Code review comment for ~info-martin-konrad/epics-base:callbackQueueStatus

I like the idea of adding queue health status.

Number of overruns is good and useful data.

Given the common usage pattern of those queues (a large number of entries is put on the queue by a driver or at IOC init for PINI processing, then a number of threads are working on taking items off) I think that an iocStats or manual check of the current usage does not tell much as it most probably misses the moment of high usage.
A simple high water mark mechanism might generate data that is a lot more useful.

I would also consider resetting the stats at every read of their values for easy one-record reporting by iocStats. (Cumulative values are not nice in the archiver.)