def run(self):
""" Run all instances. """
# Store run statistics if needed
before, after = None, None
if self.in_developer_mode and self.name != AGENT_METRICS_CHECK_NAME:
try:
before = AgentCheck._collect_internal_stats()
except Exception: # It's fine if we can't collect stats for the run, just log and proceed
self.log.debug(
"Failed to collect Agent Stats before check {0}".format(
self.name))
instance_statuses = []
for i, instance in enumerate(self.instances):
try:
min_collection_interval = instance.get(
'min_collection_interval',
self.init_config.get('min_collection_interval',
self.DEFAULT_MIN_COLLECTION_INTERVAL))
now = time.time()
if now - self.last_collection_time[i] < min_collection_interval:
self.log.debug(
"Not running instance #{0} of check {1} as it ran less than {2}s ago"
.format(i, self.name, min_collection_interval))
continue
self.last_collection_time[i] = now
check_start_time = None
if self.in_developer_mode:
check_start_time = timeit.default_timer()
self.check(copy.deepcopy(instance))
instance_check_stats = None
if check_start_time is not None:
instance_check_stats = {
'run_time': timeit.default_timer() - check_start_time
}
if self.has_warnings():
instance_status = check_status.InstanceStatus(
i,
check_status.STATUS_WARNING,
warnings=self.get_warnings(),
instance_check_stats=instance_check_stats)
else:
instance_status = check_status.InstanceStatus(
i,
check_status.STATUS_OK,
instance_check_stats=instance_check_stats)
except Exception, e:
self.log.exception("Check '%s' instance #%s failed" %
(self.name, i))
instance_status = check_status.InstanceStatus(
i,
check_status.STATUS_ERROR,
error=str(e),
tb=traceback.format_exc())
finally:
def run(self):
""" Run all instances. """
instance_statuses = []
for i, instance in enumerate(self.instances):
try:
min_collection_interval = instance.get('min_collection_interval',
self.init_config.get('min_collection_interval', self.DEFAULT_MIN_COLLECTION_INTERVAL))
now = time.time()
if now - self.last_collection_time[i] < min_collection_interval:
self.log.debug("Not running instance #{0} of check {1} as it ran less than {2}s ago".format(i, self.name, min_collection_interval))
continue
self.last_collection_time[i] = now
self.check(copy.deepcopy(instance))
if self.has_warnings():
instance_status = check_status.InstanceStatus(i,
check_status.STATUS_WARNING,
warnings=self.get_warnings()
)
else:
instance_status = check_status.InstanceStatus(i, check_status.STATUS_OK)
except Exception, e:
self.log.exception("Check '%s' instance #%s failed" % (self.name, i))
instance_status = check_status.InstanceStatus(i,
check_status.STATUS_ERROR,
error=e,
tb=traceback.format_exc()
)
instance_statuses.append(instance_status)
def run(self):
""" Run all instances. """
instance_statuses = []
for i, instance in enumerate(self.instances):
try:
self.check(instance)
instance_status = check_status.InstanceStatus(
i, check_status.STATUS_OK)
except Exception, e:
self.log.exception("Check '%s' instance #%s failed" %
(self.name, i))
instance_status = check_status.InstanceStatus(
i, check_status.STATUS_ERROR, e)
instance_statuses.append(instance_status)
def run(self):
""" Run all instances. """
instance_statuses = []
for i, instance in enumerate(self.instances):
try:
self.check(instance)
if self.has_warnings():
instance_status = check_status.InstanceStatus(
i,
check_status.STATUS_WARNING,
warnings=self.get_warnings())
else:
instance_status = check_status.InstanceStatus(
i, check_status.STATUS_OK)
except Exception, e:
self.log.exception("Check '%s' instance #%s failed" %
(self.name, i))
instance_status = check_status.InstanceStatus(
i,
check_status.STATUS_ERROR,
error=e,
tb=traceback.format_exc())
instance_statuses.append(instance_status)
def run(self):
""" Run all instances. """
# Store run statistics if needed
before, after = None, None
if self.in_developer_mode and self.name != AGENT_METRICS_CHECK_NAME:
try:
before = AgentCheck._collect_internal_stats()
except Exception: # It's fine if we can't collect stats for the run, just log and proceed
self.log.debug(
"Failed to collect Agent Stats before check {0}".format(
self.name))
instance_statuses = []
for i, instance in enumerate(self.instances):
try:
min_collection_interval = instance.get(
'min_collection_interval', self.min_collection_interval)
now = time.time()
if now - self.last_collection_time[i] < min_collection_interval:
self.log.debug(
"Not running instance #{0} of check {1} as it ran less than {2}s ago"
.format(i, self.name, min_collection_interval))
continue
self.last_collection_time[i] = now
check_start_time = None
if self.in_developer_mode:
check_start_time = timeit.default_timer()
self.check(copy.deepcopy(instance))
instance_check_stats = None
if check_start_time is not None:
instance_check_stats = {
'run_time': timeit.default_timer() - check_start_time
}
if self.has_warnings():
instance_status = check_status.InstanceStatus(
i,
check_status.STATUS_WARNING,
warnings=self.get_warnings(),
instance_check_stats=instance_check_stats)
else:
instance_status = check_status.InstanceStatus(
i,
check_status.STATUS_OK,
instance_check_stats=instance_check_stats)
except Exception as e:
self.log.exception("Check '%s' instance #%s failed" %
(self.name, i))
instance_status = check_status.InstanceStatus(
i,
check_status.STATUS_ERROR,
error=str(e),
tb=traceback.format_exc())
finally:
self._roll_up_instance_metadata()
# Discard any remaining warning so that next instance starts clean
self.get_warnings()
instance_statuses.append(instance_status)
if self.in_developer_mode and self.name != AGENT_METRICS_CHECK_NAME:
try:
after = AgentCheck._collect_internal_stats()
if self.allow_profiling:
self._set_internal_profiling_stats(before, after)
log.info(
"\n \t %s %s" %
(self.name,
pretty_statistics(self._internal_profiling_stats)))
except Exception: # It's fine if we can't collect stats for the run, just log and proceed
self.log.debug(
"Failed to collect Agent Stats after check {0}".format(
self.name))
return instance_statuses
def run(self):
""" Run all instances. """
inst = {}
hosts = []
for i, instance in enumerate(self.instances):
inst[instance['addr']] = instance
hosts.append(instance['addr'])
instance_statuses = [None] * len(hosts)
fping = Fping(hosts, self._ping_timeout)
# record elapsed time for fping
check_start_time = timeit.default_timer()
elapsed_time = 0
num = 0
failures = {}
while elapsed_time < self._last_check_time:
result = fping.run()
exec_time = timeit.default_timer()
elapsed_time = exec_time - check_start_time
num += 1
instance_check_stats = {
'run_time': timeit.default_timer() - check_start_time
}
for addr, v in result.items():
instance = inst[addr]
if v is None:
self._increment_with_tags('loss_cnt', instance)
failures[addr] = failures.get(addr, 0) + 1
if num == 1:
instance_status = check_status.InstanceStatus(
hosts.index(addr),
check_status.STATUS_WARNING,
warnings=self.get_warnings(),
instance_check_stats=instance_check_stats)
else:
self.histogram('%s.rtt' % self._basename,
v,
tags=self._instance_tags(instance))
if num == 1:
instance_status = check_status.InstanceStatus(
hosts.index(addr),
check_status.STATUS_OK,
instance_check_stats=instance_check_stats)
self._increment_with_tags('total_cnt', instance)
self._roll_up_instance_metadata()
if num == 1:
instance_statuses[hosts.index(addr)] = instance_status
for addr in failures.keys():
self.event({
'timestamp':
int(exec_time),
'event_type':
self._basename,
'msg_title':
'fping timeout',
'msg_text':
'ICMP Network Unreachable for ICMP Echo sent to %s %d times' %
(addr, failures[addr]),
'aggregation_key':
md5(addr).hexdigest()
})
elapsed_time = timeit.default_timer() - check_start_time
self.log.info("elapsed_time:%s[sec] check_times: %d" %
(round(elapsed_time, 2), num))
return instance_statuses
