2014-05-14 10:51 AM
Does maximum vol size for device include WAFL reserve or something nonobvious like that? Looks like the usable max vol size for 8.1.4 on a 3240 is actually quite a bit less than 50T as claimed in HW Universe:
> vol size vol4
vol size: Flexible volume 'vol4' has size 48318381160k.
> vol size vol4 +1m
vol size: Request to grow volume 'vol4' failed because the resulting volume size is greater than the maximum size.
That's only about 44T in 1024x1024x1024 units. We recently upgraded OnTAP from 8.0.2 to 8.1.4 to get a bigger aggregate size, but 8.1.4 upgrade only got me an additional 1TB max volume size. We were able to grow the aggregate to 48TB, so there is plenty of space in the aggregate. There are about 20 qtrees on the volume, but quota is off for the volume. SIs is on for the system, but not configured for this volume.
Snap reserve for this volume is 0 and nosnap is ON:
> vol options vol4
nosnap=on, nosnapdir=off, minra=off, no_atime_update=off, nvfail=off,
ignore_inconsistent=off, snapmirrored=off, create_ucode=on,
convert_ucode=on, maxdirsize=73400, schedsnapname=ordinal,
fs_size_fixed=off, guarantee=volume, svo_enable=off, svo_checksum=off,
svo_allow_rman=off, svo_reject_errors=off, no_i2p=off,
fractional_reserve=100, extent=off, try_first=volume_grow,
read_realloc=off, snapshot_clone_dependency=off, dlog_hole_reserve=off,
chesdfiler1a> vol status vol4
Volume State Status Options
vol4 online raid_dp, flex nosnap=on, create_ucode=on, convert_ucode=on
Volume UUID: 4f3ec690-9acb-11e0-95ee-00a09816961c
Containing aggregate: 'aggr2'
> aggr show_space aggr2
Total space WAFL reserve Snap reserve Usable space BSR NVLOG A-SIS Smtape
57279822336KB 5727982232KB 0KB 51551840104KB 0KB 0KB 0KB
Space allocated to volumes in the aggregate
Volume Allocated Used Guarantee
vol4 48586816616KB 47413626408KB volume
Aggregate Allocated Used Avail
Total space 48586816616KB 47413626408KB 2961725708KB
Snap reserve 0KB 0KB 0KB
WAFL reserve 5727982232KB 400065692KB 5327916540KB
Solved! SEE THE SOLUTION
2014-05-14 11:12 AM
> df -Vh vol4
Filesystem total used avail capacity Mounted on
/vol/vol4/ 44TB 44TB 1013GB 98% /vol/vol4/
/vol/vol4/.snapshot 0GB 21GB 0GB ---% /vol/vol4/.snapshot
> df -Ah
Aggregate total used avail capacity
aggr2 48TB 45TB 2824GB 94%
aggr2/.snapshot 0TB 0TB 0TB ---%
2014-05-15 12:46 PM
The max volume size includes the 1% aggregate overhead and 10% volume overhead, or at least it did in 7.3. I worked out a spreadsheet once in order to determine the optimum backend lun size for a V-Series needed to maximize the volume usable space, and had to take both figures into account. I also had to take into account the checksum overhead - though I hear that, with native NetApp disk, that has gone away.
2014-05-19 09:24 AM
There is some fine print. At the bottom of each hardware universe page"
The maximum usable aggregate capacity is roughly 90% of the values shown in the "Aggregate Size 32 bit(TB)" rows.
The maximum usable aggregate capacity is roughly 90% of the values shown in the "Aggregate Size 64 bit(TB)" rows.
The maximum usable FlexVol capacity is roughly 90% of the values shown in the "Flex Volume Size 32 bit(TB)" rows.
The maximum usable FlexVol capacity is roughly 90% of the values shown in the "Flex Volume Size 64 bit(TB)" rows.
This is misleading and inaccurate. I have created volumes that fill up 100% of the space of an aggregate (as sown by df -A), and volumes fill up to 100% of "usable space" all the time. Perhaps the fine print means "maximum size" instead of "maximum usable capacity"? In which case 88% for volumes is a more reasonable figure.