error in pod
"2024-03-12T08:39:43.053682465Z /usr/bin/python3: Error while finding module specification for 'vllm.entrypoints.openai.api_server' (ModuleNotFoundError: No module named 'vllm')"
i always run this in A6000, but it is getting error, why this is happening??
19 Replies
Impossible to say since you don't mention a template or anything.
it's my custom template, which i'm using same for months
system logs
container logs:-
@ashleyk please help me out, i have a demo.
n i'm using pod
Looks like a problem with the docker image.
but the same is working from last month
in between i got the same error, but trying after sometime, it was working fine.
You shouldn't use
latest tag, you should contact the devs of the Docker image, its not a RunPod issue and nothing RunPod can do about it. Or alternatively, build your own Docker image so that you are not at the mercy of someone else.ok let me try once
Now i tried again on new pod on the A 6000 GPU, it is working fine , So how can you say that this is the problem in docker image, may be problem is in the gpu,
It's happening cause that docker image uses /workspace as work dir
ok, so what should i use?
just change place where you mount volume storage
it is in my workspace
aka change /workspace to something else
example /workspace2 and in startup command change download dir to /workspace2
okay got it, will try the same.
Thank you for your time
@Papa Madiator so its some race condition that determines which version of /workspace gets mounted first?