Path: utzoo!utgpu!news-server.csri.toronto.edu!mailrus!uunet!aplcen!samsung!xylogics!merk!alliant!linus!think!barmar
From: barmar@think.com (Barry Margolin)
Newsgroups: comp.protocols.tcp-ip
Subject: Re: UDP checksums
Keywords: UDP checksum sun
Message-ID: <40819@think.Think.COM>
Date: 19 Jul 90 06:46:38 GMT
References: <5@wotk.UUCP> <1990Jul17.003242.1556@ultra.com> <1812@hsi86.hsi.UUCP>
Sender: news@Think.COM
Organization: Thinking Machines Corporation, Cambridge MA, USA
Lines: 20

In article <1812@hsi86.hsi.UUCP> stevens@hsi.com (Richard Stevens) writes:
>In article <1990Jul17.003242.1556@ultra.com> beau@ultra.com (Beau James writes:
>>UDP checksums must be enabled on both sending and receiving

>Newer systems should start conforming to the Host Requirements RFC (1122),
>meaning the receiver has to honor a checksum that was set by the sender:

The question was specifically about enabling checksums in current BSD
systems, in which case the caveat is correct.  Is this bug fixed in 4.4bsd?

Also, if the systems were conforming to the HR RFC it wouldn't be necessary
to turn on UDP checksums.  I believe the RFC also specifies that they MUST
be enabled by default, and SHOULD provide the ability for the application
or system administrator to disable them.

--
Barry Margolin, Thinking Machines Corp.

barmar@think.com
{uunet,harvard}!think!barmar