Autoresponder and default address

casey

Well-Known Member
Jan 17, 2003
2,288
0
191
I have an issue with the autoresponder. I have my default address set to :fail:, and I have an autoresponder [email protected]. tos is not a mailbox, only an autoresponder. The autoresponder works, but I also get a "Mail Delivery Failed" message. Now, obviously, I don't want this failure message. What gives?
 

casey

Well-Known Member
Jan 17, 2003
2,288
0
191
eos1 said:
Did you try :blackhole: instead of :fail:?
No, because I don't want to waste the bandwidth. I should add that forwarders work fine. It's just the autoresponders that give this message...unless there's an actual account.
 

casey

Well-Known Member
Jan 17, 2003
2,288
0
191
SarcNBit said:
What happens if you set the default address to [email protected] and then create a forwarder for [email protected] pointing to :fail:? Do the messages still generate bounces?
Interestingly enough, the same thing happens. The autoresponse is received, but so is a message saying there was a delivery failure.

One more piece of information that I just noticed: if I do the autoresponder, I get a mail delivery failure message from my server. If I send to a non-existant e-mail address, however, the error is from the mailer-daemon of the smtp server I am sending from.
 

SarcNBit

Well-Known Member
Oct 14, 2003
1,001
3
168
I have not had a chance to test this yet (I will later today if the problem persists for you). One more thing to test: create a forwarder with the same address as your auto-responder and set it to :blackhole:.
 

casey

Well-Known Member
Jan 17, 2003
2,288
0
191
SarcNBit said:
I have not had a chance to test this yet (I will later today if the problem persists for you). One more thing to test: create a forwarder with the same address as your auto-responder and set it to :blackhole:.
Weird. I guess the forwarders overrule the autoresponders. If I do that the autoresponder ceases to function. This is very strange.

If worse comes to worst I'll just tell my clients they have to use real mailboxes with the autoresponders.
 

SarcNBit

Well-Known Member
Oct 14, 2003
1,001
3
168
I just tested this and saw the same results. Here is a tail of the exim log (names changed to implicate the guilty):

2004-08-25 21:56:19 1C09VJ-0007QP-GY <= [email protected] H=(mproxy.gmail.com) [17.233.170.249] P=esmtp S=840 [email protected]
2004-08-25 21:56:19 1C09VJ-0007QP-GY ** [email protected] R=virtual_aliases:
2004-08-25 21:56:19 1C09VJ-0002zP-ML <= [email protected] U=cpdom P=local S=659
2004-08-25 21:56:19 1C09VJ-0007QP-GY => |/usr/local/cpanel/bin/autorespond [email protected] /home/cpdom/.autorespond ([email protected]) <[email protected]> R=virtual_aliases_nostar T=virtual_address_pipe
2004-08-25 21:56:19 1C09VJ-0004Qv-Op <= <> R=3C09VJ-0007QR-GY U=mailnull P=local S=1656
2004-08-25 21:56:19 1C09VJ-0007QP-GY Completed
2004-08-25 21:56:20 1C09VJ-0002zP-ML => [email protected] R=lookuphost T=remote_smtp H=gsmtp171.google.com [17.233.171.74]
2004-08-25 21:56:20 1C09VJ-0002zP-ML Completed
2004-08-25 21:56:20 1C09VJ-0004Qv-Op => [email protected] R=lookuphost T=remote_smtp H=gsmtp171.google.com [17.233.171.74]
2004-08-25 21:56:20 1C09VJ-0004Qv-Op Completed

Seems like a bug to me. I have not had anyone complain about it (surprisingly), but I do not think that is the way it should (or used to for that matter) work. I would suggest submitting to bugzilla and posting back the ticket # :)
 

casey

Well-Known Member
Jan 17, 2003
2,288
0
191
SarcNBit said:
I just tested this and saw the same results. Here is a tail of the exim log (names changed to implicate the guilty):
Seems like a bug to me. I have not had anyone complain about it (surprisingly), but I do not think that is the way it should (or used to for that matter) work. I would suggest submitting to bugzilla and posting back the ticket # :)
Will do. Thanks for proving I'm not crazy. :)


Logged as Bugzilla #1096
 
Last edited:

SarcNBit

Well-Known Member
Oct 14, 2003
1,001
3
168
Has this bug been fixed in the most recent builds? If not, it should be added to bdraco's STABLE release bug thread.
 

SarcNBit

Well-Known Member
Oct 14, 2003
1,001
3
168
We are not running the latest edge build anywhere at the moment so I have no idea if the issue is ongoing. If it is, then I think it should be added to this thread.
 

casey

Well-Known Member
Jan 17, 2003
2,288
0
191
Changelog says it is fixed, but it is not. It still doesn't work properly for me.

9.9.8 (build 80) Sat Nov 6 14:16:48 2004

Resolve a problem where autoresponders would fail if there was no pop account
at the same address when the default address for the domain was set to :fail:
 

makeityourname

Registered
Apr 23, 2005
1
0
151
Bugzilla says fixed in 9.9, but is it back in x?

I've read just about every message regarding this autoresponder problem and have yet to find an actual resolution, which is quite frustrating :(

I'm not a reseller nor do we host sites other than our own, and we are getting bombarded with 100's of failed messages (from forged headers of course) to all the autoresponders we use.

- I've set the Default address to :fail:
- Also created Forwarders for each of the autoresponders thinking
it would solve the failed messages, but all that does is allow
the sp*m to get through
- Tried to create a Forwarder that goes to :blackhole:, but then
the autoresponder doesn't work

If we were hosting customers, I can't imagine the amount of complaints we would be getting!

I've put in a support ticket and all they really suggest is to use filters through our email client, which is not acceptable as we'll still be downloading 100's of messages every few hours.

With our last host, they were able to setup a throw_away address so that autoresponders could still get sent and if there's a forged header, it quietly disappears.

Is there an exim hack? Does anyone have the answer for this very annoying problem?

Thank you,

Barb