Security Basics mailing list archives
Re: TCP/IP CRC question
From: Miles Stevenson <miles () mstevenson org>
Date: Thu, 7 Oct 2004 16:57:18 -0400
Jorge, I see that you have a lot of different, sometimes incorrect answers to this already. Alex was correct with his response though. Think of it like this... When the IP protocol was designed, it needed to run over any underlying technology, from Ethernet, to Token Ring, to whatever. Basically, it couldn't have been designed to "trust" the lower layers that it was getting correct data, which is why it implemented its own checksum. TCP took the same stance when it was designed, as did protocols on different layers of the stack. So, while you are right that it IS redundant to perform checksums on each layer of the stack instead of just doing it on one layer that encapsulates the whole packet (such as Ethernet), the protocols had to take this approach to remain independant of the other protocols. Good question though. -- Miles Stevenson miles () mstevenson org PGP FP: 035F 7D40 44A9 28FA 7453 BDF4 329F 889D 767D 2F63
Attachment:
_bin
Description:
Current thread:
- TCP/IP CRC question Jorge Mendez Bonini (Oct 06)
- RE: TCP/IP CRC question David Gillett (Oct 07)
- Re: TCP/IP CRC question Jonathan Loh (Oct 07)
- RE: TCP/IP CRC question Shaineel Singh (Oct 07)
- Re: TCP/IP CRC question Alex V . Lukyanenko (Oct 07)
- Re: TCP/IP CRC question Fernando Gont (Oct 07)
- RE: TCP/IP CRC question Shaineel Singh (Oct 07)
- Re: TCP/IP CRC question Ansgar -59cobalt- Wiechers (Oct 07)
- Re: TCP/IP CRC question Fernando Gont (Oct 07)
- Re: TCP/IP CRC question Miles Stevenson (Oct 07)
- Re: TCP/IP CRC question Miles Stevenson (Oct 08)
- RE: TCP/IP CRC question Clement Dupuis (Oct 12)
- Re: TCP/IP CRC question Miles Stevenson (Oct 08)
- <Possible follow-ups>
- Re: TCP/IP CRC question Don Parker (Oct 07)
- RE: TCP/IP CRC question Ted A (Oct 07)
- RE: TCP/IP CRC question Simon Zuckerbraun (Oct 13)