我尝试使用以下Dockerfile构建docker镜像
FROM jupyter/datascience-notebook
RUN pip install jupyterhub==0.8.0 && \
pip install s3contents && \
pip install nose dask toolz
# Install R kernel
RUN conda install -c r r-irkernel r-essentials r-base -y
# Install Data Science Package
RUN pip install sagemaker && \
conda install -y pandas
COPY jupyter_notebook_config.py /etc/jupyter/
CMD ["jupyterhub-singleuser"]
# build
# docker build -t singleuser --build-arg BASE_IMAGE=jupyter/scipy-notebook --build-arg JUPYTERHUB_VERSION=0.8.0 .
在具有32G内存，80G磁盘空间的机器上，仅用于此构建
但错误
failed to register layer: Error processing tar file(exit status 1): write /opt/julia/v0.6/HDF5/deps/usr/include/H5overflow.h: no space left on device
df -a
的结果Filesystem 1K-blocks Used Available Use% Mounted on
proc 0 0 0 - /proc
sysfs 0 0 0 - /sys
devtmpfs 16461060 56 16461004 1% /dev
devpts 0 0 0 - /dev/pts
tmpfs 16471592 0 16471592 0% /dev/shm
/dev/xvda1 8123812 1319984 6703580 17% /
devpts 0 0 0 - /dev/pts
none 0 0 0 - /proc/sys/fs/binfmt_misc
cgroup 0 0 0 - /cgroup/blkio
cgroup 0 0 0 - /cgroup/cpu
cgroup 0 0 0 - /cgroup/cpuacct
cgroup 0 0 0 - /cgroup/cpuset
cgroup 0 0 0 - /cgroup/devices
cgroup 0 0 0 - /cgroup/freezer
cgroup 0 0 0 - /cgroup/hugetlb
cgroup 0 0 0 - /cgroup/memory
cgroup 0 0 0 - /cgroup/perf_event
/dev/xvda1 8123812 1319984 6703580 17% /var/lib/docker