On Friday, 02/10/2006 at 06:55 CST, Marcy Cortes
<[EMAIL PROTECTED]> wrote:
> I guess I don't see anyone breaking into the class G guest if the class
> G guest doesn't a password and is logon by only by the systems
> programmers with her password.  I guess in theory they could somehow get
> to cp by linux, but then he'd already have the NIC anyway.

If you have only one VSWITCH defined, then the risk is obviously minor.
But what if you had two VSWITCHes, one that faced the Internet and another
that faced your intranet?  The risk of breakin may be low, but the
consequences of the Linux guest connecting to both VSWITCHes without your
explicit permission would be significant.  It would be like giving some
distributed server access to a trunk port on a switch with authorization
to all VLANs.  <shudder>

But I appreciate that in non-security-conscious environments the
RESTRICTED nature of a VSWITCH can be annoying.  But maybe it's only
annoying because it is easy to forget to authorize the access?

A final thought...  The best security controls are in an ESM.  You can lay
your system bare if that's what you want to do.  That is to say, an ESM
provides not only extra security, but extra non-security if so configured.
 With RACF, a single generic profile could be defined with UACC(UPDATE),
allowing every user on the system to connect to any Guest LAN or VSWITCH.

Philosophy #37: It should be HARD to get your system into a wide-open
state.  No accidental tourists.  In fact, security standards are moving in
the direction of requiring *two* privileged users to deactivate the
security controls.  (You know, both have to insert and turn their keys at
the same time.)

Alan Altmark
z/VM Development
IBM Endicott

----------------------------------------------------------------------
For LINUX-390 subscribe / signoff / archive access instructions,
send email to [EMAIL PROTECTED] with the message: INFO LINUX-390 or visit
http://www.marist.edu/htbin/wlvindex?LINUX-390

Reply via email to