Solaris telnetd failure with Heimdal client

classic Classic list List threaded Threaded
5 messages Options
Reply | Threaded
Open this post in threaded view
|

Solaris telnetd failure with Heimdal client

Dave Love
The Heimdal telnet client with -x causes the SEAM telnetd on Solaris 8
and 9 to dump core.  With Solaris 10's telnetd it fails with `Bad
encryption type'.  There's clearly a SEAM bug if it crashes, but does
anyone know if the Heimdal client is at fault somehow, as others must
have tried this combination?  (It works with the MIT daemon, giving
similar authdebug output, and the Solaris client works with the same
tickets/keytab/krb5.conf.)

For what it's worth, this is with Heimdal 0.6 and 0.7, the latest (but
old) SEAM servers and all Solaris 10 updates from a month or two ago
when the patch server last let me update.  [I made a support call, of
course, but our contract appears useless when a non-Sun client crashes
a network process.]

________________________________________________
Kerberos mailing list           [hidden email]
https://mailman.mit.edu/mailman/listinfo/kerberos
Reply | Threaded
Open this post in threaded view
|

OT: Re: Solaris telnetd failure with Heimdal client

Fredrik Tolf
On Thu, 2005-11-03 at 16:57 +0000, Dave Love wrote:
> [I made a support call, of
> course, but our contract appears useless when a non-Sun client crashes
> a network process.]

Off-topic reply: How extraordinarily weird. Does that mean that you
won't get support if someone attacks a DoS flaw?

Fredrik Tolf


________________________________________________
Kerberos mailing list           [hidden email]
https://mailman.mit.edu/mailman/listinfo/kerberos
Reply | Threaded
Open this post in threaded view
|

Re: OT: Re: Solaris telnetd failure with Heimdal client

Will Fiveash
On Thu, Nov 03, 2005 at 07:03:53PM +0100, Fredrik Tolf wrote:
> On Thu, 2005-11-03 at 16:57 +0000, Dave Love wrote:
> > [I made a support call, of
> > course, but our contract appears useless when a non-Sun client crashes
> > a network process.]
>
> Off-topic reply: How extraordinarily weird. Does that mean that you
> won't get support if someone attacks a DoS flaw?

The folks are Sun are aware of this and looking into it.  I do not know
the details of the interaction between Sun support and Dave so I can not
comment on that.  In regards to a DoS, I thought telnetd was spawned by
inetd so I don't see the DoS.

--
Will Fiveash
Sun Microsystems Inc.
Austin, TX, USA (TZ=CST6CDT)
________________________________________________
Kerberos mailing list           [hidden email]
https://mailman.mit.edu/mailman/listinfo/kerberos
Reply | Threaded
Open this post in threaded view
|

Re: Solaris telnetd failure with Heimdal client

Will Fiveash
In reply to this post by Dave Love
On Thu, Nov 03, 2005 at 04:57:27PM +0000, Dave Love wrote:
> The Heimdal telnet client with -x causes the SEAM telnetd on Solaris 8
> and 9 to dump core.  With Solaris 10's telnetd it fails with `Bad
> encryption type'.  There's clearly a SEAM bug if it crashes, but does
> anyone know if the Heimdal client is at fault somehow, as others must
> have tried this combination?  (It works with the MIT daemon, giving
> similar authdebug output, and the Solaris client works with the same
> tickets/keytab/krb5.conf.)

Can you use rlogin -x instead of telnet?

--
Will Fiveash
Sun Microsystems Inc.
Austin, TX, USA (TZ=CST6CDT)
________________________________________________
Kerberos mailing list           [hidden email]
https://mailman.mit.edu/mailman/listinfo/kerberos
Reply | Threaded
Open this post in threaded view
|

Re: OT: Re: Solaris telnetd failure with Heimdal client

Dave Love-3
In reply to this post by Will Fiveash
Will Fiveash <[hidden email]> writes:

> The folks are Sun are aware of this and looking into it.

[Not according to the support person.]

> I do not know the details of the interaction between Sun support and
> Dave so I can not comment on that.

For what it's worth, it subsequently appears to have been a
`misunderstanding' of a quite explicit report, but it was consistent
with previous experience, so I came to engineer land immediately.

> In regards to a DoS, I thought telnetd was spawned by
> inetd so I don't see the DoS.

Yes, but it is remotely crashing a process running as root, and might
be exploitable for all I know.  I don't doubt the engineers care
anyway, thanks.

________________________________________________
Kerberos mailing list           [hidden email]
https://mailman.mit.edu/mailman/listinfo/kerberos