Description of problem: Documentation doesn't reflect setting of cluster name as necessary step for installation of dashboard in conjunction with custom cluster names and the installation fails due to missing /etc/ceph/ceph.conf. Version-Release number of selected component (if applicable): cephmetrics-ansible-2.0.2-1.el7cp.x86_64 https://access.redhat.com/documentation/en-us/red_hat_ceph_storage/3/html-single/container_guide/index#installing-the-red-hat-ceph-storage-dashboard-container How reproducible: Always when using custom cluster name. Steps to Reproduce: 1. install Ceph cluster using custom cluster name (here: ceph32c) 2. add dashboard section to /etc/ansible/hosts 3. cd /usr/share/cephmetrics-ansible; ansible-playbook -v playbook.yml Actual results: TASK [ceph-mgr : Enable mgr prometheus module] ******************************************************************** fatal: [ceph32c-osd1]: FAILED! => {"changed": true, "cmd": ["docker", "exec", "ceph-mgr-ceph32c-osd1", "ceph", "--cluster", "ceph", "mgr", "module", "enable", "prometheus"], "delta": "0:00:00.205653", "end": "2019-06-18 05:12:17.193327", "msg": "non-zero return code", "rc": 1, "start": "2019-06-18 05:12:16.987674", "stderr": "2019-06-18 05:12:17.178216 7ff1e2768700 -1 Errors while parsing config file!\n2019-06-18 05:12:17.178239 7ff1e2768700 -1 parse_file: cannot open /etc/ceph/ceph.conf: (2) No such file or directory\n2019-06-18 05:12:17.178240 7ff1e2768700 -1 parse_file: cannot open ~/.ceph/ceph.conf: (2) No such file or directory\n2019-06-18 05:12:17.178241 7ff1e2768700 -1 parse_file: cannot open ceph.conf: (2) No such file or directory\nError initializing cluster client: ObjectNotFound('error calling conf_read_file',)", "stderr_lines": ["2019-06-18 05:12:17.178216 7ff1e2768700 -1 Errors while parsing config file!", "2019-06-18 05:12:17.178239 7ff1e2768700 -1 parse_file: cannot open /etc/ceph/ceph.conf: (2) No such file or directory", "2019-06-18 05:12:17.178240 7ff1e2768700 -1 parse_file: cannot open ~/.ceph/ceph.conf: (2) No such file or directory", "2019-06-18 05:12:17.178241 7ff1e2768700 -1 parse_file: cannot open ceph.conf: (2) No such file or directory", "Error initializing cluster client: ObjectNotFound('error calling conf_read_file',)"], "stdout": "", "stdout_lines": []} [WARNING]: Could not create retry file '/usr/share/cephmetrics-ansible/playbook.retry'. [Errno 13] Permission denied: u'/usr/share/cephmetrics-ansible/playbook.retry' PLAY RECAP ******************************************************************************************************** ceph32c-metrics : ok=1 changed=0 unreachable=0 failed=0 ceph32c-osd1 : ok=11 changed=1 unreachable=0 failed=1 ceph32c-osd2 : ok=1 changed=0 unreachable=0 failed=0 ceph32c-osd3 : ok=1 changed=0 unreachable=0 failed=0 ceph32c-osd4 : ok=1 changed=0 unreachable=0 failed=0 ceph32c-osd5 : ok=1 changed=0 unreachable=0 failed=0 Expected results: TASK [ceph-mgr : Enable mgr prometheus module] ******************************************************************** changed: [ceph32c-osd1] => {"changed": true, "cmd": ["docker", "exec", "ceph-mgr-ceph32c-osd1", "ceph", "--cluster", "ceph32c", "mgr", "module", "enable", "prometheus"], "delta": "0:00:00.644984", "end": "2019-06-18 05:40:45.196568", "rc": 0, "start": "2019-06-18 05:40:44.551584", "stderr": "", "stderr_lines": [], "stdout": "", "stdout_lines": []} Additional info: The needed change is to add configuration of cluster name via group_vars/all.yml configuration: [ansible@ceph32c-metrics cephmetrics-ansible]$ sudo cp group_vars/all.yml.sample group_vars/all.yml [ansible@ceph32c-metrics cephmetrics-ansible]$ sudo vi all.yml [ansible@ceph32c-metrics cephmetrics-ansible]$ sudo vi group_vars/all.yml [ansible@ceph32c-metrics cephmetrics-ansible]$ cat group_vars/all.yml dummy: cluster_name: ceph32c # containerized: true # Set the backend options, mgr+prometheus or cephmetrics+graphite #backend: # metrics: mgr # mgr, cephmetrics # storage: prometheus # prometheus, graphite # Turn on/off devel_mode #devel_mode: true # Set grafana admin user and password # You need to change these in the web UI on an already deployed machine, first # New deployments work fine #grafana: # admin_user: admin # admin_password: admin
I have updated the Release Notes so they explain custom cluster names are not actually supported: https://access.redhat.com/documentation/en-us/red_hat_ceph_storage/3.2/html/release_notes/bug-fixes#the_literal_ceph_volume_literal_utility_2