XPost: comp.os.linux.networking   
   From: m@rtij.nl.invlalid   
      
   On Sun, 26 Sep 2010 12:21:16 +0000, Mark Hobley wrote:   
      
   > On Sun, 26 Sep 2010 12:23:58 +0200, Ersek, Laszlo wrote:   
   >   
   >> The discovery and fragmentation work in the IP layer. The HTTP proxy   
   >> works in the TCP (and application) layers. Once Machine B is reached,   
   >> the IP layer's job is done. That job recommences towards Machine C.   
   >   
   >> Machine A and Machine B are constant links in the above chain. Machine   
   >> B connects to lots of different HTTP servers (or further proxies)   
   >> scattered all over the net. Machine C is variable, and thus each type   
   >> of B-C link is different. Therefore it makes sense for the discovered   
   >> A-B and B-C PMTU's to differ and to have different lifetimes.   
   >   
   > If Machine B is receiving data from Machine A in 1500 byte segments, and   
   > then retransmitting this in smaller segments to suit C, then PMTUD has   
   > not really prevented fragmentation. Wouldn't it have been better to just   
   > do away with PMTUD and let the fragmentation and reassembly take place   
   > downstream towards C?   
      
   No, no, those are different TCP connections. They have nothing to do with   
   each other. They are separate connections with their own session   
   identifiers, their own endpoints. The only connection between them is   
   that the connection from B to C is initiated as a result from the   
   connection from A to B, but that is all they have to do with each other.   
      
   Let's say a user telnets from A to B. Another user telnets from B to C.   
   Would you expect the MSSes from these connections to influence each   
   other? Of course not. Yet this is exactly the same scenario, two separate   
   TCP connections, one from A to B and one from B to C.   
      
   HTH,   
   M4   
      
   --- SoupGate-Win32 v1.05   
    * Origin: you cannot sedate... all the things you hate (1:229/2)