[298] in Zephyr Mailing List

home help back first fref pref prev next nref lref last post

Re: undead zwgcs

daemon@ATHENA.MIT.EDU (Steven Freed)
Tue Sep 15 13:21:44 1998

Date: Tue, 15 Sep 1998 11:11:22 -0600 (MDT)
From: Steven Freed <sfreed@gilasoft.com>
Reply-To: sfreed@gilasoft.com
To: Greg Hudson <ghudson@MIT.EDU>
Cc: zephyr@MIT.EDU
In-Reply-To: <199809151458.KAA00323@nephthys.grey17.org>

On Tue, 15 Sep 1998, Greg Hudson wrote:

> > Are there any plans to release a more current version of zephyr?
> 
> Unfortunately, there's a bit of a packaging problem on our end.
> Zephyr development here has moved back into the regular Athena tree,
> and the in-tree versions of com_err and ss went away in the process.
> So if we shipped what we have right now, it would be somewhat less
> convenient to build than 2.0.4 is.  It's not like it's impossible to
> work around this, but it's no longer a matter of simply wrapping up
> what we have and sending it out.

I understand.

> I'm actually not sure what the failure mode is for running the zephyr
> server on a multi-homed host, and I doubt the current sources do
> anything to help.

Our mail server has 3 network interfaces. Only one out of 3 messages gets
delivered. The zephyr server gets the following message:
 
zephyrd[16486]: sendit unauthentic fake packet: claimed 204.134.12.100,\
                                                   real 192.160.121.100
zephyrd[16486]: sendit unauthentic fake packet: claimed 192.160.121.161,\
                                                   real 192.160.121.100

What's happening is each time it it querying namservice and getting a
different answer back in a round-robbin. So it works every 3rd time.

> [There are also, unfortunately, security problems in 2.0.4, and I feel
> pretty lame for not addressing them in the external release somehow.]

Why don't we try to put a release together? ...something that will do for
another threee years or so :-). Let's try to fix the known issues and make
a release. I'd be happy to help. I've got the ss and et libs here, but I
don't know how this has all changed in the athena tree. The multi-homed
problem shouldn't be too difficult to fix. 

If someone at MIT is willing to work with me and make available to me the
files I need to build zephyr outside the Athena tree, I can make the build
package to put on the athena-dist ftp site.


-- 

Steven.



home help back first fref pref prev next nref lref last post