<div dir="ltr">Actually, I don't know what other people on the defense side think of when someone says "Indicators of Compromise" but I don't think about hashes or file names or registry keys at all. <div>
I think about anomalous login times, unusual traffic destinations/sources/volumes, unusual file accesses (to file servers, not file access time on a potentially compromised client), patterns of exploration or spreading changes in behavior that might indicate a system is under control by some other source.</div>
<div><br></div><div style>I'm not looking for the indications that a system has been owned, I'm looking for indications that an attacker has compromised the environment and is now pursuing their goals. </div><div style>
And I'm doing it with data that isn't stored locally on the compromised system because as Dave noted, that can all be changed in real time by any sort of serious attacker.</div><div style><br></div><div style>The things Dave described aren't "indicators of compromise", they are forensically relevant fragments that might be left on some systems as a result of being compromised that might be used to help fill out the details of how a compromise occurred _after_ it has been detected through some other means.</div>
<div style><br></div><div style>Toby</div></div><div class="gmail_extra"><br><br><div class="gmail_quote">On Wed, Jun 12, 2013 at 7:10 AM, Dave Aitel <span dir="ltr"><<a href="mailto:dave@immunityinc.com" target="_blank">dave@immunityinc.com</a>></span> wrote:<br>
<blockquote class="gmail_quote" style="margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex">Hackers spend a lot of time looking at what's coming down the technology<br>
road at them. In a sense, this business is about learning how to stare<br>
down the barrel of a gun and not blinking for decades at a time. When<br>
you blink, you end up a CISSP. Richer financially, but poorer in 0days,<br>
the only currency that matters to someone with your particular addiction.<br>
<br>
Terminology can reveal a lot, as can business strategies. I spent some<br>
time on the phone yesterday with a high level executive in the incident<br>
response industry, and he poo-pooed Immunity's offensive skills, which<br>
made me focus on the industry for a while while watching Covert Affairs<br>
after the kids went to bed.<br>
<br>
First of all, here's what's next in the incident response world:<br>
"Indicators of Compromise". And when people say that, they right now<br>
mean MD5s, file names, registry addresses, dns addresses, what addresses<br>
a trojan hooks, and that sort of thing. All of these things can be<br>
changed AT RUN TIME, by your better trojans.<br>
<br>
In other words, we have an industry focused highly on "indicators of<br>
compromise", whereas modern high-level attackers have leapfrogged the<br>
entire concept. The only true indicator of compromise is "computer is<br>
doing something I probably didn't want it to do", and that's not<br>
something you can codify in XML.<br>
<br>
Something to think about. :><br>
<span class="HOEnZb"><font color="#888888"><br>
-dave<br>
<br>
<br>
</font></span><br>_______________________________________________<br>
Dailydave mailing list<br>
<a href="mailto:Dailydave@lists.immunityinc.com">Dailydave@lists.immunityinc.com</a><br>
<a href="https://lists.immunityinc.com/mailman/listinfo/dailydave" target="_blank">https://lists.immunityinc.com/mailman/listinfo/dailydave</a><br>
<br></blockquote></div><br></div>