rcvd_nxt and IP checksum

Vernon Schryver vjs@calcite.rhyolite.com
Mon Apr 10 17:28:07 UTC 2006


> From: Sidney Markowitz 

> It seems like my most recent change to SpamAssassin is correct as far as it
> goes, but I also need to have the -a option added to dccproc and to make sure
> that dccifd is retried sufficiently often.

Yes, checking for dccifd would be particularly good.  For example, I
suspect that if utk.edu would install such a version of SpamAssassin
with a modern version of the DCC code, the number of completely useless
and wasted packets sent from puma.mail.utk.edu, aspirin.dii.utk.edu,
leopard.mail.utk.edu, jaguar.mail.utk.edu, panther.mail.utk.edu, and
cheetah.mail.utk.edu to the public DCC servers would decrease from
nearly 5,000,000 per day to about 1000 per day.

The University of Tennessee is a poster child for badly configured,
orphan installations.  They seem to use an ancient version of dccproc
with a broken firewall that blocks responses from UDP port 6277 on the
public DCC servers to their DCC requests.  It's not just the ~500 MBytes
or ~2 GBit of their own and the public DCC servers' bandwidth they waste
daily.  All of their incoming mail messages are being delayed by 4 to
15 seconds, depending on which dccproc and SpamAssassin timeouts rule.


thanks for fixing SpamAssassin,
Vernon Schryver    vjs@rhyolite.com



More information about the DCC mailing list

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