Skip to content

Latest commit

 

History

History
133 lines (96 loc) · 5.98 KB

README.md

File metadata and controls

133 lines (96 loc) · 5.98 KB
Raw

Docker

Using This Image

You will need a PostgreSQL instance URL in the format postgresql://[username]:[password]@[host]:[port]/[database_name]. Check the ./docker-compose.yml file for an example. With this URL, you can create containers for both the web application (circ-webapp) and for the background cron jobs that import and update books and otherwise keep the app running smoothly (circ-scripts). Either container can be used to initialize or migrate the database. Database initialization uses a PostgreSQL Advisory Lock to make sure that only one instance is updating the schema at a time. Database migration uses Alembic to update the schema to the latest version. This initialization or migration is done automatically when the container is started.

circ-webapp

Once the webapp Docker image is built, we can run it in a container with the following command.

# See the section "Environment Variables" below for more information
# about the values listed here and their alternatives.
$ docker run --name webapp -d \
    --p 80:80 \
    -e SIMPLIFIED_PRODUCTION_DATABASE='postgresql://[username]:[password]@[host]:[port]/[database_name]' \
    ghcr.io/natlibfi/ekirjasto-circ-webapp:main

If the database and OpenSearch(OS) are running in containers, use the --link option to let the webapp docker container to access them as bellow:

docker run \
--link pg --link os \
--name circ \
-e SIMPLIFIED_PRODUCTION_DATABASE='postgresql://[username]:[password]@[host]:[port]/[database_name]' \
-d -p 6500:80 \
ghcr.io/natlibfi/ekirjasto-circ-webapp:main

Navigate to http://localhost/admin in your browser to visit the web admin for the Circulation Manager. In the admin, you can add or update configuration information. If you have not yet created an admin authorization protocol before, you'll need to do that before you can set other configuration.

circ-scripts

Once the scripts Docker image is built, we can run it in a container with the following command.

# See the section "Environment Variables" below for more information
# about the values listed here and their alternatives.
$ docker run --name scripts -d \
    -e TZ='YOUR_TIMEZONE_STRING' \
    -e SIMPLIFIED_PRODUCTION_DATABASE='postgresql://[username]:[password]@[host]:[port]/[database_name]' \
    ghcr.io/natlibfi/ekirjasto-circ-scripts:main

Using docker exec -it scripts /bin/bash in your console, navigate to /var/log/simplified in the container. After 5-20 minutes, you'll begin to see log files populate that directory.

circ-exec

This image builds containers that will run a single script and stop. It's useful in conjunction with a tool like Amazon ECS Scheduled Tasks, where you can run script containers on a cron-style schedule.

Unlike the circ-scripts image, which runs constantly and executes every possible maintenance script--whether or not your configuration requires it--circ-exec offers more nuanced control of your Library Simplified Circulation Manager jobs. The most accurate place to look for recommended jobs and their recommended frequencies is the existing circ-scripts crontab.

Because containers based on circ-exec are built, run their job, and are destroyed, it's important to configure an external log aggregator if you wish to capture logs from the job.

# See the section "Environment Variables" below for more information
# about the values listed here and their alternatives.
$ docker run --name search_index_refresh -it \
    -e SIMPLIFIED_SCRIPT_NAME='refresh_materialized_views' \
    -e SIMPLIFIED_PRODUCTION_DATABASE='postgresql://[username]:[password]@[host]:[port]/[database_name]' \
    ghcr.io/natlibfi/ekirjasto-circ-exec:main

Environment Variables

Environment variables can be set with the -e VARIABLE_KEY='variable_value' option on the docker run command. SIMPLIFIED_PRODUCTION_DATABASE is the only required environment variable.

SIMPLIFIED_PRODUCTION_DATABASE

Required. The URL of the production PostgreSQL database for the application.

SIMPLIFIED_TEST_DATABASE

Optional. The URL of a PostgreSQL database for tests. This optional variable allows unit tests to be run in the container.

TZ

Optional. Applies to circ-scripts only. The time zone that cron should use to run scheduled scripts--usually the time zone of the library or libraries on the circulation manager instance. This value should be selected according to Debian-system time zone options. This value allows scripts to be run at ideal times.

UWSGI_PROCESSES

Optional. The number of processes to use when running uWSGI. This value can be updated in docker-compose.yml or added directly in Dockerfile under webapp stage. Defaults to 6.

UWSGI_THREADS

Optional. The number of threads to use when running uWSGI. This value can be updated in docker-compose.yml or added directly in Dockerfile under webapp stage. Defaults to 2.

Building new images

If you plan to work with stable versions of the Circulation Manager, we strongly recommend using the latest stable versions of circ-webapp and circ-scripts published to the GitHub Container Registry. However, there may come a time in development when you want to build Docker containers for a particular version of the Circulation Manager. If so, please use the instructions below.

We recommend you install at least version 18.06 of the Docker engine.

webapp and scripts images

Determine which image you would like to build and update the tag and Dockerfile listed below accordingly. Run the build command from the root of the repository not the docker folder. Use target option to determine which docker image to build as bellow:

docker build --tag circ --file docker/Dockerfile --target scripts .

See docker/Dockerfile for details.

Feel free to change the image tag as you like.