tripleo-heat-templates/deployment
2019-06-20 16:45:14 +00:00
..
aide Move Aide to deployments 2019-03-28 08:24:40 -04:00
aodh Use RpcPort for container healthchecks 2019-05-10 09:24:44 -04:00
apache Disable a directory listing of /icons in httpd. 2019-04-01 14:14:47 +09:00
barbican Move containers-common.yaml into deployment 2019-04-14 18:15:12 -04:00
cavium Move containers-common.yaml into deployment 2019-04-14 18:15:12 -04:00
ceilometer ceilometer_agent_notification: disable-panko.yaml 2019-04-29 15:11:59 -04:00
ceph-ansible ceph-base: Update ceph-ansible default playbook 2019-05-30 19:33:08 +00:00
cinder Fix bare reference of cinder_enable_iscsi_backend 2019-06-17 14:15:26 +00:00
clients Include python-panko client. 2019-03-18 14:49:28 -04:00
container-image-prepare Add parameter ContainerImageRegistryCredentials 2019-04-08 12:44:54 +12:00
database ensure /var/run/redis is present upon reboot 2019-05-01 14:46:42 +00:00
deprecated Add panko_api_cron container 2019-06-10 18:25:06 -04:00
ec2 Move containers-common.yaml into deployment 2019-04-14 18:15:12 -04:00
etcd Set setype on etcd's service directory 2019-03-07 14:36:08 -05:00
experimental Move containers-common.yaml into deployment 2019-04-14 18:15:12 -04:00
glance Add more settings for glance image cache 2019-06-08 09:52:51 +00:00
gnocchi Move containers-common.yaml into deployment 2019-04-14 18:15:12 -04:00
haproxy Fix haproxy stats network binding 2019-05-31 06:11:16 +00:00
heat Activate health checks for cron containers 2019-05-20 11:08:19 +00:00
horizon Re-enable manila dashboard 2019-06-11 16:02:02 +00:00
image-serve We have to allow httpd to listen on those ports in some cases. 2019-03-21 08:57:59 +01:00
ironic Add healthcheck to ironic_pxe_http 2019-06-06 16:53:48 +00:00
iscsid Disable iscsi.service to avoid iscsid on host from getting started 2019-06-18 08:46:58 +09:00
keepalived Move containers-common.yaml into deployment 2019-04-14 18:15:12 -04:00
kernel Set arp_notify to match ndisc_notify 2019-05-20 20:58:54 +00:00
keystone Activate health checks for cron containers 2019-05-20 11:08:19 +00:00
logging Change setype for non container /var/log/nova-placement 2019-05-02 11:08:21 -06:00
login-defs Move login-defs to deployment directory 2019-01-22 13:48:57 -07:00
logrotate Activate health checks for cron containers 2019-05-20 11:08:19 +00:00
manila Move Manila backends into deployment 2019-04-24 22:15:05 +00:00
memcached Move containers-common.yaml into deployment 2019-04-14 18:15:12 -04:00
messaging Move containers-common.yaml into deployment 2019-04-14 18:15:12 -04:00
metrics Avoid issues with non-existing directories 2019-05-03 08:08:18 +02:00
mistral Copy keys for tripleo-admin user 2019-05-07 16:44:19 +00:00
multipathd Move containers-common.yaml into deployment 2019-04-14 18:15:12 -04:00
neutron Configure Neutron API for Nova Placement... 2019-06-06 02:22:04 +00:00
nova Fix NovaNfs role parameters 2019-06-14 15:10:31 +02:00
octavia Merge "Configure server_certs_key_passphrase for Octavia" into stable/stein 2019-06-13 02:40:00 +00:00
ovn Override ovn::controller::hostname to use hiera:fqdn_canonical 2019-06-20 12:49:10 +05:30
pacemaker Move containers-common.yaml into deployment 2019-04-14 18:15:12 -04:00
podman Clean up leftover mount point after docker stop. 2019-05-07 10:29:18 +02:00
qdr Move containers-common.yaml into deployment 2019-04-14 18:15:12 -04:00
rabbitmq Move containers-common.yaml into deployment 2019-04-14 18:15:12 -04:00
sahara Use RpcPort for container healthchecks 2019-05-10 09:24:44 -04:00
securetty Move securetty to deployment dir 2019-01-22 13:45:40 -07:00
selinux Move selinux to deployment folder 2019-01-22 13:21:13 -07:00
snmp Snmp - Use net_cidr_map for firewall rules 2019-01-06 18:21:54 +01:00
sshd Avoid dangling firewall rule for ssh access 2019-01-22 14:49:46 +01:00
swift Evaluating ansible_check_mode as a boolean 2019-04-22 12:48:32 +00:00
tacker Move containers-common.yaml into deployment 2019-04-14 18:15:12 -04:00
time rename rsyslogd service to rsyslog and deal with lack of reload. 2019-03-15 14:55:00 +01:00
timesync Fix service naming conventions 2018-12-13 08:12:36 -07:00
tripleo-firewall Initialize ip(6)tables "raw" table 2019-05-28 17:21:32 +00:00
tripleo-packages Put on hold the Ansible package update until is fixed 2019-06-17 12:19:22 +02:00
tuned flatten tuned service configuration 2018-12-14 13:13:40 -05:00
undercloud Put on hold the Ansible package update until is fixed 2019-06-17 12:19:22 +02:00
zaqar Enable zaqar healthchecks 2019-05-09 13:26:30 +02:00
containers-common.yaml Use RpcPort for container healthchecks 2019-05-10 09:24:44 -04:00
README.rst Update/combine docker/services/README.rst 2019-04-14 18:15:14 -04:00

TripleO Deployments

This directory contains files that represent individual service deployments, orchestration tools, and the configuration tools used to deploy them.

Directory Structure

Each logical grouping of services will have a directory. Example: 'timesync'. Within this directory related timesync services would exist to for example configure timesync services on baremetal or via containers.

Filenaming conventions

As a convention each deployments service filename will reflect both the deployment engine (baremetal, or containers) along with the config tool used to deploy that service.

The convention is <service-name>-<engine>-<config management tool>.

Examples:

deployment/aodh/aodh-api-container-puppet.yaml (containerized Aodh service configured with Puppet)

deployment/aodh/aodh-api-container-ansible.yaml (containerized Aodh service configured with Ansible)

deployment/timesync/chrony-baremetal-ansible.yaml (baremetal Chrony service configured with Ansible)

deployment/timesync/chrony-baremetal-puppet.yaml (baremetal Chrony service configured with Puppet)

Building Kolla Images

TripleO currently relies on Kolla(Dockerfile) containers. Kolla supports container customization and we are making use of this feature within TripleO to inject puppet (our configuration tool of choice) into the Kolla base images. A variety of other customizations are being made via the tripleo-common/container-images/tripleo_kolla_template_overrides.j2 file.

To build Kolla images for TripleO adjust your kolla config1 to build your centos base image with puppet using the example below:

$ cat template-overrides.j2 {% extends parent_template %} {% set base_centos_binary_packages_append = ['puppet'] %} {% set nova_scheduler_packages_append = ['openstack-tripleo-common'] %}

kolla-build --base centos --template-override template-overrides.j2

Containerized Deployment Template Structure

Each deployment template may define a set of output values control the underlying service deployment in a variety of ways. These output sections are specific to the TripleO deployment architecture. The following sections are available for containerized services.

  • config_settings: This section contains service specific hiera data can be used to generate config files for each service. This data is ultimately processed via the container-puppet.py tool which generates config files for each service according to the settings here.

  • kolla_config: Contains YAML that represents how to map config files into the kolla container. This config file is typically mapped into the container itself at the /var/lib/kolla/config_files/config.json location and drives how kolla's external config mechanisms work.

  • docker_config: Data that is passed to paunch tool to configure a container, or step of containers at each step. See the available steps documented below which are implemented by TripleO's cluster deployment architecture.

  • puppet_config: This section is a nested set of key value pairs that drive the creation of config files using puppet. Required parameters include:

    • puppet_tags: Puppet resource tag names that are used to generate config files with puppet. Only the named config resources are used to generate a config file. Any service that specifies tags will have the default tags of 'file,concat,file_line,augeas,cron' appended to the setting. Example: keystone_config
    • config_volume: The name of the volume (directory) where config files will be generated for this service. Use this as the location to bind mount into the running Kolla container for configuration.
    • config_image: The name of the container image that will be used for generating configuration files. This is often the same container that the runtime service uses. Some services share a common set of config files which are generated in a common base container.
    • step_config: This setting controls the manifest that is used to create container config files via puppet. The puppet tags below are used along with this manifest to generate a config directory for this container.
  • container_puppet_tasks: This section provides data to drive the container-puppet.py tool directly. The task is executed only once within the cluster (not on each node) and is useful for several puppet snippets we require for initialization of things like keystone endpoints, database users, etc. See container-puppet.py for formatting.

Deployment steps

Similar to baremetal containers are brought up in a stepwise manner. The current architecture supports bringing up baremetal services alongside of containers. For each step the baremetal puppet manifests are executed first and then any containers are brought up afterwards.

Steps correlate to the following:

Pre) Containers config files generated per hiera settings. 1) Load Balancer configuration baremetal a) step 1 baremetal b) step 1 containers 2) Core Services (Database/Rabbit/NTP/etc.) a) step 2 baremetal b) step 2 containers 3) Early Openstack Service setup (Ringbuilder, etc.) a) step 3 baremetal b) step 3 containers 4) General OpenStack Services a) step 4 baremetal b) step 4 containers c) Keystone containers post initialization (tenant,service,endpoint creation) 5) Service activation (Pacemaker), online data migration a) step 5 baremetal b) step 5 containers

Update steps:

All services have an associated update_tasks output that is an ansible snippet that will be run during update in an rolling update that is expected to run in a rolling update fashion (one node at a time)

For Controller (where pacemaker is running) we have the following states:
  1. Step=1: stop the cluster on the updated node;
  2. Step=2: Pull the latest image and retag the it pcmklatest
  3. Step=3: yum upgrade happens on the host.
  4. Step=4: Restart the cluster on the node
  5. Step=5: Verification: Currently we test that the pacemaker services are running.

Then the usual deploy steps are run which pull in the latest image for all containerized services and the updated configuration if any.

Note: as pacemaker is not containerized, the points 1 and 4 happen in puppet/services/pacemaker.yaml.

Fast-forward Upgrade Steps

Each service template may optionally define a fast_forward_upgrade_tasks key, which is a list of Ansible tasks to be performed during the fast-forward upgrade process. As with Upgrade steps each task is associated to a particular step provided as a variable and used along with a release variable by a basic conditional that determines when the task should run.

Steps are broken down into two categories, prep tasks executed across all hosts and bootstrap tasks executed on a single host for a given role.

The individual steps then correspond to the following tasks during the upgrade:

Prep steps:

  • Step=0: Check running services
  • Step=1: Stop the service
  • Step=2: Stop the cluster
  • Step=3: Update repos

Bootstrap steps:

  • Step=4: DB backups
  • Step=5: Pre package update commands
  • Step=6: Package updates
  • Step=7: Post package update commands
  • Step=8: DB syncs
  • Step=9: Verification

  1. See the override file which can be used to build Kolla packages that work with TripleO.↩︎