Module usage with Tower

The Ansible modules should work normally in either Ansible Tower or Ansible AWX. There are, however, some things you should be aware of when using them within these environments.

Workers run in restricted environments

If you make use of any of the modules which download files to the Ansible controller (such as bigip_qkview or bigip_ucs, you should be aware that the Ansible worker that executes your job runs in a restricted environment.

In Tower, this environment is called a PRoot and functions nearly identically to a chroot. The net effect is that when files are saved by the job, they are saved (by default) into this chrooted environment. When the job finishes, that environment is cleaned up (ie, deleted) and your downloaded files will be deleted with it.

More background on the above can be found here and here.

In AWX, the situation is slightly different. AWX is distributed in the form of a set of Docker containers. Roughly the same restriction on the worker applies here, except that the worker is run in a container instead of a chroot.

In the end, the same situation applies though. The job will run on the container and the downloaded file will be stored in the container.

To remedy the above problems, you may want to change the Tower/AWX environment to put the files you download in a more durable location. This process is different for each of the products, so let’s look at some configuration

Tower

The Tower case is relatively easy to change.

First, we will be editing the /etc/tower/settings.py file. You can open that file for editing, and at the bottom of it, add the following

AWX_PROOT_SHOW_PATHS = ['/scratch']

Next, create that directory on your Tower VM and change the permissions of it to mirror those of the /tmp directory.

$ mkdir /scratch
$ chmod 777

Finally, restart Tower and wait for it to come back online

$ ansible-tower-service restart

When Tower comes back online, its job runner will be able to write files into the /scratch directory of the Tower VM. This is one possible way to make the system able to persist downloads longer than the lifetime of a particular job. Additionally, you maintain some of the safety against accidents that the chroot provides.

AWX

The AWX case is slightly different because it is distributed (by default) in container form.

For the following solution, I will refer to the Docker Compose method of running AWX because it is relatively simple and the same ideas can be applied to all major Container Orchestration Environments.

Provided you instructed the AWX setup script to create a docker-compose.yaml file, you can find the file in /var/lib/awx. To enable the system to write files to a more durable location on the AWX host, you can include a new bind mount to a location of your choice.

For instance, in the definition of my task service, I can specify the following

volumes:
  - "/scratch:/scratch:rw"

This will allow my task container (which runs the AWX jobs) to write to a durable location on the AWX VM.

Note that I additionally include the bind-mount shown below so that I can make use of the docker modules from within my AWX installation

volumes:
  - /var/run/docker.sock:/var/run/docker.sock

The above is handy to have.

After changing the docker-compose.yaml file, you can restart AWX by restarting the containers using docker-compose.

$ docker-compose stop
$ docker-compose start