def update_rally_checkfiles(self):
if not self.is_rally_enabled:
return
# Copy run_rally.sh to /usr/local/bin
rally_script = os.path.join(hookenv.charm_dir(), 'files',
'run_rally.py')
host.rsync(rally_script, self.scripts_dir, options=['--executability'])
ostestsfile = os.path.join('/home', self._rallyuser, 'ostests.txt')
render(source='ostests.txt.j2',
target=ostestsfile,
context=self._get_rally_checks_context(),
owner=self._rallyuser,
group=self._rallyuser)
proxy_settings = hookenv.env_proxy_settings()
if proxy_settings:
content = '\n'.join([
'{}={}'.format(proxy_var, proxy_var_val)
for proxy_var, proxy_var_val in proxy_settings.items()
])
else:
content = ''
context = {
'schedule': self.rally_cron_schedule,
'user': self._rallyuser,
'cmd': os.path.join(self.scripts_dir, 'run_rally.py'),
}
content += '\n#\n{schedule} {user} timeout -k 840s -s SIGTERM 780s {cmd}'.format(
**context)
with open(self.rally_cron_file, 'w') as fd:
fd.write('# Juju generated - DO NOT EDIT\n{}\n\n'.format(content))
def migrate_to_mount(new_path):
"""Invoked when new mountpoint appears. This function safely migrates
MySQL data from local disk to persistent storage (only if needed)
"""
old_path = '/var/lib/mysql'
if os.path.islink(old_path):
hookenv.log('{} is already a symlink, skipping migration'.format(
old_path))
return True
# Ensure our new mountpoint is empty. Otherwise error and allow
# users to investigate and migrate manually
files = os.listdir(new_path)
try:
files.remove('lost+found')
except ValueError:
pass
if files:
raise RuntimeError('Persistent storage contains old data. '
'Please investigate and migrate data manually '
'to: {}'.format(new_path))
os.chmod(new_path, 0o700)
if os.path.isdir('/etc/apparmor.d/local'):
render('apparmor.j2', '/etc/apparmor.d/local/usr.sbin.mysqld',
context={'path': os.path.join(new_path, '')})
host.service_reload('apparmor')
host.service_stop('mysql')
host.rsync(os.path.join(old_path, ''), # Ensure we have trailing slashes
os.path.join(new_path, ''),
options=['--archive'])
shutil.rmtree(old_path)
os.symlink(new_path, old_path)
host.service_start('mysql')
def install_hugepages():
""" Configure hugepages """
hugepage_config = config('hugepages')
if hugepage_config:
# TODO: defaults to 2M - this should probably be configurable
# and support multiple pool sizes - e.g. 2M and 1G.
hugepage_size = 2048
if hugepage_config.endswith('%'):
import psutil
mem = psutil.virtual_memory()
hugepage_config_pct = hugepage_config.strip('%')
hugepage_multiplier = float(hugepage_config_pct) / 100
hugepages = int((mem.total * hugepage_multiplier) / hugepage_size)
else:
hugepages = int(hugepage_config)
mnt_point = '/run/hugepages/kvm'
hugepage_support(
'nova',
mnt_point=mnt_point,
group='root',
nr_hugepages=hugepages,
mount=False,
set_shmmax=True,
)
if subprocess.call(['mountpoint', mnt_point]):
fstab_mount(mnt_point)
rsync(
charm_dir() + '/files/qemu-hugefsdir',
'/etc/init.d/qemu-hugefsdir'
)
subprocess.check_call('/etc/init.d/qemu-hugefsdir')
subprocess.check_call(['update-rc.d', 'qemu-hugefsdir', 'defaults'])
def render_checks(self, creds):
render(source='keystone.yaml',
target=self.oscreds,
context=creds,
owner='nagios',
group='nagios')
nrpe = NRPE()
if not os.path.exists(self.plugins_dir):
os.makedirs(self.plugins_dir)
charm_plugin_dir = os.path.join(hookenv.charm_dir(), 'files',
'plugins/')
host.rsync(charm_plugin_dir,
self.plugins_dir,
options=['--executability'])
contrail_check_command = os.path.join(self.plugins_dir,
'check_contrail_alarms.py')
nrpe.add_check(
shortname='contrail_alarms',
description='Check Contrail alarms',
check_cmd=contrail_check_command,
)
nrpe.write()
def update_nrpe_config():
# python-dbus is used by check_upstart_job
apt_install(['python-dbus', 'lockfile-progs'])
log('Refreshing nagios checks')
if os.path.isdir(NAGIOS_PLUGINS):
rsync(
os.path.join(os.getenv('CHARM_DIR'), 'files', 'nagios',
'check_ceph_status.py'),
os.path.join(NAGIOS_PLUGINS, 'check_ceph_status.py'))
script = os.path.join(SCRIPTS_DIR, 'collect_ceph_status.sh')
rsync(
os.path.join(os.getenv('CHARM_DIR'), 'files', 'nagios',
'collect_ceph_status.sh'), script)
cronjob = "{} root {}\n".format('*/5 * * * *', script)
write_file(STATUS_CRONFILE, cronjob)
# Find out if nrpe set nagios_hostname
hostname = nrpe.get_nagios_hostname()
current_unit = nrpe.get_nagios_unit_name()
nrpe_setup = nrpe.NRPE(hostname=hostname)
check_cmd = 'check_ceph_status.py -f {} --degraded_thresh {}' \
' --misplaced_thresh {}' \
' --recovery_rate {}'.format(STATUS_FILE,
config('nagios_degraded_thresh'),
config('nagios_misplaced_thresh'),
config('nagios_recovery_rate'))
if config('nagios_ignore_nodeepscub'):
check_cmd = check_cmd + ' --ignore_nodeepscrub'
nrpe_setup.add_check(
shortname="ceph",
description='Check Ceph health {{{}}}'.format(current_unit),
check_cmd=check_cmd)
nrpe_setup.write()
def install():
juju_log('**********install.real')
rsync(
charm_dir() + '/packages/vsm-dep-repo',
'/opt'
)
rsync(
charm_dir() + '/packages/vsmrepo',
'/opt'
)
rsync(
charm_dir() + '/files/apt.conf',
'/etc/apt'
)
rsync(
charm_dir() + '/files/vsm.list',
'/etc/apt/sources.list.d'
)
rsync(
charm_dir() + '/files/vsm-dep.list',
'/etc/apt/sources.list.d'
)
apt_update()
apt_install(VSM_PACKAGES)
juju_log('**********finished to install vsm vsm-dashboard python-vsmclient')
add_source(config('ceph-source'), config('ceph-key'))
apt_update(fatal=True)
apt_install(packages=PRE_INSTALL_PACKAGES, fatal=True)
def update_nrpe_config():
# python-dbus is used by check_upstart_job
apt_install('python-dbus')
log('Refreshing nagios checks')
if os.path.isdir(NAGIOS_PLUGINS):
rsync(
os.path.join(os.getenv('CHARM_DIR'), 'files', 'nagios',
'check_ceph_status.py'),
os.path.join(NAGIOS_PLUGINS, 'check_ceph_status.py'))
script = os.path.join(SCRIPTS_DIR, 'collect_ceph_status.sh')
rsync(
os.path.join(os.getenv('CHARM_DIR'), 'files', 'nagios',
'collect_ceph_status.sh'), script)
cronjob = "{} root {}\n".format('*/5 * * * *', script)
write_file(STATUS_CRONFILE, cronjob)
# Find out if nrpe set nagios_hostname
hostname = nrpe.get_nagios_hostname()
current_unit = nrpe.get_nagios_unit_name()
nrpe_setup = nrpe.NRPE(hostname=hostname)
nrpe_setup.add_check(
shortname="ceph",
description='Check Ceph health {%s}' % current_unit,
check_cmd='check_ceph_status.py -f {}'.format(STATUS_FILE))
nrpe_setup.write()
def migrate(self, src_dir, subdir):
assert self.needs_remount()
assert subdir, 'Can only migrate to a subdirectory on a mount'
config = hookenv.config()
config['live_mountpoint'] = self.mountpoint
if self.mountpoint is None:
hookenv.log('External storage AND DATA gone.'
'Reverting to original local storage', WARNING)
return
dst_dir = os.path.join(self.mountpoint, subdir)
if os.path.exists(dst_dir):
hookenv.log('{} already exists. Not migrating data.'.format(
dst_dir))
return
# We are migrating the contents of src_dir, so we want a
# trailing slash to ensure rsync's behavior.
if not src_dir.endswith('/'):
src_dir += '/'
# We don't migrate data directly into the new destination,
# which allows us to detect a failed migration and recover.
tmp_dst_dir = dst_dir + '.migrating'
hookenv.log('Migrating data from {} to {}'.format(
src_dir, tmp_dst_dir))
host.rsync(src_dir, tmp_dst_dir, flags='-av')
hookenv.log('Moving {} to {}'.format(tmp_dst_dir, dst_dir))
os.rename(tmp_dst_dir, dst_dir)
assert not self.needs_remount()
def install_ntpmon():
"""
Install package dependencies, source files, and startup configuration.
"""
install_dir = layer.options.get('ntpmon', 'install-dir')
service_name = layer.options.get('ntpmon', 'service-name')
using_systemd = host.init_is_systemd()
if install_dir:
log('installing ntpmon')
host.mkdir(os.path.dirname(install_dir))
host.rsync('src/', '{}/'.format(install_dir))
if service_name:
if using_systemd:
systemd_config = '/etc/systemd/system/' + service_name + '.service'
log('installing systemd service: {}'.format(service_name))
with open(systemd_config, 'w') as conffile:
conffile.write(templating.render('src/' + service_name + '.systemd', layer.options.get('ntpmon')))
subprocess.call(['systemd', 'daemon-reload'])
else:
upstart_config = '/etc/init/' + service_name + '.conf'
log('installing upstart service: {}'.format(service_name))
with open(upstart_config, 'w') as conffile:
conffile.write(templating.render('src/' + service_name + '.upstart', layer.options.get('ntpmon')))
set_flag('ntpmon.installed')
clear_flag('ntpmon.configured')
def update_nrpe_config():
# python-dbus is used by check_upstart_job
apt_install('python-dbus')
log('Refreshing nagios checks')
if os.path.isdir(NAGIOS_PLUGINS):
rsync(os.path.join(os.getenv('CHARM_DIR'), 'files', 'nagios',
'check_ceph_status.py'),
os.path.join(NAGIOS_PLUGINS, 'check_ceph_status.py'))
script = os.path.join(SCRIPTS_DIR, 'collect_ceph_status.sh')
rsync(os.path.join(os.getenv('CHARM_DIR'), 'files',
'nagios', 'collect_ceph_status.sh'),
script)
cronjob = "{} root {}\n".format('*/5 * * * *', script)
write_file(STATUS_CRONFILE, cronjob)
# Find out if nrpe set nagios_hostname
hostname = nrpe.get_nagios_hostname()
current_unit = nrpe.get_nagios_unit_name()
nrpe_setup = nrpe.NRPE(hostname=hostname)
nrpe_setup.add_check(
shortname="ceph",
description='Check Ceph health {%s}' % current_unit,
check_cmd='check_ceph_status.py -f {}'.format(STATUS_FILE)
)
nrpe_setup.write()
def install_charm_files(service_name):
""" Install files shipped with charm """
nag_dirs = [
'/etc/nagios/nrpe.d/', '/usr/local/lib/nagios/plugins',
'/var/lib/nagios/export/'
]
for nag_dir in nag_dirs:
if not os.path.exists(nag_dir):
host.mkdir(nag_dir, perms=0o755)
charm_file_dir = os.path.join(hookenv.charm_dir(), 'files')
charm_plugin_dir = os.path.join(charm_file_dir, 'plugins')
pkg_plugin_dir = '/usr/lib/nagios/plugins/'
local_plugin_dir = '/usr/local/lib/nagios/plugins/'
shutil.copy2(os.path.join(charm_file_dir, 'nagios_plugin.py'),
pkg_plugin_dir + '/nagios_plugin.py')
shutil.copy2(os.path.join(charm_file_dir, 'default_rsync'),
'/etc/default/rsync')
shutil.copy2(os.path.join(charm_file_dir, 'rsyncd.conf'),
'/etc/rsyncd.conf')
host.rsync(charm_plugin_dir,
'/usr/local/lib/nagios/',
options=['--executability'])
if not os.path.exists(local_plugin_dir + 'nagios_plugin.py'):
os.symlink(pkg_plugin_dir + 'nagios_plugin.py',
local_plugin_dir + 'nagios_plugin.py')
def render_nimsoft_robot_config():
"""Create the nimbus.conf config file.
Renders the appropriate template for the Nimbus Robot
"""
# The v5 template is compatible with all versions < 6
cfg_original_hash = file_hash(NIMBUS_ROBOT_CONFIG)
context = {
'hub': config("hub"),
'domain': config("domain"),
'hubip': config("hubip"),
'hub_robot_name': config("hub-robot-name"),
'secondary_domain': config("secondary-domain"),
'secondary_hubip': config("secondary-hubip"),
'secondary_hub': config("secondary-hub"),
'secondary_hub_robot_name': config("secondary-hub-robot-name"),
'private_address': unit_private_ip(),
'hostname': os.uname()[1]
}
render('robot.cfg', NIMBUS_ROBOT_CONFIG, context=context)
cfg_new_hash = file_hash(NIMBUS_ROBOT_CONFIG)
rsync(charm_dir() + '/files/request_linux_prod.cfg',
'/opt/nimsoft/request.cfg')
# Install the nimbus service
rsync(charm_dir() + '/files/nimbus.service',
'/lib/systemd/system/nimbus.service')
if cfg_original_hash != cfg_new_hash:
service('restart', 'nimbus')
status.active('nimbus ready.')
def migrate_to_mount(new_path):
"""Invoked when new mountpoint appears. This function safely migrates
MySQL data from local disk to persistent storage (only if needed)
"""
old_path = '/var/lib/mysql'
if os.path.islink(old_path):
hookenv.log('{} is already a symlink, skipping migration'.format(
old_path))
return True
# Ensure our new mountpoint is empty. Otherwise error and allow
# users to investigate and migrate manually
files = os.listdir(new_path)
try:
files.remove('lost+found')
except ValueError:
pass
if files:
raise RuntimeError('Persistent storage contains old data. '
'Please investigate and migrate data manually '
'to: {}'.format(new_path))
os.chmod(new_path, 0o700)
if os.path.isdir('/etc/apparmor.d/local'):
render('apparmor.j2', '/etc/apparmor.d/local/usr.sbin.mysqld',
context={'path': os.path.join(new_path, '')})
host.service_reload('apparmor')
host.service_stop('mysql')
host.rsync(os.path.join(old_path, ''), # Ensure we have trailing slashes
os.path.join(new_path, ''),
options=['--archive'])
shutil.rmtree(old_path)
os.symlink(new_path, old_path)
host.service_start('mysql')
def install_files():
# this part lifted from haproxy charm hooks.py
src = os.path.join(os.environ["CHARM_DIR"], "files/thirdparty/")
dst = '/srv/{}/parts/'.format(CHARM_NAME)
maint('Copying scripts from %s to %s' % (src, dst))
host.mkdir(dst, perms=0o755)
for fname in glob.glob(os.path.join(src, "*")):
host.rsync(fname, os.path.join(dst, os.path.basename(fname)))
# Template files may have changed in an upgrade, so we need to rewrite
# them
config_changed()
# Package prerequisites for k8s-kpi-scripts/thirdparty/*
apt_install([
'python-configparser',
'python-prometheus-client',
'python-cssselect',
'python-yaml',
'python-urllib3',
'python3-dev',
'python-click',
'python-swiftclient',
'python-keystoneclient',
])
def install():
juju_log('**********install.real')
rsync(
charm_dir() + '/packages/vsm-dep-repo',
'/opt'
)
rsync(
charm_dir() + '/packages/vsmrepo',
'/opt'
)
rsync(
charm_dir() + '/files/apt.conf',
'/etc/apt'
)
rsync(
charm_dir() + '/files/vsm.list',
'/etc/apt/sources.list.d'
)
rsync(
charm_dir() + '/files/vsm-dep.list',
'/etc/apt/sources.list.d'
)
apt_update()
apt_install(VSM_PACKAGES)
juju_log('**********finished to install vsm')
add_source(config('ceph-source'), config('ceph-key'))
apt_update(fatal=True)
apt_install(packages=PRE_INSTALL_PACKAGES, fatal=True)
def update_nrpe_checks():
if os.path.isdir(NAGIOS_PLUGINS):
rsync(os.path.join(os.getenv('CHARM_DIR'), 'scripts',
'check_rabbitmq.py'),
os.path.join(NAGIOS_PLUGINS, 'check_rabbitmq.py'))
user = '******'
vhost = 'nagios'
password_file = os.path.join(RABBIT_DIR, '%s.passwd' % user)
if os.path.exists(password_file):
password = open(password_file).read().strip()
else:
cmd = ['pwgen', '64', '1']
password = subprocess.check_output(cmd).strip()
with open(password_file, 'wb') as out:
out.write(password)
rabbit.create_vhost(vhost)
rabbit.create_user(user, password)
rabbit.grant_permissions(user, vhost)
nrpe_compat = NRPE()
nrpe_compat.add_check(
shortname=rabbit.RABBIT_USER,
description='Check RabbitMQ',
check_cmd='{}/check_rabbitmq.py --user {} --password {} --vhost {}'
''.format(NAGIOS_PLUGINS, user, password, vhost)
)
nrpe_compat.write()
def update_nrpe_config():
# python-dbus is used by check_upstart_job
apt_install(['python-dbus', 'lockfile-progs'])
log('Refreshing nagios checks')
if os.path.isdir(NAGIOS_PLUGINS):
rsync(
os.path.join(os.getenv('CHARM_DIR'), 'files', 'nagios',
'check_ceph_status.py'),
os.path.join(NAGIOS_PLUGINS, 'check_ceph_status.py'))
script = os.path.join(SCRIPTS_DIR, 'collect_ceph_status.sh')
rsync(
os.path.join(os.getenv('CHARM_DIR'), 'files', 'nagios',
'collect_ceph_status.sh'), script)
cronjob = "{} root {}\n".format('*/5 * * * *', script)
write_file(STATUS_CRONFILE, cronjob)
# Find out if nrpe set nagios_hostname
hostname = nrpe.get_nagios_hostname()
current_unit = nrpe.get_nagios_unit_name()
nrpe_setup = nrpe.NRPE(hostname=hostname)
check_cmd = 'check_ceph_status.py -f {} --degraded_thresh {}' \
' --misplaced_thresh {}' \
' --recovery_rate {}'.format(STATUS_FILE,
config('nagios_degraded_thresh'),
config('nagios_misplaced_thresh'),
config('nagios_recovery_rate'))
if config('nagios_raise_nodeepscrub'):
check_cmd = check_cmd + ' --raise_nodeepscrub'
nrpe_setup.add_check(
shortname="ceph",
description='Check Ceph health {{{}}}'.format(current_unit),
check_cmd=check_cmd)
if config('nagios_additional_checks'):
additional_critical = config('nagios_additional_checks_critical')
x = ast.literal_eval(config('nagios_additional_checks'))
for key, value in x.items():
name = "ceph-{}".format(key.replace(" ", ""))
log("Adding check {}".format(name))
check_cmd = 'check_ceph_status.py -f {}' \
' --additional_check \\\"{}\\\"' \
' {}'.format(STATUS_FILE, value,
"--additional_check_critical"
if additional_critical is True else "")
nrpe_setup.add_check(
shortname=name,
description='Additional Ceph checks {{{}}}'.format(
current_unit),
check_cmd=check_cmd)
if config('nagios_check_num_osds'):
check_cmd = 'check_ceph_status.py -f {} --check_num_osds'.format(
STATUS_FILE)
nrpe_setup.add_check(
shortname='ceph_num_osds',
description='Check whether all OSDs are up and in',
check_cmd=check_cmd)
nrpe_setup.write()
def install():
# this part lifted from haproxy charm hooks.py
src = os.path.join(os.environ["CHARM_DIR"], "scripts")
dst = '/srv/charmbuild-kpi-import/parts/'
maint('Copying scripts from %s to %s' % (src, dst))
host.mkdir(dst, perms=0o755)
for fname in glob.glob(os.path.join(src, "*")):
host.rsync(fname, os.path.join(dst, os.path.basename(fname)))
set_flag('charmbuild.installed')
def rsync_nrpe_checks(plugins_dir):
if not os.path.exists(plugins_dir):
os.makedirs(plugins_dir)
charm_plugin_dir = os.path.join(charm_dir(),
'files',
'plugins/')
rsync(charm_plugin_dir,
plugins_dir,
options=['--executability'])
def agent_changed(rid=None, unit=None):
if 'shared-db' not in CONFIGS.complete_contexts():
juju_log('shared-db relation incomplete. Peer not ready?')
return
if 'amqp' not in CONFIGS.complete_contexts():
juju_log('amqp relation incomplete. Peer not ready?')
return
with open('/etc/manifest/server.manifest') as server_manifest:
flag = 'token-tenant' in server_manifest.read()
if flag:
rel_settings = relation_get(rid=rid, unit=unit)
key = rel_settings.get('ssh_public_key')
juju_log("**********key is %s" % str(key))
if not key:
juju_log('peer did not publish key?')
return
ssh_controller_key_add(key, rid=rid, unit=unit)
host = unit_get('private-address')
hostname = get_hostname(host)
hostaddress = get_host_ip(host)
juju_log("**********host is %s" % host)
juju_log("**********hostname is %s" % hostname)
juju_log("**********hostaddress is %s" % hostaddress)
with open('/etc/hosts', 'a') as hosts:
hosts.write('%s %s' % (hostaddress, hostname) + '\n')
token_tenant = rel_settings.get('token_tenant')
juju_log("**********token_tenant is %s" % token_tenant)
rsync(
charm_dir() + '/files/server.manifest',
'/etc/manifest/server.manifest'
)
c_hostaddress = rel_settings.get('hostaddress')
juju_log("**********controller_hostaddress is %s" % c_hostaddress)
subprocess.check_call(['sudo', 'sed', '-i', 's/^controller_ip/%s/g' % c_hostaddress,
'/etc/manifest/server.manifest'])
subprocess.check_call(['sudo', 'sed', '-i', 's/token-tenant/%s/g' % token_tenant,
'/etc/manifest/server.manifest'])
subprocess.check_call(['sudo', 'service', 'vsm-agent', 'stop'])
subprocess.check_call(['sudo', 'service', 'vsm-agent', 'start'])
subprocess.check_call(['sudo', 'service', 'vsm-physical', 'stop'])
subprocess.check_call(['sudo', 'service', 'vsm-physical', 'start'])
juju_log("**********start vsm-agent")
juju_log("**********start vsm-physical")
def setup_ocf_files():
"""Setup OCF resrouce agent files
"""
# TODO (thedac) Eventually we want to package the OCF files.
# Bundle with the charm until then.
mkdir('/usr/lib/ocf/resource.d/ceph')
mkdir('/usr/lib/ocf/resource.d/maas')
# Xenial corosync is not creating this directory
mkdir('/etc/corosync/uidgid.d')
rsync('files/ocf/ceph/rbd', '/usr/lib/ocf/resource.d/ceph/rbd')
rsync('files/ocf/maas/dns', '/usr/lib/ocf/resource.d/maas/dns')
rsync('files/ocf/maas/maas_dns.py', '/usr/lib/heartbeat/maas_dns.py')
rsync('files/ocf/maas/maasclient/', '/usr/lib/heartbeat/maasclient/')
rsync('files/ocf/maas/maas_stonith_plugin.py',
'/usr/lib/stonith/plugins/external/maas')
def setup_ocf_files():
"""Setup OCF resrouce agent files
"""
# TODO (thedac) Eventually we want to package the OCF files.
# Bundle with the charm until then.
mkdir('/usr/lib/ocf/resource.d/ceph')
mkdir('/usr/lib/ocf/resource.d/maas')
# Xenial corosync is not creating this directory
mkdir('/etc/corosync/uidgid.d')
rsync('ocf/ceph/rbd', '/usr/lib/ocf/resource.d/ceph/rbd')
rsync('ocf/maas/dns', '/usr/lib/ocf/resource.d/maas/dns')
rsync('ocf/maas/maas_dns.py', '/usr/lib/heartbeat/maas_dns.py')
rsync('ocf/maas/maasclient/', '/usr/lib/heartbeat/maasclient/')
rsync(
'ocf/maas/maas_stonith_plugin.py',
'/usr/lib/stonith/plugins/external/maas')
def render_nimsoft_robot_config():
"""
Create the required config files.
Renders the appropriate template for the Nimbus Robot
"""
# The v5 template is compatible with all versions < 6
cfg_original_hash = file_hash(ROBOT_CONFIG)
context = {
'hub': config("hub"),
'domain': config("domain"),
'hubip': config("hubip"),
'hub_robot_name': config("hub-robot-name"),
'secondary_domain': config("secondary-domain"),
'secondary_hubip': config("secondary-hubip"),
'secondary_hub': config("secondary-hub"),
'secondary_hub_robot_name': config("secondary-hub-robot-name"),
'private_address': unit_private_ip(),
'hostname': os.uname()[1],
'aa_profile_mode': config("aa-profile-mode")
}
# Render robot.cfg
render(ROBOT_CONFIG, ROBOT_CONFIG_PATH, context=context)
cfg_new_hash = file_hash(ROBOT_CONFIG)
# Render request.cfg
render(DIST_REQ, DIST_REQ_PATH, context=context)
# Install the nimbus service
rsync(charm_dir() + '/files/nimbus.service',
'/lib/systemd/system/nimbus.service')
# Render AppArmor profile
render(NIMBUS_AA_PROFILE, NIMBUS_AA_PROFILE_PATH, context=context)
# Set AppArmor context
NimbusAppArmorContext().setup_aa_profile()
if cfg_original_hash != cfg_new_hash:
service('restart', 'nimbus')
hookenv.status_set('active', 'ready')
def config_vsm_controller():
if 'shared-db' in CONFIGS.complete_contexts() and \
'amqp' in CONFIGS.complete_contexts() and \
'identity-service' in CONFIGS.complete_contexts():
juju_log("**********config vsm controller")
service_host = auth_token_config('identity_uri').split('/')[2].split(':')[0]
net = '.'.join(service_host.split('.')[0:3]) + ".0\/24"
subprocess.check_call(['sudo', 'sed', '-i', 's/^192.168.*/%s/g' % net,
'/etc/manifest/cluster.manifest'])
subprocess.check_call(['sudo', 'service', 'vsm-api', 'restart'])
subprocess.check_call(['sudo', 'service', 'vsm-scheduler', 'restart'])
subprocess.check_call(['sudo', 'service', 'vsm-conductor', 'restart'])
# TODO fix the hardcode of vsm-dashboard.
subprocess.check_call(['sudo', 'sed', '-i', "s/'service'/'services'/g",
'/usr/share/vsm-dashboard/vsm_dashboard/api/vsm.py'])
keystone_vsm_service_password = auth_token_config('admin_password')
local_settings = "/usr/share/vsm-dashboard/vsm_dashboard/local/local_settings.py"
etc_local_settings = "/etc/vsm-dashboard/local_settings"
rsync(
charm_dir() + '/files/local_settings.template',
local_settings
)
subprocess.check_call(['sudo', 'sed', '-i',
's/^KEYSTONE_VSM_SERVICE_PASSWORD =*.*/KEYSTONE_VSM_SERVICE_PASSWORD = "******"/g' % keystone_vsm_service_password,
local_settings])
subprocess.check_call(['sudo', 'sed', '-i', 's/^OPENSTACK_HOST =*.*/OPENSTACK_HOST = "%s"/g' % service_host,
local_settings])
subprocess.check_call(['sudo', 'sed', '-i', 's/^OPENSTACK_KEYSTONE_DEFAULT_ROLE =*.*/OPENSTACK_KEYSTONE_DEFAULT_ROLE = "_member_"/g',
local_settings])
subprocess.check_call(['sudo', 'rm', '-rf', etc_local_settings])
subprocess.check_call(['sudo', 'ln', '-sf', local_settings, etc_local_settings])
rsync(
charm_dir() + '/scripts/https',
'/tmp/https'
)
subprocess.check_call(['sudo', 'bash', '/tmp/https'])
subprocess.check_call(['sudo', 'service', 'apache2', 'restart'])
open_port('443')
open_port('80')
def install_hugepages():
""" Configure hugepages """
hugepage_config = config('hugepages')
if hugepage_config:
mnt_point = '/run/hugepages/kvm'
hugepage_support(
'nova',
mnt_point=mnt_point,
group='root',
nr_hugepages=get_hugepage_number(),
mount=False,
set_shmmax=True,
)
# Remove hugepages entry if present due to Bug #1518771
Fstab.remove_by_mountpoint(mnt_point)
if subprocess.call(['mountpoint', mnt_point]):
service_restart('qemu-kvm')
rsync(charm_dir() + '/files/qemu-hugefsdir',
'/etc/init.d/qemu-hugefsdir')
subprocess.check_call('/etc/init.d/qemu-hugefsdir')
subprocess.check_call(['update-rc.d', 'qemu-hugefsdir', 'defaults'])
def test_rsyncs_a_path(self, log, check_output):
from_path = '/from/this/path/foo'
to_path = '/to/this/path/bar'
check_output.return_value = b' some output ' # Spaces will be stripped
result = host.rsync(from_path, to_path)
self.assertEqual(result, 'some output')
check_output.assert_called_with([
'/usr/bin/rsync', '-r', '--delete', '--executability',
'/from/this/path/foo', '/to/this/path/bar'
])
def install_hugepages():
""" Configure hugepages """
hugepage_config = config('hugepages')
if hugepage_config:
mnt_point = '/run/hugepages/kvm'
hugepage_support(
'nova',
mnt_point=mnt_point,
group='root',
nr_hugepages=get_hugepage_number(),
mount=False,
set_shmmax=True,
)
if subprocess.call(['mountpoint', mnt_point]):
fstab_mount(mnt_point)
rsync(
charm_dir() + '/files/qemu-hugefsdir',
'/etc/init.d/qemu-hugefsdir'
)
subprocess.check_call('/etc/init.d/qemu-hugefsdir')
subprocess.check_call(['update-rc.d', 'qemu-hugefsdir', 'defaults'])
def install_charm_files(service_name):
"""Install files shipped with charm."""
nag_dirs = [
"/etc/nagios/nrpe.d/",
"/usr/local/lib/nagios/plugins",
"/var/lib/nagios/export/",
]
for nag_dir in nag_dirs:
if not os.path.exists(nag_dir):
Path(nag_dir).mkdir(mode=0o755, parents=True)
charm_file_dir = os.path.join(hookenv.charm_dir(), "files")
charm_plugin_dir = os.path.join(charm_file_dir, "plugins")
if OS_RELEASE_CTXT['ID'] == 'ubuntu':
pkg_plugin_dir = "/usr/lib/nagios/plugins/"
else:
pkg_plugin_dir = "/usr/lib64/nagios/plugins/"
local_plugin_dir = "/usr/local/lib/nagios/plugins/"
shutil.copy2(
os.path.join(charm_file_dir, "nagios_plugin.py"),
pkg_plugin_dir + "nagios_plugin.py",
)
shutil.copy2(
os.path.join(charm_file_dir, "nagios_plugin3.py"),
pkg_plugin_dir + "nagios_plugin3.py",
)
shutil.copy2(os.path.join(charm_file_dir, "default_rsync"),
"/etc/default/rsync")
shutil.copy2(os.path.join(charm_file_dir, "rsyncd.conf"),
"/etc/rsyncd.conf")
host.mkdir("/etc/rsync-juju.d", perms=0o755)
host.rsync(charm_plugin_dir,
"/usr/local/lib/nagios/",
options=["--executability"])
for nagios_plugin in ("nagios_plugin.py", "nagios_plugin3.py"):
if not os.path.exists(local_plugin_dir + nagios_plugin):
os.symlink(pkg_plugin_dir + nagios_plugin,
local_plugin_dir + nagios_plugin)
def update_nrpe_config():
# python-dbus is used by check_upstart_job
apt_install('python-dbus')
log('Refreshing nrpe checks')
if not os.path.exists(NAGIOS_PLUGINS):
mkpath(NAGIOS_PLUGINS)
rsync(os.path.join(os.getenv('CHARM_DIR'), 'files', 'nrpe-external-master',
'check_swift_storage.py'),
os.path.join(NAGIOS_PLUGINS, 'check_swift_storage.py'))
rsync(os.path.join(os.getenv('CHARM_DIR'), 'files', 'nrpe-external-master',
'check_swift_service'),
os.path.join(NAGIOS_PLUGINS, 'check_swift_service'))
rsync(os.path.join(os.getenv('CHARM_DIR'), 'files', 'sudo',
'swift-storage'),
os.path.join(SUDOERS_D, 'swift-storage'))
# Find out if nrpe set nagios_hostname
hostname = nrpe.get_nagios_hostname()
current_unit = nrpe.get_nagios_unit_name()
nrpe_setup = nrpe.NRPE(hostname=hostname)
# check the rings and replication
nrpe_setup.add_check(
shortname='swift_storage',
description='Check swift storage ring hashes and replication'
' {%s}' % current_unit,
check_cmd='check_swift_storage.py {}'.format(
config('nagios-check-params'))
)
nrpe.add_init_service_checks(nrpe_setup, SWIFT_SVCS, current_unit)
nrpe_setup.write()
def update_nrpe_config():
# python-dbus is used by check_upstart_job
# python-psutil is used by check_ntpmon
fetch.apt_install(['python-dbus', 'python-psutil'])
nagios_ntpmon_checks = hookenv.config('nagios_ntpmon_checks').split(" ")
if os.path.isdir(NAGIOS_PLUGINS):
host.rsync(
os.path.join(os.getenv('CHARM_DIR'), 'files', 'nagios',
'check_ntpmon.py'),
os.path.join(NAGIOS_PLUGINS, 'check_ntpmon.py'))
hostname = nrpe.get_nagios_hostname()
current_unit = nrpe.get_nagios_unit_name()
nrpe_setup = nrpe.NRPE(hostname=hostname)
nrpe.add_init_service_checks(nrpe_setup, ['ntp'], current_unit)
allchecks = set(['offset', 'peers', 'reachability', 'sync'])
# remove any previously-created ntpmon checks
nrpe_setup.remove_check(shortname="ntpmon")
for c in allchecks:
nrpe_setup.remove_check(shortname="ntpmon_%s" % c)
# If all checks are specified, combine them into a single check to reduce
# Nagios noise.
if set(nagios_ntpmon_checks) == allchecks:
nrpe_setup.add_check(
shortname="ntpmon",
description='Check NTPmon {}'.format(current_unit),
check_cmd='check_ntpmon.py')
else:
for nc in nagios_ntpmon_checks:
if len(nc) > 0:
nrpe_setup.add_check(
shortname="ntpmon_%s" % nc,
description='Check NTPmon %s {%s}' % (nc, current_unit),
check_cmd='check_ntpmon.py --check %s' % nc)
nrpe_setup.write()
def install_hugepages():
""" Configure hugepages """
hugepage_config = config('hugepages')
if hugepage_config:
mnt_point = '/run/hugepages/kvm'
hugepage_support(
'nova',
mnt_point=mnt_point,
group='root',
nr_hugepages=get_hugepage_number(),
mount=False,
set_shmmax=True,
)
# Remove hugepages entry if present due to Bug #1518771
Fstab.remove_by_mountpoint(mnt_point)
if subprocess.call(['mountpoint', mnt_point]):
service_restart('qemu-kvm')
rsync(
charm_dir() + '/files/qemu-hugefsdir',
'/etc/init.d/qemu-hugefsdir'
)
subprocess.check_call('/etc/init.d/qemu-hugefsdir')
subprocess.check_call(['update-rc.d', 'qemu-hugefsdir', 'defaults'])
def main():
cmd = subprocess.Popen(['lsmod'],
stdout=subprocess.PIPE,
stderr=subprocess.PIPE)
stdout, stederr = cmd.communicate()
if re.compile('(megaraid).*').findall(stdout.decode('utf-8')):
megaraid = True
if os.path.isdir(NAGIOS_PLUGINS):
rsync(
os.path.join(os.getenv('CHARM_DIR'), 'files', 'nagios',
'check_lsi_raid'),
os.path.join(NAGIOS_PLUGINS, 'check_lsi_raid'))
if os.path.isdir(SUDOERS_DIR):
rsync(
os.path.join(os.getenv('CHARM_DIR'), 'files', 'nagios',
'nagios_sudoers'),
os.path.join(SUDOERS_DIR, 'nagios_sudoers'))
if os.path.isdir(NAGIOS_PLUGINS):
rsync(
os.path.join(os.getenv('CHARM_DIR'), 'files', 'nagios',
'check_bond'),
os.path.join(NAGIOS_PLUGINS, 'check_bond'))
hostname = nrpe.get_nagios_hostname()
current_unit = nrpe.get_nagios_unit_name()
nrpe_setup = nrpe.NRPE(hostname=hostname)
# Install megaraid tools
# And add megaraid nagios check
if megaraid:
install_packages(['storcli', 'libfile-which-perl'])
nrpe_setup.add_check(shortname='lsi-raid',
description='LSI Raid Check {%s}' % current_unit,
check_cmd=(os.path.join(NAGIOS_PLUGINS,
'check_lsi_raid')))
# Install checks for the network bonds
if os.path.isfile('/proc/net/bonding/bond0') and \
os.path.isfile('/proc/net/bonding/bond1'):
nrpe_setup.add_check(
shortname='bond0',
description='Bond0 check {%s}' % current_unit,
check_cmd=(os.path.join(NAGIOS_PLUGINS, 'check_bond') +
' -i bond0 -p eth2'))
nrpe_setup.add_check(
shortname='bond0',
description='Bond1 check {%s}' % current_unit,
check_cmd=(os.path.join(NAGIOS_PLUGINS, 'check_bond') +
' -i bond1 -p eth3'))
nrpe_setup.write()
reactive.set_state('raidcheck_installed')
hookenv.status_set('active', 'Unit is ready')
def update_nrpe_checks():
if os.path.isdir(NAGIOS_PLUGINS):
rsync(os.path.join(os.getenv('CHARM_DIR'), 'scripts',
'check_rabbitmq.py'),
os.path.join(NAGIOS_PLUGINS, 'check_rabbitmq.py'))
rsync(os.path.join(os.getenv('CHARM_DIR'), 'scripts',
'check_rabbitmq_queues.py'),
os.path.join(NAGIOS_PLUGINS, 'check_rabbitmq_queues.py'))
if config('stats_cron_schedule'):
script = os.path.join(SCRIPTS_DIR, 'collect_rabbitmq_stats.sh')
cronjob = CRONJOB_CMD.format(schedule=config('stats_cron_schedule'),
timeout=config('cron-timeout'),
command=script)
rsync(os.path.join(charm_dir(), 'scripts',
'collect_rabbitmq_stats.sh'), script)
write_file(STATS_CRONFILE, cronjob)
elif os.path.isfile(STATS_CRONFILE):
os.remove(STATS_CRONFILE)
# Find out if nrpe set nagios_hostname
hostname = nrpe.get_nagios_hostname()
myunit = nrpe.get_nagios_unit_name()
# create unique user and vhost for each unit
current_unit = local_unit().replace('/', '-')
user = '******' % current_unit
vhost = 'nagios-%s' % current_unit
password = rabbit.get_rabbit_password(user, local=True)
rabbit.create_vhost(vhost)
rabbit.create_user(user, password)
rabbit.grant_permissions(user, vhost)
nrpe_compat = nrpe.NRPE(hostname=hostname)
nrpe_compat.add_check(
shortname=rabbit.RABBIT_USER,
description='Check RabbitMQ {%s}' % myunit,
check_cmd='{}/check_rabbitmq.py --user {} --password {} --vhost {}'
''.format(NAGIOS_PLUGINS, user, password, vhost)
)
if config('queue_thresholds'):
cmd = ""
# If value of queue_thresholds is incorrect we want the hook to fail
for item in yaml.safe_load(config('queue_thresholds')):
cmd += ' -c "{}" "{}" {} {}'.format(*item)
nrpe_compat.add_check(
shortname=rabbit.RABBIT_USER + '_queue',
description='Check RabbitMQ Queues',
check_cmd='{}/check_rabbitmq_queues.py{} {}'.format(
NAGIOS_PLUGINS, cmd, STATS_DATAFILE)
)
nrpe_compat.write()
def update_nrpe_checks():
if os.path.isdir(NAGIOS_PLUGINS):
rsync(os.path.join(os.getenv('CHARM_DIR'), 'scripts',
'check_rabbitmq.py'),
os.path.join(NAGIOS_PLUGINS, 'check_rabbitmq.py'))
rsync(os.path.join(os.getenv('CHARM_DIR'), 'scripts',
'check_rabbitmq_queues.py'),
os.path.join(NAGIOS_PLUGINS, 'check_rabbitmq_queues.py'))
if config('stats_cron_schedule'):
script = os.path.join(SCRIPTS_DIR, 'collect_rabbitmq_stats.sh')
cronjob = "{} root {}\n".format(config('stats_cron_schedule'), script)
rsync(os.path.join(charm_dir(), 'scripts',
'collect_rabbitmq_stats.sh'), script)
write_file(STATS_CRONFILE, cronjob)
elif os.path.isfile(STATS_CRONFILE):
os.remove(STATS_CRONFILE)
# Find out if nrpe set nagios_hostname
hostname = nrpe.get_nagios_hostname()
myunit = nrpe.get_nagios_unit_name()
# create unique user and vhost for each unit
current_unit = local_unit().replace('/', '-')
user = '******' % current_unit
vhost = 'nagios-%s' % current_unit
password = rabbit.get_rabbit_password(user, local=True)
rabbit.create_vhost(vhost)
rabbit.create_user(user, password)
rabbit.grant_permissions(user, vhost)
nrpe_compat = nrpe.NRPE(hostname=hostname)
nrpe_compat.add_check(
shortname=rabbit.RABBIT_USER,
description='Check RabbitMQ {%s}' % myunit,
check_cmd='{}/check_rabbitmq.py --user {} --password {} --vhost {}'
''.format(NAGIOS_PLUGINS, user, password, vhost)
)
if config('queue_thresholds'):
cmd = ""
# If value of queue_thresholds is incorrect we want the hook to fail
for item in yaml.safe_load(config('queue_thresholds')):
cmd += ' -c "{}" "{}" {} {}'.format(*item)
nrpe_compat.add_check(
shortname=rabbit.RABBIT_USER + '_queue',
description='Check RabbitMQ Queues',
check_cmd='{}/check_rabbitmq_queues.py{} {}'.format(
NAGIOS_PLUGINS, cmd, STATS_DATAFILE)
)
nrpe_compat.write()
def install_ntpmon():
"""
Install package dependencies, source files, and startup configuration.
"""
hookenv.log('installing ntpmon dependencies')
apt_install(['python3-psutil'])
hookenv.log('installing ntpmon')
host.rsync('src/', ntpmon_dir)
if host.init_is_systemd():
hookenv.log('installing ntpmon systemd configuration')
host.rsync('src/' + service_name + '.systemd', systemd_config)
subprocess.call(['systemd', 'daemon-reload'])
else:
hookenv.log('installing ntpmon upstart configuration')
host.rsync('src/' + service_name + '.upstart', upstart_config)
set_state('ntpmon.installed')
remove_state('ntpmon.configured')
def deploy_scripts():
rsync(path.join(getenv("CHARM_DIR"), "scripts", "auto_lint.py"),
path.join(USR_LIB, "auto_lint.py"))
def update_nrpe_config():
# python-dbus is used by check_upstart_job
apt_install('python-dbus')
log('Refreshing nrpe checks')
if not os.path.exists(NAGIOS_PLUGINS):
mkpath(NAGIOS_PLUGINS)
rsync(
os.path.join(os.getenv('CHARM_DIR'), 'files', 'nrpe-external-master',
'check_swift_storage.py'),
os.path.join(NAGIOS_PLUGINS, 'check_swift_storage.py'))
rsync(
os.path.join(os.getenv('CHARM_DIR'), 'files', 'nrpe-external-master',
'check_timed_logs.pl'),
os.path.join(NAGIOS_PLUGINS, 'check_timed_logs.pl'))
rsync(
os.path.join(os.getenv('CHARM_DIR'), 'files', 'nrpe-external-master',
'check_swift_replicator_logs.sh'),
os.path.join(NAGIOS_PLUGINS, 'check_swift_replicator_logs.sh'))
rsync(
os.path.join(os.getenv('CHARM_DIR'), 'files', 'nrpe-external-master',
'check_swift_service'),
os.path.join(NAGIOS_PLUGINS, 'check_swift_service'))
rsync(
os.path.join(os.getenv('CHARM_DIR'), 'files', 'sudo', 'swift-storage'),
os.path.join(SUDOERS_D, 'swift-storage'))
# Find out if nrpe set nagios_hostname
hostname = nrpe.get_nagios_hostname()
current_unit = nrpe.get_nagios_unit_name()
nrpe_setup = nrpe.NRPE(hostname=hostname)
# check the rings and replication
nrpe_setup.add_check(
shortname='swift_storage',
description='Check swift storage ring hashes and replication'
' {%s}' % current_unit,
check_cmd='check_swift_storage.py {}'.format(
config('nagios-check-params')))
object_port = config('object-server-port')
container_port = config('container-server-port')
account_port = config('account-server-port')
nrpe_setup.add_check(
shortname="swift-object-server-api",
description="Check Swift Object Server API availability",
check_cmd="/usr/lib/nagios/plugins/check_http \
-I localhost -u /recon/version -p {} \
-e \"OK\"".format(object_port))
nrpe_setup.add_check(
shortname="swift-container-server-api",
description="Check Swift Container Server API availability",
check_cmd="/usr/lib/nagios/plugins/check_http \
-I localhost -u /recon/version -p {} \
-e \"OK\"".format(container_port))
nrpe_setup.add_check(
shortname="swift-account-server-api",
description="Check Swift Account Server API availability",
check_cmd="/usr/lib/nagios/plugins/check_http \
-I localhost -u /recon/version -p {} \
-e \"OK\"".format(account_port))
if config('nagios-replication-check-params'):
nrpe_setup.add_check(
shortname='swift_replicator_health',
description='Check swift object replicator log reporting',
check_cmd='check_swift_replicator_logs.sh {}'.format(
config('nagios-replication-check-params')))
else:
nrpe_setup.remove_check(shortname='swift_replicator_health')
nrpe.add_init_service_checks(nrpe_setup, SWIFT_SVCS, current_unit)
nrpe_setup.write()
def update_nrpe_checks():
if os.path.isdir(NAGIOS_PLUGINS):
rsync(os.path.join(charm_dir(), 'scripts',
'check_rabbitmq.py'),
os.path.join(NAGIOS_PLUGINS, 'check_rabbitmq.py'))
rsync(os.path.join(charm_dir(), 'scripts',
'check_rabbitmq_queues.py'),
os.path.join(NAGIOS_PLUGINS, 'check_rabbitmq_queues.py'))
if config('management_plugin'):
rsync(os.path.join(charm_dir(), 'scripts',
'check_rabbitmq_cluster.py'),
os.path.join(NAGIOS_PLUGINS, 'check_rabbitmq_cluster.py'))
if config('stats_cron_schedule'):
script = os.path.join(SCRIPTS_DIR, 'collect_rabbitmq_stats.sh')
cronjob = CRONJOB_CMD.format(schedule=config('stats_cron_schedule'),
timeout=config('cron-timeout'),
command=script)
rsync(os.path.join(charm_dir(), 'scripts',
'collect_rabbitmq_stats.sh'), script)
write_file(STATS_CRONFILE, cronjob)
elif os.path.isfile(STATS_CRONFILE):
os.remove(STATS_CRONFILE)
# Find out if nrpe set nagios_hostname
hostname = nrpe.get_nagios_hostname()
myunit = nrpe.get_nagios_unit_name()
# create unique user and vhost for each unit
current_unit = local_unit().replace('/', '-')
user = '******' % current_unit
vhost = 'nagios-%s' % current_unit
password = rabbit.get_rabbit_password(user, local=True)
rabbit.create_vhost(vhost)
rabbit.create_user(user, password, ['monitoring'])
rabbit.grant_permissions(user, vhost)
nrpe_compat = nrpe.NRPE(hostname=hostname)
if config('ssl') in ['off', 'on']:
cmd = ('{plugins_dir}/check_rabbitmq.py --user {user} '
'--password {password} --vhost {vhost}')
cmd = cmd.format(plugins_dir=NAGIOS_PLUGINS, user=user,
password=password, vhost=vhost)
nrpe_compat.add_check(
shortname=rabbit.RABBIT_USER,
description='Check RabbitMQ {%s}' % myunit,
check_cmd=cmd
)
if config('ssl') in ['only', 'on']:
log('Adding rabbitmq SSL check', level=DEBUG)
cmd = ('{plugins_dir}/check_rabbitmq.py --user {user} '
'--password {password} --vhost {vhost} '
'--ssl --ssl-ca {ssl_ca} --port {port}')
cmd = cmd.format(plugins_dir=NAGIOS_PLUGINS,
user=user,
password=password,
port=int(config('ssl_port')),
vhost=vhost,
ssl_ca=SSL_CA_FILE)
nrpe_compat.add_check(
shortname=rabbit.RABBIT_USER + "_ssl",
description='Check RabbitMQ (SSL) {%s}' % myunit,
check_cmd=cmd
)
if config('queue_thresholds'):
cmd = ""
# If value of queue_thresholds is incorrect we want the hook to fail
for item in yaml.safe_load(config('queue_thresholds')):
cmd += ' -c "{}" "{}" {} {}'.format(*item)
nrpe_compat.add_check(
shortname=rabbit.RABBIT_USER + '_queue',
description='Check RabbitMQ Queues',
check_cmd='{}/check_rabbitmq_queues.py{} {}'.format(
NAGIOS_PLUGINS, cmd, STATS_DATAFILE)
)
if config('management_plugin'):
# add NRPE check
nrpe_compat.add_check(
shortname=rabbit.RABBIT_USER + '_cluster',
description='Check RabbitMQ Cluster',
check_cmd='{}/check_rabbitmq_cluster.py --port {} --user {} --password {}'.format(
NAGIOS_PLUGINS,
rabbit.get_managment_port(),
user,
password
)
)
nrpe_compat.write()
def update_nrpe_checks():
if os.path.isdir(NAGIOS_PLUGINS):
rsync(os.path.join(charm_dir(), 'scripts',
'check_rabbitmq.py'),
os.path.join(NAGIOS_PLUGINS, 'check_rabbitmq.py'))
rsync(os.path.join(charm_dir(), 'scripts',
'check_rabbitmq_queues.py'),
os.path.join(NAGIOS_PLUGINS, 'check_rabbitmq_queues.py'))
if config('management_plugin'):
rsync(os.path.join(charm_dir(), 'scripts',
'check_rabbitmq_cluster.py'),
os.path.join(NAGIOS_PLUGINS, 'check_rabbitmq_cluster.py'))
if config('stats_cron_schedule'):
script = os.path.join(SCRIPTS_DIR, 'collect_rabbitmq_stats.sh')
cronjob = CRONJOB_CMD.format(schedule=config('stats_cron_schedule'),
timeout=config('cron-timeout'),
command=script)
rsync(os.path.join(charm_dir(), 'scripts',
'collect_rabbitmq_stats.sh'), script)
write_file(STATS_CRONFILE, cronjob)
elif os.path.isfile(STATS_CRONFILE):
os.remove(STATS_CRONFILE)
if config('management_plugin'):
rsync(os.path.join(charm_dir(), 'scripts',
'check_rabbitmq_cluster.py'),
os.path.join(NAGIOS_PLUGINS, 'check_rabbitmq_cluster.py'))
# Find out if nrpe set nagios_hostname
hostname = nrpe.get_nagios_hostname()
myunit = nrpe.get_nagios_unit_name()
# create unique user and vhost for each unit
current_unit = local_unit().replace('/', '-')
user = '******' % current_unit
vhost = 'nagios-%s' % current_unit
password = rabbit.get_rabbit_password(user, local=True)
rabbit.create_vhost(vhost)
rabbit.create_user(user, password, ['monitoring'])
rabbit.grant_permissions(user, vhost)
nrpe_compat = nrpe.NRPE(hostname=hostname)
if config('ssl') in ['off', 'on']:
cmd = ('{plugins_dir}/check_rabbitmq.py --user {user} '
'--password {password} --vhost {vhost}')
cmd = cmd.format(plugins_dir=NAGIOS_PLUGINS, user=user,
password=password, vhost=vhost)
nrpe_compat.add_check(
shortname=rabbit.RABBIT_USER,
description='Check RabbitMQ {%s}' % myunit,
check_cmd=cmd
)
if config('ssl') in ['only', 'on']:
log('Adding rabbitmq SSL check', level=DEBUG)
cmd = ('{plugins_dir}/check_rabbitmq.py --user {user} '
'--password {password} --vhost {vhost} '
'--ssl --ssl-ca {ssl_ca} --port {port}')
cmd = cmd.format(plugins_dir=NAGIOS_PLUGINS,
user=user,
password=password,
port=int(config('ssl_port')),
vhost=vhost,
ssl_ca=SSL_CA_FILE)
nrpe_compat.add_check(
shortname=rabbit.RABBIT_USER + "_ssl",
description='Check RabbitMQ (SSL) {%s}' % myunit,
check_cmd=cmd
)
if config('queue_thresholds'):
cmd = ""
# If value of queue_thresholds is incorrect we want the hook to fail
for item in yaml.safe_load(config('queue_thresholds')):
cmd += ' -c "{}" "{}" {} {}'.format(*item)
nrpe_compat.add_check(
shortname=rabbit.RABBIT_USER + '_queue',
description='Check RabbitMQ Queues',
check_cmd='{}/check_rabbitmq_queues.py{} {}'.format(
NAGIOS_PLUGINS, cmd, STATS_DATAFILE)
)
if config('management_plugin'):
# add NRPE check
nrpe_compat.add_check(
shortname=rabbit.RABBIT_USER + '_cluster',
description='Check RabbitMQ Cluster',
check_cmd='{}/check_rabbitmq_cluster.py --port {} --user {} --password {}'.format(
NAGIOS_PLUGINS,
rabbit.get_managment_port(),
user,
password
)
)
nrpe_compat.write()
def update_plugins(self):
charm_plugin_dir = os.path.join(hookenv.charm_dir(), 'files',
'plugins/')
host.rsync(charm_plugin_dir,
self.plugins_dir,
options=['--executability'])
def update_nrpe_checks():
if os.path.isdir(NAGIOS_PLUGINS):
rsync(os.path.join(charm_dir(), 'files', 'check_rabbitmq.py'),
os.path.join(NAGIOS_PLUGINS, 'check_rabbitmq.py'))
rsync(os.path.join(charm_dir(), 'files', 'check_rabbitmq_queues.py'),
os.path.join(NAGIOS_PLUGINS, 'check_rabbitmq_queues.py'))
if config('management_plugin'):
rsync(
os.path.join(charm_dir(), 'files',
'check_rabbitmq_cluster.py'),
os.path.join(NAGIOS_PLUGINS, 'check_rabbitmq_cluster.py'))
if config('stats_cron_schedule'):
script = os.path.join(SCRIPTS_DIR, 'collect_rabbitmq_stats.sh')
cronjob = CRONJOB_CMD.format(schedule=config('stats_cron_schedule'),
timeout=config('cron-timeout'),
command=script)
rsync(os.path.join(charm_dir(), 'files', 'collect_rabbitmq_stats.sh'),
script)
write_file(STATS_CRONFILE, cronjob)
elif os.path.isfile(STATS_CRONFILE):
os.remove(STATS_CRONFILE)
# Find out if nrpe set nagios_hostname
hostname = nrpe.get_nagios_hostname()
myunit = nrpe.get_nagios_unit_name()
# create unique user and vhost for each unit
current_unit = local_unit().replace('/', '-')
user = '******'.format(current_unit)
vhosts = [{'vhost': user, 'shortname': rabbit.RABBIT_USER}]
password = rabbit.get_rabbit_password(user, local=True)
nrpe_compat = nrpe.NRPE(hostname=hostname)
rabbit.create_user(user, password, ['monitoring'])
if config('check-vhosts'):
for other_vhost in config('check-vhosts').split(' '):
if other_vhost:
item = {
'vhost': other_vhost,
'shortname': 'rabbit_{}'.format(other_vhost)
}
vhosts.append(item)
for vhost in vhosts:
rabbit.create_vhost(vhost['vhost'])
rabbit.grant_permissions(user, vhost['vhost'])
if config('ssl') in ['off', 'on']:
cmd = ('{}/check_rabbitmq.py --user {} --password {} '
'--vhost {}'.format(NAGIOS_PLUGINS, user, password,
vhost['vhost']))
log('Adding rabbitmq non-SSL check for {}'.format(vhost['vhost']),
level=DEBUG)
description = 'Check RabbitMQ {} {}'.format(myunit, vhost['vhost'])
nrpe_compat.add_check(shortname=vhost['shortname'],
description=description,
check_cmd=cmd)
if config('ssl') in ['only', 'on']:
cmd = ('{}/check_rabbitmq.py --user {} --password {} '
'--vhost {} --ssl --ssl-ca {} --port {}'.format(
NAGIOS_PLUGINS, user, password, vhost['vhost'],
SSL_CA_FILE, int(config('ssl_port'))))
log('Adding rabbitmq SSL check for {}'.format(vhost['vhost']),
level=DEBUG)
description = 'Check RabbitMQ (SSL) {} {}'.format(
myunit, vhost['vhost'])
nrpe_compat.add_check(shortname=vhost['shortname'] + "_ssl",
description=description,
check_cmd=cmd)
if config('queue_thresholds'):
cmd = ""
# If value of queue_thresholds is incorrect we want the hook to fail
for item in yaml.safe_load(config('queue_thresholds')):
cmd += ' -c "{}" "{}" {} {}'.format(*item)
nrpe_compat.add_check(
shortname=rabbit.RABBIT_USER + '_queue',
description='Check RabbitMQ Queues',
check_cmd='{}/check_rabbitmq_queues.py{} {}'.format(
NAGIOS_PLUGINS, cmd, STATS_DATAFILE))
if config('management_plugin'):
# add NRPE check
_check_cmd = (
'{}/check_rabbitmq_cluster.py --port {} --user {} --password {}'.
format(NAGIOS_PLUGINS, rabbit.get_managment_port(), user,
password))
nrpe_compat.add_check(shortname=rabbit.RABBIT_USER + '_cluster',
description='Check RabbitMQ Cluster',
check_cmd=_check_cmd)
nrpe_compat.write()
