Fixed bug where a single packet in a quantum caused an infinite loop. Fixed a bug where only socket buffer size updated in a quanum caused an infinite loop. Outstanding issue: Odd emulab side latency. Way too low.

Added support for the new buffer-exhaustion method of bandwidth measurement. It appears to do the right thing, but I still need to test it using tg.

Added a capped pending buffer which retains write and timing information. Minor reporting fixes. When I checked out start-experiment, it said that there was an RCS version and merged the changes into my version. Therefore checking in this version should be correct. Mike should doublecheck.

things like inter-write timings.

are the same size as iperf's . This will be addressed in a better way
in the future.

Fixed initialization bug in the bandwidth-measurement argument code. Fixed a bug where throughputTick was getting called too often. Added more data for DELAY_DETAIL logging. Removed Nagle's algorithm from the control traffic on the stub side. Added a new bandwidth option. Now there are three. Changed the naming scheme to be more specific (it was too generic before).

according to the input option: -m <method>. Currently when <method> =
"base" the goodput measurement using the whole packet size will be used;
when no -m is given, the optimized method drawn from TCP vegas will be
used by default.

this means we should now be able to invoke them via the program agent.

Fixed a time-calculation bug in the monitor. Added bandwidth and base-rtt calculations based on TCP-vegas. Miscellaneous cleanups.

Fixed a transient error discovered by Junxing. Rather, I've fixed a bug that could account for it and the error hasn't resurfaced after many runs. The transient symptom was when pcap only captured the first three packets and failed from then on.

All measurement times are now based on packet timestamps rather than gettimeofday(), throughput measurement is much more consistent now, though a more thorough evaluation is still required.

Delay is averaged over a quantum. This is probably not the right thing to do, but it is better than the alternative (latest measurement). I am still discussion with Sneha about what to do here.

automatically.

Now automatically determines the number of peers in the experiment.

run-monitor-libnetmon.sh automatically generates the IP mapping file
now.

Auto-create /local/logs if it doesn't exist (it doesn't on our
devbox image)

Add two new 'auto' scripts - these are intended to be run as part
of the automatic experiments. They properly wait for the stubs
to start before starting the monitors, etc. They log the outputs
of the stub and monitor to /local/logs, where loghole can pick
them up. They also kill their children if they are killed (to be
used with the program agent)

The auto-* scripts need a little more work in term of properly waiting
for their children to exit.

Added standalone support for multiple streams. Added a logging infrastructure. Added a trivial application for testing purposes (dumb-client and dumb-server). Added the capability for running the standalone stub with a tcpdump replay. Added support for changing buffer sizes. Added shell files for experiment management, reset.sh which is to reset the dummynet settings between runs and run-iperf.sh which encapsulates a set of options to iperf.

names and IP addresses for you.

Updated run-monitor-libnetmon.sh to use better information from
common-env.sh

New run-stub.sh to automatically get the interface name for you

Fixed several minor bugs in Rob's scripts. Added the capability for multiple streams along the same path to the monitor/stub loop. Replaced the linear searches with logarithmic searches. Cleaned up various associated parts of the code.

TODO: Finish connection recycling capability
TODO: Remove commented out older code after more debugging

reporting does not start until the first quanta in which  traffic is
recieved, and exits when the number of bytes in a quanta goes back
down to zero.

Also change quanta to 500ms, as suggested by Jon.

mode, or the db entry gets marked as invalid.

get, you simply run stubd with the '-s' (standalone) option.

intervals of exactly QUANTA.

However, it's my view that having stubm and stubd seperate is a bad
idea, since one is basically a copy of the other. They should be
merged back together so we don't have to fix the same bugs in both
places.

interface  on a node in an Emulab experiment. Assumes the name of
the 'Internet' LAN is 'planetc'.

change stub-monitor.c to work with the new stubd
fixed one typo in stub-pcap.c