Fetch failed when using serverless driver running many function instances
I’m receiving ‘fetch failed’ for some of my functions using drizzle in a GCP cloud function.
We’re running a batch process that runs up to about 300 instances and some of them end up with ‘fetch failed’ error.
I’m using drizzle with the pooled connection, connecting exactly how the documentation says. I don’t understand why we are getting this error. We’re on the pro plan and I’ve enabled autoscaling up to 1 cpu which, according to the docs should allow me to have more than enough connections.
I would really appreciate some help on how to debug this or just a hint on why could this be happening.
3 Replies
optimistic-gold•2y ago
@eriklissen if you're on the pro plan, you should file a ticket with support through the console.
That is how you can get priority over free tier users
fascinating-indigoOP•2y ago
Will do, thanks @Tristan Partin
optimistic-gold•2y ago
Discord support is really only intended for free tier users. We want to give pro tier users 1:1 time with our support team since that is part of what you pay us for 🙂