Skip to content
Switch branches/tags
Go to file
Cannot retrieve contributors at this time
title: Use configuration management
last_reviewed_on: 2021-03-05
review_in: 6 months
# <%= %>
Use [configuration management][] to manage, automate and standardise your infrastructure. When using configuration management you store your [infrastructure as code][] in a version control system such as Git.
## Puppet
The use of [Puppet][] at GDS is diminishing as we move more of our infrastructure to containers and higher level services. It's mainly still in use on [GOV.UK]( but this will decline as more services are moved over to AWS ECS.
If your environment consists of a simple deployment artefact like an [Amazon Machine Image (AMI)][], Puppet may not be necessary, but the process for building that artefact must still be codified and version controlled.
## Terraform
Use [Terraform][] to configure third party cloud infrastructure like Amazon Web Services (AWS) or [Fastly][].
Terraform supports a [large number of providers][], and you can configure it to support multiple environments with different parameters. See the [govuk-aws][] repository as an example.
### Versioning
Due to the high rate of change in many cloud provider offerings we
recommend you keep your Terraform versions and codebases up to date. A
version manager such as [tfenv](,
already used by a number of GDS teams, can help you with supporting
multiple versions.
### Code analysis
There are a number of Terraform focused static analysis tools in use at
GDS. While none of them are yet ubiquitous they can help ensure your
code is more idiomatic, consistent and secure and you should consider the
benefits they could bring to your build pipelines.
* [checkov]( - "detects security and compliance misconfigurations"
* [tfsec]( - "spots potential security issues"
* [tflint]( "linter focused on possible errors, best practices and so on."
## Further reading
Find out more about configuration management in the [Service Manual][].
[configuration management]:
[infrastructure as code]:
[Amazon Elastic Compute Cloud (Amazon EC2)]:
[Amazon Machine Image (AMI)]:
[Service Manual]:
[large number of providers]: