Atlantis

Atlantis is a tool for safe collaboration on Terraform repositories.

Maintainers

Name	Email	Url
lkysow
jamengual
chenrui333
nitrocode
genpage
gmartinez-sisti

Links

• Introduction
• Prerequisites
• Required Configuration
• Additional manifests
• Values
• Upgrading
  • From 4.* to 5.*
  • From 2.* to 3.*
  • From 1.* to 2.*
• Testing the Deployment
• Update documentation
• Run unit tests

Introduction

This chart creates a single pod in a StatefulSet running Atlantis. Atlantis persists Terraform plan files and lockfiles to disk for the duration of a Pull/Merge Request. These files are stored in a PersistentVolumeClaim to survive Pod failures.

Prerequisites

• Kubernetes 1.9+
• PersistentVolume support

Required Configuration

In order for Atlantis to start and run successfully:

1. At least one of the following sets of credentials must be defined:

   • github
   • gitlab
   • bitbucket
   • azuredevops

   Refer to values.yaml for detailed examples. They can also be provided directly through a Kubernetes Secret, use the variable vcsSecretName to reference it.

2. Supply a value for orgAllowlist, e.g. github.com/myorg/*.

Additional manifests

It is possible to add additional manifests into a deployment, to extend the chart. One of the reason is to deploy a manifest specific to a cloud provider ( BackendConfig on GKE for example ).

extraManifests:
  - apiVersion: cloud.google.com/v1beta1
    kind: BackendConfig
    metadata:
      name: ".Release.Name-test"
    spec:
      securityPolicy:
        name: "gcp-cloud-armor-policy-test"

Values

Key	Type	Default	Description
affinity	object	{}
allowDraftPRs	bool	false	Enables atlantis to run on a draft Pull Requests.
allowForkPRs	bool	false	Enables atlantis to run on a fork Pull Requests.
api	object	{}	Optionally specify an API secret to enable the API. Check values.yaml for examples.
apiSecretName	string	""	If managing secrets outside the chart for the API secret, use this variable to reference the secret name.
atlantisDataDirectory	string	"/atlantis-data"	Path to the data directory for the volumeMount.
atlantisUrl	string	""	An option to override the atlantis url, if not using an ingress, set it to the external IP. Check values.yaml for examples.
aws	object	{}	To specify AWS credentials to be mapped to ~/.aws or to aws.directory. Check values.yaml for examples.
awsSecretName	string	""	To reference an already existing Secret object with AWS credentials
azuredevops	object	{}	If using Azure DevOps, please enter your values as follows. The chart will perform the base64 encoding for you for values that are stored in secrets. Check values.yaml for examples.
basicAuth	object	{"password":"","username":""}	Optionally specify an username and a password for basic authentication.
basicAuthSecretName	string	""	If managing secrets outside the chart for the Basic Auth secret, use this variable to reference the secret name.
bitbucket	object	{}	If using Bitbucket, there are two approaches: Bitbucket Server, deployed in your own infrastructure and Cloud available at (https://Bitbucket.org). The chart will perform the base64 encoding for you for values that are stored in secrets. Check values.yaml for examples.
command	list	[]	Override the command field of the Atlantis container.
commonLabels	object	{}	Common Labels for all resources created by this chart.
config	string	""	Override atlantis main configuration by config map, ref: https://www.runatlantis.io/docs/using-slack-hooks.html#configuring-atlantis. Check values.yaml for examples.
containerSecurityContext	object	{}	Check values.yaml for examples.
customPem	string	""	Allows to override the /etc/ssl/certs/ca-certificates.cer with your custom one. You have to create a secret with the specified name.
dataStorage	string	""	DEPRECATED - Disk space available to check out repositories. Example: 5Gi.
defaultTFVersion	string	""	Sets the default terraform version to be used in atlantis server. Check values.yaml for examples.
disableApply	bool	false	Disables running atlantis apply regardless of which flags are sent with it.
disableApplyAll	bool	false	Disables running atlantis apply without any flags.
disableRepoLocking	bool	false	Stops atlantis locking projects and or workspaces when running terraform.
dnsConfig	object	{}	Optionally specify dnsConfig for the Atlantis pod. Check values.yaml for examples.
dnsPolicy	string	"ClusterFirst"	Optionally specify dnsPolicy parameter to specify a DNS policy for a pod Check https://kubernetes.io/docs/concepts/services-networking/dns-pod-service/#pod-s-dns-policy
enableDiffMarkdownFormat	bool	false	Use Diff Markdown Format for color coding diffs.
enableKubernetesBackend	bool	false	Optionally deploy rbac to allow for the serviceAccount to manage terraform state via the kubernetes backend.
environment	object	{}	Environtment values to add to the Atlantis pod. Check values.yaml for examples.
environmentRaw	list	[]	Optionally specify additional environment variables in raw yaml format. Useful to specify variables refering to k8s objects. Check values.yaml for examples.
environmentSecrets	list	[]	Optionally specify additional environment variables to be populated from Kubernetes secrets. Useful for passing in TF_VAR_foo or other secret environment variables from Kubernetes secrets. Check values.yaml for examples.
extraAnnotations	object	{}
extraArgs	list	[]	Optionally specify extra arguments for the Atlantis pod. Check values.yaml for examples.
extraContainers	list	[]	Optionally specify extra containers for the Atlantis pod. Check values.yaml for examples.
extraManifests	list	[]	Optionally specify additional manifests to be created. Check values.yaml for examples.
extraPath	string	""	Additional path (: separated) that will be appended to the system PATH environment variable.
extraVolumeMounts	list	[]	Optionally specify additional volume mounts for the container. Check values.yaml for examples.
extraVolumes	list	[]	Optionally specify additional volumes for the pod. Check values.yaml for examples.
fullnameOverride	string	""	Provide a name to substitute for the full names of resources.
gitconfig	string	""	When referencing Terraform modules in private repositories, it may be helpful (necessary?) to use redirection in a .gitconfig. Check values.yaml for examples.
gitconfigReadOnly	bool	true	When true gitconfig file is mounted as read only. When false, the gitconfig value will be copied to '/home/atlantis/.gitconfig' before starting the atlantis process, instead of being mounted as a file.
gitconfigSecretName	string	""	If managing secrets outside the chart for the gitconfig, use this variable to reference the secret name
github	object	{}	If using GitHub, please enter your values as follows. The chart will perform the base64 encoding for values that are stored in secrets. The 'hostname' key is exclusive to GitHub Enterprise installations. Check values.yaml for examples.
githubApp	object	{}	If using a GitHub App, please enter your values as follows. The chart will perform the base64 encoding for you for values that are stored in secrets. installationId is necessary when there are multiple installs of the Github App. Check values.yaml for examples.
gitlab	object	{}	If using GitLab, please enter your values as follows. The 'hostname' key is exclusive to GitLab Enterprise installations. The chart will perform the base64 encoding for you for values that are stored in secrets. Check values.yaml for examples.
googleServiceAccountSecrets	list	[]	Optionally specify google service account credentials as Kubernetes secrets. If you are using the terraform google provider you can specify the credentials as "${file("/var/secrets/some-secret-name/key.json")}". Check values.yaml for examples.
hidePrevPlanComments	bool	false	Enables atlantis to hide previous plan comments.
hideUnchangedPlanComments	bool	false	Enables atlantis to hide no-changes plan comments from the pull request.
hostAliases	list	[]	Optionally specify hostAliases for the Atlantis pod. Check values.yaml for examples.
hostNetwork	bool	false
image.pullPolicy	string	"Always"
image.repository	string	"ghcr.io/runatlantis/atlantis"
image.tag	string	""	If not set appVersion field from Chart.yaml is used
imagePullSecrets	list	[]	Optionally specify an array of imagePullSecrets. Secrets must be manually created in the namespace. ref: https://kubernetes.io/docs/tasks/configure-pod-container/pull-image-private-registry/. Check values.yaml for examples.
ingress.annotations	object	{}	Check values.yaml for examples.
ingress.apiVersion	string	""
ingress.enabled	bool	true
ingress.host	string	""
ingress.hosts	list	[]	Used when several hosts are required. Check values.yaml for examples.
ingress.ingressClassName	string	nil
ingress.labels	object	{}
ingress.path	string	"/*"	Use / for nginx.
ingress.pathType	string	"ImplementationSpecific"
ingress.paths	list	[]	Used when several paths under the same host, with different backend services, are required. Check values.yaml for examples.
ingress.tls	list	[]	Check values.yaml for examples.
initConfig.containerSecurityContext	object	{}	Security context for the container.
initConfig.enabled	bool	false	Install providers/plugins into a path shared with the Atlantis pod.
initConfig.image	string	"alpine:latest"
initConfig.imagePullPolicy	string	"IfNotPresent"
initConfig.script	string	Check values.yaml.	Script to run on the init container.
initConfig.sharedDir	string	"/plugins"	SharedDir is set as env var INIT_SHARED_DIR.
initConfig.sharedDirReadOnly	bool	true
initConfig.sizeLimit	string	"300Mi"	Size for the shared volume.
initConfig.workDir	string	"/tmp"
initContainers	list	[]	Optionally specify init containers manifests to be added to the Atlantis pod. Check values.yaml for examples.
lifecycle	object	{}	Set lifecycle hooks. https://kubernetes.io/docs/tasks/configure-pod-container/attach-handler-lifecycle-event/.
livenessProbe.enabled	bool	true
livenessProbe.failureThreshold	int	5
livenessProbe.initialDelaySeconds	int	5
livenessProbe.periodSeconds	int	60	We only need to check every 60s since Atlantis is not a high-throughput service.
livenessProbe.scheme	string	"HTTP"
livenessProbe.successThreshold	int	1
livenessProbe.timeoutSeconds	int	5
loadEnvFromConfigMaps	list	[]	Optionally specify additional Kubernetes ConfigMaps to load environment variables from. All key-value pairs within these ConfigMaps will be set as environment variables. Note that any variables set here will be ignored if also defined in the env block of the atlantis statefulset. For example, providing ATLANTIS_ALLOW_FORK_PRS here and defining a value for allowForkPRs will result in the allowForkPRs value being used. Check values.yaml for examples.
loadEnvFromSecrets	list	[]	Optionally specify additional Kubernetes secrets to load environment variables from. All key-value pairs within these secrets will be set as environment variables. Note that any variables set here will be ignored if also defined in the env block of the atlantis statefulset. For example, providing ATLANTIS_GH_USER here and defining a value for github.user will result in the github.user value being used. Check values.yaml for examples.
lockingDbType	string	""	Set the desired Locking DB type Accepts boltdb or redis.
logLevel	string	""	Specify the log level for Atlantis. Accepts: debug, info, warn, or error.
nameOverride	string	""	Provide a name to substitute for the name of the chart.
netrc	string	""	When referencing Terraform modules in private repositories or registries (such as Artfactory) configuing a .netrc file for authentication may be required. Check values.yaml for examples.
netrcSecretName	string	""	If managing secrets outside the chart for the netrc file, use this variable to reference the secret name
nodeSelector	object	{}
orgAllowlist	string	"<replace-me>"	Replace this with your own repo allowlist.
orgWhitelist	string	"<deprecated>"	Deprecated in favor of orgAllowlist.
podMonitor	object	{"enabled":false,"interval":"30s"}	Enable this if you're using Google Managed Prometheus.
podTemplate.annotations	object	{}	Check values.yaml for examples.
podTemplate.labels	object	{}
readinessProbe.enabled	bool	true
readinessProbe.failureThreshold	int	5
readinessProbe.initialDelaySeconds	int	5
readinessProbe.periodSeconds	int	60
readinessProbe.scheme	string	"HTTP"
readinessProbe.successThreshold	int	1
readinessProbe.timeoutSeconds	int	5
redis	object	{}	Configure Redis Locking DB. lockingDbType value must be redis for the config to take effect. Check values.yaml for examples.
redisSecretName	string	""	When managing secrets outside the chart for the Redis secret, use this variable to reference the secret name.
replicaCount	int	1	Replica count for Atlantis pods.
repoConfig	string	""	Use Server Side Repo Config, ref: https://www.runatlantis.io/docs/server-side-repo-config.html. Check values.yaml for examples.
resources	object	{}	Resources for Atlantis. Check values.yaml for examples.
service.annotations	object	{}
service.externalTrafficPolicy	string	nil
service.loadBalancerIP	string	nil
service.loadBalancerSourceRanges	list	[]
service.nodePort	string	nil
service.port	int	80
service.portName	string	"atlantis"
service.targetPort	int	4141
service.type	string	"NodePort"
serviceAccount.annotations	object	{}	Annotations for the Service Account. Check values.yaml for examples.
serviceAccount.create	bool	true	Specifies whether a ServiceAccount should be created.
serviceAccount.mount	bool	true	If false, no kubernetes service account token will be mounted to the pod.
serviceAccount.name	string	nil	The name of the ServiceAccount to use. If not set and create is true, a name is generated using the fullname template.
serviceAccountSecrets	object	{}	To keep backwards compatibility only. Deprecated (see googleServiceAccountSecrets). To be used for mounting credential files (when using google provider). Check values.yaml for examples.
servicemonitor.additionalLabels	object	{}	Prometheus ServiceMonitor labels.
servicemonitor.auth.basicAuth	object	{"enabled":false}	If auth is enabled on Atlantis, use one of the following mechanism.
servicemonitor.auth.basicAuth.enabled	bool	false	Authentication from the secret generated with the basicAuth values this will reference the username and password keys from the atlantis-basic-auth secret.
servicemonitor.auth.externalSecret.enabled	bool	false	Authentication based on an external secret
servicemonitor.auth.externalSecret.keys	object	{}	Check values.yaml for examples.
servicemonitor.auth.externalSecret.name	string	""
servicemonitor.enabled	bool	false	To enable a Prometheus servicemonitor, set enabled to true, and enable the metrics in this file's repoConfig by setting a value for metrics.prometheus.endpoint.
servicemonitor.interval	string	"30s"
servicemonitor.path	string	"/metrics"
statefulSet.annotations	object	{}
statefulSet.labels	object	{}
statefulSet.priorityClassName	string	""
statefulSet.securityContext.fsGroup	int	1000
statefulSet.securityContext.fsGroupChangePolicy	string	"OnRootMismatch"
statefulSet.securityContext.runAsUser	int	100	It is not recommended to run atlantis as root.
statefulSet.shareProcessNamespace	bool	false	Option to share process namespace with atlantis container.
statefulSet.updateStrategy	object	{}
storageClassName	string	""	DEPRECATED - Storage class name for Atlantis disk.
terminationGracePeriodSeconds	int	default depends on the kubernetes version.	Optionally customize the termination grace period in seconds.
test.annotations	object	{}
test.enabled	bool	true	Enables test container.
test.image	string	"bats/bats"
test.imageTag	string	"1.9.0"
tlsSecretName	string	""	TLS Secret Name for Atlantis pod.
tolerations	list	[]
topologySpreadConstraints	list	[]	You can use topology spread constraints to control how Pods are spread across your cluster among failure-domains such as regions, zones, nodes, and other user-defined topology domains. (requires Kubernetes >= 1.19). Check values.yaml for examples.
vcsSecretName	string	""	If managing secrets outside the chart for the webhook, use this variable to reference the secret name
volumeClaim.accessModes[0]	string	"ReadWriteOnce"
volumeClaim.dataStorage	string	"5Gi"	Disk space available to check out repositories.
volumeClaim.enabled	bool	true
volumeClaim.storageClassName	string	""	Storage class name (if possible, use a resizable one).
webhook_ingress.annotations	object	{}	Check values.yaml for examples.
webhook_ingress.apiVersion	string	""
webhook_ingress.enabled	bool	false	When true creates a secondary webhook.
webhook_ingress.host	string	""
webhook_ingress.hosts	list	[]	Used when several hosts are required. Check values.yaml for examples.
webhook_ingress.ingressClassName	string	nil
webhook_ingress.labels	object	{}
webhook_ingress.path	string	"/*"	Use / for nginx.
webhook_ingress.pathType	string	"ImplementationSpecific"
webhook_ingress.paths	list	[]	Used when several paths under the same host, with different backend services, are required. Check values.yaml for examples.
webhook_ingress.tls	list	[]	TLS configuration. Check values.yaml for examples.

Upgrading

From 4.* to 5.*

A breaking change was merged on #304. This change moves the atlantis data volume to a separate PVC to allow changing the volume size without removing the statefulset. Unmistakingly, this change only bumped the minor version, released as 4.22.0, instead of incrementing the major version. The release 5.0.0 addresses this issue. Please find the required upgrade manual steps below.

Upgrade steps

Trying to upgrade the release directly will return the following error:

Error: UPGRADE FAILED: cannot patch "atlantis" with kind StatefulSet: StatefulSet.apps "atlantis" is invalid: spec: Forbidden: updates to statefulset spec for fields other than 'replicas', 'ordinals', 'template', 'updateStrategy', 'persistentVolumeClaimRetentionPolicy' and 'minReadySeconds' are forbidden

To allow the upgrade to proceed, first we need to remove the statefulset controller, copy all the data from the old volume to the new one and then cleanup.

NOTE: These steps will ensure no data is lost, however during the copy process the atlantis server is unavailable and any webhooks sent to atlantis during that window are missed and won't be processed.

Upgrade steps:

1. For atlantis deployments managed by GitOps solutions such as ArgoCD, disabling the automatic synchronization is required

2. Create a temporary helm configuration file named migration-4-to-5.yaml, the configuration includes the required settings to allow the atlantis data to be copied to the new volume before the atlantis server is started. The suggested configuration implies that the default command for atlantis has not been customized. If required, adjust the persistentVolumeClaim.claimName value.

   command:
     - sh
     - -c
     - 'cp -Rv /atlantis-data-old/* "${ATLANTIS_DATA_DIR}" && exec /usr/local/bin/docker-entrypoint.sh server'
   
   extraVolumes:
     - name: atlantis-data-old
       persistentVolumeClaim:
         claimName: atlantis-data-atlantis-0
   
   extraVolumeMounts:
     - name: atlantis-data-old
       mountPath: /atlantis-data-old

3. Delete the current statefulset without removing the atlantis pod. This can be achieved with the parameter --cascade=orphan:

   NOTE: If required, adjust the namespace and the statefulset name.

   kubectl delete statefulsets.apps atlantis --namespace atlantis --cascade=orphan

4. Deploy the new atlantis version, this will recreate the statefulset and update the atlantis pod acordingly. E.g.:

   helm upgrade -i atlantis runatlantis/atlantis --version 5.0.0 -f my-config.yaml -f migration-4-to-5.yaml

5. Check the logs for the atlantis pod, it should show the copy process followed by start of the atlantis server. E.g.:

   '/atlantis-data-old/atlantis.db' -> '/atlantis-data/atlantis.db'
   '/atlantis-data-old/bin' -> '/atlantis-data/bin'
   '/atlantis-data-old/existing-data.example' -> '/atlantis-data/existing-data.example'
   '/atlantis-data-old/plugin-cache' -> '/atlantis-data/plugin-cache'
   No files found in /docker-entrypoint.d/, skipping
   {"level":"info","ts":"2024-05-01T17:17:03.406Z","caller":"server/server.go:447","msg":"Utilizing BoltDB","json":{}}
   {"level":"info","ts":"2024-05-01T17:17:03.411Z","caller":"policy/conftest_client.go:153","msg":"failed to get default conftest version. Will attempt request scoped lazy loads DEFAULT_CONFTEST_VERSION not set","json":{}}
   {"level":"info","ts":"2024-05-01T17:17:03.413Z","caller":"server/server.go:985","msg":"Atlantis started - listening on port 4141","json":{}}
   {"level":"info","ts":"2024-05-01T17:17:03.413Z","caller":"scheduled/executor_service.go:51","msg":"Scheduled Executor Service started","json":{}}

6. Run a second deployment without the extra configuration file. E.g.:

   helm upgrade -i atlantis runatlantis/atlantis --version 5.0.0 -f my-config.yaml

7. Remove the old persistentVolumeClaim:

   kubectl delete persistentvolumeclaims atlantis-data-atlantis-0 --namespace atlantis

8. For atlantis deployments managed by GitOps solutions such as ArgoCD, the automatic synchronization can now be enabled

From 4.0.* to 4.1.*

• The following value are deprecated:

  • dataStorage
  • storageClassName

• In favor of the new working way:

  • volumeClaim.enabled
  • volumeClaim.dataStorage
  • volumeClaim.storageClassName

From 2.* to 3.*

• The following value names have been removed. They are replaced by Server-side Repository Configuration
  • requireApproval
  • requireMergeable
  • allowRepoConfig

To replicate your previous configuration, run Atlantis locally with your previous flags and Atlantis will print out the equivalent repo-config, for example:

$ atlantis server --allow-repo-config --require-approval --require-mergeable --gh-user=foo --gh-token=bar --repo-allowlist='*'
WARNING: Flags --require-approval, --require-mergeable and --allow-repo-config have been deprecated.
Create a --repo-config file with the following config instead:

---
repos:
- id: /.*/
  apply_requirements: [approved, mergeable]
  allowed_overrides: [apply_requirements, workflow]
  allow_custom_workflows: true

or use --repo-config-json='{"repos":[{"id":"/.*/", "apply_requirements":["approved", "mergeable"], "allowed_overrides":["apply_requirements","workflow"], "allow_custom_workflows":true}]}'

Then use this YAML in the new repoConfig value:

repoConfig: |
  ---
  repos:
  - id: /.*/
    apply_requirements: [approved, mergeable]
    allowed_overrides: [apply_requirements, workflow]
    allow_custom_workflows: true

From 1.* to 2.*

• The following value names have changed:
  • allow_repo_config => allowRepoConfig
  • atlantis_data_storage => dataStorage NOTE: more than just a snake_case change
  • atlantis_data_storageClass => storageClassName NOTE: more than just a snake_case change
  • bitbucket.base_url => bitbucket.baseURL

Testing the Deployment

To perform a smoke test of the deployment (i.e. ensure that the Atlantis UI is up and running):

1. Install the chart. Supply your own values file or use test-values.yaml, which has a minimal set of values required in order for Atlantis to start.

   helm repo add runatlantis https://runatlantis.github.io/helm-charts
   helm install -f test-values.yaml my-atlantis runatlantis/atlantis --debug

2. Run the tests:

   helm test my-atlantis

Update documentation

Documentation is auto-generated using helm-docs.

To update run the following (from the root path of the repository):

1. If required, update charts/atlantis/README.md.gotmpl
2. Run make docs

Run unit tests

From the root of the repository, run:

make unit-test-run-atlantis

---

Autogenerated from chart metadata using helm-docs v1.14.2