CORE-1336 Support to skip instances != ACTIVE

This commit adds a new module, nova_facts, in order to determine the
state of an instane prior to beginning, which results in the instance
being skipped until the rebuild step.

Change-Id: Iade8103349b83973eaedbb11e26b5036e970dbd5
This commit is contained in:
Julia Kreger 2014-08-25 14:30:38 -04:00
parent 770d76f171
commit 8c4660b2f9
5 changed files with 251 additions and 19 deletions
library/cloud
playbooks
plugins/inventory

185
library/cloud/nova_facts Normal file

@ -0,0 +1,185 @@
#!/usr/bin/env python
# This code is part of Ansible, but is an independent component.
# This particular file snippet, and this file snippet only, is BSD licensed.
# Modules you write using this snippet, which is embedded dynamically by Ansible
# still belong to the author of the module, and may assign their own license
# to the complete work.
#
# Copyright (c) 2014 Hewlett-Packard Development Company, L.P.
# All rights reserved.
#
# Redistribution and use in source and binary forms, with or without modification,
# are permitted provided that the following conditions are met:
#
# * Redistributions of source code must retain the above copyright
# notice, this list of conditions and the following disclaimer.
# * Redistributions in binary form must reproduce the above copyright notice,
# this list of conditions and the following disclaimer in the documentation
# and/or other materials provided with the distribution.
#
# THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS" AND
# ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE IMPLIED
# WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE DISCLAIMED.
# IN NO EVENT SHALL THE COPYRIGHT HOLDER OR CONTRIBUTORS BE LIABLE FOR ANY DIRECT, INDIRECT,
# INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO,
# PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS
# INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT
# LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE
# USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.
import os
try:
from novaclient.v1_1 import client as nova_client
from novaclient import exceptions
import time
except ImportError:
print("failed=True msg='novaclient is required for this module'")
DOCUMENTATION = '''
---
module: nova_facts
short_description: Return facts from Nova
description:
- Returns instance state and metadata from Openstack Nova.
options:
login_username:
description:
- login username to authenticate to keystone
required: true
default: admin
login_password:
description:
- Password of login user
required: true
default: 'yes'
login_tenant_name:
description:
- The tenant name of the login user
required: true
default: 'yes'
auth_url:
description:
- The keystone url for authentication
required: false
default: 'http://127.0.0.1:35357/v2.0/'
region_name:
description:
- Name of the region
required: false
default: None
name:
description:
- Name of the instance for which facts will be retrieved.
default: None
instance_id:
description:
- Instance ID of the instance for which facts will be retrieved.
default: None
requirements: ["novaclient"]
'''
EXAMPLES = '''
# Rebuilds an existing VM with a new image
- nova_facts:
login_username: admin
login_password: admin
login_tenant_name: admin
name: vm1
image_id: 4f905f38-e52a-43d2-b6ec-754a13ffb529
'''
# The following openstack_ is a copy paste from an upcoming
# core module "lib/ansible/module_utils/openstack.py" Once that's landed,
# these should be replaced with a line at the bottom of the file:
# from ansible.module_utils.openstack import *
def openstack_argument_spec():
# Consume standard OpenStack environment variables.
# This is mainly only useful for ad-hoc command line operation as
# in playbooks one would assume variables would be used appropriately
OS_AUTH_URL = os.environ.get('OS_AUTH_URL', 'http://127.0.0.1:35357/v2.0/')
OS_PASSWORD = os.environ.get('OS_PASSWORD', None)
OS_REGION_NAME = os.environ.get('OS_REGION_NAME', None)
OS_USERNAME = os.environ.get('OS_USERNAME', 'admin')
OS_TENANT_NAME = os.environ.get('OS_TENANT_NAME', OS_USERNAME)
spec = dict(
login_username=dict(default=OS_USERNAME),
auth_url=dict(default=OS_AUTH_URL),
region_name=dict(default=OS_REGION_NAME),
availability_zone=dict(default=None),
)
if OS_PASSWORD:
spec['login_password'] = dict(default=OS_PASSWORD)
else:
spec['login_password'] = dict(required=True)
if OS_TENANT_NAME:
spec['login_tenant_name'] = dict(default=OS_TENANT_NAME)
else:
spec['login_tenant_name'] = dict(required=True)
return spec
def _nova_facts(module, nova):
server = None
try:
if module.params.get('instance_id') is None:
servers = nova.servers.list(True, {'name': module.params['name']})
if servers:
# the {'name': module.params['name']} will also return servers
# with names that partially match the server name, so we have to
# strictly filter here
servers = [
x for x in servers if x.name == module.params['name']
]
if servers:
server = servers[0]
else:
server = nova.servers.get(module.params['instance_id'])
except Exception, e:
module.fail_json(
msg="Error in getting the server list: %s" % e.message
)
if not server:
module.exit_json(changed=False, result="not present")
try:
server = nova.servers.get(server.id)
except Exception, e:
module.fail_json(msg="not present")
facts = {'ansible_facts': {}}
facts['ansible_facts']['instance_status'] = server.status
facts['ansible_facts']['instance_metadata'] = server.metadata
module.exit_json(**facts)
def main():
argument_spec = openstack_argument_spec()
argument_spec.update(dict(
name=dict(),
instance_id=dict(),
))
module = AnsibleModule(argument_spec=argument_spec)
nova = nova_client.Client(module.params['login_username'],
module.params['login_password'],
module.params['login_tenant_name'],
module.params['auth_url'],
region_name=module.params['region_name'],
service_type='compute')
try:
nova.authenticate()
except exceptions.Unauthorized, e:
module.fail_json(
msg="Invalid OpenStack Nova credentials.: %s" % e.message
)
except exceptions.AuthorizationFailure, e:
module.fail_json(msg="Unable to authorize user: %s" % e.message)
_nova_facts(module, nova)
# this is magic, see lib/ansible/module_common.py
from ansible.module_utils.basic import *
main()

@ -13,9 +13,6 @@
# See the License for the specific language governing permissions and
# limitations under the License.
---
- name: "Change /mnt to be a read-only mount-point"
sudo: yes
command: mount -o remount,ro /mnt
- name: "Call nova_rebuild"
nova_rebuild:
instance_id: "{{ instance_id }}"

18
playbooks/remount-ro.yml Normal file

@ -0,0 +1,18 @@
# Copyright (c) 2014 Hewlett-Packard Development Company, L.P.
#
# Licensed under the Apache License, Version 2.0 (the "License");
# you may not use this file except in compliance with the License.
# You may obtain a copy of the License at
#
# http://www.apache.org/licenses/LICENSE-2.0
#
# Unless required by applicable law or agreed to in writing, software
# distributed under the License is distributed on an "AS IS" BASIS,
# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or
# implied.
# See the License for the specific language governing permissions and
# limitations under the License.
---
- name: "Change /mnt to be a read-only mount-point"
sudo: yes
command: mount -o remount,ro /mnt

@ -14,52 +14,62 @@
# limitations under the License.
---
- hosts: all
gather_facts: no
tasks:
- include: disable_os_collect_config.yml
when: instance_status == "ACTIVE"
- hosts: undercloud
name: Disable Undercloud
sudo: yes
gather_facts: no
tasks:
- service: name={{ item }} enabled=no state=stopped
with_items: helion_undercloud_services
when: helion is defined
when: helion is defined and instance_status == "ACTIVE"
- service: name={{ item }} enabled=no state=stopped
with_items: undercloud_services
when: helion is not defined
when: helion is not defined and instance_status == "ACTIVE"
- hosts: nova-compute
name: Disable Overcloud Compute
sudo: yes
gather_facts: no
tasks:
- service: name={{ item }} enabled=no state=stopped
with_items: helion_overcloud_compute_services
when: helion is defined
when: helion is defined and instance_status == "ACTIVE"
- service: name={{ item }} enabled=no state=stopped
with_items: overcloud_compute_services
with_items: overcloud_compute_services and instance_status == "ACTIVE"
when: helion is not defined
- hosts: swift-storage
name: swift-storage
sudo: yes
gather_facts: no
tasks:
- service: name={{ item }} enabled=no state=stopped
with_items: helion_overcloud_swift_services
when: helion is defined
when: helion is defined and instance_status == "ACTIVE"
- service: name={{ item }} enabled=no state=stopped
with_items: overcloud_swift_services
when: helion is not defined and instance_status == "ACTIVE"
- hosts: controller
name: Disable Overcloud Controller
sudo: yes
gather_facts: no
tasks:
- service: name={{ item }} enabled=no state=stopped
with_items: helion_overcloud_controller_services
when: helion is defined
when: helion is defined and instance_status == "ACTIVE"
- service: name={{ item }} enabled=no state=stopped
with_items: overcloud_controller_services
when: helion is not defined
when: helion is not defined and instance_status == "ACTIVE"
- hosts: controllermgmt
name: Disable Overcloud Controller Mgmt node
sudo: yes
gather_facts: no
tasks:
- service: name={{ item }} enabled=no state=stopped
with_items: helion_overcloudmgmt_controller_services
when: helion is defined
when: helion is defined and instance_status == "ACTIVE"
# Critically, we need to select a single node of the galera cluster to
# be the 'last'. So controllermgmt fits that bill for now. We will have
# to select one to be the "special" node eventually, we can do that with
@ -68,26 +78,36 @@
- hosts: controller
name: Stop MySQL on controller nodes
sudo: yes
gather_facts: no
tasks:
- include: galera_status.yml
when: instance_status == "ACTIVE"
- service: name=mysql enabled=no state=stopped
when: galera_status == 'Synced'
when: galera_status == 'Synced' and instance_status == "ACTIVE"
- fail: msg="Galera Replication is out of sync - cannot safely proceed"
when: galera_status != 'Synced'
when: galera_status != 'Synced' and instance_status == "ACTIVE"
- hosts: controllermgmt
name: Stop MySQL on Overcloud Controller Mgmt node
sudo: yes
gather_facts: no
tasks:
- include: galera_status.yml
when: instance_status == "ACTIVE"
- fail: msg="Galera Replication is out of sync - cannot safely proceed"
when: galera_status == 'Out of Sync'
when: galera_status == 'Out of Sync' and instance_status == "ACTIVE"
- service: name=mysql enabled=no state=stopped
when: instance_status == "ACTIVE"
- hosts: undercloud
name: Rebuild and Refresh Undercloud
gather_facts: no
tasks:
- include: preserve_ssh_host_keys.yml
when: instance_status == "ACTIVE"
- include: cleanup_cinder_volumes.yml
- { include: rebuild.yml, instance_id: "{{ instance_id }}", rebuild_image_id: "{{ undercloud_rebuild_image_id }}" }
when: instance_status == "ACTIVE"
- include: remount-ro.yml
when: instance_status == "ACTIVE"
- { include: rebuild.yml, instance_id: "{{ instance_id }}", rebuild_image_id: "{{ undercloud_rebuild_image_id }}", when: instance_status != "REBUILD" }
- local_action: wait_for port=22 timeout=600 host="{{ inventory_hostname }}" search_regex=OpenSSH delay=10
- include: refresh_config.yml
- hosts: undercloud
@ -99,10 +119,13 @@
with_items: undercloud_services
- hosts: controllermgmt
name: Rebuild and Refresh ControllerMgmt
gather_facts: no
tasks:
- include: preserve_ssh_host_keys.yml
when: instance_status == "ACTIVE"
- include: cleanup_cinder_volumes.yml
- { include: rebuild.yml, instance_id: "{{ instance_id }}", rebuild_image_id: "{{ controllermgmt_rebuild_image_id }}" }
when: instance_status == "ACTIVE"
- { include: rebuild.yml, instance_id: "{{ instance_id }}", rebuild_image_id: "{{ controllermgmt_rebuild_image_id }}", when: instance_status != "REBUILD" }
- local_action: wait_for port=22 timeout=600 host="{{ inventory_hostname }}" search_regex=OpenSSH delay=10
- hosts: controllermgmt
name: Start initial Galera cluster node
@ -110,10 +133,13 @@
- service: name=mysql enabled=yes state=started arguments=--wsrep_cluster_address="gcomm://"
- hosts: controller
name: Rebuild and Refresh Controller
gather_facts: no
tasks:
- include: preserve_ssh_host_keys.yml
when: instance_status == "ACTIVE"
- include: cleanup_cinder_volumes.yml
- { include: rebuild.yml, instance_id: "{{ instance_id }}", rebuild_image_id: "{{ controller_rebuild_image_id }}" }
when: instance_status == "ACTIVE"
- { include: rebuild.yml, instance_id: "{{ instance_id }}", rebuild_image_id: "{{ controller_rebuild_image_id }}", when: instance_status != "REBUILD" }
- local_action: wait_for port=22 timeout=600 host="{{ inventory_hostname }}" search_regex=OpenSSH delay=10
- hosts: controller:controllermgmt
name: Re-Refresh Controller
@ -146,9 +172,11 @@
with_items: overcloud_controller_services
- hosts: swift-storage
name: Rebuild and Refresh swift-storage
gather_facts: no
tasks:
- include: preserve_ssh_host_keys.yml
- { include: rebuild.yml, instance_id: "{{ instance_id }}", rebuild_image_id: "{{ swift_storage_rebuild_image_id }}" }
when: instance_status == "ACTIVE"
- { include: rebuild.yml, instance_id: "{{ instance_id }}", rebuild_image_id: "{{ swift_storage_rebuild_image_id }}", when: instance_status != "REBUILD" }
- local_action: wait_for port=22 timeout=600 host="{{ inventory_hostname }}" search_regex=OpenSSH delay=10
- include: refresh_config.yml
- hosts: swift-storage
@ -163,9 +191,11 @@
with_items: overcloud_swift_services
- hosts: nova-compute
name: Rebuild and Refresh Nova Compute
gather_facts: no
tasks:
- include: preserve_ssh_host_keys.yml
- { include: rebuild.yml, instance_id: "{{ instance_id }}", rebuild_image_id: "{{ nova_compute_rebuild_image_id }}" }
when: instance_status == "ACTIVE"
- { include: rebuild.yml, instance_id: "{{ instance_id }}", rebuild_image_id: "{{ nova_compute_rebuild_image_id }}", when: instance_status != "REBUILD" }
- local_action: wait_for port=22 timeout=600 host="{{ inventory_hostname }}" search_regex=OpenSSH delay=10
- include: refresh_config.yml
- hosts: nova-compute

@ -169,6 +169,7 @@ class HeatInventory(object):
res.resource_name
)
}
hostvars[addr]['instance_status'] = server.status
hostvars[addr]['instance_id'] = res.physical_resource_id
inventory = {'_meta': {'hostvars': hostvars}}
inventory.update(groups)
@ -195,6 +196,7 @@ class HeatInventory(object):
res.resource_name
)
}
hostvars['instance_status'] = server.status
hostvars['instance_id'] = res.physical_resource_id
print(json.dumps(hostvars, indent=2))
break