isn't really a good reason for keeping them seperate. All we really
care about is putting the unroutable addresses in one file and the
routable ones in another.

So, instead of generating three forward zone files, we only have to
generate two. They are not as pretty as the ones named_setup used to
generate, but the code to make them is much, much cleaner and more
comprehensible.

Also makes named_setup about twice as fast - 2s vs 4s. Still too
slow, but better!

This is typically irritating to figure out in the parser!

Previously we used a hack that just assumed the first component of the path
was the mount point, now we actually gather and parse mount info from the
file server.

files. Then use it in the newosid page to allow setting of the
nextosid field (by admin people only). This will make it a tiny tiny bit
easier to set up new testbeds.

module-specific data.  Also, fixed numbering skew that started at v 1.223
(someone decremented the version by 100).

I'll come along for a closer cut in the future.

* Modularized the plab communications 'adaptor' interface and moved the
  dslice- and PLC-specific code into their own modules.

* Wrote an API definition README

* Separated out generic routines from libplab into their own library modules
  (libtestbed.py and libdb.py)

Functionally, not much has changed - this was just a massive re-org with some
other cleanup.  Should be much easier to code up new PLAB interfaces as the
plab folks flail around in their attempt to standardize on something.

XXX: may want to re-think where the generic library modules should go.  If
more python code enters Elab, we'll probably want to move 'em to more standard
locations.

This isn't the end of the cleanup - I would eventually like to go back and
rethink the class structures, beef up the comments, and extend the API.

are not "active" (perhaps frozen).

Also, as a sanity check and to prevent mountd from barfing, make sure
that any directory we try to export, actually exists. This is done via
the local mount points on boss, and is perhaps not such a great idea
cause it is via NFS. Keep an eye out for problems.

except for a few that we consider to be 'safe'.

Also, fixed a message to use @USERNODE@ instead of hardcoding to
ops.emulab.net .

Don't expect there to be any routes in ipassign's output!

of the interfaces table from bps (which they are in the
interface_types table) into Mbps.

Units for bandwidths in assign_wrapper are a total mess!

The intent is to avoid lots of silly stated warnings about virtual
nodes that do not exist outside the life of an experiment, and for a
number of reasons, can end up in odd event states while.

Add better slot data checking by calling TBcheck_dbslot() for each of
the values that will be inserted into the DB.

options which are no longer references anyplace.

batchexp, and forgot to finish the changes! The result was a fairly
broken batch system, which is not hopefully fixed!

Took the opportunity to remove the -x (expires) and -l (priority)
options which are no longer references anyplace.

Fix up email message so that idle/auto swap times are in hours not
minutes.

Provide a proper usage() function that describes the morass of
options (for interactive use from ops).

"realmode" stuff that Mac is doing. Change to -c instead for reconfig.

passable manner. Mostly this change overwrites Mike's changes for
dealing with the pump race. Rather than setting ONBOOT="yes" for all
interfaces (which is what triggers the pump race), only set the control
interface to ONBOOT="yes", all the others get ONBOOT="no" so that
ifup will not be called on them. Also remove the hostname set from the
ifcfg scripts since that totally messes up dhclient-script, which is what
is used in Redhat 9.0 (instead of pump).

The problem with ONBOOT="no" is that the interfaces will not get
probed if ifup is not called. This messes up testbed ifconfig later.
So, added an rc.linux script that does just the probe for all interfaces.
This script is called from rc.bootsetup, if rc."osname" exists.

(we want to record "myip" so we can use it in the frisbee command line)

was getting added twice with my previous commit.

Previously this happened only if jail virt nodes were requested.

one unfilled message.  No semantic changes in the body.
Sending to globalinfotek.

controllers.  Need to merge in Leigh's tipserver setup text.

Long name; sorry.  If you can think of a sufficiently shorter but still
descriptive one, feel free to re-add as that and delete old.

totally untested! Will have to get someone to test out an IXP.

It now properly sets ipassign's arguments following Jon's latest update.