def probe(self):
"""Query system state and return metrics.
:return: generator that emits
:class:`~nagiosplugin.metric.Metric` objects
"""
# Execute `systemctl is-active <service>` and get output
# - active
# - inactive (by unkown unit file)
# - failed
try:
p = subprocess.Popen(['systemctl', 'is-active', self.unit],
stderr=subprocess.PIPE,
stdin=subprocess.PIPE,
stdout=subprocess.PIPE)
stdout, stderr = p.communicate()
except OSError as e:
raise nagiosplugin.CheckError(e)
if stderr:
raise nagiosplugin.CheckError(stderr)
if stdout:
for line in io.StringIO(stdout.decode('utf-8')):
active = line.strip()
yield Metric(name=self.unit, value=active, context='unit')
def probe(self):
# Execute systemctl is-active and get output
try:
p = subprocess.Popen(['systemctl', 'is-active', self.service],
stderr=subprocess.PIPE,
stdin=subprocess.PIPE,
stdout=subprocess.PIPE)
pres, err = p.communicate()
except OSError as e:
raise nagiosplugin.CheckError(e)
if err:
raise nagiosplugin.CheckError(err)
if pres:
result = ""
for line in io.StringIO(pres.decode('utf-8')):
result = "%s %s" % (result, line.split(' ')[0])
result = result.strip()
if result == "active":
return [
nagiosplugin.Metric('systemd', (True, None),
context='systemd')
]
else:
return [
nagiosplugin.Metric('systemd', (False, self.service),
context='systemd')
]
return [
nagiosplugin.Metric('systemd', (False, "No Service given"),
context='systemd')
]
def execute_cli(args: typing.Union[str, typing.Iterator[str]]) -> str:
"""Execute a command on the command line (cli = command line interface))
and capture the stdout. This is a wrapper around ``subprocess.Popen``.
:param args: A list of programm arguments.
:raises nagiosplugin.CheckError: If the command produces some stderr output
or if an OSError exception occurs.
:return: The stdout of the command.
"""
try:
p = subprocess.Popen(args,
stderr=subprocess.PIPE,
stdin=subprocess.PIPE,
stdout=subprocess.PIPE)
stdout, stderr = p.communicate()
except OSError as e:
raise nagiosplugin.CheckError(e)
if p.returncode != 0:
raise nagiosplugin.CheckError('The command exits with a none-zero'
'return code ({})'.format(p.returncode))
if stderr:
raise nagiosplugin.CheckError(stderr)
if stdout:
stdout = stdout.decode('utf-8')
return stdout
def probe(self):
log.debug("Graylog2Throughput.probe started")
try:
# /system/throughput sometimes returns 0 (which is normal),
# retry max_retry times before returns 0
for _ in xrange(self._max_retry):
log.debug("try #%d", _)
r = requests.get(
self._api_url, auth=(self._username, self._password))
log.debug("response: %s", r.content)
throughput = r.json()['throughput']
log.debug("throughput : %s", str(throughput))
if throughput != 0:
break
time.sleep(1) # take a break before retry
except requests.ConnectionError:
log.warn("Could not conect to server: %s", self._api_url)
raise nagiosplugin.CheckError(
'Could not connect to graylog2 server: {}'.format(
self._api_url))
except ValueError:
log.warn("Could not parse response")
raise nagiosplugin.CheckError(
'Invalid response from graylog2 server: {}'.format(
self._api_url))
log.debug("Graylog2Throughput finished")
log.debug("returning %d", int(throughput))
return [
nagiosplugin.Metric('throughput', int(throughput), min=0)
]
def probe(self):
"""Query system state and return metrics.
:return: generator that emits
:class:`~nagiosplugin.metric.Metric` objects
"""
try:
p = subprocess.Popen(['systemd-analyze'],
stderr=subprocess.PIPE,
stdin=subprocess.PIPE,
stdout=subprocess.PIPE)
stdout, stderr = p.communicate()
except OSError as e:
raise nagiosplugin.CheckError(e)
if stderr:
raise nagiosplugin.CheckError(stderr)
if stdout:
stdout = stdout.decode('utf-8')
# First line:
# Startup finished in 1.672s (kernel) + 21.378s (userspace) =
# 23.050s
# On raspian no second line
# Second line:
# graphical.target reached after 1min 2.154s in userspace
match = re.search(r'reached after (.+) in userspace', stdout)
if not match:
match = re.search(r' = (.+)\n', stdout)
yield Metric(name='startup_time',
value=format_timespan_to_seconds(match.group(1)),
context='startup_time')
def probe(self):
# Execute systemctl --failed --no-legend and get output
try:
p = subprocess.Popen(['systemctl', '--failed', '--no-legend'],
stderr=subprocess.PIPE,
stdin=subprocess.PIPE,
stdout=subprocess.PIPE)
pres, err = p.communicate()
except OSError as e:
raise nagiosplugin.CheckError(e)
if err:
raise nagiosplugin.CheckError(err)
if pres:
result = ""
for line in io.StringIO(pres.decode('utf-8')):
result = "%s %s" % (result, line.split(' ')[0])
return [
nagiosplugin.Metric('systemd', (False, result),
context='systemd')
]
return [
nagiosplugin.Metric('systemd', (True, None), context='systemd')
]
def probe(self):
"""Query system state and return metrics.
:return: generator that emits
:class:`~nagiosplugin.metric.Metric` objects
"""
# We don’t use `systemctl --failed --no-legend`, because we want to
# collect performance data of all units.
try:
p = subprocess.Popen(
['systemctl', 'list-units', '--all', '--no-legend'],
stderr=subprocess.PIPE,
stdin=subprocess.PIPE,
stdout=subprocess.PIPE)
stdout, stderr = p.communicate()
except OSError as e:
raise nagiosplugin.CheckError(e)
if stderr:
raise nagiosplugin.CheckError(stderr)
# Dictionary to store all units according their active state.
units = {
'failed': [],
'active': [],
'activating': [],
'inactive': [],
}
if stdout:
# Output of `systemctl list-units --all --no-legend`:
# UNIT LOAD ACTIVE SUB JOB DESCRIPTION
# foobar.service loaded active waiting Description text
count_units = 0
for line in io.StringIO(stdout.decode('utf-8')):
split_line = line.split()
# foobar.service
unit = split_line[0]
# failed
active = split_line[2]
# Only count not excludes units.
if unit not in self.excludes:
units[active].append(unit)
count_units += 1
for unit in units['failed']:
if unit not in self.excludes:
yield Metric(name=unit, value='failed', context='unit')
for active, unit_names in units.items():
yield Metric(name='units_{}'.format(active),
value=len(units[active]),
context='performance_data')
yield Metric(name='count_units',
value=count_units,
context='performance_data')
if len(units['failed']) == 0:
yield Metric(name='all', value=None, context='unit')
def probe(self):
"""
:return: generator that emits
:class:`~nagiosplugin.metric.Metric` objects
"""
try:
p = subprocess.Popen(['systemctl', 'list-timers', '--all'],
stderr=subprocess.PIPE,
stdin=subprocess.PIPE,
stdout=subprocess.PIPE)
stdout, stderr = p.communicate()
except OSError as e:
raise nagiosplugin.CheckError(e)
if stderr:
raise nagiosplugin.CheckError(stderr)
# NEXT LEFT
# Sat 2020-05-16 15:11:15 CEST 34min left
# LAST PASSED
# Sat 2020-05-16 14:31:56 CEST 4min 20s ago
# UNIT ACTIVATES
# apt-daily.timer apt-daily.service
if stdout:
lines = stdout.decode('utf-8').splitlines()
table_heading = lines[0]
self.column_boundaries = self.detect_column_boundaries(
table_heading)
# Remove the first line because it is the header.
# Remove the two last lines: empty line + "XX timers listed."
table_body = lines[1:-2]
state = nagiosplugin.Ok # ok
for row in table_body:
unit = self.get_column_text(row, 'UNIT')
if self.re_match(unit):
continue
next_date_time = self.get_column_text(row, 'NEXT')
if next_date_time == 'n/a':
passed_text = self.get_column_text(row, 'PASSED')
if passed_text == 'n/a':
state = nagiosplugin.Critical
else:
passed = format_timespan_to_seconds(passed_text)
if passed_text == 'n/a' or passed >= self.critical:
state = nagiosplugin.Critical
elif passed >= self.warning:
state = nagiosplugin.Warn
yield Metric(name=unit, value=state, context='dead_timers')
def probe(self):
"""
Querys the REST-API and create throughput metrics.
:return: a throughput metric.
"""
_log.info('Reading XML from: %s', self.xml_obj.build_request_url())
api_outbytes, api_inbytes = 0, 0
current_time = get_time()
soup = self.xml_obj.read()
ifnet = soup.find('ifnet')
for item in ifnet.find_all('entry'):
api_inbytes = Finder.find_item(item, 'ibytes')
api_outbytes = Finder.find_item(item, 'obytes')
_log.debug('Path to statefile: %r' % get_statefile_path())
with np.Cookie(get_statefile_path()) as cookie:
old_inbytes = cookie.get(self.host + self.interface_name + 'i', api_inbytes)
old_outbytes = cookie.get(self.host + self.interface_name + 'o', api_outbytes)
old_time = cookie.get(self.host + self.interface_name + 't', current_time)
if not api_inbytes or not api_outbytes or float(api_inbytes) < 0 or float(api_outbytes) < 0:
raise np.CheckError('Couldn\'t get a valid value!')
cookie[self.host + self.interface_name + 'i'] = api_inbytes
cookie[self.host + self.interface_name + 'o'] = api_outbytes
cookie[self.host + self.interface_name + 't'] = current_time
if float(api_inbytes) < float(old_inbytes) or float(api_outbytes) < float(old_outbytes):
raise np.CheckError('Couldn\'t get a valid value: Found throughput less then old!')
diff_time = int(current_time) - int(old_time)
if diff_time > 0:
in_bits_per_second = round(
((float(api_inbytes) - float(old_inbytes)) / diff_time) * 8, 2)
out_bits_per_second = round(
((float(api_outbytes) - float(old_outbytes)) / diff_time) * 8,
2)
else:
raise np.CheckError(
'Difference between old timestamp and new timestamp is less '
'or equal 0: If it is the first time you run the script, '
'please execute it again!')
return [
np.Metric('in_bps_' + str(self.interface_name), in_bits_per_second,
min=0),
np.Metric('out_bps_' + str(self.interface_name), out_bits_per_second,
min=0)]
def probe(self):
oid_role = '1.3.6.1.4.1.2636.3.1.8.1.12.1.1.0.0'
raw_role = str(self.argoss_snmp.fetch_oid(oid_role))
if ("Single" in raw_role):
role = "single"
elif ("node" in raw_role):
role = "node"
else:
raise nagiosplugin.CheckError("Error: this host is a slave.")
data = {}
self.argoss_snmp.gather_data(data, '1.3.6.1.4.1.2636.3.1.13.1', {
5: 'operatings',
6: 'status',
11: 'mem'
})
operatings = {}
for index, opDesc in data['operatings'].items():
if (role == "single"):
m = re.search('^Routing\ Engine (\d+)$', str(opDesc))
if m:
descr = 'mem_re' + m.group(1)
operatings[index] = [descr]
m = re.search('^FPC:.*SPC.*\ (.*)$', str(opDesc))
if m:
descr = 'mem_spc@' + m.group(1)
operatings[index] = [descr]
elif ("node" in role):
m = re.search('^' + role + '\ Routing\ Engine (\d+)$',
str(opDesc))
if m:
descr = 'mem_re' + m.group(1)
operatings[index] = [descr]
m = re.search('^' + role + '\ FPC:.*SPC.*\ (.*)$', str(opDesc))
if m:
descr = 'mem_spc@' + m.group(1)
operatings[index] = [descr]
total_load = 0
for index, descr in operatings.items():
if int(data['status'][index]) == 2:
total_load = total_load + int(data['mem'][index])
if not len(operatings):
raise nagiosplugin.CheckError(
"Impossible de trouver une carte MEM")
alert_mem_percent = total_load / len(operatings)
yield nagiosplugin.Metric('alert_mem_percent',
alert_mem_percent,
None,
context='alert_mem_percent')
def get_containers(self):
try:
result = rq.get(self.url, headers=self.headers, verify=self.verify)
except Exception as e:
raise np.CheckError(
'Failed to query API: {}'.format(e)
)
_log.debug('API Response: {}'.format(result.content))
if result.status_code != 200:
raise np.CheckError(
'http status code was {}'.format(result.status_code)
)
return result
def probe(self):
log.debug("PgSQLQuery.probe started")
try:
log.debug("connecting with postgresql")
c = psycopg2.connect(host=self.host,
port=self.port,
user=self.user,
password=self.passwd,
database=self.database)
cursor = c.cursor()
log.debug("about to execute query: %s", self.query)
cursor.execute(self.query)
records = cursor.rowcount
log.debug("resulted in %d records", records)
log.debug(records)
log.debug(cursor.fetchall())
except psycopg2.Error as err:
log.critical(err)
raise nap.CheckError(
'Something went wrong with '
'PostgreSQL query operation, Error: {}'.format(err))
log.debug("PgSQLQuery.probe finished")
log.debug("returning %d", records)
return [nap.Metric('record', records, context='records')]
def load_filesystem(self):
data = {}
self.argoss_snmp.gather_data(data,
'1.3.6.1.2.1.25.2.3.1',
{3: 'label',
4: 'alloc',
5: 'size',
6: 'used'})
id_fs = ''
for l in data['label']:
data['label'][l] = str(data['label'][l])
if(data['label'][l].split(':\\')[0].lower() ==
self.label_fs.lower()):
id_fs = l
if not id_fs:
raise nagiosplugin.CheckError('Filesystem not found')
alloc = int(data['alloc'][id_fs])
size = int(data['size'][id_fs])
used = int(data['used'][id_fs])
fs_used = used * alloc
fs_total = size * alloc
if fs_total == 0:
return fs_used, fs_total, 0
alert_fs_percent = round(100 * (used / size), 2)
return fs_used, fs_total, alert_fs_percent
def probe(self):
c = Controller(self._ctrl_addr,
self._ctrl_user,
self._ctrl_passwd,
ssl_verify=False)
for ap in c.get_aps():
if ap.get('name') == self._hostname:
if self._metric == Metrics.STATUS:
return [nagiosplugin.Metric('status', self._get_state(ap), context='null')]
elif self._metric == Metrics.CPU:
return [nagiosplugin.Metric('cpu', self._get_cpu(ap), context='cpu')]
elif self._metric == Metrics.MEMORY:
return [nagiosplugin.Metric('memory', self._get_mem(ap), context='memory')]
elif self._metric == Metrics.SPEED:
return [nagiosplugin.Metric('speed', self._get_speed(ap), context='speed')]
elif self._metric == Metrics.SATISFACTION:
return [nagiosplugin.Metric('satisfaction', self._get_satisfaction(ap), context='satisfaction')]
elif self._metric == Metrics.TEMPERATURE:
return [nagiosplugin.Metric('temperature', self._get_temperature(ap), context='temperature')]
elif self._metric == Metrics.OVERHEATING:
return [nagiosplugin.Metric('overheating', self._get_overheating(ap), context='null')]
elif self._metric == Metrics.POWER_LEVEL:
return [nagiosplugin.Metric('power_level', self._get_power_level(ap), context='power_level')]
raise nagiosplugin.CheckError('unable to find specified device: {}'.format(self._hostname))
def probe(self):
oid_desc = '1.3.6.1.2.1.47.1.1.1.1.2'
oid_indice = '1.3.6.1.4.1.9.9.109.1.1.1.1.2'
oid_value = '1.3.6.1.4.1.9.9.109.1.1.1.1.7'
fo = 0.0
ro = 0.0
i = 0
request = self.argoss_snmp.snmp_table(oid_desc)
oid_desc_found = {}
for response in request:
if 'CPU 0 of module R0' in str(response.value):
oid_desc_found["ro"] = ((str(response.oid)).split('.'))[-1]
if 'CPU 0 of module F0' in str(response.value):
oid_desc_found["fo"] = ((str(response.oid)).split('.'))[-1]
request = self.argoss_snmp.snmp_table(oid_indice)
for response in request:
if str(response.oid_index) == str(oid_desc_found["ro"]):
indice_found = ((str(response.oid)).split('.'))[-1]
ro = float(
self.argoss_snmp.fetch_oid(oid_value + "." + indice_found))
i += 1
if str(response.oid_index) == str(oid_desc_found["fo"]):
indice_found = ((str(response.oid)).split('.'))[-1]
fo = float(
self.argoss_snmp.fetch_oid(oid_value + "." + indice_found))
i += 1
if i == 2:
yield nagiosplugin.Metric('alert_cpu_percent', (fo + ro) / 2,
context='alert_cpu_percent')
else:
raise nagiosplugin.CheckError('CPU not found')
def load_memory(self):
data = {}
memory = {'used': 0, 'free': 0}
self.argoss_snmp.gather_data(data,
'1.3.6.1.4.1.9.9.221.1.1.1.1',
{6: 'valid',
18: 'used',
20: 'free'})
self.argoss_snmp.gather_data(data,
'1.3.6.1.4.1.9.9.48.1.1.1',
{4: 'valid',
5: 'used',
6: 'free'})
for label in data:
if label != 'valid':
for value in data[label]:
if int(data['valid'][value]) == 1:
memory[label] = (int(memory[label]) +
int(data[label][value]))
try:
return (
memory['used'],
memory['used']+memory['free'],
round(100*(memory['used']/(memory['free']+memory['used'])), 2))
except ZeroDivisionError:
raise nagiosplugin.CheckError('Memory pool not available')
def probe(self):
logger.debug("EventCountCheck.probe started")
# send an event for testing
dsn = self._sentry_dsn
if not dsn.startswith("requests+"):
dsn = "requests+" + dsn
client = raven.Client(dsn=dsn)
logger.debug("send monitoring messaage to sentry")
client.captureMessage("sentry monitoring")
try:
r = requests.get(self._url,
auth=(self._public_key, self._secret_key),
verify=self._verify_ssl)
data = r.json()
logger.debug("response: %s", data)
events = 0
for group in data:
events += int(group["count"])
logger.debug("number of events: %d", events)
return [nagiosplugin.Metric('number_of_events', events, min=0)]
except requests.ConnectionError as err:
raise nagiosplugin.CheckError("Could not connect to Sentry: %s",
err)
def pkg_audit(self, jail=None):
"""Run pkg audit.
We choose here to raise UNKNOWN status if we encoutered a host|jail
which in pkg audit -F has not been runned.
"""
self.audit_cmd = 'pkg audit'
if jail is not None:
self.audit_cmd = 'pkg -j %s audit' % jail
self.hostname = jail
_log.debug('querying system with "%s" command', self.audit_cmd)
stdout, stderr = _popen(self.audit_cmd.split())
if not isinstance(stderr, str): # pragma: no cover
stderr = stderr.decode()
if not isinstance(stdout, str): # pragma: no cover
stdout = stdout.decode()
if stderr:
message = stderr.splitlines()[-1]
if message.startswith('pkg: vulnxml file'):
# message = "Try running 'pkg audit -F' first"
message = stderr.split('.')[-1]
message = "%s %s" % (self.hostname, message)
_log.info(message)
raise nagiosplugin.CheckError(message)
else:
stdout = stdout.splitlines()[-1]
problems = int(stdout.split()[0])
return problems
def main():
argp = argparse.ArgumentParser(prog='check_dns', formatter_class=argparse.ArgumentDefaultsHelpFormatter)
argp.add_argument('--domain', '-d', type=str, required=True, action='store',
help='Domain to check')
argp.add_argument('--insecure-is-ok', action='store_true', default=False,
help='If the delegation is insecure, do not warn')
argp.add_argument('--expire-warn', default=72, metavar='HOURS',
help='Warn if RRSIG expiry is within this many hours')
argp.add_argument('--expire-crit', default=48, metavar='HOURS',
help='Crit if RRSIG expiry is within this many hours')
argp.add_argument('--verbose', '-v', action='count', default=0,
help='Be more verbose')
args = argp.parse_args()
if args.expire_crit > args.expire_warn:
raise nagiosplugin.CheckError('RRSIG Critical value ({}) higher than Warning value ({})'.format(
args.expire_crit, args.expire_warn
))
check = nagiosplugin.Check(
DNS(args.domain, args.insecure_is_ok),
DNSSECContext('dnssec'),
RRSIGContext('rrsig'),
RRSIGExpitationContext('rrsig_expiration',
warn_seconds=args.expire_warn * 60 * 60,
crit_seconds=args.expire_crit * 60 * 60),
DNSSummary())
check.main(args.verbose)
def probe(self):
r = None
proxies = {'http': '', 'https': ''}
try:
r = requests.get(self.url, proxies=proxies, timeout=2)
except requests.exceptions.ConnectionError:
print("Connection refused")
if r:
"""soup = BeautifulSoup(r.content, "lxml")
parsed = soup.find(re.compile('int|real'))
try:
val = float(parsed['val'])
except (KeyError, TypeError):
raise nagiosplugin.CheckError(
'Value not available')"""
parsed = xmltodict.parse(r.content)
try:
val = float(parsed[list(parsed.keys())[0]]['@val'])
except (KeyError, TypeError):
raise nagiosplugin.CheckError('Value is not available')
yield nagiosplugin.Metric(self.metric,
val,
context='alert_dogate_' + self.metric)
def __init__(self, dsn_file):
try:
loaded = unserialize_yaml(dsn_file, critical=True)
sentry_dsn = loaded["dsn"]
self._sentry_dsn = sentry_dsn
logger.debug("sentry_dsn: %s", sentry_dsn)
except IOError as err:
logger.error(
"Sentry monitoring DSN file %s does not exist"
"or is unreadable. Error: %s", dsn_file, err)
raise nagiosplugin.CheckError(
"Invalid sentry monitoring DSN file: {}".format(dsn_file))
dsn_parsed = raven.load(sentry_dsn)
self._public_key = dsn_parsed["SENTRY_PUBLIC_KEY"]
self._secret_key = dsn_parsed["SENTRY_SECRET_KEY"]
transport_options = dsn_parsed["SENTRY_TRANSPORT_OPTIONS"]
self._verify_ssl = False if (
"verify_ssl" in transport_options
and transport_options["verify_ssl"] == "0") else True
# API format: GET /api/0/projects/{project_id}/groups/
server = dsn_parsed["SENTRY_SERVERS"][0]
project_id = dsn_parsed["SENTRY_PROJECT"]
parsed = urlparse.urlparse(server)
self._url = urlparse.urlunparse((
parsed.scheme,
parsed.netloc,
"/api/0/projects/{}/groups/".format(project_id),
parsed.params,
parsed.query,
parsed.fragment,
))
def probe(self):
try:
self.fetch_status()
except pymysql.err.OperationalError as e:
raise nagiosplugin.CheckError(e)
yield nagiosplugin.Metric('status_ap',
self.status,
context='status_ap')
def probe(self):
oid = '1.3.6.1.4.1.12356.101.4.1.8'
request = argoss_libs.snmp.tools.snmp_table(oid, self.community,
self.host, self.port)
value = None
for errorIndication, errorStatus, errorIndex, varBinds in request:
if errorIndication:
raise nagiosplugin.CheckError(errorIndication)
elif errorStatus:
raise nagiosplugin.CheckError(errorStatus)
value = varBinds[0][1]
if value is None:
raise nagiosplugin.CheckError('Session not found')
yield nagiosplugin.Metric('alert_session',
value,
None,
context='alert_session')
def get_node_by_certname(self, certname):
""" gets a pypuppetdb node object given a certname"""
try:
node = self.pdb.node(certname)
except requests.exceptions.HTTPError as e:
# catch 404 - node not found; raise anything else
if e.message == '404 Client Error: Not Found':
raise nagiosplugin.CheckError(
'UNKNOWN: Node %s not found in PuppetDB on %s' %
(certname, self.puppetdb_host))
raise e
if not node:
raise nagiosplugin.CheckError(
'UNKNOWN: Node %s not found in PuppetDB on %s' %
(certname, self.puppetdb_host))
_log.info("Found node in PuppetDB")
return node
def fetch_status(self):
connection = pymysql.connect(host=self.host,
user='******',
password='******',
db='WIFI',
charset='utf8',
connect_timeout=4,
cursorclass=pymysql.cursors.DictCursor)
with connection.cursor() as cursor:
sql = 'SELECT * FROM ap_alarms WHERE name LIKE %s'
cursor.execute(sql, (self.ap))
if cursor.rowcount == 0:
raise nagiosplugin.CheckError('No equipment available')
for row in cursor:
if 'status' in row:
self.status = row['status']
else:
raise nagiosplugin.CheckError('No status available')
def evaluate(self, metric, resource):
if str(metric) == 'Up':
return nagiosplugin.result.Result(nagiosplugin.state.Ok,
'Borne OK', metric)
elif str(metric) == 'Down':
return nagiosplugin.result.Result(nagiosplugin.state.Critical,
'Borne KO', metric)
else:
raise nagiosplugin.CheckError('Unknown status')
def query_from_authority(zone):
"""Send query using external tool to fetch serial from NS servers of the zone"""
proc = subprocess.Popen([executable, zone]+parameters_from_authority.split(),
stdout=subprocess.PIPE,
stderr=subprocess.PIPE)
answers,error = proc.communicate()
if not isinstance(answers, str): # pragma: no cover
answers = answers.decode()
if not isinstance(error, str): # pragma: no cover
error = error.decode()
if answers.find("timed out") >= 0:
raise nagiosplugin.CheckError("A server timed out")
if not len(answers):
raise nagiosplugin.CheckError("No result. Domain probably does not exist")
answers = answers.strip().splitlines()
return [(int(arguments[3]), arguments[10]) for arguments in (answer.split() for answer in answers)]
def probe(self):
# Execute systemctl is-active and get output
try:
p = subprocess.Popen(['systemctl', 'is-active', self.service],
stderr=subprocess.PIPE,
stdin=subprocess.PIPE,
stdout=subprocess.PIPE)
stdout, stderr = p.communicate()
except OSError as e:
raise nagiosplugin.CheckError(e)
if stderr:
raise nagiosplugin.CheckError(stderr)
if stdout:
for line in io.StringIO(stdout.decode('utf-8')):
yield nagiosplugin.Metric(self.service,
line.strip(),
context='systemd')
def __request(self, url):
login_header = {}
if len(self.token) != 0:
login_header['Private-Token'] = self.token
try:
response = requests.get(self.url + url, headers=login_header)
except requests.exceptions.ConnectionError:
raise nagiosplugin.CheckError('connection to {0} failed'.format(
self.url))
if response.status_code != 200:
raise nagiosplugin.CheckError(
self.http_response_messages.get(response.status_code,
'unknown failure'))
return response.json()
def probe(self):
"""
Run restic and parse its return code
:return:
"""
# For some reason, check.main() is the only place where exceptions are
# printed nicely
if not self.repo and not os.environ.get('RESTIC_REPOSITORY'):
raise nagiosplugin.CheckError(
'Please specify repository location (-r, --repo or '
'$RESTIC_REPOSITORY)')
if not self.password_file and \
not (os.environ.get('RESTIC_PASSWORD') or
os.environ.get('RESTIC_PASSWORD_FILE')):
raise nagiosplugin.CheckError(
'Please specify password or its location (-p, --password-file,'
' $RESTIC_PASSWORD or $RESTIC_PASSWORD_FILE)')
cmd = [self.restic_bin, 'check', '--quiet', '--no-lock']
if self.sudo:
cmd = ['sudo'] + cmd
if self.repo:
cmd.extend(['--repo', self.repo])
if self.password_file:
cmd.extend(['--password-file', self.password_file])
_log.info('Using command: %s' % ' '.join(cmd))
try:
restic_result = subprocess.check_output(cmd,
stderr=subprocess.STDOUT)
except subprocess.CalledProcessError as e:
yield nagiosplugin.Metric(self.repo, False, context='health_check')
self.stderr = e.output.decode()
except IOError as e:
raise nagiosplugin.CheckError('Failed to run %s: %s' %
(' '.join(cmd), e))
else:
_log.debug('Got output: %s' % restic_result)
yield nagiosplugin.Metric(self.repo, True, context='health_check')
