[exim] Fwd: Failed smtp auth /24's

Startseite
Nachricht löschen
Nachricht beantworten
Autor: Todd Lyons
Datum:  
To: exim-users
Betreff: [exim] Fwd: Failed smtp auth /24's
Just thought I'd put this here since it mentions Lena's excellent code.

...Todd


---------- Forwarded message ----------
From: Todd Lyons
Date: Wed, Jul 25, 2012 at 9:45 AM
Subject: Failed smtp auth /24's
To: patternity ML


I found several swamps that are ending up with lots of smtp auth
failures. I'm thinking of just outright blocking connections from
these /24's very early on in the connection. Can anybody give me some
reasons why I shouldn't just outright block them? Only thing I can
come up with is that we might have a legitimate customer in one of
those netblocks.

I'm also going to implement part of Lena's (from the exim mailing
list) brute force detection because that forced 22 second delay after
the second and beyond failed auth attempt seems like a very good way
to slow down their brute force attempts...if they're doing that. I
have not detected a blast of smtp auth attempts, it's usually spread
out evenly over the day or week.

Some of these are accounts that I detected as being compromised last
week and earlier this week. In those cases, I just change the
password for the mailbox so that it can still receive inbound mail,
but stops all smtp auth outbound traffic. That caused a blast of
failures as some botnet was actively exploiting the mailbox. The rest
of these accounts are probably being actively probed for the password.
This is only on one mail server in a load balanced system.

Look below for log analysis that shows:
1. First group is the number of failed smtp auth connections for a
specific mailbox.
2. Second group is the number of IP addresses in each /24 that
attempted smtp auth).

CentOS58[root@ivwm51 ~]# egrep "authenticator failed.*535 "
/var/log/exim/main.log | perl failed_logins.pl --limit 20
251 => customerservice@???
280 => david@???
304 => foodorders@???
319 => health@???
68 => info@???
265 => info@???
260 => infonsales@???
331 => jordan@???
271 => jparren@???
247 => justtowels@???
348 => leslie@???
336 => ljackson@???
316 => mo3@???
329 => nb4@???
62 => newsletter@???
303 => nh10@???
75 => robert@???

110.52.10.*   => 40
110.52.6.*    => 142
110.52.7.*    => 139
110.52.9.*    => 40
110.53.24.*   => 176
110.53.25.*   => 168
110.53.26.*   => 167
110.53.27.*   => 178
110.53.30.*   => 49
110.53.31.*   => 52
115.58.132.*  => 25
115.63.10.*   => 44
115.63.11.*   => 38
115.63.12.*   => 51
115.63.13.*   => 46
115.63.14.*   => 39
115.63.15.*   => 49
115.63.8.*    => 54
115.63.9.*    => 39
125.44.240.*  => 22
125.44.241.*  => 21
125.44.242.*  => 25
125.44.243.*  => 26
125.44.247.*  => 26
222.140.163.* => 23
42.49.128.*   => 168
42.49.132.*   => 49
42.49.133.*   => 61
42.49.136.*   => 40
42.49.137.*   => 52
42.49.138.*   => 47
42.49.139.*   => 41
42.49.140.*   => 39



...Todd
--
The total budget at all receivers for solving senders' problems is $0.
If you want them to accept your mail and manage it the way you want,
send it the way the spec says to. --John Levine


--
The total budget at all receivers for solving senders' problems is $0.
If you want them to accept your mail and manage it the way you want,
send it the way the spec says to. --John Levine