We have two C5210 Controllers configured in a High Availability cluster.
The solution is configured as a Primary/Backup solution, with all the AP's homing to the first controller as the primary and the secondary as the backup.
We have just over 600 AP's.
On the second controller we see the following error for random AP's.
The AP can reach both controller so this is not a routing error as refered to in the below post: