arijeetmkh / hephaestus Goto Github PK

There needs to be a way to ensure MessageWorker threads do not block indefinitely. Possibly some timeout?

Simple HTML transport with customizable options for request type, headers, etc

When connecting with existing applications, their own logging subsystem is loaded automatically. This can conflict with the existing hephaestus logging or the client may wish to log in a different way, in which case they would need to setup custom logging configurations.

This should be doable through logging setup hooks which clients can intercept and inject their own logging configurations

Diagnostics need to be logged under hephaestus.diagnostics such as Total SQS threads, message worker threads, queue size etc to get an idea of current application state

Celery transport would be helpful in directly submitting serialisable tasks to the broker for celery workers to consume.

SIGHUP and SIGINT need to be handled and a mechanism of safe shutdown needs to be introduced

Tracked via PR: #1