Merge lp:~brad-marshall/charms/trusty/heat/add-nrpe-checks into lp:~openstack-charmers-archive/charms/trusty/heat/trunk

[Brad Marshall]

Adds nrpe-external-master interface and adds basic nrpe checks.

[Ryan Beisner (1chb1n)]

UOSCI bot says:
charm_amulet_test #368 trusty-heat for brad-marshall mp241493
    AMULET FAIL: amulet-test missing

AMULET Results (max last 5 lines):
INFO:root:Workspace dir: /var/lib/jenkins/workspace/charm_amulet_test
INFO:root:Reading file: Makefile
INFO:root:Searching for: ['@juju test']
INFO:root:Search string not found in makefile target commands.
ERROR:root:No make target was executed.

Full amulet test output: http://paste.ubuntu.com/8955708/
Build: http://10.98.191.181:8080/job/charm_amulet_test/368/

[Ryan Beisner (1chb1n)]

UOSCI bot says:
charm_lint_check #988 trusty-heat for brad-marshall mp241493
    LINT FAIL: lint-test failed

LINT Results (max last 5 lines):
ERROR:root:Make target returned non-zero.
  Running flake8 tests: hooks/heat_relations.py:163:80: E501 line too long (92 > 79 characters)
  hooks/heat_relations.py:190:22: E251 unexpected spaces around keyword / parameter equals
  hooks/heat_relations.py:190:24: E251 unexpected spaces around keyword / parameter equals
  make: *** [lint] Error 1

Full lint test output: http://paste.ubuntu.com/8955709/
Build: http://10.98.191.181:8080/job/charm_lint_check/988/

[Ryan Beisner (1chb1n)]

UOSCI bot says:
charm_unit_test #823 trusty-heat for brad-marshall mp241493
    UNIT FAIL: unit-test failed

UNIT Results (max last 5 lines):
  heat_utils 46 6 87% 70-78
  TOTAL 170 42 75%
  Ran 21 tests in 0.504s
  FAILED (errors=2)
  make: *** [test] Error 1

Full unit test output: http://paste.ubuntu.com/8955710/
Build: http://10.98.191.181:8080/job/charm_unit_test/823/

[Liam Young (gnuoy)]

Thanks again for the sorely needed nagios checks. Just a few comments...

Please steal the services() method from nova-cloud-controller and add it to hooks/heat_utils.py and use that to get a list services rather than defining a new list in update_nagios().

Please can the check_upstart_job go into charmhelpers

There appears to be some lint errors:
Running flake8 tests: hooks/heat_relations.py:163:80: E501 line too long (92 > 79 characters)
hooks/heat_relations.py:190:22: E251 unexpected spaces around keyword / parameter equals
hooks/heat_relations.py:190:24: E251 unexpected spaces around keyword / parameter equals
Makefile:5: recipe for target 'lint' failed
make: *** [lint] Error 1

review: Needs Fixing

[lp:~brad-marshall/charms/trusty/heat/add-nrpe-checks]

32. By Brad Marshall on 2014-11-17

[bradm] Fixes from pep8, used services() rather than hard coded list

[uosci-testing-bot (uosci-testing-bot)]

UOSCI bot says:
charm_lint_check #1091 trusty-heat for brad-marshall mp241493
    LINT OK: passed

LINT Results (max last 5 lines):
INFO:root:Searching for: ['@flake8']
INFO:root:command: make -f Makefile lint
  Running flake8 tests: OK
  Running charm proof: W: metadata name (heat) must match directory name (trusty-heat) exactly for local deployment.
  OK

Full lint test output: http://paste.ubuntu.com/9052189/
Build: http://10.98.191.181:8080/job/charm_lint_check/1091/

[uosci-testing-bot (uosci-testing-bot)]

UOSCI bot says:
charm_unit_test #925 trusty-heat for brad-marshall mp241493
    UNIT FAIL: unit-test failed

UNIT Results (max last 5 lines):
  heat_utils 51 10 80% 70-78, 141-144
  TOTAL 177 49 72%
  Ran 21 tests in 0.492s
  FAILED (errors=2)
  make: *** [test] Error 1

Full unit test output: http://paste.ubuntu.com/9052192/
Build: http://10.98.191.181:8080/job/charm_unit_test/925/

[uosci-testing-bot (uosci-testing-bot)]

UOSCI bot says:
charm_amulet_test #433 trusty-heat for brad-marshall mp241493
    AMULET FAIL: amulet-test missing

AMULET Results (max last 5 lines):
INFO:root:Workspace dir: /var/lib/jenkins/workspace/charm_amulet_test
INFO:root:Reading file: Makefile
INFO:root:Searching for: ['@juju test']
INFO:root:Search string not found in makefile target commands.
ERROR:root:No make target was executed.

Full amulet test output: http://paste.ubuntu.com/9052239/
Build: http://10.98.191.181:8080/job/charm_amulet_test/433/

[lp:~brad-marshall/charms/trusty/heat/add-nrpe-checks]

33. By Brad Marshall on 2014-11-18

[bradm] Removed nagios check files as they're now in n-e-m charm, added support for sysvinit daemon monitoring.

Unmerged revisions

33. By Brad Marshall on 2014-11-18

[bradm] Removed nagios check files as they're now in n-e-m charm, added support for sysvinit daemon monitoring.

32. By Brad Marshall on 2014-11-17

[bradm] Fixes from pep8, used services() rather than hard coded list

31. By Brad Marshall on 2014-11-06

[bradm] Check if host_context is defined before using it, checking if files before installing

30. By Brad Marshall on 2014-11-04

[bradm] Tweaked check to include host context and unit name

29. By Brad Marshall on 2014-10-31

[bradm] Added support to get nagios hostname from nrpe relation

28. By Brad Marshall on 2014-10-30

[bradm] Added charmsupport to charmhelpers

27. By root <email address hidden> on 2014-10-30

[bradm] initial nrpe checks

=== modified file 'charm-helpers.yaml'
--- charm-helpers.yaml	2014-08-13 13:45:18 +0000
+++ charm-helpers.yaml	2014-11-17 03:39:18 +0000
@@ -10,3 +10,4 @@
         - apache
         - cluster
     - payload.execd
+    - contrib.charmsupport

=== modified file 'config.yaml'
--- config.yaml	2014-04-08 20:05:44 +0000
+++ config.yaml	2014-11-17 03:39:18 +0000
@@ -44,3 +44,13 @@
     description: |
       By default, all services will log into their corresponding log files.
       Setting this to True will force all services to log to the syslog.
+  nagios_context:
+    default: "juju"
+    type: string
+    description: |
+      Used by the nrpe-external-master subordinate charm.
+      A string that will be prepended to instance name to set the host name
+      in nagios. So for instance the hostname would be something like:
+          juju-myservice-0
+      If you're running multiple environments with the same services in them
+      this allows you to differentiate between them.

=== added directory 'files'
=== added directory 'files/nrpe-external-master'
=== added file 'files/nrpe-external-master/check_upstart_job'
--- files/nrpe-external-master/check_upstart_job	1970-01-01 00:00:00 +0000
+++ files/nrpe-external-master/check_upstart_job	2014-11-17 03:39:18 +0000
@@ -0,0 +1,72 @@
+#!/usr/bin/python
+
+#
+# Copyright 2012, 2013 Canonical Ltd.
+#
+# Author: Paul Collins <[e-mail]>
+#
+# Based on http://www.eurion.net/python-snippets/snippet/Upstart%20service%20status.html
+#
+
+import sys
+
+import dbus
+
+
+class Upstart(object):
+    def __init__(self):
+        self._bus = dbus.SystemBus()
+        self._upstart = self._bus.get_object('com.ubuntu.Upstart',
+                                             '/com/ubuntu/Upstart')
+    def get_job(self, job_name):
+        path = self._upstart.GetJobByName(job_name,
+                                          dbus_interface='com.ubuntu.Upstart0_6')
+        return self._bus.get_object('com.ubuntu.Upstart', path)
+
+    def get_properties(self, job):
+        path = job.GetInstance([], dbus_interface='com.ubuntu.Upstart0_6.Job')
+        instance = self._bus.get_object('com.ubuntu.Upstart', path)
+        return instance.GetAll('com.ubuntu.Upstart0_6.Instance',
+                               dbus_interface=dbus.PROPERTIES_IFACE)
+
+    def get_job_instances(self, job_name):
+        job = self.get_job(job_name)
+        paths = job.GetAllInstances([], dbus_interface='com.ubuntu.Upstart0_6.Job')
+        return [self._bus.get_object('com.ubuntu.Upstart', path) for path in paths]
+
+    def get_job_instance_properties(self, job):
+        return job.GetAll('com.ubuntu.Upstart0_6.Instance',
+                          dbus_interface=dbus.PROPERTIES_IFACE)
+
+try:
+    upstart = Upstart()
+    try:
+        job = upstart.get_job(sys.argv[1])        
+        props = upstart.get_properties(job)
+
+        if props['state'] == 'running':
+            print 'OK: %s is running' % sys.argv[1]
+            sys.exit(0)
+        else:
+            print 'CRITICAL: %s is not running' % sys.argv[1]
+            sys.exit(2)
+
+    except dbus.DBusException as e:
+        instances = upstart.get_job_instances(sys.argv[1])
+        propses = [upstart.get_job_instance_properties(instance) for instance in instances]
+        states = dict([(props['name'], props['state']) for props in propses])
+        if len(states) != states.values().count('running'):
+            not_running = []
+            for name in states.keys():
+                if states[name] != 'running':
+                    not_running.append(name)
+            print 'CRITICAL: %d instances of %s not running: %s' % \
+                (len(not_running), sys.argv[1], not_running.join(', '))
+            sys.exit(2)
+        else:
+            print 'OK: %d instances of %s running' % (len(states), sys.argv[1])
+
+except dbus.DBusException as e:
+    print 'CRITICAL: failed to get properties of \'%s\' from upstart' % sys.argv[1]
+    sys.exit(2)
+

=== added directory 'hooks/charmhelpers/contrib/charmsupport'
=== added file 'hooks/charmhelpers/contrib/charmsupport/__init__.py'
=== added file 'hooks/charmhelpers/contrib/charmsupport/nrpe.py'
--- hooks/charmhelpers/contrib/charmsupport/nrpe.py	1970-01-01 00:00:00 +0000
+++ hooks/charmhelpers/contrib/charmsupport/nrpe.py	2014-11-17 03:39:18 +0000
@@ -0,0 +1,222 @@
+"""Compatibility with the nrpe-external-master charm"""
+# Copyright 2012 Canonical Ltd.
+#
+# Authors:
+#  Matthew Wedgwood <[e-mail]>
+
+import subprocess
+import pwd
+import grp
+import os
+import re
+import shlex
+import yaml
+
+from charmhelpers.core.hookenv import (
+    config,
+    local_unit,
+    log,
+    relation_ids,
+    relation_set,
+)
+
+from charmhelpers.core.host import service
+
+# This module adds compatibility with the nrpe-external-master and plain nrpe
+# subordinate charms. To use it in your charm:
+#
+# 1. Update metadata.yaml
+#
+#   provides:
+#     (...)
+#     nrpe-external-master:
+#       interface: nrpe-external-master
+#       scope: container
+#
+#   and/or
+#
+#   provides:
+#     (...)
+#     local-monitors:
+#       interface: local-monitors
+#       scope: container
+
+#
+# 2. Add the following to config.yaml
+#
+#    nagios_context:
+#      default: "juju"
+#      type: string
+#      description: |
+#        Used by the nrpe subordinate charms.
+#        A string that will be prepended to instance name to set the host name
+#        in nagios. So for instance the hostname would be something like:
+#            juju-myservice-0
+#        If you're running multiple environments with the same services in them
+#        this allows you to differentiate between them.
+#
+# 3. Add custom checks (Nagios plugins) to files/nrpe-external-master
+#
+# 4. Update your hooks.py with something like this:
+#
+#    from charmsupport.nrpe import NRPE
+#    (...)
+#    def update_nrpe_config():
+#        nrpe_compat = NRPE()
+#        nrpe_compat.add_check(
+#            shortname = "myservice",
+#            description = "Check MyService",
+#            check_cmd = "check_http -w 2 -c 10 http://localhost"
+#            )
+#        nrpe_compat.add_check(
+#            "myservice_other",
+#            "Check for widget failures",
+#            check_cmd = "/srv/myapp/scripts/widget_check"
+#            )
+#        nrpe_compat.write()
+#
+#    def config_changed():
+#        (...)
+#        update_nrpe_config()
+#
+#    def nrpe_external_master_relation_changed():
+#        update_nrpe_config()
+#
+#    def local_monitors_relation_changed():
+#        update_nrpe_config()
+#
+# 5. ln -s hooks.py nrpe-external-master-relation-changed
+#    ln -s hooks.py local-monitors-relation-changed
+
+
+class CheckException(Exception):
+    pass
+
+
+class Check(object):
+    shortname_re = '[A-Za-z0-9-_]+$'
+    service_template = ("""
+#---------------------------------------------------
+# This file is Juju managed
+#---------------------------------------------------
+define service {{
+    use                             active-service
+    host_name                       {nagios_hostname}
+    service_description             {nagios_hostname}[{shortname}] """
+                        """{description}
+    check_command                   check_nrpe!{command}
+    servicegroups                   {nagios_servicegroup}
+}}
+""")
+
+    def __init__(self, shortname, description, check_cmd):
+        super(Check, self).__init__()
+        # XXX: could be better to calculate this from the service name
+        if not re.match(self.shortname_re, shortname):
+            raise CheckException("shortname must match {}".format(
+                Check.shortname_re))
+        self.shortname = shortname
+        self.command = "check_{}".format(shortname)
+        # Note: a set of invalid characters is defined by the
+        # Nagios server config
+        # The default is: illegal_object_name_chars=`~!$%^&*"|'<>?,()=
+        self.description = description
+        self.check_cmd = self._locate_cmd(check_cmd)
+
+    def _locate_cmd(self, check_cmd):
+        search_path = (
+            '/',
+            os.path.join(os.environ['CHARM_DIR'],
+                         'files/nrpe-external-master'),
+            '/usr/lib/nagios/plugins',
+            '/usr/local/lib/nagios/plugins',
+        )
+        parts = shlex.split(check_cmd)
+        for path in search_path:
+            if os.path.exists(os.path.join(path, parts[0])):
+                command = os.path.join(path, parts[0])
+                if len(parts) > 1:
+                    command += " " + " ".join(parts[1:])
+                return command
+        log('Check command not found: {}'.format(parts[0]))
+        return ''
+
+    def write(self, nagios_context, hostname):
+        nrpe_check_file = '/etc/nagios/nrpe.d/{}.cfg'.format(
+            self.command)
+        with open(nrpe_check_file, 'w') as nrpe_check_config:
+            nrpe_check_config.write("# check {}\n".format(self.shortname))
+            nrpe_check_config.write("command[{}]={}\n".format(
+                self.command, self.check_cmd))
+
+        if not os.path.exists(NRPE.nagios_exportdir):
+            log('Not writing service config as {} is not accessible'.format(
+                NRPE.nagios_exportdir))
+        else:
+            self.write_service_config(nagios_context, hostname)
+
+    def write_service_config(self, nagios_context, hostname):
+        for f in os.listdir(NRPE.nagios_exportdir):
+            if re.search('.*{}.cfg'.format(self.command), f):
+                os.remove(os.path.join(NRPE.nagios_exportdir, f))
+
+        templ_vars = {
+            'nagios_hostname': hostname,
+            'nagios_servicegroup': nagios_context,
+            'description': self.description,
+            'shortname': self.shortname,
+            'command': self.command,
+        }
+        nrpe_service_text = Check.service_template.format(**templ_vars)
+        nrpe_service_file = '{}/service__{}_{}.cfg'.format(
+            NRPE.nagios_exportdir, hostname, self.command)
+        with open(nrpe_service_file, 'w') as nrpe_service_config:
+            nrpe_service_config.write(str(nrpe_service_text))
+
+    def run(self):
+        subprocess.call(self.check_cmd)
+
+
+class NRPE(object):
+    nagios_logdir = '/var/log/nagios'
+    nagios_exportdir = '/var/lib/nagios/export'
+    nrpe_confdir = '/etc/nagios/nrpe.d'
+
+    def __init__(self, hostname=None):
+        super(NRPE, self).__init__()
+        self.config = config()
+        self.nagios_context = self.config['nagios_context']
+        self.unit_name = local_unit().replace('/', '-')
+        if hostname:
+            self.hostname = hostname
+        else:
+            self.hostname = "{}-{}".format(self.nagios_context, self.unit_name)
+        self.checks = []
+
+    def add_check(self, *args, **kwargs):
+        self.checks.append(Check(*args, **kwargs))
+
+    def write(self):
+        try:
+            nagios_uid = pwd.getpwnam('nagios').pw_uid
+            nagios_gid = grp.getgrnam('nagios').gr_gid
+        except:
+            log("Nagios user not set up, nrpe checks not updated")
+            return
+
+        if not os.path.exists(NRPE.nagios_logdir):
+            os.mkdir(NRPE.nagios_logdir)
+            os.chown(NRPE.nagios_logdir, nagios_uid, nagios_gid)
+
+        nrpe_monitors = {}
+        monitors = {"monitors": {"remote": {"nrpe": nrpe_monitors}}}
+        for nrpecheck in self.checks:
+            nrpecheck.write(self.nagios_context, self.hostname)
+            nrpe_monitors[nrpecheck.shortname] = {
+                "command": nrpecheck.command,
+            }
+
+        service('restart', 'nagios-nrpe-server')
+
+        for rid in relation_ids("local-monitors"):
+            relation_set(relation_id=rid, monitors=yaml.dump(monitors))

=== added file 'hooks/charmhelpers/contrib/charmsupport/volumes.py'
--- hooks/charmhelpers/contrib/charmsupport/volumes.py	1970-01-01 00:00:00 +0000
+++ hooks/charmhelpers/contrib/charmsupport/volumes.py	2014-11-17 03:39:18 +0000
@@ -0,0 +1,156 @@
+'''
+Functions for managing volumes in juju units. One volume is supported per unit.
+Subordinates may have their own storage, provided it is on its own partition.
+
+Configuration stanzas:
+  volume-ephemeral:
+    type: boolean
+    default: true
+    description: >
+      If false, a volume is mounted as sepecified in "volume-map"
+      If true, ephemeral storage will be used, meaning that log data
+         will only exist as long as the machine. YOU HAVE BEEN WARNED.
+  volume-map:
+    type: string
+    default: {}
+    description: >
+      YAML map of units to device names, e.g:
+        "{ rsyslog/0: /dev/vdb, rsyslog/1: /dev/vdb }"
+      Service units will raise a configure-error if volume-ephemeral
+      is 'true' and no volume-map value is set. Use 'juju set' to set a
+      value and 'juju resolved' to complete configuration.
+
+Usage:
+    from charmsupport.volumes import configure_volume, VolumeConfigurationError
+    from charmsupport.hookenv import log, ERROR
+    def post_mount_hook():
+        stop_service('myservice')
+    def post_mount_hook():
+        start_service('myservice')
+
+    if __name__ == '__main__':
+        try:
+            configure_volume(before_change=pre_mount_hook,
+                             after_change=post_mount_hook)
+        except VolumeConfigurationError:
+            log('Storage could not be configured', ERROR)
+'''
+
+# XXX: Known limitations
+# - fstab is neither consulted nor updated
+
+import os
+from charmhelpers.core import hookenv
+from charmhelpers.core import host
+import yaml
+
+
+MOUNT_BASE = '/srv/juju/volumes'
+
+
+class VolumeConfigurationError(Exception):
+    '''Volume configuration data is missing or invalid'''
+    pass
+
+
+def get_config():
+    '''Gather and sanity-check volume configuration data'''
+    volume_config = {}
+    config = hookenv.config()
+
+    errors = False
+
+    if config.get('volume-ephemeral') in (True, 'True', 'true', 'Yes', 'yes'):
+        volume_config['ephemeral'] = True
+    else:
+        volume_config['ephemeral'] = False
+
+    try:
+        volume_map = yaml.safe_load(config.get('volume-map', '{}'))
+    except yaml.YAMLError as e:
+        hookenv.log("Error parsing YAML volume-map: {}".format(e),
+                    hookenv.ERROR)
+        errors = True
+    if volume_map is None:
+        # probably an empty string
+        volume_map = {}
+    elif not isinstance(volume_map, dict):
+        hookenv.log("Volume-map should be a dictionary, not {}".format(
+            type(volume_map)))
+        errors = True
+
+    volume_config['device'] = volume_map.get(os.environ['JUJU_UNIT_NAME'])
+    if volume_config['device'] and volume_config['ephemeral']:
+        # asked for ephemeral storage but also defined a volume ID
+        hookenv.log('A volume is defined for this unit, but ephemeral '
+                    'storage was requested', hookenv.ERROR)
+        errors = True
+    elif not volume_config['device'] and not volume_config['ephemeral']:
+        # asked for permanent storage but did not define volume ID
+        hookenv.log('Ephemeral storage was requested, but there is no volume '
+                    'defined for this unit.', hookenv.ERROR)
+        errors = True
+
+    unit_mount_name = hookenv.local_unit().replace('/', '-')
+    volume_config['mountpoint'] = os.path.join(MOUNT_BASE, unit_mount_name)
+
+    if errors:
+        return None
+    return volume_config
+
+
+def mount_volume(config):
+    if os.path.exists(config['mountpoint']):
+        if not os.path.isdir(config['mountpoint']):
+            hookenv.log('Not a directory: {}'.format(config['mountpoint']))
+            raise VolumeConfigurationError()
+    else:
+        host.mkdir(config['mountpoint'])
+    if os.path.ismount(config['mountpoint']):
+        unmount_volume(config)
+    if not host.mount(config['device'], config['mountpoint'], persist=True):
+        raise VolumeConfigurationError()
+
+
+def unmount_volume(config):
+    if os.path.ismount(config['mountpoint']):
+        if not host.umount(config['mountpoint'], persist=True):
+            raise VolumeConfigurationError()
+
+
+def managed_mounts():
+    '''List of all mounted managed volumes'''
+    return filter(lambda mount: mount[0].startswith(MOUNT_BASE), host.mounts())
+
+
+def configure_volume(before_change=lambda: None, after_change=lambda: None):
+    '''Set up storage (or don't) according to the charm's volume configuration.
+       Returns the mount point or "ephemeral". before_change and after_change
+       are optional functions to be called if the volume configuration changes.
+    '''
+
+    config = get_config()
+    if not config:
+        hookenv.log('Failed to read volume configuration', hookenv.CRITICAL)
+        raise VolumeConfigurationError()
+
+    if config['ephemeral']:
+        if os.path.ismount(config['mountpoint']):
+            before_change()
+            unmount_volume(config)
+            after_change()
+        return 'ephemeral'
+    else:
+        # persistent storage
+        if os.path.ismount(config['mountpoint']):
+            mounts = dict(managed_mounts())
+            if mounts.get(config['mountpoint']) != config['device']:
+                before_change()
+                unmount_volume(config)
+                mount_volume(config)
+                after_change()
+        else:
+            before_change()
+            mount_volume(config)
+            after_change()
+        return config['mountpoint']

=== modified file 'hooks/heat_relations.py'
--- hooks/heat_relations.py	2014-04-12 20:48:50 +0000
+++ hooks/heat_relations.py	2014-11-17 03:39:18 +0000
@@ -20,6 +20,8 @@
     charm_dir,
     log,
     relation_set,
+    relations_of_type,
+    local_unit,
     open_port,
     unit_get
 )
@@ -45,6 +47,7 @@
 from heat_utils import (
     do_openstack_upgrade,
     restart_map,
+    services,
     determine_packages,
     register_configs,
     HEAT_CONF,
@@ -54,6 +57,8 @@
 
 from charmhelpers.payload.execd import execd_preinstall
 
+from charmhelpers.contrib.charmsupport.nrpe import NRPE
+
 hooks = Hooks()
 CONFIGS = register_configs()
 
@@ -69,8 +74,9 @@
     if os.path.isdir(_files):
         for f in os.listdir(_files):
             f = os.path.join(_files, f)
-            log('Installing %s to /usr/bin' % f)
-            shutil.copy2(f, '/usr/bin')
+            if os.path.isfile(f):
+                log('Installing %s to /usr/bin' % f)
+                shutil.copy2(f, '/usr/bin')
 
     for port in API_PORTS.values():
         open_port(port)
@@ -81,6 +87,7 @@
 def config_changed():
     if openstack_upgrade_available('heat-engine'):
         do_openstack_upgrade(CONFIGS)
+    update_nrpe_config()
     CONFIGS.write_all()
 
 
@@ -154,6 +161,39 @@
     CONFIGS.write_all()
 
 
[e-mail]('nrpe-external-master-relation-joined',
+            'nrpe-external-master-relation-changed')
+def update_nrpe_config():
+    # Find out if nrpe set nagios_hostname
+    hostname = None
+    host_context = None
+    for rel in relations_of_type('nrpe-external-master'):
+        if 'nagios_hostname' in rel:
+            hostname = rel['nagios_hostname']
+            host_context = rel['nagios_host_context']
+            break
+    nrpe = NRPE(hostname=hostname)
+    apt_install('python-dbus')
+
+    if host_context:
+        current_unit = "%s:%s" % (host_context, local_unit())
+    else:
+        current_unit = local_unit()
+
+    services_to_monitor = services()
+
+    for service in services_to_monitor:
+        upstart_init = '/etc/init/%s.conf' % service
+        if os.path.exists(upstart_init):
+            nrpe.add_check(
+                shortname=service,
+                description='process check {%s}' % current_unit,
+                check_cmd='check_upstart_job %s' % service,
+                )
+
+    nrpe.write()
+
+
 def main():
     try:
         hooks.execute(sys.argv)

=== modified file 'hooks/heat_utils.py'
--- hooks/heat_utils.py	2014-04-12 21:40:16 +0000
+++ hooks/heat_utils.py	2014-11-17 03:39:18 +0000
@@ -134,3 +134,11 @@
         if svcs:
             _map.append((f, svcs))
     return OrderedDict(_map)
+
+
+def services():
+    ''' Returns a list of services associate with this charm '''
+    _services = []
+    for v in restart_map().values():
+        _services = _services + v
+    return list(set(_services))

=== added symlink 'hooks/nrpe-external-master-relation-changed'
=== target is u'heat_relations.py'
=== added symlink 'hooks/nrpe-external-master-relation-joined'
=== target is u'heat_relations.py'
=== modified file 'metadata.yaml'
--- metadata.yaml	2014-04-16 08:19:26 +0000
+++ metadata.yaml	2014-11-17 03:39:18 +0000
@@ -5,6 +5,10 @@
   Heat is the main project in the OpenStack Orchestration program. It implements an
   orchestration engine to launch multiple composite cloud applications based on
   templates in the form of text files that can be treated like code.
+provides:
+  nrpe-external-master:
+    interface: nrpe-external-master
+    scope: container
 categories:
   - openstack
 requires: