Indeed, thanks to everyone for the help. We're investigating the possibility of a "virtual NIC" sitting outside the campus firewall since there doesn't appear to be any easier way to resolve the issue, but I'm not sure yet how that will work or if we'll get approval from the security dragons. ;)<div>
<br>Out of curiosity, is there a reason why the DNS resolution is done at the server for the UDP region handshake instead of sending the hostname to the client as seems to be done with the TCP traffic? I'm guessing it can't be an easy thing to change or someone probably would have done by now..</div>
<div><br></div><div>Thanks again though, it's good to know how that bit actually works!</div><div><br></div><div>Sincerely,</div><div><br></div><div>- Chris/Fleep</div><div><br></div><div><br></div><div>Chris M. Collins (SL: Fleep Tuque)</div>
<div>Project Manager, UC Second Life </div><div>Second Life Ambassador, Ohio Learning Network </div><div>UCit Instructional & Research Computing</div><div>University of Cincinnati </div><div>406E Zimmer Hall</div><div>
PO Box 210088</div><div>Cincinnati, OH 45221-0088</div><div>(513)556-3018</div><div><a href="mailto:chris.collins@uc.edu">chris.collins@uc.edu</a></div><div><br></div><div>UC Second Life: <a href="http://homepages.uc.edu/secondlife">http://homepages.uc.edu/secondlife</a></div>
<div>OLN Second Life: <a href="http://www.oln.org/emerging_technologies/emtech.php">http://www.oln.org/emerging_technologies/emtech.php</a></div><div><br></div><div><br></div><div><br><div class="gmail_quote">On Fri, Apr 1, 2011 at 9:40 PM, Justin Clark-Casey <span dir="ltr"><<a href="mailto:jjustincc@googlemail.com">jjustincc@googlemail.com</a>></span> wrote:<br>
<blockquote class="gmail_quote" style="margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex;">Thanks for the info Simon, that's cleared that up for me. And thanks to the others contributing to this thread.<br>
<br>
I think that a utility to test a connection and provide an explicit diagnosis (rather than the clues provided by the client just hanging in various places) would be rather nice to have.<div><div></div><div class="h5"><br>
<br>
On 01/04/11 12:59, Simon Slavin wrote:<br>
<blockquote class="gmail_quote" style="margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex">
<br>
On 1 Apr 2011, at 12:43am, Justin Clark-Casey wrote:<br>
<br>
<blockquote class="gmail_quote" style="margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex">
Some people on Stack Overflow think that IPAddress.Any means listen on all NICs (<a href="http://stackoverflow.com/questions/1777629/how-to-listen-on-multiple-ip-addresses" target="_blank">http://stackoverflow.com/questions/1777629/how-to-listen-on-multiple-ip-addresses</a>). But my reading of the MS SDK reference above means that it only binds to one. Anybody able to comment on this?<br>
</blockquote>
<br>
Stack Overflow is right. My reading of that SDK page is that it's wrong, and should be corrected, but other MS documentation is clearer on what '::Any' means, for example<br>
<br>
<a href="http://msdn.microsoft.com/en-us/library/system.net.ipaddress.any.aspx" target="_blank">http://msdn.microsoft.com/en-us/library/system.net.ipaddress.any.aspx</a><br>
<br>
Returning to the standards, IP address 0.0.0.0 is reserved for specific purposes for both sending and receiving. It's called the 'anonymous' address (for historical reasons) or the 'broadcast' address (for current reasons).<br>
<br>
If a computer SENDS a packet to 0.0.0.0 then it is multibroadcasting: sending one message to every computer that can hear it. This is done most often to announce the (un)availability of a service, for instance that a printer service has come online. Sending to address 0.0.0.0 is done by, for example, DHCP and zeroconf (what Apple calls 'Bonjour'). Things like routers are usually set up to drop packets SENT to 0.0.0.0 so that you don't announce to the entire world what address your printer can be found on.<br>
<br>
When a computer LISTENS to the network interface bound to 0.0.0.0 then it is telling its TCP stack that it doesn't care which network interface a message comes in on, it wants it anyway. Almost every Internet application does this, especially now many have both Ethernet sockets and WiFi capabilities: an app doesn't care what its user is using right now, it just wants to 'use the internet'. Under normal circumstances the only programs which /don't/ listen on 0.0.0.0 are techie programs like network utilities, or a web server on a gateway computer which needs to present a web site to internal users and make sure it isn't available to external users.<br>
<br>
I tried to find an RFC to point to as reference but nothing seems to spell this out. The nearest thing i could find was RFC950.<br>
<br>
Simon.<br>
_______________________________________________<br>
Opensim-users mailing list<br>
<a href="mailto:Opensim-users@lists.berlios.de" target="_blank">Opensim-users@lists.berlios.de</a><br>
<a href="https://lists.berlios.de/mailman/listinfo/opensim-users" target="_blank">https://lists.berlios.de/mailman/listinfo/opensim-users</a><br>
<br>
</blockquote>
<br>
<br></div></div><div class="im">
-- <br>
Justin Clark-Casey (justincc)<br>
<a href="http://justincc.org/blog" target="_blank">http://justincc.org/blog</a><br>
<a href="http://twitter.com/justincc" target="_blank">http://twitter.com/justincc</a><br>
_______________________________________________<br></div><div><div></div><div class="h5">
Opensim-users mailing list<br>
<a href="mailto:Opensim-users@lists.berlios.de" target="_blank">Opensim-users@lists.berlios.de</a><br>
<a href="https://lists.berlios.de/mailman/listinfo/opensim-users" target="_blank">https://lists.berlios.de/mailman/listinfo/opensim-users</a><br>
</div></div></blockquote></div><br></div>