This is a discussion on Excepting mail ONLY for valid users within the alt.comp.mail.qmail forums, part of the Mail Servers and Related category; Hello, I've been experiencing a problem lately with spammers forging their Reply-To and From headers and inserting '<...
|
|||||||
| FAQ | Members List | Calendar | Search | Today's Posts | Mark Forums Read |
|
|||
|
Hello,
I've been experiencing a problem lately with spammers forging their Reply-To and From headers and inserting '<random_username>@mydomain.com' in these fields. Obviously, most of their email fails to be delivered and the recepient's mail server replies to this non-existent user notifying him of the failure. This in turn bounces to postmaster (me) reporting that <random_username>@mydomain.com doesn't exist, which is of course true since the bloody spammer just made it up. Is there a patch for qmail to check for the validity of the destination email address before accepting the message? I realise this may give the server a performance hit, but I'm more than willing to pay that price in exchange of the 60 failure notices I've received today. Thanks, Mark |
|
|||
|
Mark Kaufer wrote:
> > Hello, > > I've been experiencing a problem lately with spammers forging their > Reply-To and From headers and inserting '<random_username>@mydomain.com' > in these fields. > > Obviously, most of their email fails to be delivered and the recepient's > mail server replies to this non-existent user notifying him of the > failure. This in turn bounces to postmaster (me) reporting that > <random_username>@mydomain.com doesn't exist, which is of course true > since the bloody spammer just made it up. > > Is there a patch for qmail to check for the validity of the destination > email address before accepting the message? I realise this may give the > server a performance hit, but I'm more than willing to pay that price in > exchange of the 60 failure notices I've received today. There is an interesting read on this subject here: http://groups.google.com/groups?thre...5.ctd.ornl.gov |
|
|||
|
On Wed, 21 Jan 2004 07:30:25 -0500, Peter Kleiner <kleinerp@NOSPAM.smbmicro.com> wrote to alt.comp.mail.qmail:
> Mark Kaufer wrote: >> >> Hello, >> >> I've been experiencing a problem lately with spammers forging their >> Reply-To and From headers and inserting '<random_username>@mydomain.com' >> in these fields. >> >> Obviously, most of their email fails to be delivered and the recepient's >> mail server replies to this non-existent user notifying him of the >> failure. This in turn bounces to postmaster (me) reporting that >> <random_username>@mydomain.com doesn't exist, which is of course true >> since the bloody spammer just made it up. >> >> Is there a patch for qmail to check for the validity of the destination >> email address before accepting the message? I realise this may give the >> server a performance hit, but I'm more than willing to pay that price in >> exchange of the 60 failure notices I've received today. > > There is an interesting read on this subject here: > http://groups.google.com/groups?thre...5.ctd.ornl.gov Thanks for the reply, Peter. I've implemented Dave Sill's control/doublebounceto, but unfortunately, it doeesn't look like it worked. Is there a patch that needs to be applied for that to function properly? I actually have already patched my qmail w/ qmail-1.03-mfcheck.3.patch and that works great. http://www.interazioni.it/qmail/ looks very interesting; I'll have to try that out. Cheers, Mark |
|
|||
|
Mark Kaufer <mark.kaufer@NO.SPAM.bloodyhell.us> writes:
> I've implemented Dave Sill's control/doublebounceto, but unfortunately, > it doeesn't look like it worked. Is there a patch that needs to be > applied for that to function properly? No. What did you do, exactly? How is it working? -- Dave Sill Oak Ridge National Lab, Workstation Support Author, The qmail Handbook <http://web.infoave.net/~dsill> <http://lifewithqmail.org/>: Almost everything you always wanted to know. |
|
|||
|
On Thu, 22 Jan 2004 09:38:12 -0500, Dave Sill <MaxFreedom@sws5.ornl.gov> wrote to alt.comp.mail.qmail:
> Mark Kaufer <mark.kaufer@NO.SPAM.bloodyhell.us> writes: > >> I've implemented Dave Sill's control/doublebounceto, but unfortunately, >> it doeesn't look like it worked. Is there a patch that needs to be >> applied for that to function properly? > > No. What did you do, exactly? How is it working? I created /var/qmail/control/doublebounceto which is owned by root.root and chmod 644: -rw-r--r-- 1 root root 13 Jan 21 10:13 doublebounceto In it is 'doublebounce': # cat doublebounceto doublebounce Then in ~alias, I created .qmail-doublebounce which is owned by root.qmail and is also 644: -rw-r--r-- 1 root qmail 2 Jan 21 10:13 .qmail-doublebounce In ~alias/.qmail-doublebounce, is a single hash: # cat ~alias/.qmail-doublebounce # From what I read, it looked like that was pretty much it. Am I missing something? Cheers, Mark |
|
|||
|
Mark Kaufer <mark.kaufer@NO.SPAM.bloodyhell.us> writes:
> I created /var/qmail/control/doublebounceto which is owned by root.root > and chmod 644: > > -rw-r--r-- 1 root root 13 Jan 21 10:13 doublebounceto > > In it is 'doublebounce': > > # cat doublebounceto > doublebounce > > Then in ~alias, I created .qmail-doublebounce which is owned by > root.qmail and is also 644: > > -rw-r--r-- 1 root qmail 2 Jan 21 10:13 .qmail-doublebounce > > In ~alias/.qmail-doublebounce, is a single hash: > > # cat ~alias/.qmail-doublebounce > # > > From what I read, it looked like that was pretty much it. Am I missing > something? Nope, that's it. Now show me the logs for a double bounce. You can create one by injecting a message via telnet to port 25 with a bogus sender and recipient. See step 7 in TEST.deliver. -- Dave Sill Oak Ridge National Lab, Workstation Support Author, The qmail Handbook <http://web.infoave.net/~dsill> <http://lifewithqmail.org/>: Almost everything you always wanted to know. |