Integration tests
The ansible integration system.
Tests for playbooks, by playbooks.
Some tests may require credentials. Credentials may be specified with credentials.yml
.
Tests should be run as root.
Quick Start
To get started quickly using Docker containers for testing, see Tests in Docker containers.
Configuration
Making your own version of integration_config.yml
can allow for setting some
tunable parameters to help run the tests better in your environment. Some
tests (e.g. cloud) will only run when access credentials are provided. For
more information about supported credentials, refer to credentials.template
.
Prerequisites
The tests will assume things like hg, svn, and git are installed and in path.
(Complete list pending)
Non-destructive Tests
These tests will modify files in subdirectories, but will not do things that install or remove packages or things outside of those test subdirectories. They will also not reconfigure or bounce system services.
Run as follows:
make non_destructive
You can select specific tests with the --tags parameter.
TEST_FLAGS="--tags test_vars_blending" make
Destructive Tests
These tests are allowed to install and remove some trivial packages. You will likely want to devote these to a virtual environment. They won't reformat your filesystem, however :)
make destructive
Cloud Tests
Cloud tests exercise capabilities of cloud modules (e.g. ec2_key). These are not 'tests run in the cloud' so much as tests that leverage the cloud modules and are organized by cloud provider.
In order to run cloud tests, you must provide access credentials in a file
named credentials.yml
. A sample credentials file named
credentials.template
is available for syntax help.
Provide cloud credentials:
cp credentials.template credentials.yml
${EDITOR:-vi} credentials.yml
Run the tests: make cloud
WARNING running cloud integration tests will create and destroy cloud resources. Running these tests may result in additional fees associated with your cloud account. Care is taken to ensure that created resources are removed. However, it is advisable to inspect your AWS console to ensure no unexpected resources are running.
Windows Tests
These tests exercise the winrm connection plugin and Windows modules. You'll need to define an inventory with a remote Windows 2008 or 2012 Server to use for testing, and enable PowerShell Remoting to continue.
Running these tests may result in changes to your Windows host, so don't run them against a production/critical Windows environment.
Enable PowerShell Remoting (run on the Windows host via Remote Desktop): Enable-PSRemoting -Force
Define Windows inventory:
cp inventory.winrm.template inventory.winrm
${EDITOR:-vi} inventory.winrm
Run the tests: make test_winrm
Tests in Docker containers
If you have a Linux system with Docker installed, running integration tests using the same Docker containers used by the Ansible continuous integration (CI) system is recommended.
Using Docker Engine to run Docker on a non-Linux host is not recommended. Some tests, such as those that manage services or use local SSH connections are known to fail in such an environment. For best results, install Docker on a full Linux distribution such as Ubuntu, running on real hardware or in a virtual machine.
Running Integration Tests
To run all integration test targets with the default settings in a Centos 7 container, run make integration
from the repository root.
You can also run specific tests or select a different Linux distribution.
For example, to run the test test_ping
from the non_destructive target on a Ubuntu 14.04 container:
- go to the repository root
- and execute
make integration IMAGE=ansible/ansible:ubuntu1404 TARGET=non_destructive TEST_FLAGS='--tags test_ping'
Container Images
Use the prefix ansible/ansible:
with the image names below.
Running
make integration
will automatically download the container image you have specified, if it is not already available. However, you will be responsible for keeping the container images up-to-date usingdocker pull
.
Python 2
Most container images are for testing with Python 2:
- centos6
- centos7
- fedora-rawhide
- fedora23
- opensuseleap
- ubuntu1204 (requires
PRIVILEGED=true
) - ubuntu1404 (requires
PRIVILEGED=true
) - ubuntu1604
Python 3
To test with Python 3 you must set PYTHON3=1
and use the following images:
- ubuntu1604py3
Additional Options
There are additional environment variables that can be used. A few of the more useful ones:
KEEP_CONTAINERS=onfailure
- Containers will be preserved if tests fail.KEEP_CONTAINERS=1
- Containers will always be preserved.SHARE_SOURCE=1
- Changes to source from the host or container will be shared between host and container. CAUTION: Files created by the container will be owned by root on the host.
Network Tests
Note: From Ansible 2.3, for any new Network Module to be accepted it must be accompanied by a corresponding test.
For further help with this please contact gundalow
in #ansible-devel
on FreeNode IRC.
$ ANSIBLE_ROLES_PATH=targets ansible-playbook network-all.yaml
NOTE To run the network tests you will need a number of test machines and sutabily configured inventory file, a sample is included in test/integration/inventory.network
NOTE As with the rest of the integration tests, they can be found grouped by module in test/integration/targets/MODULENAME/
To filter a set of test cases set limit_to
to the name of the group, generally this is the name of the module:
$ ANSIBLE_ROLES_PATH=targets ansible-playbook -i inventory.network network-all.yaml -e "limit_to=eos_command"
To filter a singular test case set the tags options to eapi or cli, set limit_to to the test group, and test_cases to the name of the test:
$ ANSIBLE_ROLES_PATH=targets ansible-playbook -i inventory.network network-all.yaml --tags="cli" -e "limit_to=eos_command test_case=notequal"
Contributing Test Cases
Test cases are added to roles based on the module being testing. Test cases
should include both cli
and eapi
test cases. Cli test cases should be
added to test/integration/targets/modulename/tests/cli
and eapi tests should be added to
test/integration/targets/modulename/tests/eapi
.
In addition to positive testing, negative tests are required to ensure user friendly warnings & errors are generated, rather than backtraces, for example:
- name: test invalid subset (foobar)
eos_facts:
provider: "{{ cli }}"
gather_subset:
- "foobar"
register: result
ignore_errors: true
- assert:
that:
# Failures shouldn't return changes
- "result.changed == false"
# It's a failure
- "result.failed == true"
# Sensible Failure message
- "'Subset must be one of' in result.msg"
Conventions
-
Each test case should generally follow the pattern:
setup —> test —> assert —> test again (idempotent) —> assert —> -teardown (if needed) -> done
This keeps test playbooks from becoming monolithic and difficult to troubleshoot.
-
Include a name for each task that is not an assertion. (It's OK to add names to assertions too. But to make it easy to identify the broken task within a failed test, at least provide a helpful name for each task.)
-
Files containing test cases must end in
.yaml
Adding a new Network Platform
A top level playbook is required such as ansible/test/integration/eos.yaml
which needs to be references by ansible/test/integration/network-all.yaml