[MUD-Dev] strong encryption for authentication

J C Lawrence claw at 2wire.com
Fri Jul 13 15:27:07 New Zealand Standard Time 2001


On Fri, 13 Jul 2001 11:28:10 -0600 (MDT) 
Fred Clift <fclift at verio.net> wrote:

> On Thu, 12 Jul 2001, Fred Clift wrote:
>> Oh, and I'd love to get my hands on some crypto software -- It'd
>> definitely be a fun toy :).

> Uh - I mean hardware -- I have plenty of software :).

<phew>

> My main motivation here is to not allow someone to leverage mud
> access into local shell access (and thence into root access on the
> box).

Are you running in a chroot()ed environment?

> I have tools in my mud that (if I have bugs in my software) could
> be used to change files/state on the server, outside the mud (ie
> db acccess, load/save of mob scripts etc).  I dont think there are
> holes, but I've been wrong many many times before.

Presuming you are running Linux, I would be tempted to install one
of the (many) capabilities systems and then carefully restrict both
the capabilities of your game server, its children, the account it
runs under as well root itself (I normally like bolting down root to
the point that it can't actually do anything without a controlled
reboot with console access to bring the system back up in a mode
where root can actually do things).

> For now, I ssh to the box and run a client there connecting to
> localhost -- probably good enough for me, but I know some players
> who would like it.

What clients are they using on what platforms?  For *nix systems an
SSH port forward from localhost to your server, over SSH v2 (to
remove the requirement of a shell account on the server) would work
just fine as would stunnel (see previously noted caveats).  

I'd be more tempted to setup MindTerm's Java SSH client under an SSL
webpage such that it logs directly into your game server.

> I might try and convince the administrative people on the mud to
> always use encrypted connections.... 

An aspect which I did not address in my last message is marginal
gains in password security.  Its not an absolute gain -- the
passwords are no more secure -- but they are more resistant to
trivial interception and this compromise (eg a password sniffer
installed on a cracked system at a CyberCafe).

  ObNote: I've found password sniffers in 3 out of the 4 cybercafes
  I've visited (the exception was the now-defunct coffeenet here in
  SF)

Arguably its an appreciable gain for a system which is already
fragile.

> Or at least not use the same password for shell access that they
> use for their mud logins :)

Uhh, yeah, right.

--
J C Lawrence                               ("`-''-/").___..--''"`-._         
---------(*)                                 `6_ 6  )   `-.  (     ).`-.__.`)
claw at kanga.nu                               (_Y_.)'  ._   )  `._ `. ``-..-'  
http://www.kanga.nu/~claw/                _..`--'_..-_/  /--'_.' ,'         
I never claimed I was human             (il),-''  (li),'  ((!.-'           
_______________________________________________
MUD-Dev mailing list
MUD-Dev at kanga.nu
https://www.kanga.nu/lists/listinfo/mud-dev



More information about the MUD-Dev mailing list