Dockerfile (5) - Linux Manuals

Dockerfile: automate the steps of creating a Docker image


Dockerfile - automate the steps of creating a Docker image


The Dockerfile is a configuration file that automates the steps of creating a Docker image. It is similar to a Makefile. Docker reads instructions from the Dockerfile to automate the steps otherwise performed manually to create an image. To build an image, create a file called Dockerfile.

The Dockerfile describes the steps taken to assemble the image. When the Dockerfile has been created, call the docker build command, using the path of directory that contains Dockerfile as the argument.



For example:

FROM image


A Dockerfile is a file that automates the steps of creating a Docker image. A Dockerfile is similar to a Makefile.


docker build .

-- Runs the steps and commits them, building a final image.
  The path to the source repository defines where to find the context of the
  build. The build is run by the Docker daemon, not the CLI. The whole
  context must be transferred to the daemon. The Docker CLI reports
  "Sending build context to Docker daemon" when the context is sent to the

  docker build -t repository/tag .

-- specifies a repository and tag at which to save the new image if the build
  succeeds. The Docker daemon runs the steps one-by-one, committing the result
  to a new image if necessary, before finally outputting the ID of the new
  image. The Docker daemon automatically cleans up the context it is given.

Docker re-uses intermediate images whenever possible. This significantly
  accelerates the docker build process.


FROM image

FROM image:tag

FROM image [at] digest

-- The FROM instruction sets the base image for subsequent instructions. A
  valid Dockerfile must have FROM as its first instruction. The image can be any
  valid image. It is easy to start by pulling an image from the public

-- FROM must be the first non-comment instruction in Dockerfile.

-- FROM may appear multiple times within a single Dockerfile in order to create
  multiple images. Make a note of the last image ID output by the commit before
  each new FROM command.

-- If no tag is given to the FROM instruction, Docker applies the
  latest tag. If the used tag does not exist, an error is returned.

-- If no digest is given to the FROM instruction, Docker applies the
  latest tag. If the used tag does not exist, an error is returned.

  -- MAINTAINER sets the Author field for the generated images.
  Useful for providing users with an email or url for support.

  -- RUN has two forms:

  # the command is run in a shell - /bin/sh -c
  RUN <command>

  # Executable form
  RUN ["executable", "param1", "param2"]

-- The RUN instruction executes any commands in a new layer on top of the current
  image and commits the results. The committed image is used for the next step in

-- Layering RUN instructions and generating commits conforms to the core
  concepts of Docker where commits are cheap and containers can be created from
  any point in the history of an image. This is similar to source control.  The
  exec form makes it possible to avoid shell string munging. The exec form makes
  it possible to RUN commands using a base image that does not contain /bin/sh.

Note that the exec form is parsed as a JSON array, which means that you must
  use double-quotes (") around words not single-quotes (').

  -- CMD has three forms:

  # Executable form
  CMD ["executable", "param1", "param2"]`

  # Provide default arguments to ENTRYPOINT
  CMD ["param1", "param2"]`

  # the command is run in a shell - /bin/sh -c
  CMD command param1 param2

-- There should be only one CMD in a Dockerfile. If more than one CMD is listed, only
  the last CMD takes effect.
  The main purpose of a CMD is to provide defaults for an executing container.
  These defaults may include an executable, or they can omit the executable. If
  they omit the executable, an ENTRYPOINT must be specified.
  When used in the shell or exec formats, the CMD instruction sets the command to
  be executed when running the image.
  If you use the shell form of the CMD, the <command> executes in /bin/sh -c:

Note that the exec form is parsed as a JSON array, which means that you must
  use double-quotes (") around words not single-quotes (').

  FROM ubuntu
  CMD echo "This is a test." | wc -

-- If you run command without a shell, then you must express the command as a
  JSON array and give the full path to the executable. This array form is the
  preferred form of CMD. All additional parameters must be individually expressed
  as strings in the array:

  FROM ubuntu
  CMD ["/usr/bin/wc","--help"]

-- To make the container run the same executable every time, use ENTRYPOINT in
  combination with CMD.
  If the user specifies arguments to docker run, the specified commands
  override the default in CMD.
  Do not confuse RUN with CMDRUN runs a command and commits the result.
  CMD executes nothing at build time, but specifies the intended command for
  the image.

  -- LABEL <key>=<value> [<key>=<value> ...]or

  LABEL <key>[ <value>]
  LABEL <key>[ <value>]

The LABEL instruction adds metadata to an image. A LABEL is a
  key-value pair. To specify a LABEL without a value, simply use an empty
  string. To include spaces within a LABEL value, use quotes and
  backslashes as you would in command-line parsing.

  LABEL com.example.vendor="ACME Incorporated"
  LABEL com.example.vendor "ACME Incorporated"
  LABEL ""

An image can have more than one label. To specify multiple labels, separate
  each key-value pair by a space.

Labels are additive including LABELs in FROM images. As the system
  encounters and then applies a new label, new keys override any previous
  labels with identical keys.

To display an image's labels, use the docker inspect command.


-- STOPSIGNAL <signal>
  The STOPSIGNAL instruction sets the system call signal that will be sent
  to the container to exit. This signal can be a signal name in the format
  SIG, for instance SIGKILL, or an unsigned number that matches a
  position in the kernel's syscall table, for instance 9. The default is
  SIGTERM if not defined.

The image's default stopsignal can be overridden per container, using the
  --stop-signal flag on docker-run(1) and docker-create(1).

  -- EXPOSE <port> [<port>...]
  The EXPOSE instruction informs Docker that the container listens on the
  specified network ports at runtime. Docker uses this information to
  interconnect containers using links and to set up port redirection on the host

  -- ENV <key> <value>
  The ENV instruction sets the environment variable  to
  the value <value>. This value is passed to all future
  RUNENTRYPOINT, and CMD instructions. This is
  functionally equivalent to prefixing the command with <key>=<value>.  The
  environment variables that are set with ENV persist when a container is run
  from the resulting image. Use docker inspect to inspect these values, and
  change them using docker run --env <key>=<value>.

Note that setting "ENV DEBIAN_FRONTEND=noninteractive" may cause
  unintended consequences, because it will persist when the container is run
  interactively, as with the following command: docker run -t -i image bash

  -- ADD has two forms:

  ADD <src> <dest>

  # Required for paths with whitespace
  ADD ["<src>",... "<dest>"]

The ADD instruction copies new files, directories
  or remote file URLs to the filesystem of the container at path <dest>.
  Multiple <src> resources may be specified but if they are files or directories
  then they must be relative to the source directory that is being built
  (the context of the build). The <dest> is the absolute path, or path relative
  to WORKDIR, into which the source is copied inside the target container.
  If the <src> argument is a local file in a recognized compression format
  (tar, gzip, bzip2, etc) then it is unpacked at the specified <dest> in the
  container's filesystem.  Note that only local compressed files will be unpacked,
  i.e., the URL download and archive unpacking features cannot be used together.
  All new directories are created with mode 0755 and with the uid and gid of 0.

  -- COPY has two forms:

  COPY <src> <dest>

  # Required for paths with whitespace
  COPY ["<src>",... "<dest>"]

The COPY instruction copies new files from <src> and
  adds them to the filesystem of the container at path . The <src> must be
  the path to a file or directory relative to the source directory that is
  being built (the context of the build) or a remote file URL. The <dest> is an
  absolute path, or a path relative to WORKDIR, into which the source will
  be copied inside the target container. If you COPY an archive file it will
  land in the container exactly as it appears in the build context without any
  attempt to unpack it.  All new files and directories are created with mode 0755
  and with the uid and gid of 0.

  -- ENTRYPOINT has two forms:

  # executable form
  ENTRYPOINT ["executable", "param1", "param2"]`

  # run command in a shell - /bin/sh -c
  ENTRYPOINT command param1 param2

-- An ENTRYPOINT helps you configure a
  container that can be run as an executable. When you specify an ENTRYPOINT,
  the whole container runs as if it was only that executable.  The ENTRYPOINT
  instruction adds an entry command that is not overwritten when arguments are
  passed to docker run. This is different from the behavior of CMD. This allows
  arguments to be passed to the entrypoint, for instance docker run <image> -d
  passes the -d argument to the ENTRYPOINT Specify parameters either in the
  ENTRYPOINT JSON array (as in the preferred exec form above), or by using a CMD
  statement.  Parameters in the ENTRYPOINT are not overwritten by the docker run
  arguments.  Parameters specified via CMD are overwritten by docker run
  arguments.  Specify a plain string for the ENTRYPOINT, and it will execute in
  /bin/sh -c, like a CMD instruction:

  FROM ubuntu
  ENTRYPOINT wc -l -

This means that the Dockerfile's image always takes stdin as input (that's
  what "-" means), and prints the number of lines (that's what "-l" means). To
  make this optional but default, use a CMD:

  FROM ubuntu
  CMD ["-l", "-"]
  ENTRYPOINT ["/usr/bin/wc"]

  -- VOLUME ["/data"]
  The VOLUME instruction creates a mount point with the specified name and marks
  it as holding externally-mounted volumes from the native host or from other

  -- USER daemon
  Sets the username or UID used for running subsequent commands.

The USER instruction can optionally be used to set the group or GID. The
  followings examples are all valid:
  USER [user | user:group | uid | uid:gid | user:gid | uid:group ]

Until the USER instruction is set, instructions will be run as root. The USER
  instruction can be used any number of times in a Dockerfile, and will only affect
  subsequent commands.

  -- WORKDIR /path/to/workdir
  The WORKDIR instruction sets the working directory for the RUNCMD,
  ENTRYPOINTCOPY and ADD Dockerfile commands that follow it. It can
  be used multiple times in a single Dockerfile. Relative paths are defined
  relative to the path of the previous WORKDIR instruction. For example:

  RUN pwd

In the above example, the output of the pwd command is a/b/c.

   -- ARG [=]

The ARG instruction defines a variable that users can pass at build-time to
  the builder with the docker build command using the --build-arg
  <varname>=<value> flag. If a user specifies a build argument that was not
  defined in the Dockerfile, the build outputs a warning.

  [Warning] One or more build-args [foo] were not consumed

The Dockerfile author can define a single variable by specifying ARG once or many
  variables by specifying ARG more than once. For example, a valid Dockerfile:

  FROM busybox
  ARG user1
  ARG buildno

A Dockerfile author may optionally specify a default value for an ARG instruction:

  FROM busybox
  ARG user1=someuser
  ARG buildno=1

If an ARG value has a default and if there is no value passed at build-time, the
  builder uses the default.

An ARG variable definition comes into effect from the line on which it is
  defined in the Dockerfile not from the argument's use on the command-line or
  elsewhere.  For example, consider this Dockerfile:

  1 FROM busybox
  2 USER ${user:-some_user}
  3 ARG user
  4 USER $user

A user builds this file by calling:

  $ docker build --build-arg user=what_user Dockerfile

The USER at line 2 evaluates to some_user as the user variable is defined on the
  subsequent line 3. The USER at line 4 evaluates to what_user as user is
  defined and the what_user value was passed on the command line. Prior to its definition by an
  ARG instruction, any use of a variable results in an empty string.

Warning: It is not recommended to use build-time variables for
 passing secrets like github keys, user credentials etc. Build-time variable
 values are visible to any user of the image with the docker history command.

You can use an ARG or an ENV instruction to specify variables that are
  available to the RUN instruction. Environment variables defined using the
  ENV instruction always override an ARG instruction of the same name. Consider
  this Dockerfile with an ENV and ARG instruction.

  1 FROM ubuntu
  3 ENV CONT_IMG_VER=v1.0.0
  4 RUN echo $CONT_IMG_VER

Then, assume this image is built with this command:

  $ docker build --build-arg CONT_IMG_VER=v2.0.1 Dockerfile

In this case, the RUN instruction uses v1.0.0 instead of the ARG setting
  passed by the user:v2.0.1 This behavior is similar to a shell
  script where a locally scoped variable overrides the variables passed as
  arguments or inherited from environment, from its point of definition.

Using the example above but a different ENV specification you can create more
  useful interactions between ARG and ENV instructions:

  1 FROM ubuntu
  4 RUN echo $CONT_IMG_VER

Unlike an ARG instruction, ENV values are always persisted in the built
  image. Consider a docker build without the --build-arg flag:

  $ docker build Dockerfile

Using this Dockerfile example, CONT_IMG_VER is still persisted in the image but
  its value would be v1.0.0 as it is the default set in line 3 by the ENV instruction.

The variable expansion technique in this example allows you to pass arguments
  from the command line and persist them in the final image by leveraging the
  ENV instruction. Variable expansion is only supported for a limited set of
  Dockerfile instructions. <#environment-replacement>

Docker has a set of predefined ARG variables that you can use without a
  corresponding ARG instruction in the Dockerfile.


To use these, pass them on the command line using --build-arg flag, for

  $ docker build --build-arg HTTPS_PROXY= .

  The ONBUILD instruction adds a trigger instruction to an image. The
  trigger is executed at a later time, when the image is used as the base for
  another build. Docker executes the trigger in the context of the downstream
  build, as if the trigger existed immediately after the FROM instruction in
  the downstream Dockerfile.

You can register any build instruction as a trigger. A trigger is useful if
  you are defining an image to use as a base for building other images. For
  example, if you are defining an application build environment or a daemon that
  is customized with a user-specific configuration.

Consider an image intended as a reusable python application builder. It must
  add application source code to a particular directory, and might need a build
  script called after that. You can't just call ADD and RUN now, because
  you don't yet have access to the application source code, and it is different
  for each application build.

-- Providing application developers with a boilerplate Dockerfile to copy-paste
  into their application is inefficient, error-prone, and
  difficult to update because it mixes with application-specific code.
  The solution is to use ONBUILD to register instructions in advance, to
  run later, during the next build stage.


*May 2014, Compiled by Zac Dover (zdover at redhat dot com) based on Dockerfile documentation. *Feb 2015, updated by Brian Goff (cpuguy83 [at] for readability *Sept 2015, updated by Sally O'Malley (somalley [at] *Oct 2016, updated by Addam Hardy (addam.hardy [at]