Jonathan A. Zdziarski wrote:
> I'm sure other people have noticed this; I've been trying to track down
> where it came from - if you use LMTP/SMTP, and a message quarantines
> with the latest version, you wind up with ^Ms in the mbox, which causes
> all messages to appear blank in the UI.
> 
> Here's the patch for that

Good, that saves me a little trouble (since I've been meaning to find/fix that
for ages).  The patch was malformed (your mailer wrapped two lines) and it is
easier to read context diffs, but that may just be a personal preference.

Maybe you will spare a little brainpower for the next problem I've just recently
discovered:

When releasing messages from quarantine to e.g. a SMTP server, the envelope
sender (MAIL FROM: in the transaction) is blank and hence all messages appear to
come from <>.  Normally, no one would notice this, but I am filtering messages
that go to internal distribution lists (mini-listserv's) and the listserv
software refuses to accept mail from bounces (<>).

What I've done so far is to simply hardcode the MAIL FROM: field thusly:

Index: client.c
===================================================================
--- client.c    (revision 26)
+++ client.c    (working copy)
@@ -728,9 +728,11 @@

   if (proto == DDP_LMTP) {
     snprintf(buf, sizeof(buf), "MAIL FROM:<%s> SIZE=%ld",
-             ATX->mailfrom, (long) strlen(msg));
+             "[EMAIL PROTECTED]", (long) strlen(msg));
+/*             ATX->mailfrom, (long) strlen(msg)); */
   } else {
-    snprintf(buf, sizeof(buf), "MAIL FROM:<%s>", ATX->mailfrom);
+/*    snprintf(buf, sizeof(buf), "MAIL FROM:<%s>", ATX->mailfrom); */
+    snprintf(buf, sizeof(buf), "MAIL FROM:<%s>", "[EMAIL PROTECTED]");
   }

   if (send_socket(&TTX, buf)<=0) {

but the more correct fix would be to store the original envelope sender as the
FROM line in the mbox and retrieve that when releasing the message back.
Ideally, dspam should appear to be completely transparent to the original SMTP
transaction whenever possible (which would also require rewriting the code which
unfolds the header lines).

Thanks

John

-- 
John Peacock
Director of Information Research and Technology
Rowman & Littlefield Publishing Group
4501 Forbes Blvd
Suite H
Lanham, MD 20706
301-459-3366 x.5010
fax 301-429-5747

Reply via email to