imjasonh/gcloud-help
1
0
Fork 0
mirror of https://github.com/imjasonh/gcloud-help synced 2026-07-10 19:31:55 +00:00
Code Activity
gcloud-help/gcloud/beta/container/clusters/create
Automated 39dc808917 gcloud: Wed May 13 13:12:23 UTC 2026
2026-05-13 13:12:23 +00:00

3066 lines
144 KiB
Text
Raw Blame History

NAME
gcloud beta container clusters create - create a cluster for running
containers
SYNOPSIS
gcloud beta container clusters create NAME
[--accelerator=[type=TYPE,[count=COUNT,
gpu-driver-version=GPU_DRIVER_VERSION,
gpu-partition-size=GPU_PARTITION_SIZE,
gpu-sharing-strategy=GPU_SHARING_STRATEGY,
max-shared-clients-per-gpu=MAX_SHARED_CLIENTS_PER_GPU],...]]
[--addons=[ADDON[=ENABLED|DISABLED],...]] [--allow-route-overlap]
[--alpha-cluster-feature-gates=[FEATURE=true|false,...]]
[--anonymous-authentication-config=ANONYMOUS_AUTHENTICATION_CONFIG]
[--async] [--auto-monitoring-scope=AUTO_MONITORING_SCOPE]
[--autopilot-general-profile=AUTOPILOT_GENERAL_PROFILE]
[--autopilot-privileged-admission=[ALLOWLIST_PATHS,...]]
[--autopilot-workload-policies=WORKLOAD_POLICIES]
[--autoprovisioning-enable-insecure-kubelet-readonly-port]
[--autoprovisioning-network-tags=TAGS,[TAGS,...]]
[--autoprovisioning-resource-manager-tags=[KEY=VALUE,...]]
[--autoscaling-profile=AUTOSCALING_PROFILE]
[--boot-disk-kms-key=BOOT_DISK_KMS_KEY]
[--cloud-run-config=[load-balancer-type=EXTERNAL,...]]
[--cluster-ipv4-cidr=CLUSTER_IPV4_CIDR]
[--cluster-secondary-range-name=NAME]
[--cluster-version=CLUSTER_VERSION]
[--confidential-node-type=CONFIDENTIAL_NODE_TYPE]
[--containerd-config-from-file=PATH_TO_FILE]
[--control-plane-egress=CONTROL_PLANE_EGRESS]
[--create-subnetwork=[KEY=VALUE,...]]
[--data-cache-count=DATA_CACHE_COUNT]
[--database-encryption-key=DATABASE_ENCRYPTION_KEY]
[--default-max-pods-per-node=DEFAULT_MAX_PODS_PER_NODE]
[--disable-default-snat] [--disable-l4-lb-firewall-reconciliation]
[--disable-multi-nic-lustre] [--disable-pod-cidr-overprovision]
[--disk-size=DISK_SIZE] [--disk-type=DISK_TYPE]
[--enable-agent-sandbox] [--enable-ambient-networking]
[--enable-authorized-networks-on-private-endpoint] [--enable-auto-ipam]
[--enable-autorepair] [--no-enable-autoupgrade]
[--enable-cilium-clusterwide-network-policy] [--enable-cloud-logging]
[--enable-cloud-monitoring] [--enable-cloud-run-alpha]
[--enable-confidential-nodes] [--enable-confidential-storage]
[--enable-cost-allocation] [--enable-dataplane-v2]
[--enable-default-compute-class] [--enable-dns-access] [--enable-fleet]
[--enable-fqdn-network-policy] [--enable-gke-oidc]
[--enable-google-cloud-access] [--enable-gvnic]
[--enable-identity-service] [--enable-image-streaming]
[--enable-insecure-kubelet-readonly-port]
[--enable-intra-node-visibility] [--enable-ip-access]
[--enable-ip-alias] [--enable-k8s-certs-via-dns]
[--enable-k8s-tokens-via-dns]
[--enable-kernel-module-signature-enforcement]
[--enable-kubernetes-alpha]
[--enable-kubernetes-unstable-apis=API,[API,...]]
[--enable-l4-ilb-subsetting] [--enable-legacy-authorization]
[--enable-legacy-lustre-port] [--enable-logging-monitoring-system-only]
[--enable-managed-mldiagnostics] [--enable-managed-prometheus]
[--enable-master-global-access] [--enable-multi-networking]
[--enable-nested-virtualization] [--enable-network-policy]
[--enable-pod-security-policy] [--enable-ray-cluster-logging]
[--enable-ray-cluster-monitoring] [--enable-scheduled-upgrades]
[--enable-service-externalips] [--enable-shielded-nodes]
[--enable-stackdriver-kubernetes] [--enable-vertical-pod-autoscaling]
[--fleet-project=PROJECT_ID_OR_NUMBER] [--gateway-api=GATEWAY_API]
[--hpa-profile=HPA_PROFILE] [--identity-provider=IDENTITY_PROVIDER]
[--image-type=IMAGE_TYPE]
[--in-transit-encryption=IN_TRANSIT_ENCRYPTION]
[--ipv6-access-type=IPV6_ACCESS_TYPE] [--issue-client-certificate]
[--istio-config=[auth=MTLS_PERMISSIVE,...]] [--labels=[KEY=VALUE,...]]
[--local-ssd-encryption-mode=LOCAL_SSD_ENCRYPTION_MODE]
[--logging=[COMPONENT,...]] [--logging-variant=LOGGING_VARIANT]
[--machine-type=MACHINE_TYPE, -m MACHINE_TYPE]
[--managed-otel-scope=MANAGED_OTEL_SCOPE]
[--max-nodes-per-pool=MAX_NODES_PER_POOL]
[--max-pods-per-node=MAX_PODS_PER_NODE]
[--max-surge-upgrade=MAX_SURGE_UPGRADE; default=1]
[--max-unavailable-upgrade=MAX_UNAVAILABLE_UPGRADE]
[--membership-type=MEMBERSHIP_TYPE]
[--metadata=KEY=VALUE,[KEY=VALUE,...]]
[--metadata-from-file=KEY=LOCAL_FILE_PATH,[...]]
[--min-cpu-platform=PLATFORM] [--monitoring=[COMPONENT,...]]
[--network=NETWORK]
[--network-performance-configs=[PROPERTY1=VALUE1,...]]
[--node-architecture-taint-behavior=NODE_ARCHITECTURE_TAINT_BEHAVIOR]
[--node-labels=[NODE_LABEL,...]]
[--node-pool-upgrade-concurrency-config=[max-count=MAX-COUNT]]
[--node-taints=[NODE_TAINT,...]] [--node-version=NODE_VERSION]
[--notification-config=[pubsub=ENABLED|DISABLED,
pubsub-topic=TOPIC,...]] [--num-nodes=NUM_NODES; default=3]
[--patch-update=[PATCH_UPDATE]]
[--performance-monitoring-unit=PERFORMANCE_MONITORING_UNIT]
[--placement-policy=PLACEMENT_POLICY] [--placement-type=PLACEMENT_TYPE]
[--preemptible] [--private-endpoint-subnetwork=NAME]
[--private-ipv6-google-access-type=PRIVATE_IPV6_GOOGLE_ACCESS_TYPE]
[--release-channel=CHANNEL] [--resource-manager-tags=[KEY=VALUE,...]]
[--security-group=SECURITY_GROUP] [--security-posture=SECURITY_POSTURE]
[--services-ipv4-cidr=CIDR] [--services-secondary-range-name=NAME]
[--shielded-integrity-monitoring] [--shielded-secure-boot] [--spot]
[--stack-type=STACK_TYPE] [--storage-pools=STORAGE_POOL,[...]]
[--subnetwork=SUBNETWORK] [--system-config-from-file=PATH_TO_FILE]
[--tags=TAG,[TAG,...]] [--threads-per-core=THREADS_PER_CORE]
[--tier=TIER] [--workload-metadata=WORKLOAD_METADATA]
[--workload-pool=WORKLOAD_POOL]
[--workload-vulnerability-scanning=WORKLOAD_VULNERABILITY_SCANNING]
[--additional-zones=ZONE,[ZONE,...] | --node-locations=ZONE,[ZONE,...]]
[--aggregation-ca=CA_POOL_PATH --cluster-ca=CA_POOL_PATH
--control-plane-disk-encryption-key=KEY --etcd-api-ca=CA_POOL_PATH
--etcd-peer-ca=CA_POOL_PATH --gkeops-etcd-backup-encryption-key=KEY
--service-account-signing-keys=KEY_VERSION,[KEY_VERSION,...]
--service-account-verification-keys=KEY_VERSION,[KEY_VERSION,...]]
[--binauthz-policy-bindings=[name=BINAUTHZ_POLICY]
--binauthz-evaluation-mode=BINAUTHZ_EVALUATION_MODE
| --enable-binauthz]
[--boot-disk-provisioned-iops=BOOT_DISK_PROVISIONED_IOPS
--boot-disk-provisioned-throughput=BOOT_DISK_PROVISIONED_THROUGHPUT]
[--cluster-dns=CLUSTER_DNS --cluster-dns-domain=CLUSTER_DNS_DOMAIN
--cluster-dns-scope=CLUSTER_DNS_SCOPE
--additive-vpc-scope-dns-domain=ADDITIVE_VPC_SCOPE_DNS_DOMAIN
| --disable-additive-vpc-scope]
[--dataplane-v2-observability-mode=DATAPLANE_V2_OBSERVABILITY_MODE
| --disable-dataplane-v2-flow-observability
| --enable-dataplane-v2-flow-observability]
[--disable-dataplane-v2-metrics | --enable-dataplane-v2-metrics]
[--disable-pod-snapshots | --enable-pod-snapshots]
[--enable-autoprovisioning --autoprovisioning-config-file=PATH_TO_FILE
| --autoprovisioning-image-type=AUTOPROVISIONING_IMAGE_TYPE
--autoprovisioning-locations=ZONE,[ZONE,...]
--autoprovisioning-min-cpu-platform=PLATFORM --max-cpu=MAX_CPU
--max-memory=MAX_MEMORY --min-cpu=MIN_CPU --min-memory=MIN_MEMORY
--autoprovisioning-max-surge-upgrade=AUTOPROVISIONING_MAX_SURGE_UPGRADE --autoprovisioning-max-unavailable-upgrade=AUTOPROVISIONING_MAX_UNAVAILABLE_UPGRADE --autoprovisioning-node-pool-soak-duration=AUTOPROVISIONING_NODE_POOL_SOAK_DURATION --autoprovisioning-standard-rollout-policy=[batch-node-count=BATCH_NODE_COUNT,
batch-percent=BATCH_NODE_PERCENTAGE,
batch-soak-duration=BATCH_SOAK_DURATION,...]
--enable-autoprovisioning-blue-green-upgrade
| --enable-autoprovisioning-surge-upgrade
--autoprovisioning-scopes=[SCOPE,...]
--autoprovisioning-service-account=AUTOPROVISIONING_SERVICE_ACCOUNT
--enable-autoprovisioning-autorepair
--enable-autoprovisioning-autoupgrade
[--max-accelerator=[type=TYPE,count=COUNT,...]
: --min-accelerator=[type=TYPE,count=COUNT,...]]]
[--enable-autoscaling --location-policy=LOCATION_POLICY
--max-nodes=MAX_NODES --min-nodes=MIN_NODES
--total-max-nodes=TOTAL_MAX_NODES --total-min-nodes=TOTAL_MIN_NODES]
[--enable-insecure-binding-system-authenticated
--enable-insecure-binding-system-unauthenticated]
[--enable-master-authorized-networks
--master-authorized-networks=NETWORK,[NETWORK,...]]
[--enable-network-egress-metering
--enable-resource-consumption-metering
--resource-usage-bigquery-dataset=RESOURCE_USAGE_BIGQUERY_DATASET]
[--enable-private-endpoint --enable-private-nodes
--master-ipv4-cidr=MASTER_IPV4_CIDR --private-cluster]
[--enable-secret-manager --enable-secret-manager-rotation
--secret-manager-rotation-interval=SECRET_MANAGER_ROTATION_INTERVAL]
[--enable-secret-sync --enable-secret-sync-rotation
--secret-sync-rotation-interval=SECRET_SYNC_ROTATION_INTERVAL]
[--ephemeral-storage[=[local-ssd-count=LOCAL-SSD-COUNT]]
| --ephemeral-storage-local-ssd[=[count=COUNT]]
| --local-nvme-ssd-block[=[count=COUNT]]
| --local-ssd-count=LOCAL_SSD_COUNT]
[--location=LOCATION | --region=REGION | --zone=ZONE, -z ZONE]
[--maintenance-minor-version-disruption-interval=MAINTENANCE_MINOR_VERSION_DISRUPTION_INTERVAL --maintenance-patch-version-disruption-interval=MAINTENANCE_PATCH_VERSION_DISRUPTION_INTERVAL]
[--maintenance-window=START_TIME | --maintenance-window-end=TIME_STAMP
--maintenance-window-recurrence=RRULE
--maintenance-window-start=TIME_STAMP]
[--password=PASSWORD --enable-basic-auth
| --username=USERNAME, -u USERNAME]
[--reservation=RESERVATION --reservation-affinity=RESERVATION_AFFINITY]
[--scopes=[SCOPE,...];
default="gke-default" --service-account=SERVICE_ACCOUNT]
[GCLOUD_WIDE_FLAG ...]
DESCRIPTION
(BETA) Create a cluster for running containers.
EXAMPLES
To create a cluster with the default configuration, run:
$ gcloud beta container clusters create sample-cluster
POSITIONAL ARGUMENTS
NAME
The name of the cluster to create.
The name may contain only lowercase alphanumerics and '-', must start
with a letter and end with an alphanumeric, and must be no longer than
40 characters.
FLAGS
--accelerator=[type=TYPE,[count=COUNT,gpu-driver-version=GPU_DRIVER_VERSION,gpu-partition-size=GPU_PARTITION_SIZE,gpu-sharing-strategy=GPU_SHARING_STRATEGY,max-shared-clients-per-gpu=MAX_SHARED_CLIENTS_PER_GPU],...]
Attaches accelerators (e.g. GPUs) to all nodes.
type
(Required) The specific type (e.g. nvidia-tesla-t4 for NVIDIA T4)
of accelerator to attach to the instances. Use gcloud compute
accelerator-types list to learn about all available accelerator
types.
count
(Optional) The number of accelerators to attach to the instances.
The default value is 1.
gpu-driver-version
(Optional) The NVIDIA driver version to install. GPU_DRIVER_VERSION
must be one of:
`default`: Install the default driver version for this GKE version. For GKE version 1.30.1-gke.1156000 and later, this is the default option.
`latest`: Install the latest driver version available for this GKE version.
Can only be used for nodes that use Container-Optimized OS.
`disabled`: Skip automatic driver installation. You must manually install a
driver after you create the cluster. For GKE version 1.30.1-gke.1156000 and earlier, this is the default option.
To manually install the GPU driver, refer to https://cloud.google.com/kubernetes-engine/docs/how-to/gpus#installing_drivers.
gpu-partition-size
(Optional) The GPU partition size used when running multi-instance
GPUs. For information about multi-instance GPUs, refer to:
https://cloud.google.com/kubernetes-engine/docs/how-to/gpus-multi
gpu-sharing-strategy
(Optional) The GPU sharing strategy (e.g. time-sharing) to use. For
information about GPU sharing, refer to:
https://cloud.google.com/kubernetes-engine/docs/concepts/timesharing-gpus
max-shared-clients-per-gpu
(Optional) The max number of containers allowed to share each GPU
on the node. This field is used together with gpu-sharing-strategy.
--addons=[ADDON[=ENABLED|DISABLED],...]
Addons
(https://cloud.google.com/kubernetes-engine/docs/reference/rest/v1/projects.locations.clusters#Cluster.AddonsConfig)
are additional Kubernetes cluster components. Addons specified by this
flag will be enabled. The others will be disabled. Default addons:
HttpLoadBalancing, HorizontalPodAutoscaling. The Istio addon is
deprecated and removed. For more information and migration, see
https://cloud.google.com/istio/docs/istio-on-gke/migrate-to-anthos-service-mesh.
ADDON must be one of: HttpLoadBalancing, HorizontalPodAutoscaling,
KubernetesDashboard, NetworkPolicy, NodeLocalDNS, ConfigConnector,
GcePersistentDiskCsiDriver, GcpFilestoreCsiDriver, BackupRestore,
GcsFuseCsiDriver, ParallelstoreCsiDriver, HighScaleCheckpointing,
LustreCsiDriver, RayOperator, SlurmOperator, NodeReadinessController,
Istio, CloudRun.
--allow-route-overlap
Allows the provided cluster CIDRs to overlap with existing routes that
are less specific and do not terminate at a VM.
When enabled, --cluster-ipv4-cidr must be fully specified (e.g.
10.96.0.0/14 , but not /14). If --enable-ip-alias is also specified,
both --cluster-ipv4-cidr and --services-ipv4-cidr must be fully
specified.
Must be used in conjunction with '--enable-ip-alias' or
'--no-enable-ip-alias'.
--alpha-cluster-feature-gates=[FEATURE=true|false,...]
Selectively enable or disable Kubernetes alpha and beta
kubernetesfeature gates on alpha GKE cluster. Alpha clusters are not
covered by the Kubernetes Engine SLA and should not be used for
production workloads.
--anonymous-authentication-config=ANONYMOUS_AUTHENTICATION_CONFIG
Enable or restrict anonymous access to the cluster. When enabled,
anonymous users will be authenticated as system:anonymous with the
group system:unauthenticated. Limiting access restricts anonymous
access to only the health check endpoints /readyz, /livez, and
/healthz.
ANONYMOUS_AUTHENTICATION_CONFIG must be one of:
ENABLED
'ENABLED' enables anonymous calls.
LIMITED
'LIMITED' restricts anonymous access to the cluster. Only calls to
the health check endpoints are allowed anonymously, all other calls
will be rejected.
--async
Return immediately, without waiting for the operation in progress to
complete.
--auto-monitoring-scope=AUTO_MONITORING_SCOPE
Enables Auto-Monitoring for a specific scope within the cluster. ALL:
Enables Auto-Monitoring for all supported workloads within the cluster.
NONE: Disables Auto-Monitoring. AUTO_MONITORING_SCOPE must be one of:
ALL, NONE.
--autopilot-general-profile=AUTOPILOT_GENERAL_PROFILE
Sets the Autopilot general profile for the cluster; possible values are
none and no-performance. If none is used, the cluster will use the
Autopilot default configuration. AUTOPILOT_GENERAL_PROFILE must be one
of: none, no-performance.
--autopilot-privileged-admission=[ALLOWLIST_PATHS,...]
Specifies which privileged workload allowlist paths can be referenced
and installed by AllowlistSynchronizers in Autopilot modes.
The value is a comma-separated list of paths in the format:
◆ gke://<partner_name>/<app_name>/<allowlist_path> for Autopilot
partner allowlists
◆ gs://<bucket_name>/<allowlist_path> for user allowlists
By default, all GKE-managed allowlists (gke://*) are authorized. See
https://cloud.google.com/kubernetes-engine/docs/resources/autopilot-partners
for all supported Autopilot partner allowlists. When setting this flag,
be careful to explicitly specify gke://* in addition to other entries
if you rely on this default behavior.
Wildcards (*) are supported. For example, if gke://* is authorized,
then AllowlistSynchronizers can be used to install
gke://partner1/allowlist1.yaml and gke://partner2/allowlist2.yaml.
Note: Use of user allowlists (gs://) requires special permissions and
is only available to a subset of high tier customers. Please contact
your account team for more information.
Examples:
Allow all GKE-managed allowlists (default behavior):
$ gcloud beta container clusters create \
--autopilot-privileged-admission=gke://*
Authorize only allowlists from a GKE Autopilot partner:
$ gcloud beta container clusters create \
--autopilot-privileged-admission=gke://my-partner/*
Authorize only a singular user-owned allowlist
$ gcloud beta container clusters create \
--autopilot-privileged-admission=gs://my-bucket/allowlists/\
my-allowlist.yaml
Authorize all user-owned allowlists under a given path:
$ gcloud beta container clusters create \
--autopilot-privileged-admission=gs://my-bucket/*
Authorize all GKE-managed allowlists and a specific user-owned
allowlist:
$ gcloud beta container clusters create \
--autopilot-privileged-admission=gke://*,gs://my-bucket/\
allowlists/my-allowlist.yaml
Disable allowlist installation entirely:
$ gcloud beta container clusters create \
--autopilot-privileged-admission=""
Exercise caution when using this flag on an existing cluster. Upon
updates, existing AllowlistSynchronizers will uninstall allowlists that
are no longer authorized.
For instructions on installing allowlists in the cluster after
authorization, please refer to:
https://cloud.google.com/kubernetes-engine/docs/how-to/run-autopilot-partner-workloads
--autopilot-workload-policies=WORKLOAD_POLICIES
Add Autopilot workload policies to the cluster.
Examples:
$ gcloud beta container clusters create example-cluster \
--autopilot-workload-policies=allow-net-admin
The only supported workload policy is 'allow-net-admin'.
--autoprovisioning-enable-insecure-kubelet-readonly-port
Enables the Kubelet's insecure read only port for Autoprovisioned Node
Pools.
If not set, the value from nodePoolDefaults.nodeConfigDefaults will be
used.
To disable the readonly port
--no-autoprovisioning-enable-insecure-kubelet-readonly-port.
--autoprovisioning-network-tags=TAGS,[TAGS,...]
Applies the given Compute Engine tags (comma separated) on all nodes in
the auto-provisioned node pools of the new Standard cluster or the new
Autopilot cluster.
Examples:
$ gcloud beta container clusters create example-cluster \
--autoprovisioning-network-tags=tag1,tag2
New nodes in auto-provisioned node pools, including ones created by
resize or recreate, will have these tags on the Compute Engine API
instance object and can be used in firewall rules. See
https://cloud.google.com/sdk/gcloud/reference/compute/firewall-rules/create
for examples.
--autoprovisioning-resource-manager-tags=[KEY=VALUE,...]
Applies the specified comma-separated resource manager tags that has
the GCE_FIREWALL purpose to all nodes in the new Autopilot cluster or
all auto-provisioned nodes in the new Standard cluster.
Examples:
$ gcloud beta container clusters create example-cluster \
--autoprovisioning-resource-manager-tags=tagKeys/\
1234=tagValues/2345
$ gcloud beta container clusters create example-cluster \
--autoprovisioning-resource-manager-tags=my-project/key1=value1
$ gcloud beta container clusters create example-cluster \
--autoprovisioning-resource-manager-tags=12345/key1=value1,\
23456/key2=value2
$ gcloud beta container clusters create example-cluster \
--autoprovisioning-resource-manager-tags=
All nodes in an Autopilot cluster or all auto-provisioned nodes in a
Standard cluster, including nodes that are resized or re-created, will
have the specified tags on the corresponding Instance object in the
Compute Engine API. You can reference these tags in network firewall
policy rules. For instructions, see
https://cloud.google.com/firewall/docs/use-tags-for-firewalls.
--autoscaling-profile=AUTOSCALING_PROFILE
Set autoscaling behaviour, choices are 'optimize-utilization' and
'balanced'. Default is 'balanced'.
--boot-disk-kms-key=BOOT_DISK_KMS_KEY
The Customer Managed Encryption Key used to encrypt the boot disk
attached to each node in the node pool. This should be of the form
projects/[KEY_PROJECT_ID]/locations/[LOCATION]/keyRings/[RING_NAME]/cryptoKeys/[KEY_NAME].
For more information about protecting resources with Cloud KMS Keys
please see:
https://cloud.google.com/compute/docs/disks/customer-managed-encryption
--cloud-run-config=[load-balancer-type=EXTERNAL,...]
Configurations for Cloud Run addon, requires --addons=CloudRun for
create and --update-addons=CloudRun=ENABLED for update.
load-balancer-type
(Optional) Type of load-balancer-type EXTERNAL or INTERNAL.
Examples:
$ gcloud beta container clusters create example-cluster \
--cloud-run-config=load-balancer-type=INTERNAL
--cluster-ipv4-cidr=CLUSTER_IPV4_CIDR
The IP address range for the pods in this cluster in CIDR notation
(e.g. 10.0.0.0/14). Prior to Kubernetes version 1.7.0 this must be a
subset of 10.0.0.0/8; however, starting with version 1.7.0 can be any
RFC 1918 IP range.
If you omit this option, a range is chosen automatically. The
automatically chosen range is randomly selected from 10.0.0.0/8 and
will not include IP address ranges allocated to VMs, existing routes,
or ranges allocated to other clusters. The automatically chosen range
might conflict with reserved IP addresses, dynamic routes, or routes
within VPCs that peer with this cluster. You should specify
--cluster-ipv4-cidr to prevent conflicts.
This field is not applicable in a Shared VPC setup where the IP address
range for the pods must be specified with
--cluster-secondary-range-name
--cluster-secondary-range-name=NAME
Set the secondary range to be used as the source for pod IPs. Alias
ranges will be allocated from this secondary range. NAME must be the
name of an existing secondary range in the cluster subnetwork.
Cannot be specified unless '--enable-ip-alias' option is also
specified. Cannot be used with '--create-subnetwork' option.
--cluster-version=CLUSTER_VERSION
The Kubernetes version to use for the master and nodes. Defaults to
server-specified.
The default Kubernetes version is available using the following
command.
$ gcloud container get-server-config
--confidential-node-type=CONFIDENTIAL_NODE_TYPE
Enable confidential nodes for the cluster. Enabling Confidential Nodes
will create nodes using Confidential VM
https://docs.cloud.google.com/compute/docs/about-confidential-vm.
CONFIDENTIAL_NODE_TYPE must be one of: sev, sev_snp, tdx.
--containerd-config-from-file=PATH_TO_FILE
Path of the YAML file that contains containerd configuration entries
like configuring access to private image registries.
For detailed information on the configuration usage, please refer to
https://cloud.google.com/kubernetes-engine/docs/how-to/customize-containerd-configuration.
Note: Updating the containerd configuration of an existing cluster or
node pool requires recreation of the existing nodes, which might cause
disruptions in running workloads.
Use a full or relative path to a local file containing the value of
containerd_config.
--control-plane-egress=CONTROL_PLANE_EGRESS
Configures the egress policy for the GKE control plane to control
outbound traffic from the kube-apiserver. CONTROL_PLANE_EGRESS must be
one of:
NONE
(Recommended) Provides maximum security. This mode removes the
control plane's public IP address and blocks all outbound traffic
from the kube-apiserver by default, preventing unexpected data
exfiltration. Webhooks that use clientConfig.url will be disabled.
Essential GKE-managed services are still permitted to function via
an internal allowlist.
VIA_CONTROL_PLANE
(Default) Maintains backward compatibility. The control plane
retains its public IP address and allows egress traffic from the
kube-apiserver.
--create-subnetwork=[KEY=VALUE,...]
Create a new subnetwork for the cluster. The name and range of the
subnetwork can be customized via optional 'name' and 'range' key-value
pairs.
'name' specifies the name of the subnetwork to be created.
'range' specifies the IP range for the new subnetwork. This can either
be a netmask size (e.g. '/20') or a CIDR range (e.g. '10.0.0.0/20'). If
a netmask size is specified, the IP is automatically taken from the
free space in the cluster's network.
Examples:
Create a new subnetwork with a default name and size.
$ gcloud beta container clusters create --create-subnetwork ""
Create a new subnetwork named "my-subnet" with netmask of size 21.
$ gcloud beta container clusters create \
--create-subnetwork name=my-subnet,range=/21
Create a new subnetwork with a default name with the primary range of
10.100.0.0/16.
$ gcloud beta container clusters create \
--create-subnetwork range=10.100.0.0/16
Create a new subnetwork with the name "my-subnet" with a default range.
$ gcloud beta container clusters create \
--create-subnetwork name=my-subnet
Cannot be specified unless '--enable-ip-alias' option is also
specified. Cannot be used in conjunction with '--subnetwork' option.
--data-cache-count=DATA_CACHE_COUNT
Specifies the number of local SSDs to be utilized for GKE Data Cache in
the cluster.
--database-encryption-key=DATABASE_ENCRYPTION_KEY
Enable Database Encryption.
Enable database encryption that will be used to encrypt Kubernetes
Secrets at the application layer. The key provided should be the
resource ID in the format of
projects/[KEY_PROJECT_ID]/locations/[LOCATION]/keyRings/[RING_NAME]/cryptoKeys/[KEY_NAME].
For more information, see
https://cloud.google.com/kubernetes-engine/docs/how-to/encrypting-secrets.
--default-max-pods-per-node=DEFAULT_MAX_PODS_PER_NODE
The default max number of pods per node for node pools in the cluster.
This flag sets the default max-pods-per-node for node pools in the
cluster. If --max-pods-per-node is not specified explicitly for a node
pool, this flag value will be used.
Must be used in conjunction with '--enable-ip-alias'.
--disable-default-snat
Disable default source NAT rules applied in cluster nodes.
By default, cluster nodes perform source network address translation
(SNAT) for packets sent from Pod IP address sources to destination IP
addresses that are not in the non-masquerade CIDRs list. For more
details about SNAT and IP masquerading, see:
https://cloud.google.com/kubernetes-engine/docs/how-to/ip-masquerade-agent#how_ipmasq_works
SNAT changes the packet's source IP address to the node's internal IP
address.
When this flag is set, GKE does not perform SNAT for packets sent to
any destination. You must set this flag if the cluster uses privately
reused public IPs.
The --disable-default-snat flag is only applicable to private GKE
clusters, which are inherently VPC-native. Thus, --disable-default-snat
requires that you also set --enable-ip-alias and
--enable-private-nodes.
--disable-l4-lb-firewall-reconciliation
Disable reconciliation on the cluster for L4 Load Balancer VPC
firewalls targeting ingress traffic.
--disable-multi-nic-lustre
Disable the Lustre CSI driver to automatically detect and configure all
suitable network interfaces on a node for Lustre IO.
--disable-pod-cidr-overprovision
Disables pod cidr overprovision on nodes. Pod cidr overprovisioning is
enabled by default.
--disk-size=DISK_SIZE
Size for node VM boot disks in GB. Defaults to 100GB.
--disk-type=DISK_TYPE
Type of the node VM boot disk. For version 1.24 and later, defaults to
pd-balanced. For versions earlier than 1.24, defaults to pd-standard.
DISK_TYPE must be one of: pd-standard, pd-ssd, pd-balanced,
hyperdisk-balanced, hyperdisk-extreme, hyperdisk-throughput.
--enable-agent-sandbox
Enable the Agent Sandbox feature on the cluster. Use
--no-enable-agent-sandbox to disable.
--enable-ambient-networking
Enable Ambient Networking on the cluster.
--enable-authorized-networks-on-private-endpoint
Enable enforcement of --master-authorized-networks CIDR ranges for
traffic reaching cluster's control plane via private IP.
--enable-auto-ipam
Enable the Auto IP Address Management (Auto IPAM) feature for the
cluster.
--enable-autorepair
Enable node autorepair feature for a cluster's default node pool(s).
$ gcloud beta container clusters create example-cluster \
--enable-autorepair
Node autorepair is enabled by default for clusters using COS,
COS_CONTAINERD, UBUNTU or UBUNTU_CONTAINERD as a base image, use
--no-enable-autorepair to disable.
See
https://cloud.google.com/kubernetes-engine/docs/how-to/node-auto-repair
for more info.
--enable-autoupgrade
Sets autoupgrade feature for a cluster's default node pool(s).
$ gcloud beta container clusters create example-cluster \
--enable-autoupgrade
See https://cloud.google.com/kubernetes-engine/docs/node-auto-upgrades
for more info.
Enabled by default, use --no-enable-autoupgrade to disable.
--enable-cilium-clusterwide-network-policy
Enable Cilium Clusterwide Network Policies on the cluster. Disabled by
default.
--enable-cloud-logging
(DEPRECATED) Automatically send logs from the cluster to the Google
Cloud Logging API.
Legacy Logging and Monitoring is deprecated. Thus, flag
--enable-cloud-logging is also deprecated and will be removed in an
upcoming release. Please use --logging (optionally with --monitoring).
For more details, please read:
https://cloud.google.com/kubernetes-engine/docs/concepts/about-logs and
https://cloud.google.com/kubernetes-engine/docs/how-to/configure-metrics.
--enable-cloud-monitoring
(DEPRECATED) Automatically send metrics from pods in the cluster to the
Google Cloud Monitoring API. VM metrics will be collected by Google
Compute Engine regardless of this setting.
Legacy Logging and Monitoring is deprecated. Thus, flag
--enable-cloud-monitoring is also deprecated. Please use --monitoring
(optionally with --logging). For more details, please read:
https://cloud.google.com/kubernetes-engine/docs/how-to/configure-metrics
and
https://cloud.google.com/kubernetes-engine/docs/concepts/about-logs.
--enable-cloud-run-alpha
Enable Cloud Run alpha features on this cluster. Selecting this option
will result in the cluster having all Cloud Run alpha API groups and
features turned on.
Cloud Run alpha clusters are not covered by the Cloud Run SLA and
should not be used for production workloads.
--enable-confidential-nodes
Enable confidential nodes for the cluster. Enabling Confidential Nodes
will create nodes using Confidential VM
https://docs.cloud.google.com/compute/docs/about-confidential-vm.
--enable-confidential-storage
Enable confidential storage for the cluster. Enabling Confidential
Storage will create boot disk with confidential mode
--enable-cost-allocation
Enable the cost management feature.
When enabled, you can get informational GKE cost breakdowns by cluster,
namespace and label in your billing data exported to BigQuery
(https://cloud.google.com/billing/docs/how-to/export-data-bigquery).
--enable-dataplane-v2
Enables the new eBPF dataplane for GKE clusters that is required for
network security, scalability and visibility features.
--enable-default-compute-class
Enable the default compute class to use for the cluster.
To disable Default Compute Class in an existing cluster, explicitly set
flag --no-enable-default-compute-class.
--enable-dns-access
Enable access to the cluster's control plane over DNS-based endpoint.
DNS-based control plane access is recommended.
--enable-fleet
Set cluster project as the fleet host project. This will register the
cluster to the same project. To register the cluster to a fleet in a
different project, please use --fleet-project=FLEET_HOST_PROJECT.
Example: $ gcloud beta container clusters create --enable-fleet
--enable-fqdn-network-policy
Enable FQDN Network Policies on the cluster. FQDN Network Policies are
disabled by default.
--enable-gke-oidc
(DEPRECATED) Enable GKE OIDC authentication on the cluster.
When enabled, users would be able to authenticate to Kubernetes cluster
after properly setting OIDC config.
GKE OIDC is by default disabled when creating a new cluster. To disable
GKE OIDC in an existing cluster, explicitly set flag
--no-enable-gke-oidc.
GKE OIDC is being replaced by Identity Service across Anthos and GKE.
Thus, flag --enable-gke-oidc is also deprecated. Please use
--enable-identity-service to enable the Identity Service component
--enable-google-cloud-access
When you enable Google Cloud Access, any public IP addresses owned by
Google Cloud can reach the public control plane endpoint of your
cluster.
--enable-gvnic
Enable the use of GVNIC for this cluster. Requires re-creation of nodes
using either a node-pool upgrade or node-pool creation.
--enable-identity-service
Enable Identity Service component on the cluster.
When enabled, users can authenticate to Kubernetes cluster with
external identity providers.
Identity Service is by default disabled when creating a new cluster. To
disable Identity Service in an existing cluster, explicitly set flag
--no-enable-identity-service.
--enable-image-streaming
Enable Image Streaming for the cluster, allowing nodes to stream
container image data from Artifact Registry on demand to reduce
container start times. This flag sets the default for new node pools.
It is enabled by default on Autopilot clusters.
See Image Streaming documentation
(https://cloud.google.com/kubernetes-engine/docs/how-to/image-streaming)
for full requirements (including version, API enablement and Artifact
Registry usage). To disable Image Streaming for the cluster, use
--no-enable-image-streaming.
--enable-insecure-kubelet-readonly-port
Enables the Kubelet's insecure read only port.
To disable the readonly port on a cluster or node-pool set the flag to
--no-enable-insecure-kubelet-readonly-port.
--enable-intra-node-visibility
Enable Intra-node visibility for this cluster.
Enabling intra-node visibility makes your intra-node pod-to-pod traffic
visible to the networking fabric. With this feature, you can use VPC
flow logging or other VPC features for intra-node traffic.
Enabling it on an existing cluster causes the cluster master and the
cluster nodes to restart, which might cause a disruption.
--enable-ip-access
Enable access to the cluster's control plane over private IP and public
IP if --enable-private-endpoint is not enabled.
--enable-ip-alias
--enable-ip-alias creates a VPC-native cluster. If you set this option,
you can optionally specify the IP address ranges to use for Pods and
Services. For instructions, see
https://cloud.google.com/kubernetes-engine/docs/how-to/alias-ips.
--no-enable-ip-alias creates a routes-based cluster. This type of
cluster routes traffic between Pods using Google Cloud Routes. This
option is not recommended; use the default VPC-native cluster type
instead. For instructions, see
https://cloud.google.com/kubernetes-engine/docs/how-to/routes-based-cluster
Note: For IPv6-only clusters, these flags are a no-op as IP Aliases do
not apply, and any specified IP address ranges for Pods and Services
will be ignored.
You can't specify both --enable-ip-alias and --no-enable-ip-alias. If
you omit both --enable-ip-alias and --no-enable-ip-alias, the default
is a VPC-native cluster.
--enable-k8s-certs-via-dns
Enable K8s client certificates Authentication to the cluster's control
plane over DNS-based endpoint.
--enable-k8s-tokens-via-dns
Enable K8s Service Account tokens Authentication to the cluster's
control plane over DNS-based endpoint.
--enable-kernel-module-signature-enforcement
Enforces that kernel modules are signed on all new nodes in the cluster
unless explicitly overridden with
--no-enable-kernel-module-signature-enforcement when creating the
nodepool. Use --no-enable-kernel-module-signature-enforcement to
disable.
Examples:
$ gcloud beta container clusters create example-cluster \
--enable-kernel-module-signature-enforcement
--enable-kubernetes-alpha
Enable Kubernetes alpha features on this cluster. Selecting this option
will result in the cluster having all Kubernetes alpha API groups and
features turned on. Cluster upgrades (both manual and automatic) will
be disabled and the cluster will be automatically deleted after 30
days.
Alpha clusters are not covered by the Kubernetes Engine SLA and should
not be used for production workloads.
--enable-kubernetes-unstable-apis=API,[API,...]
Enable Kubernetes beta API features on this cluster. Beta APIs are not
expected to be production ready and should be avoided in
production-grade environments.
--enable-l4-ilb-subsetting
Enable Subsetting for L4 ILB services created on this cluster.
--enable-legacy-authorization
Enables the legacy ABAC authentication for the cluster. User rights are
granted through the use of policies which combine attributes together.
For a detailed look at these properties and related formats, see
https://kubernetes.io/docs/admin/authorization/abac/. To use RBAC
permissions instead, create or update your cluster with the option
--no-enable-legacy-authorization.
--enable-legacy-lustre-port
Allow the Lustre CSI driver to initialize LNet (the virtual network
layer for Lustre kernel module) using port 6988. This flag is required
to workaround a port conflict with the gke-metadata-server on GKE
nodes.
--enable-logging-monitoring-system-only
(DEPRECATED) Enable Cloud Operations system-only monitoring and
logging.
The --enable-logging-monitoring-system-only flag is deprecated and will
be removed in an upcoming release. Please use --logging and
--monitoring instead. For more information, please read:
https://cloud.google.com/kubernetes-engine/docs/concepts/about-logs and
https://cloud.google.com/kubernetes-engine/docs/how-to/configure-metrics.
--enable-managed-mldiagnostics
Enables managed machine learning diagnostics in the cluster. Use
--no-enable-managed-mldiagnostics to disable. See
https://docs.cloud.google.com/tpu/docs/ml-diagnostics/gke for more
info.
--enable-managed-prometheus
Enables managed collection for Managed Service for Prometheus in the
cluster.
See
https://cloud.google.com/stackdriver/docs/managed-prometheus/setup-managed#enable-mgdcoll-gke
for more info.
Enabled by default for cluster versions 1.27 or greater, use
--no-enable-managed-prometheus to disable.
--enable-master-global-access
Use with private clusters to allow access to the master's private
endpoint from any Google Cloud region or on-premises environment
regardless of the private cluster's region.
--enable-multi-networking
Enables multi-networking on the cluster. Multi-networking is disabled
by default.
--enable-nested-virtualization
Enables the use of nested virtualization on the default initial node
pool. Defaults to false. Can only be enabled on UBUNTU_CONTAINERD base
image or COS_CONTAINERD base image with version 1.28.4-gke.1083000 and
above.
--enable-network-policy
Enable network policy enforcement for this cluster. If you are enabling
network policy on an existing cluster the network policy addon must
first be enabled on the master by using
--update-addons=NetworkPolicy=ENABLED flag.
--enable-pod-security-policy
Enables the pod security policy admission controller for the cluster.
The pod security policy admission controller adds fine-grained pod
create and update authorization controls through the PodSecurityPolicy
API objects. For more information, see
https://cloud.google.com/kubernetes-engine/docs/how-to/pod-security-policies.
--enable-ray-cluster-logging
Enable automatic log processing sidecar for Ray clusters.
--enable-ray-cluster-monitoring
Enable automatic metrics collection for Ray clusters.
--enable-scheduled-upgrades
Enables the scheduled upgrades feature for the cluster.
When enabled, automatic cluster control plane upgrades will only occur
when you receive notice at least 72 hours beforehand.
--enable-service-externalips
Enables use of services with externalIPs field.
--enable-shielded-nodes
Enable Shielded Nodes for this cluster. Enabling Shielded Nodes will
enable a more secure Node credential bootstrapping implementation.
Starting with version 1.18, clusters will have Shielded GKE nodes by
default.
--enable-stackdriver-kubernetes
(DEPRECATED) Enable Cloud Operations for GKE.
The --enable-stackdriver-kubernetes flag is deprecated and will be
removed in an upcoming release. Please use --logging and --monitoring
instead. For more information, please read:
https://cloud.google.com/kubernetes-engine/docs/concepts/about-logs and
https://cloud.google.com/kubernetes-engine/docs/how-to/configure-metrics.
Flags for vertical pod autoscaling:
At most one of these can be specified:
--enable-vertical-pod-autoscaling
Enable vertical pod autoscaling for a cluster.
--fleet-project=PROJECT_ID_OR_NUMBER
Sets fleet host project for the cluster. If specified, the current
cluster will be registered as a fleet membership under the fleet host
project.
Example: $ gcloud beta container clusters create
--fleet-project=my-project
--gateway-api=GATEWAY_API
Enables GKE Gateway controller in this cluster. The value of the flag
specifies which Open Source Gateway API release channel will be used to
define Gateway resources. GATEWAY_API must be one of:
disabled
Gateway controller will be disabled in the cluster.
standard
Gateway controller will be enabled in the cluster. Resource
definitions from the standard OSS Gateway API release channel will
be installed.
--hpa-profile=HPA_PROFILE
Set Horizontal Pod Autoscaler behavior. Accepted values are: none,
performance. For more information, see
https://cloud.google.com/kubernetes-engine/docs/how-to/horizontal-pod-autoscaling#hpa-profile.
--identity-provider=IDENTITY_PROVIDER
Enable 3P identity provider on the cluster.
--image-type=IMAGE_TYPE
The image type to use for the cluster. Defaults to server-specified.
Image Type specifies the base OS that the nodes in the cluster will run
on. If an image type is specified, that will be assigned to the cluster
and all future upgrades will use the specified image type. If it is not
specified the server will pick the default image type.
The default image type and the list of valid image types are available
using the following command.
$ gcloud container get-server-config
--in-transit-encryption=IN_TRANSIT_ENCRYPTION
Enable Dataplane V2 in-transit encryption. Dataplane v2 in-transit
encryption is disabled by default. IN_TRANSIT_ENCRYPTION must be one
of: inter-node-transparent, none.
--ipv6-access-type=IPV6_ACCESS_TYPE
IPv6 access type of the subnetwork. Defaults to 'external'.
IPV6_ACCESS_TYPE must be one of: external, internal.
--issue-client-certificate
Issue a TLS client certificate with admin permissions.
When enabled, the certificate and private key pair will be present in
MasterAuth field of the Cluster object. For cluster versions before
1.12, a client certificate will be issued by default. As of 1.12,
client certificates are disabled by default.
--istio-config=[auth=MTLS_PERMISSIVE,...]
(REMOVED) Configurations for Istio addon, requires --addons contains
Istio for create, or --update-addons Istio=ENABLED for update.
auth
(Optional) Type of auth MTLS_PERMISSIVE or MTLS_STRICT.
Examples:
$ gcloud beta container clusters create example-cluster \
--istio-config=auth=MTLS_PERMISSIVE
The --istio-config flag is no longer supported. For more
information and migration, see
https://cloud.google.com/istio/docs/istio-on-gke/migrate-to-anthos-service-mesh.
--labels=[KEY=VALUE,...]
Labels to apply to the Google Cloud resources in use by the Kubernetes
Engine cluster. These are unrelated to Kubernetes labels.
Examples:
$ gcloud beta container clusters create example-cluster \
--labels=label_a=value1,label_b=,label_c=value3
--local-ssd-encryption-mode=LOCAL_SSD_ENCRYPTION_MODE
Encryption mode for Local SSDs on the cluster.
LOCAL_SSD_ENCRYPTION_MODE must be one of: STANDARD_ENCRYPTION,
EPHEMERAL_KEY_ENCRYPTION.
--logging=[COMPONENT,...]
Set the components that have logging enabled. Valid component values
are: SYSTEM, WORKLOAD, API_SERVER, CONTROLLER_MANAGER, SCHEDULER,
KCP_HPA, NONE
For more information, see
https://cloud.google.com/kubernetes-engine/docs/concepts/about-logs#available-logs
Examples:
$ gcloud beta container clusters create --logging=SYSTEM
$ gcloud beta container clusters create \
--logging=SYSTEM,API_SERVER,WORKLOAD
$ gcloud beta container clusters create --logging=NONE
--logging-variant=LOGGING_VARIANT
Specifies the logging variant that will be deployed on all the nodes in
the cluster. Valid logging variants are MAX_THROUGHPUT, DEFAULT. If no
value is specified, DEFAULT is used. LOGGING_VARIANT must be one of:
DEFAULT
'DEFAULT' variant requests minimal resources but may not guarantee
high throughput.
MAX_THROUGHPUT
'MAX_THROUGHPUT' variant requests more node resources and is able
to achieve logging throughput up to 10MB per sec.
--machine-type=MACHINE_TYPE, -m MACHINE_TYPE
The type of machine to use for nodes. Defaults to e2-medium. The list
of predefined machine types is available using the following command:
$ gcloud compute machine-types list
You can also specify custom machine types by providing a string with
the format "custom-CPUS-RAM" where "CPUS" is the number of virtual CPUs
and "RAM" is the amount of RAM in MiB.
For example, to create a node pool using custom machines with 2 vCPUs
and 12 GB of RAM:
$ gcloud beta container clusters create high-mem-pool \
--machine-type=custom-2-12288
--managed-otel-scope=MANAGED_OTEL_SCOPE
Specifies the scope of Managed OpenTelemetry within the cluster.
COLLECTION_AND_INSTRUMENTATION_COMPONENTS: Enables GKE Managed
OpenTelemetry Collector and other Instrumentations components like
Instrumentation CRD within the cluster. NONE: Disables the GKE Managed
OpenTelemetry.
MANAGED_OTEL_SCOPE must be one of:
COLLECTION_AND_INSTRUMENTATION_COMPONENTS, NONE.
--max-nodes-per-pool=MAX_NODES_PER_POOL
The maximum number of nodes to allocate per default initial node pool.
Kubernetes Engine will automatically create enough nodes pools such
that each node pool contains less than --max-nodes-per-pool nodes.
Defaults to 1000 nodes, but can be set as low as 100 nodes per pool on
initial create.
--max-pods-per-node=MAX_PODS_PER_NODE
The max number of pods per node for this node pool.
This flag sets the maximum number of pods that can be run at the same
time on a node. This will override the value given with
--default-max-pods-per-node flag set at the cluster level.
Must be used in conjunction with '--enable-ip-alias'.
--max-surge-upgrade=MAX_SURGE_UPGRADE; default=1
Number of extra (surge) nodes to be created on each upgrade of a node
pool.
Specifies the number of extra (surge) nodes to be created during this
node pool's upgrades. For example, running the following command will
result in creating an extra node each time the node pool is upgraded:
$ gcloud beta container clusters create example-cluster \
--max-surge-upgrade=1 --max-unavailable-upgrade=0
Must be used in conjunction with '--max-unavailable-upgrade'.
--max-unavailable-upgrade=MAX_UNAVAILABLE_UPGRADE
Number of nodes that can be unavailable at the same time on each
upgrade of a node pool.
Specifies the number of nodes that can be unavailable at the same time
while this node pool is being upgraded. For example, running the
following command will result in having 3 nodes being upgraded in
parallel (1 + 2), but keeping always at least 3 (5 - 2) available each
time the node pool is upgraded:
$ gcloud beta container clusters create example-cluster \
--num-nodes=5 --max-surge-upgrade=1 \
--max-unavailable-upgrade=2
Must be used in conjunction with '--max-surge-upgrade'.
--membership-type=MEMBERSHIP_TYPE
Specify a membership type for the cluster's fleet membership. Example:
$ gcloud beta container clusters create \
--membership-type=LIGHTWEIGHT. MEMBERSHIP_TYPE must be (only \ one
value is supported):
LIGHTWEIGHT
Fleet membership representing this cluster will be lightweight.
--metadata=KEY=VALUE,[KEY=VALUE,...]
Compute Engine metadata to be made available to the guest operating
system running on nodes within the node pool.
Each metadata entry is a key/value pair separated by an equals sign.
Metadata keys must be unique and less than 128 bytes in length. Values
must be less than or equal to 32,768 bytes in length. The total size of
all keys and values must be less than 512 KB. Multiple arguments can be
passed to this flag. For example:
--metadata key-1=value-1,key-2=value-2,key-3=value-3
Additionally, the following keys are reserved for use by Kubernetes
Engine:
◆ cluster-location
◆ cluster-name
◆ cluster-uid
◆ configure-sh
◆ enable-os-login
◆ gci-update-strategy
◆ gci-ensure-gke-docker
◆ instance-template
◆ kube-env
◆ startup-script
◆ user-data
Google Kubernetes Engine sets the following keys by default:
◆ serial-port-logging-enable
See also Compute Engine's documentation
(https://cloud.google.com/compute/docs/storing-retrieving-metadata) on
storing and retrieving instance metadata.
--metadata-from-file=KEY=LOCAL_FILE_PATH,[...]
Same as --metadata except that the value for the entry will be read
from a local file.
--min-cpu-platform=PLATFORM
When specified, the nodes for the new cluster's default node pool will
be scheduled on host with specified CPU architecture or a newer one.
Examples:
$ gcloud beta container clusters create example-cluster \
--min-cpu-platform=PLATFORM
To list available CPU platforms in given zone, run:
$ gcloud beta compute zones describe ZONE \
--format="value(availableCpuPlatforms)"
CPU platform selection is available only in selected zones.
--monitoring=[COMPONENT,...]
Set the components that have monitoring enabled. Valid component values
are: SYSTEM, WORKLOAD (Deprecated), NONE, API_SERVER,
CONTROLLER_MANAGER, SCHEDULER, DAEMONSET, DEPLOYMENT, HPA, POD,
STATEFULSET, STORAGE, CADVISOR, KUBELET, DCGM, JOBSET
Note: DAEMONSET, DEPLOYMENT, HPA, POD, STATEFULSET, STORAGE, CADVISOR,
KUBELET, DCGM, and JOBSET require Google Managed Prometheus to be
enabled.
For more information, see
https://cloud.google.com/kubernetes-engine/docs/how-to/configure-metrics#available-metrics
Examples:
$ gcloud beta container clusters create \
--monitoring=SYSTEM,API_SERVER,POD
$ gcloud beta container clusters create --monitoring=NONE
--network=NETWORK
The Compute Engine Network that the cluster will connect to. Google
Kubernetes Engine will use this network when creating routes and
firewalls for the clusters. Defaults to the 'default' network.
--network-performance-configs=[PROPERTY1=VALUE1,...]
Configures network performance settings for the cluster. Node pools can
override with their own settings.
total-egress-bandwidth-tier
Total egress bandwidth is the available outbound bandwidth from a
VM, regardless of whether the traffic is going to internal IP or
external IP destinations. The following tier values are allowed:
[TIER_UNSPECIFIED,TIER_1].
See
https://cloud.google.com/compute/docs/networking/configure-vm-with-high-bandwidth-configuration
for more information.
--node-architecture-taint-behavior=NODE_ARCHITECTURE_TAINT_BEHAVIOR
Control how architecture taint should be applied to nodes in default
node pool(s) in new cluster.
Supported values:
* unspecified: Default behavior, currently the same as `arm`.
* arm: kubernetes.io/arch=arm:NoSchedule taint will be added for ARM nodes.
* none: No architecture taint will be applied.
Examples:
$ gcloud beta container clusters create example-cluster \
--node-architecture-taint-behavior=none
To read more about node-taints, see
https://cloud.google.com/kubernetes-engine/docs/node-taints.
NODE_ARCHITECTURE_TAINT_BEHAVIOR must be one of: unspecified, arm,
none.
--node-labels=[NODE_LABEL,...]
Applies the given Kubernetes labels on all nodes in the new node pool.
Examples:
$ gcloud beta container clusters create example-cluster \
--node-labels=label-a=value1,label-2=value2
Updating the node pool's --node-labels flag applies the labels to the
Kubernetes Node objects for existing nodes in-place; it does not
re-create or replace nodes. New nodes, including ones created by
resizing or re-creating nodes, will have these labels on the Kubernetes
API Node object. The labels can be used in the nodeSelector field. See
https://kubernetes.io/docs/concepts/scheduling-eviction/assign-pod-node/
for examples.
Note that Kubernetes labels, intended to associate cluster components
and resources with one another and manage resource lifecycles, are
different from Google Kubernetes Engine labels that are used for the
purpose of tracking billing and usage information.
--node-pool-upgrade-concurrency-config=[max-count=MAX-COUNT]
Configure the concurrency settings for node pool auto upgrades.
Examples:
$ gcloud beta container clusters create example-cluster \
--node-pool-upgrade-concurrency-config=max-count=10
max-count
The maximum number of node pools to upgrade concurrently. The
maximum allowed value is 100.
--node-taints=[NODE_TAINT,...]
Applies the given kubernetes taints on all nodes in default node
pool(s) in new cluster, which can be used with tolerations for pod
scheduling.
Examples:
$ gcloud beta container clusters create example-cluster \
--node-taints=key1=val1:NoSchedule,key2=val2:PreferNoSchedule
To read more about node-taints, see
https://cloud.google.com/kubernetes-engine/docs/node-taints.
--node-version=NODE_VERSION
The Kubernetes version to use for nodes. Defaults to server-specified.
The default Kubernetes version is available using the following
command.
$ gcloud container get-server-config
--notification-config=[pubsub=ENABLED|DISABLED,pubsub-topic=TOPIC,...]
The notification configuration of the cluster. GKE supports publishing
cluster upgrade notifications to any Pub/Sub topic you created in the
same project. Create a subscription for the topic specified to receive
notification messages. See https://cloud.google.com/pubsub/docs/admin
on how to manage Pub/Sub topics and subscriptions. You can also use the
filter option to specify which event types you'd like to receive from
the following options: SecurityBulletinEvent, UpgradeEvent,
UpgradeInfoEvent, UpgradeAvailableEvent.
Examples:
$ gcloud beta container clusters create example-cluster \
--notification-config=pubsub=ENABLED,pubsub-topic=projects/\
{project}/topics/{topic-name}
$ gcloud beta container clusters create example-cluster \
--notification-config=pubsub=ENABLED,pubsub-topic=projects/\
{project}/topics/{topic-name},\
filter="SecurityBulletinEvent|UpgradeEvent"
The project of the Pub/Sub topic must be the same one as the cluster.
It can be either the project ID or the project number.
--num-nodes=NUM_NODES; default=3
The number of nodes to be created in each of the cluster's zones.
--patch-update=[PATCH_UPDATE]
The patch update to use for the cluster.
Setting to 'accelerated' automatically upgrades the cluster to the
latest patch available within the cluster's current minor version and
release channel. Setting to 'default' automatically upgrades the
cluster to the default patch upgrade targetversion available within the
cluster's current minor version and release channel.
PATCH_UPDATE must be one of: accelerated, default.
--performance-monitoring-unit=PERFORMANCE_MONITORING_UNIT
Sets the Performance Monitoring Unit level. Valid values are
architectural, standard and enhanced. PERFORMANCE_MONITORING_UNIT must
be one of:
architectural
Enables architectural PMU events tied to non last level cache (LLC)
events.
enhanced
Enables most documented core/L2 and LLC PMU events.
standard
Enables most documented core/L2 PMU events.
--placement-policy=PLACEMENT_POLICY
Indicates the desired resource policy to use.
$ gcloud beta container clusters create node-pool-1 \
--cluster=example-cluster --placement-policy my-placement
--placement-type=PLACEMENT_TYPE
Placement type allows to define the type of node placement within the
default node pool of this cluster.
UNSPECIFIED - No requirements on the placement of nodes. This is the
default option.
COMPACT - GKE will attempt to place the nodes in a close proximity to
each other. This helps to reduce the communication latency between the
nodes, but imposes additional limitations on the node pool size.
$ gcloud beta container clusters create example-cluster \
--placement-type=COMPACT
PLACEMENT_TYPE must be one of: UNSPECIFIED, COMPACT.
--preemptible
Create nodes using preemptible VM instances in the new cluster.
$ gcloud beta container clusters create example-cluster --preemptible
New nodes, including ones created by resize or recreate, will use
preemptible VM instances. See
https://cloud.google.com/kubernetes-engine/docs/preemptible-vm for more
information on how to use Preemptible VMs with Kubernetes Engine.
--private-endpoint-subnetwork=NAME
Sets the subnetwork GKE uses to provision the control plane's private
endpoint.
--private-ipv6-google-access-type=PRIVATE_IPV6_GOOGLE_ACCESS_TYPE
Sets the type of private access to Google services over IPv6.
PRIVATE_IPV6_GOOGLE_ACCESS_TYPE must be one of:
bidirectional
Allows Google services to initiate connections to GKE pods in this
cluster. This is not intended for common use, and requires previous
integration with Google services.
disabled
Default value. Disables private access to Google services over IPv6.
outbound-only
Allows GKE pods to make fast, secure requests to Google services
over IPv6. This is the most common use of private IPv6 access.
$ gcloud alpha container clusters create \
--private-ipv6-google-access-type=disabled
$ gcloud alpha container clusters create \
--private-ipv6-google-access-type=outbound-only
$ gcloud alpha container clusters create \
--private-ipv6-google-access-type=bidirectional
PRIVATE_IPV6_GOOGLE_ACCESS_TYPE must be one of: bidirectional,
disabled, outbound-only.
--release-channel=CHANNEL
Release channel a cluster is subscribed to.
If left unspecified and a version is specified, the cluster is enrolled
in the most mature release channel where the version is available
(first checking STABLE, then REGULAR, and finally RAPID). Otherwise, if
no release channel and no version is specified, the cluster is enrolled
in the REGULAR channel with its default version. When a cluster is
subscribed to a release channel, Google maintains both the master
version and the node version. Node auto-upgrade is enabled by default
for release channel clusters and can be controlled via upgrade-scope
exclusions
(https://cloud.google.com/kubernetes-engine/docs/concepts/maintenance-windows-and-exclusions#scope_of_maintenance_to_exclude).
CHANNEL must be one of:
None
Use 'None' to opt-out of any release channel.
extended
Clusters subscribed to 'extended' can remain on a minor version for
24 months from when the minor version is made available in the
Regular channel.
rapid
'rapid' channel is offered on an early access basis for customers
who want to test new releases.
WARNING: Versions available in the 'rapid' channel may be subject
to unresolved issues with no known workaround and are not subject
to any SLAs.
regular
Clusters subscribed to 'regular' receive versions that are
considered GA quality. 'regular' is intended for production users
who want to take advantage of new features.
stable
Clusters subscribed to 'stable' receive versions that are known to
be stable and reliable in production.
--resource-manager-tags=[KEY=VALUE,...]
Applies the specified comma-separated resource manager tags that has
the GCE_FIREWALL purpose to all nodes in the new default node pool(s)
of a new cluster.
Examples:
$ gcloud beta container clusters create example-cluster \
--resource-manager-tags=tagKeys/1234=tagValues/2345
$ gcloud beta container clusters create example-cluster \
--resource-manager-tags=my-project/key1=value1
$ gcloud beta container clusters create example-cluster \
--resource-manager-tags=12345/key1=value1,23456/key2=value2
$ gcloud beta container clusters create example-cluster \
--resource-manager-tags=
All nodes, including nodes that are resized or re-created, will have
the specified tags on the corresponding Instance object in the Compute
Engine API. You can reference these tags in network firewall policy
rules. For instructions, see
https://cloud.google.com/firewall/docs/use-tags-for-firewalls.
--security-group=SECURITY_GROUP
The name of the RBAC security group for use with Google security groups
in Kubernetes RBAC
(https://kubernetes.io/docs/reference/access-authn-authz/rbac/).
To include group membership as part of the claims issued by Google
during authentication, a group must be designated as a security group
by including it as a direct member of this group.
If unspecified, no groups will be returned for use with RBAC.
--security-posture=SECURITY_POSTURE
Sets the mode of the Kubernetes security posture API's off-cluster
features.
To enable advanced mode explicitly set the flag to
--security-posture=enterprise.
To enable in standard mode explicitly set the flag to
--security-posture=standard
To disable in an existing cluster, explicitly set the flag to
--security-posture=disabled.
For more information on enablement, see
https://cloud.google.com/kubernetes-engine/docs/concepts/about-security-posture-dashboard#feature-enablement.
SECURITY_POSTURE must be one of: disabled, standard, enterprise.
--services-ipv4-cidr=CIDR
Set the IP range for the services IPs.
Can be specified as a netmask size (e.g. '/20') or as in CIDR notion
(e.g. '10.100.0.0/20'). If given as a netmask size, the IP range will
be chosen automatically from the available space in the network.
If unspecified, the services CIDR range will be chosen with a default
mask size.
Cannot be specified unless '--enable-ip-alias' option is also
specified.
--services-secondary-range-name=NAME
Set the secondary range to be used for services (e.g. ClusterIPs). NAME
must be the name of an existing secondary range in the cluster
subnetwork.
Cannot be specified unless '--enable-ip-alias' option is also
specified. Cannot be used with '--create-subnetwork' option.
--shielded-integrity-monitoring
Enables monitoring and attestation of the boot integrity of the
instance. The attestation is performed against the integrity policy
baseline. This baseline is initially derived from the implicitly
trusted boot image when the instance is created.
--shielded-secure-boot
The instance will boot with secure boot enabled.
--spot
Create nodes using spot VM instances in the new cluster.
$ gcloud beta container clusters create example-cluster --spot
New nodes, including ones created by resize or recreate, will use spot
VM instances.
--stack-type=STACK_TYPE
IP stack type of the cluster nodes. STACK_TYPE must be one of: ipv4,
ipv4-ipv6, ipv6.
--storage-pools=STORAGE_POOL,[...]
A list of storage pools where the cluster's boot disks will be
provisioned.
STORAGE_POOL must be in the format
projects/project/zones/zone/storagePools/storagePool
--subnetwork=SUBNETWORK
The Google Compute Engine subnetwork
(https://cloud.google.com/compute/docs/subnetworks) to which the
cluster is connected. The subnetwork must belong to the network
specified by --network.
Cannot be used with the "--create-subnetwork" option.
--system-config-from-file=PATH_TO_FILE
Path of the YAML/JSON file that contains the node configuration,
including Linux kernel parameters (sysctls) and kubelet configs.
Examples:
kubeletConfig:
cpuManagerPolicy: static
memoryManager:
policy: Static
topologyManager:
policy: BestEffort
scope: pod
crashLoopBackOff:
maxContainerRestartPeriod: '300s'
linuxConfig:
accurateTimeConfig:
enablePtpKvmTimeSync: true
sysctl:
net.core.somaxconn: '2048'
net.ipv4.tcp_rmem: '4096 87380 6291456'
hugepageConfig:
hugepage_size2m: '1024'
hugepage_size1g: '2'
swapConfig:
enabled: true
bootDiskProfile:
swapSizeGib: 8
cgroupMode: 'CGROUP_MODE_V2'
nodeKernelModuleLoading:
policy: 'ENFORCE_SIGNED_MODULES'
List of supported kubelet configs in 'kubeletConfig'.
KEY VALUE
cpuManagerPolicy either 'static' or 'none'
cpuCFSQuota true or false (enabled by
default)
cpuCFSQuotaPeriod interval (e.g., '100ms'. The
value must be between 1ms and
1 second, inclusive.)
memoryManager specify memory manager policy
topologyManager specify topology manager
policy and scope
podPidsLimit integer (The value must be
greater than or equal to 1024
and less than 4194304.)
containerLogMaxSize positive number plus unit
suffix (e.g., '100Mi',
'0.2Gi'. The value must be
between 10Mi and 500Mi,
inclusive.)
containerLogMaxFiles integer (The value must be
between [2, 10].)
imageGcLowThresholdPercent integer (The value must be
between [10, 85], and lower
than
imageGcHighThresholdPercent.)
imageGcHighThresholdPercent integer (The value must be
between [10, 85], and greater
than
imageGcLowThresholdPercent.)
imageMinimumGcAge interval (e.g., '100s', '1m'.
The value must be less than
'2m'.)
imageMaximumGcAge interval (e.g., '100s', '1m'.
The value must be greater than
imageMinimumGcAge.)
evictionSoft specify eviction soft
thresholds
evictionSoftGracePeriod specify eviction soft grace
period
evictionMinimumReclaim specify eviction minimum
reclaim thresholds
evictionMaxPodGracePeriodSeconds integer (Max grace period for
pod termination during
eviction, in seconds. The
value must be between [0,
300].)
shutdownGracePeriodSeconds integer (Grace period for pods
terminating on node shutdown,
in seconds. Allowed values: 0,
30, 120.)
shutdownGracePeriodCriticalPodsSeconds integer (Grace period for
critical pods terminating on
node shutdown, in seconds. The
value must be between [0, 120]
and less than
shutdownGracePeriodSeconds.)
allowedUnsafeSysctls list of sysctls (Allowlisted
groups: 'kernel.shm*',
'kernel.msg*', 'kernel.sem',
'fs.mqueue.*', and 'net.*',
and sysctls under the groups.)
singleProcessOomKill true or false
maxParallelImagePulls integer (The value must be
between [2, 5].)
crashLoopBackOff specify crashloopbackoff
thresholds
List of supported keys in memoryManager in 'kubeletConfig'.
KEY VALUE
policy either 'Static' or 'None'
List of supported keys in topologyManager in 'kubeletConfig'.
KEY VALUE
policy either 'none' or
'best-effort' or
'single-numa-node' or
'restricted'
scope either 'pod' or
'container'
List of supported keys in evictionSoft in 'kubeletConfig'.
KEY VALUE
memoryAvailable quantity (e.g., '100Mi', '1Gi'. Represents
the amount of memory available before soft
eviction. The value must be at least 100Mi
and less than 50% of the node's memory.)
nodefsAvailable percentage (e.g., '20%'. Represents the
nodefs available before soft eviction. The
value must be between 10% and 50%,
inclusive.)
nodefsInodesFree percentage (e.g., '20%'. Represents the
nodefs inodes free before soft eviction.
The value must be between 5% and 50%,
inclusive.)
imagefsAvailable percentage (e.g., '20%'. Represents the
imagefs available before soft eviction. The
value must be between 15% and 50%,
inclusive.)
imagefsInodesFree percentage (e.g., '20%'. Represents the
imagefs inodes free before soft eviction.
The value must be between 5% and 50%,
inclusive.)
pidAvailable percentage (e.g., '20%'. Represents the pid
available before soft eviction. The value
must be between 10% and 50%, inclusive.)
List of supported keys in evictionSoftGracePeriod in 'kubeletConfig'.
KEY VALUE
memoryAvailable duration (e.g., '30s', '1m'. The grace
period for soft eviction for this resource.
The value must be positive and no more than
'5m'.)
nodefsAvailable duration (e.g., '30s', '1m'. The grace
period for soft eviction for this resource.
The value must be positive and no more than
'5m'.)
nodefsInodesFree duration (e.g., '30s', '1m'. The grace
period for soft eviction for this resource.
The value must be positive and no more than
'5m'.)
imagefsAvailable duration (e.g., '30s', '1m'. The grace
period for soft eviction for this resource.
The value must be positive and no more than
'5m'.)
imagefsInodesFree duration (e.g., '30s', '1m'. The grace
period for soft eviction for this resource.
The value must be positive and no more than
'5m'.)
pidAvailable duration (e.g., '30s', '1m'. The grace
period for soft eviction for this resource.
The value must be positive and no more than
'5m'.)
List of supported keys in evictionMinimumReclaim in 'kubeletConfig'.
KEY VALUE
memoryAvailable percentage (e.g., '5%'. Represents the
minimum reclaim threshold for memory
available. The value must be positive and
no more than 10%.)
nodefsAvailable percentage (e.g., '5%'. Represents the
minimum reclaim threshold for nodefs
available. The value must be positive and
no more than 10%.)
nodefsInodesFree percentage (e.g., '5%'. Represents the
minimum reclaim threshold for nodefs inodes
free. The value must be positive and no
more than 10%.)
imagefsAvailable percentage (e.g., '5%'. Represents the
minimum reclaim threshold for imagefs
available. The value must be positive and
no more than 10%.)
imagefsInodesFree percentage (e.g., '5%'. Represents the
minimum reclaim threshold for imagefs
inodes free. The value must be positive and
no more than 10%.)
pidAvailable percentage (e.g., '5%'. Represents the
minimum reclaim threshold for pid
available. The value must be positive and
no more than 10%.)
List of supported keys in crashLoopBackOff in 'kubeletConfig'.
KEY VALUE
maxContainerRestartPeriod duration (e.g., '30s',
'1m'. The maximum duration
the backoff delay can
accrue to for container
restarts. The value must
be between [1s, 300s].)
List of supported sysctls in 'linuxConfig'.
KEY VALUE
net.core.netdev_max_backlog Any positive
integer, less than
2147483647
net.core.rmem_default Must be between
[2304, 2147483647]
net.core.rmem_max Must be between
[2304, 2147483647]
net.core.wmem_default Must be between
[4608, 2147483647]
net.core.wmem_max Must be between
[4608, 2147483647]
net.core.optmem_max Any positive
integer, less than
2147483647
net.core.somaxconn Must be between
[128, 2147483647]
net.ipv4.tcp_rmem Any positive
integer tuple
net.ipv4.tcp_wmem Any positive
integer tuple
net.ipv4.tcp_tw_reuse Must be {0, 1, 2}
net.ipv4.tcp_mtu_probing Must be {0, 1, 2}
net.ipv4.tcp_max_orphans Must be between
[16384, 262144]
net.ipv4.tcp_max_tw_buckets Must be between
[4096, 2147483647]
net.ipv4.tcp_syn_retries Must be between
[1, 127]
net.ipv4.tcp_ecn Must be {0, 1, 2}
net.ipv4.tcp_congestion_control Supported values
for COS: 'reno',
'cubic', 'bbr',
'lp', 'htcp'.
Supported values
for Ubuntu:
'reno', 'cubic',
'bbr', 'lp',
'htcp', 'vegas',
'dctcp', 'bic',
'cdg',
'highspeed',
'hybla',
'illinois', 'nv',
'scalable',
'veno',
'westwood',
'yeah'.
net.ipv4.neigh.default.gc_thresh1 Must be between
[0, 262144]
net.ipv4.neigh.default.gc_thresh2 Must be between
[512, 524288]
net.ipv4.neigh.default.gc_thresh3 Must be between
[1024, 1048576]
net.netfilter.nf_conntrack_max Must be between
[65536, 4194304]
net.netfilter.nf_conntrack_buckets Must be between
[65536, 524288].
Recommend setting:
nf_conntrack_max =
nf_conntrack_bucke
ts * 4
net.netfilter.nf_conntrack_tcp_timeout_close_wait Must be between
[60, 3600]
net.netfilter.nf_conntrack_tcp_timeout_time_wait Must be between
[1, 600]
net.netfilter.nf_conntrack_tcp_timeout_established Must be between
[600, 86400]
net.netfilter.nf_conntrack_acct Must be {0, 1}
kernel.keys.maxkeys Must be between
[200, 1048576]
kernel.keys.maxbytes Must be between
[20000, 2097152]
kernel.shmmni Must be between
[4096, 32768]
kernel.shmmax Must be between
[0,
184467440736927743
99]
kernel.shmall Must be between
[0,
184467440736927743
99]
kernel.perf_event_paranoid Must be {-1, 0, 1,
2, 3}
kernel.sched_rt_runtime_us Must be [-1,
1000000]
kernel.softlockup_panic Must be {0, 1}
kernel.yama.ptrace_scope Must be {0, 1, 2,
3}
kernel.kptr_restrict Must be {0, 1, 2}
kernel.dmesg_restrict Must be {0, 1}
kernel.sysrq Must be [0, 511]
fs.aio-max-nr Must be between
[65536, 4194304]
fs.file-max Must be between
[104857, 67108864]
fs.inotify.max_user_instances Must be between
[8192, 1048576]
fs.inotify.max_user_watches Must be between
[8192, 1048576]
fs.nr_open Must be between
[1048576,
2147483584]
vm.dirty_background_ratio Must be between
[1, 100]
vm.dirty_background_bytes Must be between
[0, 68719476736]
vm.dirty_expire_centisecs Must be between
[0, 6000]
vm.dirty_ratio Must be between
[1, 100]
vm.dirty_bytes Must be between
[0, 68719476736]
vm.dirty_writeback_centisecs Must be between
[0, 1000]
vm.max_map_count Must be between
[65536,
2147483647]
vm.overcommit_memory Must be one of {0,
1, 2}. Not
supported on
machines with less
than 15 GB memory.
vm.overcommit_ratio Must be between
[0, 100]
vm.vfs_cache_pressure Must be between
[0, 100]
vm.swappiness Must be between
[0, 200]
vm.watermark_scale_factor Must be between
[10, 3000]
vm.min_free_kbytes Must be between
[67584, 1048576]
List of supported keys in 'accurateTimeConfig' under 'linuxConfig'.
KEY VALUE
enablePtpKvmTimeSync boolean
List of supported hugepage size in 'hugepageConfig'.
KEY VALUE
hugepage_size2m Number of 2M huge pages, any positive integer
hugepage_size1g Number of 1G huge pages, any positive integer
List of supported keys in 'swapConfig' under 'linuxConfig'.
KEY VALUE
enabled boolean
encryptionConfig specify encryption
settings for the swap
space
bootDiskProfile specify swap on the node's
boot disk
ephemeralLocalSsdProfile specify swap on the local
SSD shared with pod
ephemeral storage
dedicatedLocalSsdProfile specify swap on a new,
separate local NVMe SSD
exclusively for swap
List of supported keys in 'encryptionConfig' under 'swapConfig'.
KEY VALUE
disabled boolean
List of supported keys in 'bootDiskProfile' under 'swapConfig'.
KEY VALUE
swapSizeGib integer
swapSizePercent integer
List of supported keys in 'ephemeralLocalSsdProfile' under
'swapConfig'.
KEY VALUE
swapSizeGib integer
swapSizePercent integer
List of supported keys in 'dedicatedLocalSsdProfile' under
'swapConfig'.
KEY VALUE
diskCount integer
List of supported keys in 'nodeKernelModuleLoading'.
KEY VALUE
policy ENFORCE_SIGNED_MODULES,
DO_NOT_ENFORCE_SIGNED_MODU
LES, POLICY_UNSPECIFIED
The upper limit for total allocated hugepage size differs based upon
machine size.
◆ On machines with less than 30 GB memory: 60% of the total memory.
For example, on e2-standard-2 machine with 8 GB of memory, you can't
allocate more than 4.8 GB for hugepages.
◆ On machines with more than 30 GB memory: 80% of the total memory.
For example, on c4a-standard-8 machines with 32 GB of memory,
hugepages cannot exceed 25.6 GB.
1G hugepages are only available in following machine familes: c3, m2,
c2d, c3d, h3, m3, a2, a3, g2.
Supported values for 'cgroupMode' under 'linuxConfig'.
◆ CGROUP_MODE_V1: Use cgroupv1 on the node pool.
◆ CGROUP_MODE_V2: Use cgroupv2 on the node pool.
◆ CGROUP_MODE_UNSPECIFIED: Use the default GKE cgroup configuration.
Supported values for 'transparentHugepageEnabled' under 'linuxConfig'
which controls transparent hugepage support for anonymous memory.
◆ TRANSPARENT_HUGEPAGE_ENABLED_ALWAYS: Transparent hugepage is
enabled system wide.
◆ TRANSPARENT_HUGEPAGE_ENABLED_MADVISE: Transparent hugepage is
enabled inside MADV_HUGEPAGE regions. This is the default kernel
configuration.
◆ TRANSPARENT_HUGEPAGE_ENABLED_NEVER: Transparent hugepage is
disabled.
◆ TRANSPARENT_HUGEPAGE_ENABLED_UNSPECIFIED: Default value. GKE will
not modify the kernel configuration.
Supported values for 'transparentHugepageDefrag' under 'linuxConfig'
which defines the transparent hugepage defrag configuration on the
node.
◆ TRANSPARENT_HUGEPAGE_DEFRAG_ALWAYS: It means that an application
requesting THP will stall on allocation failure and directly reclaim
pages and compact memory in an effort to allocate a THP immediately.
◆ TRANSPARENT_HUGEPAGE_DEFRAG_DEFER: It means that an application
will wake kswapd in the background to reclaim pages and wake
kcompactd to compact memory so that THP is available in the near
future. It is the responsibility of khugepaged to then install the
THP pages later.
◆ TRANSPARENT_HUGEPAGE_DEFRAG_DEFER_WITH_MADVISE: It means that an
application will enter direct reclaim and compaction like always, but
only for regions that have used madvise(MADV_HUGEPAGE); all other
regions will wake kswapd in the background to reclaim pages and wake
kcompactd to compact memory so that THP is available in the near
future.
◆ TRANSPARENT_HUGEPAGE_DEFRAG_MADVISE: It means that an application
will enter direct reclaim and compaction like always, but only for
regions that have used madvise(MADV_HUGEPAGE); all other regions will
wake kswapd in the background to reclaim pages and wake kcompactd to
compact memory so that THP is available in the near future.
◆ TRANSPARENT_HUGEPAGE_DEFRAG_NEVER: It means that an application
will never enter direct reclaim or compaction.
◆ TRANSPARENT_HUGEPAGE_DEFRAG_UNSPECIFIED: Default value. GKE will
not modify the kernel configuration.
Supported values for 'policy' under 'nodeKernelModuleLoading'.
◆ POLICY_UNSPECIFIED: Default behavior. GKE selects the image based
on node type. For CPU and TPU nodes, the image will not allow loading
external kernel modules. For GPU nodes, the image will allow loading
any module, whether it is signed or not.
◆ ENFORCE_SIGNED_MODULES: Enforced signature verification: Node pools
will use a Container-Optimized OS image configured to allow loading
of Google-signed external kernel modules. Loadpin is enabled but
configured to exclude modules, and kernel module signature checking
is enforced.
◆ DO_NOT_ENFORCE_SIGNED_MODULES: Do not enforce kernel module
signature enforcement. Mirrors existing DEFAULT behavior.
Note, updating the system configuration of an existing node pool
requires recreation of the nodes which which might cause a disruption.
Use a full or relative path to a local file containing the value of
system_config.
--tags=TAG,[TAG,...]
Applies the given Compute Engine tags (comma separated) on all nodes in
the new node-pool.
Examples:
$ gcloud beta container clusters create example-cluster \
--tags=tag1,tag2
New nodes, including ones created by resize or recreate, will have
these tags on the Compute Engine API instance object and can be used in
firewall rules. See
https://cloud.google.com/sdk/gcloud/reference/compute/firewall-rules/create
for examples.
--threads-per-core=THREADS_PER_CORE
The number of visible threads per physical core for each node. To
disable simultaneous multithreading (SMT) set this to 1.
--tier=TIER
(DEPRECATED) Set the desired tier for the cluster.
The --tier flag is deprecated. More info:
https://cloud.google.com/kubernetes-engine/docs/release-notes#September_02_2025.
TIER must be one of: standard, enterprise.
--workload-metadata=WORKLOAD_METADATA
Type of metadata server available to pods running in the node pool.
WORKLOAD_METADATA must be one of:
EXPOSED
[DEPRECATED] Pods running in this node pool have access to the
node's underlying Compute Engine Metadata Server.
GCE_METADATA
Pods running in this node pool have access to the node's underlying
Compute Engine Metadata Server.
GKE_METADATA
Run the Kubernetes Engine Metadata Server on this node. The
Kubernetes Engine Metadata Server exposes a metadata API to
workloads that is compatible with the V1 Compute Metadata APIs
exposed by the Compute Engine and App Engine Metadata Servers. This
feature can only be enabled if Workload Identity is enabled at the
cluster level.
GKE_METADATA_SERVER
[DEPRECATED] Run the Kubernetes Engine Metadata Server on this
node. The Kubernetes Engine Metadata Server exposes a metadata API
to workloads that is compatible with the V1 Compute Metadata APIs
exposed by the Compute Engine and App Engine Metadata Servers. This
feature can only be enabled if Workload Identity is enabled at the
cluster level.
SECURE
[DEPRECATED] Prevents pods not in hostNetwork from accessing
certain VM metadata, specifically kube-env, which contains Kubelet
credentials, and the instance identity token. This is a temporary
security solution available while the bootstrapping process for
cluster nodes is being redesigned with significant security
improvements. This feature is scheduled to be deprecated in the
future and later removed.
--workload-pool=WORKLOAD_POOL
Enable Workload Identity on the cluster.
When enabled, Kubernetes service accounts will be able to act as Cloud
IAM Service Accounts, through the provided workload pool.
Currently, the only accepted workload pool is the workload pool of the
Cloud project containing the cluster, PROJECT_ID.svc.id.goog.
For more information on Workload Identity, see
https://cloud.google.com/kubernetes-engine/docs/how-to/workload-identity
--workload-vulnerability-scanning=WORKLOAD_VULNERABILITY_SCANNING
Sets the mode of the Kubernetes security posture API's workload
vulnerability scanning.
To enable Advanced vulnerability insights mode explicitly set the flag
to --workload-vulnerability-scanning=enterprise.
To enable in standard mode explicitly set the flag to
--workload-vulnerability-scanning=standard.
To disable in an existing cluster, explicitly set the flag to
--workload-vulnerability-scanning=disabled.
For more information on enablement, see
https://cloud.google.com/kubernetes-engine/docs/concepts/about-security-posture-dashboard#feature-enablement.
WORKLOAD_VULNERABILITY_SCANNING must be one of: disabled, standard,
enterprise.
At most one of these can be specified:
--additional-zones=ZONE,[ZONE,...]
(DEPRECATED) The set of additional zones in which the specified node
footprint should be replicated. All zones must be in the same region
as the cluster's primary zone. If additional-zones is not specified,
all nodes will be in the cluster's primary zone.
Note that NUM_NODES nodes will be created in each zone, such that if
you specify --num-nodes=4 and choose one additional zone, 8 nodes
will be created.
Multiple locations can be specified, separated by commas. For
example:
$ gcloud beta container clusters create example-cluster \
--zone us-central1-a \
--additional-zones us-central1-b,us-central1-c
This flag is deprecated. Use --node-locations=PRIMARY_ZONE,[ZONE,...]
instead.
--node-locations=ZONE,[ZONE,...]
The set of zones in which the specified node footprint should be
replicated. All zones must be in the same region as the cluster's
master(s), specified by the -location, --zone, or --region flag.
Additionally, for zonal clusters, --node-locations must contain the
cluster's primary zone. If not specified, all nodes will be in the
cluster's primary zone (for zonal clusters) or spread across three
randomly chosen zones within the cluster's region (for regional
clusters).
Note that NUM_NODES nodes will be created in each zone, such that if
you specify --num-nodes=4 and choose two locations, 8 nodes will be
created.
Multiple locations can be specified, separated by commas. For
example:
$ gcloud beta container clusters create example-cluster \
--location us-central1-a \
--node-locations us-central1-a,us-central1-b
Control Plane Keys
--aggregation-ca=CA_POOL_PATH
The Certificate Authority Service caPool that will back the aggregation
CA
--cluster-ca=CA_POOL_PATH
The Certificate Authority Service caPool that will back the cluster CA
--control-plane-disk-encryption-key=KEY
The Cloud KMS symmetric encryption cryptoKey that will be used to
encrypt the control plane disks
--etcd-api-ca=CA_POOL_PATH
The Certificate Authority Service caPool that will back the etcd API CA
--etcd-peer-ca=CA_POOL_PATH
The Certificate Authority Service caPool that will back the etcd peer
CA
--gkeops-etcd-backup-encryption-key=KEY
The Cloud KMS symmetric encryption cryptoKey that will be used to
encrypt the disaster recovery etcd backups for the cluster
--service-account-signing-keys=KEY_VERSION,[KEY_VERSION,...]
A Cloud KMS asymmetric signing cryptoKeyVersion that will be used to
sign service account tokens
--service-account-verification-keys=KEY_VERSION,[KEY_VERSION,...]
A Cloud KMS asymmetric signing cryptoKeyVersion that will be used to
verify service account tokens. Maybe specified multiple times.
Flags for Binary Authorization:
--binauthz-policy-bindings=[name=BINAUTHZ_POLICY]
The relative resource name of the Binary Authorization policy to audit
and/or enforce. GKE policies have the following format:
projects/{project_number}/platforms/gke/policies/{policy_id}.
At most one of these can be specified:
--binauthz-evaluation-mode=BINAUTHZ_EVALUATION_MODE
Enable Binary Authorization for this cluster.
BINAUTHZ_EVALUATION_MODE must be one of: disabled, policy-bindings,
policy-bindings-and-project-singleton-policy-enforce,
project-singleton-policy-enforce.
--enable-binauthz
(DEPRECATED) Enable Binary Authorization for this cluster.
The --enable-binauthz flag is deprecated. Please use
--binauthz-evaluation-mode instead.
Configure boot disk options.
--boot-disk-provisioned-iops=BOOT_DISK_PROVISIONED_IOPS
Configure the Provisioned IOPS for the node pool boot disks. Only valid
for hyperdisk-balanced boot disks.
--boot-disk-provisioned-throughput=BOOT_DISK_PROVISIONED_THROUGHPUT
Configure the Provisioned Throughput for the node pool boot disks. Only
valid for hyperdisk-balanced boot disks.
ClusterDNS
--cluster-dns=CLUSTER_DNS
DNS provider to use for this cluster. CLUSTER_DNS must be one of:
clouddns
Selects Cloud DNS as the DNS provider for the cluster.
default
Selects the default DNS provider (kube-dns) for the cluster.
kubedns
Selects Kube DNS as the DNS provider for the cluster.
--cluster-dns-domain=CLUSTER_DNS_DOMAIN
DNS domain for this cluster. The default value is cluster.local. This
is configurable when --cluster-dns=clouddns and --cluster-dns-scope=vpc
are set. The value must be a valid DNS subdomain as defined in RFC
1123.
--cluster-dns-scope=CLUSTER_DNS_SCOPE
DNS scope for the Cloud DNS zone created - valid only with
--cluster-dns=clouddns. Defaults to cluster.
CLUSTER_DNS_SCOPE must be one of:
cluster
Configures the Cloud DNS zone to be private to the cluster.
vpc
Configures the Cloud DNS zone to be private to the VPC Network.
At most one of these can be specified:
--additive-vpc-scope-dns-domain=ADDITIVE_VPC_SCOPE_DNS_DOMAIN
The domain used in Additive VPC scope. Only works with Cluster Scope.
--disable-additive-vpc-scope
Disables Additive VPC Scope.
At most one of these can be specified:
--dataplane-v2-observability-mode=DATAPLANE_V2_OBSERVABILITY_MODE
(REMOVED) Select Advanced Datapath Observability mode for the
cluster. Defaults to DISABLED.
Advanced Datapath Observability allows for a real-time view into
pod-to-pod traffic within your cluster.
Examples:
$ gcloud beta container clusters create \
--dataplane-v2-observability-mode=DISABLED
$ gcloud beta container clusters create \
--dataplane-v2-observability-mode=INTERNAL_VPC_LB
$ gcloud beta container clusters create \
--dataplane-v2-observability-mode=EXTERNAL_LB
Flag --dataplane-v2-observability-mode has been removed.
DATAPLANE_V2_OBSERVABILITY_MODE must be one of:
DISABLED
Disables Advanced Datapath Observability.
EXTERNAL_LB
Makes Advanced Datapath Observability available to the external
network.
INTERNAL_VPC_LB
Makes Advanced Datapath Observability available from the VPC
network.
--disable-dataplane-v2-flow-observability
Disables Advanced Datapath Observability.
--enable-dataplane-v2-flow-observability
Enables Advanced Datapath Observability which allows for a real-time
view into pod-to-pod traffic within your cluster.
At most one of these can be specified:
--disable-dataplane-v2-metrics
Stops exposing advanced datapath flow metrics on node port.
--enable-dataplane-v2-metrics
Exposes advanced datapath flow metrics on node port.
At most one of these can be specified:
--disable-pod-snapshots
Disable the Pod Snapshot feature on the cluster.
--enable-pod-snapshots
Enable the Pod Snapshot feature on the cluster.
Node autoprovisioning
--enable-autoprovisioning
Enables node autoprovisioning for a cluster.
Cluster Autoscaler will be able to create new node pools. Requires
maximum CPU and memory limits to be specified.
At most one of these can be specified:
--autoprovisioning-config-file=PATH_TO_FILE
Path of the JSON/YAML file which contains information about the
cluster's node autoprovisioning configuration. Currently it contains
a list of resource limits, identity defaults for autoprovisioning,
node upgrade settings, node management settings, minimum cpu
platform, image type, node locations for autoprovisioning, disk type
and size configuration, Shielded instance settings, and
customer-managed encryption keys settings.
Resource limits are specified in the field 'resourceLimits'. Each
resource limits definition contains three fields: resourceType,
maximum and minimum. Resource type can be "cpu", "memory" or an
accelerator (e.g. "nvidia-tesla-t4" for NVIDIA T4). Use gcloud
compute accelerator-types list to learn about available accelerator
types. Maximum is the maximum allowed amount with the unit of the
resource. Minimum is the minimum allowed amount with the unit of the
resource.
Identity default contains at most one of the below fields:
serviceAccount: The Google Cloud Platform Service Account to be used
by node VMs in autoprovisioned node pools. If not specified, the
project's default service account is used. scopes: A list of scopes
to be used by node instances in autoprovisioned node pools. Multiple
scopes can be specified, separated by commas. For information on
defaults, look at:
https://cloud.google.com/sdk/gcloud/reference/container/clusters/create#--scopes
Node Upgrade settings are specified under the field
'upgradeSettings', which has the following fields: maxSurgeUpgrade:
Number of extra (surge) nodes to be created on each upgrade of an
autoprovisioned node pool. maxUnavailableUpgrade: Number of nodes
that can be unavailable at the same time on each upgrade of an
autoprovisioned node pool.
Node Management settings are specified under the field 'management',
which has the following fields: autoUpgrade: A boolean field that
indicates if node autoupgrade is enabled for autoprovisioned node
pools. autoRepair: A boolean field that indicates if node autorepair
is enabled for autoprovisioned node pools.
minCpuPlatform (deprecated): If specified, new autoprovisioned nodes
will be scheduled on host with specified CPU architecture or a newer
one. Note: Min CPU platform can only be specified in Beta and Alpha.
Autoprovisioned node image is specified under the 'imageType' field.
If not specified the default value will be applied.
Autoprovisioning locations is a set of zones where new node pools can
be created by Autoprovisioning. Autoprovisioning locations are
specified in the field 'autoprovisioningLocations'. All zones must be
in the same region as the cluster's master(s).
Disk type and size are specified under the 'diskType' and
'diskSizeGb' fields, respectively. If specified, new autoprovisioned
nodes will be created with custom boot disks configured by these
settings.
Shielded instance settings are specified under the
'shieldedInstanceConfig' field, which has the following fields:
enableSecureBoot: A boolean field that indicates if secure boot is
enabled for autoprovisioned nodes. enableIntegrityMonitoring: A
boolean field that indicates if integrity monitoring is enabled for
autoprovisioned nodes.
Customer Managed Encryption Keys (CMEK) used by new auto-provisioned
node pools can be specified in the 'bootDiskKmsKey' field.
Use a full or relative path to a local file containing the value of
autoprovisioning_config_file.
Or at least one of these can be specified:
Flags to configure autoprovisioned nodes
--autoprovisioning-image-type=AUTOPROVISIONING_IMAGE_TYPE
Node Autoprovisioning will create new nodes with the specified
image type
--autoprovisioning-locations=ZONE,[ZONE,...]
Set of zones where new node pools can be created by
autoprovisioning. All zones must be in the same region as the
cluster's master(s). Multiple locations can be specified, separated
by commas.
--autoprovisioning-min-cpu-platform=PLATFORM
(DEPRECATED) If specified, new autoprovisioned nodes will be
scheduled on host with specified CPU architecture or a newer one.
The --autoprovisioning-min-cpu-platform flag is deprecated and will
be removed in an upcoming release. More info:
https://cloud.google.com/kubernetes-engine/docs/release-notes#March_08_2022
--max-cpu=MAX_CPU
Maximum number of cores in the cluster.
Maximum number of cores to which the cluster can scale. Required to
be set when --enable-autoprovisioning is used.
--max-memory=MAX_MEMORY
Maximum memory in the cluster.
Maximum number of gigabytes of memory to which the cluster can
scale. Required to be set when --enable-autoprovisioning is used.
--min-cpu=MIN_CPU
Minimum number of cores in the cluster.
Minimum number of cores to which the cluster can scale.
--min-memory=MIN_MEMORY
Minimum memory in the cluster.
Minimum number of gigabytes of memory to which the cluster can
scale.
Flags to specify upgrade settings for autoprovisioned nodes:
--autoprovisioning-max-surge-upgrade=AUTOPROVISIONING_MAX_SURGE_UPGRADE
Number of extra (surge) nodes to be created on each upgrade of an
autoprovisioned node pool.
--autoprovisioning-max-unavailable-upgrade=AUTOPROVISIONING_MAX_UNAVAILABLE_UPGRADE
Number of nodes that can be unavailable at the same time on each
upgrade of an autoprovisioned node pool.
--autoprovisioning-node-pool-soak-duration=AUTOPROVISIONING_NODE_POOL_SOAK_DURATION
Time in seconds to be spent waiting during blue-green upgrade
before deleting the blue pool and completing the update. This
argument should be used in conjunction with
--enable-autoprovisioning-blue-green-upgrade to take effect.
--autoprovisioning-standard-rollout-policy=[batch-node-count=BATCH_NODE_COUNT,batch-percent=BATCH_NODE_PERCENTAGE,batch-soak-duration=BATCH_SOAK_DURATION,...]
Standard rollout policy options for blue-green upgrade. This
argument should be used in conjunction with
--enable-autoprovisioning-blue-green-upgrade to take effect.
Batch sizes are specified by one of, batch-node-count or
batch-percent. The duration between batches is specified by
batch-soak-duration.
Example:
--standard-rollout-policy=batch-node-count=3,batch-soak-duration=60s
--standard-rollout-policy=batch-percent=0.05,batch-soak-duration=180s
Flag group to choose the top level upgrade option:
At most one of these can be specified:
--enable-autoprovisioning-blue-green-upgrade
Whether to use blue-green upgrade for the autoprovisioned node
pool.
--enable-autoprovisioning-surge-upgrade
Whether to use surge upgrade for the autoprovisioned node pool.
Flags to specify identity for autoprovisioned nodes:
--autoprovisioning-scopes=[SCOPE,...]
The scopes to be used by node instances in autoprovisioned node
pools. Multiple scopes can be specified, separated by commas. For
information on defaults, look at:
https://cloud.google.com/sdk/gcloud/reference/container/clusters/create#--scopes
--autoprovisioning-service-account=AUTOPROVISIONING_SERVICE_ACCOUNT
The Google Cloud Platform Service Account to be used by node VMs in
autoprovisioned node pools. If not specified, the project default
service account is used.
Flags to specify node management settings for autoprovisioned nodes:
--enable-autoprovisioning-autorepair
Enable node autorepair for autoprovisioned node pools. Use
--no-enable-autoprovisioning-autorepair to disable.
This flag argument must be specified if any of the other arguments
in this group are specified.
--enable-autoprovisioning-autoupgrade
Enable node autoupgrade for autoprovisioned node pools. Use
--no-enable-autoprovisioning-autoupgrade to disable.
This flag argument must be specified if any of the other arguments
in this group are specified.
Arguments to set limits on accelerators:
--max-accelerator=[type=TYPE,count=COUNT,...]
Sets maximum limit for a single type of accelerators (e.g. GPUs) in
cluster.
type
(Required) The specific type (e.g. nvidia-tesla-t4 for NVIDIA
T4) of accelerator for which the limit is set. Use gcloud
compute accelerator-types list to learn about all available
accelerator types.
count
(Required) The maximum number of accelerators to which the
cluster can be scaled.
This flag argument must be specified if any of the other
arguments in this group are specified.
--min-accelerator=[type=TYPE,count=COUNT,...]
Sets minimum limit for a single type of accelerators (e.g. GPUs) in
cluster. Defaults to 0 for all accelerator types if it isn't set.
type
(Required) The specific type (e.g. nvidia-tesla-t4 for NVIDIA
T4) of accelerator for which the limit is set. Use gcloud
compute accelerator-types list to learn about all available
accelerator types.
count
(Required) The minimum number of accelerators to which the
cluster can be scaled.
Cluster autoscaling
--enable-autoscaling
Enables autoscaling for a node pool.
Enables autoscaling in the node pool specified by --node-pool or the
default node pool if --node-pool is not provided. If not already,
--max-nodes or --total-max-nodes must also be set.
--location-policy=LOCATION_POLICY
Location policy specifies the algorithm used when scaling-up the node
pool.
◆ BALANCED - Is a best effort policy that aims to balance the sizes
of available zones.
◆ ANY - Instructs the cluster autoscaler to prioritize utilization of
unused reservations, and reduces preemption risk for Spot VMs.
LOCATION_POLICY must be one of: BALANCED, ANY.
--max-nodes=MAX_NODES
Maximum number of nodes per zone in the node pool.
Maximum number of nodes per zone to which the node pool specified by
--node-pool (or default node pool if unspecified) can scale. Ignored
unless --enable-autoscaling is also specified.
--min-nodes=MIN_NODES
Minimum number of nodes per zone in the node pool.
Minimum number of nodes per zone to which the node pool specified by
--node-pool (or default node pool if unspecified) can scale. Ignored
unless --enable-autoscaling is also specified.
--total-max-nodes=TOTAL_MAX_NODES
Maximum number of all nodes in the node pool.
Maximum number of all nodes to which the node pool specified by
--node-pool (or default node pool if unspecified) can scale. Ignored
unless --enable-autoscaling is also specified.
--total-min-nodes=TOTAL_MIN_NODES
Minimum number of all nodes in the node pool.
Minimum number of all nodes to which the node pool specified by
--node-pool (or default node pool if unspecified) can scale. Ignored
unless --enable-autoscaling is also specified.
--enable-insecure-binding-system-authenticated
Allow using system:authenticated as a subject in ClusterRoleBindings
and RoleBindings. Allowing bindings that reference system:authenticated
is a security risk and is not recommended.
To disallow binding system:authenticated in a cluster, explicitly set
the --no-enable-insecure-binding-system-authenticated flag instead.
--enable-insecure-binding-system-unauthenticated
Allow using system:unauthenticated and system:anonymous as subjects in
ClusterRoleBindings and RoleBindings. Allowing bindings that reference
system:unauthenticated and system:anonymous are a security risk and is
not recommended.
To disallow binding system:authenticated in a cluster, explicitly set
the --no-enable-insecure-binding-system-unauthenticated flag instead.
Master Authorized Networks
--enable-master-authorized-networks
Allow only specified set of CIDR blocks (specified by the
--master-authorized-networks flag) to connect to Kubernetes master
through HTTPS. Besides these blocks, the following have access as well:
1) The private network the cluster connects to if
`--enable-private-nodes` is specified.
2) Google Compute Engine Public IPs if `--enable-private-nodes` is not
specified.
Use --no-enable-master-authorized-networks to disable. When disabled,
public internet (0.0.0.0/0) is allowed to connect to Kubernetes master
through HTTPS.
--master-authorized-networks=NETWORK,[NETWORK,...]
The list of CIDR blocks (up to 100 for private cluster, 50 for public
cluster) that are allowed to connect to Kubernetes master through
HTTPS. Specified in CIDR notation (e.g. 1.2.3.4/30). Cannot be
specified unless --enable-master-authorized-networks is also specified.
Exports cluster's usage of cloud resources
--enable-network-egress-metering
Enable network egress metering on this cluster.
When enabled, a DaemonSet is deployed into the cluster. Each DaemonSet
pod meters network egress traffic by collecting data from the conntrack
table, and exports the metered metrics to the specified destination.
Network egress metering is disabled if this flag is omitted, or when
--no-enable-network-egress-metering is set.
--enable-resource-consumption-metering
Enable resource consumption metering on this cluster.
When enabled, a table will be created in the specified BigQuery dataset
to store resource consumption data. The resulting table can be joined
with the resource usage table or with BigQuery billing export.
Resource consumption metering is enabled unless --no-enable-resource-
consumption-metering is set.
--resource-usage-bigquery-dataset=RESOURCE_USAGE_BIGQUERY_DATASET
The name of the BigQuery dataset to which the cluster's usage of cloud
resources is exported. A table will be created in the specified dataset
to store cluster resource usage. The resulting table can be joined with
BigQuery Billing Export to produce a fine-grained cost breakdown.
Examples:
$ gcloud beta container clusters create example-cluster \
--resource-usage-bigquery-dataset=example_bigquery_dataset_name
Private Clusters
--enable-private-endpoint
Cluster is managed using the private IP address of the master API
endpoint.
--enable-private-nodes
Cluster is created with no public IP addresses on the cluster nodes.
--master-ipv4-cidr=MASTER_IPV4_CIDR
IPv4 CIDR range to use for the master network. This should have a
netmask of size /28 and should be used in conjunction with the
--enable-private-nodes flag.
--private-cluster
(DEPRECATED) Cluster is created with no public IP addresses on the
cluster nodes.
The --private-cluster flag is deprecated and will be removed in a
future release. Use --enable-private-nodes instead.
Flags for Secret Manager configuration:
--enable-secret-manager
Enables the Secret Manager CSI driver provider component. See
https://secrets-store-csi-driver.sigs.k8s.io/introduction
https://github.com/GoogleCloudPlatform/secrets-store-csi-driver-provider-gcp
--enable-secret-manager-rotation
Enables the rotation of secrets in the Secret Manager CSI driver
provider component.
--secret-manager-rotation-interval=SECRET_MANAGER_ROTATION_INTERVAL
Set the rotation period for secrets in the Secret Manager CSI driver
provider component. If you don't specify a time interval for the
rotation, it will default to a rotation period of two minutes.
Flags for Secret Sync configuration:
--enable-secret-sync
Enables the Secret Sync component. See
https://cloud.google.com/secret-manager/docs/sync-k8-secrets
--enable-secret-sync-rotation
Enables the rotation of secrets in the Secret Sync component. provider
component.
--secret-sync-rotation-interval=SECRET_SYNC_ROTATION_INTERVAL
Set the rotation period for secrets in the Secret Sync component.
At most one of these can be specified:
--ephemeral-storage[=[local-ssd-count=LOCAL-SSD-COUNT]]
Parameters for the ephemeral storage filesystem. If unspecified,
ephemeral storage is backed by the boot disk.
Examples:
$ gcloud beta container clusters create example_cluster \
--ephemeral-storage local-ssd-count=2
'local-ssd-count' specifies the number of local SSDs to use to back
ephemeral storage. Local SDDs use NVMe interfaces. For first- and
second-generation machine types, a nonzero count field is required
for local ssd to be configured. For third-generation machine types,
the count field is optional because the count is inferred from the
machine type.
See https://cloud.google.com/compute/docs/disks/local-ssd for more
information.
--ephemeral-storage-local-ssd[=[count=COUNT]]
Parameters for the ephemeral storage filesystem. If unspecified,
ephemeral storage is backed by the boot disk.
Examples:
$ gcloud beta container clusters create example_cluster \
--ephemeral-storage-local-ssd count=2
'count' specifies the number of local SSDs to use to back ephemeral
storage. Local SDDs use NVMe interfaces. For first- and
second-generation machine types, a nonzero count field is required
for local ssd to be configured. For third-generation machine types,
the count field is optional because the count is inferred from the
machine type.
See https://cloud.google.com/compute/docs/disks/local-ssd for more
information.
--local-nvme-ssd-block[=[count=COUNT]]
Adds the requested local SSDs on all nodes in default node pool(s) in
the new cluster.
Examples:
$ gcloud beta container clusters create example_cluster \
--local-nvme-ssd-block count=2
'count' must be between 1-8
New nodes, including ones created by resize or recreate, will have
these local SSDs.
For first- and second-generation machine types, a nonzero count field
is required for local ssd to be configured. For third-generation
machine types, the count field is optional because the count is
inferred from the machine type.
See https://cloud.google.com/compute/docs/disks/local-ssd for more
information.
--local-ssd-count=LOCAL_SSD_COUNT
The number of local SSD disks to provision on each node, formatted
and mounted in the filesystem.
Local SSDs have a fixed 375 GB capacity per device. The number of
disks that can be attached to an instance is limited by the maximum
number of disks available on a machine, which differs by compute
zone. See https://cloud.google.com/compute/docs/disks/local-ssd for
more information.
At most one of these can be specified:
--location=LOCATION
Compute zone or region (e.g. us-central1-a or us-central1) for the
cluster. Overrides the default compute/region or compute/zone value
for this command invocation. Prefer using this flag over the --region
or --zone flags.
--region=REGION
Compute region (e.g. us-central1) for a regional cluster. Overrides
the default compute/region property value for this command
invocation.
--zone=ZONE, -z ZONE
Compute zone (e.g. us-central1-a) for a zonal cluster. Overrides the
default compute/zone property value for this command invocation.
Flags for cluster disruption budget configuration:
--maintenance-minor-version-disruption-interval=MAINTENANCE_MINOR_VERSION_DISRUPTION_INTERVAL
Set the minimum interval of time between minor version cluster
upgrades.
--maintenance-patch-version-disruption-interval=MAINTENANCE_PATCH_VERSION_DISRUPTION_INTERVAL
Set the minimum interval of time between patch version cluster
upgrades.
One of either maintenance-window or the group of maintenance-window flags
can be set.
At most one of these can be specified:
--maintenance-window=START_TIME
Set a time of day when you prefer maintenance to start on this
cluster. For example:
$ gcloud beta container clusters create example-cluster \
--maintenance-window=12:43
The time corresponds to the UTC time zone, and must be in HH:MM
format.
Non-emergency maintenance will occur in the 4 hour block starting at
the specified time.
This is mutually exclusive with the recurring maintenance windows and
will overwrite any existing window. Compatible with maintenance
exclusions.
Or at least one of these can be specified:
Set a flexible maintenance window by specifying a window that recurs
per an RFC 5545 RRULE. Non-emergency maintenance will occur in the
recurring windows.
Examples:
For a 9-5 Mon-Wed UTC-4 maintenance window:
$ gcloud beta container clusters create example-cluster \
--maintenance-window-start=2000-01-01T09:00:00-04:00 \
--maintenance-window-end=2000-01-01T17:00:00-04:00 \
--maintenance-window-recurrence='FREQ=WEEKLY;BYDAY=MO,TU,WE'
For a daily window from 22:00 - 04:00 UTC:
$ gcloud beta container clusters create example-cluster \
--maintenance-window-start=2000-01-01T22:00:00Z \
--maintenance-window-end=2000-01-02T04:00:00Z \
--maintenance-window-recurrence=FREQ=DAILY
--maintenance-window-end=TIME_STAMP
The end time for calculating the duration of the maintenance
window, as expressed by the amount of time after the START_TIME, in
the same format. The value for END_TIME must be in the future,
relative to START_TIME. This only calculates the duration of the
window, and doesn't set when the maintenance window stops
recurring. Maintenance windows only stop recurring when they're
removed. See $ gcloud topic datetimes for information on time
formats.
This flag argument must be specified if any of the other arguments
in this group are specified.
This flag argument must be specified if any of the other arguments
in this group are specified.
--maintenance-window-recurrence=RRULE
An RFC 5545 RRULE, specifying how the window will recur. Note that
minimum requirements for maintenance periods will be enforced. Note
that FREQ=SECONDLY, MINUTELY, and HOURLY are not supported.
This flag argument must be specified if any of the other arguments
in this group are specified.
--maintenance-window-start=TIME_STAMP
Start time of the first window (can occur in the past). The start
time influences when the window will start for recurrences. See $
gcloud topic datetimes for information on time formats.
This flag argument must be specified if any of the other arguments
in this group are specified.
Basic auth
--password=PASSWORD
The password to use for cluster auth. Defaults to a server-specified
randomly-generated string.
Options to specify the username.
At most one of these can be specified:
--enable-basic-auth
Enable basic (username/password) auth for the cluster.
--enable-basic-auth is an alias for --username=admin;
--no-enable-basic-auth is an alias for --username="". Use --password
to specify a password; if not, the server will randomly generate one.
For cluster versions before 1.12, if neither --enable-basic-auth nor
--username is specified, --enable-basic-auth will default to true.
After 1.12, --enable-basic-auth will default to false.
--username=USERNAME, -u USERNAME
The user name to use for basic auth for the cluster. Use --password
to specify a password; if not, the server will randomly generate one.
Specifies the reservation for the default initial node pool.
--reservation=RESERVATION
The name of the reservation, required when
--reservation-affinity=specific.
--reservation-affinity=RESERVATION_AFFINITY
The type of the reservation for the default initial node pool.
RESERVATION_AFFINITY must be one of: any, none, specific.
Options to specify the node identity.
Scopes options.
--scopes=[SCOPE,...]; default="gke-default"
Specifies scopes for the node instances.
Examples:
$ gcloud beta container clusters create example-cluster \
--scopes=https://www.googleapis.com/auth/devstorage.read_only
$ gcloud beta container clusters create example-cluster \
--scopes=bigquery,storage-rw,compute-ro
Multiple scopes can be specified, separated by commas. Various scopes
are automatically added based on feature usage. Such scopes are not
added if an equivalent scope already exists.
◆ monitoring-write: always added to ensure metrics can be written
◆ logging-write: added if Cloud Logging is enabled
(--enable-cloud-logging/--logging)
◆ monitoring: added if Cloud Monitoring is enabled
(--enable-cloud-monitoring/--monitoring)
◆ gke-default: added for Autopilot clusters that use the default
service account
◆ cloud-platform: added for Autopilot clusters that use any other
service account
SCOPE can be either the full URI of the scope or an alias. Default
scopes are assigned to all instances. Available aliases are:
Alias URI
bigquery https://www.googleapis.com/auth/bigquery
cloud-platform https://www.googleapis.com/auth/cloud-platform
cloud-source-repos https://www.googleapis.com/auth/source.full_control
cloud-source-repos-ro https://www.googleapis.com/auth/source.read_only
compute-ro https://www.googleapis.com/auth/compute.readonly
compute-rw https://www.googleapis.com/auth/compute
datastore https://www.googleapis.com/auth/datastore
default https://www.googleapis.com/auth/devstorage.read_only
https://www.googleapis.com/auth/logging.write
https://www.googleapis.com/auth/monitoring.write
https://www.googleapis.com/auth/pubsub
https://www.googleapis.com/auth/service.management.readonly
https://www.googleapis.com/auth/servicecontrol
https://www.googleapis.com/auth/trace.append
gke-default https://www.googleapis.com/auth/devstorage.read_only
https://www.googleapis.com/auth/logging.write
https://www.googleapis.com/auth/monitoring
https://www.googleapis.com/auth/service.management.readonly
https://www.googleapis.com/auth/servicecontrol
https://www.googleapis.com/auth/trace.append
logging-write https://www.googleapis.com/auth/logging.write
monitoring https://www.googleapis.com/auth/monitoring
monitoring-read https://www.googleapis.com/auth/monitoring.read
monitoring-write https://www.googleapis.com/auth/monitoring.write
pubsub https://www.googleapis.com/auth/pubsub
service-control https://www.googleapis.com/auth/servicecontrol
service-management https://www.googleapis.com/auth/service.management.readonly
sql (deprecated) https://www.googleapis.com/auth/sqlservice
sql-admin https://www.googleapis.com/auth/sqlservice.admin
storage-full https://www.googleapis.com/auth/devstorage.full_control
storage-ro https://www.googleapis.com/auth/devstorage.read_only
storage-rw https://www.googleapis.com/auth/devstorage.read_write
taskqueue https://www.googleapis.com/auth/taskqueue
trace https://www.googleapis.com/auth/trace.append
userinfo-email https://www.googleapis.com/auth/userinfo.email
DEPRECATION WARNING: https://www.googleapis.com/auth/sqlservice account
scope and sql alias do not provide SQL instance management capabilities
and have been deprecated. Please, use
https://www.googleapis.com/auth/sqlservice.admin or sql-admin to manage
your Google SQL Service instances.
--service-account=SERVICE_ACCOUNT
The Google Cloud Platform Service Account to be used by the node VMs.
If a service account is specified, the cloud-platform and
userinfo.email scopes are used. If no Service Account is specified, the
project default service account is used.
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 beta and might change without notice. These
variants are also available:
$ gcloud container clusters create
$ gcloud alpha container clusters create
View git blame Copy permalink