Keith Ivey wrote:
I had another situation where a message was looping forever, being
repeatedly delivered from my server to itself, and causing a high load.
The problem is that the MX record looks like this:
conservation.com. 3600 IN MX 1000 0.0.0.0.
I've previously encountered the same problem with other domains. Since
the message is coming from inside my network and has a local FROM
address, it's accepted for relaying. Then because of the MX record it
gets "relayed" from the server to itself and queued for relaying again
and again.
With qmail-smtpd, this wouldn't go on for long, because there's
primitive loop detection, counting the number of "Received" and
"Delivered-To" lines. But qpsmtpd is missing that.
I've attached a plugin that duplicates the loop-checking in qmail-smtpd.
Actually, not detecting 0.0.0.0 as a local address is considered a "bug"
in qmail and a patch to fix it is one of the Recommended Patches [1] on
qmail.org.
[1] http://qmail.org/top.html#patches
Perhaps a better approach to fix this problem would be to add similar
checking to the qpsmtpd core code? It's only a couple of lines in qmail:
--- qmail-1.03/ipme.c Mon Jun 15 06:53:16 1998
+++ qmail-1.03-sg/ipme.c Mon Jan 29 02:27:38 2001
@@ -46,6 +46,11 @@
ipme.len = 0;
ix.pref = 0;
+ /* 0.0.0.0 is a special address which always refers to
+ * "this host, this network", according to RFC 1122, Sec. 3.2.1.3a.
+ */
+ byte_copy(&ix.ip,4,"\0\0\0\0");
+ if (!ipalloc_append(&ipme,&ix)) { return 0; }
if ((s = socket(AF_INET,SOCK_STREAM,0)) == -1) return -1;
len = 256;
I'm not sure where the corresponding change should be made in qpsmtpd.
R.
--
http://robinbowes.com
If a man speaks in a forest,
and his wife's not there,
is he still wrong?
