Deleted aggregate but disks still show container-type = old aggregate | NetApp | Page 1

nimble whale Sep 30, 2025, 11:34 AM

#

Hello,

We need to recreate all the aggregates on a FAS8200 (7 shelves).

Phase 1:
I moved all the data, took the aggregates offline, and then deleted them using:
storage aggregate delete -aggregate ...

Phase 2:
I wanted to reassign the disks to the FAS nodes with:
storage disk assign -owner Node01 -disk 00001

But when I run:
storage disk show -disk 00001

the container-type field is not spare but still shows the name of the old aggregate that I already deleted.

As a result, I cannot reuse my disks.
According to NetApp documentation, it seems that the disk metadata still contains the name of the old aggregate.

I haven’t been able to find a method to wipe these metadata and reuse my disks.

Does anyone have any advice or suggestions?

mental anvil Sep 30, 2025, 12:31 PM

#

Do you still see the seemingly deleted aggregate with aggr show -aggregate [aggrname]?
Are these aggregates you created in this cluster? Or did you add shelves with aggregates from other clusters?

#

If the aggr info is still in WAFL, you could try the following:

set advanced
storage aggregate remove-stale-record -aggregate [aggrname] -nodename [nodename]

nimble whale Sep 30, 2025, 12:48 PM

#

Thank you. No, I don’t see the aggregate anymore when I run aggr show -aggregate.
However, the name still appears in the container-type field of the disks.

dry stag Sep 30, 2025, 1:11 PM

#

check the vreport -

#

set d; debug vreport show

nimble whale Sep 30, 2025, 1:24 PM

#

I just ran debug vreport show and I can still see 3 of the aggregates I deleted.
They are in a Failed state, and it says “Present only in WAFL”.

dry stag Sep 30, 2025, 1:33 PM

#

thx. Generally you can run a fix and remove them, but without knowing more. i'd recomend opening a case to verify it's all good an these are infact only stale entries.

nimble whale Sep 30, 2025, 1:52 PM

#

Thanks, I’ll open a NetApp case and share the outcome. Meanwhile, if anyone has other ideas, I’m all ears.

mental anvil Sep 30, 2025, 4:33 PM

#

Did you do the aggregate remove-stale-record? Should do the same as debug vreport but at least it's a supported cmd.

#

But yeah always better to go the support case route.

nimble whale Oct 1, 2025, 6:58 AM

#

Hello, yesterday debug vreport show listed 3 aggregates as Failed (Present only in WAFL), but this morning only 1 remains.
Is there some internal WAFL cleanup process?
Can I force a refresh or resync?

nimble whale Oct 1, 2025, 3:03 PM

#

We can still see the phantom aggregates in vreport even if the disks aren’t assigned.
After some testing, I found that aggregate remove-stale-record only works when the disks are assigned to the node.

Now I can reuse my disks and rebuild the aggregates.
Thanks for the help, we can close this thread.

#Deleted aggregate but disks still show container-type = old aggregate