GitLab

creates a new Geni style event, which are re-injected back into the local
(boss only) event stream. The new events have a SITE setting (which we have
never used before) to the URN of the aggregate, and other non-emulab
things, such as a json encoded representation of the sliver status whenever
a node changes state.

CLUSTER_PUBSUBD_PORT is the SSL port that pubsubd listens on at the Portal,
and CLUSTER_PORTAL is the hostname of the portal (boss.emulab.net) that
tells clusters to forward geni events.

Geni style events to the portal pubsubd (on the SSL port).

receiving Geni style events from event enabled clusters.

On clusters where CLUSTER_PORTAL is defined, start up an SSL enabled pubsub
notification forwarder, to send geni style events to the portal pubsubd.

for the benefit of the Geni watcher that is sending out events.

an ajax failure handler.

mothership branch. Yes, this is going to annoy someone.

of the Utah clusters.

for emulab-ops and testbed for some testing.

option to the mapper, which passes along a -G option to ptopgen when
it should remove all 10G links.

are updated (this is a temporary hack until I change how accounts are
updated on the geni path).

1. Do not allow guest users to use anything but the APT cluster. We had
   talked about this a while back, and today it caused a problem:

2. Because a guest tried to use the Mothership (cause of a URN in the
   profile), we had GeniUser lookup confusion. We store guest users in the
   geni-sa geni_users table, but because PROTOGENI_LOCALUSER=1, we end up
   creating a nonlocal account on the Geni path, and that conflicts.
   Changed how we do lookups.

Status 204 (RC_NO_CONTENT) is actually the correct one, since we are not
retuning any content, but apparently gitlab is (no longer?) smart enough
to recoginze all 2xx codes as being okay, and seens to want to see 200

via the Geni API, which have hard to guess project names).

	boss> wap perl prereserve emulab.net:testbed 1

At the remote cluster, you prepend the domain of the SA (for the portal,
that is emulab.net) to the project name at the portal.  prereserve will
find the correct project, assuming at least one experiment has ever been
created on the remote cluster. If not, then this trivial little hack will
not work.

Useful when executing a bunch of these in a row.

This is useful for the new r430s where the control net location varies
depending on how many 1Gb experimental interfaces the node has.

shellinaboxd daemon and/or the wrapper is doing something strange that
makes apache unhappy.