Skip site navigation (1)Skip section navigation (2)
Date:      Fri, 18 Jun 1999 09:54:39 -0500 (CDT)
From:      James Wyatt <jwyatt@RWSystems.net>
To:        Darren Reed <avalon@coombs.anu.edu.au>
Cc:        Brendan Conoboy <synk@swcp.com>, freebsd-security@FreeBSD.ORG
Subject:   Re: ipf howto, tada
Message-ID:  <Pine.BSF.4.05.9906180939200.6084-100000@kasie.rwsystems.net>
In-Reply-To: <199906181039.UAA22257@cheops.anu.edu.au>

next in thread | previous in thread | raw e-mail | index | archive | help
On Fri, 18 Jun 1999, Darren Reed wrote:
	[ ... ]
> > I prefer to run ipmon with as "ipmon -s" so it
> > syslogs logged packets instead of having them dump to stdout.
> 
> "ipmon /var/log/iplog" will save log entries direct to that file.  ipmon
> also handles SIGHUP as you would expect, closing and re-opening the log
> file to allow for rotation.  With newsyslog, this should be possible
> without too much hassle.

Who says you want to syslog to a plain ASCII file?

> FWIW, you might like to mention the "log-or-block" option where it will
> block a packet to be pass'd and logged if it cannot log it due to the
> log buffer being too full.
> 
> i.e.
> pass in log first or-block on vx0 proto tcp from any to any port = 80 flags S/SA keep state
> 
> Here we say only log the first packet for this connection as recorded by
> "keep state", but if it can't be logged, then block it.

Neat trick! Could this easily be used for DOS? I like, this idea, but want
to understand it. If you filled the syslogs with dummy attempts, would it
block access, preventing you from cycling syslog files?

	[ ... ]
> > pass out quick proto tcp from 200.200.200.1/32 to any keep state
> 
> This can be in or out...essentially when the packet first crosses your
> perimeter.  If it is a locally made connection going out then the above
> is correct.  If it is from another host on your LAN going through your
> IP Filter firewall, then it should be "pass in" on the LAN interface.

I've been considering the value of a 'firewall' interface that could track
how long a TCP session was open. If you get in on anything but ssh/rsh and
telnet for more than 5 min, it would trip... I've also wondered if it's
keeping state, if it could feed start/end time when the session closed?
Even if waiting for 5 min to tell you you're hacked wasn't a good idea, it
might be nice to have such stats. - Jy@



To Unsubscribe: send mail to majordomo@FreeBSD.org
with "unsubscribe freebsd-security" in the body of the message




Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?Pine.BSF.4.05.9906180939200.6084-100000>