Re: PermitRootLogin=yes versus su
From: Simon Tatham (anakin@pobox.com)
Date: 01/27/03
- Next message: Richard Caley: "Re: PermitRootLogin=yes versus su"
- Previous message: Sebastian Hans: "Re: PermitRootLogin=yes versus su"
- In reply to: Sebastian Hans: "Re: PermitRootLogin=yes versus su"
- Next in thread: Sebastian Hans: "Re: PermitRootLogin=yes versus su"
- Reply: Sebastian Hans: "Re: PermitRootLogin=yes versus su"
- Messages sorted by: [ date ] [ thread ] [ subject ] [ author ] [ attachment ]
From: Simon Tatham <anakin@pobox.com> Date: 27 Jan 2003 11:38:01 +0000 (GMT)
Sebastian Hans <hanss@in.tum.de> wrote:
> The su binary isn't /just/ an additional point of vulnerability in any
> case. To exploit su I would have to gain access as the regular user in
> the first place. If you protect the login of the regular user the same
> way as you would protect root's, su adds another layer of security - it
> doesn't take one away.
But if su has a vulnerability, then the attacker now need only break
an ordinary user's account - and that's likely to be much easier,
since on any machine with a large number of users there's bound to
be _one_ who's done something stupid. The root account is likely to
be guarded by someone who actually knows what they're doing.
In addition, if su has a local-root vulnerability, any of your local
users who you don't completely trust becomes a direct threat, so
external attackers are suddenly the least of your worries!
-- Simon Tatham "The distinction between the enlightened and the <anakin@pobox.com> terminally confused is only apparent to the latter."
- Next message: Richard Caley: "Re: PermitRootLogin=yes versus su"
- Previous message: Sebastian Hans: "Re: PermitRootLogin=yes versus su"
- In reply to: Sebastian Hans: "Re: PermitRootLogin=yes versus su"
- Next in thread: Sebastian Hans: "Re: PermitRootLogin=yes versus su"
- Reply: Sebastian Hans: "Re: PermitRootLogin=yes versus su"
- Messages sorted by: [ date ] [ thread ] [ subject ] [ author ] [ attachment ]
Relevant Pages
|