This forces imagehash to hash pieces on aligned sector boundaries rather
than just doing it from the start of each chunk. Result now is that there
may be a partial hash piece at the beginning of each range and another
partial piece at the end. Ideally, we would avoid those partial pieces,
but we are only using this as a point of comparison right now.

Scroll doesn't zoom map when mouse is over an info window. Slices in the map legend update with status and when selected. Added buttons to refresh slice, slices or everything. Fixed offline support. Added ability to use cache when logged in to use cached managers but live data for everything else.

Randomized exponential backoff on busy. Shows spinner until Slice has been created and status has been determined.  Sends full RSPEC to all managers. If a CreateSliver? fails, the others will not be cancelled. Give user more feedback on errors during slice creation. RSPEC now included along with the XML-RPC call in log messages.

Redeem fails. This was causing ReleaseTicket() failures later.

first arg, so all other args off by one. This stuff would never have
worked right.

Namely, attempting to upload a project-wide shared image from a node in
an experiment in a subgroup of the project.

Includes support for adding new nodes/interfaces/interface-types/wires.
Also includes sample xml files.

Conflicts:
	protogeni/flack/src/protogeni/communication/RequestTicketRedeem.as

Disabled link buttons on the slice canvas for managers without links. Switched delay node to be at the end of the draggable nodes on the slice window. Dragging a manager onto the canvas now adds a node or prompts the user to add one. Bolded the map markers.  General bug fixes various places.

Conflicts:
	protogeni/flack/src/protogeni/communication/RequestTicketRedeem.as
	protogeni/flack/src/protogeni/display/SliceWindow.mxml
	protogeni/flack/src/protogeni/resources/VirtualNode.as

frisuploadd will move <foo>.tmp to <foo> when it completes. Since create_image
monitors the size of the former, it might get an undefined value if it checks
the size right after it have been renamed. Compensate.

Attempts to load an image from a subgroup of a project didn't work because
I was only setting the primary group ID based on the node's experiment's
project. Now we use setgroups() to establish all groups that the swapper
of the experiment belongs to.

It goes in $TB/etc/extracerts.bundle.  Right now, there's no way to
similarly maintain local CRLs, CA blacklists (certificates _not_ to allow
from the CH), or to refuse certificates from the CH by default, but we
can find a way to do those things if there's any need.

If nodes.pxe_boot_path is set to '/tftpboot/pxelinux/<something>', then
dhcpd_makeconf will set the (pxeboot) filename to /tftpboot/pxelinux.0
and symlink the node's config file (/tftpboot/pxelinux.cfg/<mac>) to
/tftpboot/pxelinux.cfg/<something>.

In other words, we can customize pxelinux to some small degree, using one
of some small number of pre-existing configurations. We were using pxelinux
before for plab-in-elab and we will also need it for loading WinPE for
configuring Windows7 images. For the latter we will set the pxe_boot_path
to /tftpboot/pxelinux/winpe.

Anyway, ideally we would allow the user to specify a pxelinux config file
through the NS file, but need to think about the implications of that some
more. Small steps...

Also make sure that datetime in XML is rfc 3339 compliant (tack on Z
to indicate GMT).

indicate GMT).

Right now the servers that are spawned have just the primary gid of the user.
Determining the right set of gids requires rummaging around in the DB for the
Emulab config.

The mfs directory has been moved to its own public repository:

git://git-public.flux.utah.edu/linux-mfs.git

For ease of debugging.