1. 11 Aug, 2006 3 commits
  2. 10 Aug, 2006 13 commits
    • Mike Hibler's avatar
      Add TBDBDisconnect to go with TBDBConnect. · d6dd8938
      Mike Hibler authored
    • Dan Gebhardt's avatar
      Major re-do of the initial condition gathering. · e0420109
      Dan Gebhardt authored
      Available data elements in initial condition structure:
      - Exponential average for bandwidth and latency,
      - Number of samples used
      - Number of error-val samples
      - Number of sequential error-val from newest measurement, backwards
      - timestamp of most recent measurement
      - source node
      - destination node
      Testing needed.
    • Kirk Webb's avatar
      · 7901885c
      Kirk Webb authored
      The other half of the changes that cause the plab event proxy to now try
      to get the routable IP of the node from tmcd rather than relying on
      the success of a hostname lookup.  It will still fall back to trying a
      hostname lookup if it can't get the IP from tmcd.
    • Kirk Webb's avatar
      · e087f217
      Kirk Webb authored
      Send along the IP address of the plab node in the return string from
      the 'plabconfig' command.  We can't trust that the node will have
      a resolvable hostname (or have working DNS even..), so slap down the IP
      we have on record in the DB into a file.  This will be used by the
      event proxy, which needs to know the node's routable IP in order to
      subscripe to elvind on ops properly.
    • Leigh B. Stoller's avatar
      Minor fixes. · 5766e95e
      Leigh B. Stoller authored
    • Robert Ricci's avatar
    • Jonathon Duerig's avatar
    • Robert Ricci's avatar
      Add the timestamp at which the connect() occurs for Jon. Uses a new, · 10fa9140
      Robert Ricci authored
      function, fprintTime() which will be used to standardize the time
      Also added const to some declarations to keep the compiler happy.
    • Robert Ricci's avatar
      Use the C99 standard · c4e21e5a
      Robert Ricci authored
    • Leigh B. Stoller's avatar
      Okay, now we can view graphs from the historical data (template record). · 0c1b1a23
      Leigh B. Stoller authored
      A couple of things to note:
      * When requesting a graph, we have to have a checkout of the archive
        (the DB dump file) so that we can create a temporary DB with the data.
        This is done on demand, and the DB is left in place since its a
        fairly time consuming operation to do the checkout and the dbload.
        I do not delete the DBs though; we will need to age them out as needed.
      * Even so, when returning to a page we end up getting the graphs
        again, and that still takes more time then I like to wait. Perhaps
        add a refresh button so that the user has to force a redraw. Might
        need to add a time/date stamp to the graph.
    • Mike Hibler's avatar
      First crack at surviving down planetlab nodes. If the master barrier sync · 5f413b47
      Mike Hibler authored
      node sits in the stub or monitor barrier sync for more than the SYNCTIMO
      timeout value in common-env.sh, it will send a HUP to syncd which will
      knock all the other nodes out of their barrier sync.  If that happens,
      all nodes will print a warning message and continue.
      All nodes wait for both a stub sync and a monitor sync, so if one plab node
      is down, they will timeout on both barrier syncs.  Race conditions?  Sure.
      If for example everyone times out on the stub barrier due to a slow node,
      and then that node reaches the barrier, it will hang there while everyone
      else waits on the monitor barrier.  When the latter times out, it will
      kick the slow node out of the stub sync and it will then proceed to hang
      in the monitor sync until the experiment is stopped.  Got that?
      As an aside, it would be nice if the initializer of a barrier could specify
      a timeout value, and return a special error code to everyone if it timed out,
      but that would require an incompatible change to the sync protocol.
    • Mike Hibler's avatar
      Minor tweaks: · 4c5005da
      Mike Hibler authored
      * add getopt processing
      * adjust delay to be one way before calling tevc
    • Leigh B. Stoller's avatar
      Next checkpoint of graphing code. On a currently active template · 79ae0bfe
      Leigh B. Stoller authored
      instance there are graphs on the instance show page and on the
      individual run show pages. On the run pages, the graphs select just
      the packets between start and stop of the run. I also added drop down
      menus to select particular source and destination vnodes.
  3. 09 Aug, 2006 15 commits
  4. 08 Aug, 2006 8 commits
  5. 07 Aug, 2006 1 commit