ops will boot more quickly if boss is also booting or not booted
yet.

This turns out to be less obvious then first thought, since
with PAM built into sshd, also need to turn off the
ChallengeResponseAuthentication setting.

everything figured out for setting up the private vlan on both
switches. There is a wire from eth1 on the control node to a port on
the control switch. Another wire from eth2 to a port on the
experimental switch. The complication on the control switch is that we
have no serial console, and so we actually have another wire from eth3
to the control switch, so that we can dhcp and then log in and do a
switcherroo to get the private vlan configured.

little room.

Need to add a desire on the lannode that says it is okay to put
it on a shared node, otherwise it gets a desires violation even though
it maps okay.

Thanks to Rob for figuring this out!

Avoids a problem on machines with no serial port.

no longer included in their original ports.

Add xauth since it is needed for proper x11 agent forwarding, and
should always be installed.

them to be unique in the shared node case, but for good measure lets
just generate random/unique macs, and then also set the "locally
administered" bit (0x02 in the upper octet) cause linux requires it,
and we do not want the client side to change the mac if we can avoid
it.

them to be unique in the shared node case, but for good measure lets
just generate random/unique macs, and then also set the "locally
administered" bit (0x02 in the upper octet) cause linux requires it,
and we do not want the client side to change the mac if we can avoid
it.

updated.

This is necessary for a "make install" to succeed without a previous
"make [all]".

them to be unique in the shared node case, but for good measure lets
just generate random/unique macs, and then also set the "locally
administered" bit (0x02 in the upper octet) cause linux requires it,
and we do not want the client side to change the mac if we can avoid
it.

might otherwise get lost.

comments in the code.