On Tue, Feb 4, 2014 at 1:20 PM, George Clark <geomh@fenachrone.com> wrote:
>  I see those two addresses scattered throughout.  I also see a long
> string of 1515.... which may be caused by the Hub saying "whoa, way to
> much data at once, slow down."

Yes during sync operations I see messages pop up about PLM being
"extremely busy"  off and on.
>
> My guess is that someone bytes are being lost.  Given what I see, I
> would suspect that the bytes are being lost in the socket connection
> between the hub and MH.  Why I am not sure.
>
It's hard to believe that it's the TCP connection.  That should be
extremely reliable, as TCP will recover any lost or corrupted packets,
assuming it's implemented correctly.

Agree, it would be strange.  But I don't really have any other explanation as to why the Hub would be sending that many erroneous messages.  Since you are seeing a lot of "extremely busy" messages, we should try and slow down how fast we talk to the Hub.  It could be that we are overloading it causing it to drop bytes somehow.

Do you still have Insteon_PLM_xmit_delay set to 1?  Can you try and increase this to see if the "extremely busy" messages go away?