On 6/7/22 14:32, Harald Jensas wrote:
On 6/7/22 11:49, Anirudh Gupta wrote:
Hi Harald/Julia/Openstack Team
I had successfully installed Ironic Service in Tripleo Train, Ussuri Release with the help of openstack discuss team members.
https://lists.openstack.org/pipermail/openstack-discuss/2022-February/027047... <https://lists.openstack.org/pipermail/openstack-discuss/2022-February/027047.html>
I had custom network configuration for which I had passed Service net map as below in network-environment file.
parameter_defaults: ServiceNetMap: IronicApiNetwork: oc_provisioning IronicNetwork: oc_provisioning
But with the Wallaby Release, since the deployment format has changed There is no network-environment file.
So I tried passing it in a separate environment.yaml file along with the templates. Command to deploy in wallaby:
openstack overcloud deploy --templates \ --networks-file /home/stack/templates/custom_network_data.yaml \ --vip-file /home/stack/templates/custom_vip_data.yaml \ --baremetal-deployment /home/stack/templates/overcloud-baremetal-deploy.yaml --deployed-server \ --network-config \ -e /home/stack/templates/environment.yaml \ -e /usr/share/openstack-tripleo-heat-templates/environments/services/ironic-conductor.yaml \ -e /usr/share/openstack-tripleo-heat-templates/environments/services/ironic-inspector.yaml \ -e /usr/share/openstack-tripleo-heat-templates/environments/services/ironic-overcloud.yaml \ -e /home/stack/templates/ironic-config.yaml \ -e /usr/share/openstack-tripleo-heat-templates/environments/services/ptp.yaml \ -e /usr/share/openstack-tripleo-heat-templates/environments/docker-ha.yaml \ -e /usr/share/openstack-tripleo-heat-templates/environments/podman.yaml \ -e /home/stack/containers-prepare-parameter.yaml
Without Service Net Map, deployment was successful but TFTP of image was an issue since we are going with custom network approach.
Passing Service Net map details in a seperate file environment.yaml, is giving the below error:
24d-f43f-6f76-000000006ce9 | TASK | Create containers managed by Podman for /var/lib/tripleo-config/container-startup-config/step_3 2022-06-07 11:16:46.806982 | | WARNING | ERROR: Can't run container ironic_db_sync stderr: Error: statfs /var/lib/config-data/puppet-generated/ironic_api: no such file or directory 2022-06-07 11:16:46.809252 | 525400fd-f24d-f43f-6f76-000000006c79 | FATAL | Create containers managed by Podman for /var/lib/tripleo-config/container-startup-config/step_3 | overcloud-controller-1 | error={"changed": false, "msg": "Failed containers: ironic_db_sync"} 2022-06-07 11:16:46.810197 | 525400fd-f24d-f43f-6f76-000000006c79 | TIMING | tripleo_container_manage : Create containers managed by Podman for /var/lib/tripleo-config/container-startup-config/step_3 | overcloud-controller-1 | 0:11:40.574944 | 10.62s 2022-06-07 11:16:47.683400 | | WARNING | ERROR: Can't run container ironic_db_sync stderr: Error: statfs /var/lib/config-data/puppet-generated/ironic_api: no such file or directory 2022-06-07 11:16:47.684455 | 525400fd-f24d-f43f-6f76-000000006ce9 | FATAL | Create containers managed by Podman for /var/lib/tripleo-config/container-startup-config/step_3 | overcloud-controller-2 | error={"changed": false, "msg": "Failed containers: ironic_db_sync"}
Can you please suggest the place we can pass the service net map details in Wallaby Release to resolve this issue?
You should be able to pass it in any of the environment files, like you did. Can you try adding the following to the environment file where you override the ServiceNetMap?
parameter_merge_strategies: ServiceNetMap: merge
Actually, my bad. (Thanks Rabi for reminding me on IRC.) The merge strategies is after a change in master. With Wallaby we still have ServiceNetMapDefaults which is merged with overrides in ServiceNetMap. AFICT from this error ...
2022-06-07 11:16:47.683400 | | WARNING | ERROR: Can't run container ironic_db_sync stderr: Error: statfs /var/lib/config-data/puppet-generated/ironic_api: no such file or directory
... it did not execute $THT/common/container-puppet.sh for the ironic_api service correctly. Could it be because you don't use a custom roles file? The default[1] role file does not have the `oc_provisioning` provisioning network. I wonder if some task is skipped because the node is not on the network the service is mapped to. Can you try using a custom roles file with the `oc_provisioning` network defined on the controller role? I think we probably need more information to understand the problem. Would it be possible to share more data? i.e full deploy log, all custom environment files? Thanks! [1] https://opendev.org/openstack/tripleo-heat-templates/src/branch/stable/walla...