Description of problem: Undeploy prometheus 3.9, throws out error playbooks/openshift-prometheus/private/uninstall_prometheus.yaml is missing TASK [openshift_prometheus : include_tasks] ********************************************************************************************************************************************* task path: /usr/share/ansible/openshift-ansible/roles/openshift_prometheus/tasks/main.yaml:28 fatal: [ec2-54-158-112-3.compute-1.amazonaws.com]: FAILED! => { "reason": "Unable to retrieve file contents\nCould not find or access '/usr/share/ansible/openshift-ansible/playbooks/openshift-prometheus/private/uninstall_prometheus.yaml'" } Version-Release number of selected component (if applicable): $ rpm -qa | grep openshift-ansible openshift-ansible-docs-3.9.0-0.34.0.git.0.c7d9585.el7.noarch openshift-ansible-playbooks-3.9.0-0.34.0.git.0.c7d9585.el7.noarch openshift-ansible-3.9.0-0.34.0.git.0.c7d9585.el7.noarch openshift-ansible-roles-3.9.0-0.34.0.git.0.c7d9585.el7.noarch How reproducible: Always Steps to Reproduce: 1. Undeploy prometheus 3.9 2. 3. Actual results: playbooks/openshift-prometheus/private/uninstall_prometheus.yaml is missing Expected results: Could be undeployed successfully Additional info:
This was merged to master last week: https://github.com/openshift/openshift-ansible/pull/6811 Scott, Is master targeted to 3.10, or does it need a special label?
Master still hasn't forked. I moved this bug's target release to 3.10 because I didn't htink it was critical enough to block 3.9 release, but since you say it's fixed on master moving it back to 3.9.
Still the same error TASK [openshift_prometheus : include_tasks] ********************************************************************************* task path: /usr/share/ansible/openshift-ansible/roles/openshift_prometheus/tasks/main.yaml:28 fatal: [host-8-245-46.host.centralci.eng.rdu2.redhat.com]: FAILED! => { "reason": "Unable to retrieve file contents\nCould not find or access '/usr/share/ansible/openshift-ansible/playbooks/openshift-prometheus/private/uninstall_prometheus.yaml'" } to retry, use: --limit @/usr/share/ansible/openshift-ansible/playbooks/openshift-prometheus/config.retry ansible version # rpm -qa | grep openshift-ansible openshift-ansible-playbooks-3.9.0-0.39.0.git.0.fea6997.el7.noarch openshift-ansible-roles-3.9.0-0.39.0.git.0.fea6997.el7.noarch openshift-ansible-3.9.0-0.39.0.git.0.fea6997.el7.noarch openshift-ansible-docs-3.9.0-0.39.0.git.0.fea6997.el7.noarch
The name of the uninstall playbook is: playbooks/openshift-prometheus/private/uninstall.yml And actually, the one that should be called is: playbooks/openshift-prometheus/uninstall.yml which calls the private playbook. The name aligns with config.yml (which is not called config_prometheus.yaml) I believe this should resolve the issue. Please verify
# cat playbooks/openshift-prometheus/private/uninstall.yml --- - name: Uninstall Prometheus hosts: masters[0] tasks: - name: Run the Prometheus Uninstall Role Tasks include_role: name: openshift_prometheus tasks_from: uninstall ************************************************** # cat playbooks/openshift-prometheus/uninstall.yml --- - import_playbook: private/uninstall.yml ************************************************** # ls -alR playbooks/openshift-prometheus playbooks/openshift-prometheus: total 16 drwxr-xr-x. 3 root root 80 Feb 6 23:13 . drwxr-xr-x. 27 root root 4096 Feb 6 21:22 .. -rw-r--r--. 1 root root 49 Feb 6 23:07 config.retry -rw-r--r--. 1 root root 79 Feb 6 17:20 config.yml drwxr-xr-x. 2 root root 58 Feb 6 23:13 private -rw-r--r--. 1 root root 45 Feb 6 17:20 uninstall.yml playbooks/openshift-prometheus/private: total 8 drwxr-xr-x. 2 root root 58 Feb 6 23:13 . drwxr-xr-x. 3 root root 80 Feb 6 23:13 .. -rw-r--r--. 1 root root 714 Feb 6 17:20 config.yml lrwxrwxrwx. 1 root root 14 Feb 6 21:22 roles -> ../../../roles -rw-r--r--. 1 root root 191 Feb 6 17:20 uninstall.yml ************************************************** # cat playbooks/openshift-prometheus/config.yml --- - import_playbook: ../init/main.yml - import_playbook: private/config.yml
Great. Can you move to verified?
(In reply to Zohar Gal-Or from comment #6) > Great. > Can you move to verified? No, Comment 5 is current playbook files in openshift-ansible-3.9.0-0.39.0, I just show you the fix is in this openshift-ansible version, but it still throw out "Could not find or access '/usr/share/ansible/openshift-ansible/playbooks/openshift-prometheus/private/uninstall_prometheus.yaml" when undeploy prometheus, see Comment 3
I don't understand. What exactly are you running that results in the error output?
Created attachment 1392565 [details] ansible running log
I see the problem now. last PR broke that path. I'll fix it
PR with fix: https://github.com/openshift/openshift-ansible/pull/7044
https://github.com/openshift/openshift-ansible/pull/7044 merged to master
Fix is in openshift-ansible-3.9.0-0.46.0
Tested with openshift-ansible-3.9.0-0.47.0, prometheus can be undeployed successfully now. # rpm -qa | grep openshift-ansible openshift-ansible-docs-3.9.0-0.47.0.git.0.f8847bb.el7.noarch openshift-ansible-roles-3.9.0-0.47.0.git.0.f8847bb.el7.noarch openshift-ansible-3.9.0-0.47.0.git.0.f8847bb.el7.noarch openshift-ansible-playbooks-3.9.0-0.47.0.git.0.f8847bb.el7.noarch
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-2018:3748