On Tue, Dec 1, 2009 at 8:53 PM, Joseph Sinclair
<plug-discussion@stcaz.net> wrote:
It's not going to find everything, and it's definitely not a fully-automated tool, but I find the SQLInjectMe plugin for Firefox to be a very useful tool for SQL injection testing.
For more automated scanning, you might try Wikto (http://www.sensepost.com/research/wikto/), although I don't know much about it...
Joe wrote:
> Hey all,
>
> Can anyone (Lisa, I'm looking in your direction) recommend a decent SQL
> injection scanner? I don't really care if it's server-side or
> client-side since it's my server, and I don't need to *exploit* the
> injection points, I just need an easy way to find them. I'd like it to
> be easy to figure out, generate output or reports that are easy to
> follow and not require too much to be installed on the server.
I suggest that you test the way "they" will.
> The reason I'm looking for something is that the server on which my
> company hosts its websites has been compromised and I've been putting in
> some considerable hours trying to fix things. I've removed malicious
> scripts, fixed or removed the exploited code and changed all of our
> passwords (from ssh to mysql to user accounts).
Keyloggers, puppet or cfengine might assist to trap them in real time, or annoy them by restoring all the files changed on a server every few minutes?
> Today, I happened to catch a SQL injection scan and now I'm trying to
> look down that path some more. Basically, they used one of our (many)
> poorly escaped queries to poll password data for our site login (among
> other things). Luckily, I shut the scan down before they got the
> passwords so I didn't have to have users reset them *again*.
UG! Did you IPTABLE/ACL their source subnets?
Generally doing that you see the same traffic from another source IP, as they usually attack from many sites, but watching logs for a string that matches the original signature (like SNORT inline would) and automagically iptable denying them, might help for the immediate, while you get it together to run a full scan and get the developers and dba's to evaluate the results. That bash shell script is easy to build integrated with iptables.
> I've cleaned up a bunch of the sql code over the past could days, but
> I'm wondering if there's a way for me to scan for injections myself and
> attack code that is "more vulnerable" than others. I found sqlsus
> (
http://sqlsus.sourceforge.net/), which looked pretty impressive, but it
> didn't run properly and it wasn't really a scanning tool so much as it
> was an exploiting tool. I also found Pixy
> (
http://pixybox.seclab.tuwien.ac.at/pixy/), which looked pretty
> comprehensive, but the output looked a little intimidating. Plus, the
> little I read of the docs wasn't really clear about how to actually use it.
>
> Anything else anyone would recommend?
Go through the full list of exploits and check your installations against the known holes by version. Then start with the code. Many PCI compliant applications must purchase a layer 7 application switch because code rewrites are too invasive.
I would start with the comprehesive examples from OWASP.
---------------------------------------------------
PLUG-discuss mailing list - PLUG-discuss@lists.plug.phoenix.az.us
To subscribe, unsubscribe, or to change your mail settings:
http://lists.PLUG.phoenix.az.us/mailman/listinfo/plug-discuss