this function is backwards; DBQueryFatal() should call DBQuery(), not
the other way around!

Remove some debugging stuff that broke the library. Kirk can add it
back later when he decides what to do.

before the /proj/pid/exp/eid dir is create.

Watch for transient NFS error that causes this directory to not be
removed, and send email.

Also, remove another ball of lint.

	- check intervals driven by sitevars delivered by TMCD command
	- handles rusage stats return on plab nodes
It is now a single process and executes any auxilliary scripts synchronously.
This may prove to be unwieldy in the face of long running scripts like
update.  If so, we'll have to add all that fork/exec/waitpid mucky-muck.

Random values were causing random failures on plab.
Also cleaned up some double colons in error messages.

* If -p option given, use that project to find the image. Fatal error
  if no such image in that project.

* If no -p option:

     * If in -e (pid,eid) mode, then first look in the same project as
       the experiment (pid), and then look in emulab-ops.

     * If not in -e mode, then look only in the emulab-ops project.

     * If not obvious, -p option always overrides the search.

updates to README.sa for last snapshot
add -I option to client to allow setting of redodelay (redo "I"nterval)

email is always sent of course.

the command line, and use that class. The default is "experiment" if no
arg given. Paperbag passes the argument, and it corresponds to the -m
argument in the client.

people with real shells on boss can use the RPC interface; Everyone
else goes through paperbag where its easy to arrange for the right
stuff to happen. The rest of us need some tricks to make it work.

* created "-w" vnode_setup option that specifies how long to wait (per-vnode)
  for setup to complete before giving up.
* added sitevars for plab batch parallelism size and vnode setup timeout
* modified os_setup to use above sitevars when invoking vnode_setup for an
  experiment containing plab vnodes.

* Added comments
* Added Emulab copyright
* made mod_PLC handle the "not assigned" error case in freeNode()
  - optimization and less log clutter.
* bug fix in plabmonitord (ISUP decection)

if the pubkey is too large to fit in the default buffer.

Also added a F_MAXLOG value that you can set to force maximum logging
for a particular command.

Fix an off-by-one in the OUTPUT macro

* Changed the way options are parsed in the python scripts so that modules
  can easily add and use their own options independent of top-level scripts.

* Added --noIS and --pollNodes module options.

* Added batch option to vnode_setup (degree of parallelization)
  - defaults to 10

* Major updates to plamonitord
  - batches testing, currently to 40

server, turning xmlrpc/module into "program module".

I view this as temporary; paperbag exists and allows me to quickly export
the server. Long term, we need a different solution.

Add a little bit more output to help in determining why keys get
rejected.

back to page that invoked it; user was not getting any feedback about
a key being in wrong format.

watchdog process based on sitevars.

Also added a command flags field to take care of a couple of misc. attributes
like whether a command can be allowed from a remote site via UDP or whether
we should not perform as much logging.

(Plab support is down right now, and people want to use it.)