Environment configuration reference

Open Object can be run as a Docker container or a VPS / dedicated server. It relies on other services, such as database and cache backends, which can be configured through environment variables.

Available environment variables

Required

SECRET_KEY: Secret key that’s used for certain cryptographic utilities. .
ALLOWED_HOSTS: a comma separated (without spaces!) list of domains that serve the installation. Used to protect against Host header attacks. Defaults to: (empty string).
CACHE_DEFAULT: redis cache address for the default cache (this MUST be set when using Docker). Defaults to: localhost:6379/0.
CACHE_AXES: redis cache address for the brute force login protection cache (this MUST be set when using Docker). Defaults to: localhost:6379/0.
EMAIL_HOST: hostname for the outgoing e-mail server (this MUST be set when using Docker). Defaults to: localhost.
SITE_DOMAIN: Defines the primary domain where the application is hosted.

Database

DB_NAME: name of the PostgreSQL database. Defaults to: objects.
DB_USER: username of the database user. Defaults to: objects.
DB_PASSWORD: password of the database user. Defaults to: objects.
DB_HOST: hostname of the PostgreSQL database. Defaults to db for the docker environment, otherwise defaults to localhost.
DB_PORT: port number of the database. Defaults to: 5432.
DB_CONN_MAX_AGE: The lifetime of a database connection, as an integer of seconds. Use 0 to close database connections at the end of each request — Django’s historical behavior. This setting is ignored if connection pooling is used. Defaults to: 60.
DB_POOL_ENABLED: Experimental: Whether to use connection pooling. This feature is not yet recommended for production use. See the documentation for details: https://open-api-framework.readthedocs.io/en/latest/connection_pooling.html. Defaults to: False.
DB_POOL_MIN_SIZE: The minimum number of connection the pool will hold. The pool will actively try to create new connections if some are lost (closed, broken) and will try to never go below min_size. Defaults to: 4.
DB_POOL_MAX_SIZE: The maximum number of connections the pool will hold. If None, or equal to min_size, the pool will not grow or shrink. If larger than min_size, the pool can grow if more than min_size connections are requested at the same time and will shrink back after the extra connections have been unused for more than max_idle seconds. Defaults to: None.
DB_POOL_TIMEOUT: The default maximum time in seconds that a client can wait to receive a connection from the pool (using connection() or getconn()). Note that these methods allow to override the timeout default. Defaults to: 30.
DB_POOL_MAX_WAITING: Maximum number of requests that can be queued to the pool, after which new requests will fail, raising TooManyRequests. 0 means no queue limit. Defaults to: 0.
DB_POOL_MAX_LIFETIME: The maximum lifetime of a connection in the pool, in seconds. Connections used for longer get closed and replaced by a new one. The amount is reduced by a random 10% to avoid mass eviction. Defaults to: 3600.
DB_POOL_MAX_IDLE: Maximum time, in seconds, that a connection can stay unused in the pool before being closed, and the pool shrunk. This only happens to connections more than min_size, if max_size allowed the pool to grow. Defaults to: 600.
DB_POOL_RECONNECT_TIMEOUT: Maximum time, in seconds, the pool will try to create a connection. If a connection attempt fails, the pool will try to reconnect a few times, using an exponential backoff and some random factor to avoid mass attempts. If repeated attempts fail, after reconnect_timeout second the connection attempt is aborted and the reconnect_failed() callback invoked. Defaults to: 300.
DB_POOL_NUM_WORKERS: Number of background worker threads used to maintain the pool state. Background workers are used for example to create new connections and to clean up connections when they are returned to the pool. Defaults to: 3.
DB_DISABLE_SERVER_SIDE_CURSORS: Whether or not server side cursors should be disabled for Postgres connections. Setting this to true is required when using a connection pooler in transaction mode (like PgBouncer). WARNING: the effect of disabling server side cursors on performance has not been thoroughly tested yet. Defaults to: False.

Logging

LOG_STDOUT: whether to log to stdout or not. Defaults to: True.
LOG_LEVEL: control the verbosity of logging output. Available values are CRITICAL, ERROR, WARNING, INFO and DEBUG. Defaults to: INFO.
LOG_QUERIES: enable (query) logging at the database backend level. Note that you must also set DEBUG=1, which should be done very sparingly!. Defaults to: False.
LOG_REQUESTS: enable structured logging of requests. Defaults to: True.
LOG_FORMAT_CONSOLE: The format for the console logging handler, possible options: json, plain_console. Defaults to: json.
LOG_OUTGOING_REQUESTS_EMIT_BODY: Whether or not outgoing request bodies should be logged. Defaults to: True.
LOG_OUTGOING_REQUESTS_DB_SAVE: Whether or not outgoing request logs should be saved to the database. Defaults to: False.
LOG_OUTGOING_REQUESTS_DB_SAVE_BODY: Whether or not outgoing request bodies should be saved to the database. Defaults to: True.
LOG_OUTGOING_REQUESTS_MAX_AGE: The amount of time after which request logs should be deleted from the database. Defaults to: 7.
LOG_OUTGOING_REQUESTS: enable logging of the outgoing requests. This must be enabled along with LOG_OUTGOING_REQUESTS_DB_SAVE to save outgoing request logs in the database. Defaults to: False.

Celery

CELERY_LOGLEVEL: control the verbosity of logging output for celery, independent of LOG_LEVEL. Available values are CRITICAL, ERROR, WARNING, INFO and DEBUG. Defaults to: INFO.
CELERY_RESULT_BACKEND: the URL of the backend/broker that will be used by Celery to send the notifications. Defaults to: redis://localhost:6379/1.
CELERY_BROKER_URL: the URL of the broker that will be used by Celery to send the notifications. Defaults to: redis://localhost:6379/1.
CELERY_RESULT_EXPIRES: How long the results of tasks will be stored in Redis (in seconds), this can be set to a lower duration to lower memory usage for Redis. Defaults to: 3600.
CELERY_TASK_HARD_TIME_LIMIT: Task hard time limit in seconds. The worker processing the task will be killed and replaced with a new one when this is exceeded. Defaults to: 900.

Elastic APM

ELASTIC_APM_SERVER_URL: URL where Elastic APM is hosted. Defaults to: None.
ELASTIC_APM_SERVICE_NAME: Name of the service for this application in Elastic APM. Defaults to objects - <environment>.
ELASTIC_APM_SECRET_TOKEN: Token used to communicate with Elastic APM. Defaults to: default.
ELASTIC_APM_TRANSACTION_SAMPLE_RATE: By default, the agent will sample every transaction (e.g. request to your service). To reduce overhead and storage requirements, set the sample rate to a value between 0.0 and 1.0. Defaults to: 0.1.

Content Security Policy

CSP_EXTRA_DEFAULT_SRC: Extra default source URLs for CSP other than self. Used for img-src, style-src and script-src. Defaults to: [].
CSP_EXTRA_FORM_ACTION: Additional form-action sources. Defaults to: [].
CSP_FORM_ACTION: Override the default form-action sources. Defaults to: ['"\'self\'"'].
CSP_EXTRA_IMG_SRC: Extra img-src sources. Defaults to: [].
CSP_OBJECT_SRC: object-src sources. Defaults to: ['"\'none\'"'].
CSP_REPORT_URI: URI for CSP report-uri directive. Defaults to: None.
CSP_REPORT_PERCENTAGE: Fraction (between 0 and 1) of requests to include report-uri directive. Defaults to: 0.0.

Optional

SITE_ID: The database ID of the site object. You usually won’t have to touch this. Defaults to: 1.
DEBUG: Only set this to True on a local development environment. Various other security settings are derived from this setting!. Defaults to: False.
USE_X_FORWARDED_HOST: whether to grab the domain/host from the X-Forwarded-Host header or not. This header is typically set by reverse proxies (such as nginx, traefik, Apache…). Note: this is a header that can be spoofed and you need to ensure you control it before enabling this. Defaults to: False.
IS_HTTPS: Used to construct absolute URLs and controls a variety of security settings. Defaults to the inverse of DEBUG.
EMAIL_PORT: port number of the outgoing e-mail server. Note that if you’re on Google Cloud, sending e-mail via port 25 is completely blocked and you should use 487 for TLS. Defaults to: 25.
EMAIL_HOST_USER: username to connect to the mail server. Defaults to: (empty string).
EMAIL_HOST_PASSWORD: password to connect to the mail server. Defaults to: (empty string).
EMAIL_USE_TLS: whether to use TLS or not to connect to the mail server. Should be True if you’re changing the EMAIL_PORT to 487. Defaults to: False.
DEFAULT_FROM_EMAIL: The default email address from which emails are sent. Defaults to: objects@example.com.
SESSION_COOKIE_AGE: For how long, in seconds, the session cookie will be valid. Defaults to: 1209600.
SESSION_COOKIE_SAMESITE: The value of the SameSite flag on the session cookie. This flag prevents the cookie from being sent in cross-site requests thus preventing CSRF attacks and making some methods of stealing session cookie impossible.Currently interferes with OIDC. Keep the value set at Lax if used. Defaults to: Lax.
CSRF_COOKIE_SAMESITE: The value of the SameSite flag on the CSRF cookie. This flag prevents the cookie from being sent in cross-site requests. Defaults to: Strict.
ENVIRONMENT: An identifier for the environment, displayed in the admin depending on the settings module used and included in the error monitoring (see SENTRY_DSN). The default is set according to DJANGO_SETTINGS_MODULE.
SUBPATH: If hosted on a subpath, provide the value here. If you provide /gateway, the component assumes its running at the base URL: https://somedomain/gateway/. Defaults to an empty string. Defaults to: None.
RELEASE: The version number or commit hash of the application (this is also sent to Sentry).
NUM_PROXIES: the number of reverse proxies in front of the application, as an integer. This is used to determine the actual client IP adres. On Kubernetes with an ingress you typically want to set this to 2. Defaults to: 1.
CSRF_TRUSTED_ORIGINS: A list of trusted origins for unsafe requests (e.g. POST). Defaults to: [].
NOTIFICATIONS_DISABLED: indicates whether or not notifications should be sent to the Notificaties API for operations on the API endpoints. Defaults to True for the dev environment, otherwise defaults to False.
SENTRY_DSN: URL of the sentry project to send error reports to. Default empty, i.e. -> no monitoring set up. Highly recommended to configure this.
DISABLE_2FA: Whether or not two factor authentication should be disabled. Defaults to: False.
OBJECTS_ADMIN_SEARCH_DISABLED: Indicates whether or not searching in the Objects admin should be disabled. Defaults to: False.
ENABLE_CLOUD_EVENTS: EXPERIMENTAL: indicates whether or not cloud events should be sent to the configured endpoint for specific operations on Zaak (not ready for use in production). Defaults to: False.
NOTIFICATIONS_SOURCE: EXPERIMENTAL: the identifier of this application to use as the source in notifications and cloudevents. Defaults to: (empty string).

Initial superuser creation (Docker only)

A clean installation of Open Object comes without pre-installed or pre-configured admin user by default.

Users can opt-in to provision an initial superuser via environment variables. The user will only be created if it doesn’t exist yet.

OBJECTS_SUPERUSER_USERNAME: specify the username of the superuser to create. Setting this to a non-empty value will enable the creation of the superuser. Default empty.
OBJECTS_SUPERUSER_EMAIL: specify the e-mail address to configure for the superuser. Defaults to admin@admin.org. Only has an effect if OBJECTS_SUPERUSER_USERNAME is set.
OBJECTS_SUPERUSER_PASSWORD: specify the password for the superuser. Default empty, which means the superuser will be created _without_ password. Only has an effect if OBJECTS_SUPERUSER_USERNAME is set.

Initial configuration

Open Object supports the setup_configuration management command, which allows configuration via environment variables. All these environment variables are described at command line.

Specifying the environment variables

There are two strategies to specify the environment variables:

provide them in a .env file
start the component processes (with uwsgi/gunicorn/celery) in a process manager that defines the environment variables

Providing a .env file

This is the most simple setup and easiest to debug. The .env file must be at the root of the project - i.e. on the same level as the src directory ( NOT in the src directory).

The syntax is key-value:

SOME_VAR=some_value
OTHER_VAR="quoted_value"

Provide the envvars via the process manager

If you use a process manager (such as supervisor/systemd), use their techniques to define the envvars. The component will pick them up out of the box.