gcloud alpha compute network-endpoint-groups create

NAME: gcloud alpha compute network-endpoint-groups create - create a Google Compute Engine network endpoint group
SYNOPSIS: gcloud alpha compute network-endpoint-groups create NAME [--default-port=DEFAULT_PORT] [--network=NETWORK] [--network-endpoint-type=NETWORK_ENDPOINT_TYPE; default="gce-vm-ip-port"] [--producer-port=PRODUCER_PORT] [--psc-target-service=PSC_TARGET_SERVICE] [--subnet=SUBNET] [--cloud-function-name=CLOUD_FUNCTION_NAME --cloud-function-url-mask=CLOUD_FUNCTION_URL_MASK | --cloud-run-service=CLOUD_RUN_SERVICE --cloud-run-tag=CLOUD_RUN_TAG --cloud-run-url-mask=CLOUD_RUN_URL_MASK | --[no-]app-engine-app --app-engine-service=APP_ENGINE_SERVICE --app-engine-url-mask=APP_ENGINE_URL_MASK --app-engine-version=APP_ENGINE_VERSION | --serverless-deployment-platform=SERVERLESS_DEPLOYMENT_PLATFORM --serverless-deployment-resource=SERVERLESS_DEPLOYMENT_RESOURCE --serverless-deployment-url-mask=SERVERLESS_DEPLOYMENT_URL_MASK --serverless-deployment-version=SERVERLESS_DEPLOYMENT_VERSION] [--global | --region=REGION | --zone=ZONE] [GCLOUD_WIDE_FLAG …]
DESCRIPTION: (ALPHA) Create a Google Compute Engine network endpoint group.
EXAMPLES: To create a network endpoint group:
gcloud alpha compute network-endpoint-groups create my-neg --zone=us-central1-a --network=my-network --subnet=my-subnetwork
POSITIONAL ARGUMENTS: NAME

Name of the network endpoint group to operate on.
FLAGS: --default-port=DEFAULT_PORT

The default port to use if the port number is not specified in the network endpoint.
If this flag isn't specified for a NEG with endpoint type gce-vm-ip-port, gce-vm-ip-portmap or non-gcp-private-ip-port, then every network endpoint in the network endpoint group must have a port specified. For a global NEG with endpoint type internet-ip-port and internet-fqdn-port if the default port is not specified, the well-known port for your backend protocol is used (80 for HTTP, 443 for HTTPS).
This flag is not supported for NEGs with endpoint type serverless.
This flag is not supported for NEGs with endpoint type private-service-connect.

--network=NETWORK

Name of the network in which the NEG is created. default project network is used if unspecified.
This is only supported for NEGs with endpoint type gce-vm-ip-port, non-gcp-private-ip-port, gce-vm-ip, private-service-connect, internet-ip-port, internet-fqdn-port, or gce-vm-ip-portmap.
For Private Service Connect NEGs, you can optionally specify --network and --subnet if --psc-target-service points to a published service. If --psc-target-service points to the regional service endpoint of a Google API, do not specify --network or --subnet.

--network-endpoint-type=NETWORK_ENDPOINT_TYPE; default="gce-vm-ip-port"

Determines the spec of endpoints attached to this group.

gce-vm-ip-port

Endpoint IP address must belong to a VM in Compute Engine (either the primary IP or as part of an aliased IP range). The --default-port must be specified or every network endpoint in the network endpoint group must have a port specified.

internet-ip-port

Endpoint IP address must be a publicly routable address. If specified, the default port is used. If unspecified, the well-known port for your backend protocol is used as the default port (80 for HTTP, 443 for HTTPS).

internet-fqdn-port

Endpoint FQDN must be resolvable to a public IP address via public DNS. The default port is used, if specified. If the default port is not specified, the well-known port for your backend protocol is used as the default port (80 for HTTP, 443 for HTTPS).
After creating a NEG of this type, you can use the gcloud compute network-endpoint-groups update command with the --add-endpoint flag. Example: --add-endpoint="fqdn=backend.example.com,port=443"

non-gcp-private-ip-port

Endpoint IP address must belong to a VM not in Compute Engine and must be routable using a Cloud Router over VPN or an Interconnect connection. In this case, the NEG must be zonal. The --default-port must be specified or every network endpoint in the network endpoint group must have a port specified.

serverless

The network endpoint is handled by specified serverless infrastructure, such as Cloud Run, App Engine, or Cloud Function. Default port, network, and subnet are not effective for serverless endpoints.

private-service-connect

The network endpoint corresponds to a service outside the VPC, accessed via Private Service Connect.

gce-vm-ip

Endpoint must be the IP address of a VM's network interface in Compute Engine. Instance reference is required. The IP address is optional. If unspecified, the primary NIC address is used. A port must not be specified.

gce-vm-ip-portmap

Endpoint IP address must be a primary IP of a VM's network interface in Compute Engine. The --default-port must be specified or every network endpoint in the network endpoint group must have a port specified.
NETWORK_ENDPOINT_TYPE must be one of: gce-vm-ip-port, internet-ip-port, internet-fqdn-port, non-gcp-private-ip-port, serverless, gce-vm-ip, private-service-connect, gce-vm-ip-portmap.

--producer-port=PRODUCER_PORT

The producer port to use when a consumer PSC NEG connects to a producer's internal network load balancer. If this flag isn't specified for a NEG with endpoint type private-service-connect, the PSC NEG will connect to port 443 or the first available port in the PSC producer port range, or to port 1 if the PSC producer's forwarding rule ports flag is set to all-ports.
This flag is not supported for NEGs with endpoint type other than private-service-connect.

--psc-target-service=PSC_TARGET_SERVICE

PSC target service name to add to the private service connect network endpoint groups (NEG).

--subnet=SUBNET

Name of the subnet to which all network endpoints belong.
If not specified, network endpoints may belong to any subnetwork in the region where the network endpoint group is created.
This is only supported for NEGs with endpoint type gce-vm-ip-port, gce-vm-ip, private-service-connect, or gce-vm-ip-portmap. For Private Service Connect NEGs, you can optionally specify --network and --subnet if --psc-target-service points to a published service. If --psc-target-service points to the regional service endpoint of a Google API, do not specify --network or --subnet.

The serverless routing configurations are only valid when endpoint type of the network endpoint group is serverless. At most one of these can be specified:

Configuration for a Cloud Function network endpoint group. Cloud Function name must be provided explicitly or in the URL mask.

--cloud-function-name=CLOUD_FUNCTION_NAME

Cloud Function name to add to the Serverless NEG. The function must be in the same project and the same region as the Serverless network endpoint groups (NEG).

--cloud-function-url-mask=CLOUD_FUNCTION_URL_MASK

A template to parse function field from a request URL. URL mask allows for routing to multiple Cloud Functions without having to create multiple network endpoint groups and backend services.

Configuration for a Cloud Run network endpoint group. Cloud Run service must be provided explicitly or in the URL mask. Cloud Run tag is optional, and may be provided explicitly or in the URL mask.

--cloud-run-service=CLOUD_RUN_SERVICE

Cloud Run service name to add to the Serverless network endpoint groups (NEG). The service must be in the same project and the same region as the Serverless NEG.

--cloud-run-tag=CLOUD_RUN_TAG

Cloud Run tag represents the "named revision" to provide additional fine-grained traffic routing configuration.

--cloud-run-url-mask=CLOUD_RUN_URL_MASK

A template to parse service and tag fields from a request URL. URL mask allows for routing to multiple Run services without having to create multiple network endpoint groups and backend services.

Configuration for an App Engine network endpoint group. Both App Engine service and version are optional, and may be provided explicitly or in the URL mask. The app-engine-app flag is only used for default routing. The App Engine app must be in the same project as the Serverless network endpoint groups (NEG).

--[no-]app-engine-app

If set, the default routing is used. Use --app-engine-app to enable and --no-app-engine-app to disable.

--app-engine-service=APP_ENGINE_SERVICE

Optional serving service to add to the Serverless NEG.

--app-engine-url-mask=APP_ENGINE_URL_MASK

A template to parse service and version fields from a request URL. URL mask allows for routing to multiple App Engine services without having to create multiple network endpoint groups and backend services.

--app-engine-version=APP_ENGINE_VERSION

Optional serving version to add to the Serverless NEG.

Configuration for a Serverless network endpoint group. Serverless NEGs support all serverless backends and are the only way to setup a network endpoint group for Cloud API Gateways.
To create a serverless NEG with a Cloud Run, Cloud Functions or App Engine endpoint, you can either use the previously-listed Cloud Run, Cloud Functions or App Engine-specific properties, OR, you can use the following generic properties that are compatible with all serverless platforms, including API Gateway: serverless-deployment-platform, serverless-deployment-resource, serverless-deployment-url-mask, and serverless-deployment-version.

--serverless-deployment-platform=SERVERLESS_DEPLOYMENT_PLATFORM

The platform of the NEG backend target(s). Possible values:

API Gateway: apigateway.googleapis.com

App Engine: appengine.googleapis.com

Cloud Functions: cloudfunctions.googleapis.com

Cloud Run: run.googleapis.com

--serverless-deployment-resource=SERVERLESS_DEPLOYMENT_RESOURCE

The user-defined name of the workload/instance. This value must be provided explicitly or using the --serverless-deployment-url-mask option. The resource identified by this value is platform-specific and is as follows:

API Gateway: The gateway ID

App Engine: The service name

Cloud Functions: The function name

Cloud Run: The service name

--serverless-deployment-url-mask=SERVERLESS_DEPLOYMENT_URL_MASK

A template to parse platform-specific fields from a request URL. URL mask allows for routing to multiple resources on the same serverless platform without having to create multiple network endpoint groups and backend resources. The fields parsed by this template are platform-specific and are as follows:

API Gateway: The 'gateway' ID

App Engine: The 'service' and 'version'

Cloud Functions: The 'function' name

Cloud Run: The 'service' and 'tag'

--serverless-deployment-version=SERVERLESS_DEPLOYMENT_VERSION

The optional resource version. The version identified by this value is platform-specific and is as follows:

API Gateway: Unused

App Engine: The service version

Cloud Functions: Unused

Cloud Run: The service tag

At most one of these can be specified:

--global

If set, the network endpoint group is global.

--region=REGION

Region of the network endpoint group to operate on. If not specified, you might be prompted to select a region (interactive mode only).
To avoid prompting when this flag is omitted, you can set the compute/region property:
gcloud config set compute/region REGION

A list of regions can be fetched by running:
gcloud compute regions list

To unset the property, run:
gcloud config unset compute/region

Alternatively, the region can be stored in the environment variable CLOUDSDK_COMPUTE_REGION.

--zone=ZONE

Zone of the network endpoint group to operate on. If not specified and the compute/zone property isn't set, you might be prompted to select a zone (interactive mode only).
To avoid prompting when this flag is omitted, you can set the compute/zone property:
gcloud config set compute/zone ZONE

A list of zones can be fetched by running:
gcloud compute zones list

To unset the property, run:
gcloud config unset compute/zone

Alternatively, the zone can be stored in the environment variable CLOUDSDK_COMPUTE_ZONE.
GCLOUD WIDE FLAGS: 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.
NOTES: This command is currently in alpha and might change without notice. If this command fails with API permission errors despite specifying the correct project, you might be trying to access an API with an invitation-only early access allowlist. These variants are also available:
gcloud compute network-endpoint-groups create

gcloud beta compute network-endpoint-groups create