Terraform, a tool created by Hashicorp in 2014, written in Go, aims to build, change and version control your infrastructure. This tool have a powerfull and very intuitive Command Line Interface.
$ curl -O https://releases.hashicorp.com/terraform/ 1.4.6/terraform_1.4.6_darwin_amd64.zip $ sudo terraform_1.4.6_darwin_amd64.zip -d /usr/local/bin/ $ rm terraform_1.4.6_darwin_amd64.zip
First of all, download the tfenv binary and put it in your PATH.
$ git clone https://github.com/tfutils/tfenv.git --depth=1 ~/.tfenv $ echo 'export PATH="$HOME/.tfenv/bin:$PATH"' >> ~/.bash_profile
Then, you can install and use desired version of terraform:
$ tfenv install 1.4.6 $ tfenv use 1.4.6
$ terraform init
It’s the first command you need to execute. Unless, terraform plan, apply, destroy and import will not work. The command terraform init will install :
-
terraform modules
-
eventually a backend
-
and provider(s) plugins
$ terraform init -backend-config=cfg/s3.dev.tf -reconfigure
-reconfigure is used in order to tell terraform to not copy the existing state to the new remote state location.
This command is useful when you have defined some modules. Modules are vendored so when you edit them, you need to get again modules content.
$ terraform get -update=true
When you use modules, the first thing you’ll have to do is to do a terraform
get. This pulls modules into the .terraform directory. Once you do that, unless
you do another terraform get -update=true, you’ve essentially vendored those
modules.
The plan step check configuration to execute and write a plan to apply to target infrastructure provider.
$ terraform plan -out plan.out
It’s an important feature of Terraform that allows a user to see which actions Terraform will perform prior to making any changes, increasing confidence that a change will have the desired effect once applied.
When you execute terraform plan command, terraform will scan all *.tf files in your directory and create the plan.
Now you have the desired state so you can execute the plan.
$ terraform apply plan.out
Good to know: Since terraform v0.11+, in an interactive mode (non
CI/CD/autonomous pipeline), you can just execute terraform apply command which
will print out which actions TF will perform.
By generating the plan and applying it in the same command, Terraform can guarantee that the execution plan won’t change, without needing to write it to disk. This reduces the risk of potentially-sensitive data being left behind, or accidentally checked into version control.
$ terraform apply
$ terraform apply -auto-approve
-var tags-repository_url=${GIT_URL}
$ terraform destroy
Delete all the resources!
A deletion plan can be created before:
$ terraform plan –destroy
-target option allow to destroy only one resource, for example a S3 bucket :
$ terraform destroy -target aws_s3_bucket.my_bucket
The Terraform console command is useful for testing interpolations before using them in configurations. Terraform console will read configured state even if it is remote.
$ echo "aws_iam_user.notif.arn" | terraform console arn:aws:iam::123456789:user/notif
Set the log to DEBUG level and save the log in an output external file.
$ TF_LOG_PATH=mylogfile.txt TF_LOG=debug terraform apply
Validate command is used to validate/check the syntax of the Terraform files. A syntax check is done on all the terraform files in the directory, and will display an error if any of the files doesn’t validate. The syntax check does not cover every syntax common issues.
$ terraform validate
You can use a lot of providers/plugins in your terraform definition resources, so it can be useful to have a tree of providers used by modules in your project.
$ terraform providers
.
├── provider.aws ~> 1.24.0
├── module.my_module
│ ├── provider.aws (inherited)
│ ├── provider.null
│ └── provider.template
└── module.elastic
└── provider.aws (inherited)
$ terraform state push
This command is usefull if for example you originally use a local tf state and then you define a backend storage, in S3 or Consul…
If you moved an existing resource in a module, you need to update the state:
$ terraform state mv aws_iam_role.role1 module.mymodule
To manage multiple distinct sets of infrastructure resources/environments.
Instead of create a directory for each environment to manage, we need to just create needed workspace and use them:
This command create a new workspace and then select it
$ terraform workspace new dev
jq is a lightweight command-line JSON processor. Combined with terraform output it can be powerful.
For example, we define outputs in a module and when we execute terraform apply outputs are displayed:
$ terraform apply ... Apply complete! Resources: 0 added, 0 changed, 0 destroyed. Outputs: elastic_endpoint = vpc-toto-12fgfd4d5f4ds5fngetwe4. eu-central-1.es.amazonaws.com
We can extract the value that we want in order to use it in a script for example. With jq it’s easy:
$ terraform output -json
{
"elastic_endpoint": {
"sensitive": false,
"type": "string",
"value": "vpc-toto-12fgfd4d5f4ds5fngetwe4.
eu-central-1.es.amazonaws.com"
}
}
$ terraform output -json | jq '.elastic_endpoint.value'
"vpc-toto-12fgfd4d5f4ds5fngetwe4.eu-central-1.
es.amazonaws.com"