Well, 420 TCP Read Error is not a spamdyke error message. The remote
server seems to be running Novell GroupWise -- Google says this error is
caused by a slow/flaky connection between the GWIA server and the internet.
The only way I can imagine spamdyke being responsible is if you're
running a version earlier than 3.0.0, before I implemented the multibyte
writes. I can only think of three ways to try to troubleshoot this.
First, you could turn on full logging in spamdyke to see what happens
when these messages come in. If that doesn't work, you could use a
packet sniffer instead. Lastly, you could remove spamdyke entirely to
see if the errors stop.
-- Sam Clippinger
Joe Nelson wrote:
I've had an interesting issue recently where I've been unable to
receive emails from a particular email server. Unfortunately it's a
state email server that many of my users receive email from. They've
all reported the following error: 420 TCP Read Error. The server is
at 168.180.96.58. I've verified that I can telnet to port 25 from my
mail server so I know that there's no firewall blocking going on.
I've whitelisted their IP address just to see if that will bypass
anything in Spamdyke but users are still reporting the 420 TCP Read
Error message. Is this something that others have seen? If so, where
should I start looking for a solution?
Here's my spamdyke.conf for reference:
log-level=3
local-domains-file=/var/qmail/control/rcpthosts
rdns-whitelist-file=/var/qmail/control/spamdyke/whitelist_rdns_names
ip-whitelist-file=/var/qmail/control/spamdyke/whitelist_ips
ip-in-rdns-keyword-file=/var/qmail/control/spamdyke/blacklist_rdns_keywords
recipient-whitelist-file=/var/qmail/control/spamdyke/whitelist_recipient
sender-whitelist-file=/var/qmail/control/spamdyke/whitelist_sender
check-dnsrbl=sbl-xbl.spamhaus.org
check-dnsrbl=bl.spamcop.net
reject-missing-sender-mx
reject-ip-in-cc-rdns
reject-empty-rdns
___
spamdyke-users mailing list
spamdyke-users@spamdyke.org
http://www.spamdyke.org/mailman/listinfo/spamdyke-users
