Ericson578 wrote:
Does anyone else watch those like a hawk looking for spikes?
Not really. I measure a lot of things, but the only thing I watch really closely is actual performance of the service being provided. If that starts tanking, then I look at all the graphs.
(That's not to say I avoid looking at the graphs unless there's a problem; I've got a browser tab open to munin and a tail -F running on the log server most of the time, but between Nagios, Pingdom, and the best users in the industry, it's not the first line of defense.)
Quote:
I looked into firewall rules to deny all from certain countries (especially china), but I hear that's only effective against those not smart enough to run through a proxy (which so far is most of the snoopers in my logs).
There are no humans behind the ssh connections, nor are they isolated to particular geographic regions. It's a worm (or, more accurately, a number of different worms). It spreads on its own, and while it might report its findings somewhere, it's autonomous. Do not attempt to anthropomorphize it, nor should you attempt to out-clever it. Only way you can "win" is by making it impossibly difficult.
Also remember that "once in a million years" happens annually per each million computers.
_________________
Code:
/* TODO: need to add signature to posts */