monitoring-check-systemd-service/check-systemd-service

#!/usr/bin/python3
# -*- coding: utf-8 -*-

"""Nagios plugin to check a systemd service on different properties"""

import argparse
import logging
import collections

try:
        import nagiosplugin
except ImportError as e:
    print("Please install python3-nagiosplugin")
    raise e

try:
        from gi.repository.Gio import DBusProxy, BusType
except ImportError as e:
    print("Please install python3-gi")
    raise e

_log = logging.getLogger('nagiosplugin')

class Systemd_Service(nagiosplugin.Resource):
    """One Systemd Service"""

    def __init__(self, **kwords):
        _log.debug('Initializing Systemd_Service with %r', (kwords))
        for key, value in kwords.items():
            self.__setattr__(key, value)
        self.normalize()

    def connect_systemd(self):
        # initializing systemd dbus connection
        systemd = DBusProxy.new_for_bus_sync(BusType.SYSTEM,
                                             0,
                                             None,
                                             'org.freedesktop.systemd1',
                                             '/org/freedesktop/systemd1',
                                             'org.freedesktop.systemd1.Manager',
                                             None)
        try:
            loadedUnit = systemd.LoadUnit('(s)', self.unit)
        except Exception as e:
            _log.error(e)
            raise e

        service = DBusProxy.new_for_bus_sync(BusType.SYSTEM,
                                             0,
                                             None,
                                             'org.freedesktop.systemd1',
                                             loadedUnit,
                                             'org.freedesktop.systemd1.Unit',
                                             None)
        self.service = service
        self.activeStateD = collections.defaultdict(lambda: None, {
            'active':       1.0,
            'reloading':    0.9,
            'activating':   0.8,
            'deactivating': 0.4,
            'inactive':     0.0,
            'failed':       0.0,
        })

    def normalize(self):
        if '.' in self.unit:
            _log.debug('Found \'.\' in ServiceName, so assuming you know what youre asking for')
        else:
            self.unit = self.unit + '.service'
            _log.debug('Normalized unitname to check to %r', self.unit)

    @property
    def name(self):
        """formatting the Testname (will be formatted as uppercase letters)"""
        return "SYSTEMD SERVICE %s SERVICESTATE" % (self.unit.split('.service')[0])


    def activestate(self):
        """
        ActiveState contains a state value that reflects whether the unit is
        currently active or not. The following states are currently defined:
        active, reloading, inactive, failed, activating, deactivating. active
        indicates that unit is active (obviously...). reloading indicates
        that the unit is active and currently reloading its configuration.
        inactive indicates that it is inactive and the previous run was
        successful or no previous run has taken place yet. failed indicates
        that it is inactive and the previous run was not successful (more
        information about the reason for this is available on the unit type
        specific interfaces, for example for services in the Result property,
        see below). activating indicates that the unit has previously been
        inactive but is currently in the process of entering an active state.
        Conversely deactivating indicates that the unit is currently in the
        process of deactivation.
        """
        t = self.service.get_cached_property('ActiveState').unpack()
        _log.debug('ServiceState of %r is %r', self.service, t)
        return t

    def substate(self):
        """
        SubState encodes states of the same state machine that ActiveState
        covers, but knows more fine-grained states that are unit-type-specific.
        Where ActiveState only covers six high-level states, SubState covers
        possibly many more low-level unit-type-specific states that are mapped
        to the six high-level states. Note that multiple low-level states might
        map to the same high-level state, but not vice versa. Not all
        high-level states have low-level counterparts on all unit types. At
        this point the low-level states are not documented here, and are more
        likely to be extended later on than the common high-level states
        explained above.
        """
        t = self.service.get_cached_property('SubState').unpack()
        _log.debug('Substate of %r is %r', self.service, t)
        return t

    def probe(self):
        """ Create check metric for Systemd Service"""
        self.connect_systemd()
        self.running_service_found = self.activeStateD[self.activestate()]
        self.service_state = (self.activestate(), self.substate())
        yield  nagiosplugin.Metric('service_state', self.service_state)

class Service_Context(nagiosplugin.Context):

    def evaluate(self, metric, recource):
        # possible Values are:
        # nagiosplugin.Ok, nagiosplugin.Warn, nagiosplugin.Critical, nagiosplugin.Unknown
        resultD = collections.defaultdict( lambda: nagiosplugin.Unknown, {
            'active':       nagiosplugin.Ok,
            'reloading':    nagiosplugin.Ok,
            'activating':   nagiosplugin.Ok,
            'deactivating': nagiosplugin.Warn,
            'inactive':     nagiosplugin.Critical,
            'failed':       nagiosplugin.Critical,
        })
        return self.result_cls(resultD[metric.value[0]], metric=metric)


@nagiosplugin.guarded
def main():
    argp = argparse.ArgumentParser(description=__doc__,
                      formatter_class=argparse.RawTextHelpFormatter,
           )
    argp.add_argument('-v', '--verbose', action='count', default=0,
                      help='increase output verbosity (use up to 3 times)')
    argp.add_argument('-t', '--timeout', default=10,
                      help='abort execution after TIMEOUT seconds')
    argp.add_argument('unit', help='Check this Unit')
    args = argp.parse_args()
    _log.debug('Found arguments %r', args)
    check = nagiosplugin.Check(
        Systemd_Service(**vars(args)),
        Service_Context('service_state', fmt_metric="ServiceState is {value[0]}({value[1]})"),
    )
    check.main(args.verbose, args.timeout)


if __name__ == '__main__':
    main()