Volumes:

By default docker doesnot persist the data. For example,

You run the mysql server in a docker and DB gets stored in a file. As long as Docker runs you will not have any problem. Due to some unexpected things the docker instance got removed . Also the DB is gone.

To avoid such stuff, Persistence Data needs to be outside of the docker container. So when the docker instance removed, you can launch a another docker container with this persistence data, so that it will retain the state.

Major places where the persistence data is required in,

• DBs (SQL, NoSQL or any sort of DB apps such as MySQL, Mongo, Cassandra etc)
• WebServers (apache2, ngnix,etc)
• Application needs to Log the Files which needs to be persisted.

To Persist the data in docker:

• Create a docker volume
• Attach the docker volume to the container at the time of container creation.

Volume is just a directory located in the host and will be mapped to the docker instance.

Docker Volume Creation :

docker volume create <volume name>

The volume will be created in /var/lib/docker/volumes folder.

To List :

docker volume list

To see the details :

docker inspect <volume name>

$ sudo docker volume create myvol1
myvol1
$ sudo docker volume list
DRIVER              VOLUME NAME
local               myvol
local               myvol1
$ sudo docker inspect myvol1
[
    {
        "CreatedAt": "2017-10-23T05:52:38Z",
        "Driver": "local",
        "Labels": {},
        "Mountpoint": "/var/lib/docker/volumes/myvol1/_data",
        "Name": "myvol1",
        "Options": {},
        "Scope": "local"
    }
]

Create a docker container with volume:

The mount option in the container creation can be used for volume.

--mount source=<volume name>,target=<target directory in the container>

<volume name> is the name of the volume

<target directory> - In the container, the volume will be mounted in the specified directory.

Example:

docker run -dit --name my-apache-app -p 8080:80 --mount source=myvol,target=/usr/local/apache2/htdocs/ httpd:2.4

Now you can query on port 8080 from the docker machine