Re: Security process

[Walter <walter@xxxxxxx>]

Pierre Abbat wrote:
> On Monday 08 March 2010 15:33:11 Walter wrote:
> > I got curious about BSD (DragonFly, specifically) security and
> > wondered why there wasn't a security process that processed all
> > security-relevant error messages which could then be used to
> > block IPs, disable user accounts, and kill processes.  At least
> > it'd be a step to automating *some* obvious security measures
> > rather than requiring root action.  Things like repeated login-
> > in failures from external (as in China) IPs.  Anyone?
>
> How would you write a program to process error messages and decide which user 
> accounts to disable?

I don't know what error messages are available for when
an account tries to do things for which it has no privilege,
causes a system/privileged program to crash... whatever.
I was hoping there was some security trail for things that
might open a door...  I don't know.. too many errors, a
temporary privilege being granted for too long might mean
someone hacked a system utility... I don't know.

> As to blocking repeated login failures, there are such things. I wrote one 
> myself and have it running on my Linux box (the DragonFly box is a laptop and 
> isn't publicly visible). It doesn't care whether the source of the logins is 
> in China or my net-door neighbor (or even the laptop, which looks like the 
> router's external IP the way it's forwarded).

China was just an example.

Can you direct me to where/how I can block IPs based
upon error messages (login failures, etc.)?