Skip to main content
?

System Variable

System variables are configurable in 3 ways.

  1. In the configuration file (by default this is the autoflow.conf file in the data folder within the install location)
  2. On the settings page within the products UI (this writes the values into the configuration file)
  3. As shell variables with the prefixΒ AUTOFLOW_Β (for instance AUTOFLOW_CONFIG_FILE)

The CONFIG_FILE variable is unique in that it is only usable as a shell variable since the configuration file itself defines the other variables. Shell variables override the variables in the configuration file.

ADMIN_PASSWORD​

Used to set the initial password and bypass the first user page. Requires ADMIN_USERNAME to be set as well.

ADMIN_USERNAME​

Used to set an initial username and bypass the first user page. Requires ADMIN_PASSWORD to be set as well.

BOOT_FILE​

This file is used to automatically load solutions during bootup. This variable is predominantly used in docker deployments. The file itself can be created by downloading a config from an existing installation. This value will be ignored if configuration already exists but can be forced to be used with BOOT_FILE_REPLACES_CONFIG.

BOOT_FILE_REPLACES_CONFIG​

Determines whether the BOOT_FILE should replace existing config. The value can be 'true' or 'false'.

CLUSTER_HOSTS​

A comma separated list of hosts to include in the cluster. Only used if CLUSTER_MODE is 'hosts'.

CLUSTER_KUBERNETES_POLLING_INTERVAL​

The polling interval to maintain the cluster in kubernetes. Only used if CLUSTER_MODE is 'kubernetes'.

CLUSTER_KUBERNETES_SERVICE​

The name of the kubernetes headless service. Only used if CLUSTER_MODE is 'kubernetes'.

CLUSTER_MODE​

This determines the clustering strategy. Values can be 'kubernetes', 'hosts', or 'dynamic'. Only used if DEPLOYMENT_MODE is 'cluster'.

CONFIG_FILE​

The customizable environment variables for the product. Since this value is an environment variable itself, it can only be used as a command line argument; otherwise, the default CONFIG_FILE will be used within the DATA_DIRECTORY.

The file is structured as KEY=VALUE. For example:

NAME=My Application
UI_PORT=8080

DATA_DIRECTORY​

The default location for product specific files. For instance, the product CONFIG_FILE and LICENSE_FILE are product specific files. Default: data

DEPLOYMENT_MODE​

The deployment mode to use. Values can be 'solo' and 'cluster'.

DIRECTORY​

The directory for allΒ relativeΒ file system access.

LICENSE_ADDRESS​

The server address to use for license validation.

LICENSE_FILE​

The file to use for license validation. Should contain license information provided by the license server.

LICENSE_PORT​

The server port to use for license validation.

LOG_LEVEL​

The log level to use during operation. Values can be 'debug', 'info', 'warn', and 'error'.

LOG_MODE​

The log mode to use during operation. Values can be 'json' and 'plain'. The 'json' option is useful for integration with logging tools like fluentd/logstash.

NAME​

The name to use for this installation. The name will be listed in the UI browser title.

PORT_OFFSET​

This is an offset to apply to all ports configured in the solution. This can be useful if you want to deploy multiple instances of the same soluton on the same system but use different ports.

UI_PORT​

The port to serve the UI. Default: 4000