You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
SAQ has a UI which is an aiohttp Application object. Unfortunately these are not asgi apps: aio-libs/aiohttp#2902
There is https://github.com/mosquito/aiohttp-asgi, or might need an alternate implementation of the UI that could be delivered as a starlite application config plugin.
The text was updated successfully, but these errors were encountered:
I´m sorry if I´m talking non-sense as I´m a beginner. But maybe wouldn´t it be better to separate the worker logic from starlite application?
I followed this tutorial https://github.com/fcakyon/flask-redis-docker and found it very easy to implement changes. If your worker is doing ML, sometimes you need to specify a GPU to it.
Also, creating a new worker was very easy. All I had to do was add a new container with same image and change command with a new name.
Hey @albcunha - if your workers are doing CPU bound work it definitely would make sense to have your worker in a separate process. If the worker does mainly IO bound stuff, then this pattern should be just fine.
Goldziher
changed the title
Feat: mount the SAQ UI
Enhancement: mount the SAQ UI
Jul 4, 2023
SAQ has a UI which is an aiohttp
Application
object. Unfortunately these are not asgi apps: aio-libs/aiohttp#2902There is https://github.com/mosquito/aiohttp-asgi, or might need an alternate implementation of the UI that could be delivered as a starlite application config plugin.
The text was updated successfully, but these errors were encountered: