Created attachment 1384680 [details] playbook.log Description of problem: cephmetrics-ansible installation with non-default password using vars.yml fails. Version-Release number of selected component (if applicable): [root@magna024 cephmetrics-ansible]# ceph -v ceph version 10.2.10-14.el7cp (b05aecbcede52a89a654e9f2bec179c2050ddff6) [root@magna024 cephmetrics-ansible]# rpm -qa | grep ansible ceph-ansible-3.0.18-1.el7cp.noarch cephmetrics-ansible-1.0-8.el7cp.x86_64 ansible-2.4.1.0-1.el7ae.noarch How reproducible: Always Steps to Reproduce: 1. Followed the document https://doc-stage.usersys.redhat.com/documentation/en-us/red_hat_ceph_storage/3/html-single/administration_guide/#monitoring-ceph-clusters-with-the-red-hat-ceph-storage-dashboard, for installation 2. Created a vars.yml file as mentioned in the document. 3. Playbook fails with the following error. TASK [ceph-grafana : Set grafana_data_source] ********************************************************************************************************************************************************************* fatal: [magna024.ceph.redhat.com]: FAILED! => {"failed": true, "msg": "The task includes an option with an undefined variable. The error was: 'dict object' has no attribute 'datasource'\n\nThe error appears to have been in '/usr/share/cephmetrics-ansible/roles/ceph-grafana/tasks/configure_grafana.yml': line 51, column 3, but may\nbe elsewhere in the file depending on the exact syntax problem.\n\nThe offending line appears to be:\n\n\n- name: Set grafana_data_source\n ^ here\n\nexception type: <class 'ansible.errors.AnsibleUndefinedVariable'>\nexception: 'dict object' has no attribute 'datasource'"} Actual results: Fails with Error Expected results: The playbook should complete successfully. Additional info: [root@magna024 cephmetrics-ansible]# cat vars.yml grafana: admin_password: CGqf5HhUaZ
Apparently, this was kinda by design, you were supposed to deploy with the default password and then change it in the web UI. Afterwards, you had to change the admin_password variable to be able to redeploy when updating dashboards, etc. This should still work. I suppose this is not very well documented (it is in the code as a commentary). I have a patch for this staged for 3.1 where you could directly deploy with a non-default password in this PR: https://github.com/ceph/cephmetrics/pull/165 Given that there is no customer case attached, I'd probably rather keep this staged for 3.1 (the back-port would not apply cleanly due to recent changes and the PR is not merged, yet). Any objections to pushing this to 3.1?
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-2018:2819