effect a single node, but are different from 'class' fds in that they
are stateful - ie. the socre you get depends on all other vnodes
currently mapped to the pnodes. These are identified by prefixing
them with a '?'. ('*' for global, mnemoic 'matches all'; '?' for
local, mnemonic 'matches one')

The currently implemented local fd is the additive one, or '?+'. We
add the desire weight for all vnodes on the pnode together, and if
this number is greater than the pnode's weight for the feature, a
violation is flagged.

The idea here is that we can give a pnode a feature like: ?+memory:256
and vnodes desires like ?+memory:16 or ?+memory:64 to pack vnodes
onto pnodes in a manner that appropriately reflects their resource
usage.

terminating experiments that are not already swapped out.

tbswap, since there are parts of the state machine that are outside
these scripts.

with. Currently 'emulab' or 'plab', defaults to emulab. This will
replace the current plab_user bit that Rob added, but I will remove
that later.

on boss, this cuts the size of the testsuite's 'clean' db from > 100MB
to a paltry 12MB.

should start with. Currently 'emulab' or 'plab', defaults to emulab.

to the node_types_auxtypes and node_auxtypes tables. The goal is to
not have a gaggle of fake types in the node_types table, but rather to
store the virtual types that a node (type) can take on, in the two aux
tables (node_auxtypes holds the count). This removes part of the
special handling of virtual types from ptopgen, and cuts down the
number of bogus entries in the node_types table.

I also did some cleanup. Got rid of a bunch of separate hash tables
and combined into a single hash of "records" indexed by type. This
script could use quite a bit more such cleanup.

node's primary virtual type.

node's primary virtual type.

what it means a little more.

the message instead of tbops.

in the dev tree. I'm tired of problems on the mini that wastes
my time.

Changes include 2 new tmcd commands: tmcc routelist returns
the routes for all the vnodes hosted on a pnode. tmcc role
returns the role of a reserved node, like 'virthost' or
'simhost.  tmcc ifconfig now reports an RTABID field which
is calculated in assign wrapper. All the new changes
in assign wrapper will be checked in after I finish testing.

All the DB changes are in: simnode_capacity in node_types, rtabid in
interfaces and veth_interfaces. New NSE event_objtype and NSEEVENT
event_eventtype. Changed the erole field in the reserved table
to have 'simhost' instead of 'simnode'. Changed the correspoding
libdb subroutines.

use pcplab nodes.

containing the set of node types a user is allowed to use. Add
a NODETYPE_ALLOWED() function to check against this hash.

to get a list of what types a project is allowed to use.

interface to the backend. mkproj now looks at the pcremote_ok set
and makes the proper calls to grantnodetype. This reduces the amount
of hardwired goo in the web interface.

Still, there is a bit of hardwired stuff in mkproj. At present we do
not form a relationship between a phys node type and the types we
assign to the virtual nodes. Thats is, nothing says that a pcplabphys
implies the right to use pcplabinet, etc. With only 3 remote phys
types, I just hardwired it into mkproj calling grantnodetype with type
pcplab (the class for the virtnodes) for pcplabphys. Same for pcron
and pcwa, (both get pcvwa). Ultimately we need a better type system.
In general the type system is pretty screwy.

physical plab nodes.

Grant permission to use types/class of nodes. If given a specific
"type", then grant permission to use just that type. If its a class, then
permission to use all of the types in that class, the class itself, and any
aux nodetypes for the type/class (node_types_auxtypes table).

For example:

	wap grantnodetype -p testbed pc2000
	wap grantnodetype -p testbed pcvm
	wap grantnodetype -p testbed pcplab

features to be lumped togther in one pclass. We were checking to make
sure that the second node passed to pclass_equiv() had all the
features of the first, but not that the first has all the features of
the second.

of bad passwords and freeze the user after 15 attempts, or the IP
after 25 attempts.

someone with an actual quota.

the user may have visited the last page with http. If they did, send them
back through https

more liberal about which links don't need delays. Before, we were
putting delays on anything that wasn't 100Mbps.

The other change is that for delay nodes, we set virtual link
bandwidths by rounding up to the next fastest speed of interface,
instead of hardwiring 100Mbps. Similarly, we set the speed of
physical delay interfaces to the card's 'native' speed instead of
hardwiring 100Mbps.

These changes work for our nodes and links, but are still not totally
general. Really, what needs to happen is that assign needs to be the
one to insert delay nodes, because it's the only place where we really
know which interfaces of what speeds are available on which nodes.
But, that is a very big project, since both assign wrapper and assign
need to be changed.

of the same speed. This is obviously not right, but I think it's
better than the old hack, which was that we assume switch ports are
100Mbps.

Some stuff Chris is working on should remove the need for this hack
because we will start keeping track of the speeds of switch ports.

did this with a simple macro called OUTPUT() that does the check and
bails from the current function. This was as boring and tedious as
life gets, but we should be protected from stack smashing now.

Also fixed up a couple cases where we take data from the client;
make sure that the sscanf calls are bounded.

interactive shell is allowed or not - paperpag will just exit
silently if interactive shells are not allowed, but no command was
given on the command line.

This variable is now off to prevent interactive shells, but all the
interactive code is still in place, in case we ever have a reason to
want it.

into $CHECKLOGIN_STATUS