We’re currently using a dedicated instance in Dgraph Cloud. When making requests from the client, we constantly get this error message while querying and mutating. We are currently beta testing our product and hence we do not have too many users.
Just from testing it with one client, we are being frequently hit with this error. Is there a way we can resolve this?
Thanks in advance.
Hi,
Sorry that you are running into the 429 Error with your dedicated backend. Could you please share the ID or the email associated with the backend so that we can investigate the issue.
Thanks
-Megha
Hi. Thank you for reaching out.
Here’s the email that we used to create the dedicated backend instance: tech@livnlife.io.
Hi,
Dgraph Cloud backends have max pending query limits set to prevent the backend from getting overloaded. This value is set to 16 for 1M backend which we believe is the right amount of concurrency a 1M backend can handle without running into resource limits. I have just updated it to 32 allowing for more concurrency during your Beta testing.
In production, you should pick a higher tier depending on the load you anticipate.
Please let me know if this resolves the issue.
Thanks
Hi. We do not see any improvement. 2/3 of the requests return a 429.
Hi,
We do not see any 429 Errors in our logs for your backends and we see that there were hundreds(776 total)
Are you doing retries in client side? It is best approach from the client-side is to enable a retry mechanism if a 429 is received from the server.Would you be able to email us (megha@hypermode.com) your client code for us to take a look?
Thanks
-Megha
That’s correct. We added retry mechanism in the client side along with caching. That seems to have fixed the problem. Thank you for your support
Thanks for confirming that the issue is now resolved.
