@Luc1d20@Kinsi55 Ein Kollege von mir hatte dieses Problem auch. Im Taskmanager gab es einen "nvidia-container" Prozess, welcher Scans durchfรผhrt. Da kann man irgendwo einen Haken rausmachen, dass dieser Scan nicht mehr lรคuft. Damit waren die Probleme bei ihm weg. Ich frag mal nach.
@levelsio@postmarkapp For future problems one could implement a message queue instead of fire & forget with backchecks for delivers. Not sure if postmarkapp provides a check if the mail has been sent but other services offer checks on the current delivery status.
@anthonynsimon@nurodev Secret sauce here is that they are using hardware from I think it was 2017 with around half the performance the current AMD plan is offering cpu-wise.
So if you don't need that much CPU and rely more on memory - go for it.
@pscheit The solution we build is a connector between Stripe and sevDesk to automate this whole invoicing and tax process for businesses in Germany, too.
We are currently serving a few customers. It's basically plug and play if you have some requirements like Strip + sevDesk.
@arvidkahl I can only recommend using sth. like Healthchecks and Tunnel for automated failovers. That's where Cloudflare is not "providing" help. If your origin is offline but still in the DNS entries, it will timeout (after rr) within 30s.
@_lhermann@SocketIO Hey Lukas,
I am working with a notifier service that injects a message into the message queue instead of letting the server itself send the message.
@amosbastian Damn. I checked the docs. You could take a look at this: https://t.co/O1AEGRcMyu
If I read it correctly this should work for you if you use a โsmallโ api which initiates a transform. Combined with Cloudflare caching rules, this should fit your usecase.