with disk images being corrupted.

correct (non-xen) kernel.

so they get propagated when clonging an image.

booting XEN VM, let the library function deal with that, since
there is now code in there to watch a dead control network and
retry.

Currently it is configured (hardwired) to run every 15 minutes, even that
may be too frequent as things don't happen too fast in lease-world.

Before it defaulted to no auto-destroy and you specified -A to turn it on.
Now we default to auto-destroy and you specify -N to turn it off.

I was being overly paranoid and you can accomplish the same thing by setting
the autodestroy sitevar for each lease type to 0.

Shorter default client_ids will reduce the odds of us running into hostname limitations. Green should now not show until things are actually done according to sliver status.

We want the lease to appear unused up until it is first mapped.
Also, fixed some typos in comments.

For persistent blockstores, is based on the value of the "readonly"
virt_blockstore_attributes attribute if it exists. The RO attribute
is set by libvtop when an attempt is made to use a lease that is in
the 'grace' state.

network to see if we can catch when a node is never going to boot
cause XEN created a broken control vif; tear down the VM and try
again, which seems to solve the problem most of the time.

for elabinelab (multicast, apod).

say to do that instead of bridge mode. Sadly, the xen script that does
this is broken and so here is our version.

the resource mapper. To add a node restriction:

boss> node_exclude pcXXX

To remove the restriction:

boss> node_exclude -r pcXXX

start of a page to create new profiles, lots of other changes
and additions.