def check(self, request):
if os.geteuid() != 0:
os.seteuid(0)
p = PingPing(request.alloutput, request.host, request.timeout,
request.packetsize)
if p.is_unknown_host:
return nagios.Result("PING", nagios.Status.UNKNOWN,
"Unknown host: %s" % (request.host),
request.appname)
p.run(request.count)
# ATTENTION: if nothing receive, it's critical anyway
if p.receive_count == 0:
status_code = nagios.Status.CRITICAL
else:
status_code = self.verdict(p.lost_rate, request.warn, request.crit)
msg = "Packet loss = %s %%, RTA = %.3f ms" % (p.lost_rate, p.rta)
# save traceroute(request, tr_msg)
if status_code > nagios.Status.OK:
tr_msg = self.check_traceroute(request)
self.save_output(request, tr_msg)
msg += " " + tr_msg
# generate result
r = nagios.Result("PING", status_code, msg, request.appname)
r.add_performance_data("lost_rate",
p.lost_rate,
UOM="%",
warn=request.warn,
crit=request.crit)
r.add_performance_data("rta", p.rta, UOM="ms")
return r
def get_bytes_transfer(self, request):
# read data from command line, calculate and verdict
attrs = ["Bytes_received", "Bytes_sent"]
values = []
total = 0
status_code = nagios.Status.OK
for attr in attrs:
v = float(self.get_delta_value(attr, request)) / 1024 / 1024
values.append(v)
total += v
status_code = self.superimpose(status_code, v, request.warn,
request.crit)
# build result
r = nagios.Result(request.option, status_code, '%sMB in total' % total,
request.appname)
r.add_performance_data('total',
total,
'MB',
warn=request.warn,
crit=request.crit)
r.add_performance_data('bytes_received',
values[0],
'MB',
warn=request.warn,
crit=request.crit)
r.add_performance_data('bytes_sent',
values[1],
'MB',
warn=request.warn,
crit=request.crit)
return r
def get_transactions(self, request):
# read data from command line, calculate and verdict
attrs = ["Handler_commit", "Handler_rollback"]
values = []
total = 0
status_code = nagios.Status.OK
for attr in attrs:
v = self.get_delta_value(attr, request)
values.append(v)
total += v
status_code = self.superimpose(status_code, v, request.warn,
request.crit)
# build result
r = nagios.Result(request.option, status_code,
'%s transactions' % total, request.appname)
r.add_performance_data('total',
total,
warn=request.warn,
crit=request.crit)
r.add_performance_data('commit',
values[0],
warn=request.warn,
crit=request.crit)
r.add_performance_data('rollback',
values[1],
warn=request.warn,
crit=request.crit)
return r
def get_queue_length(self, request):
stats = {}
query = "smembers resque:queues"
output = self.run_query(request, query)
total = 0
status_code = nagios.Status.OK
query_pattern = "llen resque:queue:%s"
for q in output.split('\n'):
if q:
v = self.run_query(request, query_pattern % q)
stats[q] = int(v)
total += stats[q]
sc = self.verdict(v, request.warn, request.crit)
if sc == nagios.Status.WARNING and status_code == nagios.Status.OK:
status_code = nagios.Status.WARNING
elif sc == nagios.Status.CRITICAL:
status_code = nagios.Status.CRITICAL
r = nagios.Result(request.option, status_code,
'%s jobs in queues' % total, request.appname)
r.add_performance_data('total',
total,
warn=request.warn,
crit=request.crit)
for k, v in stats.iteritems():
r.add_performance_data(k, v, warn=request.warn, crit=request.crit)
return r
def get_overall_health(self, request):
disklist = self._get_disks(request)
# load the SMART info of adaptec raid controller
status_code, message = self.check_health_status(request, disklist)
r = nagios.Result(request.option, status_code, message,
request.appname)
return r
def test_result_str(self):
# a result w/out pf data
r = nagios.Result('SLOW_QUERIES', nagios.Status.OK, '2 queries',
'MYSQL')
expected = 'SLOW_QUERIES OK: 2 queries'
self.assertEqual(expected, str(r))
# one performance data added
r.add_performance_data('slow_queries', 2, warn=10, crit=20)
expected = 'SLOW_QUERIES OK: 2 queries | slow_queries=2;10;20'
self.assertEqual(expected, str(r))
# two performance data added
r.add_performance_data('slow_queries_rate', 0.2, warn=0.5, crit=1)
expected = 'SLOW_QUERIES OK: 2 queries | slow_queries=2;10;20 slow_queries_rate=0.2;0.5;1'
self.assertEqual(expected, str(r))
def get_select_stats(self, request):
# read data from command line, calculate and verdict
attrs = [
"Select_full_join", "Select_full_range_join", "Select_range",
"Select_range_check", "Select_scan"
]
values = []
total = 0
status_code = nagios.Status.OK
for attr in attrs:
v = self.get_delta_value(attr, request)
values.append(v)
total += v
status_code = self.superimpose(status_code, v, request.warn,
request.crit)
# build result
r = nagios.Result(request.option, status_code, '%s joins' % total,
request.appname)
r.add_performance_data('total',
total,
warn=request.warn,
crit=request.crit)
r.add_performance_data('select_full_join',
values[0],
warn=request.warn,
crit=request.crit)
r.add_performance_data('select_full_range_join',
values[1],
warn=request.warn,
crit=request.crit)
r.add_performance_data('select_range',
values[2],
warn=request.warn,
crit=request.crit)
r.add_performance_data('select_range_check',
values[3],
warn=request.warn,
crit=request.crit)
r.add_performance_data('select_scan',
values[4],
warn=request.warn,
crit=request.crit)
return r
def get_result(self, request, sub_perfs, message):
status_code = nagios.Status.OK
r = nagios.Result(request.option, status_code, message,
request.appname)
critical = request.crit
for disk, attribute in sub_perfs.iteritems():
if not critical:
critical = attribute.threshold
status_code = self.superimpose(status_code,
attribute.value,
request.warn,
critical,
reverse=True)
r.add_performance_data(disk,
attribute.value,
warn=request.warn,
crit=critical)
r.set_status_code(status_code)
return r
def get_adaptec_health(self, request):
if sys.platform == "win32":
raise nagios.StatusUnknownError(
request, "Adaptec Health only supported on linux.")
disklist = self._get_disks(request)
diskdict = self._detect_adaptec(disklist, request)
if not diskdict:
raise nagios.StatusUnknownError(
request, "No Adaptec Raid Controller detected.")
message = ""
cmd = nagios.rootify("/usr/StorMan/arcconf getlogs 1 stats")
output = commands.getoutput(cmd)
if not self._validate_arcconf_output(request, output):
return
xml = output[output.index("<ControllerLog"
):output.index("</ControllerLog>") + 16]
dom = parseString(xml)
status_code = nagios.Status.OK
if not request.warn:
warn = 1
else:
warn = request.warn
sub_perfs = []
for statsnodee in dom.getElementsByTagName("physicaldrivestats"):
value = int(statsnodee.getAttribute("smartWarnCnt"))
disk = diskdict[statsnodee.getAttribute("id")]
if value > 0:
if message == "":
message = "smart warnings:"
message += " %s=%s" % (disk, value)
status_code = self.superimpose(status_code, value, warn,
request.crit)
sub_perfs.append((disk, value))
if message == "":
message = "no smart warning"
r = nagios.Result(request.option, status_code, message,
request.appname)
for disk, value in sub_perfs:
r.add_performance_data(disk, value, warn=warn, crit=request.crit)
return r
def get_row_opertions(self, request):
# read data from command line, calculate and verdict
attrs = [
"Innodb_rows_deleted", "Innodb_rows_inserted",
"Innodb_rows_updated", "Innodb_rows_read"
]
values = []
total = 0
status_code = nagios.Status.OK
for attr in attrs:
v = self.get_delta_value(attr, request)
values.append(v)
total += v
status_code = self.superimpose(status_code, v, request.warn,
request.crit)
# build result
r = nagios.Result(request.option, status_code,
'%s row operations' % total, request.appname)
r.add_performance_data('total',
total,
warn=request.warn,
crit=request.crit)
r.add_performance_data('rows_deleted',
values[0],
warn=request.warn,
crit=request.crit)
r.add_performance_data('rows_inserted',
values[1],
warn=request.warn,
crit=request.crit)
r.add_performance_data('rows_updated',
values[2],
warn=request.warn,
crit=request.crit)
r.add_performance_data('rows_read',
values[3],
warn=request.warn,
crit=request.crit)
return r
def test_wrapper():
result = nagios.Result("STATSWRAPPER", 0, appname="nagios")
result.add_performance_data("total", 1)
return result
def testfunc3(*arg, **kwargs):
result = nagios.Result("TIMER_COMMAND", nagios.Status.OK)
result.add_performance_data("total", 10)
return result
def get_replication(self, request):
return nagios.Result(request.option, nagios.Status.UNKNOWN,
"mysterious status", request.appname)
def get_network_traffic(self, request):
return nagios.Result(request.option, nagios.Status.UNKNOWN,
"mysterious status", request.appname)
def get_command_frequency(self, request):
return nagios.Result(request.option, nagios.Status.UNKNOWN,
"mysterious status", request.appname)
