GitLab

added to the testbed. Useful when we know that all of the old images
will work on the new types.

(If the reservation being approved isn't in the list returned by LookupAll(),
presumably because it was previously pending, then append it during
feasibility checking.)

We have had issues with uploading images to boss where they are then written
across NFS to ops. That seems to be a network hop too far on CloudLab Utah
where we have a 10Gb control network. We get occasional transcient timeouts
from somewhere in the TCP code. With the convoluted path through real and
virtual NICs, some with offloading, some without, packets wind up getting
out of order and someone gets far enough behind to cause problems.

So we work around it.

If IMAGEUPLOADTOFS is defined in the defs-* file, we will run a frisbee
master server on the fs (ops) node and the image creation path directs the
nodes to use that server. There is a new hack configuration for the master
server "upload-only" which is extremely specific to ops: it validates the
upload with the boss master server and, if allowed, fires up an upload
server for the client to talk to. The image will thus be directly uploaded
to the local (ZFS) /proj or /groups filesystems on ops. This seems to be
enough to get around the problem.

Note that we could allow this master server to serve downloads as well to
avoid the analogous problem in that direction, but this to date has not
been a problem.

NOTE: the ops node must be in the nodes table in the DB or else boss will
not validate proxied requests from it. The standard install procedure is
supposed to add ops, but we have a couple of clusters where it is not in
the table!

infrastructure switch

to clear a prereserve without revoking it since that invariably means it
never gets revoked cause there is no record of it.

Also need to specify the slice (partition) number using the old
argument rather than the "PART=" keyvalue. Otherwise you wind up
with a full disk image masquerading as a partition image.

munge it, and made it too long.

(Along the way, adding an option to "predict" to compute the minimum free
node count over a type period.)

Closes #373.

If we support provenance but not deltas, then we do not use the
newer create-versioned-image when creating images from Xen vnodes.
However, we had a bug in that path where we would then not pass the
imageid argument to the old script, resulting in us spewing the image
out to stdout which got put in the logfile.

the way it was; note that an approved reservation cannot be shrunk,
but an unapproved reservation can shrink or grow, and we want to
keep it in the same state. This overrides the -a and -p flags.

directories from the proj and groups tree.

of never seeing it.

1. Add WebTask support for communicating back to the web interface and
   the Geni interface. Especially helpful for the case that a dataset
   needs admin approval.

2. Some changes related to above to make it easier to pass back the
   reason for nor approving.

3. Small semantic change to -C (always create) which I added for the
   Geni path a while back. It now means always create but respect the
   approval decision. In other words, if a dataset would need approval
   cause it is too big, create the dataset but mark it as needing
   approval and send mail. -C used to mean always create and approve.
   The new approach is a better fit for the Portal/Geni path.

remove them. Send email too.

1. -E schedules cancellation of a reservation.

2. -O aborts scheduled cancellation of a reservation.

3. -y runs Tidy() to clear out expired reservations.

careful so lets make it harder.

poke the portal so it knows that happened.

a reservation, so we can easily restore its announcement if the cancel
is later canceled.