Server Fault is a question and answer site for system and network administrators. Join them; it only takes a minute:

Sign up
Here's how it works:
  1. Anybody can ask a question
  2. Anybody can answer
  3. The best answers are voted up and rise to the top

I've issues with a continuous, concerted cracking attempt on a website (coded in php). The main problem is sql-injection attempts, running on a Debian server.

A secondary effect of the problem is being spidered or repeatedly spammed with urls that, though a security hole has been closed, are still obviously related attempts to crack the site, and continue to add load to the site, and thus should be blocked.

So what measures can I take to:

A: Block known intruders/known attack machines (notably making themselves anonymous via botnet or relaying servers) to prevent their repeated, continuous, timed access from affecting the load of the site,

and B: report & respond to the attack (I'm aware that the reporting to law enforcement is almost certainly futile, as may be reporting to the ip/machine where the attacks are originating, but other responses to take would be welcome).

share|improve this question

For option A you can investigate mod_security, however this is something you should thoroughly test, initially in a reporting mode.

I presume your application is reasonably hardened:

  • the database user php connects as only has enough permissions to do what they require
  • your php config is hardened

And that the box it is running on is fully patched and has all inessential services disabled.

Furthermore you could considering running the database on a different host.

share|improve this answer
Uh, why would a different host be desirable? – Kzqai Mar 15 '11 at 1:05
You can put the DB box in a different firewall segment – jamespo Mar 15 '11 at 11:15
I'd assume the DB isn't directly accessible from the net (if it is, he's got major problem). Separating it to a different host isn't going to do much if the hacker uses sql-injection, or roots the server. – Chris S Mar 15 '11 at 12:20

In my experience there isn't much you can do but block the offending IPs. It probably wouldn't hurt to get an external company to do a security audit if you've become a regular target and weaknesses have been found in the past. We've used a company called 7safe to do this.

If the IP originates from your own country, and the hack attempt has been successful, involving the police isn't a bad idea. We've had a bit of luck with that here in the UK.

I was watching some hack attempts unfold against my own site earlier today. Needless to say they didn't get anywhere! But hacking is a fact of life on the internet.

share|improve this answer

My standard response to this remains the same: You can't stop people from trying to do bad things on the internet. The attacks you're seeing are undoubtedly automated, and wide-band: they aren't after your site particularly, they're after any vulnerable site anywhere.

You can spend a huge amount of effort chasing down 'known bad' hosts, following up leads, doing the detective bit; but the only ultimately useful thing you can do is secure your system against its application (app runs as non-privileged user, possibly chrooted, reasonable ulimits, up-to-date kernel, etc), secure the app against its users (reliable authentication and authorization, placeholder SQL query building, pen-testing with web security tools), and monitor for successful attempts at unauthorized access.

share|improve this answer

In my experience reporting the behaviour to the netblock owner in a polite email, while not very often effective, is still likely to have a more satisfactory outcome than relying on either your local or the attacker's local law enforcement. Particularly if they provide an 'abuse' address.

Just provide them with log entries showing the problem (remember to provide details of any time zone conversions for dates/times).

While you can start blocking access from those addresses, you might want to consider implementing something like 'fail2ban' via a PHP autoprepend - so if you detect someone being naughty you can automatically block their access using the OS firewall.

share|improve this answer
+1 Report to netblock owner. Be sure to include who you are, the source IP, example(s) of what you are reporting, and keep it as brief as possible. Do not: get emotional, assign intent, 'interpret' the evidence. Also Fail2Ban is my personal favorite, mod_security is another good idea. – Chris S Mar 15 '11 at 12:23

Your Answer


By posting your answer, you agree to the privacy policy and terms of service.

Not the answer you're looking for? Browse other questions tagged or ask your own question.