User Tools

Site Tools


Testbed testing

The objective of this page is to have a summary of the status of maturity of the testbed and provide links to other resources. It should help developers and testbed operations to have an overview of all, and to researchers to know the current global status of the testbed.

This entails several interdependent tasks and resources (with examples and people involved):

  • Controller centric (e.g. slice CRUD, sliver CRUD)
    • Description: the controller implements all main features and many more
    • Status: all works, testing features, what's validated and known to work,
    • What's left: list, add your name to volunteer for it.
    • Contact: if you find any issue report it to the bug tracker or contact or
  • Node centric (e.g. keeping RD nodes up, up-to-date with latest image, which nodes are “usable”)
    • Description: there are 46 nodes already registered, how many with latest version? All the nodes in the lab have the last image that controller provides. To check nodes&slices IPs:
    • Status: VCT images repo (axel upload to repo), Node images repo (Axel/Pau compiles and upload to Confine repo), node updates (Albert/Jorge do manual update of the nodes in the lab and external ones when needed)
    • What's left: list, add your name to volunteer for it.
      • VCT debugging.
      • Use a well-defined (at least documented) sliver deployment procedure (proposal), or at least document the current one so that researchers know what to expect.
    • Contact: General issues: / Technical issues:
  • Experiment centric (e.g. running a trivial experiment)
    • Description: BT experiment, routing experiment, any other?
    • Status: BT experiment works with current testbed sw? (Davide?), routing experiment works with the VCT (pending to test controller-nodes interaction)
    • What's left: list, add your name to volunteer for it.
      • Routing experiment (Ester): test directly with the controller without using VCT
    • Contact: if you find any issue go HERE or contact HIM
  • Monitoring/mgmt centric (e.g. having a monitoring service running)
    • Description: Monitoring ready for public deployment
    • Status: Ready for public deployment
    • What's left: add monitoring sw to nodes (e.g. as an update in a package, as part of the node image ? by Navaneeth with help from WHO?), deploy the monitoring server in a VM (by Navaneeth with help from Marc)
    • Contact: if you find any issue go HERE or contact Navaneeth.

If you want to / you should know that:

  • Register to the testbed: only for Confine partners here, it is not yet open for external usage.
  • Run an experiment you should know that you can do … and not yet … on the nodes marked as …
  • Add/update a node …
  • See the status of the testbed …
  • Get support …
soft/testbedtest.txt · Last modified: 2013/04/09 10:04 by avellana