目录
https://blog.csdn.net/sqzhao/article/details/71307518
不用交互式shell，直接执行命令。（注意：命令不能加”“。。。另外，好像也不能有&&之类的[可以后续再探索探索。。。]）
其中：
cat /home/disk0/daiwk_img_data/demo_scrapy/lp_mining/lp_spider/containerid | xargs -i docker exec -dt {} /bin/bash /home/data/demo_scrap
y/lp_mining/lp_spider/run_all_images.sh
cat /home/disk0/daiwk_img_data/demo_scrapy/lp_mining/lp_spider/containerid | xargs -i docker exec {} tail /home/data/demo_scrapy/lp_min
ing/lp_spider/crawl.images8.log
当然，也可以交互式啦~~~！！
docker ps
CONTAINER ID IMAGE COMMAND CREATED STATUS PORTS NAMES
e56da0385825 tensorflow/tensorflow:latest "/bin/bash" 3 months ago Up 3 months 6006/tcp, 8888/tcp boring_gates
然后直接如下就行了，即使退出了，再重进也是一样的啦
[work@myhost ~] docker exec -it e56da0385825 /bin/bash
root@e56da0385825:/notebooks# ll
total 416
drwxr-xr-x 2 root root 4096 Oct 11 06:40 ./
drwxr-xr-x 22 root root 4096 Jul 3 12:43 ../
-rw-rw-r-- 1 root root 25023 Apr 28 00:37 1_hello_tensorflow.ipynb
-rw-rw-r-- 1 root root 164559 Apr 28 00:37 2_getting_started.ipynb
-rw-rw-r-- 1 root root 209951 Apr 28 00:37 3_mnist_from_scratch.ipynb
-rw-rw-r-- 1 root root 119 Apr 28 00:37 BUILD
-rw-rw-r-- 1 root root 586 Apr 28 00:37 LICENSE
-rw-r--r-- 1 root root 0 Oct 11 06:40 x
root@e56da0385825:/notebooks# exit
exit
[work@myhost ~] docker exec -it e56da0385825 /bin/bash
root@e56da0385825:/notebooks# ll
total 416
drwxr-xr-x 2 root root 4096 Oct 11 06:40 ./
drwxr-xr-x 22 root root 4096 Jul 3 12:43 ../
-rw-rw-r-- 1 root root 25023 Apr 28 00:37 1_hello_tensorflow.ipynb
-rw-rw-r-- 1 root root 164559 Apr 28 00:37 2_getting_started.ipynb
-rw-rw-r-- 1 root root 209951 Apr 28 00:37 3_mnist_from_scratch.ipynb
-rw-rw-r-- 1 root root 119 Apr 28 00:37 BUILD
-rw-rw-r-- 1 root root 586 Apr 28 00:37 LICENSE
-rw-r--r-- 1 root root 0 Oct 11 06:40 x
root@e56da0385825:/notebooks#
注：mac似乎不能用此方法，因为那个bin是在linux下编的
docker pull jpetazzo/nsenter
docker run --rm -v /usr/bin:/target jpetazzo/nsenter
docker run --rm -v /usr/local/bin:/target jpetazzo/nsenter
wget -P ~ https://raw.githubusercontent.com/yeasy/docker_practice/docker-legacy/_local/.bashrc_docker --no-check-certificate;
echo "[ -f ~/.bashrc_docker ] && . ~/.bashrc_docker" >> ~/.bashrc; source ~/.bashrc
使用时，直接找到对应的containerid，然后：
docker-enter ffd32a5b82f7
docker run -idt -v /home/work/daiwenkai:/home/data 390b21e493af /bin/bash
docker images
REPOSITORY TAG IMAGE ID CREATED VIRTUAL SIZE
ubuntu ru latest f5bb94a8fac4 11 days ago 117.3 MB
xxx.baidu.com/daiwenkai/deep-learning-factory 1.0.2 390b21e493af 7 weeks ago 2.174 GB
tensorflow/tensorflow latest 05a46e5af4d3 8 weeks ago 1.028 GB
xxx.baidu.com/public/centos6u3 1.0.1 9a6d077da3a3 23 months ago 1.363 GB
docker ps
CONTAINER ID IMAGE COMMAND CREATED STATUS PORTS NAMES
ffd32a5b82f7 ubuntu:latest "/bin/bash" About a minute ago Up About a minute loving_torvalds
166ab859e45c 390b21e493af:latest "/bin/bash" 6 weeks ago Up 6 weeks 6006/tcp, 8888/tcp lonely_bartik
kill并rm所有container
docker ps -aq | xargs docker rm -f
docker pull ubuntu
docker commit -m "1.0.1 version" ffd32a5b82f7 xxx.baidu.com/daiwenkai/scrapy-framework:1.0.1
docker push xxx.baidu.com/daiwenkai/scrapy-framework:1.0.1
导出image
docker save -o scrapy-framework.docker.tar 8ce4bbc775d2
导入image
docker load -i scrapy-framework.docker.tar
导出container
docker export ffd32a5b82f7 > scrapy-framework.1.0.1.tar
导入container对应的image，启动时，和之前的container一样
cat scrapy-framework.tar | docker import - daiwk/scrapy-framework
cat deep-learning-factory.1.0.3.tar | docker import - test/deep-learning-factory
在win7安装时，要用docker toolbox(https://download.docker.com/win/stable/DockerToolbox.exe)
win10以上时，用docker-ce(https://store.docker.com/editions/community/docker-ce-desktop-windows?tab=description)
发现有的container没法stop/kill/rm的时候，例如
docker rm -f 316f2eca7a3b
Error response from daemon: Could not kill running container, cannot remove - [2] Container does not exist: container destroyed
Error: failed to remove containers: [316f2eca7a3b]
尝试找到docker的daemon进程，并kill，然后重启：
ps aux | grep docker
root 4157 0.0 0.0 105352 832 pts/3 S+ 13:26 0:00 grep docker
root 32816 0.0 0.0 1562760 42500 ? Sl Jun02 1:36 /usr/bin/docker -d -b docker0 -H tcp://0.0.0.0:2375 -H unix:///var/run/docker.sock --insecure-registry registry.xxxxx.com -g /home/work/docker
kill -9 32816
nohup /usr/bin/docker -d -b docker0 -H tcp://0.0.0.0:2375 -H unix:///var/run/docker.sock --insecure-registry registry.xxxxx.com -g /home/work/docker &
这个时候就发现container的状态已经是exited (137) 7 seconds ago了，再去rm，就可以啦