I've decided to be a little in-your-face with the Net.gods.  I've got an HP
printer connected via Ethernet switch cable to ae0, and mc0 connected to
the JPL network.  Netatalk is installed and basically working.

/usr/local/atalkd.conf contains:
mc0 -phase 2 -net 13278 -addr 13278.147 -zone "JPL-301-2-Enet"
ae0 -seed -phase 2 -net 15000 -addr 15000.147 -zone "Hanks Printer"

I can see the printer inside Chooser from my powermac on the same zone as
the NetBSD box, but it takes 3-4 seconds to show up.  Other people on
different subnets/zones don't seem to be able to see it, even though they
can see the zone.

Any suggestions for how to debug this situation?

More info:

The NetBSD box registers itself as a printer and printing to it from mc0
results in printout on the ae0-connected HP (and an error message about no
.ppd file).  I have no config files for the printer daemons anywhere.

The machine generally hangs in _domaininit during boot, just after "fpu0 at
mainbus0" as documented in PR kern/5118.  A workaround is to disconnect mc0
during boot until after this point.  If it is left disconnected then the
atalkd.conf file gets trashed.  (Seems like -seed information should not be
changed by atalkd.)

---------------------------------------------------

Separate Issue

What was the fix for the boot hang while setting ttyflags?  This one only
bites me occasionally, but it's the number 2 reason my machine won't boot
itself.

My /etc/ttys file contains:

console "/usr/libexec/getty Pc"         vt220   off secure
ttye0   "/usr/libexec/getty Pc"         vt220   on secure

as the possibly relevant lines.  I have nothing currently connected to
either serial port.

Signature failed Preliminary Design Review.
Feasibility of a new signature is currently being evaluated.
h.b.hotz@jpl.nasa.gov, or hbhotz@oxy.edu