Description of problem: Install RHVH via Anaconda GUI, the LVs generated by auto partitioning are like: [root@dhcp-10-111 ~]# lvs -a LV VG Attr LSize Pool Origin Data% Meta% Move Log Cpy%Sync Convert home rhvh_dhcp-10-111 Vwi-aotz-- 1.00g pool00 4.79 [lvol0_pmspare] rhvh_dhcp-10-111 ewi------- 96.00m pool00 rhvh_dhcp-10-111 twi-aotz-- <181.44g 3.55 0.42 [pool00_tdata] rhvh_dhcp-10-111 Twi-ao---- <181.44g [pool00_tmeta] rhvh_dhcp-10-111 ewi-ao---- 1.00g rhvh-4.1-0.20170714.0 rhvh_dhcp-10-111 Vwi---tz-k <154.44g pool00 root rhvh-4.1-0.20170714.0+1 rhvh_dhcp-10-111 Vwi-aotz-- <154.44g pool00 rhvh-4.1-0.20170714.0 2.61 root rhvh_dhcp-10-111 Vwi-a-tz-- <154.44g pool00 2.73 swap rhvh_dhcp-10-111 -wi-ao---- <3.88g tmp rhvh_dhcp-10-111 Vwi-aotz-- 1.00g pool00 4.81 var rhvh_dhcp-10-111 Vwi-aotz-- 15.00g pool00 2.87 var_log rhvh_dhcp-10-111 Vwi-aotz-- 8.00g pool00 3.26 var_log_audit rhvh_dhcp-10-111 Vwi-aotz-- 2.00g pool00 4.77 The LVs generated by manual partitioning(clicking '+' to add partitions manually) are like: [root@dhcp-10-111 ~]# lvs -a LV VG Attr LSize Pool Origin Data% Meta% Move Log Cpy%Sync Convert home rhvh_dhcp-10-111 Vwi-aotz-- 1.00g pool00 4.79 [lvol0_pmspare] rhvh_dhcp-10-111 ewi------- 12.00m pool00 rhvh_dhcp-10-111 twi-aotz-- <179.09g 2.76 0.35 [pool00_tdata] rhvh_dhcp-10-111 Twi-ao---- <179.09g [pool00_tmeta] rhvh_dhcp-10-111 ewi-ao---- 1.00g rhvh-4.1-0.20170714.0 rhvh_dhcp-10-111 Vwi---tz-k 164.10g pool00 root rhvh-4.1-0.20170714.0+1 rhvh_dhcp-10-111 Vwi-aotz-- 164.10g pool00 rhvh-4.1-0.20170714.0 2.01 root rhvh_dhcp-10-111 Vwi-a-tz-- 164.10g pool00 1.97 swap rhvh_dhcp-10-111 -wi-ao---- 6.40g tmp rhvh_dhcp-10-111 Vwi-aotz-- 2.00g pool00 4.79 var rhvh_dhcp-10-111 Vwi-aotz-- <14.99g pool00 3.41 var-log rhvh_dhcp-10-111 Vwi-aotz-- 8.00g pool00 3.22 var-log-audit rhvh_dhcp-10-111 Vwi-aotz-- 2.00g pool00 3.99 There are mainly two kinds of differences: 1. For auto partitioning, the size of tmp is 1G, but for manual partitioning, it's 2G. 2. For auto partitioning, the name of partitions mounted on /var/log and /var/log/audit are var_log, and var_log_audit, respectively. For manual partitioning, they are var-log, and var-log-audit, which replaced the "_" with "-". As in both auto partitioning mode and manual partitioning mode, the NIST LVs are generated by program, users can't specify them explicitly, so it's better to keep them consistent. Version-Release number of selected component (if applicable): redhat-virtualization-host-4.1-20170714.1 How reproducible: 100% Steps to Reproduce: 1. Install redhat-virtualization-host-4.1-20170714.1 via Anaconda GUI, and choose auto partitioning, finish all needed steps. 2. Reboot and log into RHVH, run "lvs -a" 3. Reinstall redhat-virtualization-host-4.1-20170714.1 via Anaconda GUI, and choose manual partitioning, click "+" to add /boot, /var, /, and swap, finish all needed steps. 4. Reboot and log into RHVH, run "lvs -a" Actual results: 1. The sizes of tmp are different, just as described above. 2. The names of partitions mounted on /var/log, /var/log/audit are different, just as described above. Expected results: 1. The sizes of tmp could be the same. 2. The names of NIST LVs could keep consistent. Additional info:
The /tmp size was already fixed on master (see https://gerrit.ovirt.org/#/c/75707/), and the naming issue is on POST in another patch (https://gerrit.ovirt.org/#/c/79565/)
Verify Versions: redhat-virtualization-host-4.1-20170815.0 imgbased-0.9.43-0.1.el7ev.noarch Verify Steps: As the steps in comment #0 Results: 1. The size of /tmp is 1G, no matter choosing auto partitioning or manual partitioning. 2. The names of partitions mounted on /var/log, /var/log/audit are unified to var_log, var_log_audit, respectively. So the bug is fixed, set the status to VERIFIED.