Odd behavior with WAFL Reserve

[christian_keilfit]

Good Morning,

I am currently performing some burn in Testing with our new FAS3160.

After delting 2 Volumes around 12TB each, (i used them for FS Testing with alot of Small Files)

I turned both Volumes offline and then Destroyed them via CLI.

So far so good, when I now check the free space, i am stilling missing those 12TB.

fitn01p001-b01*> aggr show_space -h
Aggregate 'aggr1'

    Total space    WAFL reserve    Snap reserve    Usable space       BSR NVLOG           A-SIS          Smtape
           32TB          3310GB           595GB            28TB             0KB            21MB             0KB

Space allocated to volumes in the aggregate

Volume                          Allocated            Used       Guarantee
vol0                                301GB          2861MB          volume

Aggregate                       Allocated            Used           Avail
Total space                         301GB          2861MB            16TB
Snap reserve                        595GB             0KB           595GB
WAFL reserve                       3310GB            12TB             0KB

Aggregate 'aggr0'

    Total space    WAFL reserve    Snap reserve    Usable space       BSR NVLOG           A-SIS          Smtape
           32TB          3335GB           600GB            28TB             0KB             0KB             0KB

This aggregate does not contain any volumes

Aggregate                       Allocated            Used           Avail
Total space                           0KB             0KB            17TB
Snap reserve                        600GB           434GB           165GB
WAFL reserve                       3335GB            11TB             0KB

seems like WAFL reserve is still using up all the space, is there any way to free that up by cli ?

Kind Regards

Christian

[aborzenkov · Accepted solution]

Reclaiming space from deleted files is done in background and can take quite some time. Check with “df –Ak” with couple of minutes in between. If available space grows steadily, nothing to worry about – it will be there eventually.

[christian_keilfit]

thanks for the heads up, free space is indeed growing in KB steps

[nqlcadmin]

I'm having the same issue.

Have deleted all volumes/snapshots and I've reclaimed some of the space but there I'm still missing 1.17TB of space which is used by WAFL. It's reclaimed about 896gig but the remaining does not seem to be given back. Not sure why.

Aggregate                total       used      avail capacity

aggr0               2195696472KB 1255210676KB 940485796KB      57%

aggr0/.snapshot     115562972KB 45810772KB 69752200KB      40%

Aggregate 'aggr0'

    Total space    WAFL reserve    Snap reserve    Usable space       BSR NVLOG           A-SIS

   2568066048KB     256806604KB     115562972KB    2195696472KB             0KB       7954552KB

Space allocated to volumes in the aggregate

Volume                          Allocated            Used       Guarantee

vol0                           27095324KB       1313496KB          volume

Aggregate                       Allocated            Used           Avail

Total space                    27095324KB       1313496KB     940485796KB

Snap reserve                  115562972KB      45810772KB      69752200KB

WAFL reserve                  256806604KB    1247569104KB             0KB

Can I force the WAFL reserve to delete or reclaim? There is no data/volumes on that aggr anymore so nothing is needed. Running a df -Ak does not show the data being reclaimed, it sits there at the same value.

Thanks