Needs to be a distinct type in the client scripts since its control
interface is not the same as a true pc850.

interfaces table to determine which interfaces to clear.

When inserting new interface entries, set the role appropriately,
based on control_iface setting in the node_types table.

are couple left in the newnodes code that Rob will need to help me
with.

is legal (and common!) in node vnames.

and updating the status of existing ones now actually works.  It
performs a named update after new nodes have been added.  Node adding
has been reordered to avoid race conditions with people requesting
Plab nodes while they are being added to the DB.  Vnode status updates
are now done in one SQL swoop, instead of one at a time.  Lots of
little tweaks.

so the freeing first waits for the exec to finish in the other prong
of the fork.

appropriate.  The node status information now reports SSHD port
separately; doesn't report IP port ranges for Plab nodes; does report
RPM's for virtual nodes; only reports BIOS version if it's specified
in the DB; and, if a control net IP isn't found for a node (the case
with virtnodes), it reports the control net IP of the physical node.

modify newwanode to set that for the new wa nodes (control interface).

       control_iface join with node_types, since new nodes have more
       than just expt and ctrl interfaces (ixps). The control_net slot
       is bogus anyway. Note that 'jail' is a placeholder for later.
       A 'fake' interface is not to be confused with veth interfaces; it
       is a psuedo interface like that found between an ixp and the host.

       alter table interfaces add \
		role enum('ctrl', 'expt', 'jail', 'fake', 'other') \
		NOT NULL default 'other' after iface;

install no longer depend on them.

main zone file (which were also being inserted with bad syntax).

(necessary in order to play nicely with vnode_setup).  rc.vinit now
figures out he nodeid based on the file in which libsetup stores it.

* Various improvements to new node stuff, including reworking node
  status updates so that they use the right table, and don't update
  vnodes that are alive (since their watchdog will do this).

* Added renewal code to automatically renew all leases that are doing
  to expire within two days.

* Moved Emulabification directly into the node abstraction.  Now the
  libplab wrapper scripts are all just plain wrapper scripts, instead
  of having the knowledge spread out

* Switched from using a Plab-specific keypair to using the normal
  Emulab one, which makes it possible to use sshtb to Plab nodes.

* Removed node booting code, since vnode_setup takes care of this.

destroyed if necessary, and nodes are allocated into the slice when
needed.  This should be optimized a bit better so that the Plab
manager can obtain as many tickets as possible in one operation,
instead of per node.

that jails do.  sshtb can now handle vnodeid's, and must be given a
vnodeid if it's a Plab node, because the username will differ
depending on this.

really working this time.

I will not check things in without running the testsuite
I will not check things in without running the testsuite
I will not check things in without running the testsuite
I will not check things in without running the testsuite
I will not check things in without running the testsuite
I will not check things in without running the testsuite
I will not check things in without running the testsuite
I will not check things in without running the testsuite
I will not check things in without running the testsuite
I will not check things in without running the testsuite

I was getting the IP from the hostname, so I switched to the trick
we use on the testbed nodes; use /etc/dhclient-exit-hooks to write
the data into /var/run, and use that in register.sh when the IP=DHCP.