The goal of this project is to develop open source validation environment that will allow fully automated validation of DHCPv6 protocols compliance.


Requirements proposition:

  • program will consist of two parts, each working on separate virtual machines:
    • first, running in background on server machine responsible for:
      • installing specified version of DHCP server,
      • starting it with particular config (received from client),
    • second, running on other machine, responsible for:
      • DHCPv6 server job management (sending config files)
      • running/reporting/adding tests,
  • It's going to be extensible (in future tests not only for DHCPv6?)


Performing tests:

  • program will work only with the console (GUI not available)
  • no interactive interface
  • running tests, specify options by program arguments (e.g. ./program -r -t --RFC ...)
  • saving user option set (for future use):
    • e.g. ./ -r -t -RFC --SAVE my_set_1 then user can start program with: ./ my_set_1
  • choosing tests by:
    • by ID - single ID or set (10-30)
    • by rfc
    • by rfc section
    • by name
    • by type
  • tests will be carried out for a specific version of the server (specified before testing and identified by git ID). Each specific version will have it's own set of results.

first set of tests: program will make directory named as tested DHCP sever version and:

  • gitID.html <- list of all performed test (with basic information like: date, numbers of tests..)
  • testX.html <- information about this set of tests and their results. (X is set number)

next sets:

  • add to gitID.html basic information about this set
  • create:
    • testX.html <- information about this set of tests and their results.

informations contained in gitID.html (for each set of tests):

  • date and time
  • OS
  • number of performed tests
  • number of passed tests

informations contained in testX.html for each test in set:

  • name, ID, RFC reference
  • outcome
  • if test is failed, information about point of test in which test was failed

Test schema:

test identification

  • id
  • name
  • type
  • rfc
  • rfc section

test features:

  • short description
  • additional config file requirement
  • requirement for additional test if server works properly (in case of testing server 'no response')
  • pass criteria

test scenario

  • all test steps

the whole set of test will be here

Adding new tests:

Adding new test is still optional

  • different ways:
    • manually write full test
    • generation test file with blank space, just fill it and run.
    • automatic file parser (optional for now - no details yet)
  • program should have some 'uni-test' for new test validation, last think we want is running broken test
Last modified 7 years ago Last modified on Jan 29, 2013, 6:23:55 PM

Attachments (1)

Download all attachments as: .zip