Can you cat this file /etc/ansible/roles/gluster.infra/roles/backend_setup/tasks/vg_create.yml
It seems that the VG creation is not idempotent .As a workaround, delete the VG 'gluster_vg_sdb' on all Gluster nodes: vgremove gluster_vg_sdb Best Regards,Strahil Nikolov
Strahil, Thank you for your response, I have added the file for you, but with my continued efforts over the past day I have finally managed to get oVirt to install but not without issues. Over a period of 2 weeks I have kicked off the install process over 20 times. Occasionally it would return the error posted here, but most of the time the install process would hang on this step. I posted the error as that was the only log/error I could get, when it hung no log files or any footprint was created it just sat there until I either reboot the host or cancelled the install. The problem I have found is that nodectl doesn't return while the installer is in progress, so I assume that when the installer tries to ssh to the localhost it never gets to a shell because nodectl is waiting indefinitely for something. So I removed nodectl-motd.sh and nodectl-run-banner.sh from /etc/profile.d and now the Gluster install wizards works perfectly. Next, the Cockpit wizard refused to identify any of my network devices, but the command line installer was fine, so I have now got a self hosted engine running on one node via: hosted-engine --deploy However my next issue now is that when I try to login to the Administration Portal with the admin user, it is giving me "Invalid username or password". I can log into the Monitoring Portal just fine, but the Administration and VM Portal doesn't like the admin credentials. So now to work out why authentication isn't working. - Dave. ----- /etc/ansible/roles/gluster.infra/roles/backend_setup/tasks/vg_create.yml ----- --- # We have to set the dataalignment for physical volumes, and physicalextentsize # for volume groups. For JBODs we use a constant alignment value of 256K # however, for RAID we calculate it by multiplying the RAID stripe unit size # with the number of data disks. Hence in case of RAID stripe_unit_size and data # disks are mandatory parameters. - name: Check if valid disktype is provided fail: msg: "Unknown disktype. Allowed disktypes: JBOD, RAID6, RAID10, RAID5." when: gluster_infra_disktype not in [ 'JBOD', 'RAID6', 'RAID10', 'RAID5' ] # Set data alignment for JBODs, by default it is 256K. This set_fact is not # needed if we can always assume 256K for JBOD, however we provide this extra # variable to override it. - name: Set PV data alignment for JBOD set_fact: pv_dataalign: "{{ gluster_infra_dalign | default('256K') }}" when: gluster_infra_disktype == 'JBOD' # Set data alignment for RAID # We need KiB: ensure to keep the trailing `K' in the pv_dataalign calculation. - name: Set PV data alignment for RAID set_fact: pv_dataalign: > {{ gluster_infra_diskcount|int * gluster_infra_stripe_unit_size|int }}K when: > gluster_infra_disktype == 'RAID6' or gluster_infra_disktype == 'RAID10' or gluster_infra_disktype == 'RAID5' - name: Set VG physical extent size for RAID set_fact: vg_pesize: > {{ gluster_infra_diskcount|int * gluster_infra_stripe_unit_size|int }}K when: > gluster_infra_disktype == 'RAID6' or gluster_infra_disktype == 'RAID10' or gluster_infra_disktype == 'RAID5' - include_tasks: get_vg_groupings.yml vars: volume_groups: "{{ gluster_infra_volume_groups }}" when: gluster_infra_volume_groups is defined and gluster_infra_volume_groups is not none and gluster_infra_volume_groups|length >0 - name: Record for missing devices for phase 2 set_fact: gluster_phase2_has_missing_devices: true loop: "{{ vg_device_exists.results }}" when: item.stdout_lines is defined and "0" in item.stdout_lines - name: Print the gateway for each host when defined ansible.builtin.debug: msg: vg names {{ gluster_volumes_by_groupname }} # Tasks to create a volume group # The devices in `pvs' can be a regular device or a VDO device # Please take note; only the first item per volume group will define the actual configuraton! #TODO: fix pesize // {{ ((item.value | first).vg_pesize || vg_pesize) | default(4) }} - name: Create volume groups register: gluster_changed_vgs command: vgcreate --dataalignment {{ item.value.pv_dataalign | default(pv_dataalign) }} -s {{ vg_pesize | default(4) }} {{ (item.value | first).vgname }} {{ item.value | ovirt.ovirt.json_query('[].pvname') | unique | join(',') }} # lvg: # state: present # vg: "{{ (item.value | first).vgname }}" # pvs: "{{ item.value | json_query('[].pvname') | unique | join(',') }}" # pv_options: "--dataalignment {{ item.value.pv_dataalign | default(pv_dataalign) }}" # pesize is 4m by default for JBODs # pesize: "{{ vg_pesize | default(4) }}" loop: "{{gluster_volumes_by_groupname | default({}) | dict2items}}" when: gluster_volumes_by_groupname is defined and item.value|length>0 - name: update LVM fact's setup: filter: 'ansible_lvm' when: gluster_changed_vgs.changed