Runpod•4mo ago

CUDA 12.8 Serverless fails

Hey so I just tried pushing a serverless worker for WAN 2.1 Running on cuda 12.8 however it seems to fail tests

Error response from daemon: failed to create task for container: failed to create shim task: OCI runtime create failed: runc create failed: unable to start container process: error during container init: error running hook #0: error running hook: exit status 1, stdout: , stderr: Auto-detected mode as 'legacy' nvidia-container-cli: requirement error: unsatisfied condition: cuda>=12.8, please update your driver to a newer version, or use an earlier cuda container: unknown

Error response from daemon: failed to create task for container: failed to create shim task: OCI runtime create failed: runc create failed: unable to start container process: error during container init: error running hook #0: error running hook: exit status 1, stdout: , stderr: Auto-detected mode as 'legacy' nvidia-container-cli: requirement error: unsatisfied condition: cuda>=12.8, please update your driver to a newer version, or use an earlier cuda container: unknown

6 Replies

Unknown User•4mo ago

Message Not Public

blirdOP•4mo ago

Using 12.8 for base This is what I'm using as base FROM nvidia/cuda:12.8.1-cudnn-devel-ubuntu24.04 AS base but it seems like its not respected during Tests

initialization

Error response from daemon: failed to create task for container: failed to create shim task: OCI runtime create failed: runc create failed: unable to start container process: error during container init: error running hook #0: error running hook: exit status 1, stdout: , stderr: Auto-detected mode as 'legacy' nvidia-container-cli: requirement error: unsatisfied condition: cuda>=12.8, please update your driver to a newer version, or use an earlier cuda container: unknown

initialization

Error response from daemon: failed to create task for container: failed to create shim task: OCI runtime create failed: runc create failed: unable to start container process: error during container init: error running hook #0: error running hook: exit status 1, stdout: , stderr: Auto-detected mode as 'legacy' nvidia-container-cli: requirement error: unsatisfied condition: cuda>=12.8, please update your driver to a newer version, or use an earlier cuda container: unknown

blirdOP•4mo ago

blirdOP•4mo ago

Even when setting this in Advanced settings Seems like the Test environment doesn't support it

Unknown User•4mo ago

Message Not Public

Henky!!•4mo ago

Serverless is #⚡｜serverless You posted in the one for pods

Gaming

Programming

CUDA 12.8 Serverless fails

Did you find this page helpful?