This call appears to always reselect the DB, which might be a problem
when the DB is busy.

the actual query.

since we are saving just a single slice, and 0 is actually the MBR (by
local convention).

reporting in isalive status, and since we are returning a "need to
update" flag in the message, and since we are now initiating widearea
account updates by setting the flag for each node, and since tmcd
decrements that flag when the node picks up accounts, it is a simple
matter to do update for remote nodes! Simply, bump the flag in this
script, and then wait a little while to see whose flag does not change
back! Report those in the email message. Works very slick.

This mechanism is going to be generalized some to handle other kinds
of updates. Either via the isalive return value, or via a specialized
TMCD call to return the list of things that need to be updated. The
next most needed update mechanism we need is for the testbed software,
followed closely by user software.

doesn't change anything, but could be a mild information leak.

when there is only a single control node, that acts as both boss and
ops.

time in between retries.  This should help out when we have dozens
of experiments swapping in and out at once.

nodes.
Add handler for remote nodes, which now report keepalive once a
minute. Simply look at the timestamp and change status from up to
down if it has not responded in more than 90 seconds. This is kinda
crude, and mostly for the benefit of the web page. The wan solver
will use the timestamps directly.

(and so, has been renamed setVlansOnTrunk). Part of the work necessary
to get the stack-level setVlanOnTrunks function to accept multiple
VLANs in one call.

performance analysis or validation done yet.

First, we'll no longer have seperate sup releases for each revision,
so the supfiles use the release 'current'

Second, since we no longer have seperate sup releases, the prepare
scripts no longer need to remove checkouts files.

* Add ability to control timeout when invoking tmcc, and whether to
  fatally die if tmcc fails. Related change was to allow more options
  to be passed into tmcc by way of an additional options argument.

* Change to pubkey handling. tmcd now returns a list of pubkeys for
  each login. The entire list is written to the authkeys files, and
  the old is backed up (with mod time preserved).

* tmcd now returns a serial number for each use to indicate when
  user info has changed (via the web pages). Only users with updated
  serial numbers get modified (and new authkeys file created). This
  should vastly reduce the passwd/authkeys churning on the ron nodes,
  which periodically check for account updates. The little db file now
  records the serial number too.

* Change to os_usermod; Change the actual password string! Was not
  doing that before.

* Conditionalize all the tmcc.foo stuff that Shashi added; don't want
  any of this on remote nodes yet.

* Remove a bunch of printfs to reduce output clutter.

actually doing that on local nodes before. I guess I'm very ssh
centric (I never type passwords) but that appears not to be the case
for most people!

keys file.

well. We use a UDP packet to keep it lightweight. If it does not get
through, thats okay, obviously. The return value is just a yes/no flag
that says an update needs to run. Right now, thats just accounts.
This allows us to churn a little less on accounts.
Other cleanups.

when remote is set timeouts when trying to update accounts, and fail
more gracefully.

caused by hupping mountd are not getting propogated back until later
writes. Must do an fsync to catch the error. Too many fsyncs slows
down imagezip by a factor of three though, so I reorged things so that
we create the 1MB chunks all in memory, and then write out each one
out as finished. Much better; only a 10% slowdown.

to lose them!

Return the update_accounts flag so that the client knows to
update accounts. This flag is set in node_update for remote nodes.
Once the client picks up accounts, decrement the update_accounts flag.
This functions a simple barrier (up/down counter so that clients
do not miss).

mode.

started.) Instead of avoiding sending an event when we think the
node will be sending it, go back to always sending it. Nodes
sending the events on shutdown just isn't reliable enough. Instead,
we'll just make two REBOOT events in a row legal.

indicating when the image was last updated. This gets updated by
create_image .

The hope is that it will give us some idea of which revision of
our standard images custom images are based off of.

fixe's the config.status bug.

Fix broken calculation of hours since last swap request was sent. (had problems when the time was from yesterday.)

Add idle_ignore column to db. Makes idlecheck completely ignore it when looking for idle expts. Unswappable expts still get checked and possibly harassed if they go idle.

Fix up idle view better for slothd now that we trust it more. Add info about # of swap requests and last time of a swap request too. Took out lastlogins column for idle view, but the sanity check still happens.

swapin.

defaults for other fields.