On Tuesday 19 September 2006 07:30, Michael Christie wrote:
> I'm in the throes of applying for level 2 Broadband for Health payment.
> I'm supposed to check my firewall regularly.

The mere questions they ask reveal that the B4H guys have no clue.

A few things I do:

0.) "root" and other "well known" users are excluded from remote connections 
(e.g. via ssh) - attempts to connect under such user ID leads to IP number 
banned from retrying - see below

1.) "fail2ban" - a small script that checks logs in fixed intervals (2 minutes 
here) for failed login attempts - any IP address that tried to log into any 
of the services (open ports) more than 3 times unsuccesfully gets banned for 
10 minutes via automatic modification of the firewall rules.  Those who retry 
after ten minutes I notify their ISP, still have to automate that part. It is 
usually people who are "owned", that is their computers are used as slaves 
for such attacks, and they are usually grateful for the hint.

2.) "chkrootkit" - another script that checks the system in fixed intervals 
for known rootkits and the network interfaces for unexpected activity /  
promiscuity. It sends me a report by email every 6 hours

3.) "logcheck" - another script that examines the system log and all 
application logs for abnormalities, and emails me the summary (if any) every 
hour

4.) netstat -lt & netstat -t output redirected into a file periodically, and a 
script mails me any active ports (both connection established or just 
listening) that are not in an "allowed" list

5.) temp directory is mounted as "non-executable" now = if anybody manages to 
exploit a web server hole and gains access to /tmp, he can do bugger all 
(other than filling /tmp with trash, but he cannot upload any executables and 
execute them)

All these scripts and features are either part of Debian Linux or can be 
installed via simple "apt-get install fail2ban chkrootkit logcheck" if you 
are connected to the internet, and they get updated automatically via 
security,debian.org

So all I have to do is read my email if I want to know if anything got past 
the firewall. All such email goes to a dedicated email account which always 
gets attended either by myself or my son (we both get copies via 
forwarders) - so we would know within an hour or less if there was any 
intrusion or persistent intrusion attempt (I am only interested in persistent 
attempts, general attempts happen literally every second of the day)

Despite all this we had a recent intrusion via a security hole in TikiWiki on 
one server that was not quite up to scratch yet

Horst
_______________________________________________
Gpcg_talk mailing list
[email protected]
http://ozdocit.org/cgi-bin/mailman/listinfo/gpcg_talk

Reply via email to