ollyswanson / ruok Goto Github PK

The architecture is very simple, but explicitly documenting how the two actors interact would be useful.

Add actions for testing and linting, and monitoring versions of dependencies.

Update Service to include optional authentication headers

Adding tracing with tokio-rs/tracing would be useful for debugging, and generating useful data about uptimes.

Create aditional notification types and messages appropriate to the given platform.

Startup

The errors produced on startup (opening and parsing config file) can be improved to be more user friendly. Instead of using Boxed error, I would live to investigate more robust error handling that allows better granularity with the error types.

Running

Most of the unwraps in checker and notifier are fine, but a lot of the results from requests are thrown away, it might be useful to tie error handling here to logging so that it's easier to see when and why a request (or other fallible action) has failed.

Slack

The documentation for formatting Slack messages sent via webhook can be found here, the quality of the current notification format can be drastically improved.

Use an HTTP mocking library such as mockito to test the behaviour of each actor. Due to the time based nature of events, how to test tokio based applications also needs to be investigated.