This setup is intended to run multiple builds against a specific Indy instance.

To install, you may want to use a virtualenv:

    $ virtualenv ./venv
    $ source ./venv/bin/activate

Then, you can install in "developer mode" using:

    $ pip install -e .

To run a test, the user must capture the build parameters in a YAML file. This file should be in its own directory, since the output / logs / other data related to the build execution will also be captured there. This allows the user to keep track of the circumstances under which a particular type of build was run, and the result of that test.

The basic structure is:

    ./my-test-build
    +- test.yaml

You can find a sample of the build specification in sample-testfile.yaml. The build specification allows the user to specify the following:

• HTTProx proxy port (if used)
• section for build, containing:
  • number of build threads
  • number of total builds
  • project directory to use as a location for cloning build sources
• section for report verification, containing:
  • number of threads for verifying folo tracking reports
• a vagrant section, containing test initialization parameters for the VMs, with the following sections:
  • pre-build
  • post-build
  • pre-report
  • post-report

Each of the vagrant sub-sections outlined above can contain a copy section and a run section. The copy section looks like this:

    copy:
      - <src>: <dest>
      ...

The run section is a little more involved:

    run:
      -
        host: nfs
        commands:
          - 'sudo ls -alh /tmp'
          ...
      -
        host: indy
        commands:
          - 'sudo do-something-cool'
        wait-for-indy: true
      ...

This section allows you to specify a script of actions and optionally wait for Indy to start responding again (in case you restart it).

The user must provide the following arguments to the multibuild command:

• TESTFILE - A path to the YAML file, which is outlined above. This file's directory will collect the build results
• INDY_URL - An Indy URL, with everything setup to work with the public group

To use this with the Vagrant configuration contained in this project, run the VMs. You can use the instructions in the root README to fine-tune these VMs for your test.

Use vagrant ssh-config indy in the root directory of this repository to get the host IP address for the Indy instance. Once you have this, you can use http://<IP>:8080 as the Indy URL for input into the multibuild script.

Starting from the project root directory (directory above this one):

    $ vagrant up
    $ export INDY=$(vagrant ssh-config indy | grep HostName | awk '{print $2}')

    [CONNECT TO INDY UI AT http://${INDY}:8080 and configure it appropriately]

    $ cd multibuild
    $ git clone https://github.com/Commonjava/indy project
    $ multibuild my-test/test.yaml http://${INDY}:8080

    [BUILDS RUN]