ton o' dcc failures lately

Vernon Schryver
Tue Feb 27 20:24:01 UTC 2007

> From: "Jeff A. Earickson" 

> I've been getting a ton of these in my syslogs since upgrading
> to 1.3.51:
> continue not asking DCC 88 seconds after failure
> where the number varies.  Anything that I am doing wrong?
> Setup: Solaris 10, sendmail, MailScanner 4.58.9

Those messages say that your DCC client, dccifd, dccproc, or dccm,
is not getting answers from DCC servers and is erring on the side
of passing all mail including spam instead of delaying all mail for
4 seconds and still passing it along.

The command `cdcc info` dumps the /var/dcc/map file including the
current average round trip times and queuing delays for the known
DCC servers.

Is there any chance taht during the upgrade to 1.3.51, a change
to a firewall was made?  Blocking UDP port 6277 is the most common
reason why clients of the public DCC servers don't work.

I hope you did not use explicit IP addresses or particular host names
for public DCC servers instead of the generic names.  If so, it would
be best to delete the explicit addresses or host names with
   cdcc "delete whatever"
and then add the generic names with

    cdcc "add	RTT+1000 ms	anon"
    cdcc "add	RTT+1000 ms	anon"
    cdcc "add	RTT+1000 ms	anon"
    cdcc "add	RTT+1000 ms	anon"
    cdcc "add	RTT+1000 ms	anon"

I'm assuming you are not using a private DCC server.  If you have
a private server, you should ensure that all of your DCC clients
know about it with
    cdcc "add RTT-1000 ms 32768 secret1"
where 32768 and secret1 are a client-ID and matching password from
/var/dcc/ids on the server.

Vernon Schryver

More information about the DCC mailing list

Contact by mail or use the form.