worker exited with exit code 0

[pliant slate]

Hello team, I'm trying to host my remotion video rendering on Runpod serverless built with nodejs via docker.

the build completes but when I shoot a request, it never moves out of job queue, worker starts and gives error worker exited with exit code 0 and never shuts down and the video didn't get's rendered

every time I've to terminate the worker and purge the queue.

What is going wrong?

worker id - j9spmt6ba47py6
endpoint - https://api.runpod.ai/v2/1bqxtx4casdrc9/run

[gloomy sand]

can you check your dockerfifle

you must execute your handler.py, and call serverless start

if you did that, check for errors in your code, add debugging logs to check for errors in the process

[pliant slate]

Okay @gloomy sand, thank you for your response. adding and checking it

[pliant slate]

Hey @gloomy sand , I did this, still the same error. It just gives error exited with code 0, nothing else

[gloomy sand]

Try using. Working example from runpod worker github instead

Then change your way from there

If something goes wrong reverse back to the working one

[robust stump]

why exit code 0 tho

maybe your docker CMD doesn't wait forever

[gloomy sand]

Yeah many reasons, just means the program is finished Google says

[pliant slate]

understood, will try and revert back here

I'm still not able to solve this, can anyone help me with this as a paid gig?

[gloomy sand]

Send your files here dockerfile, handler

[pliant slate]

here it is - Thanks @gloomy sand 🙌🏻

[gloomy sand]

well you cant really use .js file as a handler actually

usually people use .py, as far as i know thats what's supported

but im sure easily with LLM's now they can guide you with migrating these two files fast!

just send the runpod docs for creating handlers / find a reference like this
https://github.com/blib-la/runpod-worker-comfy/tree/main

[pliant slate]

Thank you for sharing this! checking it out.
But just want to know that initially this code was made for the server and it worked if I manually started the worker and the same code also worked initially for a few renders (videos were rendering but the worker didn't terminate) but stopped working after that, bit weird

[gloomy sand]

"manually started the worker" how?

[gloomy sand]

yep, this is why because if there's no runpod-specific library, it cannot stop the worker automatically then.

[pliant slate]

By triggering via a sample request via the runpod dashboard, it started by an internal server. It communicates with Supabase realtime to render videos

[pliant slate]

got it

[robust stump]

really?

[gloomy sand]

i think so, unless you copy the whole python library into correctly supporting js

isnt t?

[gloomy sand]

i see yep

[pliant slate]

I'm using Remotion for video rendering. Remotion is built specifically for the Node.js. so in this case I don't think Runpod will be the right platform to host if .js handler is not supported?

[gloomy sand]

There is another way actually

Like from python you can run any programs

.sh..js any that runs in linux

By using subprocess or something else, so you may try that

And just pass the input as program arguments

[robust stump]

yeah

[pliant slate]

okay, let me check if we can do this. thanks 🙌

[robust stump]

just make a rest-api ish thing with node andthen

invoke it with python

[gloomy sand]

Yeah or that

[pliant slate]

Understood

[gloomy sand]

You can see an example in this repo

How they use requests library to communicate with other program's http api