無料でビルドを開始
CircleCI.comアカデミーブログコミュニティサポート

Installation reference

2 months ago11 min read
Server v4.2
Server Admin
このページの内容
  • Example manifests
  • AWS
  • GCP
  • All Helm values.yaml options

Example manifests

The following are example manifests that contain the basic required parameters necessary to spin up the circleci-server Helm installation.

AWS

The below is an example manifest of the necessary parameters for an installation of circleci-server in an AWS environment. Note that this installation uses IAM roles for service accounts (IRSA), which is recommended. Fields with base64 encoding are marked as such.

global:
  domainName: "<full-domain-name-of-your-install>"
  license: '<license>'
  container:
    registry: cciserver.azurecr.io
    org:

apiToken: "<circleci-api-token>"
sessionCookieKey: "<session-cookie-key>"

keyset:
  signing: '<generated-signing-key>'
  encryption: '<generated-encryption-key>'

nomad:
  server:
    gossip:
      encryption:
        key: "<nomad-gossip-encryption-key>"
    rpc:
      mTLS:
        enabled: true
        CACertificate: "<nomad-mtls-base64-ca>"
        certificate: "<nomad-mtls-base64-cert>"
        privateKey: "<nomad-mtls-base64-key>"

object_storage:
  bucketName: '<s3-bucket-name>'
  s3:
    enabled: true
    endpoint: "<aws-region-url>" # ex: https://s3.us-east-1.amazonaws.com
    region: "<aws-region>" # defaults to us-east-1, Update this value if you are using a different region
    irsaRole: "<arn-of-irsa-role>"

github:
  clientId: "<generated-github-client-id>"
  clientSecret: "<generated-github-client-secret>"

vm_service:
  providers:
    ec2:
      enabled: true
      region: "<aws-region>"
      subnets:
      - "<subnet-id>"
      securityGroupId: "<security-group-id>"
      irsaRole: "<arn-of-irsa-role>"

mongodb:
  auth:
    rootPassword: "<mongodb-root-password>"
    password: "<mongodb-password>"

postgresql:
  auth:
    postgresPassword: "<postgres-password>"

pusher:
  secret: "<pusher-secret>"

rabbitmq:
  auth:
    password: "<rabbitmq-password>"
    erlangCookie: "<rabbitmq-erlang-cookie>"

GCP

The below is an example manifest of the necessary parameters for an installation of circleci-server in a GCP environment. Note that this installation uses Workload Identity, which is recommended. Fields with base64 encoding are marked as such.

global:
  domainName: "<full-domain-name-of-your-install>"
  license: '<license-for-circleci-server>'
  container:
    registry: cciserver.azurecr.io
    org:

apiToken: "<circleci-api-token>"
sessionCookieKey: "<session-cookie-key>"
keyset:
  signing: '<generated-signing-key>'
  encryption: '<generated-encryption-key>'

github:
  clientId: "<generated-github-client-id>"
  clientSecret: "<generated-github-client-secret>"
object_storage:
  bucketName: "<gcs-bucket-name>"
  gcs:
    enabled: true
    workloadIdentity: "<service-account-email-with-gcs-access>"

mongodb:
  auth:
    rootPassword: "<mongodb-root-password>"
    password: "<mongodb-password>"
vm_service:
  providers:
    gcp:
      enabled: true
      project_id: <gcp-project-id>
      network_tags:
      - <network-tag>
      zone: <gcp-zone>
      network: "<gcp-network>"
      subnetwork: "" # leave blank for auto-subnetting
      workloadIdentity: "<service-account-email-with-compute-access>"

pusher:
  secret: "<pusher-secret>"
postgresql:
  auth:
    postgresPassword: "<postgres-password>"
rabbitmq:
  auth:
    password: "<rabbitmq-password>"
    erlangCookie: "<rabbitmq-erlang-cookie>"
nomad:
  server:
    gossip:
      encryption:
        key: "<nomad-gossip-encryption-key>"
    rpc:
      mTLS:
        enabled: true
        CACertificate: "<nomad-mtls-base64-ca>"
        certificate: "<nomad-mtls-base64-cert>"
        privateKey: "<nomad-mtls-base64-key>"

All Helm values.yaml options

KeyTypeDefaultDescription

api_service.replicas

int

1

Number of replicas to deploy for the api-service deployment.

audit_log_service.replicas

int

1

Number of replicas to deploy for the audit-log-service deployment.

branch_service.replicas

int

1

Number of replicas to deploy for the branch-service deployment.

builds_service.replicas

int

1

Number of replicas to deploy for the builds-service deployment.

contexts_service.replicas

int

1

Number of replicas to deploy for the contexts-service deployment.

cron_service.replicas

int

1

Number of replicas to deploy for the cron-service deployment.

dispatcher.replicas

int

1

Number of replicas to deploy for the dispatcher deployment.

distributor.agent_base_url

string

"https://circleci-binary-releases.s3.amazonaws.com/circleci-agent"

location of the task-agent. When air-gapped, the task-agent will need to be hosted within the air gap and this value updated

distributor.launch_agent_base_url

string

"https://circleci-binary-releases.s3.amazonaws.com/circleci-launch-agent"

Location of the launch-agent. When air-gapped, the launch-agent will need to be hosted within the air gap and this value updated

distributor_cleaner.replicas

int

1

Number of replicas to deploy for the distributor-dispatcher deployment.

distributor_dispatcher.replicas

int

1

Number of replicas to deploy for the distributor-dispatcher deployment.

distributor_external.replicas

int

1

Number of replicas to deploy for the distributor-external deployment.

distributor_internal.replicas

int

1

Number of replicas to deploy for the distributor-internal deployment.

domain_service.replicas

int

1

Number of replicas to deploy for the domain-service deployment.

frontend.replicas

int

1

Number of replicas to deploy for the frontend deployment.

GitHub

object

{"clientId":"","clientSecret":"","enterprise":false,"fingerprint":null,"hostname":"ghe.example.com","scheme":"https","unsafeDisableWebhookSSLVerification":false}

VCS Configuration details (currently limited to GitHub Enterprise and GitHub.com)

github.clientId

string

""

Client ID for OAuth Login via GitHub (2 Options).
Option 1: Set the value here and CircleCI will create the secret automatically.
Option 2: Leave this blank, and create the secret yourself. CircleCI will assume it exists.

Create on by Navigating to Settings > Developer Settings > OAuth Apps. Your homepage should be set to {{ .Values.global.scheme }}://{{ .Values.global.domainName }} and callback should be {{ .Value.scheme }}://{{ .Values.global.domainName }}/auth/github.

github.clientSecret

string

""

Client Secret for OAuth Login via Github (2 Options).
Option 1: Set the value here and CircleCI will create the secret automatically.
Option 2: Leave this blank, and create the secret yourself. CircleCI will assume it exists. + Retrieved from the same location as specified in github.clientID.

github.enterprise

bool

false

Set to true for Github Enterprise and false for Github.com

github.fingerprint

string

nil

Required when it is not possible to directly ssh-keyscan a GitHub Enterprise instance. It is not possible to proxy ssh-keyscan.

github.hostname

string

"ghe.example.com"

Github hostname. Ignored on Github.com. This is the hostname of your Github Enterprise installation.

github.scheme

string

"https"

One of ‘http’ or ‘https’. Ignored on Github.com. Set to ‘http’ if your Github Enterprise installation is not using TLS.

github.unsafeDisableWebhookSSLVerification

bool

false

Disable SSL Verification in webhooks. This is not safe and shouldn’t be done in a production scenario. This is required if your Github installation does not trust the certificate authority that signed your Circle server certificates (e.g they were self signed).

global.container.org

string

""

The registry organization to pull all images from (if in use), defaults to none.

global.container.registry

string

"cciserver.azurecr.io"

The registry to pull all images from, defaults to “cciserver.azurecr.io”.

global.domainName

string

""

Domain name of your CircleCI install

global.imagePullSecrets[0].name

string

"regcred"

global.license

string

""

License for your CircleCI install

global.scheme

string

"https"

Scheme for your CircleCI install

global.tracing.collector_host

string

""

global.tracing.enabled

bool

false

global.tracing.sample_rate

float

1

insights_service.dailyCronHour

int

3

Defaults to 3AM local server time.

insights_service.hourlyCronMinute

int

35

Defaults to 35 minutes past the hour.

insights_service.isEnabled

bool

true

Whether or not to enable the insights-service deployment.

insights_service.replicas

int

1

Number of replicas to deploy for the insights-service deployment.

internal_zone

string

"server.circleci.internal"

keyset

object

{"encryption":"","signing":""}

Keysets (2 Options) used to encrypt and sign artifacts generated by CircleCI. You need these values to configure server.
Option 1: Set the values keyset.signing and keyset.encryption here and CircleCI will create the secret automatically.
Option 2: Leave this blank, and create the secret yourself. CircleCI will assume it exists.
The secret must be named ‘signing-keys’ and have the keys; signing-key, encryption-key.

keyset.encryption

string

""

Encryption Key: To generate an artifact ENCRYPTION key run: docker run circleci/server-keysets:latest generate encryption -a stdout

keyset.signing

string

""

Signing Key: To generate an artifact SIGNING key run: docker run circleci/server-keysets:latest generate signing -a stdout

kong.acme.email

string

"your-email@example.com"

kong.acme.enabled

bool

false

This setting will fetch and renew Let’s Encrypt certs for you. It defaults to false as this only works when there’s a valid DNS entry for your domain (and the app. sub domain) - so you will need to deploy with this turned off and set the DNS records first. You can then set this to true and run helm upgrade with the updated setting if you want.

kong.debug_level

string

"notice"

Debug level for Kong. Available levels: debug, info, warn, error, crit. Default is “notice”.

kong.replicas

int

1

kong.resources.limits.cpu

string

"3072m"

kong.resources.limits.memory

string

"3072Mi"

kong.resources.requests.cpu

string

"512m"

kong.resources.requests.memory

string

"512Mi"

kong.status_page

bool

false

Set to true for public health check page (kong) for loadbalancers to hit

legacy_notifier.replicas

int

1

Number of replicas to deploy for the legacy-notifier deployment.

mongodb.architecture

string

"standalone"

mongodb.auth.database

string

"admin"

mongodb.auth.existingSecret

string

""

mongodb.auth.mechanism

string

"SCRAM-SHA-1"

mongodb.auth.password

string

""

mongodb.auth.rootPassword

string

""

mongodb.auth.username

string

"root"

mongodb.fullnameOverride

string

"mongodb"

mongodb.hosts

string

"mongodb:27017"

MongoDB host. This can be a comma-separated list of multiple hosts for sharded instances.

mongodb.image.tag

string

"3.6.22-debian-9-r38"

mongodb.internal

bool

true

Set to false if you want to use an externalized MongoDB instance.

mongodb.labels.app

string

"mongodb"

mongodb.labels.layer

string

"data"

mongodb.options

string

""

mongodb.persistence.size

string

"8Gi"

mongodb.podAnnotations.”backup.velero.io/backup-volumes”

string

"datadir"

mongodb.podLabels.app

string

"mongodb"

mongodb.podLabels.layer

string

"data"

mongodb.ssl

bool

false

mongodb.tlsInsecure

bool

false

If using an SSL connection with custom CA or self-signed certs, set this to true

mongodb.useStatefulSet

bool

true

nginx.annotations.”service.beta.kubernetes.io/aws-load-balancer-cross-zone-load-balancing-enabled”

string

"true"

nginx.annotations.”service.beta.kubernetes.io/aws-load-balancer-type”

string

"nlb"

Use “nlb” for Network Load Balancer and “clb” for Classic Load Balancer see https://aws.amazon.com/elasticloadbalancing/features/ for feature comparison

nginx.aws_acm.enabled

bool

false

⚠️ WARNING: Enabling this will recreate frontend’s service which will recreate the load balancer. If you are updating your deployed settings, then you will need to route your frontend domain to the new loadbalancer. You will also need to add service.beta.kubernetes.io/aws-load-balancer-ssl-cert: <acm-arn> to the nginx.annotations block.

nginx.loadBalancerIp

string

""

Load Balancer IP: To use a static IP for the provisioned load balancer with GCP, set to a reserved static ipv4 address

nginx.private_load_balancers

bool

false

nginx.replicas

int

1

nginx.resources.limits.cpu

string

"3000m"

nginx.resources.limits.memory

string

"3072Mi"

nginx.resources.requests.cpu

string

"500m"

nginx.resources.requests.memory

string

"512Mi"

nomad.auto_scaler.aws.accessKey

string

""

AWS Authentication Config (3 Options).
Option 1: Set accessKey and secretKey here, and CircleCI will create the secret for you.
Option 2: Leave accessKey and secretKey blank, and create the secret yourself. CircleCI will assume it exists.
Option 3: Leave accessKey and secretKey blank, and set the irsaRole field (IAM roles for service accounts).

nomad.auto_scaler.aws.autoScalingGroup

string

"asg-name"

nomad.auto_scaler.aws.enabled

bool

false

nomad.auto_scaler.aws.irsaRole

string

""

nomad.auto_scaler.aws.region

string

"some-region"

nomad.auto_scaler.aws.secretKey

string

""

nomad.auto_scaler.enabled

bool

false

nomad.auto_scaler.gcp.enabled

bool

false

nomad.auto_scaler.gcp.mig_name

string

"some-managed-instance-group-name"

nomad.auto_scaler.gcp.project_id

string

"some-project"

nomad.auto_scaler.gcp.region

string

""

The GCP region where the Managed Instance Group resides. Providing this parameter indicates the MIG is regional. If set, do not provide a zone

nomad.auto_scaler.gcp.service_account

object

{"project_id":"... ...","type":"service_account"}

GCP Authentication Config (3 Options).
Option 1: Set service_account with the service account JSON (raw JSON, not a string), and CircleCI will create the secret for you.
Option 2: Leave the service_account field as its default, and create the secret yourself. CircleCI will assume it exists.
Option 3: Leave the service_account field as its default, and set the workloadIdentity field with a service account email to use workload identities.

nomad.auto_scaler.gcp.workloadIdentity

string

""

nomad.auto_scaler.gcp.zone

string

""

The GCP zone where the Managed Instance Group resides. Providing this parameter indicates the MIG is zonal. If set, do not provide a region

nomad.auto_scaler.image.repository

string

"hashicorp/nomad-autoscaler"

nomad.auto_scaler.scaling.max

int

5

nomad.auto_scaler.scaling.min

int

1

nomad.auto_scaler.scaling.node_drain_deadline

string

"5m"

nomad.buildAgentImage

string

"circleci/picard"

By default, Dockerhub is assumed to be the image registry unless otherwise specified eg: registry.example.com/organization/repository

nomad.clients

object

{}

nomad.clusterDomain

string

"cluster.local"

nomad.server.gossip.encryption.enabled

bool

true

nomad.server.pdb.enabled

bool

true

nomad.server.pdb.minAvailable

int

2

nomad.server.replicas

int

3

nomad.server.rpc.mTLS

object

{"CACertificate":"","certificate":"","enabled":false,"privateKey":""}

mTLS is strongly suggested for RPC communication. It encrypts traffic but also authenticates clients to ensure no unauthenticated clients can join the cluster as workers. Base64 encoded PEM encoded certificates are expected here.

nomad.server.rpc.mTLS.CACertificate

string

""

base64 encoded nomad mTLS certificate authority

nomad.server.rpc.mTLS.certificate

string

""

base64 encoded nomad mTLS certificate

nomad.server.rpc.mTLS.privateKey

string

""

base64 encoded nomad mTLS private key

nomad.server.service.unsafe_expose_api

bool

false

object_storage

object

{"bucketName":"","expireAfter":0,"gcs":{"enabled":false,"service_account":{"project_id":"... ...","type":"service_account"},"workloadIdentity":""},"s3":{"accessKey":"","enabled":false,"endpoint":"https://s3.us-east-1.amazonaws.com","irsaRole":"","secretKey":""}}

Object storage for build artifacts, audit logs, test results and more. One of object_storage.s3.enabled or object_storage.gcs.enabled must be true for the chart to function.

object_storage.expireAfter

int

0

number of days after which artifacts will expire

object_storage.gcs.service_account

object

{"project_id":"... ...","type":"service_account"}

GCP Storage (GCS) Authentication Config (3 Options).
Option 1: Set service_account with the service account JSON (raw JSON, not a string), and CircleCI will create the secret for you.
Option 2: Leave the service_account field as its default, and create the secret yourself. CircleCI will assume it exists.
Option 3: Leave the service_account field as its default, and set the workloadIdentity field with a service account email to use workload identities.

object_storage.s3

object

{"accessKey":"","enabled":false,"endpoint":"https://s3.us-east-1.amazonaws.com","irsaRole":"","secretKey":""}

S3 Configuration for Object Storage. Authentication methods: AWS Access/Secret Key, and IRSA Role

object_storage.s3.accessKey

string

""

AWS Authentication Config (3 Options).
Option 1: Set accessKey and secretKey here, and CircleCI will create the secret for you.
Option 2: Leave accessKey and secretKey blank, and create the secret yourself. CircleCI will assume it exists.
Option 3: Leave accessKey and secretKey blank, and set the irsaRole field (IAM roles for service accounts), also set region: “your-aws-region”.

object_storage.s3.endpoint

string

"https://s3.us-east-1.amazonaws.com"

API endpoint for S3. If in AWS us-west-2, for example, this would be the regional endpoint https://s3.us-west-2.amazonaws.com. If using S3 compatible storage, specify the API endpoint of your object storage server

orb_service.replicas

int

1

Number of replicas to deploy for the orb-service deployment.

output_processor.replicas

int

2

Number of replicas to deploy for the output-processor deployment.

permissions_service.replicas

int

1

Number of replicas to deploy for the permissions-service deployment.

policy_service.replicas

int

1

Number of replicas to deploy for the policy-service deployment.

policy_service_internal.replicas

int

1

Number of replicas to deploy for the policy-service deployment.

postgresql.auth.existingSecret

string

""

postgresql.auth.password

string

""

Use only when postgresql.internal is false, this is the password of your externalized postgres user Ignored if auth.existingSecret with key password is provided

postgresql.auth.postgresPassword

string

""

Use only when postgresql.internal is true. This is the password for the internal postgres instance. Ignored if auth.existingSecret with key postgres-password is provided.

postgresql.auth.username

string

""

Use only when postgresql.internal is false, then this is the username used to connect with your externalized postgres instance

postgresql.fullnameOverride

string

"postgresql"

postgresql.image.tag

string

"12.6.0"

postgresql.internal

bool

true

postgresql.postgresqlHost

string

"postgresql"

postgresql.postgresqlPort

int

5432

postgresql.primary.extendedConfiguration

string

"max_connections = 500\nshared_buffers = 300MB\n"

postgresql.primary.persistence.existingClaim

string

""

postgresql.primary.persistence.size

string

"8Gi"

postgresql.primary.podAnnotations.”backup.velero.io/backup-volumes”

string

"data"

prometheus.alertmanager.enabled

bool

false

prometheus.enabled

bool

false

prometheus.extraScrapeConfigs

string

"- job_name: 'telegraf-metrics'\n scheme: http\n metrics_path: /metrics\n static_configs:\n - targets:\n - \"telegraf:9273\"\n labels:\n service: telegraf\n"

prometheus.fullnameOverride

string

"prometheus"

prometheus.nodeExporter.fullnameOverride

string

"node-exporter"

prometheus.pushgateway.enabled

bool

false

prometheus.server.emptyDir.sizeLimit

string

"8Gi"

prometheus.server.fullnameOverride

string

"prometheus-server"

prometheus.server.persistentVolume.enabled

bool

false

proxy.enabled

bool

false

If false, all proxy settings are ignored

proxy.http

object

{"auth":{"enabled":false,"password":null,"username":null},"host":"proxy.example.com","port":3128}

Proxy for HTTP requests

proxy.https

object

{"auth":{"enabled":false,"password":null,"username":null},"host":"proxy.example.com","port":3128}

Proxy for HTTPS requests

proxy.no_proxy

list

[]

List of hostnames, IP CIDR blocks exempt from proxying. Loopback and intra-service traffic is never proxied.

pusher.key

string

"circle"

rabbitmq.auth.erlangCookie

string

""

Either Provide the password or secret name for existingErlangSecret

rabbitmq.auth.existingErlangSecret

string

""

Secret must contain a value for rabbitmq-erlang-cookie key

rabbitmq.auth.existingPasswordSecret

string

""

Must contain a value for rabbitmq-password key

rabbitmq.auth.password

string

""

Either Provide the password or secret name for existingPasswordSecret

rabbitmq.auth.username

string

"circle"

rabbitmq.fullnameOverride

string

"rabbitmq"

rabbitmq.image.tag

string

"3.11.16-debian-11-r0"

rabbitmq.podAnnotations.”backup.velero.io/backup-volumes”

string

"data"

rabbitmq.podLabels.app

string

"rabbitmq"

rabbitmq.podLabels.layer

string

"data"

rabbitmq.replicaCount

int

1

rabbitmq.statefulsetLabels.app

string

"rabbitmq"

rabbitmq.statefulsetLabels.layer

string

"data"

redis.cluster.enabled

bool

true

redis.cluster.slaveCount

int

1

redis.fullnameOverride

string

"redis"

redis.image.tag

string

"6.2.1-debian-10-r13"

redis.master.persistence.size

string

"8Gi"

To increase PVC size, follow this guide: https://circleci.com/docs/server/v4.2/operator/expanding-internal-database-volumes

redis.master.podAnnotations.”backup.velero.io/backup-volumes”

string

"redis-data"

redis.podLabels.app

string

"redis"

redis.podLabels.layer

string

"data"

redis.slave.persistence.size

string

"8Gi"

To increase PVC size, follow this guide: https://circleci.com/docs/server/v4.2/operator/expanding-internal-database-volumes

redis.slave.podAnnotations.”backup.velero.io/backup-volumes”

string

"redis-data"

redis.statefulset.labels.app

string

"redis"

redis.statefulset.labels.layer

string

"data"

redis.usePassword

bool

false

schedulerer.replicas

int

1

Number of replicas to deploy for the schedulerer deployment.

serveUnsafeArtifacts

bool

false

⚠️ WARNING: Changing this to true will serve HTML artifacts instead of downloading them. This can allow specially-crafted artifacts to gain control of users’ CircleCI accounts.

smtp

object

{"host":"smtp.example.com","notificationUser":"builds@circleci.com","password":"secret-smtp-passphrase","port":25,"tls":true,"user":"notification@example.com"}

Email notification settings

smtp.port

int

25

Outbound connections on port 25 are blocked on most cloud providers. Should you select this default port, be aware that your notifications may fail to send.

smtp.tls

bool

true

StartTLS is used to encrypt mail by default. Only disable this if you can otherwise guarantee the confidentiality of traffic.

soketi.replicas

int

1

Number of replicas to deploy for the soketi deployment.

telegraf.args[0]

string

"--config-directory"

telegraf.args[1]

string

"/etc/telegraf/telegraf.d"

telegraf.args[2]

string

"--watch-config"

telegraf.args[3]

string

"poll"

telegraf.config.agent.flush_interval

string

"60s"

telegraf.config.agent.interval

string

"30s"

telegraf.config.agent.omit_hostname

bool

true

telegraf.config.custom_config_file

string

""

telegraf.config.inputs

list

[{"statsd":{"datadog_extensions":true,"max_ttl":"12h","metric_separator":".","percentile_limit":1000,"percentiles":[50,95,99],"service_address":":8125"}}]

To specify additional config for your install, paste it under .Values.telegraf.config.custom_config_file. Do NOT modify inputs/outputs below.

telegraf.config.outputs[0].prometheus_client.listen

string

":9273"

telegraf.fullnameOverride

string

"telegraf"

telegraf.mountPoints[0].mountPath

string

"/etc/telegraf/telegraf.d"

telegraf.mountPoints[0].name

string

"telegraf-config"

telegraf.resources.limits.memory

string

"512Mi"

telegraf.resources.requests.cpu

string

"200m"

telegraf.resources.requests.memory

string

"256Mi"

telegraf.volumes[0].configMap.name

string

"telegraf-config"

telegraf.volumes[0].name

string

"telegraf-config"

test_results_service.replicas

int

1

Number of replicas to deploy for the test-results-service deployment.

tink

object

{"enabled":false,"keyset":""}

Tink Configuration: + Tink is given precedence over vault. If tink.enabled is true, vault will not be deployed. Tink or Vault must be set once at install and cannot be changed.
Option 1: Set the values tink.keyset here and CircleCI will create the secret automatically.
Option 2: Leave this blank, and create the secret yourself. CircleCI will assume it exists.
The secret must be named ‘tink’ and have the key; keyset. generate a keyset via: tinkey create-keyset --key-template XCHACHA20_POLY1305

tls.certificate

string

""

base64 encoded certificate, leave empty to use self-signed certificates

tls.certificates

list

[]

List of base64’d certificates that will be imported into the system

tls.import

list

[]

List of host:port from which to import certificates

tls.privateKey

string

""

base64 encoded private key, leave empty to use self-signed certificates

vault

object

{"internal":true,"podAnnotations":{"backup.velero.io/backup-volumes":"data"},"token":"","transitPath":"transit","url":"http://vault:8200"}

External Services configuration

vault.internal

bool

true

Disables this charts Internal Vault instance

vault.token

string

""

This token is required when internal: false.

vault.transitPath

string

"transit"

When internal: true, this value is used for the vault transit path.

vm_gc.replicas

int

1

Number of replicas to deploy for the vm-gc deployment.

vm_scaler.prescaled

list

[{"count":0,"cron":"","docker-engine":true,"image":"docker-default","type":"l1.medium"},{"count":0,"cron":"","docker-engine":false,"image":"default","type":"l1.medium"},{"count":0,"cron":"","docker-engine":false,"image":"docker","type":"l1.large"},{"count":0,"cron":"","docker-engine":false,"image":"windows-default","type":"windows.medium"}]

Configuration options for, and numbers of, prescaled instances.

vm_scaler.replicas

int

1

Number of replicas to deploy for the vm-scaler deployment.

vm_service.dlc_lifespan_days

int

3

Number of days to keep DLC volumes before pruning them.

vm_service.enabled

bool

true

vm_service.providers

object

{"ec2":{"accessKey":"","assignPublicIP":false,"enabled":false,"irsaRole":"","linuxAMI":"","region":"us-west-1","secretKey":"","securityGroupId":"sg-8asfas76","subnets":["subnet-abcd1234"],"tags":["key","value"],"windowsAMI":"ami-mywindowsami"},"gcp":{"assignPublicIP":true,"enabled":false,"linuxImage":"","network":"my-server-vpc","network_tags":["circleci-vm"],"project_id":"my-server-project","service_account":{"project_id":"... ...","type":"service_account"},"subnetwork":"my-server-vm-subnet","windowsImage":"","workloadIdentity":"","zone":"us-west2-a"}}

Provider configuration for the VM service.

vm_service.providers.ec2.accessKey

string

""

EC2 Authentication Config (3 Options).
Option 1: Set accessKey and secretKey here, and CircleCI will create the secret for you.
Option 2: Leave accessKey and secretKey blank, and create the secret yourself. CircleCI will assume it exists.
Option 3: Leave accessKey and secretKey blank, and set the irsaRole field (IAM roles for service accounts).

vm_service.providers.ec2.enabled

bool

false

Set to enable EC2 as a virtual machine provider

vm_service.providers.ec2.subnets

list

["subnet-abcd1234"]

Subnets must be in the same availability zone

vm_service.providers.ec2.tags

list

["key","value"]

List of tags to apply to all VMs; “key”,“value”,“foo”,“bar” will turn into “key”: “value”, “foo”: “bar”

vm_service.providers.gcp.enabled

bool

false

Set to enable GCP Compute as a VM provider

vm_service.providers.gcp.service_account

object

{"project_id":"... ...","type":"service_account"}

GCP Compute Authentication Config (3 Options).
Option 1: Set service_account with the service account JSON (raw JSON, not a string), and CircleCI will create the secret for you.
Option 2: Leave the service_account field as its default, and create the secret yourself. CircleCI will assume it exists.
Option 3: Leave the service_account field as its default, and set the workloadIdentityField with a service account email to use workload identities.

vm_service.providers.gcp.subnetwork

string

"my-server-vm-subnet"

Put an empty string here if you use auto-subnetting

vm_service.providers.gcp.windowsImage

string

""

If you don’t want or need windows executors, leave this blank. Otherwise, put the name of your windows executor image here. You will need to build this yourself using the image builder repo https://github.com/CircleCI-Public/circleci-server-windows-image-builder

vm_service.replicas

int

1

Number of replicas to deploy for the vm-service deployment.

web_ui.replicas

int

1

Number of replicas to deploy for the web-ui deployment.

web_ui_404.replicas

int

1

Number of replicas to deploy for the web-ui-404 deployment.

web_ui_insights.replicas

int

1

Number of replicas to deploy for the web-ui-insights deployment.

web_ui_onboarding.replicas

int

1

Number of replicas to deploy for the web-ui-onboarding deployment.

web_ui_org_settings.replicas

int

1

Number of replicas to deploy for the web-ui-org-settings deployment.

web_ui_project_settings.replicas

int

1

Number of replicas to deploy for the web-ui-project-settings deployment.

web_ui_server_admin.replicas

int

1

Number of replicas to deploy for the web-ui-server-admin deployment.

web_ui_user_settings.replicas

int

1

Number of replicas to deploy for the web-ui-user-settings deployment.

webhook_service.isEnabled

bool

true

webhook_service.replicas

int

1

Number of replicas to deploy for the webhook-service deployment.

workflows_conductor_event_consumer.replicas

int

1

Number of replicas to deploy for the workflows-conductor-event-consumer deployment.

workflows_conductor_grpc.replicas

int

1

Number of replicas to deploy for the workflows-conductor-grpc deployment.


Suggest an edit to this page

Make a contribution
Learn how to contribute