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 12,914 of 14,669    |
|    pk to Alan S    |
|    Re: Path MTU    |
|    03 Jun 09 18:57:44    |
      From: pk@pk.invalid              On Wednesday 3 June 2009 18:21, Alan S wrote:              >>Right, or another problem is when there is some device in the return path       >>that blocks such ICMP packets, so the source doesn't receive the PMTU       >>information, thus breaking the PMTU discovery process.       >       > Which is correct here?              PMTU discovery is done by the endpoints only, and it works fine if there are       no black holes in the middle. If there are broken hosts in the middle (that       don't send or block ICMP messages on which PMTU discovery relies upon),       PMTU discovery won't work, as the endpoints won't be able to gather the       necessary information. An alternative algorithm to do PMTU discovery which       does not rely on ICMP is described in RFC 4821.              --- 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