Help Needed: RunPod Pod Creation & Proxy Connection Issues for Stable Diffusion
Hi RunPod community, I'm facing persistent issues with pod creation and proxy connections for Stable Diffusion batch image generation on my Ubuntu server. Here's a summary of the problems:
Pod Creation Failures: Frequent "Something went wrong" errors during pod creation via GraphQL and SDK, even with fallback logic for GPU types, regions, and network volume settings.
Proxy Connectivity Timeouts: Connection timeouts to the Stable Diffusion Web UI proxy URLs (e.g., <pod-id>.proxy.runpod.net:7860), tried both HTTP and HTTPS across multiple ports (3000, 7860).
Infrastructure Delays: Pods often stuck in "RUNNING (waiting for runtime)" state without container startup.
I've implemented extensive fallback logic for GPU selection and connection protocols, but the issues persist. My setup uses the runpod/stable-diffusion:web-ui-10.2.1 image with network volume for faster model loading when possible. Logs show repeated connection attempts failing with "ServerTimeoutError".
Has anyone encountered similar infrastructure or proxy issues? Any advice on troubleshooting or alternative configurations for stable pod creation and connection? (Note: API keys and sensitive data have been removed from this post.)
Thanks for any insights!
2 Replies
Unknown User•4mo ago
Message Not Public
Sign In & Join Server To View
Thank Jason, Way longer than 100 seconds. Ill try the TCP ports. Thanks. Really appreciated.