home bbs files messages ]

Forums before death by AOL, social media and spammers... "We can't have nice things"

   comp.protocols.tcp-ip      TCP and IP network protocols.      14,669 messages   

[   << oldest   |   < older   |   list   |   newer >   |   newest >>   ]

   Message 13,596 of 14,669   
   Jorgen Grahn to Morten Reistad   
   Re: Extending IPv4 with source translati   
   15 Sep 10 20:37:20   
   
   From: grahn+nntp@snipabacken.se   
      
   On Tue, 2010-09-14, Morten Reistad wrote:   
   > In article ,   
   > Jorgen Grahn   wrote:   
   >>On Mon, 2010-09-13, Morten Reistad wrote:   
   >>> In article ,   
   >>> Jorgen Grahn   wrote:   
   >>>>On Thu, 2010-09-09, glen herrmannsfeldt wrote:   
   >>>>> In comp.dcom.lans.ethernet Jorgen Grahn    
   wrote:   
   >>>>> (snip)   
   >   
   >>> All the UDP-based realtime media protocols do this when implemented   
   >>> to support this NAT-mode. SIP, RTP/RTCP, IAX, etc   
   >>>   
   >>> So do all the tunnel protocols, so people like us can have our own   
   >>> addressed network on top of the Internet infrastructure, no matter now   
   >>> much NAT stuff the underlying routers deal with.   
   >>   
   >>Uh, *which* NAT mode are we talking about, here? I was under the   
   >>impression that we were discussing a scenario where you could send a   
   >>number of packets from A to B, and B would see source address C on   
   >>some of them and D on others, depending which path they took.   
   >>   
   >>I hope no NAT works like that.   
   >   
   > Your hopes are not confirmed by reality. This is a regular occurrance,   
   > and RTP/RTCP, SIP, RTSP, IAX and H.323 have "extensions" that deal   
   > successfully with this. The ipv4 internet has decayed way below what   
   > we would reasonably have expected to deal with in 1992.   
   >   
   > Yes, NAT devices, as implemented in reality, regularly drop state   
   > and give you a new externally visible session.   
      
   Ugh. In that case, I'm happy the only NAT I'm behind is my own, very   
   stateful, Linux box.   
      
   > The IPv4 internet has decayed beyond guaranteeing long time persistence   
   > of ip-based sessions. TCP sessions regularly die, unless we use a next   
   > layer of sessions. This will only get worse.   
      
   If it cannot handle ssh, it's not the internet as far as I'm   
   concerned.  It's something else.   
      
   /Jorgen   
      
   --   
     // Jorgen Grahn    O  o   .   
      
   --- SoupGate-Win32 v1.05   
    * Origin: you cannot sedate... all the things you hate (1:229/2)   

[   << oldest   |   < older   |   list   |   newer >   |   newest >>   ]


(c) 1994,  bbs@darkrealms.ca