>> UDP is for AF_INET and AF_INET6, not AF_LOCAL (nee AF_UNIX).
> I agree, but I struggle to underestand how he can really be using an
> AF_LOCAL socket for remote management polling, which is what I
> thought the application was described as.

That was one of the things that puzzled me about the original message,
but it appeared to me on a careful reading that the problem was not
with communication with remote machines but rather with a local process
communicating with a local server, to collect information _about_
remote machines.

Reinforcing this was the circumstantial evidence that the code quoted
used a variable name ending _sau for the sockaddr for sendto().  To me
that suffix says "sockaddr_un", not "sockaddr_in".  And this message
clearly stated AF_LOCAL, and while it also clearly stated UDP, the code
will not say UDP anywhere, whereas it will say AF_LOCAL, so the
AF_LOCAL is more likely to be right - and getting nothing from
udp_usrreq debugging seems like confirmation.

/~\ The ASCII				der Mouse
\ / Ribbon Campaign
 X  Against HTML	       mouse@rodents.montreal.qc.ca
/ \ Email!	     7D C8 61 52 5D E7 2D 39  4E F1 31 3E E8 B3 27 4B