> Benjamin Scott wrote:
> > Ummm, I believe conventional wisdom says that with modern algorithms, session
> > encryption keys longer then 100 bits or so is just a waste of resources. In
> > fact, I just checked, and the FreeS/WAN website makes reference to this.
I don't think that statement is true... GPG recommends a key length
of 1024 bits, and suggests that keys larger than that probably don't
really enhance security at all, but slow down decryption substantially
(and use more CPU).
Of course, the encryption mechanism itself may make the key size
irrelevant. For example, both versions of Netscape's SSL encryption
(meaning 40-bit and 128-bit) were cracked, due to a weakness in the
algorithm that generates the keys. Details of this crack can be found
fairly easily by doing a google search; I wasn't able to discover
whether or not the latest versions of SSL were vulnerable to the same
sort of attacks. These reports were fairly old (1995 and 1996) so
this may be limited to older versions of Netscape.
Note that details of the 40-bit hack were fairly widespread; you'll
likely have to hunt harder to find stories about the 128-bit hack.
I've also read that both were compromised by brute force with the
equivalent of a beowulf cluster, and someone told me that the 128-bit
encryption was compromised by an MIT grad student in about four hours
using an array of DSPs. I haven't been able to verify that one
though, so take it with a grain of salt (or do the research yourself)!
If you do find such a report, please send it to me... :)
That frood Kenneth E. Lussier sassed:
> The box: a PIII 600MHz system, 256MB RAM, yadda yadda yadda..... A good
> system. Using a moderate IPChains script, port-forwarding, NAT, etc. And
> let's say that there are about 1000 hits to the firewall per hour (not
> counting port-scans and other enomolies). This is a fairly stable
> system. Traffic flows well. No real bottlenecks. Now, on top of that we
> add FreeS/WAN. Since we want FS to be secure, we use RSA (`CUZ WE
> CAN!!!!!) 4096-bit authentication keys, rekey the session keys every 15
> minutes, and let's say that you have 5 simultanious tunnels per
> connection (this is actually a low number of tunnels per connection).
> You have just multiplied your CPU usage 12x.
>
> Now, this doesn't make the box insecure, but under heavier traffic, it
> could become unstable. It introduces a bottleneck on the firewall, where
> you probably don't it. Also, if the system goes down because of the
> load, then the one route out of the network is down.
But when setting up such a system, you also need to consider what
level of risk is acceptable, and how much you are willing to pay to
keep your data secure. For most circumstances, a key length of 4096
bits is just silly, and you should know that before you go ahead and
select it. If you really NEED that level of protection, chances are
you can afford the hardware that you'll need to pull it off.
Also, in the context of the original poster's message, does this make
any sense? I'm still not really sure what Ferenc wanted to do either,
but from the sounds of it he just wanted to connect one network to the
outside world (either to the internet via NAT firewall, or perhaps to
a corporate VPN). We're probably not talking about anything even
close to the level of activity you're describing, and you yourself
told me that a single P75 would perform acceptably both as a firewall
and as the endpoint of a single VPN tunnel.
> > There is a far too common attitude that simply by placing things behind a
> > firewall, you are secure. That is bogus. Most exploits are made possible by
> > bugs or mis-configurations in network software.
[SNIP]
> > wouldn't connect to the Internet in the first place). Such bugs can be
> > exploited just as well through a firewall.
>
> This is 100% true. If the service is exploitable, it's exploitable no
> amtter where is is, as long as the service is available.
This is an important point, which one must keep in mind when
protecting one's network. I can't count the number of times a manager
has said to me "but we're behind a firewall, so we're ok, right?"
That's a huge misconception. But one should also not take this as
proof positive that firewalls are a waste of time. Here's a real life
example:
As many of you may remember, my own firewall was rooted some months
ago. The attacker got in through a hole in the name server (since
plugged!), broke the login program so that root didn't require a
password, and enabled telnetd from inetd. Fortunately for me, the
attacker wasn't smart enough to realize that I had ipchains blocking
the telnet port, so they were still not able to get in to my system.
I was also fortunate in that the attack took place at precisely the
time I wake up in the morning, and I happened to check my e-mail,
which I was unable to do because of the attack. The machine had just
been rebooted, and that's how I discovered it. I logged in as root,
which required no password, and immediately I knew something was amiss.
Point being, IPCHAINS (and my timely awakening) saved me from further
penetration, because the attacker was not able to exploit the hole
they left for themselves. Granted, with a little more time, and a
different choice of WHEN the attack happened, things would probably
been different, but having those rules in place gave me enough time to
discover the break-in without any further damage occuring.
Now, as any security expert will tell you, you shouldn't run any
unnecessary services on your firewall, and this incident is definitely
proof of that. Had I not gotten up when I did, this attacker could
have gained access to the rest of the machine, and the rest of my
network, and made my day rather a lot more unpleasant than it already
was. Needless to say, a lot of things have changed on my network
since then...
=8^)
P.S. While I'm on the topic, for those of you out there who are
running a redhat 6.x install (and probably most other distributions of
Linux as well) on your Internet-connected firewall or other Linux box,
and also have other Internet services running on that machine, YOU ARE
AT RISK. If all you have done is install the default OS, and have not
made any effort to secure your system (i.e. by getting the latest
security updates, and/or setting up named to run as a non-root user,
to name just a couple of things), then you are definitely in danger of
having your system compromised. Well-known exploits exist for a
variety of software that ships with the OS. And, especially if you
have a permanent or semi-permanent IP address (like DSL or Cable), you
are a target, because of the power available to an attacker on your
typical Linux machine.
At the very least, make sure you keep up with security updates for
your OS, and I'd also suggest you watch security focus or the bugtraq
mailing list for security alerts that might affect you.
As I often point out, a great source of info on how to secure your
machine is "Maximum Linux Security" by Anonymous. It goes into great
detail about how to make your machine more secure, and exactly what
kinds of things people can do to your machine. If you have no
exposure to Internet security, it's a real eye-opener.
--
We sometimes catch a window, a glimpse of what's beyond
Was it just imagination stringing us along?
---------------------------------------------------
Derek Martin | Unix/Linux geek
[EMAIL PROTECTED] | GnuPG Key ID: 0x81CFE75D
Retrieve my public key at http://pgp.mit.edu
**********************************************************
To unsubscribe from this list, send mail to
[EMAIL PROTECTED] with the following text in the
*body* (*not* the subject line) of the letter:
unsubscribe gnhlug
**********************************************************
