The revived Plab interface is here!

Lots of updates to the plab backend, including improved plab <-> elab node
id translation and update handling.  Includes support for the current PLC
API, and the new pl_conf node manager interface API.  Several more db library
routines were ported from the perl library to the python one to support the
new code (mostly the node_id tracking stuff).  Fixes to the client side and
also a rootball creation cleanup (binaries removed from the CVS repo).

There are also enhancements to the experiment view page for experiments
including plab nodes: site and widearea hostname are now displayed along
with the other node information.

Note that the way setup timeout for vnodes is calculated has been changed a
bit.  Instead of using a hardwired base timeout, the base timeout is now
based on the reload_waittime database field, which comes from the 'OS'
(e.g., FBSD-JAIL, RHL-PLAB) the vnode runs.

The default max duration for a plab slice created through the plab_ez interface
is set to 1 year, and linktest is currently disabled and hidden through
the ez interface.

There is still work to do, but this checkin brings with it a functional
plab portal!

archive!

studly users in the testbed project on the mainsite.

experiment, still likely to change.

Doh!  sshtb needs to exit with the status of the child process when
trying sshv2 and able to connect.

be muncged to work on older perl we use on boss.

(this was done quite a while back in our running MFS, just commiting
the change now so it doesn't get lost)

Remove the hardwired "-1" from an invocation of sshtb

change, and for now is only active at Utah.

the emulab-ops project only, so that we can more easily take nodes out
of the general pool, but still be able to use them.  This slot
actually takes a pid, so we could potentially use it more generally
for reserving nodes to particular projects, but I am not sure that is
the best approach yet.

 * remove db/entropy files to avoid sharing entropy
 * move update of locate DB to the end
slicefix:
 * update /entropy file where it exists in attempt to give every
   node a fresh burst of entropy

Add a note about /root/.ssh/config on boss and ops.

Added instructions for creating a passphraseless v2 RSA keypair for root
on boss, and explained where it needed to be added (on ops, boss and images).
Also described how to set Procotol 2 as the default everywhere possible.
THere is also a note about old user images and compatibility.

Update SSH_ARGS to remove "Protocol 1" specification.  Everyone needs to
reconfig their object trees!

significance (an audit file under Linux that was several MB)

can run.

users.

Be default sshtb now tries to ssh to the specified node via protocol version 2.
It will try protocol version 1 if a v2 connection/auth fails.  This was a
little messy to implement since sshtb now has to fork the first ssh attempt and
wait to see if it succeeds (all the fun of dealing with signals and killing
off the child ssh process and whatnot).

* Add creation of no-passphrase Protocol 2 RSA key in addition to
  Protocol 1 key. Currently Protocol 1 will continue to be generated,
  until we figure out an acceptable way to conditionalize this for old
  and new sites.

* No longer generate authorized_keys2 file. All keys go in the main
  file, and the authorized_keys2 file is deleted if it exists, after
  successful creation of the main file.

* When regenerating the Emulab keys, read the current .pub file in and
  delete the existing keys from the DB.

causing the program agent to run as last swapper.

This was a little tricky cause of a poor decision to share the
usage of the last_swap_uid in the stats gathering code, which wants
the to set the last swapper late so that the previous swapper gets
charged appropriately. Rather then mess with the stats code too much,
I moved things around a bit, setting the swapper earlier and adding
code in libdb to capture the original swapper at the begining of a
swapmod for accounting, and then adding code in swapexp to reset the
swapper if a swapmod fails.

Should fix the stats code at some point to have its own idea of
swapper.

recorded for every preloaded experiment. I'll have to clean up the
stats records, but its an easy fix (will need a small script to do it
though).

 * allow requesting status by naming a node on a controller rather than
   needing to know the controller name
 * make output of status command more grep-able

to use a delay value that is too low.