Docker Containers

• The Docker image is collection of layers,
• This image is mounted as one file system when the container starts. For this docker has a filesystems such as overlay or union which understand image layers
• Storage Driver allows docker to understand and mount multiple layers image and one read write layer as a file system inside container
• All the changes made by the container i.e. w.r.t file creations, modifications and deletions are stored in R/W Layer specific to the container
• When the container is deleted the R/W Layer also gets deleted so the changes done by container are also lost.
  
• Docker volumes will solve the problem of preserving data which we will look into very soon.
• Docker has some storage drivers
• overlay2
• aufs
• devicemapper
• btrfs
• zfs
• Refer Here for the article around storage drivers
• using Docker container we would run our applications inside containers and there are two kinds of applications
• stateless applications
  • These applications will not use local storage for storing any information, they would use remote systems such as databases, shared filesystems etc
• stateful applications:
  • These applications will use local storage for storing some application state/data
  • examples: databases, traditional applications
• To make stateful applications work with docker, we need a way to persist the data created in the R/W Layer which is important for your application.
• To do this we need to understand docker volumes Refer Here for the article about docker volumes
• Lets do a small experiment. Create 3 containers as shown below
  
  
  
  

Docker Volumes

• To persist the data from the Docker container we will be using Docker volumes.
• There are three types of docker volumes
• Volumes:
  • Stored in the host system and manged by Docker.
  • Generally these volumes are stored in /var/lib/docker/volumes on Linux
  • Non-Docker processes should not modify this and this is docker specific location.
• Bind Mounts:
  • Can store the data any where on the host system by creating a mapping as shown in the below image
    
    
    
  • Non-Docker processes can easily modify the contents which have been created by your container
• tmpfs mounts
• Working with Docker Volumes
  
  
  
  
• Now inspect docker container docker container inspect c1
  
  
• Now we can use this volume even after the container is deleted to mount to other container with same image
  
  
• Lets think of a database like mysql, I want to run mysql inside container, We need to preserve the data stored in the database. But where does mysql store the data
• Lets observe the Dockerfile of mysql Refer Here
  
• Now lets create a volume for mysql
  
• Now lets create a mysql container

docker container run --name mysql  -v mysqldb:/var/lib/mysql -p 3306:3306 -e MYSQL_ROOT_PASSWORD=qwert456 -d mysql