GitLab

can tell which interface (APT or CloudLab) was used to create an instance.
Don't want to have to go back and parse email.

* Add a uuid to the profile itself. So now we have a uuid for each specific
  version, and a uuid that points to the profile as a whole.

* On the manage page, move the URL to the left hand panel, and add a second
  URL. One is the URL of the specific version, and the other is a URL to
  the entire profile. Add popovers to explain the difference between.

* On the instantiate page, when you use this URL, we instantiate the most
  recently published version of the profile.

* Add a bit of verbiage to the publish modal.

support code.

get the URL stuff figured out.

allows the user to create a new disk image in place, instead of having
to create a new profile (Clone).

which starts an image capture in the background on behalf of a user
using the APT web interface. The webtask holds the state of the image
capture as to told to us by the aggregate manager, and conferred to
the user via ajax calls which read the state of the webtask.

only visible to admin users. The control flow is not how we
want it, and I don't want to spend too much time on it until
Jon gets his stuff committed, which I think will include some
better ajax code.

a routeable IP, then use "basic" instead of closed so that ssh is
allowed in (on the alternate port of course).

of the rspec.

* Add a .htaccess file that does the rewrites, instead of in the httpd
  confile file. Added Rob's stuff for rewriting urls to hide the .php
  although not sure this is working correctly yet.

* Add simple MyExperiments page so that logged in users can find their
  way back to running profiles.

* Move the DB table holding the running experiment records from the
  geni-sa DB into the main Emulab DB. Lots of little changes for that.

* Change logout to plain link instead of ajax call. That was a silly
  thing to do.

* Bug fixes to ssh keys and shell login from the status page.

and libraries. Rough, needs plenty more work.