Could you be more specific with the message? Are you referring to some errors from during a disk scrub? There are background tasks that frequently check the disks for any issues and will prefail the disk if there are serious issues. Pre-failing will trigger a copy to a spare and then the disk will be listed as broken and need to be replaced.
If you want further info, provide some of the outputs you are seeing.
Do you have any spare disks? You can assign the not owned disks to one of the nodes to set as spares so the aggregates can try to reconstruct themselves. Do you have support with NetApp? One node looks like it is down as well, is there other hardware issues happening?
