1. 18 Dec, 2009 1 commit
    • Leigh B. Stoller's avatar
      Changes to support the SPP nodes. My approach was a little odd. · fd015646
      Leigh B. Stoller authored
      What I did was create node table entries for the three SPP nodes.
      These are designated as local, shared nodes, reserved to a holding
      experiment. This allowed me to use all of the existing shared node
      pool support, albeit with a couple of tweaks in libvtop that I will
      not bother to mention since they are hideous (another thing I need to
      fix).
      
      The virtual nodes that are created on the spp nodes are figments; they
      will never be setup, booted or torn down. They exist simply as place
      holders in the DB, in order hold the reserved bandwidth on the network
      interfaces. In other words, you can create as many of these imaginary
      spp nodes (in different slices if you like) as there are interfaces on
      the spp node. Or you can create a single spp imaginary node with all
      of the interfaces. You get the idea; its the reserved bandwidth that
      drives the allocation.
      
      There are also some minor spp specific changes in vnode_setup.in to
      avoid trying to generalize things. I will return to this later as
      needed.
      
      See this wiki page for info and sample rspecs:
      
      https://www.protogeni.net/trac/protogeni/wiki/SPPNodes
      fd015646
  2. 17 Dec, 2009 1 commit
  3. 15 Dec, 2009 2 commits
  4. 14 Dec, 2009 1 commit
  5. 11 Dec, 2009 1 commit
  6. 03 Dec, 2009 1 commit
  7. 02 Dec, 2009 1 commit
  8. 13 Nov, 2009 1 commit
  9. 11 Nov, 2009 1 commit
  10. 10 Nov, 2009 2 commits
  11. 09 Nov, 2009 1 commit
  12. 07 Nov, 2009 1 commit
    • Leigh B. Stoller's avatar
      Change to infodir (/usr/testbed/expinfo) handling; experiment · 41d34103
      Leigh B. Stoller authored
      directories are now placed in a project subdirectory, to avoid
      blowing out the max number of subdirs (32K in FreeBSD). Dirs are
      now called $pid/$eid/$idx.
      
      Added sanity checks to batchexp, swapexp, and endexp to watch for the
      case that testbed admin installed the new code but did not run the
      fixup script as instructed in doc/UPDATING.
      41d34103
  13. 29 Oct, 2009 2 commits
  14. 28 Oct, 2009 2 commits
  15. 27 Oct, 2009 1 commit
    • Mike Hibler's avatar
      Strange, for some reason I started getting warnings from the SNMP code · 25c8af39
      Mike Hibler authored
      about "mib-2" and some others.  This is defined in SNMPv2-SMI which wasn't
      in the list of MIB files and is needed by SNMPv2-MIB.  But, it probably
      shouldn't need to be explicitly included, since there is an IMPORT of the
      symbols in question at the beginning of SNMPv2-MIB.  But I added it to
      the SNMP::addMibFiles() arguments anyway.
      
      This came about because I was messing with PowerNet-MIB.txt to support the
      7960 power controllers (we have an 8 year old version of that file).
      25c8af39
  16. 22 Oct, 2009 2 commits
  17. 21 Oct, 2009 1 commit
  18. 20 Oct, 2009 3 commits
    • Keith Sklower's avatar
      This version includes heavy duty machinery for parallelizing activities · c8781411
      Keith Sklower authored
      among all switches in a stack.  Does check to see if the required additional
      perl modules are installed and will revert to standard behavior if not there.
      
      Currently only looking up vlan numbers from all the switches in the stack
      is actually parallelized; this alone results in a speed up by a factor of 5
      on DETER.
      
      Additional speedups could be gained if we assign all the vlan numbers for
      vlans in snmpit -t at the time it is run thereby allowing us to
      construct different vlans in different switches in an overlapping way.
      
      We haven't implemented this yet on DETER awaiting synchronizing with the
      mothership to make use of VLan->setVlanTag().
      c8781411
    • Keith Sklower's avatar
      add --vlan_tag option to snmpit (for use with dragon signaling) · 59ae8faa
      Keith Sklower authored
      handle quieting backends in a different way - (to minimized changes
      to nortel, foundry, and hp support), use perl "select" with one arg
      to change default for "print" with no explicit fileglob.
      
      print STDERR, STDOUT, still works; won't interfere with tblog() and friends.
      will put parallization mods in a separate CVS change.
      59ae8faa
    • Jonathon Duerig's avatar
  19. 18 Oct, 2009 1 commit
  20. 16 Oct, 2009 1 commit
    • David Johnson's avatar
      Make shared vnodes reloadable. This whole thing sucks for modifies · abcc783c
      David Johnson authored
      because we (vnode_setup) needs to go out to the nodes and run vnodesetup
      to trigger the reload, but os_setup needs to setup the reload.  So for
      now, os_setup sets up the reload but does not wait nor reboot the vnode;
      vnode_setup does that like normal.  Probably there are going to be timeout
      problems, but it's good enough for my needs right now.
      abcc783c
  21. 13 Oct, 2009 1 commit
  22. 12 Oct, 2009 3 commits
    • David Johnson's avatar
      Whoops, fix silly query. · 984379c6
      David Johnson authored
      984379c6
    • David Johnson's avatar
      Make the mapper work with Leigh's latest libvtop checkpoint... probably · c0c52c68
      David Johnson authored
      will get worked out better after the latest checkpoint :-).
      c0c52c68
    • David Johnson's avatar
      Add the ability to load images on virtnodes. For now, we just overload · c6c57bc9
      David Johnson authored
      the tb-set-node-os command with a second optional argument; if that is
      present, the first arg is the child OS and the second is the parent OS.
      We add some new features in ptopgen (OS-parentOSname-childOSname) based
      off a new table that maps which child OSes can run on which parents, and
      the right desires get added to match.  We setup the reloads in os_setup
      along with the parents.  Also needed a new opmode, RELOAD-PCVM, to handle
      all this.
      
      For now, users only have to specify that their images can run on pcvms, a
      special hack for which type the images can run on.  This makes sense in
      general since there is no point conditionalizing childOS loading on
      hardware type at the moment, but rather on parentOS.  Hopefully this stuff
      wiill mostly work on shared nodes too, although we'll have to be more
      aggressive on the client side garbage collecting old frisbee'd images for
      long-lived shared hosts.
      
      I only made these changes in libvtop, so assign_wrapper folks are left in
      the dark.
      
      Currently, the client side supports frisbee.  Only in openvz for now, and
      this probably breaks libvnode_xen.pm.  Also in here are some openvz
      improvements, like ability to sniff out which network is the public
      control net, and which is the fake virtual control net.
      c6c57bc9
  23. 07 Oct, 2009 1 commit
  24. 05 Oct, 2009 1 commit
  25. 02 Oct, 2009 1 commit
  26. 29 Sep, 2009 1 commit
  27. 27 Sep, 2009 1 commit
  28. 25 Sep, 2009 2 commits
  29. 24 Sep, 2009 2 commits