.. This work is licensed under a Creative Commons Attribution 4.0 International .. License. .. http://creativecommons.org/licenses/by/4.0 .. (c) OPNFV, Huawei Technologies Co.,Ltd and others. =========================== Dovetail Tool Installation =========================== Abstract ######## The Dovetail tool supports installation either directly on the Linux host, or within a Docker container. The detailed installation procedure on the Linux host or via the Docker container are shown in the following sections. There is a need to config the following parameters before Dovetail tool running on the SUT (System Under Test): :: SUT_TYPE, SUT type, e.g., apex, compass, fuel, joid, etc SUT_IP, SUT external network IP, e.g., 192.168.200.2 NODE_NAME, this can be shown in the test result for users to see which pod the dovetail tool runs DEPLOY_SCENARIO, deployment scenario, e.g., os-nosdn-nofeature-ha BUILD_TAG, this can be shown in the test result for users to identify logs CI_DEBUG, true for debug information printed and false for not printed DEPLOY_TYPE, baremetal or virtual Dovetail tool installation on local Linux host environment ########################################################## In order to install Dovetail into a local Linux host environment, the following steps should be performed: Downloading Dovetail tool -------------------------- Source code downloading: :: git clone https://gerrit.opnfv.org/gerrit/dovetail Environment preparation ----------------------- Dovetail configuration: :: cd {dovetail_path}/dovetail/conf vim functest_config.yml vim yardstick_config.yml To modify the ``envs`` parameters, e.g., SUT_TYPE, SUT_IP, NODE_NAME, DEPLOY_SCENARIO, BUILD_TAG, CI_DEBUG, DEPLOY_TYPE, in the yml files. System dependencies and project configuration information are listed in ``requirements.txt``, ``setup.py`` and ``setup.cfg``, in Ubuntu environment, to run :: apt-get -y install python-pip pip install -e . in ``centos`` and ``redhat`` environment, to run :: yum -y install python-pip pip install -e . in ``fedora`` environment, to run :: dnf -y install python-pip pip install -e . then the setup requirements are installed and **dovetail** commmand-line can be used, by using ``dovetail --help``, **dovetail** command-line details are shown. The tool ``tox`` is used, which configuration information is defined in ``tox.ini``, unitttest, flake8-checking, coverage check, etc, can be done by running: :: tox Compliance and certification test cases ---------------------------------------- The compliance and certification test cases can be defined under the ``/dovetail/compliance`` directory, which is defined in yaml format. A sample file named ``compliance_set.yml`` is provided as follows: :: compliance_set: name: compliance_set testcases_list: - dovetail.ipv6.tc001 The testcase listed here is dovetail.ipv6.tc001, defined within ``dovetail/testcase``. Running Dovetail tool --------------------- After environment preparation is complete and test cases added, the Dovetail tool can be run with :: python run.py --testsuite compliance_set The value ``compliance_set`` passed to the ``testsuite`` flag can be replaced with the test cases yaml file. If not argument is given, the compliance_set testsuite will be run as the default. Moreover, the testcases in given testarea can be run with ``testarea`` command line argument, such as testarea ``ipv6`` in ``compliance_set`` :: python run.py --testsuite compliance_set --testarea ipv6 Dovetail provides some sets, ``debug``, ``proposed_tests`` and ``compliance_set``, ``debug`` is used for locally and Continuous Integration(CI) developing purpose, which provides typical testcase examples, feel free to edit it when develops locally, such as only to run a testcase which only takes minutes. ``proposed_tests`` is the testcase candidate which mainly comes from the wiki link https://wiki.opnfv.org/display/dovetail/Dovetail+Test+Areas+and+Test+Cases. ``compliance_set`` is used for compliance. Moreover, dovetail tool can be easily extended to support more complicated compliance requirements, such as feature set based or scenario based compliance. If you want to run ``debug``, just run with :: python run.py --testsuite debug Running Dovetail in a Docker container ######################################## The Dovetail tool can be run in a Docker container by utilizing the following steps: Pull Dovetail Docker image from public Dockerhub ------------------------------------------------ :: sudo docker pull opnfv/dovetail: here is the version, 'latest' is used for the master branch. (Optional) Create Docker image from Dockerfile ----------------------------------------------- Instead of an official Docker image, you can build your own Docker image from Dockerfile(s) located under the ``docker/`` directory. The ``Dockerfile`` is based on Ubuntu and the ``Dockerfile.centos7`` file is for a CentOS based Docker image. :: sudo docker build -t -f . Dovetail Docker container creation ---------------------------------- Next, create the ``dovetail-docker-env`` file to define the environment parameters :: INSTALLER_TYPE=compass INSTALLER_IP=192.168.200.2 DEPLOY_TYPE=baremetal DEPLOY_SCENARIO=ha-nosdn CI_DEBUG=true Then to instantiate the Dovetail Docker container, execute:: sudo docker run --privileged=true --rm -t \ --env-file dovetail-docker-env \ -v /home/opnfv/dovetail/results:/home/opnfv/dovetail/results \ -v /var/run/docker.sock:/var/run/docker.sock \ --name \ opnfv/dovetail: /bin/bash To attach dovetail container and Running test cases ---------------------------------------------------- Before connecting to the container, you can check the container status by running :: docker ps -a Attach to the container by starting it and obtaining a bash prompt with :: docker exec -it bash Inside the container the following commands can be executed to trigger the testcases :: cd /home/opnfv/dovetail/dovetail python run.py --testsuite compliance_set Results Output ############### The running log is stored in ``/home/opnfv/dovetail/results/dovetail.log``. The compliance report is stored in ``/home/opnfv/dovetail/results/dovetail_report.txt``. Dovetail Version and Release ############################ Dovetail version tag is shown in ``setup.cfg``, which will also shown in the ``dovetail report``. At the time of version release, just to set the version value in ``setup.cfg``. # TO DO: (which should be discussed) 1)how to pubish version, such as both the online and offline package in some website or somewhere. 2)provide version download address, userguide, etc.