Bug 1661232
Summary: | [openstack] node-groups variables missmatch when using groups | ||
---|---|---|---|
Product: | OpenShift Container Platform | Reporter: | Eduardo Minguez <eminguez> |
Component: | Installer | Assignee: | Russell Teague <rteague> |
Installer sub component: | openshift-ansible | QA Contact: | Johnny Liu <jialiu> |
Status: | CLOSED WONTFIX | Docs Contact: | |
Severity: | unspecified | ||
Priority: | unspecified | CC: | aos-bugs, eduen, gpei, jokerman, mmccomas, pablo.iranzo, wmeng |
Version: | 3.11.0 | ||
Target Milestone: | --- | ||
Target Release: | 3.11.z | ||
Hardware: | Unspecified | ||
OS: | Unspecified | ||
Whiteboard: | |||
Fixed In Version: | Doc Type: | If docs needed, set a value | |
Doc Text: | Story Points: | --- | |
Clone Of: | Environment: | ||
Last Closed: | 2019-10-18 17:07:30 UTC | Type: | Bug |
Regression: | --- | Mount Type: | --- |
Documentation: | --- | CRM: | |
Verified Versions: | Category: | --- | |
oVirt Team: | --- | RHEL 7.3 requirements from Atomic Host: | |
Cloudforms Team: | --- | Target Upstream Version: | |
Embargoed: |
Description
Eduardo Minguez
2018-12-20 14:04:31 UTC
I *THINK* the issue is located here https://github.com/openshift/openshift-ansible/blob/master/playbooks/openstack/resources.py#L105-L106 I think this is somehow related to the heat stack... I've tried to run the provision playbook and canceling it while the "Check if the stack exists" and browsing the generated stack in the /tmp/XXX folder: # Per group (both openshift_node_group_name and openshift_openstack_*_group_name) $ grep -B2 node_group stack.yaml cluster_id: shiftstack.automated.lan type: master openshift_node_group_name: node-config-master -- type: node subtype: app openshift_node_group_name: node-config-compute -- type: node subtype: infra openshift_node_group_name: node-config-infra -- cluster_id: shiftstack.automated.lan type: cns openshift_node_group_name: node-config-compute # Per host $ grep -B2 node_group stack.yaml cluster_id: shiftstack.automated.lan type: master openshift_node_group_name: node-config-master -- type: node subtype: app openshift_node_group_name: node-config-compute -- type: node subtype: infra openshift_node_group_name: node-config-infra -- cluster_id: shiftstack.automated.lan type: cns openshift_node_group_name: node-config-compute I think the issue is the stack is created without taking into account the group_vars/{masters,openstack_*}.yml (as the nodes doesn't exist yet) but only the all.yml file... in this case, I've added the openshift_openstack_*_group_name to the all.yml file and now it seems to work: ... type: node subtype: app openshift_node_group_name: node-config-compute-crio image: rhel-server-7.6 ... So, I think the node_group_name should be specified in the all.yml to be taken into account for heat, but the other variables should be added in their corresponding group: all.yml: ... # Node group settings openshift_openstack_compute_group_name: node-config-compute-crio openshift_openstack_master_group_name: node-config-master-crio openshift_openstack_infra_group_name: node-config-infra $ for i in masters openstack_compute_nodes openstack_infra_nodes; do echo $i; cat inventory/group_vars/$i.yml; done masters openshift_use_crio_only: true openshift_use_crio: true openstack_compute_nodes openshift_use_crio_only: true openshift_use_crio: true openstack_infra_nodes openshift_use_crio_only: false openshift_use_crio: false I'm running a deployment with this settings now and will notify here the result. I can confirm that theory! Oddly enough, looking at the documentation you may have been the one to add the documentation that the openshift_openstack_*_group_name parameters would have to be specified in all.yml :) (In reply to Tzu-Mainn Chen from comment #5) > I can confirm that theory! Oddly enough, looking at the documentation you > may have been the one to add the documentation that the > openshift_openstack_*_group_name parameters would have to be specified in > all.yml :) PEBKAC... this scenario worked: all.yml: ... # Node group settings openshift_openstack_compute_group_name: node-config-compute-crio openshift_openstack_master_group_name: node-config-master-crio openshift_openstack_infra_group_name: node-config-infra $ for i in masters openstack_compute_nodes openstack_infra_nodes; do echo $i; cat inventory/group_vars/$i.yml; done masters openshift_use_crio_only: true openshift_use_crio: true openstack_compute_nodes openshift_use_crio_only: true openshift_use_crio: true openstack_infra_nodes openshift_use_crio_only: false openshift_use_crio: false I did created a PR to clarify it and to be merged in the release-3.11 branch here https://github.com/openshift/openshift-ansible/pull/10915 Thanks! |