If the AF inbound fee target for a channel is not a multiple of the increment, the fee will adjust and alternate back and forth from the Max Cost percentage target and the AF Increment.
Example:
Node A has a fee of 350ppm, 65% max cost percentage, and AF increment is set to 25.
The inbound fee will adjust from 0ppm -> -227ppm (based on max cost percentage). Then on the next adjustment process, it will adjust to -225pm (based on AF increment). Then it will adjust back to -227ppm (based on max cost)... and keeps going.
I think a more intuitive behaviour is for the Max Cost Percentage to override the AF increment when adjusting to be a multiple, and in the example for the inbound fee to stay at -227ppm.
If the AF inbound fee target for a channel is not a multiple of the increment, the fee will adjust and alternate back and forth from the Max Cost percentage target and the AF Increment.
Example:
Node A has a fee of 350ppm, 65% max cost percentage, and AF increment is set to 25.
The inbound fee will adjust from 0ppm -> -227ppm (based on max cost percentage). Then on the next adjustment process, it will adjust to -225pm (based on AF increment). Then it will adjust back to -227ppm (based on max cost)... and keeps going.
I think a more intuitive behaviour is for the Max Cost Percentage to override the AF increment when adjusting to be a multiple, and in the example for the inbound fee to stay at -227ppm.