol)
self.protocol = NullClient()
log.debug('New Protocol: %s' % self.protocol)
self.protocol.makeConnection(self)
self.protocol = self.original_protocol
log.debug('Reverted Protocol: %s' % self.protocol)
def _setupNullClient(self):
'''
This needs to be called anytime we didn't get a connection.
Rules processing occurs
'''
self.status = '0'
log.debug('Entered null client setup...')
self.original_protocol = self.protocol
log.debug('Original Protocol: %s' % self.original_protocol)
self.protocol = NullClient()
log.debug('New Protocol: %s' % self.protocol)
self.protocol.makeConnection(self)
self.protocol = self.original_protocol
log.debug('Reverted Protocol: %s' % self.protocol)
def runTwistedFactoryEngine(rootService):
cfg = rootService.cfg
enabledServices = cfg.getEnabledServices()
# iterate through the enabled services
for pymonService in enabledServices:
serviceName = pymonService.getName()
log.debug("Service name: " + serviceName)
factoryName = pymonService.default.factory
# for each enabled service, iterate through the checks for that
# service
for check in pymonService.checks:
log.debug("Service check uri: "+check.uri)
factory = monitors.AbstractFactory(factoryName, serviceName,
check.uri)
monitor = factory.makeMonitor(cfg.configFactory(factory.uid))
# XXX this next line could be a potential memory hog
globalRegistry.factories.update({monitor.uid:monitor})
interval = monitor.getInterval()
log.debug("Service interval: %s" % interval)
service = internet.TimerService(interval, monitor)
service.setServiceParent(rootService)
def connectionLost(self, reason):
results = self.factory.data
log.debug(results)
# push the returned data through the threshold checks
status = self.rules.check(gain)
self.workflow.checkTransition(status, self.factory.cfg)
#self.rules.setMsg(results['gain'], self.getHost())
#self.rules.setSubj(self.getHost(), results['loss'])
#if self.rules.isSendMessage():
# self.rules.sendIt()
# dump info to log file
log.debug('Service: %s' % self.factory.uid)
log.debug("Status: %s for %s" % (self.rules.status, host))
# update state information
self.updateState()
# dump info to log file
log.info('State Data: ' + str(self.factory.state.data) + '\n')
# final cleanup
ClientMixin.connectionLost(self)
def connectionLost(self, reason):
results = self.factory.data
log.debug(results)
# push the returned data through the threshold checks
status = self.rules.check(gain)
self.workflow.checkTransition(status, self.factory.cfg)
#self.rules.setMsg(results['gain'], self.getHost())
#self.rules.setSubj(self.getHost(), results['loss'])
#if self.rules.isSendMessage():
# self.rules.sendIt()
# dump info to log file
log.debug('Service: %s' % self.factory.uid)
log.debug("Status: %s for %s" % (self.rules.status, host))
# update state information
self.updateState()
# dump info to log file
log.info('State Data: '+str(self.factory.state.data)+'\n')
# final cleanup
ClientMixin.connectionLost(self)
def errorHandlerPartialPage(self, failure):
failure.trap(PartialDownloadError)
log.error("Hmmm... got a partial page...")
log.debug('Return status: %s' % self.status)
def connectionMade(self):
'''
We never got a connection, and therefore can't get a connection
lost, so we need to do the things here that would normally get
done in connectionLost.
'''
ClientMixin.connectionMade(self)
log.debug("Factory in NullClient: %s" % self.factory)
status = self.factory.cfg.app.state_definitions.failed
checked_resource = self.factory.cfg.checks.uri
log.debug("Starting rules processing...")
self.rules.check(status)
self.rules.setMsg(checked_resource, self.factory.status, self.factory.message)
self.rules.setSubj(checked_resource)
if self.rules.isSendMessage():
self.rules.sendIt()
log.debug("Finished rules processing.")
# dump info to log file
log.debug('NullClient Service: %s' % self.factory.uid)
log.info(self.rules.msg)
log.info(self.rules.subj)
log.debug("NullClient Status: %s for %s" % (status, self.getHost()))
# update state information
self.updateState()
# dump info to log file
log.debug(self.factory.state)
def errorHandlerPartialPage(self, failure):
failure.trap(PartialDownloadError)
log.error("Hmmm... got a partial page...")
log.debug('Return status: %s' % self.status)
def getPingReturn(self, results):
self.data = results
log.debug("Ping results: %s" % results)
self.disconnect()
def useDef():
interval = self.cfg.defaults.interval
log.debug('Set interval from service check defaults')
return interval
def getPingReturn(self, results):
self.data = results
log.debug('Ping results: %s' % results)
self.disconnect()
def connectionLost(self, reason):
status = self.factory.status
if status:
status = int(status)
log.debug(str(dir(self.factory)))
log.debug(str(self.factory.headers))
log.debug(str(self.factory.response_headers))
checked_resource = self.factory.checkConfig.uri
# push the returned data through the threshold checks
self.rules.check(status)
self.rules.setMsg(checked_resource, status, self.factory.message)
self.rules.setSubj(checked_resource, status)
if self.rules.isSendMessage():
self.rules.sendIt()
# dump info to log file
log.debug("Service: %s" % self.factory.uid)
log.info(self.rules.msg)
log.info(self.rules.subj)
log.debug("Status: %s for %s" % (status, self.getHost()))
# update state information
self.updateState()
# dump info to log file
log.debug("State Data: " + str(self.factory.state.data) + "\n")
def connectionMade(self):
'''
We never got a connection, and therefore can't get a connection
lost, so we need to do the things here that would normally get
done in connectionLost.
'''
ClientMixin.connectionMade(self)
log.debug("Factory in NullClient: %s" % self.factory)
status = self.factory.cfg.app.state_definitions.failed
checked_resource = self.factory.cfg.checks.uri
log.debug("Starting rules processing...")
self.rules.check(status)
self.rules.setMsg(checked_resource, self.factory.status,
self.factory.message)
self.rules.setSubj(checked_resource)
if self.rules.isSendMessage():
self.rules.sendIt()
log.debug("Finished rules processing.")
# dump info to log file
log.debug('NullClient Service: %s' % self.factory.uid)
log.info(self.rules.msg)
log.info(self.rules.subj)
log.debug("NullClient Status: %s for %s" % (status, self.getHost()))
# update state information
self.updateState()
# dump info to log file
log.debug(self.factory.state)
def connectionLost(self, reason):
status = self.factory.status
if status:
status = int(status)
log.debug(str(dir(self.factory)))
log.debug(str(self.factory.headers))
log.debug(str(self.factory.response_headers))
checked_resource = self.factory.checkConfig.uri
# push the returned data through the threshold checks
self.rules.check(status)
self.rules.setMsg(checked_resource, status, self.factory.message)
self.rules.setSubj(checked_resource, status)
if self.rules.isSendMessage():
self.rules.sendIt()
# dump info to log file
log.debug('Service: %s' % self.factory.uid)
log.info(self.rules.msg)
log.info(self.rules.subj)
log.debug("Status: %s for %s" % (status, self.getHost()))
# update state information
self.updateState()
# dump info to log file
log.debug('State Data: ' + str(self.factory.state.data) + '\n')
