Added support for queue-length-in-time. Fixed a bug where unusually small bandwidth measurements were sent during slow start.

the user has requested a per-experiment DB on ops (or wherever the DP
is). As you see, its an optional thing that will probably only be used
for templates.

the verify parse.

FIN timeout thus getting more accurate BW measurements.  From the code:

                  #
                  # Even without packet loss, there is a good chance that
                  # iperf's end-of-stream (FIN) will be lost since we are
                  # over-driving the link.  The ACK for that FIN needs to
                  # be received before iperf's timer stops, so a long
                  # timeout (for resending the FIN) on a short duration
                  # run will lower the BW reading considerably.  So we
                  # attempt to keep the timeout as short as possible,
                  # taking into account the round-trip latency of the link.
                  # We may also need to adjust the run time of the test
                  # upward, but we do not exceed the indicated max duration.
                  #
                  # 250 ms is the default timeout value for iperf.  We
                  # start with a candidate value of 50 ms which is chosen
                  # as it is sufficient for the 3% BW error on min duration
                  # (3 second) tests.  You should not pick a base acktime
                  # less than the resolution of the clock (10ms or 1ms).
                  #

wait for an ACK of its FIN.  The fixed 1/4 second was too big for us.
The option gets passed over to the server when running in tradeoff or
dual modes.

Record page lets you export the contents of the archive that corresponds
to that record, along with an XML file that describes the various DB bits
for the template and instance.

This is just a first cut so that Mike can start playing around. Subject to
change, I'm sure.

The archive is dumped to /proj/$pid/exports/$guid/$vers/$exptidx, which
is basically the last commit of the instance when it was terminated.

The xml file is called export.xml and is placed in the top level directory
of the above directory. The file is created with XML::Simple, and a typical
XML file might look like:

<instance>
  <bindings>
    <name>NodeCount</name>
    <description>Number of nodes!</description>
    <value>1</value>
  </bindings>
  <bindings>
    <name>OS</name>
    <description></description>
    <value>RHL90-STD</value>
  </bindings>
  <bindings>
    <name>ScriptArgs</name>
    <description></description>
    <value>-b</value>
  </bindings>
  <eid>NewOne-V2</eid>
  <guid>10149/2</guid>
  <metadata>
    <name>M1</name>
    <guid>10162/1</guid>
    <value>Some metadata</value>
  </metadata>
  <pid>testbed</pid>
  <runs>
    <name>1</name>
    <archive_tag>T20060526-082533-172_endexp</archive_tag>
    <description></description>
    <exptidx>110</exptidx>
    <idx>1</idx>
    <runid>NewOne-V2</runid>
    <start_time>2006-05-26 08:23:02</start_time>
    <stop_time>2006-05-26 08:25:16</stop_time>
  </runs>
  <uid>stoller</uid>
</instance>

available to the public yet.

cleanup the database before this fix is installed.

  + On the swap-in status page (showing the swap-in progress), there's a link
    to the template page but no link to the instance page.  Conversely, on the
    swap-out page ("Terminate Instance"), there's a link to the instance but
    not to the template.

zero.

See event/linktest/CHANGES for details.

use loghole to capture the log files and store them in the experiment
archive before the commit point between runs (and at swapout).

Some other small unrelated changes as well.

of the logs to the archive.

up waiting node. Turns out to be a fd problem with stdin and ssh, and
the use of popen in the script.

Sigh, gonna need new images ...

Iperf processes timeout after trying too long.

queries all totally wrong.

draw in odd places sometime.

but that stuff is still not totally hooked in yet.