how to make max tries to 0 in serverless handler or endpoint UI
my requests are processing twice even when the first response is correct
15 Replies
And not sure why execution time is negative
@yhlong00000 can you please look into it once
It looks like this request failed and was retried by the system. I assume you don’t want the system to retry any of your jobs? I’m not sure if we have the option to disable this system-level retry. I’ll need to check into that.
the execution time appears as a negative value, which is unusual and something I haven’t seen before. I’ll need to investigate further to figure out why this happened.
The request hasn't failed, I got the output
Please resolve this issue as soon as possible 🙏
Interested in same issue with retry
Hey @yhlong00000
Every request is getting retried, costing us double
We need to move to production asap
Can you atleast provide a temporary fix or something until you resolve it permanently
Waiting for your response..
replied the ticket you created.
Hi, I basically have the same problem with the serverless. Where can i see the reply to the ticket? As i want also to solve it
Ohh, tickets are private communication between staffs and one user only.. but they can share the information here if they want to
hey, can you open a ticket with your rundpod sdk version, endpoint id, logs and when it happens, how frequently and etc.
https://contact.runpod.io/hc/en-us/requests/new
My serverless retry any request, when does it fix?
Is this behavior start recently for you?
me too. all jobs run twice and sometimes I got negative executionTime. (I don't know its necessary but my container image:
nvidia/cuda:11.7.1-cudnn8-runtime-ubuntu20.04)on sdk 1.7.0-1.7.2 has this issue, try to upgrade to 1.7.3 and see if that helps.
I don’t use any cli, Is it something in dashboard?
when you build your worker handler file, you
import runpod that is SDK