the release number on them so the repo metadata will result in the most
recent version getting installed.

for fedora >= 11

Added a work-around for distros that have a /etc/dhcp configuration directory. Creates a symlink in these directories pointing to /etc/dhcpd.conf.

linux.

Since we don't use these, they are now commented out of our default config.

rsync, and rebuild the repodata somewhere, and synch that up too.

look in /usr/lib* to catch lib64 stuff.

ended up being a hack, since gnu make wouldn't recurse over implicit
rules... so we just do our own recursion by starting another make :-).

Other cleanups of the main Makefile to get rid of some unnecessary vars
for our packages.

general, this should be a collection of useful make stuff for any rpms
that must be built from a combination of sources in cvs and online.
pkg-build is a driver program that parallelizes builds, commands, source
synch'ing, to build servers.  The makefiles handle almost everything
necessary to build, except for some targets that are best triggered
manually, since they are one-shot requirements and take time to verify.
The only thing currently unhandled by the makefiles is installing packages
in the build that are required by packges yet to be built -- this is hard
because we don't necessarily know the names of the rpms that will be
built, so we kind of have to construct them by mining the spec file.
That'll come later...

diffs for snmpit_{hp,nortel,foundry}.pm

(the non-binary ones, that is) in here -- everything else gets stuck on a
webserver.

:-)).

otcl is still at 8.4 (and otcl is going to take some hacking to move
to 8.5).

  * switched over to the current perl DBI stuff in most places; no more
    Mysql DBD direct stuff.
  * platform objects (linux and bsd currently) to handle certain platform
    differences (i.e., unix security stuff like users/groups)
  * addition of new wrappers to set UID and EUID et al, since the behavior
    is *different* on BSD and Linux -- this was very nasty.
  * autoconf now finds most of the previous command binaries that were
    hardcoded
  * lots of updates to patches, or to get things to compile, on newer
    versions

Can't remember anything else that was really big.  Tons of little stuff,
of course.

platform stuff.

will get worked out better after the latest checkpoint :-).

the tb-set-node-os command with a second optional argument; if that is
present, the first arg is the child OS and the second is the parent OS.
We add some new features in ptopgen (OS-parentOSname-childOSname) based
off a new table that maps which child OSes can run on which parents, and
the right desires get added to match.  We setup the reloads in os_setup
along with the parents.  Also needed a new opmode, RELOAD-PCVM, to handle
all this.

For now, users only have to specify that their images can run on pcvms, a
special hack for which type the images can run on.  This makes sense in
general since there is no point conditionalizing childOS loading on
hardware type at the moment, but rather on parentOS.  Hopefully this stuff
wiill mostly work on shared nodes too, although we'll have to be more
aggressive on the client side garbage collecting old frisbee'd images for
long-lived shared hosts.

I only made these changes in libvtop, so assign_wrapper folks are left in
the dark.

Currently, the client side supports frisbee.  Only in openvz for now, and
this probably breaks libvnode_xen.pm.  Also in here are some openvz
improvements, like ability to sniff out which network is the public
control net, and which is the fake virtual control net.

done with a fake lan node in the rspec, but this is clearly (clear to
me at least) the wrong way to do this. But nececcary for the moment.