starting any one of our scripts can take a second or two. That time is
spent including and compiling 10000s of thousands of lines of perl
code, both from our libraries and from the perl libraries.

Mostly this is just a maintenance thing; we just never thought about
it much and we have a lot more code these days.

So I have done two things.

1) I have used SelfLoader() on some of our biggest perl modules.
   SelfLoader delays compilation until code is used. This is not as
   good as AutoLoader() though, and so I did it with just a few 
   modules (the biggest ones).

2) Mostly I reorganized things:

  a) Split libdb into an EmulabConstants module and all the rest of
     the code, which is slowly getting phased out.

  b) Move little things around to avoid including libdb or Experiment
     (the biggest files).

  c) Change "use foo" in many places to a "require foo" in the
     function that actually uses that module. This was really a big
     win cause we have dozens of cases where we would include a
     module, but use it in only one place and typically not all.

Most things are now starting up in 1/3 the time. I am hoping this will
help to reduce the load spiking we see on boss, and also help with the
upcoming Geni tutorial (which kill boss last time).

takes to include each module. Run like:

boss> perl -I /usr/testbed/devel/stoller/lib -MTraceUse foo.pl

(I like static typing.)

We never get live data from the SA about when slices expire, so simply
assume that the slice expires at the maximum of all credential expiry
times we've ever seen from that slice.

DeleteSliver had been calling StopSliver then DeleteSlice,  but the
recommentation from Utah is to skip that and just call
CMV2::DeleteSlice.

This should help to explain exactly what has happened when the slice
was renewed but for some reason the sliver was not.

Not much more to say. Shutdown was stubbed as unsupported and now it
has been implemented as a pass through to CMV2::Shutdown.

Compression is now done in one place and in one way rather than two
different places and two different ways. Added handling of an
additional error case in the second call to resolve if a slice_urn is
present. Improved comments as well.

The AM ListResources implementation had punted on the path that
included a slice_urn in the options. This is now implemented per the
GENI AM API spec. It's still a bit brittle, but working.

Need to set the magic debug sysctl on geom to allow changing active disk.

(this time for the urn). Also change the protoype to allow an error to
be returned.

(this time for the urn). Also change the protoype to allow an error to
be returned.

vlans.

internally, as the primary key in the tables, but the CM/SA APIs no
longer use them. The CH still accepts them for now. We can probably
stop putting them into manifests and advertisements at this point as
well. 

For slivers, stop using the uuid of the node as the uuid of the sliver
itself; generate a new one. As above, this is cause the uuid is the
primary key in the table, but the URN is what we use for lookups,
etc.