1. 20 Dec, 2006 7 commits
  2. 19 Dec, 2006 6 commits
  3. 18 Dec, 2006 5 commits
    • Robert Ricci's avatar
      Added preliminary support for version 3 of libnetmon output. In this · 54711915
      Robert Ricci authored
      version, the unique identifier is the pid + fd, rather than anything
      involving port numbers. UDP sockets won't necessarily have any of
      I've made a few tweaks to Jon's v3 format. The format is not
      finalized, so some details could still change.
      Added preliminary support for UDP sockets. Right now, only handles
      sockets that are connect()ed, not ones that are used with sendto().
      We need to decide what the best way to report on sendto() is.  UDP
      support can be enabled by passing the new '-u' argument to netmond.
      Needs support for handling applications that fork(), since the PID
      changes. Shouldn't be too hard, just one more syscall to wrap.
      At some point, we should stop supporting earlier output versions,
      because that will allow me to clean up a lot of code.
    • David Johnson's avatar
      Bugfix for plab nodes. This problem was tripped by Kevin. What was · e53e402f
      David Johnson authored
      happening was that when Kevin swapmod'd to get rid of failed nodes,
      he just took the bad ones out.  This forced a change in the
      vname<->vnode mapping, and the failed node got put in a state
      (RES_INIT_CLEAN) that vnode_setup couldn't handle for plab nodes.
      Basically, the problem is that vnode_setup was assuming that the
      RES_INIT_CLEAN meant that the plab vnode needed to be allocated --
      but it was already allocated in the previous swap.
    • David Johnson's avatar
      A little script that renews our plab slices that were instantiated via · ed8e2ca5
      David Johnson authored
      the planetlab web interface (instead of over xmlrpc).  If the slice is
      due to expire within a week, I renew it for 4 weeks.  The script looks
      at /usr/testbed/etc/plab/metaslices to get the slices to check and
      renew.  I've put it in cron on boss so it runs nightly.
    • Jonathon Duerig's avatar
    • Robert Ricci's avatar
  4. 15 Dec, 2006 3 commits
    • Mike Hibler's avatar
      Attempt to reconcile the definition of some program-agent environment · 448e377b
      Mike Hibler authored
      variables on cluster and plab nodes.
      NODECNET (new)
      The fully-qualified (Emulab) name of the node this program agent is running on.
      This name resolves to the IP address of the control network interface
      of the node.
      NODECNETIP (new)
      The IP address of the control network interface.
      This address should not be advertised to, or used by, applications
      within an experiment as it will cause all traffic to flow over the control
      network rather than the experimental network.
      The unqualified name of the node this program agent is running on.
      For nodes with experimental interfaces,
      this name resolves to the IP address of an experimental interface on the node.
      For nodes with more than one experimental interface,
      there is no guarantee which one it will resolve to.
      For nodes with no experimental interfaces,
      the name will not resolve.
      The IP address of the experiment network interface that NODE resolves to.
      For nodes with no experimental interfaces, this variable will not be set.
    • Mike Hibler's avatar
      Missed a PELABSUPPORT change from yesterday · 06fa8210
      Mike Hibler authored
    • Robert Ricci's avatar
  5. 14 Dec, 2006 6 commits
  6. 13 Dec, 2006 5 commits
  7. 12 Dec, 2006 5 commits
  8. 11 Dec, 2006 3 commits