Dear Lustre experts, I have configured two different Lustre instances, both using Lustre 2.5.3, one with ldiskfs on RAID-6 hardware RAID and one using ZFS and RAID-Z2, using the same type of hardware. I was wondering, why I 24 TB less space available, when I should have the same amount of parity used:
# lfs df UUID 1K-blocks Used Available Use% Mounted on fs19-MDT0000_UUID 50322916 472696 46494784 1% /testlustre/fs19[MDT:0] fs19-OST0000_UUID 51923288320 12672 51923273600 0% /testlustre/fs19[OST:0] fs19-OST0001_UUID 51923288320 12672 51923273600 0% /testlustre/fs19[OST:1] fs19-OST0002_UUID 51923288320 12672 51923273600 0% /testlustre/fs19[OST:2] fs19-OST0003_UUID 51923288320 12672 51923273600 0% /testlustre/fs19[OST:3] filesystem summary: 207693153280 50688 207693094400 0% /testlustre/fs19 UUID 1K-blocks Used Available Use% Mounted on fs18-MDT0000_UUID 47177700 482152 43550028 1% /lustre/fs18[MDT:0] fs18-OST0000_UUID 58387106064 6014088200 49452733560 11% /lustre/fs18[OST:0] fs18-OST0001_UUID 58387106064 5919753028 49547068928 11% /lustre/fs18[OST:1] fs18-OST0002_UUID 58387106064 5944542316 49522279640 11% /lustre/fs18[OST:2] fs18-OST0003_UUID 58387106064 5906712004 49560109952 11% /lustre/fs18[OST:3] filesystem summary: 233548424256 23785095548 198082192080 11% /lustre/fs18 fs18 is using ldiskfs, while fs19 is ZFS: # zpool list NAME SIZE ALLOC FREE CAP DEDUP HEALTH ALTROOT lustre-ost1 65T 18,1M 65,0T 0% 1.00x ONLINE - # zfs list NAME USED AVAIL REFER MOUNTPOINT lustre-ost1 13,6M 48,7T 311K /lustre-ost1 lustre-ost1/ost1 12,4M 48,7T 12,4M /lustre-ost1/ost1 Any idea on why my 6TB per OST went? Regards, Götz Waschk _______________________________________________ lustre-discuss mailing list lustre-discuss@lists.lustre.org http://lists.lustre.org/listinfo.cgi/lustre-discuss-lustre.org