1. 29 Jul, 2003 3 commits
  2. 22 May, 2003 1 commit
    • Leigh B. Stoller's avatar
      Reorg the batch system slightly as per Eric's request that batch mode · da97ba35
      Leigh B. Stoller authored
      experiments look more like regular experiments. Batch mode experiments
      can now be preloaded and swapped. When preloaded, they go into a
      "Pause" state. Swapping a batch mode experiment in puts them into the
      "posted" state so the batch daemon will see them. Swapping out a
      batchmode experiment does the expected; it puts them back into the
      Pause state. Terminating a batch mode experiment does the expected;
      its gone. When a batch mode experiment finishes normally, it goes back
      into the pause state, which allows batches to be reinjected as many
      times as Eric likes.
      da97ba35
  3. 30 Apr, 2003 2 commits
    • Leigh B. Stoller's avatar
      Some batch mode changes. In the early days we did not have such fancy · 0197f41d
      Leigh B. Stoller authored
      tb tools! I've changed the batch system to "preload" the experiment in
      foreground mode (results of parse spit back to user directly). The
      batch daemon now uses swapexp instead of startexp. Upon failure, the
      experiment goes back to the "swapped" state; previously its virt state
      was blasted, and rentered again next try. This is nice cause you can
      actually look at the batch experiment (vis, virt tables, etc) while it
      is posted and not running.
      
      Not sure if all the Ts are crossed. Will find out ...
      0197f41d
    • Leigh B. Stoller's avatar
      Add batch/retry_wait sitevar, defaulted to 900 seconds between · ba8103b0
      Leigh B. Stoller authored
      retries. Change batch daemon to check that variable each loop.
      ba8103b0
  4. 16 Apr, 2003 1 commit
  5. 30 Jan, 2003 1 commit
  6. 16 Sep, 2002 1 commit
  7. 11 Jul, 2002 1 commit
  8. 07 Jul, 2002 1 commit
  9. 29 Apr, 2002 1 commit
  10. 12 Feb, 2002 1 commit
  11. 12 Nov, 2001 1 commit
  12. 08 Nov, 2001 1 commit
  13. 25 Oct, 2001 1 commit
  14. 17 Oct, 2001 1 commit
    • Leigh B. Stoller's avatar
      Rework of the batch experiment code. Unified it with the immediate · 4d420b21
      Leigh B. Stoller authored
      experiment code. No longer uses another table. Rather, the experiment
      record contains a couple of extra fields for the batch system. Also
      combined some of the backend code (no longer a killbatch script).
      Also added scriptable experiments; the batchexp program in the bin
      directory can start an experiment from the command line, and in fact
      is used from the web page for both batch experiments and immediate
      experiments (-i option). All of the DB code that was in the web
      interfaces was moved to batchexp.
      4d420b21
  15. 16 Oct, 2001 1 commit
  16. 26 Sep, 2001 1 commit
  17. 25 Sep, 2001 2 commits
  18. 21 Sep, 2001 1 commit
  19. 18 Sep, 2001 1 commit
  20. 18 Jul, 2001 1 commit
    • Leigh B. Stoller's avatar
      A set of fixes and updates to batchmode, which had suffered a fair · 32b16d6e
      Leigh B. Stoller authored
      degree of bit rot. Updated the create batch web page to do
      local/remote NS file spec. Switched to libdb interface in the batch
      mode scripts. Removed estimated pc/shark slots since we don't support
      that anymore, and because the new parser returns proper exit status
      when not enough nodes are available. The DB will be updated at that
      point, and we can incorporate that info (have not done so yet).
      Incorporated node boot status into isexpdone(), since a failed node
      will never report startcommand exit status. Clean up email code so
      that email goes to user/tblogs/tbops in the proper situations.
      32b16d6e
  21. 20 Jun, 2001 1 commit
  22. 10 May, 2001 2 commits
  23. 30 Apr, 2001 1 commit
  24. 19 Apr, 2001 1 commit
    • Leigh B. Stoller's avatar
      Watch for "not enough nodes" error status from startexp. Send email · 4eba6847
      Leigh B. Stoller authored
      every now and then. Also change the way experiments are selected to
      be configured. Instead of trying to start the same experiment over and
      over every 15 seconds, use a select to pick out experiments that have
      not been tried within the last 10 minutes. This will favor brand new
      experiments the first time, but after that all failed experiments are
      treated the same. The least recently attempted experiment over 10
      minutes is selected next.
      4eba6847
  25. 04 Apr, 2001 1 commit
  26. 28 Mar, 2001 1 commit
  27. 27 Mar, 2001 2 commits
  28. 25 Mar, 2001 2 commits
  29. 23 Mar, 2001 1 commit