0195b441c7
For train branch, In upstream we have non-Ha by default but when HA is enabled (Like in Downstream we have HA by default). In multinode-1ctlr-featureset010 job: tempest fails when creating networks as neutron components fails to reach OVN VIP becasue there were no ovn-dbs resource in pcmk *nor* its VIP. ~~~ ERROR ovsdbapp.backend.ovs_idl.idlutils [-] Unable to open stream to tcp:192.168.24.19:6641 to retrieve schema: No route to host ~~~ OS::TripleO::Services::OVNDBs: ../deployment/ovn/ovn-dbs-pacemaker-puppet.yaml from environments/docker-ha.yaml[1] got overridden by latter OS::TripleO::Services::OVNDBs: ../../deployment/ovn/ovn-dbs-container-puppet.yaml in multinode-containers.yaml[2] With this patch we are removing OS::TripleO::Services::OVNDBs from multinode-containers.yaml[1] in train branch, deployment will take the defaults already present in overcloud-resource-registry-puppet.j2.yaml and both (non-ha train) upstream and (ha )downstream ci will be happy. Upstream ussuri+ already have HA as default[3][4] so its not conflicting. [1] https://github.com/openstack/tripleo-heat-templates/blob/stable/train/environments/docker-ha.yaml#L24 [2] https://github.com/openstack/tripleo-heat-templates/blob/stable/train/ci/environments/multinode-containers.yaml#L7 [3] https://review.opendev.org/#/c/359060/ [4] https://github.com/openstack/tripleo-heat-templates/blob/stable/ussuri/ci/environments/multinode-containers.yaml#L7 Change-Id: I53360b49e39578b7b9ec970a94a9b3460b2ac351 |
||
---|---|---|
.. | ||
common | ||
environments | ||
scripts | ||
README.rst |
README.rst
TripleO CI environments
TripleO CI environments are exclusively used for Continuous Integration purpose or for development usage. They should not be used in production and we don't guarantee they work outside TripleO CI.
For more informations about TripleO CI, please look: https://github.com/openstack-infra/tripleo-ci