RE: FW: xAP configuration protocol

[Kevin Hawkins]

> -----Original Message-----
> From: Stuart Booth [mailto:<a
href="/group/xAP_developer/post?postID=Mnq2pDU0tRGQTgja5i4CYz5kIZ35ToHcSFq6kXmHwsccX32wHbVYxkvK_3bQpuWtaBbTFnQZdss870ffvEkp4g">lists@x...</a>]
> Sent: 12 August 2003 15:40
> To: <a
href="/group/xAP_developer/post?postID=77PM9frpRQdoypNcf0Q7hP_B8MC-3ce83OSsTpQbBnXcMJkAPppzRLv5YFwJfKBLFeccNqledfy8_ZuWBIpaakCsSw">xAP_developer@xxxxxxx</a>
> Subject: Re: FW: [xAP_developer] xAP configuration protocol
>
> On Tue, 12 Aug 2003 14:59:09 +0100, "Kevin Hawkins"
> <<a
href="/group/xAP_developer/post?postID=RTqMo4S7QHcxLp8ovD5tpEh_6ydjA7rBsjLvfbUciw7YJ8RUyNzqRktGJufWd_FNQM58w1oqSCRN5QXGlpyBvA">lists@u...</a>>
wrote:
>
> > Yep - in fact Heartbeats are typically much more frequent -
> although
> >I think it's not currently a requirement that everyone sends them.
> There
> >will be some receiver only xAPp's but as a result they shouldn't
need
> UID's
> >anyway . How do people feel about moving heartbeats to mandatory
for
> all xAP
> >senders ?
>
> Receiver only xAPps don't need h/b's?!


What I meant here was that a xAPp implemented in an embedded device that
was
a receiver only needn't have a heartbeat - as it is a receiver only in fact
nobody knows it's there. I think the spec says heartbeats are recommended
but not mandatory - even in transmitting xAP'S - my question was really
about peoples views on moving this to mandatory for transmitters.

Receivers (only) running on a PC for example will never be hub compliant as
the hub will never know they are there - it uses the heartbeat to know
where
to forward the incoming xAP datagrams to. Only one receiver can sit on any
one port. A receive only app could run on a PC but only as the sole xAP
application monopolising port 3639.

K

>
> I figured every persistant xAP must send a h/b out at regular
> intervals. The only xAP I have not implemented any h/b transmission in
> is my simple Send utility which starts, sends a message, and exits
> again.

Hmmm - an interesting one - particularly in this context of UID conflicts -
how does it know it's UID is unique when it does this ? - or how would it
know that in the context of any autoconfiguration mechanism we could
formulate ? Perhaps we should have a special UID allocated for non
persistant devices like this - and other ranges for devices that have no
listening capability so that the network knows they are deaf.

K