kubeflow: Cannot import name 'AsyncMultiKernelManager' with custom notebook image

I’ve added new packages (ffmpeg, libsm6, libxext6) to existing tf notebook dockerfile. Pod is affected by CrashLoopBackOff, log:

Traceback (most recent call last):
  File "/usr/local/bin/jupyter-lab", line 5, in <module>
    from jupyterlab.labapp import main
  File "/usr/local/lib/python3.6/dist-packages/jupyterlab/__init__.py", line 7, in <module>
    from .labapp import LabApp
  File "/usr/local/lib/python3.6/dist-packages/jupyterlab/labapp.py", line 15, in <module>
    from jupyterlab_server import slugify, WORKSPACE_EXTENSION
  File "/usr/local/lib/python3.6/dist-packages/jupyterlab_server/__init__.py", line 4, in <module>
    from .app import LabServerApp
  File "/usr/local/lib/python3.6/dist-packages/jupyterlab_server/app.py", line 11, in <module>
    from jupyter_server.extension.application import ExtensionApp, ExtensionAppJinjaMixin
  File "/usr/local/lib/python3.6/dist-packages/jupyter_server/extension/application.py", line 21, in <module>
    from jupyter_server.serverapp import ServerApp
  File "/usr/local/lib/python3.6/dist-packages/jupyter_server/serverapp.py", line 75, in <module>
    from .services.kernels.kernelmanager import MappingKernelManager, AsyncMappingKernelManager
  File "/usr/local/lib/python3.6/dist-packages/jupyter_server/services/kernels/kernelmanager.py", line 20, in <module>
    from jupyter_client.multikernelmanager import MultiKernelManager, AsyncMultiKernelManager
ImportError: cannot import name 'AsyncMultiKernelManager'

About this issue

Original URL
State: closed
Created 3 years ago
Comments: 36 (20 by maintainers)

Most upvoted comments

@Ursanon The base images are all open source and start from Ubuntu. I linked the PR’s they came from, but they will be merged into the repo and official images will be out soon.

You can try building these and see if that helps.

FROM davidspek/kubeflow-ubuntu-pytorch-base:0.3-cuda

USER root
RUN apt-get -yq update \
         && apt-get -yq install --no-install-recommends \
         libsm6 \
         libxext6 \
         && apt-get clean \
         && rm -rf /var/lib/apt/lists/*
USER $NB_UID
RUN python3 -m pip install --quiet --no-cache-dir\
    tensorboard==2.4.1 \
    opencv-python==4.5.1.48

FROM davidspek/kubeflow-ubuntu-tensorflow-base:0.2-cuda

USER root
RUN apt-get -yq update \
         && apt-get -yq install --no-install-recommends \
         libsm6 \
         libxext6 \
         && apt-get clean \
         && rm -rf /var/lib/apt/lists/*
USER $NB_UID
RUN python3 -m pip install --quiet --no-cache-dir\
    tensorboard==2.4.1 \
    opencv-python==4.5.1.48

davidspek on Mar 8, 2021

@Ursanon Can you tell me what packages you need (and if you need tensorflow GPU support)? Then I will create a Dockerfile for you tomorrow so that you can continue your work while the new images are still being reviewed.

davidspek on Mar 5, 2021