rejecting on Subject: ?

Vernon Schryver vjs@calcite.rhyolite.com
Tue Jan 8 20:29:06 UTC 2002


> From: Francois Baligant <francois@be.wanadoo.com>

> 	Ok, I didn't setup DCCM_CKSUMS nor DCCM_XTRA_CKSUMS
> 	so Im rejecting because of Fuz1=many.
>
> 	From what I can read in the source, the decoded
> 	body of the email is 'compressed' (remove punctations, spaces)
> 	and then a checksum is calculated.
>
> 	Won't all empty messages produce the same checksum
> 	and so push it until it reaches 'many' ?
>
> 	Im right or is the calculation different ?

The fuz1 checksum is not computed for small messages.  Notice the
"if (cks->ctx_fuz1.total < 20)" statement in dcc_ck_fuz1_fin().

The body checksum is computed for small but not empty messages.  Since
the it ignores whitespace and quote-printable noise, it can cause the
rejection of small messages.

Would it be bad to reject a message containing little more than blanks
and punctuation as the same as similar messages and so very "bulky?"

I'm uncomfortable talking about the details of fuzzy checksums in
public.  Releasing source for the checksum computations means that
the bad guys could in principle figure it out for themselves.  However,
many of the bad guys are unable to read C because of the congenital
traits that cause them to send spam.


Vernon Schryver    vjs@rhyolite.com



More information about the DCC mailing list

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