Creating a pod using a Docker image built via Serverless GitHub integration

I'd like to create a pod on Runpod using the docker image I built via GitHub integration for the Serverless endpoint. I can't build it locally and push it to my registry because then the container doesn't work properly. It only works when it is built on the Runpod platform via the GitHub integration. In the build logs I can see it is published in a registry but I can't see its address. I assume it's a Runpod's registry so it'd probably have access to it during pod creation?
10 Replies
Madiator2011
Madiator20116mo ago
registry is limited to serverless only wont work with pods
fishermann
fishermannOP6mo ago
Ok I guess I'll try building my container on the pod
Unknown User
Unknown User6mo ago
Message Not Public
Sign In & Join Server To View
fishermann
fishermannOP6mo ago
I basically need a machine with CUDA to build it properly, I'll look into GHA
Unknown User
Unknown User6mo ago
Message Not Public
Sign In & Join Server To View
riverfog7
riverfog76mo ago
If you dont need cuda compute You can do it with no gpus Nvcc does not require you to have gpus
fishermann
fishermannOP5mo ago
I install a model in the container and when I build it on my mac it doesn't work on RunPod. Like it's built for a cpu device or something. Perhaps there are configs for that but I didn't wanna go deep there. I ended up getting a pod and basically running all the Dockerfile commands there to set up the environment and it worked out well for me
Unknown User
Unknown User5mo ago
Message Not Public
Sign In & Join Server To View
fishermann
fishermannOP5mo ago
Yeah I did that but I think the code that initializes the model checks for gpu available
Unknown User
Unknown User5mo ago
Message Not Public
Sign In & Join Server To View

Did you find this page helpful?