GitLab

application directly.

Add a new script, instrument.sh, that runs your app with libnetmon,
which will connect to netmond, and thus to the monitor.

Change both to allow them to be run from arbitrary directories - you
no longer have to cd into the monitor directory to run them.

   both directions at the same time again. I had put in a delay, attempting
   to see if that might help the packet loss problems; I do not think it
   does. So, loss tests will complete faster.

2) Changed the duration for the bw tests as we discussed. I did a quickie
   trial, and ended up with:

	    x > 33Mbs	3 seconds
    33Mbs > x > 10Mbs	5 seconds
    10Mbs > x > 64Kbs   7 seconds

   Then add 3 seconds if the loss if over 10%

   The above numbers are very conservative; we can probably do better, but
   it would take a longer set of trials (across a variety bw/delay
   products). Remember, all bw tests ran for 10 seconds (in each direction).

3) Add in the ethernet/ip/udp overhead to the bw results; Much nicer. I
   reduced the allowed error factor fom 6% to 2%, although its actually
   within less then 1% most of the time.

4) Allow BW checks when there is loss on the link; previously linktest
   skipped any link with loss on it. Since we UDP for the tests, we
   can fairly well estimate what the BW will be. This is not perfect
   yet, especially with low BW, high loss links. We need to run some
   trials to see where accuracy falls off too far, or how to
   compensate for it.

5) Added some more time stamps.

Parse the NS file with the real NS parser so we can make sure linktest is
doing the "right" thing.

	* configure, configure.in: Add tbsetup/nsverify files.

	* tbsetup/GNUmakefile.in: Add nsverify subdir.

	* tbsetup/tbprerun.in: Run verify-ns on the experiments NS file.

	* tbsetup/ns2ir/nstb_compat.tcl: Bring up-to-date with the current
	world.

	* tbsetup/nsverify/GNUmakefile.in: Makefile.

	* tbsetup/nsverify/ns-2.27.patch: Patch file for NS version 2.27.

	* tbsetup/nsverify/nstbparse.in: Wrapper for the NS parser.

	* tbsetup/nsverify/tb_compat.tcl: Different version of
	tb_compat.tcl that is used to verify linktest parameters.

	* tbsetup/nsverify/verify-ns.in: Script that runs on boss and
	verifies that the testbed parser worked correctly.

	* tbsetup/ns2ir/parse-ns.in, tbsetup/ns2ir/parse.proxy.in: Tweaked
	a bit so parse.proxy can be used to run the regular NS parser in
	addition to the testbed one.

OK - really exit when I say so.  Note: sys.exit() _and_ sys._exit() throw
exceptions.  Must use os._exit() to exit immediately.

More info when exiting due to a signal.

libplab.py.in:

Exit right away when signalled while trying to perform a remote command.

vnode_setup.in:

More info when a timeout occurs, and reduce the execution spacing a little.

Bump rootball revision.

Create our own passwd file inside plab vservers.  We add the slice user
to this passwd file, with the password starred out.  This fixes the gaping
security hole resulting from the passwordless root and slice user accounts
that exist in the default plab vserver environment (in conjunction with our
per-vserver sshd).

-Doc permanent behavior: No email when explicitly invoked.

from apps being monitored with libnetmon. This is so that we can
invoke netmond from the monitor, and attach whatever process we want
to it.

If LIBNETMON_SOCKPATH is set in the environment, libnetmon will now
spit results to it rather than stdout.

the other way of invoking the monitor, this one takes the application
to run on the command line.

Monitor now finishes when it gets an eof on stdin.

Allow monitor.py to skip input lines it doesn't understand, since when
run with libnetmon, it sees the stdout of the application.

Implement the write() wrapper, so that it works with wget

assign_wrapper check the OSID of each physical node in a link/lan to
make sure that they support linkdelays.

Fixed an overflow bug in the pcap code. Added a timer to make throughput more accurate (this still needs tweaking and isn't quite right).

Also, skip sockets that are not IP or not TCP.

Two bugfixes. One caused the receiver to hang. The other prevented select from noticing that the sender was able to write.

by an applicaion.