Running Wafliron on an aggregate having a vvol with same fsid can cause the system to panic.

 

Description

In WAFL, each volume is uniquely identified by an id called "File System IDentifier" aka FSID.

In a filehandle that WAFL sends to clients (such as, NFS), there are two pieces of information: a) A unique identifier (i.e., FSID) for a volume on a given host b) The inode number relative to the volume Thus, when WAFL is presented with a filehandle from an end-client, WAFL can retrieve data from the associated file. The FSIDs are nearly unique. If a vvol and parent aggregate have the same fsid and one file in vvol and another in aggregate are having the same fileid then this problem could occur because Wafliron looks for any inode on the basis of fileid and fsid.  

Workaround

Though there is a rare chance to hit this issue but before the upgrade,

"vol read_fsid <vol-name>" & "aggr read_fsid <aggr-name>" can be used to find the fsid of a vvol/aggr. If some aggr and vvol, to be ironed, have duplicate fsid, options could be: 1. Run wack in place of wafliron 2. Change fsid with an absolutely unique fsid and run wafliron     aggr rewrite_fsid <aggr-name> can be used to change the fsid. Note:These commands are available in maintenance mode only. Customers running wafliron are recommended to contact support to receive the version of Data ONTAP which contains the fix for burt 441653.