We do not want cinder-volume to be managed by Pacemaker on
BlockStorage nodes, where Pacemaker is not running at all.
This change adds a new BlockStorageCinderVolume service name
which can (and is, by default) mapped to the non Pacemaker
implementation of the service.
The error was:
Could not find dependency Exec[wait-for-settle] for
Pacemaker::Resource::Systemd[openstack-cinder-volume]
Also moves cinder::host setting into the Pacemaker specific service
definition because we only want to set a shared host= string when
the service is managed by Pacemaker.
Closes-Bug: #
1628912
Change-Id: I2f7e82db4fdfd5f161e44d65d17893c3e19a89c9
OS::TripleO::Services::CinderBackup: OS::Heat::None
OS::TripleO::Services::CinderScheduler: puppet/services/cinder-scheduler.yaml
OS::TripleO::Services::CinderVolume: puppet/services/cinder-volume.yaml
+ OS::TripleO::Services::BlockStorageCinderVolume: puppet/services/cinder-volume.yaml
OS::TripleO::Services::Core: OS::Heat::None
OS::TripleO::Services::Keystone: puppet/services/keystone.yaml
OS::TripleO::Services::GlanceApi: puppet/services/glance-api.yaml
cinder::rabbit_password: {get_param: RabbitPassword}
cinder::rabbit_port: {get_param: RabbitClientPort}
cinder::rabbit_heartbeat_timeout_threshold: 60
- cinder::host: hostgroup
cinder::cron::db_purge::destination: '/dev/null'
cinder::db::database_db_max_retries: -1
cinder::db::database_max_retries: -1
- get_attr: [CinderVolumeBase, role_data, config_settings]
- cinder::volume::manage_service: false
cinder::volume::enabled: false
+ cinder::host: hostgroup
step_config:
include ::tripleo::profile::pacemaker::cinder::volume
- name: BlockStorage
ServicesDefault:
- OS::TripleO::Services::CACerts
- - OS::TripleO::Services::CinderVolume
+ - OS::TripleO::Services::BlockStorageCinderVolume
- OS::TripleO::Services::Kernel
- OS::TripleO::Services::Ntp
- OS::TripleO::Services::Timezone