This message reflects a transient condition where the queue which accepts channel measurement information is full. The protocol around the sounding process is tolerant to these faults and will deal with errors, then recover. It shouldn't affect overall link performance.
It would be useful to take a look and examine field diags and the engineering.cgi file for this link, if you're willing to share.
How many do you have installed so far? How are you liking the performance aside from these event log messages?
Only two up so far, but one of them has seen a significant increase in bandwidth over the previous PMP450 AP. Not seeing as much change in the one which replaced a PMP 450i. We had some Ethernet issues on the other, probably due to cable length, but that one is looking good now, so we'll see how it performs over the next few days ...
If we continue to see these errors, I'll be happy to provide engineering data.
Again, these failures are logged, but most of the time, do not indicate a real issue. If you suspect that the performance of the sectors where this is happening is sub-optimal, then Cambium Support can certainly take a look at some file captures to see if it can be improved.