Skip to content

Latest commit

 

History

History
59 lines (39 loc) · 1.56 KB

README.md

File metadata and controls

59 lines (39 loc) · 1.56 KB
Raw

ustack-logging

Default logging configuration for uStack style Python applications.

Installation

pip install ustack-logging

Usage

from ustack_logging.logging_configuration import configure_logging


def main():
    configure_logging()

    # Run your application


if __name__ == "__main__":
    main()

Intent

This library is intended to codify the standard way uStudio Python applications log and report errors when running inside Kubernetes. It takes no configuration because its intent is to enforce a configuration.

However, it is built on datadog-logger, which is configurable. If you need a different configuration, it should be trivial to implement it using that library directly.

Configuration

The library configures the following things:

It sets up the logging library to log the time, log level, module name and log message for every message, and it sets the current log level to INFO.

In order to function, it uses a secret named environment-info in the ustudio-system namespace. The secret must contain the keys environment and datadog-api-key, and the pod must contain a label named role.

If any information is missing or any connections fail it logs a warning and does not error.

It configures the datadog-logger library to send ERROR and above log messages to Datadog as events, with the following tag mapping:

  • environment-info.environment -> Datadog tag environment
  • K8s Pod Namespace -> Datadog tag service
  • K8s Pod Label role -> Datadog tag role