Using `ENV` and `ARG` to handle secrets can lead to sensitive information being
disclosed
to an inappropriate sphere.

The `ARG` and `ENV` instructions in a Dockerfile are used to configure the image
build and the container environment respectively. Both can be used at image
build time,
during the execution of commands in the container, and `ENV` can also be used
at runtime.

In most cases, build-time and environment variables are used to propagate
configuration items
from the host to the image or container. A typical example for an environmental
variable is the `PATH` variable, used
to configure where system executables are searched for.

Using `ARG` and `ENV` to propagate configuration entries that contain secrets
causes a
security risk. Indeed, in most cases, artifacts of those values are kept in the
final image. The secret information
leak can happen either in the container environment itself, the image
metadata or the build environment logs.

The concrete impact of such an issue highly depends on the secret's purpose and
the exposure sphere:

* Financial impact if a paid service API key is disclosed and used.
* Application compromise if an application's secret, like a session signing
key, is disclosed.
* Infrastructure component takeover, if a system secret, like a remote access
key, is leaked.


== Ask Yourself Whether

* The variable contains a value that should be kept confidential.
* The container image or Dockerfile will be distributed to users who do not need to know the secret value.

There is a risk if you answered yes to any of those questions.


== Recommended Secure Coding Practices

* Use Buildkit's secret mount options when secrets have to be used at build
time.
* For run time secret variables, best practices would recommend only setting
them at runtime, for example with the `--env` option of the `docker run` command.

Note that, in both cases, the files exposing the secrets should be securely
stored and not exposed to a large sphere. In most cases, using a secret vault or
another similar component should be preferred. For example, *Docker Swarm*
provides a *secrets* service that can be used to handle most confidential data.


== Sensitive Code Example

[source,docker]
----
FROM example
# Sensitive
ARG ACCESS_TOKEN
# Sensitive
ENV ACCESS_TOKEN=${ACCESS_TOKEN}
CMD /run.sh
----

== Compliant Solution

For build time secrets, use https://docs.docker.com/engine/reference/builder/#run---mounttypesecret[Buildkit's secret mount type] instead:
[source,docker]
----
FROM example
RUN --mount=type=secret,id=build_secret ./installer.sh
----

For runtime secrets, leave the environment variables empty until runtime:

[source,docker]
----
FROM example
ENV ACCESS_TOKEN=""
CMD /run.sh
----

Store the runtime secrets in an https://docs.docker.com/compose/env-file/[environment file] (such as `.env`) and then start the container with the https://docs.docker.com/engine/reference/commandline/run/#env[`--env-file`] argument:

[source,docker]
----
docker run --env-file .env myImage
----

== See
* https://docs.docker.com/engine/reference/builder/#env[Dockerfile reference]  - ENV command
* https://docs.docker.com/engine/reference/builder/#arg[Dockerfile reference]  - ARG command
* https://docs.docker.com/engine/reference/builder/#run---mounttypesecret[Dockerfile reference] - RUN command secrets mount points
* https://docs.docker.com/engine/swarm/secrets/[Docker documentation] - Manage sensitive data with Docker secrets
* CWE - https://cwe.mitre.org/data/definitions/522[CWE-522 - Insufficiently Protected Credentials]


ifdef::env-github,rspecator-view[]

'''
== Implementation Specification
(visible only on this page)

== Message

When a dangerous environment variable is found:

* Make sure that using ENV to handle a secret is safe here.

When a dangerous argument variable is found:

* Make sure that using ARG to handle a secret is safe here.

== Highlighting

The environment or argument variable name.

'''

endif::env-github,rspecator-view[]