Aggregate Considered Possible Location for Additional Metadata

[swirlypillow]

Hi!

I received a perplexing message when creating two new aggregates across some DS2246 shelves that had recently been migrated from our old FAS3220 to our new FAS8200.

After assuming ownership of the disks, I got the following message after creating each aggregate:

"Job succeeded: DONE. Due to a lack of sufficient system storage, this aggregate will be considered as a possible location for additional metadata."

Any idea why these aggregates would be marked as potential metadata storage locations?  Seems my root vols for both nodes are only 10% full.

[Sahana]

Hi,

Auditing volumes were introduced in Data ONTAP 8.2, there might be a space usage issue for NAS volumes. All staging volume names begin with MDV_aud_ followed by the UUID of the aggregate containing that staging volume. Check the aggregate free space.

http://mysupport.netapp.com/NOW/cgi-bin/bol?Type=Detail&Display=816097

[swirlypillow]

Hi Sahana,

Thanks.  We did have auditing configured (but not enabled) for one SVM prior to the two new aggregates being created.  All aggregates in the cluster (with the exception of the aggr0s containing the root vols) have 36% free space or greater.  

Both the new aggrs still have nearly 100% free space, as we have not created or moved any volumes to them yet.  According to the bug detail, creating large aggregates can shut down the auditing function, but we already had it disabled prior to creating the aggregates.  However, I see that both new aggrs now contain MDV_aud vols, so I'm not sure if that indicates an actual issue or not.  Also, I noticed the bug detail shows the bug being fixed as of ONTAP 8.2.2.RC1 (we're running 9.1 GA).

Thank you for your insight!  I'm sure the issue lies with the auditing function, so I'll continue to investigate this angle.