Date: Sun, 24 Mar 1996 10:03:56 -0500 From: "Louis A. Mamakos" <louie@TransSys.COM> To: Michael Smith <msmith@atrad.adelaide.edu.au> Cc: taob@io.org (Brian Tao), freebsd-hackers@FreeBSD.ORG Subject: Re: Changing Ethernet frame size to 576 bytes? Message-ID: <199603241503.KAA21220@wa3ymh.transsys.com> In-Reply-To: Your message of "Sun, 24 Mar 1996 19:15:50 %2B1030." <199603240845.TAA24105@genesis.atrad.adelaide.edu.au>
next in thread | previous in thread | raw e-mail | index | archive | help
> You can discount the "normal Internet packet size" concept; there ain't > no such animal. As has been already observed, 576 is the minimum > permitted MSS. To be completely correct, 576 would be considered the minimum MTU. Given that, the associated TCP MSS on that path ought to be 536 bytes. > This is bogus arithmetic; lossage is a normally a point event and results > in the loss of one unit datagram around the point, regardless of its size. > This is why small-packet proocols (like kermit) fare better on noisy > uncorrected lines than large-packet protocols like Ymodem. The arithmetic is not completely bogus; the loss of a single fragment renders the remainder of the fragments useless. (That is unless the protocol stacks have become significantly smarter and TCP retransmissions of the same segments use the same IP ID. I've never actually observed this.) This same issue revisits us today when you run an IP datagram though the slicer and dicer and turn it into ATM cells. louie
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?199603241503.KAA21220>