Subject: Re: rdist vs. sup
To: Ted Lemon , Randy Terbush <firstname.lastname@example.org>
From: Ronald Khoo <ronald@cpm.COM.MY>
Date: 03/03/1995 10:39:30
In your message dated Mar 2, you write:
> My only complaint about sup versus rdist is that sup doesn't appear to
> CRC the files that it transfers, so if the IP checksum fails, you've
> got a corrupt file and no way of knowing that it's corrupt.
Um. sup uses TCP, so an IP header checksum failure won't cause
a file corruption. Note, IP only checksums its own header, not the
data. However, I _have_ known TCP _data_ checksums to regularly
fail on a particularly bad link -- it took four or five tries to
get a 1 Meg file down across that link uncorrupted! So even the TCP
checksum isn't really adequate if you're in a backwater of the Internet.
So, yes, a confidence CRC at the end of the whole file (or better
still something longer e.g. like an MD5 signature) would be useful.
/* email@example.com +60 3 241 5232 | firstname.lastname@example.org +44 181 371 1000 */