gcloud run deploy - create or update a Cloud Run service
gcloud run deploy [[SERVICE] --namespace=NAMESPACE] [--args=[ARG,...]] [--async] [--command=[COMMAND,...]] [--concurrency=CONCURRENCY] [--cpu=CPU] [--ingress=INGRESS; default="all"] [--max-instances=MAX_INSTANCES] [--memory=MEMORY] [--min-instances=MIN_INSTANCES] [--platform=PLATFORM; default="managed"] [--port=PORT] [--service-account=SERVICE_ACCOUNT] [--tag=TAG] [--timeout=TIMEOUT] [--no-traffic] [--[no-]use-http2] [--clear-env-vars | --env-vars-file=FILE_PATH | --set-env-vars=[KEY=VALUE,...] | --remove-env-vars=[KEY,...] --update-env-vars=[KEY=VALUE,...]] [--clear-labels | --remove-labels=[KEY,...] --labels=[KEY=VALUE,...] | --update-labels=[KEY=VALUE,...]] [--clear-secrets | --set-secrets=[KEY=VALUE,...] | --remove-secrets=[KEY,...] --update-secrets=[KEY=VALUE,...]] [--connectivity=CONNECTIVITY --clear-config-maps | --set-config-maps=[KEY=VALUE,...] | --remove-config-maps=[KEY,...] --update-config-maps=[KEY=VALUE,...]] [--image=IMAGE | --source=SOURCE] [--[no-]allow-unauthenticated --breakglass=JUSTIFICATION --clear-vpc-connector --[no-]cpu-throttling --description=DESCRIPTION --execution-environment=EXECUTION_ENVIRONMENT --revision-suffix=REVISION_SUFFIX --vpc-connector=VPC_CONNECTOR --vpc-egress=VPC_EGRESS --add-cloudsql-instances=[CLOUDSQL-INSTANCES,...] | --clear-cloudsql-instances | --remove-cloudsql-instances=[CLOUDSQL-INSTANCES,...] | --set-cloudsql-instances=[CLOUDSQL-INSTANCES,...] --binary-authorization=POLICY | --clear-binary-authorization --clear-key | --key=KEY --clear-post-key-revocation-action-type | --post-key-revocation-action-type=POST_KEY_REVOCATION_ACTION_TYPE] [--region=REGION | --cluster=CLUSTER --cluster-location=CLUSTER_LOCATION | --context=CONTEXT --kubeconfig=KUBECONFIG] [GCLOUD_WIDE_FLAG ...]
Creates or updates a Cloud Run service.
To deploy a container to the service my-backend on Cloud Run:
$ gcloud run deploy my-backend \ --image=us-docker.pkg.dev/project/image
You may also omit the service name. Then a prompt will be displayed with a suggested default value:
$ gcloud run deploy --image=us-docker.pkg.dev/project/image
To deploy to Cloud Run on Kubernetes Engine, you need to specify a cluster:
$ gcloud run deploy --image=us-docker.pkg.dev/project/image \ --cluster=my-cluster
- Service resource - Service to deploy to. The arguments in this group can be
used to specify the attributes of this resource.
- [SERVICE]
ID of the service or fully qualified identifier for the service. To set the service attribute:
provide the argument SERVICE on the command line;
specify the service name from an interactive prompt.
- --namespace=NAMESPACE
Specific to Cloud Run for Anthos: Kubernetes namespace for the service. To set the namespace attribute:
provide the argument SERVICE on the command line with a fully specified name;
specify the service name from an interactive prompt with a fully specified name;
provide the argument --namespace on the command line;
set the property run/namespace;
For Cloud Run on Kubernetes Engine, defaults to "default". Otherwise, defaults to project ID.;
provide the argument project on the command line;
set the property core/project.
- --args=[ARG,...]
Comma-separated arguments passed to the command run by the container image. If not specified and no '--command' is provided, the container image's default Cmd is used. Otherwise, if not specified, no arguments are passed. To reset this field to its default, pass an empty string.
- --async
Return immediately, without waiting for the operation in progress to complete.
- --command=[COMMAND,...]
Entrypoint for the container image. If not specified, the container image's default Entrypoint is run. To reset this field to its default, pass an empty string.
- --concurrency=CONCURRENCY
Set the maximum number of concurrent requests allowed per container instance. Leave concurrency unspecified or provide the special value 'default' to receive the server default value.
- --cpu=CPU
Set a CPU limit in Kubernetes cpu units.
Cloud Run (fully managed) supports values 1, 2 and 4. For Cloud Run (fully managed), 4 cpus also requires a minimum 2Gi --memory value. Examples 2, 2.0, 2000m
Cloud Run for Anthos and Knative-compatible Kubernetes clusters support fractional values. Examples .5, 500m, 2
- --ingress=INGRESS; default="all"
Set the ingress traffic sources allowed to call the service. For Cloud Run (fully managed) the --[no-]allow-unauthenticated flag separately controls the identities allowed to call the service. INGRESS must be one of:
- all
Inbound requests from all sources are allowed.
- internal
For Cloud Run (fully managed), only inbound requests from VPC networks in the same project or VPC Service Controls perimeter, as well as Pub/Sub subscriptions and Eventarc events in the same project or VPC Service Controls perimeter are allowed. All other requests are rejected. See https://cloud.google.com/run/docs/securing/ingress for full details on the definition of internal traffic for Cloud Run (fully managed). For Cloud Run for Anthos, only inbound requests from the same cluster are allowed.
- internal-and-cloud-load-balancing
Only supported for Cloud Run (fully managed). Only inbound requests from Google Cloud Load Balancing or a traffic source allowed by the internal option are allowed.
- --max-instances=MAX_INSTANCES
The maximum number of container instances of the Service to run. Use 'default' to unset the limit and use the platform default.
- --memory=MEMORY
Set a memory limit. Ex: 1024Mi, 4Gi.
- --min-instances=MIN_INSTANCES
The minimum number of container instances of the Service to run or 'default' to remove any minimum.
- --platform=PLATFORM; default="managed"
Target platform for running commands. Alternatively, set the property [run/platform]. PLATFORM must be one of:
- managed
Fully managed version of Cloud Run. Use with the --region flag or set the [run/region] property to specify a Cloud Run region.
- gke
Cloud Run for Anthos on Google Cloud. Use with the --cluster and --cluster-location flags or set the [run/cluster] and [run/cluster_location] properties to specify a cluster in a given zone.
- kubernetes
Use a Knative-compatible kubernetes cluster. Use with the --kubeconfig and --context flags to specify a kubeconfig file and the context for connecting.
- --port=PORT
Container port to receive requests at. Also sets the $PORT environment variable. Must be a number between 1 and 65535, inclusive. To unset this field, pass the special value "default".
- --service-account=SERVICE_ACCOUNT
Service account associated with the revision of the service. The service account represents the identity of the running revision, and determines what permissions the revision has. For the managed platform, this is the email address of an IAM service account. For the Kubernetes-based platforms (gke, kubernetes), this is the name of a Kubernetes service account in the same namespace as the service. If not provided, the revision will use the default service account of the project, or default Kubernetes namespace service account respectively.
- --tag=TAG
Traffic tag to assign to the newly created revision.
- --timeout=TIMEOUT
Set the maximum request execution time (timeout). It is specified as a duration; for example, "10m5s" is ten minutes, and five seconds. If you don't specify a unit, seconds is assumed. For example, "10" is 10 seconds.
- --no-traffic
True to avoid sending traffic to the revision being deployed. Setting this flag assigns any traffic assigned to the LATEST revision to the specific revision bound to LATEST before the deployment. The effect is that the revision being deployed will not receive traffic.
After a deployment with this flag the LATEST revision will not receive traffic on future deployments. To restore sending traffic to the LATEST revision by default, run the gcloud run services update-traffic command with --to-latest.
- --[no-]use-http2
Whether to use HTTP/2 for connections to the service. Use --use-http2 to enable and --no-use-http2 to disable.
- At most one of these can be specified:
- --clear-env-vars
Remove all environment variables.
- --env-vars-file=FILE_PATH
Path to a local YAML file with definitions for all environment variables. All existing environment variables will be removed before the new environment variables are added. Example YAML content:
KEY_1: "value1" KEY_2: "value 2"
- --set-env-vars=[KEY=VALUE,...]
List of key-value pairs to set as environment variables. All existing environment variables will be removed first.
- Only --update-env-vars and --remove-env-vars can be used together. If
both are specified, --remove-env-vars will be applied first.
- --remove-env-vars=[KEY,...]
List of environment variables to be removed.
- --update-env-vars=[KEY=VALUE,...]
List of key-value pairs to set as environment variables.
- At most one of these can be specified:
- --clear-labels
Remove all labels. If --update-labels is also specified then --clear-labels is applied first.
For example, to remove all labels:
$ gcloud run deploy --clear-labels
To remove all existing labels and create two new labels, foo and baz:
$ gcloud run deploy --clear-labels --update-labels foo=bar,baz=qux
- --remove-labels=[KEY,...]
List of label keys to remove. If a label does not exist it is silently ignored. If --update-labels is also specified then --update-labels is applied first.
- At most one of these can be specified:
- --labels=[KEY=VALUE,...]
List of label KEY=VALUE pairs to add.
An alias to --update-labels.
- --update-labels=[KEY=VALUE,...]
List of label KEY=VALUE pairs to update. If a label exists, its value is modified. Otherwise, a new label is created.
- Specify secrets to mount or provide as environment variables. Keys starting with
a forward slash '/' are mount paths. All other keys correspond to environment variables. Values should be in the form SECRET_NAME:SECRET_VERSION. For example: '--update-secrets=/secrets/api/key=mysecret:latest,ENV=othersecret:1' will mount a volume at '/secrets/api' containing a file 'key' with the latest version of secret 'mysecret'. An environment variable named ENV will also be created whose value is version 1 of secret 'othersecret'.
At most one of these can be specified:
- --clear-secrets
Remove all secrets.
- --set-secrets=[KEY=VALUE,...]
List of key-value pairs to set as secrets. All existing secrets will be removed first.
- Only --update-secrets and --remove-secrets can be used together. If both
are specified, --remove-secrets will be applied first.
- --remove-secrets=[KEY,...]
List of secrets to be removed.
- --update-secrets=[KEY=VALUE,...]
List of key-value pairs to set as secrets.
- Only applicable if connecting to Cloud Run for Anthos deployed on Google Cloud
or Cloud Run for Anthos deployed on VMware. Specify --platform=gke or --platform=kubernetes to use:
- --connectivity=CONNECTIVITY
(DEPRECATED) Defaults to 'external'. If 'external', the service can be invoked through the internet, in addition to through the cluster network.
The --connectivity flag is deprecated but will continue to be supported. Prefer to use the --ingress flag instead. CONNECTIVITY must be one of:
- external
Visible from outside the cluster.
- internal
Visible only within the cluster.
- Specify config map to mount or provide as environment variables. Keys starting
with a forward slash '/' are mount paths. All other keys correspond to environment variables. The values associated with each of these should be in the form CONFIG_MAP_NAME:KEY_IN_CONFIG_MAP; you may omit the key within the config map to specify a mount of all keys within the config map. For example: '--update-config-maps=/my/path=myconfig,ENV=otherconfig:key.json' will create a volume with config map 'myconfig' and mount that volume at '/my/path'. Because no config map key was specified, all keys in 'myconfig' will be included. An environment variable named ENV will also be created whose value is the value of 'key.json' in 'otherconfig. Not supported on the fully managed version of Cloud Run.
At most one of these can be specified:
- --clear-config-maps
Remove all config-maps.
- --set-config-maps=[KEY=VALUE,...]
List of key-value pairs to set as config-maps. All existing config-maps will be removed first.
- Only --update-config-maps and --remove-config-maps can be used together.
If both are specified, --remove-config-maps will be applied first.
- --remove-config-maps=[KEY,...]
List of config-maps to be removed.
- --update-config-maps=[KEY=VALUE,...]
List of key-value pairs to set as config-maps.
- At most one of these can be specified:
- --image=IMAGE
Name of the container image to deploy (e.g. gcr.io/cloudrun/hello:latest).
- --source=SOURCE
The location of the source to build. If a Dockerfile is present in the source code directory, it will be built using that Dockerfile, otherwise it will use Google Cloud buildpacks. See https://cloud.google.com/run/docs/deploying-source-code for more details. The location can be a directory on a local disk or a gzipped archive file (.tar.gz) in Google Cloud Storage. If the source is a local directory, this command skips the files specified in the --ignore-file. If --ignore-file is not specified, use.gcloudignore file. If a .gcloudignore file is absent and a .gitignore file is present in the local source directory, gcloud will use a generated Git-compatible .gcloudignore file that respects your .gitignored files. The global .gitignore is not respected. For more information on .gcloudignore, see gcloud topic gcloudignore.
- Only applicable if connecting to Cloud Run (fully managed). Specify
--platform=managed to use:
- --[no-]allow-unauthenticated
Whether to enable allowing unauthenticated access to the service. This may take a few moments to take effect. Use --allow-unauthenticated to enable and --no-allow-unauthenticated to disable.
- --breakglass=JUSTIFICATION
Justification to bypass Binary Authorization policy constraints and allow the operation. See https://cloud.google.com/binary-authorization/docs/using-breakglass for more information. Next update or deploy command will automatically clear existing breakglass justification.
- --clear-vpc-connector
Remove the VPC connector for this resource.
- --[no-]cpu-throttling
Whether to throttle the CPU when the container is not actively serving requests. Use --cpu-throttling to enable and --no-cpu-throttling to disable.
- --description=DESCRIPTION
Provides an optional, human-readable description of the service.
- --execution-environment=EXECUTION_ENVIRONMENT
Selects the execution environment where the application will run. EXECUTION_ENVIRONMENT must be one of:
- gen1
Run the application in a first generation execution environment.
- gen2
Run the application in a second generation execution environment.
- --revision-suffix=REVISION_SUFFIX
Specify the suffix of the revision name. Revision names always start with the service name automatically. For example, specifying [--revision-suffix=v1] for a service named 'helloworld', would lead to a revision named 'helloworld-v1'.
- --vpc-connector=VPC_CONNECTOR
Set a VPC connector for this resource.
- --vpc-egress=VPC_EGRESS
The outbound traffic to send through the VPC connector for this resource. This resource must have a VPC connector to set VPC egress. VPC_EGRESS must be one of:
- all
(DEPRECATED) Sends all outbound traffic through the VPC connector. Provides the same functionality as 'all-traffic'. Prefer to use 'all-traffic' instead.
- all-traffic
Sends all outbound traffic through the VPC connector.
- private-ranges-only
Default option. Sends outbound traffic to private IP addresses defined by RFC1918 through the VPC connector.
- These flags modify the Cloud SQL instances this Service connects to. You can
specify a name of a Cloud SQL instance if it's in the same project and region as your Cloud Run service; otherwise specify <project>:<region>:<instance> for the instance.
At most one of these can be specified:
- --add-cloudsql-instances=[CLOUDSQL-INSTANCES,...]
Append the given values to the current Cloud SQL instances.
- --clear-cloudsql-instances
Empty the current Cloud SQL instances.
- --remove-cloudsql-instances=[CLOUDSQL-INSTANCES,...]
Remove the given values from the current Cloud SQL instances.
- --set-cloudsql-instances=[CLOUDSQL-INSTANCES,...]
Completely replace the current Cloud SQL instances with the given values.
- At most one of these can be specified:
- --binary-authorization=POLICY
Binary Authorization policy to check against. This must be set to "default".
- --clear-binary-authorization
Remove any previously set Binary Authorization policy.
- At most one of these can be specified:
- --clear-key
Remove any previously set CMEK key reference.
- --key=KEY
CMEK key reference to encrypt the container with.
- At most one of these can be specified:
- --clear-post-key-revocation-action-type
Remove any previously set post CMEK key revocation action type.
- --post-key-revocation-action-type=POST_KEY_REVOCATION_ACTION_TYPE
Action type after CMEK key revocation. POST_KEY_REVOCATION_ACTION_TYPE must be one of:
- prevent-new
No new instances will be started after CMEK key revocation.
- shut-down
No new instances will be started and the existing instances will be shut down after CMEK key revocation.
- Arguments to locate resources, depending on the platform used.
At most one of these can be specified:
- Only applicable if connecting to Cloud Run (fully managed). Specify
--platform=managed to use:
- --region=REGION
Region in which the resource can be found. Alternatively, set the property [run/region].
- Only applicable if connecting to Cloud Run for Anthos deployed on Google Cloud.
Specify --platform=gke to use:
- Cluster resource - Kubernetes Engine cluster to connect to. The arguments in
this group can be used to specify the attributes of this resource. (NOTE) Some attributes are not given arguments in this group but can be set in other ways. To set the project attribute:
- —
provide the argument --cluster on the command line with a fully specified name;
- —
set the property run/cluster with a fully specified name;
- —
specify the cluster from a list of available clusters with a fully specified name;
- —
provide the argument --project on the command line;
- —
set the property core/project.
- --cluster=CLUSTER
ID of the cluster or fully qualified identifier for the cluster. To set the cluster attribute:
provide the argument --cluster on the command line;
set the property run/cluster;
specify the cluster from a list of available clusters.
- --cluster-location=CLUSTER_LOCATION
Zone in which the cluster is located. Alternatively, set the property [run/cluster_location]. To set the location attribute:
provide the argument --cluster on the command line with a fully specified name;
set the property run/cluster with a fully specified name;
specify the cluster from a list of available clusters with a fully specified name;
provide the argument --cluster-location on the command line;
set the property run/cluster_location;
specify the cluster location from a list of available zones.
- Only applicable if connecting to Cloud Run for Anthos deployed on VMware.
Specify --platform=kubernetes to use:
- --context=CONTEXT
The name of the context in your kubectl config file to use for connecting.
- --kubeconfig=KUBECONFIG
The absolute path to your kubectl config file. If not specified, the colon- or semicolon-delimited list of paths specified by $KUBECONFIG will be used. If $KUBECONFIG is unset, this defaults to ~/.kube/config.
These flags are available to all commands: --access-token-file, --account, --billing-project, --configuration, --flags-file, --flatten, --format, --help, --impersonate-service-account, --log-http, --project, --quiet, --trace-token, --user-output-enabled, --verbosity.
Run $ gcloud help for details.
These variants are also available:
$ gcloud alpha run deploy
$ gcloud beta run deploy