that the caller can establish his virtual identity on the phys node.
Eventually need per-vnodeid keys, but not necessary now since only
trusted people get to use vrons.

since we firewall that off, but it clutters the hell out of the
exports file and it was bugging me.

by me, and and earlier one from Kristin (td.tex).

Also included here is my defs.

Add idle expts view to experiment list page. Filters based on days idle (default is 2), and adds a column with a button that takes you directly to the confirmation page for sending a swap/terminate request. Very cool, and handy dandy for the Idle Daemon to use. Still doesn't do sorting by last login, because that would require either a monster 15-line SQL query or saving everything into an array, sorting, and then printing, instead of our current 'print in the order you get it from the db' method. Maybe later. Changed TBExpUidLastLogins to return a daysidle value as well.

interface should be configured to. Another possibility (and one we may
switch to) is to add the IP to the interfaces table so it gets
returned in the normal ifconfig, but thats going to require a number
of cascading changes that I do not want to deal with yet.

tunnels need to be created on each node; will be returned as a new
tmcd command.

Add nodeipportnum table, which is a hacky way to allocate non
overlapping port numbers on physical nodes (when multiple virtual
nodes are mapped to them). Used in assign_wrapper as a post pass over
the trafgens table and the tunnels table.

Add target_vname to virt_trafgens table as an aid in post passing.

reported.

the vlans table to determine how much of the trunk bandwidth is
currently 'reserved', and subtracts that from the trunk bandwidth
reported in the output.

This feature is disabled by default. However, you can enable it by
putting the line
TRACK_INTERSWITCH_BANDWIDTH=1
in your defs file.

the hostnames. Eventually purge the backwards compat code when the
virt_trafgens tables has been cleared of older entries (no ip addrs).
The perl code on the client side should have no problem with getting
IP addrs instead of hostnames (so he says).

to let users set the link, but for now this will cause the table to
be populated correctly for tmcd.

allow us to (eventually) use trafgens on nodes with more than one
link, but more importantly, solves an immediate problem for ron nodes;
a lack of hostname translation.

on fatal errors, and error codes >= 1 for assign errors. The batch
system depends on these error codes to determine if a batch needs
to be punted from the system.

active at a time. This avoids the situation where a person enters many
batches, and they get scheduled all at once, sucking up every node in
the testbed.

trying to start a new one (as in replay).

MODIFY command.

this was the only script still using the old log location.

removed mention of them from the boss setup documentation.

Note that this means we no longer tell people to /var/log/testbed to
/usr/testbed/log, since the former is deprecated.