It was possible for an experiment to crash stated simply by registering
the wrong node as an emulab_syncd.

requires a fully bound rspec; the other side can be left unbound, as
long as you say what CM. Each side will contact the other side to find
out the bound details.

"testing" variable, which of course, makes things break in totally
obscure ways.

No need to suck in an extra library.

It should not be necessary to do anything to you inner elab infrastructure
to make it work in the New World.  Eventually, those with elabinelabs should
update/rebuild/install their Emulab software on boss/ops and update their
MFSes so they too will use a master server.

reason I totally forgot to make it conditional. Sheesh.

One of ucast, mcast, or bcast.  This only applies when using the master
server (-F).

This way I don't have to bother with backward compat for frisbee distribution
of blobs.

vlans will later use when instantiated. Needed for protogeni.

In the short run, we leave this at pre-master-server levels
for compatibility (we still support advance startup of servers
via the localhost proxy).

We also need this at Utah while we work out some MC problems on
our control net; sometimes nodes can take minutes before they
actually hook up with the server, even with the mserver.

for running from update-testbed.

1. Add -i option to tie mfrisbeed to a particular interface and use that
   option for a multi-homed boss node (only listen on inside interface).

2. Make that option play nice with the localhost proxying features of a
   previous commit: we also listen on localhost in this case.  The localhost
   proxying feature now has the compile time option USE_LOCALHOST_PROXY
   rather than mixing it up with the USE_EMULAB_CONFIG option (even though
   it probably will only ever be used in the Emulab config).

3. Both of those were really to fix one bug: if we got a request from
   localhost to start up a frisbeed daemon, we were starting it up bound
   to the localhost interface.  Now we will make sure the server gets
   started up using the "other" interface, i.e., the one specified with -i.

Add script to start it up with the right options and tweak the rc.mksubboss
script for setting up a subboss.

Properly reset some variables prior to parsing loadinfo info for an image.
This could cause failure if loading multiple images.

Neither of which as causing any problems.

The big backward compatibility issue is that we no longer store running
frisbeed info in the DB.  This means that loadinfo could not return
address:port info to clients and thus old frisbee MFSes could no longer
work.  While not a show stopper to require people to update their MFS first,
I made a token effort to implement backward compat as follows.

When an old frisbee MFS does "tmcc loadinfo" (as identified by a tmcd
version < 33), tmcd will invoke "frisbeehelper" to startup a daemon.
Sound like frisbeelauncher?  Well sorta, but vastly simplified and I only
want this to be temporary.  The helper just uses the frisbee client to make
a "proxy" request to the localhost master server.  The Emulab configuration
of the master server now allows requests from localhost to proxy for another
node.

frisbeehelper is also used by webfrisbeekiller to kill a running daemon
(yes, just like frisbeelauncher).  It makes a proxy status request on
localhost and uses the returned info to identify the particular instance
and kill it.