POD stops working after a day or 2:
Pod stops working after a day or 2, and I have to terminate it and redeploy it and upload the models again. This takes up almost half of the day, as the models are large (more than 6 GBs)
7 Replies
stops working?
can you explain more?, did it stop or your program stopped?
The pod stops working as it shows that the GPU is no longer available, and when I open the WebUI link, it gives an error that "The port is not up yet..", and it stays that way permanently.
so like the models are suddenly gone after a day or two?
GPU is no longer available? in runpod website or what?
the port is not up yet means your appliation isnt listening to the port, what template are you using / app you installed?
"the port is not up yet" is the message showing up when I open the webui on port 3000. Also on the main template page, it is showing me the GPU count as "0 x RTX3090", which means that the gpu is not available. I am using the GPU On-Demand. The template is A1111 Stable Diffusion 1.10.0. Since the port issue remains consistent, I have to terminate the pod and deploy a new one again, hence being forced to upload the models again too.
i think you must use other template, dont run too many programs on that pod without gpus