Re: IP Banning (Using IPFW)

2006-02-09 Thread Daniel A.
On 2/9/06, Chris [EMAIL PROTECTED] wrote:
 On 07/02/06, David Scheidt [EMAIL PROTECTED] wrote:
 
  On Tue, Feb 07, 2006 at 12:40:22AM +0200, Atis wrote:
   On Sun, 5 Feb 2006 18:55:13 -0500
   David Scheidt [EMAIL PROTECTED] wrote:
  
   
Nonsense.  There may be some people that only scan well-known ports,
but it's much more common to scan every port on a machine.  If you're
running a server on a non-standard port, an attacker will find it.
   
  
   sure, but 99% of the time the machines attacking your server are zombies
   that do not care to do a full portscan. i suppose the purpose is to
   find other misconfigured, easy-to-hack computers on the network. by
   putting your services on non-standard ports you get rid of these
   mindless drones and don't pollute log files with useless garbage.
  
   now if somebody _does_ actually target your server in particular then
   this is definitely not the solution.
  
   anywayz, putting things on non-standard ports helps a lot, and is
   one of the first and easiest security measures an administrator
   may consider.
  
 
  Taking your clothes off and painting yourself blue is also one of the
  first and easiest security measures to consider.  It's even more
  effective, too.  I know of no machine that's been cracked that had a
  wheel naked and painted blue.  I've seen lots running standard
  services on non-standard ports.
 
  Security through obscurity doesn't work, it makes tracking down
  other problems harder, and creates work to maintain non-standard
  configurations.


 I understand his point, I see 2 types of problems we have to deal with.  The
 thousands of drones that scan for boxes that are vulnerable to a specific
 exploit, they will often scan ip ranges on a specific port and if its open
 see if its vulnerable.  For these types of intruders chnging ports is very
 effective since you would simply be skipped past on their scan, for most of
 us 99% of attempted intrusions are zombie based or some script a kid has
 downloaded of the web.

 The argument against changing ports is of course when you have a persistent
 hacker who wants in, he will of course scan all the ports and find the
 service and this type of protection is nullified.  In this scenario if you
 havent taken additional measures to secure the box then you may be in
 trouble,

 I personally move things like sshd of its normal port simply to stop my logs
 been flooded with brute force logins and since I am the only one who uses
 ssh there is no downside to it, I of course dont rely on this alone and keep
 my software up to date amongst other security measures it is simply an extra
 layer of skin on the onion.  For things like httpd I keep on port 80 as I
 think moving the port of that is more hassle then its worth.
I've seen someone mention how to move httpd to a non-reserved port (ie
8080), and let that change be transparent for the end-user by using
ipf. I dont know how, though.

 Chris
 ___
 freebsd-questions@freebsd.org mailing list
 http://lists.freebsd.org/mailman/listinfo/freebsd-questions
 To unsubscribe, send any mail to [EMAIL PROTECTED]

___
freebsd-questions@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-questions
To unsubscribe, send any mail to [EMAIL PROTECTED]


Re: IP Banning (Using IPFW)

2006-02-08 Thread Chris
On 07/02/06, David Scheidt [EMAIL PROTECTED] wrote:

 On Tue, Feb 07, 2006 at 12:40:22AM +0200, Atis wrote:
  On Sun, 5 Feb 2006 18:55:13 -0500
  David Scheidt [EMAIL PROTECTED] wrote:
 
  
   Nonsense.  There may be some people that only scan well-known ports,
   but it's much more common to scan every port on a machine.  If you're
   running a server on a non-standard port, an attacker will find it.
  
 
  sure, but 99% of the time the machines attacking your server are zombies
  that do not care to do a full portscan. i suppose the purpose is to
  find other misconfigured, easy-to-hack computers on the network. by
  putting your services on non-standard ports you get rid of these
  mindless drones and don't pollute log files with useless garbage.
 
  now if somebody _does_ actually target your server in particular then
  this is definitely not the solution.
 
  anywayz, putting things on non-standard ports helps a lot, and is
  one of the first and easiest security measures an administrator
  may consider.
 

 Taking your clothes off and painting yourself blue is also one of the
 first and easiest security measures to consider.  It's even more
 effective, too.  I know of no machine that's been cracked that had a
 wheel naked and painted blue.  I've seen lots running standard
 services on non-standard ports.

 Security through obscurity doesn't work, it makes tracking down
 other problems harder, and creates work to maintain non-standard
 configurations.


I understand his point, I see 2 types of problems we have to deal with.  The
thousands of drones that scan for boxes that are vulnerable to a specific
exploit, they will often scan ip ranges on a specific port and if its open
see if its vulnerable.  For these types of intruders chnging ports is very
effective since you would simply be skipped past on their scan, for most of
us 99% of attempted intrusions are zombie based or some script a kid has
downloaded of the web.

The argument against changing ports is of course when you have a persistent
hacker who wants in, he will of course scan all the ports and find the
service and this type of protection is nullified.  In this scenario if you
havent taken additional measures to secure the box then you may be in
trouble,

I personally move things like sshd of its normal port simply to stop my logs
been flooded with brute force logins and since I am the only one who uses
ssh there is no downside to it, I of course dont rely on this alone and keep
my software up to date amongst other security measures it is simply an extra
layer of skin on the onion.  For things like httpd I keep on port 80 as I
think moving the port of that is more hassle then its worth.

Chris
___
freebsd-questions@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-questions
To unsubscribe, send any mail to [EMAIL PROTECTED]


Re: IP Banning (Using IPFW)

2006-02-06 Thread Atis
On Sun, 5 Feb 2006 18:55:13 -0500
David Scheidt [EMAIL PROTECTED] wrote:

 
 Nonsense.  There may be some people that only scan well-known ports,
 but it's much more common to scan every port on a machine.  If you're
 running a server on a non-standard port, an attacker will find it.
 

sure, but 99% of the time the machines attacking your server are zombies
that do not care to do a full portscan. i suppose the purpose is to
find other misconfigured, easy-to-hack computers on the network. by
putting your services on non-standard ports you get rid of these
mindless drones and don't pollute log files with useless garbage.

now if somebody _does_ actually target your server in particular then
this is definitely not the solution.

anywayz, putting things on non-standard ports helps a lot, and is
one of the first and easiest security measures an administrator
may consider.


Atis
___
freebsd-questions@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-questions
To unsubscribe, send any mail to [EMAIL PROTECTED]


Re: IP Banning (Using IPFW)

2006-02-06 Thread David Scheidt
On Tue, Feb 07, 2006 at 12:40:22AM +0200, Atis wrote:
 On Sun, 5 Feb 2006 18:55:13 -0500
 David Scheidt [EMAIL PROTECTED] wrote:
 
  
  Nonsense.  There may be some people that only scan well-known ports,
  but it's much more common to scan every port on a machine.  If you're
  running a server on a non-standard port, an attacker will find it.
  
 
 sure, but 99% of the time the machines attacking your server are zombies
 that do not care to do a full portscan. i suppose the purpose is to
 find other misconfigured, easy-to-hack computers on the network. by
 putting your services on non-standard ports you get rid of these
 mindless drones and don't pollute log files with useless garbage.
 
 now if somebody _does_ actually target your server in particular then
 this is definitely not the solution.
 
 anywayz, putting things on non-standard ports helps a lot, and is
 one of the first and easiest security measures an administrator
 may consider.
 

Taking your clothes off and painting yourself blue is also one of the
first and easiest security measures to consider.  It's even more
effective, too.  I know of no machine that's been cracked that had a
wheel naked and painted blue.  I've seen lots running standard
services on non-standard ports.

Security through obscurity doesn't work, it makes tracking down
other problems harder, and creates work to maintain non-standard
configurations.

David
___
freebsd-questions@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-questions
To unsubscribe, send any mail to [EMAIL PROTECTED]


Re: IP Banning (Using IPFW)

2006-02-05 Thread Philip Hallstrom
I was wondering if there's some sort of port available that can actively ban 
IPs that try and bruteforce a service such as SSH or Telnet, by scanning the 
/var/log/auth.log log for Regex such as Illegal User or LOGIN FAILURES, 
and then using IPFW to essentially deny (ban) that IP for a certain period of 
time or possibly forever.


I've seen a very useful one that works for linux (fail2ban), and was 
wondering if one exists for FreeBSD's IPFW?


There are some in the ports, but you can write your own pretty easy too. 
The one thing I didn't like about the ones in the ports is the app was 
responsible for removing the rules after a set amount of time.  Which 
could be a problem if that app crashed for some reason.  You could lock 
yourself out permanently...


Here's a quick perl script I wrote that does what you want...

http://pastebin.com/540575

Combine that with these two crontab entries:

0-59/4 * * * * /sbin/ipfw delete 501 /dev/null 21
2-59/4 * * * * /sbin/ipfw delete 500 /dev/null 21

-philip
___
freebsd-questions@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-questions
To unsubscribe, send any mail to [EMAIL PROTECTED]


RE: IP Banning (Using IPFW)

2006-02-05 Thread fbsd_user
I find this kind of approach is treating the symptom and not the
cause.
The basic problem is the services have well published port numbers
and attackers beat on those known port numbers. A much simpler
approach is to change the standard port numbers to some high order
port number. See /etc/services  SSH logon command allows for a port
number and the same for telnet. Your remote users will be the only
people knowing your selected port numbers for those services. This
way a attackers port scan will show the well published port numbers
as not open so they will pass on attacking those ports on your ip
address. This way your bandwidth usage will be reduced as attackers
find your ip address as having nothing of interest.

This same kind of thing can also be done for port 80 by using the
web forwarding function of Zoneedit pointing to different port for
your web server. Only people coming to your site through dns will be
forwarded to the correct port.

The clear key here is attackers roll through a large range of ip
address port scanning for open ports. By using nonstandard port
numbers for your services you stop the attacker even finding you in
the first place.

good luck what ever you choose to do.

-Original Message-
From: [EMAIL PROTECTED]
[mailto:[EMAIL PROTECTED] Behalf Of Michael A.
Alestock
Sent: Sunday, February 05, 2006 10:42 AM
To: [EMAIL PROTECTED]
Subject: IP Banning (Using IPFW)
Importance: High


Hello,

I was wondering if there's some sort of port available that can
actively
ban IPs that try and bruteforce a service such as SSH or Telnet, by
scanning the /var/log/auth.log log for Regex such as Illegal User
or
LOGIN FAILURES, and then using IPFW to essentially deny (ban) that
IP
for a certain period of time or possibly forever.

I've seen a very useful one that works for linux (fail2ban), and was
wondering if one exists for FreeBSD's IPFW?

I've looked around in /usr/ports/security and /usr/ports/net but
can't
seem to find anything that closely resembles that.

Your help would be greatly appreciated Thanks in advance!

 Michael A., USA... Loyal FreeBSD user since 2000.
___
freebsd-questions@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-questions
To unsubscribe, send any mail to
[EMAIL PROTECTED]

___
freebsd-questions@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-questions
To unsubscribe, send any mail to [EMAIL PROTECTED]


Re: IP Banning (Using IPFW)

2006-02-05 Thread Daniel A.
On 2/5/06, fbsd_user [EMAIL PROTECTED] wrote:
 I find this kind of approach is treating the symptom and not the
 cause.
 The basic problem is the services have well published port numbers
 and attackers beat on those known port numbers. A much simpler
 approach is to change the standard port numbers to some high order
 port number. See /etc/services  SSH logon command allows for a port
 number and the same for telnet. Your remote users will be the only
 people knowing your selected port numbers for those services. This
 way a attackers port scan will show the well published port numbers
 as not open so they will pass on attacking those ports on your ip
 address. This way your bandwidth usage will be reduced as attackers
 find your ip address as having nothing of interest.

 This same kind of thing can also be done for port 80 by using the
 web forwarding function of Zoneedit pointing to different port for
 your web server. Only people coming to your site through dns will be
 forwarded to the correct port.

 The clear key here is attackers roll through a large range of ip
 address port scanning for open ports. By using nonstandard port
 numbers for your services you stop the attacker even finding you in
 the first place.

 good luck what ever you choose to do.
You just argued against yourself. If an attacker is genuinely
interested in rooting someones box, that attacker will most likely
portscan the box - And thereby discovering that you have assigned
alternative port numbers to your services.
Security through obscurity is a bad place to start.

 -Original Message-
 From: [EMAIL PROTECTED]
 [mailto:[EMAIL PROTECTED] Behalf Of Michael A.
 Alestock
 Sent: Sunday, February 05, 2006 10:42 AM
 To: [EMAIL PROTECTED]
 Subject: IP Banning (Using IPFW)
 Importance: High


 Hello,

 I was wondering if there's some sort of port available that can
 actively
 ban IPs that try and bruteforce a service such as SSH or Telnet, by
 scanning the /var/log/auth.log log for Regex such as Illegal User
 or
 LOGIN FAILURES, and then using IPFW to essentially deny (ban) that
 IP
 for a certain period of time or possibly forever.

 I've seen a very useful one that works for linux (fail2ban), and was
 wondering if one exists for FreeBSD's IPFW?

 I've looked around in /usr/ports/security and /usr/ports/net but
 can't
 seem to find anything that closely resembles that.

 Your help would be greatly appreciated Thanks in advance!

  Michael A., USA... Loyal FreeBSD user since 2000.
 ___
 freebsd-questions@freebsd.org mailing list
 http://lists.freebsd.org/mailman/listinfo/freebsd-questions
 To unsubscribe, send any mail to
 [EMAIL PROTECTED]

 ___
 freebsd-questions@freebsd.org mailing list
 http://lists.freebsd.org/mailman/listinfo/freebsd-questions
 To unsubscribe, send any mail to [EMAIL PROTECTED]

___
freebsd-questions@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-questions
To unsubscribe, send any mail to [EMAIL PROTECTED]


RE: IP Banning (Using IPFW)

2006-02-05 Thread fbsd_user
You missed to whole meaning.
Attackers only scan for the published service port numbers,
that is what is meant by portscan the box.
Those high order port numbers are dynamically
used during normal session conversation.
So any response from those port numbers if an
attacker scanned that high would be meaningless.
Please check your facts before commenting.

-Original Message-
From: [EMAIL PROTECTED]
[mailto:[EMAIL PROTECTED] Behalf Of Daniel A.
Sent: Sunday, February 05, 2006 4:58 PM
To: [EMAIL PROTECTED]
Cc: [EMAIL PROTECTED]; Michael A. Alestock
Subject: Re: IP Banning (Using IPFW)


On 2/5/06, fbsd_user [EMAIL PROTECTED] wrote:
 I find this kind of approach is treating the symptom and not the
 cause.
 The basic problem is the services have well published port numbers
 and attackers beat on those known port numbers. A much simpler
 approach is to change the standard port numbers to some high order
 port number. See /etc/services  SSH logon command allows for a
port
 number and the same for telnet. Your remote users will be the only
 people knowing your selected port numbers for those services. This
 way a attackers port scan will show the well published port
numbers
 as not open so they will pass on attacking those ports on your ip
 address. This way your bandwidth usage will be reduced as
attackers
 find your ip address as having nothing of interest.

 This same kind of thing can also be done for port 80 by using the
 web forwarding function of Zoneedit pointing to different port for
 your web server. Only people coming to your site through dns will
be
 forwarded to the correct port.

 The clear key here is attackers roll through a large range of ip
 address port scanning for open ports. By using nonstandard port
 numbers for your services you stop the attacker even finding you
in
 the first place.

 good luck what ever you choose to do.
You just argued against yourself. If an attacker is genuinely
interested in rooting someones box, that attacker will most likely
portscan the box - And thereby discovering that you have assigned
alternative port numbers to your services.
Security through obscurity is a bad place to start.

 -Original Message-
 From: [EMAIL PROTECTED]
 [mailto:[EMAIL PROTECTED] Behalf Of Michael
A.
 Alestock
 Sent: Sunday, February 05, 2006 10:42 AM
 To: [EMAIL PROTECTED]
 Subject: IP Banning (Using IPFW)
 Importance: High


 Hello,

 I was wondering if there's some sort of port available that can
 actively
 ban IPs that try and bruteforce a service such as SSH or Telnet,
by
 scanning the /var/log/auth.log log for Regex such as Illegal
User
 or
 LOGIN FAILURES, and then using IPFW to essentially deny (ban)
that
 IP
 for a certain period of time or possibly forever.

 I've seen a very useful one that works for linux (fail2ban), and
was
 wondering if one exists for FreeBSD's IPFW?

 I've looked around in /usr/ports/security and /usr/ports/net but
 can't
 seem to find anything that closely resembles that.

 Your help would be greatly appreciated Thanks in advance!

  Michael A., USA... Loyal FreeBSD user since 2000.
 ___
 freebsd-questions@freebsd.org mailing list
 http://lists.freebsd.org/mailman/listinfo/freebsd-questions
 To unsubscribe, send any mail to
 [EMAIL PROTECTED]

 ___
 freebsd-questions@freebsd.org mailing list
 http://lists.freebsd.org/mailman/listinfo/freebsd-questions
 To unsubscribe, send any mail to
[EMAIL PROTECTED]

___
freebsd-questions@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-questions
To unsubscribe, send any mail to
[EMAIL PROTECTED]

___
freebsd-questions@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-questions
To unsubscribe, send any mail to [EMAIL PROTECTED]


Re: IP Banning (Using IPFW)

2006-02-05 Thread Daniel A.
I know for a fact, that if a hacker wants to root a box, the first and
least thing he does is to
nmap -p1-65535 -Avv host
And yeah, it does detect services on unusual ports. And regardless of
what you say, assigning nondefault ports is security through
obscurity.

On 2/5/06, fbsd_user [EMAIL PROTECTED] wrote:
 You missed to whole meaning.
 Attackers only scan for the published service port numbers,
 that is what is meant by portscan the box.
 Those high order port numbers are dynamically
 used during normal session conversation.
 So any response from those port numbers if an
 attacker scanned that high would be meaningless.
 Please check your facts before commenting.

 -Original Message-
 From: [EMAIL PROTECTED]
 [mailto:[EMAIL PROTECTED] Behalf Of Daniel A.
 Sent: Sunday, February 05, 2006 4:58 PM
 To: [EMAIL PROTECTED]
 Cc: [EMAIL PROTECTED]; Michael A. Alestock
 Subject: Re: IP Banning (Using IPFW)


 On 2/5/06, fbsd_user [EMAIL PROTECTED] wrote:
  I find this kind of approach is treating the symptom and not the
  cause.
  The basic problem is the services have well published port numbers
  and attackers beat on those known port numbers. A much simpler
  approach is to change the standard port numbers to some high order
  port number. See /etc/services  SSH logon command allows for a
 port
  number and the same for telnet. Your remote users will be the only
  people knowing your selected port numbers for those services. This
  way a attackers port scan will show the well published port
 numbers
  as not open so they will pass on attacking those ports on your ip
  address. This way your bandwidth usage will be reduced as
 attackers
  find your ip address as having nothing of interest.
 
  This same kind of thing can also be done for port 80 by using the
  web forwarding function of Zoneedit pointing to different port for
  your web server. Only people coming to your site through dns will
 be
  forwarded to the correct port.
 
  The clear key here is attackers roll through a large range of ip
  address port scanning for open ports. By using nonstandard port
  numbers for your services you stop the attacker even finding you
 in
  the first place.
 
  good luck what ever you choose to do.
 You just argued against yourself. If an attacker is genuinely
 interested in rooting someones box, that attacker will most likely
 portscan the box - And thereby discovering that you have assigned
 alternative port numbers to your services.
 Security through obscurity is a bad place to start.
 
  -Original Message-
  From: [EMAIL PROTECTED]
  [mailto:[EMAIL PROTECTED] Behalf Of Michael
 A.
  Alestock
  Sent: Sunday, February 05, 2006 10:42 AM
  To: [EMAIL PROTECTED]
  Subject: IP Banning (Using IPFW)
  Importance: High
 
 
  Hello,
 
  I was wondering if there's some sort of port available that can
  actively
  ban IPs that try and bruteforce a service such as SSH or Telnet,
 by
  scanning the /var/log/auth.log log for Regex such as Illegal
 User
  or
  LOGIN FAILURES, and then using IPFW to essentially deny (ban)
 that
  IP
  for a certain period of time or possibly forever.
 
  I've seen a very useful one that works for linux (fail2ban), and
 was
  wondering if one exists for FreeBSD's IPFW?
 
  I've looked around in /usr/ports/security and /usr/ports/net but
  can't
  seem to find anything that closely resembles that.
 
  Your help would be greatly appreciated Thanks in advance!
 
   Michael A., USA... Loyal FreeBSD user since 2000.
  ___
  freebsd-questions@freebsd.org mailing list
  http://lists.freebsd.org/mailman/listinfo/freebsd-questions
  To unsubscribe, send any mail to
  [EMAIL PROTECTED]
 
  ___
  freebsd-questions@freebsd.org mailing list
  http://lists.freebsd.org/mailman/listinfo/freebsd-questions
  To unsubscribe, send any mail to
 [EMAIL PROTECTED]
 
 ___
 freebsd-questions@freebsd.org mailing list
 http://lists.freebsd.org/mailman/listinfo/freebsd-questions
 To unsubscribe, send any mail to
 [EMAIL PROTECTED]


___
freebsd-questions@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-questions
To unsubscribe, send any mail to [EMAIL PROTECTED]


Re: IP Banning (Using IPFW)

2006-02-05 Thread David Scheidt
On Sun, Feb 05, 2006 at 05:38:11PM -0500, fbsd_user wrote:
 
 You missed to whole meaning.
 Attackers only scan for the published service port numbers,
 that is what is meant by portscan the box.
 Those high order port numbers are dynamically
 used during normal session conversation.
 So any response from those port numbers if an
 attacker scanned that high would be meaningless.
 Please check your facts before commenting.

Nonsense.  There may be some people that only scan well-known ports,
but it's much more common to scan every port on a machine.  If you're
running a server on a non-standard port, an attacker will find it.

___
freebsd-questions@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-questions
To unsubscribe, send any mail to [EMAIL PROTECTED]