Docker Cookbook

2.1 Keeping Changes Made to a Container by Committing
to an Image
Problem
After making some changes inside a container, you decide that you would like to keep
those changes. You do not want to lose those changes after you exit or stop the con‐
tainer, and you would like to reuse this type of container as a basis for others.
Solution
Commit the changes that you made by using the 
docker commit
command and
define a new image.
Let’s start a container with an interactive bash shell and update the packages in it:
$ docker run -t -i ubuntu:14.04 /bin/bash
root@69079aaaaab1:/# apt-get update
When you exit the container, it stops running, but it is still available to you until you
remove it entirely with 
docker rm
. So before you do this, you can commit the
changes made to the container and create a new image, 
ubuntu:update
. The name of
the image is 
ubuntu
, and you add the tag 
update
 (see 
Recipe 2.6
) to mark the differ‐
ence from the 
ubuntu:latest
image:
$ docker commit 69079aaaaab1 ubuntu:update
13132d42da3cc40e8d8b4601a7e2f4dbf198e9d72e37e19ee1986c280ffcb97c
$ docker images
REPOSITORY TAG IMAGE ID CREATED VIRTUAL SIZE
ubuntu update 13132d42da3c 5 days ago 213 MB
...
You can now safely remove the stopped container and you will be able to start new
ones based on the 
ubuntu:update
image.
38 | Chapter 2: Image Creation and Sharing

---

Discussion
You can inspect the changes that have been made inside this container with the
docker diff
command:
$ docker diff 69079aaaaab1
C /root
A /root/.bash_history
C /tmp
C /var
C /var/cache
C /var/cache/apt
D /var/cache/apt/pkgcache.bin
D /var/cache/apt/srcpkgcache.bin
C /var/lib
C /var/lib/apt
C /var/lib/apt/lists
...
A
means that the file or directory listed was added, 
C
means that there was a change
made, and 
D
means that it was deleted.