One of my LVM volume is shown to have 372.6% data. How could this happen? Note the volume have been re-sized down from a bigger size, but the file system seems working fine. It actually impacts the pool's free space availability. Basically, even the file system and LVN have be resized to a smaller size, the space is not freed.
The question is, how to make any unused space available for use?
# lvs
LV VG Attr LSize Pool Origin Data% Move Log Cpy%Sync Convert
lv1 vg1 Vwi-aot--- 820.00g tp1 372.60
lv2 vg1 Vwi-aot--- 1000.00g tp1 1.60
The LVM tree structure looks like this:
# dmsetup ls --tree
vg1-tp1 (253:6)
└─vg1-tp1-tpool (253:5)
├─vg1-tp1_tmeta (253:1)
│ └─ (147:1)
├─vg1-tp1_tierdata_2 (253:4)
│ └─ (147:1)
├─vg1-tp1_tierdata_1 (253:3)
└─vg1-tp1_tierdata_0 (253:2)
ce_cachedev2 (253:11)
└─cachedev2 (253:10)
└─vg1-lv2 (253:9)
└─vg1-tp1-tpool (253:5)
├─vg1-tp1_tmeta (253:1)
│ └─ (147:1)
├─vg1-tp1_tierdata_2 (253:4)
│ └─ (147:1)
├─vg1-tp1_tierdata_1 (253:3)
└─vg1-tp1_tierdata_0 (253:2)
ce_cachedev1 (253:8)
└─cachedev1 (253:0)
└─vg1-lv1 (253:7)
└─vg1-tp1-tpool (253:5)
├─vg1-tp1_tmeta (253:1)
│ └─ (147:1)
├─vg1-tp1_tierdata_2 (253:4)
│ └─ (147:1)
├─vg1-tp1_tierdata_1 (253:3)
└─vg1-tp1_tierdata_0 (253:2)
