[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
Re: How to block telnet access.
On Thu, 30 Sep 1999, guy keren wrote:
>
> On Wed, 29 Sep 1999, Nir Soffer wrote:
>
> > Break into my home machine. All I have now running is sshd (Which is
> > vulnerable, I know, but at least enough people use it to clear it out of
> > the bugs, and I can keep track.) , and I'm about to suspend it too.
>
> are you really only running sshd ? are you not running the linux kernel,
> which listens for IP connections? and X window? are you aware of the fact
IP connections to what? What does it do with them? It only replies to
ICMPs, and IGMPs. I am not running any routing daemon, so I can't poison
my routes. (All I have is a static route). I am not running any DNS
server, so sending bogus messages is also out.
DOS attacks are another matter, ofcourse. And those are issues of system
security as well.
No. I'm not running X. Because of the fact it listens to connections on
port 6000.
> that by default, the X server also accepts connections via TCP on port
> 6000, and any bug in itshost acces code would render your machine
> vulnerable? you also verified that indeed you're not having redhat's linux
> conf server running, for exmaple? what about portmap, without which you
> will not be able to run any rpc-based program on your PC? and lpd is not
> also active? does it not accept connections via TCP as well?
Disabled lpd. I'm not using RedHat for exactly these reasons.
psychodad:~> rpcinfo -p localhost
rpcinfo: can't contact portmapper: RPC: Remote system error - Connection
refusedpsychodad:~>
I am also not running the portmapper.
> > I
> > don't need stuff, I disable it. When I need it, I make sure it does what I
> > it do (Again, ssh is a bad example :) ).
>
> and apprently proves my point - even your home PCthat gives services to
> no one, is not completely safe.
But at least I can keep track of vulnerabilities in that ONE application.
> again, don't get me wrong - i think it's not very likely that someone will
> actually succeed with breaking into your PC - i'm just saying that you
> cannot make it 100% secure...
Like Adam pointed out. I can to a certain threat model. I can assume that
if someone has a shell on my machine, he has root. I can also assume
that's it not very damn likely for anyone to obtain a shell on my machine
using ICMP packets or something.
I agree, however that 'not very damn likely' is not 100% secure. But it's
close enough for me. If anyone breaks into a machine using ICMP only - not
only does Alan Cox (or wheover developed the networking code in the 'Foo
OS') deserve a flogging, but I'll buy whoever did it a beer.
Regards,
Nir.
--
Nir Soffer AKA ScorpioS. scorpios @ cs.huji.ac.il
"Cry havoc, and let slip the dogs of war!"
- William Shakespeare, "Julius Caesar", Act III, Scene 1
Mail me with the subject 'get pgp key' for my PGP Public key.
=================================================================
To unsubscribe, send mail to linux-il-request@linux.org.il with
the word "unsubscribe" in the message body, e.g., run the command
echo unsubscribe | mail linux-il-request@linux.org.il