def test_scale_down_on_group_threshold_breach():
tenant = 'test_group_breach_scale_down'
config = utils.load_config()
main_client = utils.get_rest_client_using_config(
config,
tenant='default_tenant',
)
main_client.tenants.create(tenant)
client = utils.get_rest_client_using_config(
config,
tenant=tenant,
)
utils.upload_config_secrets(config, client)
installed_plugins = utils.upload_required_plugins(client)
utils.deploy_nagios(
utils.get_examples_blueprint_path('nagios-groups.yaml'),
utils.get_nagios_inputs(config),
client,
)
utils.install_blueprint(
utils.get_examples_blueprint_path('basegroup1.yaml'),
utils.get_monitored_vms_inputs(config),
'basegroup1',
client,
)
utils.install_blueprint(
utils.get_examples_blueprint_path('basegroup2.yaml'),
utils.get_monitored_vms_inputs(config),
'basegroup2',
client,
)
utils.execute_arbitrary_command(
'basegroup1',
'echo 2 > /tmp/cloudifytestinteger',
client,
)
utils.execute_arbitrary_command(
'basegroup1',
'echo {time_now}:2.0 > /tmp/cloudifytestcounter'.format(time_now=int(
time.time()), ),
client,
)
# Confirm the expected workflow runs
utils.wait_for_execution_on_deployment('scale', 'basegroup1', client)
utils.remove_deployment('basegroup1', client)
utils.delete_blueprint('basegroup1', client)
utils.remove_deployment('basegroup2', client)
utils.delete_blueprint('basegroup2', client)
utils.remove_nagios(client)
utils.delete_plugins(installed_plugins, client)
utils.remove_config_secrets(client)
main_client.tenants.delete(tenant)
def test_ignore_unreachable():
tenant = 'test_aggregate_ignore_unreachable'
config = utils.load_config()
main_client = utils.get_rest_client_using_config(
config,
tenant='default_tenant',
)
main_client.tenants.create(tenant)
client = utils.get_rest_client_using_config(
config,
tenant=tenant,
)
utils.upload_config_secrets(config, client)
installed_plugins = utils.upload_required_plugins(client)
utils.deploy_nagios(
utils.get_examples_blueprint_path('nagios-aggregates.yaml'),
utils.get_nagios_inputs(config),
client,
)
utils.install_blueprint(
utils.get_examples_blueprint_path('baseaggregate.yaml'),
utils.get_monitored_vms_inputs(config),
'baseaggregate',
client,
)
# Turn off SNMP on a node
utils.execute_arbitrary_command(
'baseaggregate', 'sudo service snmpd stop', client,
utils.get_first_node_instance('base_aggregate_host', 'baseaggregate',
client))
# Then trigger a scale down
utils.execute_arbitrary_command(
'baseaggregate',
'echo {time_now}:0 > /tmp/cloudifytestcounter'.format(time_now=int(
time.time()), ),
client,
)
# Confirm the expected workflow runs
utils.wait_for_execution_on_deployment(
'scale',
'baseaggregate',
client,
max_wait_for_start=120,
)
utils.remove_deployment('baseaggregate', client)
utils.delete_blueprint('baseaggregate', client)
utils.remove_nagios(client)
utils.delete_plugins(installed_plugins, client)
utils.remove_config_secrets(client)
main_client.tenants.delete(tenant)
def test_abort_on_unreachable():
tenant = 'test_aggregate_abort_on_unreachable'
config = utils.load_config()
main_client = utils.get_rest_client_using_config(
config,
tenant='default_tenant',
)
main_client.tenants.create(tenant)
client = utils.get_rest_client_using_config(
config,
tenant=tenant,
)
utils.upload_config_secrets(config, client)
installed_plugins = utils.upload_required_plugins(client)
utils.deploy_nagios(
utils.get_examples_blueprint_path('nagios-aggregates.yaml'),
utils.get_nagios_inputs(config),
client,
)
utils.install_blueprint(
utils.get_examples_blueprint_path('baseaggregate.yaml'),
utils.get_monitored_vms_inputs(config),
'baseaggregate',
client,
)
# Turn off SNMP on a node
utils.execute_arbitrary_command(
'baseaggregate', 'sudo service snmpd stop', client,
utils.get_first_node_instance('base_aggregate_host', 'baseaggregate',
client))
# We should see an UNKNOWN check state within 60 seconds
saw_unknown_state = False
for check in range(60):
result = utils.execute_arbitrary_command(
'nagios',
'sudo tail -n10 /var/log/nagios/check_snmp_aggregate.log '
'| grep UNKNOWN',
client,
)
if result['status'] == 0:
saw_unknown_state = True
break
time.sleep(1)
assert saw_unknown_state
utils.remove_deployment('baseaggregate', client)
utils.delete_blueprint('baseaggregate', client)
utils.remove_nagios(client)
utils.delete_plugins(installed_plugins, client)
utils.remove_config_secrets(client)
main_client.tenants.delete(tenant)
def test_external_trap_triggers_heal():
tenant = 'test_external_trap_triggers_heal'
config = utils.load_config()
main_client = utils.get_rest_client_using_config(
config,
tenant='default_tenant',
)
main_client.tenants.create(tenant)
client = utils.get_rest_client_using_config(
config,
tenant=tenant,
)
utils.upload_config_secrets(config, client)
installed_plugins = utils.upload_required_plugins(client)
utils.deploy_nagios(
utils.get_examples_blueprint_path('nagios-traps.yaml'),
utils.get_nagios_inputs(config),
client,
)
utils.install_blueprint(
utils.get_examples_blueprint_path('basetrap.yaml'),
utils.get_monitored_vms_inputs(config),
'basetrap',
client,
)
# Send the trap
utils.execute_arbitrary_command(
'nagios',
'snmptrap -v2c -c testcommunity {ip} "" {oid} '
'.1.3.6.1.4.1.52312.0.1.1 s "Test message" '
'.1.3.6.1.4.1.52312.0.1.2 s "The address is {node_address}"'.format(
ip='localhost',
oid='.1.3.6.1.4.1.52312.0.0.2',
node_address=utils.get_first_node_instance_ip(
node='base_trap_host',
deployment='basetrap',
client=client,
),
),
client,
)
# Confirm the expected workflow runs
utils.wait_for_execution_on_deployment('heal', 'basetrap', client)
utils.remove_deployment('basetrap', client)
utils.delete_blueprint('basetrap', client)
utils.remove_nagios(client)
utils.delete_plugins(installed_plugins, client)
utils.remove_config_secrets(client)
main_client.tenants.delete(tenant)
def test_heal_on_threshold_exceeded():
tenant = 'test_value_breach_heal'
config = utils.load_config()
main_client = utils.get_rest_client_using_config(
config,
tenant='default_tenant',
)
main_client.tenants.create(tenant)
client = utils.get_rest_client_using_config(
config,
tenant=tenant,
)
utils.upload_config_secrets(config, client)
installed_plugins = utils.upload_required_plugins(client)
utils.deploy_nagios(
utils.get_examples_blueprint_path('nagios-values.yaml'),
utils.get_nagios_inputs(config),
client,
)
utils.install_blueprint(
utils.get_examples_blueprint_path('basevalue.yaml'),
utils.get_monitored_vms_inputs(config),
'basevalue',
client,
)
utils.execute_arbitrary_command(
'basevalue',
'echo 42 > /tmp/cloudifytestinteger',
client,
)
# Confirm the expected workflow runs
utils.wait_for_execution_on_deployment('heal', 'basevalue', client)
utils.remove_deployment('basevalue', client)
utils.delete_blueprint('basevalue', client)
utils.remove_nagios(client)
utils.delete_plugins(installed_plugins, client)
utils.remove_config_secrets(client)
main_client.tenants.delete(tenant)
def test_trap_triggers_scale_down():
tenant = 'test_trap_triggers_scale_down'
config = utils.load_config()
main_client = utils.get_rest_client_using_config(
config,
tenant='default_tenant',
)
main_client.tenants.create(tenant)
client = utils.get_rest_client_using_config(
config,
tenant=tenant,
)
utils.upload_config_secrets(config, client)
installed_plugins = utils.upload_required_plugins(client)
utils.deploy_nagios(
utils.get_examples_blueprint_path('nagios-traps.yaml'),
utils.get_nagios_inputs(config),
client,
)
utils.install_blueprint(
utils.get_examples_blueprint_path('basetrap.yaml'),
utils.get_monitored_vms_inputs(config),
'basetrap',
client,
)
utils.execute(
'basetrap',
'scale',
client,
parameters={
'scalable_entity_name': 'base_trap_host',
'delta': '+1',
'scale_compute': True,
},
)
# Send the trap
utils.execute_arbitrary_command(
'basetrap',
'snmptrap -v2c -c testcommunity {ip} "" {oid}'.format(
ip=utils.get_nagios_internal_ip(client),
oid='.1.3.6.1.4.1.52312.0.0.1',
),
client,
)
# Confirm the expected workflow runs
utils.wait_for_execution_on_deployment('scale', 'basetrap', client)
utils.remove_deployment('basetrap', client)
utils.delete_blueprint('basetrap', client)
utils.remove_nagios(client)
utils.delete_plugins(installed_plugins, client)
utils.remove_config_secrets(client)
main_client.tenants.delete(tenant)
def test_trap_not_triggered_constraints():
tenant = 'test_trap_not_triggering_constraints'
config = utils.load_config()
main_client = utils.get_rest_client_using_config(
config,
tenant='default_tenant',
)
main_client.tenants.create(tenant)
client = utils.get_rest_client_using_config(
config,
tenant=tenant,
)
utils.upload_config_secrets(config, client)
installed_plugins = utils.upload_required_plugins(client)
utils.deploy_nagios(
utils.get_examples_blueprint_path('nagios-traps.yaml'),
utils.get_nagios_inputs(config),
client,
)
utils.install_blueprint(
utils.get_examples_blueprint_path('basetrap.yaml'),
utils.get_monitored_vms_inputs(config),
'basetrap',
client,
)
# Send the trap
utils.execute_arbitrary_command(
'basetrap',
'snmptrap -v2c -c testcommunity {ip} "" {oid}'.format(
ip=utils.get_nagios_internal_ip(client),
oid='.1.3.6.1.4.1.52312.0.0.1',
),
client,
)
# We should see a log entry stating that there was no reaction within 5
# seconds
saw_no_reaction = False
for check in range(5):
result = utils.execute_arbitrary_command(
'nagios',
'sudo tail -n10 /var/log/nagios/notify_cloudify.log '
'| grep "No reaction"',
client,
)
if result['status'] == 0:
saw_no_reaction = True
break
time.sleep(1)
assert saw_no_reaction
utils.remove_deployment('basetrap', client)
utils.delete_blueprint('basetrap', client)
utils.remove_nagios(client)
utils.delete_plugins(installed_plugins, client)
utils.remove_config_secrets(client)
main_client.tenants.delete(tenant)
def test_groups_do_not_collide():
tenant = 'test_groups_do_not_collide'
config = utils.load_config()
main_client = utils.get_rest_client_using_config(
config,
tenant='default_tenant',
)
main_client.tenants.create(tenant)
client = utils.get_rest_client_using_config(
config,
tenant=tenant,
)
utils.upload_config_secrets(config, client)
installed_plugins = utils.upload_required_plugins(client)
utils.deploy_nagios(
utils.get_examples_blueprint_path('nagios-groups-nocollide.yaml'),
utils.get_nagios_inputs(config),
client,
)
utils.install_blueprint(
utils.get_examples_blueprint_path('basegroup-nocollide.yaml'),
utils.get_monitored_vms_inputs(config),
'basegroup-nocollide',
client,
)
attempt = 0
counter_result = 'no checks'
# Until the dependent checks have run there will be a response indicating
# that there are "no checks associated"
while 'no checks' in counter_result:
# Wait up to a 30 seconds (this should be about twice as long as is
# needed)
assert attempt < 10, 'Timed out waiting for counter check'
time.sleep(3)
counter_result = utils.execute_arbitrary_command(
'nagios',
'sudo /usr/lib64/nagios/plugins/check_group_aggregate '
'--approach="arithmetic_mean" --tenant="{tenant}" '
'--group-instance="crossdeploymentcountergroup" '
'--unknown="ignore" '
'--group-type="Test check group counter"'.format(tenant=tenant, ),
client,
)['output']
attempt += 1
# This should be a reasonably large number
assert int(get_check_value(counter_result)) > 100000
attempt = 0
value_result = 'no checks'
# Until the dependent checks have run there will be a response indicating
# that there are "no checks associated"
while 'no checks' in value_result:
# Wait up to a 30 seconds (this should be about twice as long as is
# needed)
assert attempt < 10, 'Timed out waiting for value check'
time.sleep(3)
value_result = utils.execute_arbitrary_command(
'nagios',
'sudo /usr/lib64/nagios/plugins/check_group_aggregate '
'--approach="arithmetic_mean" --tenant="{tenant}" '
'--group-instance="crossdeploymentvaluegroup" --unknown="ignore" '
'--group-type="Test check group value"'.format(tenant=tenant, ),
client,
)['output']
attempt += 1
assert int(get_check_value(value_result)) == 0
utils.remove_deployment('basegroup-nocollide', client)
utils.delete_blueprint('basegroup-nocollide', client)
utils.remove_nagios(client)
utils.delete_plugins(installed_plugins, client)
utils.remove_config_secrets(client)
main_client.tenants.delete(tenant)
def test_adding_and_updating_target_types():
tenant = 'test_adding_and_updating_target_types'
config = utils.load_config()
main_client = utils.get_rest_client_using_config(
config,
tenant='default_tenant',
)
main_client.tenants.create(tenant)
client = utils.get_rest_client_using_config(
config,
tenant=tenant,
)
utils.upload_config_secrets(config, client)
installed_plugins = utils.upload_required_plugins(client)
utils.deploy_nagios(
utils.get_examples_blueprint_path('nagios-update-1.yaml'),
utils.get_nagios_inputs(config),
client,
)
# Confirm we currently have no target types
nodes = [item['type'] for item in client.nodes.list(_include=['type'])]
assert 'cloudify.nagios.nodes.TargetType' not in nodes
# We don't yet have the target type we need, so let's add it
client.blueprints.upload(
path=utils.get_examples_blueprint_path('nagios-update-2.yaml'),
entity_id='nagiosupdate2',
)
client.deployment_updates.update_with_existing_blueprint(
deployment_id='nagios',
blueprint_id='nagiosupdate2',
)
# Now that we have the target type we need, we can install our test node
utils.install_blueprint(
utils.get_examples_blueprint_path('baseupdate.yaml'),
utils.get_monitored_vms_inputs(config),
'baseupdate',
client,
)
# Now we will set the test integer to return a higher value
utils.execute_arbitrary_command(
'baseupdate',
'echo 10 > /tmp/cloudifytestinteger',
client,
)
# Allow time for the heal to start running if there is a problem
time.sleep(60)
# ...and then update the check threshold so that the check can actually run
client.blueprints.upload(
path=utils.get_examples_blueprint_path('nagios-update-3.yaml'),
entity_id='nagiosupdate3',
)
update = client.deployment_updates.update_with_existing_blueprint(
deployment_id='nagios',
blueprint_id='nagiosupdate3',
reinstall_list=[utils.get_first_node_instance(
'base_update_instance',
'nagios',
client,
)],
)
utils.wait_for_execution(update['execution_id'], client)
# Reconcile because updated target types have to be re-created
utils.execute(
'nagios',
'execute_operation',
client,
parameters={
'node_ids': ['nagios'],
'operation': 'cloudify.interfaces.reconcile.monitoring',
'allow_kwargs_override': True,
},
)
# Confirm the expected workflow runs
# If the workflow already ran then it will already have healed so we won't
# see it run again now.
utils.wait_for_execution_on_deployment('heal',
'baseupdate',
client)
utils.remove_deployment('baseupdate', client)
utils.delete_blueprint('baseupdate', client)
utils.remove_nagios(client)
utils.delete_blueprint('nagiosupdate2', client)
utils.delete_blueprint('nagiosupdate3', client)
utils.delete_plugins(installed_plugins, client)
utils.remove_config_secrets(client)
main_client.tenants.delete(tenant)