[Live-devel] Problem with ourSourceAddressForMulticast()
David Betrand
bidibulle at operamail.com
Thu Feb 1 14:28:28 PST 2007
> Once (the admittedly badly named) "ourSourceAddressForMulticast()"
> function succeeds in getting a proper IP address, it will just return
> that same address from then on (with no more multicast).
Right. I supppose you understood that I didn't remind this because it never succeeds in my case.
> Well, "99% of computers of our beautiful planet" will have no need to
> set these addresses at all, because they have only one network
> interface.
One point for you here ! Unfortunately, I'm in "rest of the world" team
> The real question here (which I don't know the answer to) is: If you
> set these addresses to a (valid) local IP address, then why does
> multicast loopback fail, and can this be fixed?
Agreed. And really dont know why it doesn't succeed. All I can say is that all servers on wich I deployed my app have apparently the same issue (all on the same hardware and OS baseline, though). Just out of curiosity, I'm very curious to know if my little test in testOnDemandRTSPServer works with you. Once you have 5 minutes to test this of course.
If we both really don't know what happens, let's forget this. Maybe one day someone will get the same problem and find out the reason.
> Do you know of any way to work around this, or else know of some
> alternative way to look up a valid IP address (rather than by
> explicitly parsing the /etc/hosts file, which would be gross) that
> will work on Linux systems (and not break on other systems)? If so,
> I could add it - as a third mechanism to try - to the
> "ourSourceAddressForMulticast()" code.
Unfortunately, no.
> I want to come up with code that works. Just throwing up my hands
> and saying "if this code doesn't work, here's how to hard-wire the
> address yourself" is not an acceptable solution (it's the ultimate in
> non-portable code :-). Of course, this is source code, so if people
> really want to do this, they can. I just don't want to encourage it.
Were are on the same wave length on this.
David
--
_______________________________________________
Search for products and services at:
http://search.mail.com
Powered by Outblaze
More information about the live-devel
mailing list