about Fuz1 and Fuz2

Vernon Schryver vjs@calcite.rhyolite.com
Tue Jun 8 15:25:11 UTC 2004

> From: Maxim Cherniavsky <maxim@mtu.ru>

> I'm trying to integrate DCC checksums into postfix, so I've ripped all 
> logic from dccproc and dccifd, and ported it into postfix-cleanup. I 
> have a question of internal meaning of Fuz1 and Fuz2 checksums - for 
> what kind of messages they  are computed? For some messages dcc server 
> reports only  body and Fuz1 checksums, and Fuz2 is not reported; for 
> some messages only Body checksum is reported. Must they exist for any 
> kind of message? I suppose that I've missed something, but cant find what :(

That is the wrong tactic.  When I change the checksumming machinery,
that code will stop working.  Based on similar efforts by other people
and judging from the question, the DCC client-server will probably be
broken and cause problems for the public DCC servers.  For example,
that is why the barracudanetworks.com test systems have long been in
the defensive blacklist used by the public DCC servers.

Instead use the Perl interface to dccifd in dccifd/dccif-test/dccif-test.pl
in the DCC source or the dccif() interface function in the dcclib.a
library as demonstrated by dccifd/dccif-test/dccif-test.c.  See 

Connecting MTAs to DCC servers is the reason dccifd exists, which makes
talk about removing code from dccifd to postfix-cleanup surprising.
Again, look at dccifd/dccif-test/dccif-test.pl and

Even better would probably be to use one of the existing implementations
of dccifd and dccproc in postfix.  See

Vernon Schryver    vjs@rhyolite.com

More information about the DCC mailing list

Contact vjs@rhyolite.com by mail or use the form.