Re: xPL announcement/description protocol -- was xPLDiag

["Mal Lansell" <mal@xxxxxxxxxxx>]

>
> Sorry, my rant wasn't directed directly at your post (should have
started a
> new thread), I am concerned about all of the posts that have flown
around
> for the last week regarding hubs and everything in general.
>

That's seems to be the way this group runs - nothing but your
excellent xPLMediaNet for months, and then an avalanche of new stuff
for a few weeks :-)


> My point was that xpl works as is (at least should work), and real
problems
> seemingly only started (and only for some users) when we added a
hub to
> applications? At least that's how it appears from my perspective.
>

That is true - I was never very comfortable with the multi-hub
solution, but we're now going for an even cleaner solution than
before with a single stand-alone hub and no built-in versions of any
kind.

However, there were good things too that were introduced with the V3
hub - we now deal properly with multiple network cards for example,
so these features are staying in the new hubs.  The sockets problems
I've been reporting are more to do with my xPLSDK-based apps
communicating with the old xPLHal hub, rather than anything to do
with my new hub.  It may appear that I'm having a nightmare, but my
new hub has actually been working fine with all the existing xPL
applications and Tom has been running it all week on his complex
setup without problems

> I have no problem with the diags stuff in general
>
> I do have a problem with things like 3 second heartbeats, but as
it doesn't
> seemingly change the protocol in practice (but for me it does in
spirit, but
> that's just my opinion) if other people want to introduce this in
their own
> app's/device's then that's fine by me.

I agree that 3 seconds is a bit excessive, but the reason for the
rapid heartbeats is a valid one, IMO.  Even if an app is already
configured and starts straight away, if it starts before the hub it
could be up to 9 minutes before the hub receives the next hbeat from
the app and starts sending xPL messages to it - and that is not good.

Personally I went for 10 seconds in my SDK, although I haven't yet
implemented the slowdown after two minutes of failed communications
(I didn't know about that at the time).

Mal