https://www.prisma.io/ logo
Join Slack
Powered by
What do you do when you get the infamous `internal...
# orm-help
t
What do you do when you get the infamous 
internal server error
on a non-local cluster that you cannot reproduce locally in a docker cluster? We still don’t have access to`prisma cluster logs` for non-local clusters so I’m curious what others have found helps them diagnose? 
Error: GraphQL Error: Whoops. Looks like an internal server error. Search your cluster logs for request ID: api:api:blahblahblah
n
so we discoverd it to be a memory issue
with the help of Nilan
we did a lot of optimization to queries and mutations
and bumped up box size
t
@noahdavis interesting, thank you, do you know how much memory prisma hosted clusters have by default?
I’ll try and repro with that locally
n
Ours had 4GB
t2.medium AWS Aurora MySQL
t
cool - do you know if the local docker containers use the same limit or is no limit specified? I haven’t looked yet.
n
not sure
t
I can figure that out myself, sorry, dumb question
thanks for your help, really appreciate it
n
no problem man
let me know what you find
we just finished our optimization and deployed to prod
so I will update you if we have success
okay @Troy Sandal - it has been sucessful
we just limited the data called in our queries and the number of items we were fetching at once
n
Please reach out to support@prisma.io if you need help with internal server errors on Prisma Cloud clusters for now. You can read more here https://github.com/graphcool/prisma-cloud-feedback/issues/94
3 Views
Open in Slack
PreviousNext
Powered by