Richard Geary
Cadet
- Joined
- May 8, 2016
- Messages
- 5
I've created a 1Tb RAID-1 pool named NASBackup, and enabled dedup & compression. I'm getting confusing numbers for the available size on the pool vs the dataset.
Pool has 209G available, dataset has only 77G available. There is only one dataset in the pool.
Some questions :
1) Why is avail 132Gb smaller on the dataset? There's no parity here.
2) Why is the dataset used property larger than the logicalused property, when compression ratio is 1.12?
3) Why is the zpool alloc size 719G when dedup is 1.26x? Wouldn't that would imply 905Gb un-deduped compressed data? (which is smaller than "used" property)
Thanks!
Pool has 209G available, dataset has only 77G available. There is only one dataset in the pool.
Code:
[root@freenas ~]# zpool list NASBackup NAME SIZE ALLOC FREE EXPANDSZ FRAG CAP DEDUP HEALTH ALTROOT NASBackup 928G 719G 209G - 62% 77% 1.26x ONLINE /mnt [root@freenas ~]# zfs list -o space NASBackup NAME AVAIL USED USEDSNAP USEDDS USEDREFRESERV USEDCHILD NASBackup 77.4G 858G 0 27K 0 858G [root@freenas ~]# zfs get "name,available,used,logicalused,compressratio" NASBac kup 2>&1|less NAME PROPERTY VALUE SOURCE NASBackup name NASBackup - NASBackup available 75.4G - NASBackup used 860G - NASBackup logicalused 844G - NASBackup compressratio 1.12x -
Some questions :
1) Why is avail 132Gb smaller on the dataset? There's no parity here.
2) Why is the dataset used property larger than the logicalused property, when compression ratio is 1.12?
3) Why is the zpool alloc size 719G when dedup is 1.26x? Wouldn't that would imply 905Gb un-deduped compressed data? (which is smaller than "used" property)
Thanks!
Last edited by a moderator: