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
First of, I LOVE your library and have been able to perfectly implement it in my project. I have been planning on scaling up though and am having some thoughts on signals originating from multiple cpus/dynos or perhaps even firing them from a scheduler. Is there a way to have signals trigger subscriptions as per the aforementioned requirements? I was thinking of firing them across channels, but it seems rather complicated and I am still fairly new to python.
What are your thoughts? Help is greatly appreciated. Thank you in advance!
The text was updated successfully, but these errors were encountered:
graphene-subscriptions actually handles sending events to multiple app instances using Channel Layers automatically. If you use the built-in post_save_subscription and post_delete_subscription signals, an event will be sent via Channel Layers so that all instances with active Subscription connections receive it and send the data to the client. If you're using Custom Events this is also handled automatically.
That's good to know. I was running a Heroku Scheduler to run some updating tasks and the signals weren't triggering graphene-subscriptions, so I assumed that they weren't being sent over channels, but this may be a separate issue.
First of, I LOVE your library and have been able to perfectly implement it in my project. I have been planning on scaling up though and am having some thoughts on signals originating from multiple cpus/dynos or perhaps even firing them from a scheduler. Is there a way to have signals trigger subscriptions as per the aforementioned requirements? I was thinking of firing them across channels, but it seems rather complicated and I am still fairly new to python.
What are your thoughts? Help is greatly appreciated. Thank you in advance!
The text was updated successfully, but these errors were encountered: