AppSignal for Python load order
This documentation outlines how the AppSignal configuration is loaded in your Python application.
There are two different methods for configuring your AppSignal integration, which can be used simultaneously:
- Config file (
__appsignal__.py
) - Environment variables
A list of available options can be found in the Configuration options documentation.
Load order
The configuration is loaded in a four-step process:
- Package defaults -
default
- System detected settings -
system
- Environment variables -
env
- Initial configuration -
initial
1. Package defaults
First, the default configuration is loaded, which sets paths and enables certain features.
This source will be listed under default
in your application's config and is visible in the Diagnose report.
2. System detected settings
The package detects what kind of system it's running on and configures itself accordingly.
For example, if running inside a container-based system such as Docker or Heroku, the configuration option running_in_container
is set to true
.
This source will be listed under system
in your application's config and is visible in the Diagnose report.
3. Environment variables
AppSignal will look for Configuration options set via Environment variables. The Environment variable value will override values for the same Configuration options defined in previous steps.
This source will be listed under env
in your application's configuration, and is visible in the Diagnose report.
4. Initial configuration given to Appsignal
Finally, the initial configuration is loaded via the Appsignal
function called from the __appsignal__.py
config file:
The values of Conig options defined in this step will override values for the same Configuration options defined in previous steps.
This source will be listed under initial
in your application's config and is visible in the Diagnose report.