Settings

Below are all settings for django-pgclone.

Note

Most settings relate to command options and are global defaults for them. Check out the basics section for more information on how default values for commands are determined.

PGCLONE_ALLOW_RESTORE

If False, running any restore results in an error.

PGCLONE_ALLOW_DUMP

If False, running any dump results in an error.

PGCLONE_ALLOW_COPY

If False, running any copy results in an error.

Default True

PGCLONE_CONFIGS

Configurations that store options for the commands. For example:

PGCLONE_CONFIGS = {
    "no_users": {
        "exclude": ["auth.User"]
    }
}

See the configurations section for more information on what options can be configured.

Default {}

PGCLONE_CONN_DB

The connection database used when running psql commands. A connection database that is different from the dumped or restored database must be used for special commands like killing connections before swapping restores.

Default postgres if there are no postgres databases in settings.DATABASES, otherwise template1 is used.

PGCLONE_DATABASE

The default database to use for commands.

Default default

PGCLONE_EXCLUDE

The tables to exclude for dumps.

Default []

PGCLONE_INSTANCE

The instance name to use in the dump key. For example, using "prod" as the instance when running production dumps.

Default Uses socket.gethostname() to generate the instance.

PGCLONE_PRE_DUMP_HOOKS

The hooks to run by default for dumps.

Default []

PGCLONE_PRE_SWAP_HOOKS

The hooks to run by default for restores before swapping happens.

Default ["migrate"]

PGCLONE_REVERSIBLE

True if the restore command should create reversible restores by default. See the reversible restores section for more information.

Default False

PGCLONE_S3_CONFIG

The environment variable overrides when using the AWS CLI. Only applicable when using the S3 storage backend.

For example:

PGCLONE_S3_CONFIG = {
    "AWS_ACCESS_KEY_ID": "acces_key",
    "AWS_SECRET_ACCESS_KEY": "secret_access_key_read_from_environment",
    "AWS_DEFAULT_REGION": "us-east-1"
}

Default: {}

PGCLONE_S3_ENDPOINT_URL

The S3 endpoint url to send requests to if using a non-standard AWS endpoint or an S3 service other than AWS (such as DigitalOcean Spaces or self-hosting an endpoint directly within your private VPC). Only applicable when using the S3 storage backend.

For example:

PGCLONE_S3_ENDPOINT_URL = "https://endpoint.example.com"

If using DigitalOcean Spaces, be sure to set PGCLONE_S3_ENDPOINT_URL and the storage location appropriately. For example, if the name of your Space is "my-backup-bucket" in the DigitalOcean SFO2 region, and the resulting endpoint is "https://my-backup-bucket.sfo2.digitaloceanspaces.com", then the following example settings would work (note we removed the Space's name from the endpoint url):

PGCLONE_S3_ENDPOINT_URL = "https://sfo2.digitaloceanspaces.com"
PGCLONE_STORAGE_LOCATION = "s3://my-backup-bucket/"

Default: None

PGCLONE_STORAGE_LOCATION

Where dumps are stored. Use relative paths to store in the local file system. Use paths that begin with s3:// to use an S3 storage backend. See storage backends.

For example:

PGCLONE_STORAGE_LOCATION = "s3://bucket/prefix/"

Default .pgclone/

PGCLONE_VALIDATE_DUMP_KEYS

False if invalid dump keys should be returned by python manage.py pgclone ls. This helps preserve backwards compatibility with version 1. Note that --instances, --databases, and --configs arguments will ignore invalid keys.

Default True

PGCLONE_STATEMENT_TIMEOUT

The statement_timeout Postgres setting to use when running core pgclone commands via psql. For example, the statement timeout will apply to any CREATE DATABASE, ALTER DATABASE, or DROP DATABASE statements, along with statements that terminate blocking queries.

Default None

PGCLONE_LOCK_TIMEOUT

The lock_timeout Postgres setting to use when running core pgclone commands via psql. Similar to PGCLONE_STATEMENT_TIMEOUT.

Default None