This is a working technology preview of running Bacula (Open Source) 7.4.1 in multiple Docker containers.
To make it easier to get started a docker-compose.yml
is provided and it's
assumed that docker-compose
is used to start the containers. For actual
deployment the docker
binary ought to be used directly (e.g. by starting
the containers as a service).
The images are available on the Docker Hub as:
- redcoolbeans/bacula-opensource-db-data
- redcoolbeans/bacula-opensource-db
- redcoolbeans/bacula-opensource-dir
- redcoolbeans/bacula-opensource-sd
- redcoolbeans/bacula-opensource-fd
The following components are required for this setup:
- Docker >= 1.9.0
- Docker Compose >= 1.5.0
Instructions on how to install these on for example CentOS, can be found here and here. Please note that the Docker version shipped with CentOS 7 is too old and cannot be used.
This setup uses the networking features of Docker, therefore docker-compose
needs to be invoked with the --x-networking
flag. The remainder of this document
assumes that a shell alias has been setup to shrink repeated shell commands:
alias docker-compose='docker-compose --x-networking'
The local network that will be created is named after the directory the docker-compose.yml
file is located in. E.g. if the full path is /home/bat/docker/bacula
then the network will
be named bacula
and this can be verified (after docker-compose up
) with docker network ls
.
Then fetch the images:
docker-compose pull
When this has succesfully completed you can start the containers:
docker-compose up
This will spawn five containers, which can be verified with docker ps
:
- bacula-db-data: data volume container; will not actually run but provide a persistent storage
- bacula-db: PostgreSQL container, using storage provided by
bacula-db-data
- bacula-dir: Bacula Director container
- bacula-sd: Bacula Storage Daemon container
- bacula-fd: Bacula File Daemon (client) container
docker-compose will run in the foreground and multiplex the container's output to the controlling terminal. To run all containers in the background:
docker-compose start
To stop all containers one can simple hit ^C
if they run in the foreground, or when running the background:
docker-compose stop
Configuration files are stored in $PWD/configs/{dir,sd}/
relative to the
directory this README file is located.
The dir
/sd
directories are mounted read-only into the container to
/opt/bacula/etc/
In order to apply new configuration the associated container has to be restarted. For example:
docker-compose restart bacula-dir
There are several options which can be modified in docker-compose.yml
.
Changing their values does require either a docker-compose restart $CONTAINER
,
or a full restart of the setup if that's prefered (docker-compose
).
The most common options:
- BACULA_DEBUG: defaults to
50
- mem_limit: limit the available memory the container is allowed to use. Size is memory + swap, in bytes.
bconsole can be invoked from outside of the container:
docker exec -ti bacula-dir bconsole
A Bacula client container is started which runs the client daemon. The following environment
variables are used to configure it's bacula-fd.conf
:
- FD_NAME
- FD_PASSWORD
- DIR_NAME
- MON_NAME
- MON_FD_PASSWORD
Of which only FD_PASSWORD
must be set, the others have default values.
These images are provided free of charge by RedCoolBeans. Various security measures have not been implemented or added in these images. For supported and security hardened images with Bacula Enterprise, please contact us.
- upgrades between Bacula versions is not yet accounted for
- e-mail notifications are not enabled in the default configuration shipped with this demonstration setup. If e-mail is desired it can be enabled and pointed to an existing mailhost.
- Mounting a local
bacula-fd.conf
into thebacula-fd
container is not supported yet. Due to the limited configuration possible in this file it is advised to set the documented environment variables for now.
MIT, please see the LICENSE file.
- Fork it
- Create your feature branch (
git checkout -b my-new-feature
) - Commit your changes (
git commit -am 'Add some feature'
) - Push to the branch (
git push origin my-new-feature
) - Create new Pull Request