Description of problem: variable for etcd_img has changed between 3.9.14-1 and 3.9.33, due to this installation fails using existing "working" inventory Line 190 of `/usr/share/ansible/openshift-ansible/playbooks/cluster- operator/aws/roles/openshift_health_checker/openshift_checks/docker_ima ge_availability.py` Version: 3.9.14-1 etcd_img = self.get_var("osm_etcd_image", default="registry.access.redhat.com/rhel7/etcd") Version: 3.9.33 etcd_img = self.get_var("etcd_image", default="registry.access.redhat.com/rhel7/etcd") Version-Release number of the following components: openshift-ansible-3.9.33-1.git.56.19ba16e.el7 How reproducible: Steps to Reproduce: 1. Using "working" inventory that installed 3.9.14 and set "osm_etcd_image" try to install version 3.9.33 2. image availability checks fail Actual results: image availability check fails Expected results: Installation to succeed Additional info: No change to this variable is in release notes for 3.9.33 Looks like this change was made upstream https://github.com/openshift/openshift-ansible/commit/1f1eb5ddd526b48fbf3351ac88cf276215cf4db4
Vadim, I think we need to refactor this so that the osm_etcd_image variable continues to work as expected since it's documented in the example inventories.
Please attach the inventory and full `ansible-playbook -vvv` output. This change would only affect default settings, so `osm_etcd_image` would be used for Enterprise setup. If `etcd_image` is set, it would override any of those.
Created PR https://github.com/openshift/openshift-ansible/pull/9461 to revert this commit - it was introduced to fix CI failures, so it should be reverted if it breaks existing inventories
Seem like this bug is already fixed.
Verified this bug with openshift-ansible-3.9.43-1.git.0.d0bc600.el7.noarch, and PASS. Trigger an container install, and set openshift_image_tag=v3.9.33 and osm_etcd_image, pre-install is passed. CHECK [docker_image_availability : host-8-252-211.host.centralci.eng.rdu2.redhat.com] *** CHECK [memory_availability : host-8-252-211.host.centralci.eng.rdu2.redhat.com] *** changed: [host-8-252-211.host.centralci.eng.rdu2.redhat.com] => {"changed": true, "checks": {"disk_availability": {}, "docker_image_availability": {"changed": true}, "docker_storage": {}, "memory_availability": {}, "package_availability": {"skipped": true, "skipped_reason": "Not active for this host"}, "package_version": {"skipped": true, "skipped_reason": "Not active for this host"}}, "failed": false, "playbook_context": "install"} Actually this is already fixed in present release version - 3.9.41.
Since the problem described in this bug report should be resolved in a recent advisory, it has been closed with a resolution of ERRATA. For information on the advisory, and where to find the updated files, follow the link below. If the solution does not work for you, open a new bug report. https://access.redhat.com/errata/RHBA-2018:2658