the top level.  This will build all the necessary binaries and then install
them.  This works on FBSD4 and RHL7.3.  It still doesn't work on FBSD5
(newer compiler that no longer supports a style of use of _FUNCTION_ in the
event lib) or RHL9 (event lib needs SSL lib which has a bad dependency
on Kerberos).  Notes:

- requires that elvin libraries be installed on nodes (they are) to build
  event agents, requires linuxthreads be installed on FBSD (it is now) to
  build imagezip (which is installed, but is not strictly necessary)

- installed event-agents and other binaries are stripped

- added a few missing files to the source tree for bsd (healthd.conf)
  and linux (healthd.conf, rc.local)

- the only thing that doesn't get rebuilt in /usr/local/etc/emulab is
  healthd, I couldn't quickly find how it gets built

- uses a scaled down version of libtb with no DB functions (since mysql
  isn't installed on nodes).  N.B. DO NOT DO A CLIENT INSTALL FROM YOUR
  REGULAR OBJ TREE OR ELSE YOU MAY WIND UP WITH A NEUTERED VERSION OF
  libtb.a!

The build-as-well-as-install semantics are counter to the regular install
targets, but this is what we gotta do for now.  Once the TB source builds
under Linux and newer BSDs, we could undo this and just require that people
do a regular "make" followed by "make client-install"  OTOH, there should
be no reason to require installation of mysql and other server-side packages
just to build clients (or make them sit through the compilation of assign),
so maybe we will keep the client build special.

common to all our machines.

removal of unnecessary TEST option. Pending change: still need to move this to standard emulab style.

Testbed levels correspond to the forthcoming documentation file.

Note: more changes are pending for the makefile related to moving to standard makefile style/suppressing errors.

named by ${vname}_startcmd, and that it can be controlled directly via
the NS "at" command or with tevc.

well known name so you can control it via the usual event mechanisms
(ns "at" and tevc). Previously, its name came from an internal
counter, and because the routine that creates the program object is
not at "toplevel", the parser would not rename the object. I'm sure
someone who understands this cruft could do better!

Anyway, now they are named as ${vname}_startcmd.

* In dosfshostid, store sfs hostid into the new table. Fix sscanf that
  was not checking for a null argument. Check string for any slashes
  and reject it (since we use the id in a symlink call). Sanitize the
  string we insert into the db with mysql_escape_string().

* Fill out dokostkeys function stub that Mike added storing keys into
  the DB. If a key is not presented, it is set to NULL in the DB. This
  may not be the most desirable approach; we will find out. Note that
  I assume a node_hostkeys table entry already exists, and that I can
  just update it.

  Note that this function is kinda ugly; a perl regex library for C
  would be so nice!

That Stoller boy had better do some ChangeLog updating too,
he has been a busy beaver.

Change headings to say matching lines. not first part of file.

soeftware called swish-e since webglimpse was broken, did not install
out of the box, no longer being maintained by the authors, and
generally a pain in the ass.

swish-e is probably overkill, but it works and can be configured and
run from a single configuration file (swish.conf.in).

I've removed the advanced search form and replaced with simple form.
I can bring back advanced stuff if needed, although better to wait
until the swish-e port is upgraded and there is better backend support
for generating the html output.

There is a backend script called websearch that is called from the php
search page. It invokes swish and spits out html that is amenable to
being inbedded in our existing page layout (we were not able to do
that with webglimpse).

directory. Add text labels to each node (node_id), although I am not
sure how useful it is. I also tried with vnames, but they are too big
and everything runs together.

of a live packet into a simulator packet must be performed right at the
time of capture before introducing the packet into the network. In 1.19,
I had optimized so that it is decapsulated just before the delivery of the
packet to the traffic agent which leads to simulation based on a wrong
packet size (i.e. the size of the encapsulated packet rather than the
simulated one).

thought I had committed these a long time ago.

internal errors.

script not in the web page.

active.

-e pid,eid option. When using -e, floor and building become optional, which
will restrict the display to just that building (and floor). Note that we
do not really have multiple building support yet, but we can pretend.

experiment. If non-zero, then add a menu option point to the wireless
floormaps for that experiment.

moment if its not "ethernet" its marked as wireless. We will need some
extra info if we ever branch out into non-ethernet, non-wireless lans.
Might also want to record what kind of wireless lans at some point?

Poller for this purpose. This works only on the FreeBSD 4.9 image in which
Mike has now incorporated a bpf fix (http://www.freebsd.org/cgi/query-pr.cgi?pr=kern/64178)
Another bugfix (http://www.freebsd.org/cgi/query-pr.cgi?pr=kern/64718)
should also be incorporated but will not cause problems with nse since we no longer
use BIOCSSEESENT flag anymore.

create_image script. Also check path; filename must translate to a path
on /proj, /users, or /groups since the image is actually written from
the node, and those are the only places it makes sense to write them to.

Minor change to web interface; email error messages to user *and* to
tbops; was going only to tbops.

links.

Now it is treaded just like ECONNRESET.

Also, don't clobber the read errno with the intermediate sigsetmask call
(though it should never fail and errno should never be affected)

Change warn() to warning() to avoid conflict with standard BSD library
routine.  This showed up when statically linking capture.