Annotations
The following are the available annotations for the injector. These annotations are organized into two sections: agent and vault. All of the annotations below change the configurations of the Vault Agent containers injected into the pod.
Agent annotations
Agent annotations change the Vault Agent containers templating configuration. For example, agent annotations allow users to define what secrets they want, how to render them, optional commands to run, etc.
vault.hashicorp.com/agent-inject
- configures whether injection is explicitly enabled or disabled for a pod. This should be set to atrue
orfalse
value. Defaults tofalse
.vault.hashicorp.com/agent-inject-status
- blocks further mutations by adding the valueinjected
to the pod after a successful mutation.vault.hashicorp.com/agent-configmap
- name of the configuration map where Vault Agent configuration file and templates can be found.vault.hashicorp.com/agent-image
- name of the Vault docker image to use. This value overrides the default image configured in the injector and is usually not needed. Defaults tohashicorp/vault:1.17.2
.vault.hashicorp.com/agent-init-first
- configures the pod to run the Vault Agent init container first iftrue
(last iffalse
). This is useful when other init containers need pre-populated secrets. This should be set to atrue
orfalse
value. Defaults tofalse
.vault.hashicorp.com/agent-inject-command
- configures Vault Agent to run a command after the template has been rendered. To map a command to a specific secret, use the same unique secret name:vault.hashicorp.com/agent-inject-command-SECRET-NAME
. For example, if a secret annotationvault.hashicorp.com/agent-inject-secret-foobar
is configured,vault.hashicorp.com/agent-inject-command-foobar
would map a command to that secret.vault.hashicorp.com/agent-inject-secret
- configures Vault Agent to retrieve the secrets from Vault required by the container. The name of the secret is any unique string aftervault.hashicorp.com/agent-inject-secret-
, such asvault.hashicorp.com/agent-inject-secret-foobar
. The value is the path in Vault where the secret is located.vault.hashicorp.com/agent-inject-template
- configures the template Vault Agent should use for rendering a secret. The name of the template is any unique string aftervault.hashicorp.com/agent-inject-template-
, such asvault.hashicorp.com/agent-inject-template-foobar
. This should map to the same unique value provided invault.hashicorp.com/agent-inject-secret-
. If not provided, a default generic template is used.vault.hashicorp.com/error-on-missing-key
- configures whether Vault Agent should exit with an error when accessing a struct or map field/key that does not exist. The name of the secret is the string aftervault.hashicorp.com/error-on-missing-key-
, and should map to the same unique value provided invault.hashicorp.com/agent-inject-secret-
. Defaults tofalse
. See Vault Agent Template Config for more details.vault.hashicorp.com/agent-inject-containers
- comma-separated list that specifies in which containers the secrets volume should be mounted. If not provided, the secrets volume will be mounted in all containers in the pod.vault.hashicorp.com/secret-volume-path
- configures where on the filesystem a secret will be rendered. To map a path to a specific secret, use the same unique secret name:vault.hashicorp.com/secret-volume-path-SECRET-NAME
. For example, if a secret annotationvault.hashicorp.com/agent-inject-secret-foobar
is configured,vault.hashicorp.com/secret-volume-path-foobar
would configure where that secret is rendered. If no secret name is provided, this sets the default for all rendered secrets in the pod.vault.hashicorp.com/agent-inject-file
- configures the filename and path in the secrets volume where a Vault secret will be written. This should be used withvault.hashicorp.com/secret-volume-path
, which mounts a memory volume to the specified path. Ifsecret-volume-path
is used, the path can be omitted from this value. To map a filename to a specific secret, use the same unique secret name:vault.hashicorp.com/agent-inject-file-SECRET-NAME
. For example, if a secret annotationvault.hashicorp.com/agent-inject-secret-foobar
is configured,vault.hashicorp.com/agent-inject-file-foobar
would configure the filename.vault.hashicorp.com/agent-inject-perms
- configures the permissions of the file to create in the secrets volume. The name of the secret is the string after "vault.hashicorp.com/agent-inject-perms-", and should map to the same unique value provided in "vault.hashicorp.com/agent-inject-secret-". The value is the octal permission, for example:0644
.vault.hashicorp.com/agent-inject-template-file
- configures the path and filename of the custom template to use. This should be used withvault.hashicorp.com/extra-secret
, which mounts a Kubernetes secret to/vault/custom
. To map a template file to a specific secret, use the same unique secret name:vault.hashicorp.com/agent-inject-template-file-SECRET-NAME
. For example, if a secret annotationvault.hashicorp.com/agent-inject-secret-foobar
is configured,vault.hashicorp.com/agent-inject-template-file-foobar
would configure the template file.vault.hashicorp.com/agent-inject-default-template
- configures the default template type for rendering secrets if no custom template is defined. Possible values includemap
andjson
. Defaults tomap
.vault.hashicorp.com/template-config-exit-on-retry-failure
- controls whether Vault Agent exits after it has exhausted its number of template retry attempts due to failures. Defaults totrue
. See Vault Agent Template Config for more details.vault.hashicorp.com/template-static-secret-render-interval
- If specified, configures how often Vault Agent Template should render non-leased secrets such as KV v2. See Vault Agent Template Config for more details.vault.hashicorp.com/template-max-connections-per-host
- If specified, limits the total number of connections that the Vault Agent templating engine can use for a particular Vault host. The connection limit includes all connections in the dialing, active, and idle states. See Vault Agent Template Config for more details.vault.hashicorp.com/agent-extra-secret
- mounts Kubernetes secret as a volume at/vault/custom
in the sidecar/init containers. Useful for custom Agent configs with auto-auth methods such as approle that require paths to secrets be present.vault.hashicorp.com/agent-inject-token
- configures Vault Agent to share the Vault token with other containers in the pod, in a file namedtoken
in the root of the secrets volume (i.e./vault/secrets/token
). This is helpful when other containers communicate directly with Vault but require auto-authentication provided by Vault Agent. This should be set to atrue
orfalse
value. Defaults tofalse
.vault.hashicorp.com/agent-limits-cpu
- configures the CPU limits on the Vault Agent containers. Defaults to500m
. Setting this to an empty string disables CPU limits.vault.hashicorp.com/agent-limits-mem
- configures the memory limits on the Vault Agent containers. Defaults to128Mi
. Setting this to an empty string disables memory limits.vault.hashicorp.com/agent-limits-ephemeral
- configures the ephemeral storage limit on the Vault Agent containers. Defaults to unset, which disables ephemeral storage limits. Also available as a command-line option (-ephemeral-storage-limit
) or environment variable (AGENT_INJECT_EPHEMERAL_LIMIT
) to set the default for all injected Agent containers. Note: Pod limits are equal to the sum of all container limits. Setting this limit without setting it for other containers will also affect the limits of other containers in the pod. See Kubernetes resources documentation for more details.vault.hashicorp.com/agent-requests-cpu
- configures the CPU requests on the Vault Agent containers. Defaults to250m
. Setting this to an empty string disables CPU requests.vault.hashicorp.com/agent-requests-mem
- configures the memory requests on the Vault Agent containers. Defaults to64Mi
. Setting this to an empty string disables memory requests.vault.hashicorp.com/agent-requests-ephemeral
- configures the ephemeral storage requests on the Vault Agent Containers. Defaults to unset, which disables ephemeral storage requests (and will default to the ephemeral limit if set). Also available as a command-line option (-ephemeral-storage-request
) or environment variable (AGENT_INJECT_EPHEMERAL_REQUEST
) to set the default for all injected Agent containers. Note: Pod requests are equal to the sum of all container requests. Setting this limit without setting it for other containers will also affect the requests of other containers in the pod. See Kubernetes resources documentation for more details.vault.hashicorp.com/agent-revoke-on-shutdown
- configures whether the sidecar will revoke it's own token before shutting down. This setting will only be applied to the Vault Agent sidecar container. This should be set to atrue
orfalse
value. Defaults tofalse
.vault.hashicorp.com/agent-revoke-grace
- configures the grace period, in seconds, for revoking it's own token before shutting down. This setting will only be applied to the Vault Agent sidecar container. Defaults to5s
.vault.hashicorp.com/agent-pre-populate
- configures whether an init container is included to pre-populate the shared memory volume with secrets prior to the containers starting. This should be set to atrue
orfalse
value. Defaults totrue
.vault.hashicorp.com/agent-pre-populate-only
- configures whether an init container is the only injected container. If true, no sidecar container will be injected at runtime of the pod. Enabling this option is recommended for workloads of typeCronJob
orJob
to ensure a clean pod termination.vault.hashicorp.com/preserve-secret-case
- configures Vault Agent to preserve the secret name case when creating the secret files. This should be set to atrue
orfalse
value. Defaults tofalse
.vault.hashicorp.com/agent-run-as-user
- sets the user (uid) to run Vault agent as. Also available as a command-line option (-run-as-user
) or environment variable (AGENT_INJECT_RUN_AS_USER
) for the injector. Defaults to 100.vault.hashicorp.com/agent-run-as-group
- sets the group (gid) to run Vault agent as. Also available as a command-line option (-run-as-group
) or environment variable (AGENT_INJECT_RUN_AS_GROUP
) for the injector. Defaults to 1000.vault.hashicorp.com/agent-set-security-context
- controls whetherSecurityContext
is set in injected containers. Also available as a command-line option (-set-security-context
) or environment variable (AGENT_INJECT_SET_SECURITY_CONTEXT
). Defaults totrue
.vault.hashicorp.com/agent-run-as-same-user
- run the injected Vault agent containers as the User (uid) of the first application container in the pod. RequiresSpec.Containers[0].SecurityContext.RunAsUser
to be set in the pod spec. Also available as a command-line option (-run-as-same-user
) or environment variable (AGENT_INJECT_RUN_AS_SAME_USER
). Defaults tofalse
.Note: If the first application container in the pod is running as root (uid 0), the
run-as-same-user
annotation will fail injection with an error.vault.hashicorp.com/agent-share-process-namespace
- sets shareProcessNamespace in the Pod spec where Vault Agent is injected. Defaults tofalse
.vault.hashicorp.com/agent-cache-enable
- configures Vault Agent to enable caching. In Vault 1.7+ this annotation will also enable a Vault Agent persistent cache. This persistent cache will be shared between the init and sidecar container to reuse tokens and leases retrieved by the init container. Defaults tofalse
.vault.hashicorp.com/agent-cache-use-auto-auth-token
- configures Vault Agent cache to authenticate on behalf of the requester. Set toforce
to enable. Disabled by default.vault.hashicorp.com/agent-cache-listener-port
- configures Vault Agent cache listening port. Defaults to8200
.vault.hashicorp.com/agent-copy-volume-mounts
- copies the mounts from the specified container and mounts them to the Vault Agent containers. The service account volume is ignored.vault.hashicorp.com/agent-service-account-token-volume-name
- the optional name of a projected volume containing a service account token for use with auto-auth against Vault's Kubernetes auth method. If the volume is mounted to another container in the deployment, the token volume will be mounted to the same location in the vault-agent containers. Otherwise it will be mounted at the default location of/var/run/secrets/vault.hashicorp.com/serviceaccount/
.vault.hashicorp.com/agent-enable-quit
- enable the/agent/v1/quit
endpoint on an injected agent. This option defaults to false, and if true will be set on the existing cache listener, or a new localhost listener with a basic cache stanza configured. The agent-cache-listener-port annotation can be used to change the port.vault.hashicorp.com/agent-telemetry
- specifies the telemetry configuration for the Vault Agent sidecar. The name of the config is any unique string aftervault.hashicorp.com/agent-telemetry-
, such asvault.hashicorp.com/agent-telemetry-prometheus_retention_time
. This annotation can be reused multiple times to configure multiple settings for the agent telemetry.vault.hashicorp.com/go-max-procs
- set theGOMAXPROCS
environment variable for injected agentsvault.hashicorp.com/agent-json-patch
- change the injected agent sidecar container using a JSON patch before it is created. This can be used to add, remove, or modify any attribute of the container. For example, setting this to[{"op": "replace", "path": "/name", "value": "different-name"}]
will update the agent container's name to bedifferent-name
instead of the defaultvault-agent
.vault.hashicorp.com/agent-init-json-patch
- same asvault.hashicorp.com/agent-json-patch
, except that the JSON patch will be applied to the injected init container instead.
Vault annotations
Vault annotations change how the Vault Agent containers communicate with Vault. For example, Vault's address, TLS certificates to use, client parameters such as timeouts, etc.
vault.hashicorp.com/auth-config
- configures additional parameters for the configured authentication method. The name of the config is any unique string aftervault.hashicorp.com/auth-config-
, such asvault.hashicorp.com/auth-config-role-id-file-path
. This annotation can be reused multiple times to configure multiple settings for the authentication method. Some authentication methods may require additional secrets and should be mounted via thevault.hashicorp.com/agent-extra-secret
annotation. For a list of valid authentication configurations, see the Vault Agent auto-auth documentation.vault.hashicorp.com/auth-path
- configures the authentication path for the Kubernetes auth method. Defaults toauth/kubernetes
.vault.hashicorp.com/auth-type
- configures the authentication type for Vault Agent. Defaults tokubernetes
. For a list of valid authentication methods, see the Vault Agent auto-auth documentation.vault.hashicorp.com/auth-min-backoff
- set the min_backoff option in the auto-auth config. Requires Vault 1.11+.vault.hashicorp.com/auth-max-backoff
- set the max_backoff option in the auto-auth configvault.hashicorp.com/agent-auto-auth-exit-on-err
- set the exit_on_err option in the auto-auth configvault.hashicorp.com/ca-cert
- path of the CA certificate used to verify Vault's TLS. This can also be set as the default for all injected Agents via theAGENT_INJECT_VAULT_CACERT_BYTES
environment variable which takes a PEM-encoded certificate or bundle.vault.hashicorp.com/ca-key
- path of the CA public key used to verify Vault's TLS.vault.hashicorp.com/client-cert
- path of the client certificate used when communicating with Vault via mTLS.vault.hashicorp.com/client-key
- path of the client public key used when communicating with Vault via mTLS.vault.hashicorp.com/client-max-retries
- configures number of Vault Agent retry attempts when certain errors are encountered. Defaults to 2, for 3 total attempts. Set this to0
or less to disable retrying. Error codes that are retried are 412 (client consistency requirement not satisfied) and all 5xx except for 501 (not implemented).vault.hashicorp.com/client-timeout
- configures the request timeout threshold, in seconds, of the Vault Agent when communicating with Vault. Defaults to60s
and accepts value types of60
,60s
or1m
.vault.hashicorp.com/log-level
- configures the verbosity of the Vault Agent log level. Default isinfo
.vault.hashicorp.com/log-format
- configures the log type for Vault Agent. Possible values arestandard
andjson
. Default isstandard
.vault.hashicorp.com/namespace
- configures the Vault Enterprise namespace to be used when requesting secrets from Vault. Also available as a command-line option (-vault-namespace
) or environment variable (AGENT_INJECT_VAULT_NAMESPACE
) to set the default namespace for all injected Agents.vault.hashicorp.com/proxy-address
- configures the HTTP proxy to use when connecting to a Vault server.vault.hashicorp.com/role
- configures the Vault role used by the Vault Agent auto-auth method. Required whenvault.hashicorp.com/agent-configmap
is not set.vault.hashicorp.com/service
- configures the Vault address for the injected Vault Agent to use. This value overrides the default Vault address configured in the injector, and may either be the address of a Vault service within the same Kubernetes cluster as the injector, or an external Vault URL.vault.hashicorp.com/tls-secret
- name of the Kubernetes secret containing TLS Client and CA certificates and keys. This is mounted to/vault/tls
.vault.hashicorp.com/tls-server-name
- name of the Vault server to verify the authenticity of the server when communicating with Vault over TLS.vault.hashicorp.com/tls-skip-verify
- if true, configures the Vault Agent to skip verification of Vault's TLS certificate. It's not recommended to set this value to true in a production environment.vault.hashicorp.com/agent-disable-idle-connections
- Comma-separated list of Vault Agent features where idle connections should be disabled. Also available as a command-line option (-disable-idle-connections
) or environment variable (AGENT_INJECT_DISABLE_IDLE_CONNECTIONS
) to set the default for all injected Agents.vault.hashicorp.com/agent-disable-keep-alives
- Comma-separated list of Vault Agent features where keep-alives should be disabled. Also available as a command-line option (-disable-keep-alives
) or environment variable (AGENT_INJECT_DISABLE_KEEP_ALIVES
) to set the default for all injected Agents.