Docker-stacks: conda environment inconsistency in scipy-notebook

[philippegr]

What docker image you are using?
jupyter/scipy-notebook

What complete docker command do you run to launch the container (omitting sensitive values)?

docker run -it jupyter/scipy-notebook bash

What steps do you take once the container is running to reproduce the issue?
conda update --all
(or condacommand that will try to install something)

What do you expect to happen?
No warmings reported by conda

What actually happens?
conda reports an environment inconsistency

Collecting package metadata: done
Solving environment: \
The environment is inconsistent, please check the package plan carefully
The following packages are causing the inconsistency:

• conda-forge/linux-64::matplotlib==3.0.3=py37_1
  done

[miramar-labs]

I am seeing this also in all-spark-notebook

[Blizzke]

I'm experiencing the same issue for my all-spark-notebook based installation and it's preventing me from adding other needed packages.
Is there any progress on this matter?

[Blizzke]

I don't know yet what the impact will be for my users and if it is a fix for anyone else, but installing matplotlib as 3.1* has allowed me to continue working (notebook blocks version of matplotlib on 3.0*).
I also don't know how to feel about sip being downgraded, but I just wanted to share things here for you all.

/opt/conda/bin/conda install matplotlib=3.1*
Collecting package metadata (current_repodata.json): done
Solving environment: failed
Collecting package metadata (repodata.json): done
Solving environment: /
The environment is inconsistent, please check the package plan carefully
The following packages are causing the inconsistency:

  - conda-forge/linux-64::matplotlib==3.0.3=py37_1
done

## Package Plan ##

  environment location: /opt/conda

  added / updated specs:
    - matplotlib=3.1

The following packages will be downloaded:

    package                    |            build
    ---------------------------|-----------------
    matplotlib-3.1.1           |           py37_0           6 KB  conda-forge
    pyqt-5.9.2                 |   py37hcca6a23_0         5.7 MB  conda-forge
    qt-5.9.7                   |       h52cfd70_2        85.9 MB  conda-forge
    sip-4.19.8                 |py37hf484d3e_1000         290 KB  conda-forge
    ------------------------------------------------------------
                                           Total:        91.9 MB

The following NEW packages will be INSTALLED:

  pyqt               conda-forge/linux-64::pyqt-5.9.2-py37hcca6a23_0
  qt                 conda-forge/linux-64::qt-5.9.7-h52cfd70_2

The following packages will be UPDATED:

  matplotlib                                   3.0.3-py37_1 --> 3.1.1-py37_0

The following packages will be DOWNGRADED:

  sip                                4.19.13-py37he1b5a44_0 --> 4.19.8-py37hf484d3e_1000


Proceed ([y]/n)? y


Downloading and Extracting Packages
pyqt-5.9.2           | 5.7 MB    | ################################################ | 100%
qt-5.9.7             | 85.9 MB   | ################################################ | 100%
sip-4.19.8           | 290 KB    | ################################################ | 100%
matplotlib-3.1.1     | 6 KB      | ################################################ | 100%
Preparing transaction: done
Verifying transaction: done
Executing transaction: done

[rubensa]

Last available scipy-notebook version (jupyter/scipy-notebook:6c3390a9292e) on dockerhub has the same environment inconsistency problem. Also looks like the installed conda version is not the last available:

Collecting package metadata: done
Solving environment: \
The environment is inconsistent, please check the package plan carefully
The following packages are causing the inconsistency:

  - conda-forge/linux-64::matplotlib==3.0.3=py37_1
done


==> WARNING: A newer version of conda exists. <==
  current version: 4.6.14
  latest version: 4.7.5

Please update conda by running

    $ conda update -n base conda

[jamesdbrock]

There is a PR for conda 4.7 https://github.com/jupyter/docker-stacks/pull/896