1 - Overview

Service Overview

Kubernetes Engine is a service that provides lightweight virtual computing and containers, as well as a Kubernetes cluster to manage them. Users can utilize the Kubernetes environment without complex preparation by installing, operating, and maintaining the Kubernetes Control Plane.

Features

  • Standard Kubernetes Environment Configuration: The standard Kubernetes environment can be used without separate configuration through the default Kubernetes Control Plane provided. It is compatible with applications in other standard Kubernetes environments, so you can use standard Kubernetes applications without modifying the code.

  • Easy Kubernetes Deployment: Provides secure communication between worker nodes and managed control planes, and quickly provisions worker nodes, allowing users to focus on building applications on the provided container environment.

  • Convenient Kubernetes Management: Provides various management features to conveniently use the created Kubernetes cluster, such as cluster information inquiry and cluster management, namespace management, and workload management through the dashboard for enterprise environments.

Service Composition Diagram

Configuration Diagram
Figure. K8s Engine Configuration Diagram

Provided Features

Kubernetes Engine provides the following features.

  • Cluster Management: You can create and manage clusters to use the Kubernetes Engine service. After creating a cluster, you can add services necessary for operation, such as nodes, namespaces, and workloads.
  • Node Management: A node is a set of machines that run containerized applications. Every cluster must have at least one worker node to deploy applications. Nodes can be defined and used by defining a node pool. Nodes belonging to a node pool must have the same server type, size, and OS image, and multiple node pools can be created to establish a flexible deployment strategy.
  • Namespace Management: Namespace is a logical separation unit within a Kubernetes cluster, and is used to specify access permissions or resource usage limits by namespace.
  • Workload Management: Workload is an application running on Kubernetes Engine. You can create a namespace, then add or delete workloads. Workloads are created and managed item by item, such as deployments, pods, stateful sets, daemon sets, jobs, and cron jobs.
  • Service and Ingress Management: Service is an abstraction method that exposes applications running in a set of pods as a network service, and Ingress is used to expose HTTP and HTTPS paths from outside the cluster to the inside. After creating a namespace, you can create or delete services, endpoints, ingresses, and ingress classes.
  • Storage Management: When using Kubernetes Engine, you can create and manage the storage to be used. Storage is created and managed by items such as PVC, PV, and storage class.
  • Configuration Management: When there is a need to manage values that change inside a container according to multiple environments such as Dev/Prod, managing them with separate images due to environment variables is inconvenient and causes significant cost waste. In Kubernetes, you can manage environment variables or configuration values as variables from the outside so that they can be inserted when a Pod is created, and at this time, ConfigMap and Secret can be used.
  • Access Control: In cases where multiple users access a Kubernetes cluster, you can grant permissions for specific APIs or namespaces to restrict access. You can apply Kubernetes’ role-based access control (RBAC) feature to set permissions for clusters or namespaces. You can create and manage cluster roles, cluster role bindings, roles, and role bindings.

Component

Control Plane

The Control Plane is the master node role in the Kubernetes Engine service. The master node is the management node of the cluster, and it plays a role in managing other nodes in the cluster. The cluster is the basic creation unit of the Kubernetes Engine service, and it is used to manage node pools, objects, controllers, and other components within it. Users set up the cluster name, control plane, network, File Storage, and other settings, and then create a node pool within the cluster to use it. The master node assigns tasks to the cluster, monitors the status of the nodes, and plays a role in data communication between nodes.

The cluster name creation rule is as follows.

  • It starts with English and can be set within 3-30 characters using English, numbers, and special characters (-).
  • The cluster name must not be duplicated with the existing one.

Worker Node

The Worker Node is a work node in the cluster, playing a role in performing the cluster’s tasks. The Worker Node receives tasks from the cluster’s master node, performs them, and reports the task results to the cluster’s master node. All nodes created within the node pool and namespace play the role of a worker node.

The creation rule of the node pool, which is a collection of worker nodes, is as follows.

  • A node pool must have at least one node to be created for application deployment to be possible.
  • Up to 100 nodes can be created in a node pool.
  • Since the maximum number of nodes is 100, if there are 100 node pools, 1 node per node pool, and if there are 50 node pools, 2 nodes per node pool, the total number of nodes can be created freely within 100 nodes.
  • It is possible to set up Block Storage connected to the node pool.
  • It is possible to set the server type, size, and OS image for nodes belonging to the node pool, and all must be the same.
  • Auto-Scaling service allows you to set automatic node pool expansion/reduction according to the requirements of the deployed application.

Preceding Service

This is a list of services that must be pre-configured before creating this service. Please refer to the guide provided for each service and prepare in advance for more details.

Service CategoryServiceDetailed Description
NetworkingVPCA service that provides an independent virtual network in a cloud environment
NetworkingSecurity GroupA virtual firewall that controls the server’s traffic
StorageFile StorageA storage that allows multiple clients to share files over the network
  • Used as a Persistant Volume
Fig. Preceding services of Kubernetes Engine

1.1 - Monitoring Metrics

Kubernetes Engine Monitoring Metrics

The following table shows the monitoring metrics of Kubernetes Engine that can be checked through Cloud Monitoring. For detailed instructions on using Cloud Monitoring, refer to the Cloud Monitoring guide.

Performance ItemDetailed DescriptionUnit
Cluster Namespaces [Active]Number of active namespacescnt
Cluster Namespaces [Total]Total number of namespaces in the clustercnt
Cluster Nodes [Ready]Number of nodes in READY statecnt
Cluster Nodes [Total]Total number of nodes in the clustercnt
Cluster Pods [Failed]Number of failed pods in the clustercnt
Cluster Pods [Pending]Number of pending pods in the clustercnt
Cluster Pods [Running]Number of running pods in the clustercnt
Cluster Pods [Succeeded]Number of succeeded pods in the clustercnt
Cluster Pods [Unknown]Number of unknown pods in the clustercnt
Instance StatusCluster statusstatus
Namespace Pods [Failed]Number of failed pods in the namespacecnt
Namespace Pods [Pending]Number of pending pods in the namespacecnt
Namespace Pods [Running]Number of running pods in the namespacecnt
Namespace Pods [Succeeded]Number of succeeded pods in the namespacecnt
Namespace Pods [Unknown]Number of unknown pods in the namespacecnt
Namespace GPU Clock FrequencySM clock frequency in the namespaceMHz
Namespace GPU Memory UsageMemory utilization in the namespace%
Namespace GPU UsageGPU utilization in the namespace%
Node CPU Size [Allocatable]Allocatable CPU in the nodecnt
Node CPU Size [Capacity]CPU capacity in the nodecnt
Node CPU UsageCPU usage in the node%
Node CPU Usage [Request]CPU request ratio in the node%
Node CPU UsedCPU utilization in the nodestatus
Node Filesystem UsageFilesystem usage in the node%
Node Memory Size [Allocatable]Allocatable memory in the nodebytes
Node Memory Size [Capacity]Memory capacity in the nodebytes
Node Memory UsageMemory utilization in the node%
Node Memory Usage [Request]Memory request ratio in the node%
Node Memory WorkingsetMemory working set in the nodebytes
Node Network In BytesNode network received bytesbytes
Node Network Out BytesNode network transmitted bytesbytes
Node Network Total BytesNode network total bytesbytes
Node Pods [Failed]Number of failed pods in the nodecnt
Node Pods [Pending]Number of pending pods in the nodecnt
Node Pods [Running]Number of running pods in the nodecnt
Node Pods [Succeeded]Number of succeeded pods in the nodecnt
Node Pods [Unknown]Number of unknown pods in the nodecnt
Pod CPU Usage [Limit]CPU usage limit ratio in the pod%
Pod CPU Usage [Request]CPU request ratio in the pod%
Pod CPU UsageCPU usage in the pod%
Pod GPU Clock FrequencySM clock frequency in the podMHz
Pod GPU Memory UsageMemory utilization in the pod%
Pod GPU UsageGPU utilization in the pod%
Pod Memory Usage [Limit]Memory usage limit ratio in the pod%
Pod Memory Usage [Request]Memory request ratio in the pod%
Pod Memory UsageMemory usage in the podbytes
Pod Network In BytesPod network received bytesbytes
Pod Network Out BytesPod network transmitted bytesbytes
Pod Network Total BytesPod network total bytesbytes
Pod Restart ContainersContainer restart count in the podcnt
Workload Pods [Running]-cnt
Table. Kubernetes Engine Monitoring Metrics

1.2 - ServiceWatch metric

Kubernetes Engine sends metrics to ServiceWatch. The metrics provided by default monitoring are data collected at a 1‑minute interval.

Reference
To check metrics in ServiceWatch, refer to the ServiceWatch guide.

Basic Indicators

The following are the basic metrics for the namespace Kubernetes Engine.

Indicator nameDetailed descriptionUnitMeaningful statistics
cluster_upCluster upCount
  • Total
cluster_node_countCluster node countCount
  • Sum
cluster_failed_node_countCluster failed node countCount
  • Total
cluster_namespace_phase_countCluster Namespace Phase CountCount
  • Total
cluster_pod_phase_countCluster pod phase countCount
  • Total
node_cpu_allocatableNode CPU allocatable-
  • Total
node_cpu_capacityNode CPU capacity-
  • Total
node_cpu_usageNode CPU usage-
  • Total
node_cpu_utilizationNode CPU Utilization-
  • Total
node_memory_allocatableNode memory allocatableBytes
  • Total
node_memory_capacityNode memory capacityBytes
  • Total
node_memory_usageNode memory usageBytes
  • Total
node_memory_utilizationNode Memory Utilization-
  • Total
node_network_rx_bytesNode Network Receive BytesBytes/Second
  • Total
node_network_tx_bytesNode network transmission bytesBytes/Second
  • Total
node_network_total_bytesNode Network Total BytesBytes/Second
  • Total
node_number_of_running_podsNode Running Pod CountCount
  • Total
namespace_number_of_running_podsNamespace running pod countCount
  • Total
namespace_deployment_pod_countNamespace deployment pod countCount
  • Total
namespace_statefulset_pod_countNamespace StatefulSet Pod CountCount
  • Total
namespace_daemonset_pod_countNamespace DaemonSet Pod CountCount
  • Total
namespace_job_active_countNamespace job active countCount
  • Total
namespace_cronjob_active_countNamespace CronJob Active CountCount
  • Total
pod_cpu_usagePod CPU usage-
  • Total
pod_memory_usagePod memory usageBytes
  • Total
pod_network_rx_bytesPod network receive bytesBytes/Second
  • Total
pod_network_tx_bytesPod network transmission bytesBytes/Second
  • Total
pod_network_total_bytesPod network total bytesCount
  • Total
container_cpu_usageContainer CPU usage-
  • Total
container_cpu_limitContainer CPU limit-
  • Total
container_cpu_utilizationContainer CPU Utilization-
  • Total
container_memory_usagecontainer memory usageBytes
  • total
container_memory_limitcontainer memory limitBytes
  • Total
container_memory_utilizationcontainer memory utilization-
  • Total
node_gpu_countNode GPU countCount
  • Total
gpu_tempGPU Temperature-
  • Total
gpu_power_usageGPU power usage-
  • Total
gpu_utilGPU utilizationPercent
  • Total
gpu_sm_clockGPU SM Clock-
  • Total
gpu_fb_usedGPU FB usageMegabytes
  • Total
gpu_tensor_activeGPU Tensor Activation Rate-
  • Total
pod_gpu_utilPod GPU Usage RatePercent
  • Total
pod_gpu_tensor_activePod GPU Tensor Activation Rate-
  • Total
Table. Kubernetes Engine Basic Metrics

2 - How-to guides

The user can enter required information for the Kubernetes Engine through the Samsung Cloud Platform Console, select detailed options, and create a service.

Create Kubernetes Engine

You can create and use the Kubernetes Engine service from the Samsung Cloud Platform Console.

You can create and manage clusters to use the Kubernetes Engine service. After creating a cluster, you can add services needed for operation such as nodes, namespaces, and workloads.

Caution
  • You can select up to 4 Security Groups in the network settings of Kubernetes Engine.

    • If you directly add a Security Group to nodes created by Kubernetes Engine on the Virtual Server service page, it may be automatically released because it is not managed by Kubernetes Engine.
    • For nodes, the Security Group must be added/managed in the network settings of the Kubernetes Engine service.
  • Managed Security Group is automatically managed in Kubernetes Engine.

  • Do not use it for user arbitrary purposes because deleting the Managed Security Group or adding/deleting rules will automatically be reverted.

Creating a Cluster

You can create and use a Kubernetes Engine cluster service from the Samsung Cloud Platform Console.

To create a Kubernetes Engine cluster, follow these steps.

  1. All Services > Container > Kubernetes Engine Click the menu. Navigate to the Service Home page of Kubernetes Engine.
  2. Click the Create Cluster button on the Service Home page. You will be taken to the Create Cluster page.
  3. Cluster Creation page, enter the information required to create the service, and select detailed options.
  • Enter service information area, input or select the required information.
    Category
    Required or not
    Detailed description
    Cluster NameRequiredCluster Name
    • Start with an English letter and use English letters, numbers, special character (-) within 3-30 characters
    Control Plane Settings > Kubernetes VersionRequiredSelect Kubernetes Version
    Control Area Settings > Private Endpoint Access ControlSelectSelect whether to use Private Endpoint Access Control
    • After selecting Use, click Add to select resources that are allowed to access the private endpoint
    • Only resources in the same Account and same region can be registered
    • Regardless of the Use setting, the nodes of the cluster can access the private endpoint
    Control Area Settings > Public Endpoint Access/Access ControlSelectSelect whether to use Public Endpoint Access/Access Control
    • After selecting Use, enter the Allowed Access IP Range as 192.168.99.0/24
      • Set the access control IP range so that external users can access the Kubernetes API server endpoint
      • If external access is not needed, you can disable it to reduce security threats
    ServiceWatch log collectionOptionalSet whether to enable log collection so that logs for the cluster can be viewed in ServiceWatch. Log storage up to 5 GB for all services within the account is provided for free, and fees are charged based on storage volume if it exceeds 5 GB.
    • If you need to check cluster logs, it is recommended to enable the ServiceWatch log collection feature
    Cloud Monitoring log collectionOptionalSet whether to enable log collection so that logs for the cluster can be viewed in Cloud Monitoring. Up to 1 GB of log storage for all services within the account is provided for free, and any amount exceeding 1 GB is deleted sequentially.
    Network SettingsRequiredNetwork connection settings for node pool
    • VPC: Select a pre-created VPC
    • Subnet: Choose a standard Subnet to use from the subnets of the selected VPC
    • Security Group: after clicking the Select button, select a Security Group in the Security Group Selection popup
      • Up to 4 Security Group can be selected
    File Storage SettingsRequiredSelect the file storage volume to be used in the cluster
    • Default Volume (NFS): Click the Search button and then select the file storage in the File Storage Selection popup. The default Volume file storage can only use the NFS format.
    Table. Kubernetes Engine Service Information Input Items
    • Additional Information Input Enter or select the required information in the area.
      Category
      Required
      Detailed description
      TagSelectAdd Tag
      • Up to 50 can be added per resource
      • After clicking the Add Tag button, enter or select Key, Value values
      Table. Kubernetes Engine Additional Information Input Items
  1. Summary Check the detailed information and estimated billing amount generated in the panel, and click the Complete button.
    • When creation is complete, check the created resources on the Cluster List page.

Check cluster detailed information

Kubernetes Engine service can view and edit the full resource list and detailed information. Cluster Details page consists of Details, Node Pools, Tags, Activity History tabs.

To view detailed cluster information, follow the steps below.

  1. All Services > Container > Kubernetes Engine Click the menu. Navigate to the Service Home page of Kubernetes Engine.
  2. Click the Cluster menu on the Service Home page. Navigate to the Cluster List page.
  3. Click the resource (cluster) you want to view detailed information for on the Cluster List page. You will be taken to the Cluster Details page.
    • Cluster Details page displays the cluster’s status information and detailed information, and consists of Details, Node Pool, Tags, Job History tabs.
      CategoryDetailed description
      Cluster StatusKubernetes Engine cluster status
      • Creating: Creating
      • Running: Created / Running
      • Error: Error occurred
      Service TerminationButton to terminate a Kubernetes Engine cluster
      • To terminate the Kubernetes Engine service, you must delete all node pools added to the cluster
      • If the service is terminated, the running service may be stopped immediately, so termination is necessary after considering the impact of service interruption
      Table. Status Information and Additional Functions

Detailed Information

Cluster List page allows you to view detailed information of the selected resource and, if necessary, edit the information.

Category
Detailed description
serviceservice name
Resource TypeResource Type
SRNUnique resource ID in Samsung Cloud Platform
Resource NameResource Name
  • In the Kubernetes Engine service, it refers to the cluster name
Resource IDUnique resource ID in the service
CreatorUser who created the service
Creation TimeTime the service was created
EditorUser who modified the service information
Modification DateTimeDate and time when service information was modified
Cluster nameCluster name
LLM EndpointLLM Endpoint information
Control Plane SettingsCheck assigned Kubernetes control plane (Control Plane) version and access permission range
  • If there is a Kubernetes version of the control plane that can be upgraded, click the Edit icon to perform Cluster Version Upgrade. See Cluster Version Upgrade for details
  • Click the Admin Kubeconfig Download/User Kubeconfig Download button of the private endpoint address to download the kubeconfig settings for each role as a yaml document
  • Click the Edit icon of private endpoint access control to modify usage and allowed resources
  • Click the Admin Kubeconfig Download/User Kubeconfig Download button of the public endpoint address to download the kubeconfig settings for each role as a yaml document
  • Click the Edit icon of public endpoint access/control to modify usage and allowed IP range
  • Click the Edit icon of ServiceWatch log collection to change usage. When log collection is enabled, view the cluster control plane’s Audit/Event logs in ServiceWatch > Log Group
  • Click the Edit icon of Cloud Monitoring log collection to change usage. When log collection is enabled, view the cluster control plane’s Audit/Event logs in Cloud Monitoring > Log Analysis
Network SettingsCheck VPC, Subnet, and Security Group information set when creating a Kubernetes Engine cluster
  • Click each setting to view detailed information on the detail page
  • If a Security Group change is needed, click the Edit icon to configure
File Storage SettingsIf you click the volume name, you can view detailed information on the storage detail page
Table. Cluster detailed information items
Reference
  • The version of Kubernetes Engine is denoted in order [major].[minor].[patch], and you can upgrade only one minor version at a time.
    • Example: version 1.11.x > 1.13.x (Not possible) / version 1.11.x > 1.12.x (Possible)
  • If you are using a Kubernetes version that has reached end of support or a version that is scheduled to reach end of support, a red exclamation mark will appear to the right of the version. If this icon appears, we recommend upgrading the Kubernetes version.

Node Pool

You can view, add, modify, or delete cluster node pool information. For detailed information on using node pools, refer to Managing Nodes.

CategoryDetailed description
Add node poolAdd node pool to the current cluster
Node Pool InformationProvides node pool list view and management functions
  • Node Information: Displays node name, version, and status information
  • Node Pool Upgrade: Node pool version upgrade
  • Node Pool Deletion: Delete node
Table. Node Pool Items
Reference

If a red exclamation mark icon appears on the version of the node pool information, the server OS of that node pool is not supported in higher versions of Kubernetes. To ensure stable service, the node pool server OS must be upgraded.

  • If you want to upgrade the node pool version, you must delete the node pool and then create a new node pool with a higher server OS version.

Tag

On the Cluster List page, you can view the tag information of the selected resource, and you can add, modify, or delete it.

CategoryDetailed description
Tag ListTag List
  • Key and Value information of tags can be checked
  • Up to 50 tags can be added per resource
  • When entering tags, search and select from the list of previously created Keys and Values
Table. Cluster Tag Tab Items

Work History

You can view the operation history of the selected resource on the Cluster List page.

CategoryDetailed description
Work History ListResource Change History
  • Work details, work date and time, resource type, resource name, work result, can check operator information
  • Work History List list, when you click the corresponding resource, the Work History Details popup opens
Table. Cluster Job History Tab Detailed Information Items

Managing Cluster Resources

To manage cluster resources, we provide cluster version upgrade, kubeconfig download, and control plane logging modification features.

Caution
To use Kubernetes Engine, you need at least read permissions for VPC, VPC Subnet, Security Group, FileStorage, and Virtual Server.
Security Group and Virtual Server are created/deleted by Kubernetes Engine for lifecycle management purposes even without create/delete permissions, and the creator/modifier is indicated as System.

Cluster Version Upgrade

If there is a version that can be upgraded from the cluster’s Kubernetes version, you can perform the upgrade on the Cluster Details page.

Reference
  • Before the cluster upgrade, check the following items.
    • Check if the cluster’s status is Running
    • Check that all node pool statuses of the cluster are Running or Deleting
    • Check that all node pool versions in the cluster are the same version as the cluster
    • Check whether automatic scaling/downsizing of all node pools in the cluster and node auto-recovery feature are unused
  • After upgrading the cluster, proceed with node pool upgrade. The control plane and node pool upgrades of the Kubernetes cluster are performed separately.
  • You can only upgrade one minor version at a time.
    • Example: version 1.12.x > 1.13.x (possible) / version 1.11.x > 1.13.x (not possible)
  • After an upgrade, you cannot perform a downgrade or rollback, so to use the previous version again you must create a new cluster.

Caution
  • Since user systems using end-of-support Kubernetes versions may have security vulnerabilities, upgrade the control plane and node pool versions directly in the Samsung Cloud Platform Console.
    • No additional cost will be incurred due to the upgrade.
  • Please perform compatibility testing for the upgrade version in advance to ensure stable system operation for users.

Cluster version upgrade preparation

There is no need to delete and recreate API objects when upgrading the cluster version. For the transitioned API, all existing API objects can be read and updated using the new API version. However, due to deprecated APIs in older Kubernetes versions, you may be unable to read or modify existing objects or create new ones. Therefore, to ensure system stability, it is recommended to migrate clients and manifests before the upgrade.

Migrate the client and manifest using the following method.

Reference
Since the deprecated API differs for each cluster version, the scope of application and system impact may also vary. For detailed explanation, refer to the Kubernetes official documentation > Deprecation Guide.

Upgrade Cluster and Node Pool Versions

To update the cluster and node pool, follow the steps below.

  1. All Services > Container > Kubernetes Engine Click the menu. Navigate to the Service Home page of Kubernetes Engines.
  2. Click the Cluster menu on the Service Home page. It navigates to the Cluster List page.
  3. Cluster List page, click the resource (cluster) to upgrade the version. It navigates to the Cluster Details page.
  4. Click the Edit icon of Kubernetes version on the Cluster Details page. Navigate to the Cluster Version Upgrade popup.
  5. Select the Kubernetes version to upgrade, and click the Confirm button.
    • It may take a few minutes for the cluster upgrade to complete.
    • During the upgrade, the cluster status is shown as Updating, and when the upgrade is complete, it is shown as Running.
  6. When the upgrade is complete, select the Node Pool tab. Go to the Node Pool page.
  7. Click the More button of the node pool item and click Node Pool Upgrade. It will move to the Node Pool Version Upgrade popup window.
  8. Node Pool Version Upgrade After checking the message in the popup window, click the Confirm button.
    • It may take a few minutes until the node pool upgrade is completed.
    • During the upgrade, the node pool status is shown as Updating, and when the upgrade is complete, it is shown as Running.

kubeconfig download

You can download the admin/user kubeconfig settings of the cluster’s public and private endpoints as a yaml document.

To download the cluster’s kubeconfig settings, follow the steps below.

  1. All Services > Container > Kubernetes Engine Click the menu. Navigate to the Service Home page of Kubernetes Engines.
  2. Click the Cluster menu on the Service Home page. You will be taken to the Cluster List page.
  3. On the Cluster List page, click the resource (cluster) to download the kubeconfig. You will be taken to the Cluster Details page.
  4. On the Cluster Details page, click the Admin kubeconfig download/User kubeconfig download button for the desired endpoint.
    • You can download the kubeconfig file in yaml format for each permission.

Edit Private Endpoint Access Control

You can change the private endpoint access control settings of the cluster.

  1. Click the All Services > Container > Kubernetes Engine menu. Navigate to the Service Home page of Kubernetes Engines.
  2. Click the Cluster menu on the Service Home page. Navigate to the Cluster List page.
  3. On the Cluster List page, click the resource (cluster) to modify private endpoint access control. You will be taken to the Cluster Details page.
  4. Cluster Details page, click the Edit icon of Private Endpoint Access Control. Navigate to the Private Endpoint Access Control Edit popup.
  5. Private Endpoint Access Control Modification in the popup window, check the Private Endpoint Access Control Usage, add the allowed access resources, and click the Confirm button.

Modify public endpoint access/access control

You can change the public endpoint access control settings of the cluster.

  1. All Services > Container > Kubernetes Engine Click the menu. Navigate to the Service Home page of Kubernetes Engines.
  2. Click the Cluster menu on the Service Home page. Navigate to the Cluster List page.
  3. Click the resource (cluster) to modify public endpoint access control on the Cluster List page. You will be taken to the Cluster Details page.
  4. Click the Edit icon of Public Endpoint Access/Access Control on the Cluster Details page. It moves to the Public Endpoint Access/Access Control Edit popup.
  5. Public Endpoint Access/Access Control Modification In the popup window, check the Public Endpoint Access Control Use status and add the allowed IP range, then click the Confirm button.

Modify control area log collection settings

You can change the log collection settings of the cluster’s control plane. Detailed logs of the cluster can be viewed in the ServiceWatch service or the Cloud Monitoring service.

Reference

Even if you set up Cloud Monitoring log collection, you can still view the cluster logs.

  • However, since the Cloud Moniotring log collection feature is scheduled for termination, we recommend using ServiceWatch log collection.

Follow the steps below to change the control plane log collection settings of the cluster.

  1. Click the All Services > Container > Kubernetes Engine menu. Navigate to the Service Home page of Kubernetes Engines.
  2. Click the Cluster menu on the Service Home page. Navigate to the Cluster List page.
  3. Cluster List page, click the resource (cluster) to modify control plane logging. Cluster Details page will be opened.
  4. Cluster Details page, click the Edit icon of ServiceWatch Log Collection. It will navigate to the ServiceWatch Log Collection popup.
    • Cloud Monitoring log collection feature can also be set the same way.
  5. In the ServiceWatch log collection popup, after checking the use of ServiceWatch log modification, click the Confirm button.
Reference

When using log collection, you can view the Audit/Event logs of the cluster control area in each service. Detailed logs can be viewed on the next page.

Security Group Edit

You can modify the cluster’s Security Group.

Caution
  • In the network settings of Kubernetes Engine, you can select multiple Security Groups. (up to 4)

    • If you directly add a Security Group on the Virtual Server service page to nodes created by Kubernetes Engine, they may be automatically released because they are not managed by Kubernetes Engine.
    • For nodes, the Security Group must be added/managed in the network settings of the Kubernetes Engine service.
  • Managed Security Group is automatically managed in Kubernetes Engine.

  • Do not use it for any user-defined purpose because if you delete a Managed Security Group or add/delete rules, it will automatically be restored.

To modify the cluster’s Security Group, follow the steps below.

  1. All Services > Container > Kubernetes Engine menu를 클릭하세요. Kubernetes Engines의 Service Home 페이지로 이동합니다.
  2. Click the Cluster menu on the Service Home page. You will be taken to the Cluster List page.
  3. Cluster List page, click the resource (cluster) whose Security Group you want to modify. You will be taken to the Cluster Details page.
  4. Click the Edit icon of Security Group on the Cluster Details page. It will navigate to the Security Group Edit popup.
  5. After selecting or deselecting the Security Group to modify, click the Confirm button.

Cancel Cluster

Caution
If you terminate the cluster, all node pools will be deleted, and all data in all pods within the cluster will be permanently deleted.

To cancel the cluster, follow the steps below.

  1. All Services > Container > Kubernetes Engine Click the menu. Navigate to the Service Home page of Kubernetes Engines.
  2. Click the Cluster menu on the Service Home page. You will be taken to the Cluster List page.
  3. Click the resource (cluster) on the Cluster List page to view detailed information. You will be taken to the Cluster Detail page.
  4. On the Cluster Details page, click Service Termination.
  5. Service termination After checking the contents in the popup window, click the Confirm button.

2.1 - Managing Namespaces

A namespace is a logical separation unit within a Kubernetes cluster, and is used to specify access permissions or resource usage limits by namespace.

Create namespace

To create a namespace, follow these steps.

  1. Click on the menu for all services > Container > Kubernetes Engine. It moves to the Service Home page of Kubernetes Engine.
  2. On the Service Home page, click the Namespace menu. It moves to the Namespace List page.
  3. On the Namespace List page, select the cluster where you want to create a namespace from the gear button at the top left, then click Create Object.
  4. Object creation popup where you enter object information and click the Confirm button.
Reference
For more information about creating objects, see the Kubernetes official documentation > Kubernetes objects.

Check namespace details

You can check the namespace status and detailed information on the namespace detail page.

To check namespace details, follow the next procedure.

  1. Click on the menu for all services > Container > Kubernetes Engine. It moves to the Service Home page of Kubernetes Engine.
  2. On the Service Home page, click the Namespace menu. It moves to the Namespace List page.
  3. On the Namespace List page, select the cluster where the namespace that requires detailed information is located from the gear button in the top left, and then click OK.
  4. On the Namespace List page, select and click the item you want to check the details for. It moves to the Namespace Details page.
ClassificationDetailed Description
Status IndicatorDisplays the current status of the namespace
Namespace DeletionYou can delete a namespace
  • A namespace that contains workloads cannot be deleted. To delete a namespace, you must delete all connected workloads
Detailed InformationCheck the namespace’s Account information and metadata information
YAMLYou can modify the namespace in the YAML editor
  • Click the Edit button, modify the namespace, and then click the Save button to apply the changes
EventCheck events that occurred within the namespace
PodCheck pod information in the namespace
Account InformationAccount name, location, creation time, etc., basic information about the Account can be checked
Metadata InformationCheck the metadata information of the namespace
Table. Namespace details items

Deleting a namespace

To delete a namespace, follow these steps.

  1. Click on the menu for all services > Container > Kubernetes Engine. It moves to the Service Home page of Kubernetes Engine.
  2. On the Service Home page, click the Namespace menu. It moves to the Namespace List page.
  3. On the Namespace List page, select the cluster where the namespace you want to delete is located from the gear button at the top left, then click the OK button.
  4. On the Namespace List page, select and click the item you want to check the details for. It moves to the Namespace Details page.
  5. Click Namespace Delete on the Namespace Details page.
  6. When the Notification Confirmation Window appears, click the OK button.
Caution
  • Select the item to be deleted from the namespace list page, then click Delete to delete the selected namespace. A namespace that contains a workload cannot be deleted. To delete a namespace, delete all associated workloads.

2.2 - Managing Workloads

The workload is an application running on Kubernetes Engine. You can create a namespace and then add or remove workloads. Workloads are created and managed item by item, such as deployments, pods, stateful sets, daemon sets, jobs, and cron jobs.

Managing Deployment

Deployment is a resource that provides updates for pods and replica sets. You can create a deployment in a workload, check its details, or delete it.

Creating Deployment

To create a deployment, follow these steps.

  1. Click on the menu for all services > Container > Kubernetes Engine. It moves to the Service Home page of Kubernetes Engine.
  2. On the Service Home page, click Deployment under the Workload menu. It moves to the List Deployment page.
  3. Deployment list page, select the cluster and namespace from the gear button at the top left, then click Create object.
  4. Object Creation Popup where you enter object information and click the Confirm button.
    • The following is an example of a .yaml file showing the required fields and object Spec for deployment creation. (application/deployment.yaml)
      Color mode
       apiVersion: apps/v1
       kind: Deployment
       metadata:
         name: nginx-deployment
       spec:
         selector:
            matchLabels:
               app: nginx
         replicas: 2 # tells deployment to run 2 pods matching the template
         template:
           metadata:
              labels:
                 app: nginx
           spec:
              containers:
              - name: nginx
                image: nginx:1.14.2
                ports:
                - containerPort: 80
       apiVersion: apps/v1
       kind: Deployment
       metadata:
         name: nginx-deployment
       spec:
         selector:
            matchLabels:
               app: nginx
         replicas: 2 # tells deployment to run 2 pods matching the template
         template:
           metadata:
              labels:
                 app: nginx
           spec:
              containers:
              - name: nginx
                image: nginx:1.14.2
                ports:
                - containerPort: 80
      Code block. Essential fields and object Spec for deployment creation
Reference
For more information about the concept of deployment and object creation, please refer to the Kubernetes official documentation > Deployment.

Check Deployment Details

To check the deployment details, follow the next procedure.

  1. Click on the menu for all services > Container > Kubernetes Engine. It moves to the Service Home page of Kubernetes Engine.
  2. On the Service Home page, click Deployment under the Workload menu. It moves to the Deployment List page.
  3. Deployment list page, select the cluster and namespace from the gear button at the top left, then click OK.
  4. Deployment List page, select the item you want to check the detailed information. It moves to the Deployment Details page.
    • Selecting Show System Objects at the top of the list displays all items except Kubernetes object entries.
  5. Click each tab to check service information.
Classification
Detailed Description
Deployment DeletionDelete the deployment
Detailed InformationCheck detailed information of deployment
YAMLYou can modify the resource file of the deployment in the YAML editor
  • Click the Edit button, modify the resource, and then click the Save button to apply the changes
EventCheck the event that occurred within the deployment
PodCheck pod information of deployment
  • Pod is the smallest computing unit that can be created, managed, and deployed in Kubernetes Engine
Account InformationAccount name, location, creation time, etc. Basic information about the Account can be checked
Metadata InformationCheck the metadata information of the deployment
Object InformationCheck object information of deployment
Table. Deployment details items

Deleting Deployment

To delete a deployment, follow these steps.

  1. Click on the menu for all services > Container > Kubernetes Engine. It moves to the Service Home page of Kubernetes Engine.
  2. On the Service Home page, click Deployment under the Workload menu. It moves to the Deployment List page.
  3. Deployment list page, select the cluster and namespace from the gear button at the top left, then click OK.
  4. Deployment List page, select the item you want to delete. It moves to the Deployment Details page.
  5. Deployment Details page, click Delete Deployment.
  6. When the Notification Confirmation Window appears, click the Confirm button.
Caution
In the deployment list page, you can delete the selected deployment by selecting the item you want to delete and then clicking Delete.

Managing Pods

A Pod is the smallest computing unit that can be created, managed, and deployed in Kubernetes, referring to a group of one or more containers. You can create pods in workloads and view or delete their details.

Creating a Pod

To create a pod, follow the next procedure.

  1. Click on the menu for all services > Container > Kubernetes Engine. It moves to the Service Home page of Kubernetes Engine.
  2. On the Service Home page, click Pods under the Workload menu. It moves to the Pod List page.
  3. Pod list page, select the cluster and namespace from the gear button at the top left, then click Create object.
  4. Object Creation Popup where you enter object information and click the Confirm button.
Reference
For more information about the concept of pods and object creation, please refer to the Kubernetes official documentation > Pods.

Check Pod Details

To check the pod details, follow the next procedure.

  1. Click on the menu for all services > Container > Kubernetes Engine. It moves to the Service Home page of Kubernetes Engine.
  2. On the Service Home page, click Pods under the Workload menu. It moves to the Pod List page.
  3. Pod list page, select the cluster and namespace from the gear button at the top left, then click OK.
  4. Pod List page, select the item you want to check the detailed information. It moves to the Pod Detail page.
    • Selecting Show System Objects at the top of the list displays all items except Kubernetes object entries.
  5. Click each tab to check the service information.
Classification
Detailed Description
Status IndicatorIndicates the current status of the pod
Delete PodDelete the pod
Detailed InformationYou can check the detailed information of the pod
YAMLYou can modify the pod’s resource file in the YAML editor
  • Click the Edit button, modify the resource, and then click the Save button to apply the changes
EventCheck the event that occurred within the pod
LogIf you select a container, you can check the container information that the pod has
Account InformationAccount name, location, creation time, etc., basic information about the Account can be checked
Metadata InformationCheck the pod’s metadata information
Object InformationCheck the object information of the pod
Initialization Container InformationCheck the initialization container information of the pod
Container InformationCheck container information of the pod
Table. Pod detailed information items

Deleting Pods

To delete a pod, follow these steps.

  1. Click on the menu for all services > Container > Kubernetes Engine. It moves to the Service Home page of Kubernetes Engine.
  2. On the Service Home page, click Pods under the Workload menu. It moves to the Pod List page.
  3. Pod List page, select the cluster and namespace from the gear button at the top left, then click OK.
  4. Select the item to delete on the Pod List page. It moves to the Pod Detail page.
  5. Pod Details page, click Delete Pod.
  6. When the Notification Confirmation Window appears, click the OK button.
Caution
In the pod list page, you can delete the selected pod by clicking Delete after selecting the item you want to delete.

Managing StatefulSets

A StatefulSet is a workload API object used to manage stateful applications, you can create, describe, or delete StatefulSet in the workload.

Creating a StatefulSet

To create a stateful set, follow these steps.

  1. Click on the menu for all services > Container > Kubernetes Engine. It moves to the Service Home page of Kubernetes Engine.
  2. On the Service Home page, click StatefulSet under the Workload menu. It moves to the Statefulset List page.
  3. StatefulSet list page, select the cluster and namespace from the gear button at the top left, then click Create object.
  4. Object creation popup where you enter object information and click the OK button.
Reference
For more information about the StatefulSet concept and object creation, see Kubernetes official documentation > StatefulSet.

Checking Detailed Information of StatefulSet

To view detailed information about a StatefulSet, follow these steps.

  1. Click on the menu for all services > Container > Kubernetes Engine. It moves to the Service Home page of Kubernetes Engine.
  2. On the Service Home page, click StatefulSet under the Workload menu. It moves to the StatefulSet List page.
  3. StatefulSet list page, select the cluster and namespace from the gear button at the top left, then click OK.
  4. StatefulSet list page, select the item you want to check the detailed information. It moves to the StatefulSet detail page.
    • Selecting Show System Objects at the top of the list displays all items except Kubernetes object entries.
  5. Click each tab to check service information.
Classification
Detailed Description
StatefulSet deletionDelete the StatefulSet
Detailed InformationCheck detailed information of StatefulSet
YAMLYou can modify the resource file of the StatefulSet in the YAML editor
  • Click the Edit button, modify the resource, and then click the Save button to apply the changes
EventCheck the event that occurred within the stateful set
PodCheck pod information of StatefulSet
Account InformationAccount name, location, creation time, etc., basic information about the Account can be checked
Metadata InformationCheck the metadata information of the StatefulSet
Object InformationCheck object information of the StatefulSet
Table. Detailed information items of StatefulSet

Deleting a StatefulSet

To delete a stateful set, follow these steps.

  1. Click on the menu for all services > Container > Kubernetes Engine. It moves to the Service Home page of Kubernetes Engine.
  2. On the Service Home page, click StatefulSet under the Workload menu. It moves to the StatefulSet List page.
  3. StatefulSet list page, select the cluster and namespace from the gear button at the top left, then click OK.
  4. StatefulSet list page, select the item you want to delete. It moves to the StatefulSet details page.
  5. StatefulSet details page, click Delete StatefulSet.
  6. When the Notification Confirmation Window appears, click the Confirm button.
Caution
In the stateful set list page, you can delete the selected stateful set by selecting the item you want to delete and then clicking Delete.

Managing DaemonSets

A daemon set is a resource that allows all nodes or some nodes to run a copy of a pod. You can create a daemon set in a workload, check its details, or delete it.

Creating a DaemonSet

To create a daemon set, follow these steps.

  1. Click on the menu for all services > Container > Kubernetes Engine. It moves to the Service Home page of Kubernetes Engine.
  2. On the Service Home page, click DaemonSet under the Workload menu. It moves to the DaemonSet list page.
  3. On the DaemonSet list page, select the cluster and namespace from the gear button in the top left, then click Create object.
  4. Object creation popup where you enter object information and click the OK button.
Reference
For more information about the concept of daemon sets and object creation, please refer to the Kubernetes official documentation > DaemonSet.

Checking DaemonSet Details

To check the details of the daemon set, follow the next procedure.

  1. Click on the menu for all services > Container > Kubernetes Engine. It moves to the Service Home page of Kubernetes Engine.
  2. On the Service Home page, click DaemonSet under the Workload menu. It moves to the DaemonSet list page.
  3. On the DaemonSet list page, select the cluster and namespace from the gear button at the top left, then click OK.
  4. Select an item to check the detailed information on the DaemonSet list page. It moves to the DaemonSet details page.
    • Selecting Show System Objects at the top of the list displays all items except Kubernetes object entries.
  5. Click each tab to check service information.
Classification
Detailed Description
Delete DaemonSetDelete DaemonSet
Detailed InformationCheck detailed information of daemon set
YAMLYou can modify the daemon set’s resource file in a YAML editor
  • Click the Edit button, modify the resource, and then click the Save button to apply the changes
EventCheck the event that occurred within the daemon set
PodCheck pod information of daemon set
Account InformationAccount name, location, creation time, etc., basic information about the Account can be checked
Metadata InformationCheck the metadata information of the daemon set
Object InformationCheck the object information of the daemon set
Table. Detailed information items of daemon set

Deleting DaemonSets

To delete a daemon set, follow these steps.

  1. Click on the menu for all services > Container > Kubernetes Engine. It moves to the Service Home page of Kubernetes Engine.
  2. On the Service Home page, click DaemonSet under the Workload menu. It moves to the DaemonSet list page.
  3. On the DaemonSet list page, select the cluster and namespace from the gear button at the top left, then click OK.
  4. Select the item to delete from the DaemonSet list page. It will move to the DaemonSet details page.
  5. DaemonSet details page, click Delete DaemonSet.
  6. When the Notification Confirmation Window appears, click the Confirm button.
Caution
In the daemon set list page, you can delete the selected daemon set by selecting the item you want to delete and then clicking Delete.

Job Management

A job is a resource that creates one or more pods and continues to run them until a specified number of pods complete successfully. You can create a job in a workload and view or delete its details.

Creating a Job

To create a job, follow these steps.

  1. Click on the menu for all services > Container > Kubernetes Engine. It moves to the Service Home page of Kubernetes Engine.
  2. On the Service Home page, click Job under the Workload menu. It moves to the Job List page.
  3. Job list page, select the cluster and namespace from the gear button at the top left, then click Create object.
  4. Object Creation Popup where you enter object information and click the Confirm button.
Reference
For more information on the concept of a job and object creation, see Kubernetes official documentation > Job.

Check Job Details

To check the job details, follow the next procedure.

  1. Click on the menu for all services > Container > Kubernetes Engine. It moves to the Service Home page of Kubernetes Engine.
  2. On the Service Home page, click Job under the Workload menu. It moves to the Job List page.
  3. Job List page, select the cluster and namespace from the gear button at the top left, then click Confirm.
  4. Job list page, select the item you want to check the details. It moves to the Job details page.
    • Selecting Show System Objects at the top of the list displays all items except Kubernetes object entries.
  5. Click each tab to check service information.
Classification
Detailed Description
Job DeleteDelete Job
Detailed InformationCheck detailed information of the book
YAMLYou can modify the job’s resource file in the YAML editor
  • Click the Edit button, modify the resource, and then click the Save button to apply the changes
EventCheck the event that occurred in the job
PodCheck the pod information of the job
Account InformationAccount name, location, creation time, etc., basic information about the Account can be checked
Metadata InformationCheck the metadata information of the book
Object InformationCheck the object information of the book
Table. Job detailed information items

Delete Job

To delete a job, follow these steps.

  1. Click on the menu for all services > Container > Kubernetes Engine. It moves to the Service Home page of Kubernetes Engine.
  2. On the Service Home page, click Job under the Workload menu. It moves to the Job List page.
  3. Job list page, select the cluster and namespace from the gear button at the top left, then click OK.
  4. Job List page, select the item you want to delete. It moves to the Job Detail page.
  5. Click Job Delete on the Job Details page.
  6. When the Notification Confirmation Window appears, click the Confirm button.
Caution
On the job list page, you can delete the selected job by selecting the item you want to delete and then clicking Delete.

Managing Cron Jobs

A cron job is a resource that runs a job periodically according to a schedule written in cron format. It can be used to execute repetitive tasks at a fixed interval, such as backup and report creation. You can create a cron job in the workload and check or delete detailed information.

Creating a Cron Job

To create a cron job, follow these steps.

  1. Click on the menu for all services > Container > Kubernetes Engine. It moves to the Service Home page of Kubernetes Engine.
  2. On the Service Home page, click CronJob under the Workload menu. It moves to the CronJob List page.
  3. CronJob list page, select the cluster and namespace from the gear button at the top left, then click Create Object.
  4. Object Creation Popup where you enter object information and click the Confirm button.
Reference
For more information on the concept of CronJob and object creation, please refer to the Kubernetes official documentation > CronJob.

Check Cron Job Details

To check the detailed information of the cron job, follow the next procedure.

  1. Click on the menu for all services > Container > Kubernetes Engine. It moves to the Service Home page of Kubernetes Engine.
  2. On the Service Home page, click CronJob under the Workload menu. It moves to the CronJob List page.
  3. CronJob list page, select the cluster and namespace from the gear button at the top left, then click OK.
  4. Cron Job List page, select the item you want to check the detailed information. It moves to the Cron Job Detail page.
    • Selecting Show System Objects at the top of the list displays all items except Kubernetes object entries.
  5. Click each tab to check the service information.
Classification
Detailed Description
Delete Cron JobDelete a cron job
Detailed InformationPossible to check detailed information of cron job
YAMLYou can modify the resource file of CronJob in YAML editor
  • Click the Edit button, modify the resource, and then click the Save button to apply the changes
EventCheck the event that occurred within the cron job
JobCheck the job information of Cron Job. If you select a job item, it moves to the job detail page
Account InformationAccount name, location, creation time, etc., basic information about the Account can be checked
Metadata InformationCheck the metadata information of the cron job
Object InformationCheck the object information of the cron job
Table. Cron job detail items

Deleting a Cron Job

To delete a cron job, follow these steps.

  1. Click on the menu for all services > Container > Kubernetes Engine. It moves to the Service Home page of Kubernetes Engine.
  2. On the Service Home page, click CronJob under the Workload menu. It moves to the CronJob List page.
  3. CronJob list page, select the cluster and namespace from the gear button at the top left, then click OK.
  4. Select the item to be deleted from the Cron Job List page. It will move to the Cron Job Details page.
  5. Cron Job Details page, click Delete Cron Job.
  6. When the Notification Confirmation Window appears, click the Confirm button.
Caution
In the cron job list page, you can delete the selected cron job by selecting the item you want to delete and then clicking Delete.

2.3 - Managing Services and Ingress

The service is an abstraction method that exposes applications running in a set of pods as a network service, and ingress is used to expose HTTP and HTTPS paths from outside the cluster to inside the cluster. After creating a namespace, you can create or delete services, endpoints, ingresses, and ingress classes.

Reference

Service, Endpoint, Ingress, IngressClass The service is set to the default cluster (namespace) selected when the service is created. Even if you select other items in the list, the default cluster (namespace) setting is maintained.

  • To select a different cluster (namespace), click the gear button on the right side of the list. In the Cluster/Namespace Settings popup window, select the cluster and namespace you want to change and click the OK button. You can see the services created in the selected cluster/namespace.

Managing Services

You can create a service, check detailed information, or delete it.

Creating a Service

To create a service, follow these steps.

  1. Click All services > Container > Kubernetes Engine menu. It moves to the Service Home page of Kubernetes Engine.
  2. On the Service Home page, click Service under the Service and Ingress menu. It moves to the Service List page.
  3. On the Service List page, select the cluster and namespace from the gear button at the top left, then click Create Object.
  4. Object Creation Popup에서 오브젝트 정보를 입력하고 Confirm 버튼을 클릭하세요.
Reference
For more information on the concept of a service and how to create an object, see the Kubernetes official documentation > Service.

Check Service Details

To check the service details, follow the next procedure.

  1. Click all services > Container > Kubernetes Engine menu. It moves to the Service Home page of Kubernetes Engine.
  2. Service Home page, click Service under the Service and Ingress menu. It moves to the Service List page.
  3. On the Service List page, select the cluster and namespace from the gear button at the top left, then click Confirm.
  4. On the Service List page, select the item you want to check the detailed information. It moves to the Service Detail page.
    • Selecting System Object View at the top of the list displays all items except Kubernetes object entries.
  5. Click each tab to check the service information.
Classification
Detailed Description
Service DeletionDelete the service
Detailed InformationPossible to check detailed information of the service
YAMLThe service’s resource file can be modified in the YAML editor
  • Click the Edit button, modify the resource, and then click the Save button to apply the changes
EventCheck the events that occurred within the service
Account InformationAccount name, location, creation time, etc., basic information about the Account can be checked
Metadata InformationCheck the metadata information of the service
Object InformationCheck the object information of the service
Table. Service Detail Information Items

Delete Service

To delete a service, follow these steps.

  1. Click all services > Container > Kubernetes Engine menu. It moves to the Service Home page of Kubernetes Engine.
  2. On the Service Home page, click Services under the Services and Ingress menu. It moves to the Service List page.
  3. On the Service List page, select the cluster and namespace from the gear button at the top left, then click OK.
  4. Select the item to be deleted on the Service List page. It moves to the Service Detail page.
  5. Click Service Delete on the Service Details page.
  6. When the Notification Confirmation Window appears, click the Confirm button.
Caution

On the service list page, after selecting the item you want to delete, clicking Delete allows you to delete the selected service.

  • Services connected to the ingress cannot be deleted. To delete a service, delete all connected ingresses.

Managing Endpoints

You can create an endpoint and check or delete detailed information.

Creating Endpoints

To create an endpoint, follow these steps.

  1. Click All services > Container > Kubernetes Engine menu. It moves to the Service Home page of Kubernetes Engine.
  2. Service Home page, click Endpoints under the Services and Ingress menu. It moves to the Endpoint List page.
  3. On the Endpoint List page, select the cluster and namespace from the gear button at the top left, then click Create Object.
  4. Object Creation Popup에서 오브젝트 정보를 입력하고 확인 버튼을 클릭하세요.

Check Endpoint Details

To check the endpoint details, follow the next procedure.

  1. Click all services > Container > Kubernetes Engine menu. It moves to the Service Home page of Kubernetes Engine.
  2. Service Home page, click Endpoints under the Services and Ingress menu. It moves to the Endpoint List page.
  3. Endpoint list page, select the cluster and namespace from the gear button at the top left, then click OK.
  4. On the Endpoint List page, select the item you want to check the detailed information. It moves to the Endpoint Detail page.
    • Selecting System Object View at the top of the list displays all items except Kubernetes object entries.
  5. Click each tab to check the service information.
Classification
Detailed Description
Endpoint DeletionDelete endpoint
Detailed InformationPossible to check detailed information of the endpoint
YAMLThe resource file of the endpoint can be modified in the YAML editor
  • Click the Edit button, modify the resource, and then click the Save button to apply the changes
EventCheck the event that occurred within the endpoint
Account InformationAccount name, location, creation time, etc., basic information about the Account can be checked
Metadata InformationCheck the metadata information of the endpoint
Object InformationCheck the object information of the endpoint
Table. Endpoint details items

Deleting Endpoints

To delete an endpoint, follow these steps.

  1. Click all services > Container > Kubernetes Engine menu. It moves to the Service Home page of Kubernetes Engine.
  2. On the Service Home page, click Endpoints under the Services and Ingress menu. It moves to the Endpoint List page.
  3. Endpoint list page, select cluster and namespace from the gear button at the top left, then click OK.
  4. Select the item to delete on the Endpoint List page. It moves to the Endpoint Detail page.
  5. Endpoint Details page, click Delete Endpoint.
  6. When the Notification Confirmation Window appears, click the Confirm button.
Reference
On the endpoint list page, you can delete the selected endpoint by selecting the item you want to delete and clicking Delete.

Managing Ingress

Ingress is an API object that manages external access (HTTP, HTTPS) to services within Kubernetes Engine, used to expose workloads to the outside, and provides L7 load balancing functionality.

Creating an Ingress

To create an ingress, follow these steps.

  1. Click All services > Container > Kubernetes Engine menu. It moves to the Service Home page of Kubernetes Engine.
  2. On the Service Home page, click Ingress under the Services and Ingress menu. It moves to the Ingress List page.
  3. Ingress List page, select the cluster and namespace from the gear button at the top left, then click Create Object.
  4. Object Creation Popup에서 오브젝트 정보를 입력하고 Confirm 버튼을 클릭하세요.
Reference
For more information on the concept of Ingress and creating Ingress objects, see Kubernetes official documentation > Ingress.

Check Ingress Details

To check the details of the ingress, follow the next procedure.

  1. Click All services > Container > Kubernetes Engine menu. It moves to the Service Home page of Kubernetes Engine.
  2. Service Home page, click Ingress under the Service and Ingress menu. It moves to the Ingress List page.
  3. Ingress list page, select the cluster and namespace from the gear button at the top left, then click OK.
  4. Ingress List page, select the item you want to check the detailed information. It moves to the Ingress Detail page.
    • Selecting System Object View at the top of the list displays all items except Kubernetes object entries.
  5. Click each tab to check service information.
Classification
Detailed Description
Ingress DeletionDelete Ingress
Detailed InformationPossible to check detailed information of Ingress
YAMLYou can modify the ingress resource file in the YAML editor
  • Click the Edit button, modify the resource, and then click the Save button to apply the changes
EventCheck the event that occurred within the ingress
Account InformationAccount name, location, creation time, etc., Check the basic information about the Account
Metadata InformationCheck the metadata information of the ingress
Object InformationCheck the object information of the ingress
Table. Ingress detailed information items

Deleting Ingress

To delete an Ingress, follow these steps.

  1. Click all services > Container > Kubernetes Engine menu. It moves to the Service Home page of Kubernetes Engine.
  2. On the Service Home page, click Ingress under the Services and Ingress menu. It moves to the Ingress List page.
  3. Ingress list page, select the cluster and namespace from the gear button at the top left, then click OK.
  4. Ingress List page, select the item you want to delete. It moves to the Ingress Detail page.
  5. Ingress Details page, click Delete Ingress.
  6. When the Notification Confirmation Window appears, click the Confirm button.
Caution
On the Ingress list page, after selecting the item you want to delete, clicking Delete allows you to delete the selected Ingress.

Managing Ingress Classes

An IngressClass is an API resource that allows you to use multiple Ingress controllers in a single cluster. Each Ingress must specify a class that includes the configuration for the IngressClass resource that it refers to, including the controller it should be implemented by.

Creating an Ingress Class

To create an IngressClass, follow these steps.

  1. Click all services > Container > Kubernetes Engine menu. It moves to the Service Home page of Kubernetes Engine.
  2. On the Service Home page, click Ingress Class under the Services and Ingress menu. It moves to the Ingress Class List page.
  3. IngressClass list page, select the cluster and namespace from the gear button in the top left, then click Create object.
  4. Object Creation Popup에서 오브젝트 정보를 입력하고 Confirm 버튼을 클릭하세요.
Reference
For more information on the concept of IngressClass and creating objects, see the Kubernetes official documentation > Ingress.

Checking Ingress Class Details

To check the details of the IngressClass, follow the next procedure.

  1. Click All services > Container > Kubernetes Engine menu. It moves to the Service Home page of Kubernetes Engine.
  2. On the Service Home page, click Ingress Class under the Services and Ingress menu. It moves to the Ingress Class List page.
  3. IngressClass list page, select the cluster and namespace from the gear button at the top left, then click OK.
  4. IngressClass list page, select the item you want to check the detailed information. It moves to the IngressClass details page.
    • Selecting System Object View at the top of the list displays all items except Kubernetes object entries.
  5. Click each tab to check the service information.
Classification
Detailed Description
Ingress Class DeletionDelete an Ingress Class
Detailed InformationPossible to check detailed information of Ingress class
YAMLThe resource file of the Ingress class can be modified in the YAML editor
  • Click the Edit button, modify the resource, and then click the Save button to apply the changes
EventCheck the event that occurred within the Ingress class
Account InformationAccount name, location, creation time, etc., basic information about the Account can be checked
Metadata InformationCheck the metadata information of the Ingress class
Object InformationCheck the object information of the Ingress class
Table. IngressClass detailed information items

Deleting IngressClass

To delete an IngressClass, follow these steps.

  1. Click all services > Container > Kubernetes Engine menu. It moves to the Service Home page of Kubernetes Engine.
  2. On the Service Home page, click Ingress Class under the Services and Ingress menu. It moves to the Ingress Class List page.
  3. IngressClass list page, select the cluster and namespace from the gear button at the top left, then click OK.
  4. Select the item to delete on the IngressClass list page. It moves to the IngressClass detail page.
  5. Ingress Class Details page, click Delete Ingress Class.
  6. When the Notification Confirmation Window appears, click the Confirm button.
Caution
In the IngressClass list page, you can delete the selected IngressClass by selecting the item you want to delete and clicking Delete.

2.4 - Managing Storage

When using the Kubernetes Engine, you can create and manage storage. Storage is created and managed by item, including PVC, PV, and storage classes.

Note

The PVC, PV, and storage class services are set to the default cluster (namespace) selected when the service was created. Even if you select a different item from the list, the default cluster (namespace) setting is maintained.

  • To select a different cluster (namespace), click the gear button on the right side of the list. In the Cluster/Namespace Settings popup window, select the cluster and namespace you want to change to, and click the OK button. You can then view the services created in the selected cluster/namespace.
Guide

The items associated with each storage type are as follows:

TypeDetailed Description
Block StorageSupports storage classes that use the volume of the Block storage product in Virtual Server
Object StorageCan be linked with Samsung Cloud Platform products or external Object Storage
  • No separate settings are required in Kubernetes Engine, and you can set it up directly in your workload (application) according to the Object Storage guide
File StorageSupports storage classes that use NFS and CIFS protocol volumes with the File Storage product
  • For NFS protocol volumes, it is mandatory to select during Kubernetes Engine creation (supports HDD and SSD disk types)
  • For CIFS protocol volumes, it can be selected during or after Kubernetes Engine creation
Table. Storage association items by type

Managing PVC

A Persistent Volume Claim (PVC) is an object that defines the storage capacity to be allocated. PVC provides high usability through abstraction and can prevent data from being deleted together with the container lifecycle (maintaining Data Persistence).

Creating a PVC

To create a PVC, follow these steps:

  1. Click All Services > Container > Kubernetes Engine. You will be taken to the Kubernetes Engine Service Home page.
  2. On the Service Home page, click Storage under the menu, then click PVC. You will be taken to the PVC List page.
  3. On the PVC List page, select the cluster and namespace from the gear button at the top left, then click Create Object.
  4. In the Create Object popup window, enter the object information and click the OK button.
Note
For more information on the concept of PVC and creating objects, refer to the Kubernetes official documentation > Persistent Volumes.

Viewing PVC Details

To view PVC details, follow these steps:

  1. Click All Services > Container > Kubernetes Engine. You will be taken to the Kubernetes Engine Service Home page.
  2. On the Service Home page, click Storage under the menu, then click PVC. You will be taken to the PVC List page.
  3. On the PVC List page, select the cluster and namespace from the gear button at the top left, then click OK.
  4. On the PVC List page, select the item you want to view details for. You will be taken to the PVC Details page.
    • Select Show System Objects at the top of the list to display Kubernetes objects.
  5. Click each tab to view the service information.
    Category
    Detailed Description
    StatusDisplays the current status of the PVC.
    • Bound: Normal connection
    Delete PVCDeletes the PVC
    DetailsDisplays detailed information about the PVC
    YAMLAllows you to modify the PVC resource file in the YAML editor
    • Click the Edit button, modify the resource, and click the Save button to apply the changes
    EventsDisplays events that occurred within the PVC
    Account InformationDisplays basic information about the account, such as the account name, location, and creation time
    Metadata InformationDisplays metadata information about the PVC
    Object InformationDisplays object information about the PVC
    Table. PVC detail items

Delete PVC

To delete a PVC, follow these steps:

  1. Click All Services > Container > Kubernetes Engine menu. Move to the Service Home page of Kubernetes Engine.
  2. On the Service Home page, click PVC under the Storage menu. Move to the PVC List page.
  3. On the PVC List page, select a cluster and namespace from the gear button at the top left, and click Confirm.
  4. On the PVC List page, select the item you want to delete. Move to the PVC Details page.
  5. On the PVC Details page, click Delete PVC.
  6. When the Notification Confirmation window appears, click the Confirm button.
Caution

You can delete the selected PVC by selecting the item you want to delete on the PVC List page and clicking Delete.

  • Before deleting a PVC, check if the PV and volume to be deleted are backed up.

Manage PV

Persistent Volume (PV) refers to a physical disk created by the system administrator in Kubernetes Engine.

Create PV

To create a PV, follow these steps:

  1. Click All Services > Container > Kubernetes Engine menu. Move to the Service Home page of Kubernetes Engine.
  2. On the Service Home page, click PV under the Storage menu. Move to the PV List page.
  3. On the PV List page, select a cluster and namespace from the gear button at the top left, and click Create Object.
  4. In the Create Object popup window, enter object information and click the Confirm button.
Reference
For more information on the concept of PV and object creation, see Kubernetes Official Documentation > Persistent Volumes.

Check PV Details

To check the PV details, follow these steps:

  1. Click All Services > Container > Kubernetes Engine menu. Move to the Service Home page of Kubernetes Engine.
  2. On the Service Home page, click PV under the Storage menu. Move to the PV List page.
  3. On the PV List page, select a cluster and namespace from the gear button at the top left, and click Confirm.
  4. On the PV List page, select the item you want to check the details of. Move to the PV Details page.
    • Select Show System Objects at the top of the list to display items other than Kubernetes objects.
  5. Click each tab to check the service information.
    CategoryDescription
    StatusDisplays the current status of the PV.
    • Bound: Normal connection
    Delete PVDelete PV
    DetailsCheck the detailed information of the PV
    YAMLModify the PV resource file in the YAML editor
    • Click the Edit button, modify the resource, and click the Save button to apply the changes
    EventsCheck the events that occurred within the PV
    Account InformationCheck the basic information of the account, such as account name, location, and creation time
    Metadata InformationCheck the metadata information of the PV
    Object InformationCheck the object information of the PV
    Table. PV Details Items

Delete PV

To delete a PV, follow these steps:

  1. Click All Services > Container > Kubernetes Engine menu. Move to the Service Home page of Kubernetes Engine.
  2. On the Service Home page, click PV under the Storage menu. Move to the PV List page.
  3. On the PV List page, select a cluster and namespace from the gear button at the top left, and click Confirm.
  4. On the PV List page, select the item you want to delete. Move to the PV Details page.
  5. On the PV Details page, click Delete PV.
  6. When the Notification Confirmation window appears, click the Confirm button.
Caution
You can delete the selected PV by selecting the item you want to delete on the PV List page and clicking Delete.

Managing Storage Classes

A Storage Class (Storage Class) is a Kubernetes resource that defines the type or performance level of storage.

Note

Kubernetes Engine provides nfs-subdir-external-sc and bs-sc storage classes by default, with the following characteristics:

  • The nfs-subdir-external-sc storage class shares file storage connected to the cluster.
    • Access mode: RWX - ReadWriteMany
    • Reclaim policy: Delete (deletes PV and stored data when PVC is deleted), Retain (keeps PV and stored data when PVC is deleted)
    • Capacity expansion: Individual PVC expansion not supported / File storage expansion allowed
  • The bs-sc storage class supports SSD-type volumes in conjunction with block storage products.
    • Access mode: RWO - ReadWriteOnce
    • Reclaim policy: Delete (deletes PV and stored data when PVC is deleted), Retain (keeps PV and stored data when PVC is deleted)
    • Capacity expansion supported: Individual PVC expansion supported (8 Gi unit volume automatic expansion)

Predefined Storage Classes

Storage ClassReclaim Policy*Volume Expansion Supported**Mount OptionsNote
nfs-subdir-external-sc (default)DeleteNot supportednfsvers=3, noresvportLinked to default volume (NFS) settings
nfs-subdir-external-sc-retainRetainNot supportednfsvers=3, noresvportLinked to default volume (NFS) settings
bs-scDeleteSupported-Linked to VirtualServer > BlockStorage products
bs-sc-retainRetainSupported-Linked to VirtualServer > BlockStorage products
  • (*) To use a non-default storage class, specify the storage class name in the spec.storageClassName of the PVC.
  • (**) Users can change the default storage class (storageclass.kubernetes.io/is-default-class: “true” annotation adjustment)
    Table. List of predefined storage classes
Caution

The characteristics of the reclaim policy are as follows:

  • Delete: When a PVC is deleted, the corresponding PV and physical data are also deleted.
  • Retain: When a PVC is deleted, the corresponding PV and physical data are not deleted and are retained. Since physical data not used by the workload can remain in storage, careful capacity management is required.
Caution

When using volume expansion, consider the following:

  1. nfs-subdir-external-sc storage class
    • The capacity of the PVC cannot be adjusted. (Volume expansion not supported)
    • All PVs share the total capacity of the file storage, so individual PVC volume expansion is not necessary.
  2. bs-sc storage class
    • The capacity of the PVC can be expanded. (No reduction function supported)
    • The capacity of the PV is not guaranteed to be the same as the capacity requested by the PVC. (8 Gi unit expansion supported)

Creating a Storage Class

To create a storage class, follow these steps:

  1. Click All Services > Container > Kubernetes Engine. The Kubernetes Engine Service Home page appears.
  2. On the Service Home page, click Storage under the Storage menu. The Storage Class List page appears.
  3. On the Storage Class List page, select the cluster and namespace from the gear button at the top left, and then click Create Object.
  4. In the Create Object popup, enter the object information and click OK.
    Note
    For more information on the concept of storage classes and object creation, see the Kubernetes official documentation > Storage Classes.

Checking Storage Class Details

To check the details of a storage class, follow these steps:

  1. Click All Services > Container > Kubernetes Engine. The Kubernetes Engine Service Home page appears.
  2. On the Service Home page, click Storage under the Storage menu. The Storage Class List page appears.
  3. On the Storage Class List page, select the cluster and namespace from the gear button at the top left, and then click OK.
  4. On the Storage Class List page, select the item for which you want to check the details. The Storage Class Details page appears.
    • Select Show System Objects at the top of the list to display items other than Kubernetes objects.
  5. Click each tab to check the service information.
    Category
    Description
    Delete Storage ClassDelete the storage class
    DetailsCheck the detailed information of the storage class
    YAMLModify the storage class resource file in the YAML editor
    • Click the Edit button, modify the resource, and click the Save button to apply the changes
    EventsCheck the events that occurred within the storage class
    Account InformationCheck the basic information of the account, such as the account name, location, and creation time
    Metadata InformationCheck the metadata information of the storage class
    Object InformationCheck the object information of the storage class
    Table. Storage class details items

Deleting a Storage Class

To delete a storage class, follow these steps:

  1. Click All Services > Container > Kubernetes Engine. The Kubernetes Engine Service Home page appears.
  2. On the Service Home page, click Storage under the Storage menu. The Storage Class List page appears.
  3. On the Storage Class List page, select the cluster and namespace from the gear button at the top left, and then click OK.
  4. On the Storage Class List page, select the item you want to delete. The Storage Class Details page appears.
  5. On the Storage Class Details page, click Delete Storage Class.
  6. In the Confirmation window, click OK.
    Caution
    On the storage class list page, you can delete the selected storage class by clicking Delete after selecting the item you want to delete.

2.5 - Configuration Management

When there is a need to manage changing values inside the container according to various environments such as development and operation, managing separate images due to environment variables is inconvenient and costly. In Kubernetes, you can manage environment variables or configuration values as variables so that they can be changed from outside, and at this time, ConfigMap and Secret can be used to insert them when a Pod is created.

Reference

ConfigMap and Secret services are set to the cluster (namespace) selected when the service is created by default. Even if you select other items in the list, the default cluster (namespace) setting is maintained.

  • To select a different cluster (namespace), click the gear button to the right of the list. In the Cluster/Namespace Settings popup window, select the cluster and namespace you want to change and click the OK button. You can see the config map and secret service created in the selected cluster/namespace.

Managing Config Maps

You can write and manage Config information used in the namespace as a config map.

Creating a Config Map

To create a configmap, follow these steps.

  1. Click all services > Container > Kubernetes Engine menu. It moves to the Service Home page of Kubernetes Engine.
  2. On the Service Home page, click ConfigMap under the Configuration menu. It moves to the ConfigMap list page.
  3. ConfigMap List page, select the cluster and namespace from the gear button at the top left, then click Create Object.
  4. Object Creation Popup에서 오브젝트 정보를 입력하고 Confirm 버튼을 클릭하세요.
Reference
For more information about the concept of ConfigMap and object creation, see Kubernetes official documentation > ConfigMap.

Checking ConfigMap Details

To check the config map details, follow the next procedure.

  1. Click all services > Container > Kubernetes Engine menu. It moves to the Service Home page of Kubernetes Engine.
  2. On the Service Home page, click ConfigMap under the Configuration menu. It moves to the ConfigMap list page.
  3. ConfigMap List page, select the cluster and namespace from the gear button at the top left, then click OK.
  4. On the Config Map List page, select the item you want to check the detailed information. It moves to the Config Map Detail page.
    • Selecting System Object View at the top of the list displays all items except for Kubernetes object entries.
  5. Click each tab to check the service information.
Classification
Detailed Description
Config Map DeletionDelete Config Map
Detailed InformationCheck detailed information of Config Map
YAMLThe resource file of ConfigMap can be modified in the YAML editor
  • Click the Edit button, modify the resource, and then click the Save button to apply the changes
EventCheck the event that occurred within the config map
Account InformationAccount name, location, creation time, etc., check the basic information about the Account
Metadata InformationCheck the metadata information of the config map
Object InformationCheck the object information of the config map
  • Data separates lines with - - - and displays value in textarea format
  • Binary Data value outputs the length value
Table. Configmap detail information items

Deleting ConfigMap

To delete a configmap, follow this procedure.

  1. Click All services > Container > Kubernetes Engine menu. It moves to the Service Home page of Kubernetes Engine.
  2. On the Service Home page, click ConfigMap under the Configuration menu. It moves to the ConfigMap list page.
  3. ConfigMap list page, select the cluster and namespace from the gear button at the top left, then click OK.
  4. Configmap List page, select the item you want to delete. It moves to the Configmap Detail page.
  5. Configmap Details page, click Delete Configmap.
  6. When the Notification Confirmation Window appears, click the Confirm button.
Caution
In the ConfigMap list page, you can delete the selected ConfigMap by selecting the item you want to delete and then clicking Delete.

Managing Secrets

Using secrets, you can safely store and manage sensitive information such as passwords, OAuth tokens, and SSH keys.

Creating a Secret

To create a secret, follow these steps.

  1. Click All services > Container > Kubernetes Engine menu. It moves to the Service Home page of Kubernetes Engine.
  2. On the Service Home page, click 시크릿 under the 구성 menu. It moves to the 시크릿 목록 page.
  3. Secret List page, select cluster and namespace from the gear button at the top left, then click Create Object.
  4. Object Creation Popup에서 오브젝트 정보를 입력하고 Confirm 버튼을 클릭하세요.
Reference
For more information on the concept of Secrets and creating Secret objects, see Kubernetes official documentation > Secret.

Check Secret Details

To check the secret details, follow the following procedure.

  1. Click All services > Container > Kubernetes Engine menu. It moves to the Service Home page of Kubernetes Engine.
  2. On the Service Home page, click 시크릿 under the 구성 menu. It moves to the 시크릿 목록 page.
  3. Secret List page, select cluster and namespace from the gear button at the top left, then click Confirm.
  4. Secret List page, select the item you want to check the detailed information. It moves to the Secret Detail page.
    • Selecting Show System Objects at the top of the list displays all items except Kubernetes object entries.
  5. Click each tab to check the service information.
Classification
Detailed Description
Secret DeletionDelete secret
Detailed InformationCheck the detailed information of Secret
YAMLThe resource file of the secret can be modified in the YAML editor
  • Click the Edit button, modify the resource, and then click the Save button to apply the changes
EventCheck the event that occurred within Secret
Account InformationAccount name, location, creation time, etc., basic information about the Account can be checked
Metadata InformationCheck the metadata information of the secret
Object InformationCheck the object information of the secret
Table. Secret detailed information items

Deleting Secrets

To delete a secret, follow these steps.

  1. Click all services > Container > Kubernetes Engine menu. It moves to the Service Home page of Kubernetes Engine.
  2. On the Service Home page, click 시크릿 under the 구성 menu. It moves to the 시크릿 목록 page.
  3. Secret List page, select the cluster and namespace from the gear button at the top left, then click OK.
  4. Secret List page, select the item you want to delete. It moves to the Secret Detail page.
  5. Secret Detail page, click Delete Secret.
  6. When the Notification Confirmation Window appears, click the Confirm button.
Caution
In the Secret list page, you can delete the selected secret by selecting the item you want to delete and clicking Delete.

2.6 - Managing Permissions

When multiple users access the Kubernetes cluster, you can grant permissions by specific API or namespace and specify the access range. You can apply the Role-Based Access Control (RBAC) feature of Kubernetes to set permissions by cluster or namespace. You can create and manage ClusterRole, ClusterRoleBinding, Role, and RoleBinding.

Reference

ClusterRole, ClusterRoleBinding, Role, RoleBinding services are set to the cluster (namespace) selected when the service is created by default. Even if you select other items in the list, the default cluster (namespace) setting is maintained.

  • To select a different cluster (namespace), click the gear button on the right side of the list. In the Cluster/Namespace Settings popup window, select the cluster and namespace you want to change and click the OK button. You can view the services created in the selected cluster/namespace.
Reference

Managing Cluster Roles

You can set and manage access permissions on a cluster unit basis. You can also set permissions for APIs or resources that are not limited to a namespace.

Creating a Cluster Role

To create a cluster role, follow these steps.

  1. Click All services > Container > Kubernetes Engine menu. It moves to the Service Home page of Kubernetes Engine.
  2. On the Service Home page, click ClusterRole under the Authorities menu. It moves to the ClusterRole list page.
  3. Cluster Role List page, select the cluster and namespace from the gear button at the top left, then click Create Object.
  4. Enter object information in the object creation popup window and click the OK button.
Reference
For more information on ClusterRole, see Kubernetes official documentation > Using RBAC Authorization.

Check Cluster Role Details

To view detailed information about the cluster role, follow these steps.

  1. Click All services > Container > Kubernetes Engine menu. It moves to the Service Home page of Kubernetes Engine.
  2. On the Service Home page, click 클러스터롤 under the 권한 menu. It moves to the 클러스터롤 목록 page.
  3. Cluster Role List page, select the cluster and namespace from the gear button at the top left, then click Confirm.
  4. Cluster Roll List page, select the item you want to check the detailed information. Move to the Cluster Roll Detail page.
    • Selecting System Object View at the top of the list displays all items except Kubernetes object entries.
  5. Click each tab to check the service information.
Classification
Detailed Description
Cluster Role DeletionDelete cluster role
Detailed InformationPossible to check detailed information of cluster role
YAMLThe resource file of the cluster role can be modified in the YAML editor
  • Click the Edit button, modify the resource, and then click the Save button to apply the changes
EventCheck the event that occurred within the cluster role
Account InformationAccount name, location, creation time, etc., basic information about the Account can be checked
Metadata InformationCheck the metadata information of the cluster role
Policy Rule InformationCheck policy rule information for ClusterRole
  • Resources: List of resources to which the rule is applied
  • Non-Resource URLs: Set of non-resource URLs that users must access
    • * is allowed, but only as the final part of the path
    • Non-resource URLs are not namespaced, so this field is only applicable to ClusterRoles referenced by ClusterRoleBindings
    • Rules can be applied to either API resources (e.g. “pods” or “secrets”) or non-resource URL paths (e.g. “/api”), but not both
  • Resource Names: Resource names are an optional whitelist of names that the rule applies to. An empty set means everything is allowed
Table. Cluster Role Detailed Information Items

Deleting a Cluster Role

To delete a cluster role, follow this procedure.

  1. Click all services > Container > Kubernetes Engine menu. It moves to the Service Home page of Kubernetes Engine.
  2. On the Service Home page, click ClusterRole under the Authority menu. It moves to the ClusterRole list page.
  3. Cluster Role List page, select the cluster and namespace from the gear button at the top left, then click OK.
  4. Select the item to delete on the Cluster Role List page. It moves to the Cluster Role Detail page.
  5. Cluster Role Detail page, click Delete Cluster Role.
  6. When the Notification Confirmation Window appears, click the Confirm button.
Caution
On the cluster role list page, you can delete the selected cluster role by selecting the item you want to delete and clicking Delete.

Managing Cluster Role Bindings

You can create and manage cluster role bindings by connecting cluster roles and specific targets.

Creating Cluster Role Binding

To create a cluster role binding, follow these steps.

  1. Click all services > Container > Kubernetes Engine menu. It moves to the Service Home page of Kubernetes Engine.
  2. On the Service Home page, click ClusterRoleBinding under the Authority menu. It moves to the ClusterRoleBinding list page.
  3. Cluster Role Binding List page, select the cluster and namespace from the gear button at the top left, then click Create Object.
  4. Object Creation Popup에서 오브젝트 정보를 입력하고 Confirm 버튼을 클릭하세요.
Reference
For more information on cluster role binding, see Kubernetes official documentation > Using RBAC Authorization.

Check Cluster Role Binding Details

To check the cluster role binding details, follow the next procedure.

  1. Click all services > Container > Kubernetes Engine menu. It moves to the Service Home page of Kubernetes Engine.
  2. On the Service Home page, click ClusterRoleBinding under the Authority menu. It moves to the ClusterRoleBinding list page.
  3. Cluster Role Binding List page, select the cluster and namespace from the gear button at the top left, then click Confirm.
  4. On the Cluster Role Binding List page, select the item you want to check the detailed information. It moves to the Cluster Role Binding Detail page.
    • Selecting Show System Objects at the top of the list displays all items except Kubernetes object entries.
  5. Click each tab to check the service information.
Classification
Detailed Description
Cluster Role Binding DeletionDelete cluster role binding
Detailed InformationCheck the detailed information of the cluster role binding
YAMLThe resource file of ClusterRoleBinding can be modified in the YAML editor
  • Click the Edit button, modify the resource, and then click the Save button to apply the changes
EventCheck the event that occurred within the cluster role binding
Account InformationAccount name, location, creation time, etc., basic information about the Account can be checked
Metadata InformationCheck the metadata information of the cluster role binding
Role/Target InformationCheck the role and target information of the cluster role
Table. Cluster Role Binding Details Items

Deleting Cluster Role Binding

To delete a cluster role binding, follow these steps.

  1. Click all services > Container > Kubernetes Engine menu. It moves to the Service Home page of Kubernetes Engine.
  2. On the Service Home page, click ClusterRoleBinding under the Authority menu. It moves to the ClusterRoleBinding list page.
  3. Cluster Role Binding List page, select the cluster and namespace from the gear button at the top left, then click Confirm.
  4. Cluster Role Binding List page, select the item you want to delete. It moves to the Cluster Role Binding Details page.
  5. Cluster Role Binding Detail page, click Delete Cluster Role Binding.
  6. When the Notification Confirmation Window appears, click the Confirm button.
Caution
On the Cluster Role Binding list page, you can delete the selected Cluster Role Binding by selecting the item you want to delete and clicking Delete.

Managing Roles

A role is a set of rules that explicitly define permissions for a specific API or resource, and it can create and manage permissions that can only be accessed within the namespace to which the role belongs.

Create Role

To create a role, follow these steps.

  1. Click All services > Container > Kubernetes Engine menu. It moves to the Service Home page of Kubernetes Engine.
  2. On the Service Home page, click under the 권한 menu. It moves to the 롤 목록 page.
  3. Roll list page, select cluster and namespace from the gear button at the top left, then click Create Object.
  4. Object Creation Popup에서 오브젝트 정보를 입력하고 Confirm 버튼을 클릭하세요.
Reference

Check Roll Details

To check the roll details, follow the next procedure.

  1. Click All services > Container > Kubernetes Engine menu. It moves to the Service Home page of Kubernetes Engine.
  2. On the Service Home page, click under the 권한 menu. It moves to the 롤 목록 page.
  3. On the Roll List page, select the cluster and namespace from the Gear button at the top left, then click OK.
  4. Role List page, select the item you want to check the detailed information. Move to the Role Detail page.
    • Selecting System Object View at the top of the list displays all items except Kubernetes object entries.
  5. Click each tab to check the service information.
Classification
Detailed Description
Role Deleteto delete a role
Detailed InformationCheck the detailed information of the roll
YAMLYou can modify the role’s resource file in the YAML editor
  • Click the Edit button, modify the resource, and then click the Save button to apply the changes
EventCheck the event that occurred in the roll
Account InformationAccount name, location, creation time, etc., basic information about the Account can be checked
Metadata InformationCheck the metadata information of the roll
Policy Rule InformationCheck the policy rule information of the role
  • Resources: List of resources to which the rule is applied
  • Non-Resource URLs: Non-resource URLs are a set of partial URLs that the user must access
    • * is allowed, but only as the final step in the path
    • Non-resource URLs are not namespaced, so this field is only applicable to ClusterRoles referenced by ClusterRoleBindings
    • Rules can be applied to either API resources (e.g. “pods” or “secrets”) or non-resource URL paths (e.g. “/api”), but not both
  • Resource Names: Resource names are an optional whitelist of names that the rule applies to, an empty set means everything is allowed
Table. Detailed information items of roles

Delete Role

To delete a role, follow these steps.

  1. Click All services > Container > Kubernetes Engine menu. It moves to the Service Home page of Kubernetes Engine.
  2. On the Service Home page, click under the 권한 menu. It moves to the 롤 목록 page.
  3. On the Roll List page, select the cluster and namespace from the gear button at the top left, then click OK.
  4. Select the item you want to delete from the Role List page. It moves to the Role Detail page.
  5. Role Details page, click Delete Role.
  6. When the Notification Confirmation Window appears, click the Confirm button.
Caution
On the roll list page, after selecting the item you want to delete, clicking Delete allows you to delete the selected roll.

Managing Roll Binding

You can create and manage role bindings by linking roles to specific targets.

Creating Roll Binding

To create a role binding, follow these steps.

  1. Click All services > Container > Kubernetes Engine menu. It moves to the Service Home page of Kubernetes Engine.
  2. On the Service Home page, click 롤바인딩 under the 권한 menu. It moves to the 롤바인딩 목록 page.
  3. Role Binding List page, select the cluster and namespace from the gear button at the top left, then click Create Object.
  4. Object Creation Popup에서 오브젝트 정보를 입력하고 확인 버튼을 클릭하세요.
Reference
For more information on RoleBinding, please refer to Kubernetes official documentation > Using RBAC Authorization.

Check Roll Binding Details

To check the details of the roll binding, follow the next procedure.

  1. Click all services > Container > Kubernetes Engine menu. It moves to the Service Home page of Kubernetes Engine.
  2. On the Service Home page, click 롤바인딩 under the 권한 menu. It moves to the 롤바인딩 목록 page.
  3. On the 롤바인딩 목록 page, select the cluster and namespace from the 톱니바퀴 button at the top left, then click 확인.
  4. On the Roll Binding List page, select the item you want to check the detailed information. It moves to the Roll Binding Details page.
    • Selecting Show System Objects at the top of the list displays all items except Kubernetes object entries.
  5. Click each tab to check the service information.
Classification
Detailed Description
Roll Binding DeleteDelete roll binding
Detailed InformationCheck the detailed information of roll binding
YAMLRollbinding’s resource file can be modified in YAML editor
  • Click the Edit button, modify the resource, and then click the Save button to apply the changes
EventCheck the event that occurred within the roll binding
Account InformationAccount name, location, creation time, etc., basic information about the Account can be checked
Metadata InformationCheck the metadata information of Roll Binding
Roll/Target InformationCheck the roll’s role and target information
Table. Rollbinding detailed information items

Deleting Roll Binding

To delete a role binding, follow these steps.

  1. Click all services > Container > Kubernetes Engine menu. It moves to the Service Home page of Kubernetes Engine.
  2. On the Service Home page, click 롤바인딩 under the 권한 menu. It moves to the 롤바인딩 목록 page.
  3. Rollbinding List page, select the cluster and namespace from the gear button at the top left, then click OK.
  4. Select the item to delete from the Roll Binding List page. It moves to the Roll Binding Details page.
  5. On the Roll Binding Details page, click Delete Roll Binding.
  6. When the Notification Confirmation Window appears, click the OK button.
Caution
On the Role Binding list page, after selecting the item you want to delete, clicking Delete allows you to delete the selected role binding.

2.7 -

3 - Using Kubernetes Engine

Configure external network communication to expose HTTP and HTTPS services from the cluster to the outside. To configure external network communication, you can create a service of type LoadBalancer.

Using Kubernetes Engine Guide

The Using Kubernetes Engine guide describes the following features. For more information, refer to the corresponding guide.

GuideDescription
Creating a LoadBalancer ServiceInstructions on how to create a LoadBalancer-type service through a service manifest file
Table. Description of Using Kubernetes Engine Guide

3.1 - Authentication and Authorization

Kubernetes Engine has Kubernetes’ authentication and RBAC authorization features applied. This explains the authentication and authorization features of Kubernetes and how to link them with Kubernetes Engine and IAM.

Kubernetes Authentication and Authorization

This explains the authentication and RBAC authorization features of Kubernetes.

Authentication

The Kubernetes API server acquires the necessary information for user or account authentication from certificates or authentication tokens and proceeds with the authentication process.

Note
For a detailed explanation of Kubernetes authentication, refer to the following document: https://kubernetes.io/docs/reference/access-authn-authz/authentication/
Note
For a detailed explanation of using kubectl and kubeconfig, refer to Accessing the Cluster.

Authorization

The Kubernetes API server checks if the user has permission for the requested action using the user information obtained through the authentication process and the RBAC-related objects. There are four types of RBAC-related objects as follows:

ObjectScopeDescription
ClusterRoleCluster-wideDefinition of permissions across all namespaces in the cluster
ClusterRoleBindingCluster-wideBinding definition between ClusterRole and user
RoleNamespaceDefinition of permissions for a specific namespace
RoleBindingNamespaceBinding definition between ClusterRole or Role and user
Table. RBAC-related objects
Note
For a detailed explanation of Kubernetes RBAC authorization, refer to the following document: https://kubernetes.io/docs/reference/access-authn-authz/rbac/

Role

Kubernetes has several predefined ClusterRoles. Some of these ClusterRoles do not have the prefix system:, which means they are intended for user use. These include the cluster-admin role that can be applied to the entire cluster using ClusterRoleBinding, and the admin, edit, and view roles that can be applied to a specific namespace using RoleBinding.

Default ClusterRoleDefault ClusterRoleBindingDescription
cluster-adminsystem:masters groupGrants superuser access to perform all actions on all resources.
  • When used in ClusterRoleBinding, it grants full control over all resources in the cluster and all namespaces.
  • When used in RoleBinding, it grants full control over the namespace and all resources in the namespace bound to the RoleBinding.
adminNoneGrants administrator access to the namespace when used with RoleBinding. When used in RoleBinding, it grants read/write access to most resources in the namespace, including the ability to create roles and role bindings. However, this role does not grant write access to resource quotas or the namespace itself.
editNoneGrants read/write access to most objects in the namespace. This role does not grant the ability to view or modify roles and role bindings. However, this role allows access to secrets, which can be used to run pods in the namespace as any account, effectively granting API access at the account level.
viewNoneGrants read-only access to most objects in the namespace. Roles and role bindings cannot be viewed. This role does not grant access to secrets, as reading secret contents would allow access to account credentials and potentially grant API access at the account level (a form of privilege escalation).
Table. Default ClusterRole and ClusterRoleBinding descriptions
Note
For a detailed explanation of user-facing roles, refer to the following document: https://kubernetes.io/docs/reference/access-authn-authz/rbac/#user-facing-roles

In addition to the predefined ClusterRoles, you can define separate roles (or ClusterRoles) as needed. For example:

Color mode
# Role that grants permission to view pods in the "default" namespace
apiVersion: rbac.authorization.k8s.io/v1
kind: Role
metadata:
  namespace: default
  name: pod-reader
rules:
- apiGroups: [""]
  resources: ["pods"]
  verbs: ["get", "list", "watch"]
# Role that grants permission to view pods in the "default" namespace
apiVersion: rbac.authorization.k8s.io/v1
kind: Role
metadata:
  namespace: default
  name: pod-reader
rules:
- apiGroups: [""]
  resources: ["pods"]
  verbs: ["get", "list", "watch"]
Code block. Role that grants permission to view pods in a namespace
Color mode
# ClusterRole that grants permission to view nodes
apiVersion: rbac.authorization.k8s.io/v1
kind: ClusterRole
metadata:
  name: node-viewer
rules:
- apiGroups: [""]
  resources: ["nodes"]
  verbs: ["get", "list", "watch"]
# ClusterRole that grants permission to view nodes
apiVersion: rbac.authorization.k8s.io/v1
kind: ClusterRole
metadata:
  name: node-viewer
rules:
- apiGroups: [""]
  resources: ["nodes"]
  verbs: ["get", "list", "watch"]
Code block. ClusterRole that grants permission to view nodes
Note
For more information about roles and cluster roles, see the following document: https://kubernetes.io/docs/reference/access-authn-authz/rbac/#role-and-clusterrole

Role Binding

To manage access to the Kubernetes Engine using Samsung Cloud Platform IAM, you need to understand the relationship between Kubernetes’ role binding and IAM. The target (subjects) of role binding (or cluster role binding) can include individual users (User) or groups (Group).

  • User matches the Samsung Cloud Platform username, and Group matches the IAM user group name.

For role binding/cluster role binding, subjects.kind can be one of the following:

  • User: Binds to a Samsung Cloud Platform individual user.
  • Group: Binds to a Samsung Cloud Platform IAM user group.
Note
In addition to the above, a service account can also be specified, but a service account is generally not for users and cannot be bound to a Samsung Cloud Platform user.

The subjects.name of role binding/cluster role binding can be specified as follows:

  • User case: Samsung Cloud Platform individual username (e.g. jane.doe)
  • Group case: Samsung Cloud Platform IAM user group name (e.g. ReadPodsGroup)
Note
subjects.name is case-sensitive.

In this way, an IAM user group is bound to a role binding (or cluster role binding) written in the Kubernetes Engine cluster. Additionally, the permission to perform API operations included in the role (or cluster role) bound to the group is granted.

Example) Role Binding read-pods #1

An example of writing a User (Samsung Cloud Platform individual user) to a role binding is as follows:

Color mode
# This role binding allows the user "jane.doe@example.com" to view pods in the "default" namespace.
# A "pod-reader" role must exist in the namespace.
apiVersion: rbac.authorization.k8s.io/v1
metadata:
  name: read-pods
  namespace: default
roleRef:
  # "roleRef" specifies the binding to a role or cluster role.
  kind: Role       # Must be Role or ClusterRole.
  name: pod-reader # Must match the name of the role or cluster role to bind.
  apiGroup: rbac.authorization.k8s.io
subjects:
# One or more "targets" can be specified.
- kind: User
  name: jane.doe
  apiGroup: rbac.authorization.k8s.io
# This role binding allows the user "jane.doe@example.com" to view pods in the "default" namespace.
# A "pod-reader" role must exist in the namespace.
apiVersion: rbac.authorization.k8s.io/v1
metadata:
  name: read-pods
  namespace: default
roleRef:
  # "roleRef" specifies the binding to a role or cluster role.
  kind: Role       # Must be Role or ClusterRole.
  name: pod-reader # Must match the name of the role or cluster role to bind.
  apiGroup: rbac.authorization.k8s.io
subjects:
# One or more "targets" can be specified.
- kind: User
  name: jane.doe
  apiGroup: rbac.authorization.k8s.io
Code block. Example of writing a User (Samsung Cloud Platform individual user) to a role binding

If a role binding like the above is created in a cluster, a user with the username jane.doe is granted the permission to perform the API actions defined in the pod-reader role.

Example) Role Binding read-pods #2

An example of writing a group (IAM user group) to a role binding is as follows:

Color mode
# This role binding allows users in the "ReadPodsGroup" group to view pods in the "default" namespace.
# A "pod-reader" role must exist in the namespace.
apiVersion: rbac.authorization.k8s.io/v1
kind: RoleBinding
metadata:
  name: read-pods
  namespace: default
roleRef:
  kind: Role
  name: pod-reader
  apiGroup: rbac.authorization.k8s.io
subjects:
# One or more "targets" can be specified.
- kind: Group
  name: ReadPodsGroup
  apiGroup: rbac.authorization.k8s.io
# This role binding allows users in the "ReadPodsGroup" group to view pods in the "default" namespace.
# A "pod-reader" role must exist in the namespace.
apiVersion: rbac.authorization.k8s.io/v1
kind: RoleBinding
metadata:
  name: read-pods
  namespace: default
roleRef:
  kind: Role
  name: pod-reader
  apiGroup: rbac.authorization.k8s.io
subjects:
# One or more "targets" can be specified.
- kind: Group
  name: ReadPodsGroup
  apiGroup: rbac.authorization.k8s.io
Code block. Example of Role binding that allows the ReadPodsGroup group to view pods

If a role binding like the above is created in the cluster, users in the IAM user group ReadPodsGroup are granted the permission to perform API operations written in the pod-reader role.

Example) Cluster Role Binding read-nodes

Color mode
# This cluster role binding allows users in the "ReadNodesGroup" group to view nodes.
# A cluster role named "node-reader" must exist.
apiVersion: rbac.authorization.k8s.io/v1
kind: ClusterRoleBinding
metadata:
  name: read-nodes
roleRef:
  kind: ClusterRole
  name: node-reader
  apiGroup: rbac.authorization.k8s.io
subjects:
- kind: Group
  name: ReadNodesGroup
  apiGroup: rbac.authorization.k8s.io
# This cluster role binding allows users in the "ReadNodesGroup" group to view nodes.
# A cluster role named "node-reader" must exist.
apiVersion: rbac.authorization.k8s.io/v1
kind: ClusterRoleBinding
metadata:
  name: read-nodes
roleRef:
  kind: ClusterRole
  name: node-reader
  apiGroup: rbac.authorization.k8s.io
subjects:
- kind: Group
  name: ReadNodesGroup
  apiGroup: rbac.authorization.k8s.io
Code block. Example of a cluster role binding that allows the ReadNodesGroup group to view nodes

When a cluster role binding like the one above is created in the cluster, users in the IAM user group ReadNodesGroup are granted the permissions to perform the API actions written in the cluster role node-reader.

Note
For more detailed explanations on role binding creation, refer to the following document: https://kubernetes.io/docs/reference/access-authn-authz/rbac/#role-binding-examples

Predefined Roles and Role Bindings for Samsung Cloud Platform

The Kubernetes Engine of Samsung Cloud Platform has predefined cluster role bindings scp-cluster-admin, scp-view, scp-namespace-view, and cluster roles scp-namespace-view. The following table shows the binding relationship between predefined roles and role bindings, and Samsung Cloud Platform users. Here, cluster roles cluster-admin and view are predefined within the Kubernetes cluster. For more detailed explanations, refer to the Roles section.

Cluster Role BindingCluster RoleSubjects (User)
scp-cluster-admincluster-admin
  • Group AdministratorGroup
  • Group OperatorGroup
  • User john.smith
scp-viewviewGroup ViewerGroup
scp-namespace-viewscp-namespace-viewAll authenticated users in the cluster
Table. Predefined Roles and Role Bindings for Samsung Cloud Platform, IAM User Groups, and User Binding Relationships
  • According to the cluster role binding scp-cluster-admin, users in the IAM user groups AdministratorGroup or OperatorGroup, as well as the Kubernetes Engine product applicant, are granted cluster administrator permissions.
  • According to the cluster role binding scp-view, users in the ViewerGroup are granted cluster viewer permissions. More precisely, since it is linked to the predefined cluster role view in Kubernetes, access permissions for cluster-scoped resources (e.g., namespaces, nodes, ingress classes, etc.) and secrets within namespaces are not included. For more detailed explanations, refer to the Roles section.
  • According to the cluster role binding scp-namespace-view, all authenticated users in the cluster are granted namespace viewer permissions.
Note
  • Predefined roles and role bindings for Samsung Cloud Platform are created only once when the cluster product is applied.
  • Users can modify or delete predefined cluster role bindings and cluster roles for Samsung Cloud Platform as needed.

The details of predefined roles and role bindings for Samsung Cloud Platform are as follows:

Cluster Role Binding scp-cluster-admin

The cluster role binding scp-cluster-admin is bound to the cluster role cluster-admin and bound to the IAM user groups AdministratorGroup, OperatorGroup, and the SCP user (Kubernetes Engine cluster creator) according to the subjects.

Color mode
apiVersion: rbac.authorization.k8s.io/v1
kind: ClusterRoleBinding
  name: scp-cluster-admin
roleRef:
  kind: ClusterRole
  name: cluster-admin
  apiGroup: rbac.authorization.k8s.io
subjects:
- kind: Group
  name: AdministratorGroup
  apiGroup: rbac.authorization.k8s.io
- kind: Group
  name: OperatorGroup
  apiGroup: rbac.authorization.k8s.io
- kind: User                 # Cluster creator
  name: jane.doe # cluster creater name
  apiGroup: rbac.authorization.k8s.io
apiVersion: rbac.authorization.k8s.io/v1
kind: ClusterRoleBinding
  name: scp-cluster-admin
roleRef:
  kind: ClusterRole
  name: cluster-admin
  apiGroup: rbac.authorization.k8s.io
subjects:
- kind: Group
  name: AdministratorGroup
  apiGroup: rbac.authorization.k8s.io
- kind: Group
  name: OperatorGroup
  apiGroup: rbac.authorization.k8s.io
- kind: User                 # Cluster creator
  name: jane.doe # cluster creater name
  apiGroup: rbac.authorization.k8s.io
Code Block. Example of Cluster Role Binding scp-cluster-admin

Cluster Role Binding scp-view

The cluster role binding scp-view is bound to the cluster role view and bound to the IAM user group ViewerGroup according to the subjects.

Color mode
apiVersion: rbac.authorization.k8s.io/v1
kind: ClusterRoleBinding
metadata:
  name: scp-view
roleRef:
  kind: ClusterRole
  name: view
  apiGroup: rbac.authorization.k8s.io
subjects:
- kind: Group
  name: ViewerGroup
  apiGroup: rbac.authorization.k8s.io
apiVersion: rbac.authorization.k8s.io/v1
kind: ClusterRoleBinding
metadata:
  name: scp-view
roleRef:
  kind: ClusterRole
  name: view
  apiGroup: rbac.authorization.k8s.io
subjects:
- kind: Group
  name: ViewerGroup
  apiGroup: rbac.authorization.k8s.io
Code Block. Example of Cluster Role Binding scp-view

Cluster Role and Cluster Role Binding scp-namespace-view

Cluster Role scp-namespace-view is a role that defines the authority to view namespaces. Cluster Role Binding scp-namespace-view is associated with Cluster Role scp-namespace-view and grants namespace view authority to all authenticated users in the cluster.

Color mode
apiVersion: rbac.authorization.k8s.io/v1
kind: ClusterRole
metadata:
  name: scp-namespace-view
rules:
- apiGroups: [""]
  resources: ["namespaces"]
  verbs: ["get", "list", "watch"]
---
apiVersion: rbac.authorization.k8s.io/v1
kind: ClusterRoleBinding
metadata:
  name: scp-namespace-view
roleRef:
  kind: ClusterRole
  name: scp-namespace-view
  apiGroup: rbac.authorization.k8s.io
subjects:
- kind: Group
  name: system:authenticated
  apiGroup: rbac.authorization.k8s.io
apiVersion: rbac.authorization.k8s.io/v1
kind: ClusterRole
metadata:
  name: scp-namespace-view
rules:
- apiGroups: [""]
  resources: ["namespaces"]
  verbs: ["get", "list", "watch"]
---
apiVersion: rbac.authorization.k8s.io/v1
kind: ClusterRoleBinding
metadata:
  name: scp-namespace-view
roleRef:
  kind: ClusterRole
  name: scp-namespace-view
  apiGroup: rbac.authorization.k8s.io
subjects:
- kind: Group
  name: system:authenticated
  apiGroup: rbac.authorization.k8s.io
Code Block. Cluster Role and Cluster Role Binding scp-namespace-view Example

IAM User Group RBAC Use Case

This chapter explains examples of granting authority by major user scenarios. The names of IAM user groups, ClusterRoleBindings/RoleBindings, and ClusterRoles presented here are examples for understanding. Administrators should define and apply appropriate names and authorities according to their needs.

ScopeUse CaseIAM User GroupClusterRoleBinding/RoleBindingClusterRoleNote
ClusterCluster AdministratorClusterAdminGroupClusterRoleBinding cluster-admin-groupcluster-adminAdministrator for a specific cluster
ClusterCluster EditorClusterEditGroupClusterRoleBinding cluster-edit-groupeditEditor for a specific cluster
ClusterCluster ViewerClusterViewGroupClusterRoleBinding cluster-view-groupviewViewer for a specific cluster
NamespaceNamespace AdministratorNamespaceAdminGroupRoleBinding namespace-admin-groupadminAdministrator for a specific namespace
NamespaceNamespace EditorNamespaceEditGroupRoleBinding namespace-edit-groupeditEditor for a specific namespace
NamespaceNamespace ViewerNamespaceViewGroupRoleBinding namespace-view-groupviewViewer for a specific namespace
Table. Predefined Roles and RoleBindings, IAM User Groups, and Binding Relationships for Samsung Cloud Platform
Note
The ClusterRoles (cluster-admin, admin, edit, view) in the table above are predefined in the Kubernetes cluster. For more information, see the Role section.

Cluster Administrator

To create a cluster administrator, follow these steps:

  1. Create an IAM user group named ClusterAdminGroup.
  2. Create a ClusterRoleBinding with the following content in the target cluster:
Color mode
apiVersion: rbac.authorization.k8s.io/v1
kind: ClusterRoleBinding
metadata:
  name: cluster-admin-group
roleRef:
  kind: ClusterRole
  name: cluster-admin
  apiGroup: rbac.authorization.k8s.io
subjects:
- kind: Group
  name: ClusterAdminGroup
  apiGroup: rbac.authorization.k8s.io
apiVersion: rbac.authorization.k8s.io/v1
kind: ClusterRoleBinding
metadata:
  name: cluster-admin-group
roleRef:
  kind: ClusterRole
  name: cluster-admin
  apiGroup: rbac.authorization.k8s.io
subjects:
- kind: Group
  name: ClusterAdminGroup
  apiGroup: rbac.authorization.k8s.io
Code Block. Create Cluster Administrator
  • It is associated with the default ClusterRole cluster-admin, granting administrator authority for the cluster.

Cluster Editor

To create a cluster editor, follow these steps:

  1. Create an IAM user group named ClusterEditGroup.
  2. Create a ClusterRoleBinding with the following content in the target cluster:
Color mode
apiVersion: rbac.authorization.k8s.io/v1
kind: ClusterRoleBinding
metadata:
  name: cluster-edit-group
roleRef:
  kind: ClusterRole
  name: edit
  apiGroup: rbac.authorization.k8s.io
subjects:
- kind: Group
  name: ClusterEditGroup
  apiGroup: rbac.authorization.k8s.io
apiVersion: rbac.authorization.k8s.io/v1
kind: ClusterRoleBinding
metadata:
  name: cluster-edit-group
roleRef:
  kind: ClusterRole
  name: edit
  apiGroup: rbac.authorization.k8s.io
subjects:
- kind: Group
  name: ClusterEditGroup
  apiGroup: rbac.authorization.k8s.io
Code Block. Create Cluster Editor
  • The default cluster role edit is associated with it, and editor permissions are granted for the cluster.

Cluster Viewer

To create a cluster viewer, follow these steps:

  1. Create an IAM user group named ClusterViewGroup.
  2. Create a cluster role binding with the following content in the target cluster.
Color mode
apiVersion: rbac.authorization.k8s.io/v1
kind: ClusterRoleBinding
metadata:
  name: cluster-view-group
roleRef:
  kind: ClusterRole
  name: view
  apiGroup: rbac.authorization.k8s.io
subjects:
- kind: Group
  name: ClusterViewGroup
  apiGroup: rbac.authorization.k8s.io
apiVersion: rbac.authorization.k8s.io/v1
kind: ClusterRoleBinding
metadata:
  name: cluster-view-group
roleRef:
  kind: ClusterRole
  name: view
  apiGroup: rbac.authorization.k8s.io
subjects:
- kind: Group
  name: ClusterViewGroup
  apiGroup: rbac.authorization.k8s.io
Code block. Create Cluster Viewer
  • The default cluster role view is associated with it, and viewer permissions are granted for the cluster.

Namespace Administrator

To create a namespace administrator, follow these steps:

  1. Create an IAM user group named NamespaceAdminGroup.
  2. Create a role binding with the following content in the target cluster.
Color mode
apiVersion: rbac.authorization.k8s.io/v1
kind: RoleBinding
metadata:
  name: namespace-admin-group
  namespace: <namespace_name>
roleRef:
  kind: ClusterRole
  name: admin
  apiGroup: rbac.authorization.k8s.io
subjects:
- kind: Group
  name: NamespaceAdminGroup
  apiGroup: rbac.authorization.k8s.io
apiVersion: rbac.authorization.k8s.io/v1
kind: RoleBinding
metadata:
  name: namespace-admin-group
  namespace: <namespace_name>
roleRef:
  kind: ClusterRole
  name: admin
  apiGroup: rbac.authorization.k8s.io
subjects:
- kind: Group
  name: NamespaceAdminGroup
  apiGroup: rbac.authorization.k8s.io
Code block. Create Namespace Administrator
  • The default cluster role admin is associated with it, and administrator permissions are granted for the namespace.

Namespace Editor

To create a namespace editor, follow these steps:

  1. Create an IAM user group named NamespaceEditGroup.
  2. Create a role binding with the following content in the target cluster.
Color mode
apiVersion: rbac.authorization.k8s.io/v1
kind: RoleBinding
metadata:
  name: namespace-edit-group
  namespace: <namespace_name>
roleRef:
  kind: ClusterRole
  name: edit
  apiGroup: rbac.authorization.k8s.io
subjects:
- kind: Group
  name: NamespaceEditGroup
  apiGroup: rbac.authorization.k8s.io
apiVersion: rbac.authorization.k8s.io/v1
kind: RoleBinding
metadata:
  name: namespace-edit-group
  namespace: <namespace_name>
roleRef:
  kind: ClusterRole
  name: edit
  apiGroup: rbac.authorization.k8s.io
subjects:
- kind: Group
  name: NamespaceEditGroup
  apiGroup: rbac.authorization.k8s.io
Code block. Create Namespace Editor
  • The default cluster role edit is associated with it, and editor permissions are granted for the namespace.

Namespace Viewer

To create a namespace viewer, follow these steps:

  1. Create an IAM user group named NamespaceViewGroup.
  2. Create a role binding with the following content in the target cluster.
Color mode
apiVersion: rbac.authorization.k8s.io/v1
kind: RoleBinding
metadata:
  name: namespace-view-group
  namespace: <namespace_name>
roleRef:
  kind: ClusterRole
  name: view
  apiGroup: rbac.authorization.k8s.io
subjects:
- kind: Group
  name: NamespaceViewGroup
  apiGroup: rbac.authorization.k8s.io
apiVersion: rbac.authorization.k8s.io/v1
kind: RoleBinding
metadata:
  name: namespace-view-group
  namespace: <namespace_name>
roleRef:
  kind: ClusterRole
  name: view
  apiGroup: rbac.authorization.k8s.io
subjects:
- kind: Group
  name: NamespaceViewGroup
  apiGroup: rbac.authorization.k8s.io
Code block. Create Namespace Viewer
  • The default cluster role view is associated with it, and viewer permissions are granted for the namespace. To create a namespace viewer, follow these steps:
  1. Create an IAM user group: Create an IAM user group named NamespaceViewGroup.
  2. Create a role binding: Create a role binding with the following content in the target cluster.
Color mode
apiVersion: rbac.authorization.k8s.io/v1
kind: RoleBinding
metadata:
  name: namespace-view-group
  namespace: <namespace_name>
roleRef:
  kind: ClusterRole
  name: view
  apiGroup: rbac.authorization.k8s.io
subjects:
- kind: Group
  name: NamespaceViewGroup
  apiGroup: rbac.authorization.k8s.io
apiVersion: rbac.authorization.k8s.io/v1
kind: RoleBinding
metadata:
  name: namespace-view-group
  namespace: <namespace_name>
roleRef:
  kind: ClusterRole
  name: view
  apiGroup: rbac.authorization.k8s.io
subjects:
- kind: Group
  name: NamespaceViewGroup
  apiGroup: rbac.authorization.k8s.io
Code Block. Create Namespace Viewer
  • The view cluster role is associated with the viewer permission for the specified namespace.

Practice Example

This chapter describes an example and procedure for applying an administrator to a specific namespace.

  • IAM user group: NamespaceAdminGroup
  • IAM policy: NamespaceAdminAccess
  • Role binding: namespace-admin-group

Create an IAM User Group

Note
For more information about IAM user groups, see IAM > User Group.

To create an IAM user group in Samsung Cloud Platform, follow these steps:

  1. Click All Services > Management > IAM. The Identity and Access Management (IAM) Service Home page appears.

  2. On the Service Home page, click User Group. The User Group List page appears.

  3. On the User Group List page, click Create User Group.

    • Enter the required information in the Basic Information, Add User, Attach Policy, and Additional Information sections.

      Category
      Required
      Description
      User Group NameRequiredEnter the user group name
      • Use Korean, English, numbers, and special characters (+=,.@-_) to enter a value between 3 and 24 characters
      • Enter NamespaceAdminGroup as the user group name
      DescriptionOptionalDescription of the user group name
      • Enter a detailed description of the user group name, up to 1,000 characters
      UserOptionalUsers to add to the user group
      • The list of users registered in the account is displayed, and the selected user’s name is displayed at the top of the screen when the checkbox is selected
      • Click the Delete button at the top of the screen or uncheck the checkbox in the user list to cancel the selection of the selected user
      • If there are no users to add, click Create User at the bottom of the user list to register a new user, and then refresh the user list to select the user
      PolicyOptionalPolicy to attach to the user group
      • The list of policies registered in the account is displayed, and the selected policy name is displayed at the top of the screen when the checkbox is selected
      • Select ViewerAccess in the policy list
      TagOptionalTags to add to the user group
      • Up to 50 tags can be added per resource
      Table. User Group Creation Information Input Items
  4. Click the Complete button. The User Group List page appears.

Note

In this practice example, the ViewerAccess policy (permission to view all resources) is attached for demonstration purposes.

  • If you do not need permission to view all resources in the Samsung Cloud Platform Console, you do not need to attach the ViewerAccess policy. Define and apply a separate policy according to your actual situation.

Create an IAM Policy

Note
If you do not need to grant Samsung Cloud Platform Console usage permissions, you do not need to perform this step.
Note
For more information about IAM policies, see IAM > Policy.

To create an IAM policy in Samsung Cloud Platform, follow these steps:

  1. Click All Services > Management > IAM. The Identity and Access Management (IAM) Service Home page appears.

  2. On the Service Home page, click Policy. The Policy List page appears.

  3. On the Policy List page, click Create Policy. The Create Policy page appears.

  4. Enter the required information in the Basic Information and Additional Information sections.

    Category
    Required
    Description
    Policy NameRequiredEnter the policy name
    • Use Korean, English, numbers, and special characters (+=,.@-_) to enter a value between 3 and 128 characters
    • Enter NamespaceAdminAccess as the policy name
    DescriptionOptionalDescription of the policy name
    • Enter a detailed description of the policy name, up to 1,000 characters
    TagOptionalTags to add to the policy
    • Up to 50 tags can be added per resource
    Table. Policy Creation Information Input Items - Basic Information and Additional Information
  5. Click the Next button. The Permission Settings section appears.

  6. Enter the required information in the Permission Settings section.

    • Select Kubernetes Engine in the Service section.

    • You can create a policy by importing an existing policy using Policy Import. For more information about Policy Import, see Policy Import.

      Category
      Required
      Description
      Control TypeRequiredSelect the policy control type
      • Allow Policy: A policy that allows defined permissions
      • Deny Policy: A policy that denies defined permissions
      The deny policy takes precedence for the same target
      ActionRequiredSelect actions provided by each service
      • Create: CreateKubernetesObject selected
      • Delete: DeleteKubernetesObject selected
      • List: ListKubernetesEngine, ListKubernetesObject selected
      • Read: DetailKubernetesObject selected
      • Update: UpdateKubernetesObject selected
      • Add Action Directly: Use wildcard * to specify multiple actions at once
      Applied ResourceRequiredResource to which the action is applied
      • All Resources: Apply to all resources for the selected action
      • Individual Resource: Apply only to the specified resource for the selected action
        • Individual resources are only possible when selecting actions that allow individual resource selection (purple actions)
        • Click the Add Resource button to specify the target resource by resource type
        • For more information on Add Resource, see Registering individual resources as applied resources
      Authentication TypeRequiredAuthentication method for the target user
      • All Authentication: Apply regardless of authentication method
      • API Key Authentication: Apply to users who use API key authentication
      • IAM Key Authentication, Console Login: Apply to users who use IAM key authentication or console login
      Applied IPRequiredIP addresses to which the policy is applied
      • User-specified IP: Register and manage IP addresses directly by the user
        • Applied IP: Register IP addresses directly by the user as IP addresses or ranges to which the policy is applied
        • Excluded IP: Register IP addresses to be excluded from Applied IP as IP addresses or ranges
      • All IP: Do not restrict IP access
        • Allow access to all IP addresses, but if exceptions are needed, register Excluded IP to restrict access to registered IP addresses
      Table. Policy creation information input items - Permission settings
Note

Permission settings provide Basic Mode and JSON Mode.

  • If you write in Basic Mode and enter JSON Mode or move to another screen, services with the same conditions will be integrated into one, and settings that are not completed will be deleted.
  • If the content written in JSON Mode does not match the JSON format, you cannot switch to Basic Mode.
  1. Click the Next button. Move to the Input Information Check page.
  2. Check the input information and click the Complete button. Move to the Policy List page.

Add a user to an IAM user group

Reference
For more information on managing IAM user groups, see IAM > Managing User Groups.

To add a user to an IAM user group in Samsung Cloud Platform, follow these steps.

  1. Click All Services > Management > IAM menu. Move to the Identity and Access Management (IAM) Service Home page.
  2. On the Service Home page, click the User menu. Move to the User List page.
  3. On the User List page, click the user to be added to the IAM user group. Move to the User Details page.
  4. On the User Details page, click the User Group tab.
  5. On the user group tab, select the Add User Group button. Move to the Add User Group page.
  6. On the Add User Group page, select the user group to be added and click the Complete button. Move to the User Details page.
    • Select NamespaceAdminGroup from the user group.

Create a role binding

Create a role binding by referring to the example below.

Color mode
apiVersion: rbac.authorization.k8s.io/v1
kind: RoleBinding
metadata:
  name: namespace-admin-group
  namespace: dev # target namespace
roleRef:
  kind: ClusterRole
  name: admin # pre-defined cluster role in Kubernetes
  apiGroup: rbac.authorization.k8s.io
subjects: 
- kind: Group
  name: NamespaceAdminGroup # IAM user group created earlier
  apiGroup: rbac.authorization.k8s.io
apiVersion: rbac.authorization.k8s.io/v1
kind: RoleBinding
metadata:
  name: namespace-admin-group
  namespace: dev # target namespace
roleRef:
  kind: ClusterRole
  name: admin # pre-defined cluster role in Kubernetes
  apiGroup: rbac.authorization.k8s.io
subjects: 
- kind: Group
  name: NamespaceAdminGroup # IAM user group created earlier
  apiGroup: rbac.authorization.k8s.io
Code block. Create a role binding

Verify the user

Verify that the user’s namespace permissions are applied normally.

To verify namespace user permissions in Samsung Cloud Platform, follow these steps.

  1. Click All Services > Container > Kubernetes Engine menu. Move to the Kubernetes Engine Service Home page.
  2. On the Service Home page, click Workload menu under Pod. Move to the Pod List page.
  3. On the Pod List page, select the cluster and namespace from the gear button at the top left and click Confirm.
  4. On the Pod List page, verify that the pod list is retrieved.
    • If you select a namespace with permissions, the pod list will be displayed.
    • If you select a namespace without permissions, a confirmation window will be displayed indicating that you do not have permission to retrieve the list.

3.2 - Accessing the Cluster

kubectl Installation and Usage Guide

After creating a Kubernetes Engine service, you can use the Kubernetes command-line tool kubectl to execute commands on a Kubernetes cluster. Using kubectl, you can deploy applications, inspect and manage cluster resources, and view logs. You can find how to install and use kubectl in the official Kubernetes documentation as follows.

Reference

You must use a kubectl version that is within the minor version difference of the cluster. For example, if the cluster version is 1.30, you can use kubectl versions 1.29, 1.30, or 1.31.

To access a Kubernetes cluster with kubectl, you need a kubeconfig file containing the Kubernetes server address and authentication information.

Reference
For detailed information on Kubernetes authentication and authorization, see Authentication and Authorization.

Kubernetes Engine supports authentication via admin certificate kubeconfig and user authentication key kubeconfig.

admin certificate kubeconfig

This kubeconfig uses the admin certificate as an authentication method when accessing the Kubernetes API.

Admin kubeconfig download

Kubernetes Engine > Cluster List > Cluster Details > Admin kubeconfig Download button to click and download the kubeconfig file.

Caution
  • Administrator kubeconfig download is only possible for Admin.
  • There are separate private endpoint and public endpoint versions, and you can download each only once.

Admin kubeconfig use

Reference
  • By default, kubectl looks for a file named config in the $HOME/.kube directory. Or you can set the KUBECONFIG environment variable or specify the kubeconfig flag to use a different kubeconfig file.
  • Private endpoints are by default only accessible from nodes of the respective cluster. For resources in the same Account and same region, you can allow access by adding them to the private endpoint access control settings.
  • If you need to access the cluster from the external internet, setting public endpoint access to enabled allows you to access using the public endpoint kubeconfig.

User authentication key kubeconfig

This kubeconfig uses the user’s Open API authentication key as the authentication method when accessing the Kubernetes API.

User kubeconfig download

Kubernetes Engine > Cluster List > Cluster Details > User kubeconfig download Click the button to download the kubeconfig file.

Caution
  • User kubeconfig download is only possible for users with cluster view permission.
  • There are separate ones for private endpoint and public endpoint.
  • Since the downloaded kubeconfig file does not contain the authentication key token, you need to add the authentication key token information before using it. (See the next paragraph)

Add authentication key token to user kubeconfig file

Below is an example of a user’s kubeconfig file. To use the kubeconfig file, you need to add the authentication key token (AUTHKEY_TOKEN) information in the token field inside the file.

Color mode
apiVersion: v1
clusters:
- cluster:
    certificate-authority-data: LS0tLS1CRUdJTiBDRVJUSUZJQ0FURS0t...
    server: https://my-cluster-a1c3e.ske.xxx.samsungsdscloud.com:6443
  name: my-cluster-a1c3e
contexts:
- context:
    cluster: my-cluster-a1c3e
    user: jane.doe
  name: jane.doe@my-cluster-a1c3e
current-context: jane.doe@my-cluster-a1c3e
kind: Config
preferences: {}
users:
- name: jane.doe
  user:
    token: <AUTHKEY_TOKEN> #### writing needed
apiVersion: v1
clusters:
- cluster:
    certificate-authority-data: LS0tLS1CRUdJTiBDRVJUSUZJQ0FURS0t...
    server: https://my-cluster-a1c3e.ske.xxx.samsungsdscloud.com:6443
  name: my-cluster-a1c3e
contexts:
- context:
    cluster: my-cluster-a1c3e
    user: jane.doe
  name: jane.doe@my-cluster-a1c3e
current-context: jane.doe@my-cluster-a1c3e
kind: Config
preferences: {}
users:
- name: jane.doe
  user:
    token: <AUTHKEY_TOKEN> #### writing needed
Code block. User kubeconfig file example

AUTHKEY_TOKEN can be generated by concatenating the authentication key’s ACCESS_KEY and SECRET_KEY with a colon (:) and then Base64 encoding it. The following is an example of creating AUTHKEY_TOKEN in a Linux environment.

Color mode
$ ACCESS_KEY=5df418813aed051548a72f4a814cf09e
$ SECRET_KEY=6ba7b810-9dad-11d1-80b4-00c04fd430c8
$ AUTHKEY_TOKEN=$(echo -n "$ACCESS_KEY:$SECRET_KEY" | base64 -w0)
$ echo $AUTHKEY_TOKEN
NWRmNDE4ODEzYWVkMDUxNTQ4YTcyZjRhODE0Y2YwOWU6NmJhN2I4MTAtOWRhZC0xMWQxLTgwYjQtMDBmMDRmZDQzMGM4r
$ ACCESS_KEY=5df418813aed051548a72f4a814cf09e
$ SECRET_KEY=6ba7b810-9dad-11d1-80b4-00c04fd430c8
$ AUTHKEY_TOKEN=$(echo -n "$ACCESS_KEY:$SECRET_KEY" | base64 -w0)
$ echo $AUTHKEY_TOKEN
NWRmNDE4ODEzYWVkMDUxNTQ4YTcyZjRhODE0Y2YwOWU6NmJhN2I4MTAtOWRhZC0xMWQxLTgwYjQtMDBmMDRmZDQzMGM4r
Code block. AUTHKEY_TOKEN value generation example
Note
  • For detailed information on authentication key generation, please refer to API Reference > Common > Samsung Cloud Platform Open API call procedure.

User kubeconfig execution example

You can see an example of executing the user kubeconfig.

When access is blocked by access control or a firewall

Color mode
$ kubectl --kubeconfig=user-kubeconfig.yaml get namespaces
Unable to connect to the server: dial tcp 123.123.123.123:6443: i/o timeout
$ kubectl --kubeconfig=user-kubeconfig.yaml get namespaces
Unable to connect to the server: dial tcp 123.123.123.123:6443: i/o timeout
Code block. Example execution when access is blocked by access control or firewall

When AUTHKEY_TOKEN does not match and authentication fails

Color mode
$ kubectl --kubeconfig=user-kubeconfig.yaml get namespaces
error: You must be logged in to the server (Unauthorized)
$ kubectl --kubeconfig=user-kubeconfig.yaml get namespaces
error: You must be logged in to the server (Unauthorized)
Code block. Example execution when authentication fails because AUTHKEY_TOKEN does not match

AUTHKEY_TOKEN When authentication succeeds

Color mode
$ kubectl --kubeconfig=user-kubeconfig.yaml get namespaces
...
kube-node-lease    Active 10d
kube-public        Active 10d
kube-system        Active 10d
$ kubectl --kubeconfig=user-kubeconfig.yaml get namespaces
...
kube-node-lease    Active 10d
kube-public        Active 10d
kube-system        Active 10d
Code block. Example execution when AUTHKEY_TOKEN authentication succeeds

AUTHKEY_TOKEN Authentication succeeded but no permission

Color mode
$ kubectl --kubeconfig=user-kubeconfig.yaml get nodes
Error from server (Forbidden): nodes is forbidden: User "jane.doe" cannot list resource "nodes" in API group "" at the cluster scope
$ kubectl --kubeconfig=user-kubeconfig.yaml get nodes
Error from server (Forbidden): nodes is forbidden: User "jane.doe" cannot list resource "nodes" in API group "" at the cluster scope
Code block. Example execution when AUTHKEY_TOKEN authentication succeeds but lacks permission
Reference
If AUTHKEY_TOKEN authentication succeeds but there is no permission, it means that the authentication process was completed correctly, but the authority to perform the requested operation was not granted (authorized). For detailed information about authorization, see Authentication and Authorization.

3.3 - type LoadBalancer Service Usage

Service Configuration Method

Service manifest file (example: my-lb-svc.yaml ) can be written and applied to configure a Service of type LoadBalancer.

Caution
  • LoadBalancer is created in the cluster Subnet by default.
  • To create a LoadBalancer in a different Subnet, use the annotation service.beta.kubernetes.io/scp-load-balancer-subnet-id. For more details, see Annotation Detailed Settings

To create and apply a type LoadBalancer Service, follow the steps below.

  1. Service manifest file my-lb-svc.yaml write.

    Color mode
    apiVersion: v1
    kind: Service
    metadata:
      name: my-service
    spec:
      selector:
        app.kubernetes.io/name: MyApp
      ports:
        - protocol: TCP
          port: 80
          targetPort: 9376
          appProtocol: tcp # Refer to the LB service protocol type setting section
      type: LoadBalancer
    apiVersion: v1
    kind: Service
    metadata:
      name: my-service
    spec:
      selector:
        app.kubernetes.io/name: MyApp
      ports:
        - protocol: TCP
          port: 80
          targetPort: 9376
          appProtocol: tcp # Refer to the LB service protocol type setting section
      type: LoadBalancer
    Code block. Service manifest file my-lb-svc.yaml example

  2. Deploy the Service manifest using the kubectl apply command.

    Color mode
      
    kubectl apply -f my-lb-svc.yaml
      
    kubectl apply -f my-lb-svc.yaml
    Code block. Deploy Service manifest with kubectl apply command

Caution
  • When a type LoadBalancer Service is created, the corresponding Load Balancer service is automatically created. It may take a few minutes for the configuration to complete.
  • Do not arbitrarily modify the automatically generated Load Balancer service and LB server group. Changes may be reverted or cause unexpected behavior.
  • For configurable detailed functions, refer to Annotation Detailed Settings.
  1. kubectl get service Use the command to check the Load Balancer configuration.
    Color mode
    # kubectl get service my-lb-svc
    NAMESPACE     NAME         TYPE           CLUSTER-IP       EXTERNAL-IP       PORT(S)         AGE
    default       my-lb-svc    LoadBalancer   172.20.49.206    123.123.123.123   80:32068/TCP    3m
    # kubectl get service my-lb-svc
    NAMESPACE     NAME         TYPE           CLUSTER-IP       EXTERNAL-IP       PORT(S)         AGE
    default       my-lb-svc    LoadBalancer   172.20.49.206    123.123.123.123   80:32068/TCP    3m
    Code block. Verify Load Balancer configuration with kubectl get service command

Protocol Type

You can create a Service manifest and use it. Here is a simple example.

Color mode
apiVersion: v1
kind: Service
metadata:
  name: my-service
spec:
  selector:
    ...
  ports:
    - port: 80
      targetPort: 9376
      protocol: TCP    # required (choose one of TCP, UDP)
      appProtocol: tcp # choice (if not entered, select one of tcp, http, https)
  type: LoadBalancer   # type LoadBalancer
apiVersion: v1
kind: Service
metadata:
  name: my-service
spec:
  selector:
    ...
  ports:
    - port: 80
      targetPort: 9376
      protocol: TCP    # required (choose one of TCP, UDP)
      appProtocol: tcp # choice (if not entered, select one of tcp, http, https)
  type: LoadBalancer   # type LoadBalancer
Code block. Service manifest writing example

The list of protocols (protocol and appProtocol) supported by the type Load Balancer Service in Kubernetes Engine, and the settings applied to the Load Balancer service accordingly, are as follows.

Category(k8s)
protocol
(k8s)
appProtocol
(LB)
Service Category
(LB)
LB Listener
(LB)
LB Server Group
(LB)
Health Check
L4 TCPTCP(tcp)L4TCP {port}TCP {nodePort}TCP {nodePort}
L4 UDPUDP-L4UDP {port}UDP {nodePort}TCP {nodePort}
L7 HTTPTCPhttpL7HTTP {port}TCP {nodePort}TCP/HTTP {nodePort}
L7 HTTPSTCPhttpsL7HTTPS {port}TCP {nodePort}TCP/HTTP {nodePort}
Table. k8s Service manifest and Load Balancer service configuration
  • k8s Service can specify multiple ports for a single service according to the manifest spec.
Caution

According to the Load Balancer service classification (L4, L7), you cannot mix protocol layers within a single Service.

  • Thus L4 (TCP, UDP) and L7 (HTTP, HTTPS) cannot be used together in a single Service.

L4 Service Manifest creation example

Color mode
apiVersion: v1
kind: Service
metadata:
  name: my-service
spec:
  selector:
    app.kubernetes.io/name: MyApp
  ports:
    - protocol: TCP
      port: 80
      targetPort: 9376
  type: LoadBalancer
apiVersion: v1
kind: Service
metadata:
  name: my-service
spec:
  selector:
    app.kubernetes.io/name: MyApp
  ports:
    - protocol: TCP
      port: 80
      targetPort: 9376
  type: LoadBalancer
Code block. L4 Service manifest writing example

L7 Service Manifest creation example

Color mode
apiVersion: v1
kind: Service
metadata:
  annotations:
    service.beta.kubernetes.io/scp-load-balancer-layer-type: "L7" # required
    service.beta.kubernetes.io/scp-load-balancer-client-cert-id: "24da35de187b450eb0cf09fb6fa146de" # required
  name: my-service
spec:
  selector:
    app.kubernetes.io/name: MyApp
  ports:
    - appProtocol: http # required
      protocol: TCP
      port: 80
      targetPort: 9376
    - appProtocol: https # required
      protocol: TCP
      port: 443
      targetPort: 9898
  type: LoadBalancer
  
apiVersion: v1
kind: Service
metadata:
  annotations:
    service.beta.kubernetes.io/scp-load-balancer-layer-type: "L7" # required
    service.beta.kubernetes.io/scp-load-balancer-client-cert-id: "24da35de187b450eb0cf09fb6fa146de" # required
  name: my-service
spec:
  selector:
    app.kubernetes.io/name: MyApp
  ports:
    - appProtocol: http # required
      protocol: TCP
      port: 80
      targetPort: 9376
    - appProtocol: https # required
      protocol: TCP
      port: 443
      targetPort: 9898
  type: LoadBalancer
  
Code block. L7 Service manifest writing example

Annotation Detailed Settings

You can add annotations to the service manifest to configure detailed functions.

Color mode
apiVersion: v1
kind: Service
metatdata:
  name: my-lb-svc
  annotations:
    service.beta.kubernetes.io/scp-load-balancer-public-ip-enabled: "true"
    service.beta.kubernetes.io/scp-load-balancer-health-check-interval: "5"
    service.beta.kubernetes.io/scp-load-balancer-health-check-timeout: "5"
    service.beta.kubernetes.io/scp-load-balancer-health-check-count: "3"
    service.beta.kubernetes.io/scp-load-balancer-session-duration-time: "300"
  spec:
  type: LoadBalancer
  ...
  
apiVersion: v1
kind: Service
metatdata:
  name: my-lb-svc
  annotations:
    service.beta.kubernetes.io/scp-load-balancer-public-ip-enabled: "true"
    service.beta.kubernetes.io/scp-load-balancer-health-check-interval: "5"
    service.beta.kubernetes.io/scp-load-balancer-health-check-timeout: "5"
    service.beta.kubernetes.io/scp-load-balancer-health-check-count: "3"
    service.beta.kubernetes.io/scp-load-balancer-session-duration-time: "300"
  spec:
  type: LoadBalancer
  ...
  
Code block. Example of adding annotation to service manifest

If no separate annotation is added to the service, the default and allowed values of the applied annotation are as follows. Also, check the precautions for each annotation.

AnnotationProtocolDefaultAllowed valuesExampleDescription
service.beta.kubernetes.io/scp-load-balancer-source-ranges-firewall-rulesAllfalsetrue, falsefalseAutomatically add firewall rules (LB source ranges → LB service IP)
service.beta.kubernetes.io/scp-load-balancer-snat-healthcheck-firewall-rulesAllfalsetrue,falsefalseAutomatically add firewall rules (LB Source NAT IP, HealthCheck IP → member IP:Port)
  • If you use this annotation, firewall rules are added for each port of a type LB service, so the number of firewall rules can become very large.
  • If having too many firewall rules is a burden, as an alternative you can add firewall rules manually without using this annotation. For example, you can add firewall rules that target the member IP’s NodePort range (30000-32767).
Table. Firewall-related settings in Kubernetes annotations
AnnotationProtocolDefaultAllowed valuesExampleDescription
service.beta.kubernetes.io/scp-load-balancer-security-group-idAll-UUID92d84b44-ee71-493d-9782-3a90481ce5f3Automatically adds rules to the Security Group corresponding to the specified ID
  • If you use this annotation, rules are added to the Security Group for each port of the type LB service, so Security Group rules can become very numerous.
  • If having too many Security Group rules is burdensome, you can alternatively add Security Group rules manually without using this annotation. For example, you can set the target address to the Load Balancer’s Source NAT IP and health check IP, and add a Security Group rule that allows ports in the NodePort range (30000-32767).
  • Security Group rules added by this annotation are not automatically deleted even if this annotation is removed or changed.
  • Multiple entries can be added separated by commas. (Example: ddc25ad8-6d3f-4242-8c86-2a059212ddc6,26ab7fe1-b3ea-4aa9-9e9d-35a7c237904e)
  • This annotation can be used together with the service.beta.kubernetes.io/scp-load-balancer-security-group-name annotation, and rules are automatically added to all Security Groups that meet the criteria.
service.beta.kubernetes.io/scp-load-balancer-security-group-nameAll-stringsecurity-group-1Automatically adds rules to the Security Group corresponding to the specified Name
  • If you use this annotation, rules are added to the Security Group for each port of the type LB service, so Security Group rules can become very numerous.
  • If having too many Security Group rules is burdensome, you can alternatively add Security Group rules manually without using this annotation. For example, you can set the target address to the Load Balancer’s Source NAT IP and health check IP, and add a Security Group rule that allows ports in the NodePort range (30000-32767).
  • Security Group rules added by this annotation are not automatically removed even if the annotation is deleted or changed.
  • Multiple can be added separated by commas (e.g., security-group-1,security-group-2)
  • This annotation can be used together with the service.beta.kubernetes.io/scp-load-balancer-security-group-id annotation, and rules are automatically added to all Security Groups that meet the conditions.
Table. Security Group related settings in Kubernetes annotations
AnnotationProtocolDefaultAllowed valuesExampleDescription
service.beta.kubernetes.io/scp-load-balancer-layer-typeAllL4L4, L7L4Specify the service type of the Load Balancer
  • When using this annotation, specify L4 if you want to use TCP or UDP, and L7 if you want to use HTTP or HTTPS.
  • Cannot be changed after initial creation. To change it, you must recreate the service.
service.beta.kubernetes.io/scp-load-balancer-subnet-idAll-ID7f05eda5e1cf4a45971227c57a6d60faSpecify the Service Subnet of the Load Balancer
  • If this annotation is not specified, the cluster’s Subnet is used.
  • It cannot be changed after initial creation. To change it, the service must be recreated.
service.beta.kubernetes.io/scp-load-balancer-service-ipAll-IP address192.168.10.7Specify the Service IP of the Load Balancer
  • Cannot be changed after initial creation. To change it, you must recreate the service.
service.beta.kubernetes.io/scp-load-balancer-public-ip-enabledAllfalsetrue, falsefalseSpecify whether to use the Load Balancer’s Public NAT IP
  • If this annotation is set to true and service.beta.kubernetes.io/scp-load-balancer-public-ip-id is not specified, an IP is automatically assigned.
  • If this annotation is set to true and service.beta.kubernetes.io/scp-load-balancer-public-ip-id is specified, the Public IP corresponding to the specified ID is applied.
service.beta.kubernetes.io/scp-load-balancer-public-ip-idAll-ID4119894bd9614cef83db6f8dda667a20Specify the ID of the Public IP to be used as the Load Balancer’s Public NAT IP
  • If service.beta.kubernetes.io/scp-load-balancer-public-ip-enabled is not set to true, this annotation is ignored.
  • If service.beta.kubernetes.io/scp-load-balancer-public-ip-enabled is set to true and this annotation is specified, the Public IP corresponding to the specified ID is applied.
Table. Load Balancer related settings in Kubernetes annotations
AnnotationProtocolDefaultAllowed valuesExampleDescription
service.beta.kubernetes.io/scp-load-balancer-response-timeoutHTTP, HTTPS00 - 12060Specify the response timeout (seconds) of the LB Listener
  • 0 means the response timeout feature is disabled.
  • After setting to 1 - 120, it cannot be changed to 0. To change, you must recreate the service.
service.beta.kubernetes.io/scp-load-balancer-session-duration-timeAll1200 - 120120Specify the session persistence time (seconds) of the LB Listener
  • 0 means the session persistence feature is disabled.
  • After setting to 1 - 120, it cannot be changed to 0. To change, you must recreate the service.
service.beta.kubernetes.io/scp-load-balancer-insert-client-ipTCPfalsetrue, falsefalseSpecify Insert Client IP of LB Listener
service.beta.kubernetes.io/scp-load-balancer-x-forwarded-protoHTTP, HTTPSfalsetrue, falsefalseSpecify whether to use the X-Forwarded-Proto header of the LB Listener
service.beta.kubernetes.io/scp-load-balancer-x-forwarded-portHTTP, HTTPSfalsetrue, falsefalseSpecify whether to use the X-Forwarded-Port header of the LB Listener
service.beta.kubernetes.io/scp-load-balancer-x-forwarded-forHTTP, HTTPSfalsetrue, falsefalseSpecify whether to use the X-Forwarded-For header of the LB Listener
service.beta.kubernetes.io/scp-load-balancer-support-http2HTTP, HTTPSfalsetrue, falsefalseSpecify whether the LB Listener supports HTTP 2.0
service.beta.kubernetes.io/scp-load-balancer-persistenceTCP, HTTP, HTTPS"""", source-ip, cookiesource-ipSpecify the LB Listener’s persistence (none, source IP, or cookie)
  • For UDP, this annotation cannot be used.
  • For TCP, you can specify "" or source-ip to use.
  • For HTTP/HTTPS, you can specify one of “”, source-ip, cookie to use.
service.beta.kubernetes.io/scp-load-balancer-client-cert-idHTTPS-UUID78b9105e00324715b63700933125fa83Specify the ID of the client SSL certificate for the LB Listener
  • HTTPS is a required field when selected.
service.beta.kubernetes.io/scp-load-balancer-client-cert-levelHTTPSHIGHHIGH, NORMAL, LOWHIGHSpecify the security level of the client SSL certificate for the LB Listener
service.beta.kubernetes.io/scp-load-balancer-server-cert-levelHTTPS-HIGH, NORMAL, LOWHIGHSpecifies the security level of the LB Listener’s server SSL certificate
Table. Settings related to LB Listener in Kubernetes annotations
AnnotationProtocolDefault valueAllowed valuesExampleDescription
service.beta.kubernetes.io/scp-load-balancer-lb-methodAllROUND_ROBINROUND_ROBIN, LEAST_CONNECTION, IP_HASHROUND_ROBINSpecify the load balancing policy of the LB server group
Table. Settings related to LB server group in Kubernetes annotations
AnnotationProtocolDefaultAllowed valuesExampleDescription
service.beta.kubernetes.io/scp-load-balancer-health-check-enabledAlltruetrue, falsetrueSpecify whether to use LB health check
service.beta.kubernetes.io/scp-load-balancer-health-check-protocolAllTCPTCP, HTTPTCPSpecify the protocol for LB health check
service.beta.kubernetes.io/scp-load-balancer-health-check-portAll{nodeport}1 - 6553430000Specify the health check port for LB health check
  • {nodeport} is set as default, so generally you don’t need to specify it.
service.beta.kubernetes.io/scp-load-balancer-health-check-countAll31 - 103Specify the detection count of LB health check
service.beta.kubernetes.io/scp-load-balancer-health-check-intervalAll51 - 1805Specifies the LB health check interval
service.beta.kubernetes.io/scp-load-balancer-health-check-timeoutAll51 - 1805Specify the LB health check’s wait time
service.beta.kubernetes.io/scp-load-balancer-health-check-http-methodHTTPGETGET, POSTGETSpecify the HTTP method for LB health check
service.beta.kubernetes.io/scp-load-balancer-health-check-urlHTTP/string/healthzSpecify the URL for LB health check
service.beta.kubernetes.io/scp-load-balancer-health-check-response-codeHTTP200200 - 500200Specify the response code for LB health check
service.beta.kubernetes.io/scp-load-balancer-health-check-request-dataHTTP-stringusername=admin&password=1234Specify the request string for LB health check
  • HTTPS is required input when specified.
service.beta.kubernetes.io/scp-load-balancer-port-{port}-health-check-enabledAlltruetrue, falsetrueSpecifies whether to use LB health check for the Service’s {port} port number
service.beta.kubernetes.io/scp-load-balancer-port-{port}-health-check-protocolAllTCPTCP, HTTPTCPSpecifies the LB health check protocol for the Service’s {port} port number
service.beta.kubernetes.io/scp-load-balancer-port-{port}-health-check-portAll-1 - 6553430000Specify the LB health check port for the Service’s {port} port number
service.beta.kubernetes.io/scp-load-balancer-port-{port}-health-check-countAll31 - 103Specifies the number of LB health check detections for the Service’s {port} port number
service.beta.kubernetes.io/scp-load-balancer-port-{port}-health-check-intervalAll51 - 1805Specifies the LB health check interval for the Service’s {port} port number
service.beta.kubernetes.io/scp-load-balancer-port-{port}-health-check-timeoutAll51 - 1805Specifies the LB health check timeout for the Service’s {port} port number
service.beta.kubernetes.io/scp-load-balancer-port-{port}-health-check-http-methodHTTPGETGET, POSTGETSpecify the LB health check HTTP method for the Service’s {port} port number
service.beta.kubernetes.io/scp-load-balancer-port-{port}-health-check-urlHTTP/string/healthzSpecifies the LB health check URL for the Service’s {port} port number
service.beta.kubernetes.io/scp-load-balancer-port-{port}-health-check-response-codeHTTP200200 - 500200Specifies the LB health check response code for the Service’s {port} port number
service.beta.kubernetes.io/scp-load-balancer-port-{port}-health-check-request-dataHTTP-Stringusername=admin&password=1234Specify the LB health check request string for the Service’s {port} port number
  • It is a required field when HTTPS is specified.
Table. Settings related to LB health check in Kubernetes annotations

Constraints

The constraints to consider when using Kubernetes annotations are as follows.

ConstraintsRelated Annotations
When changing the Security Group, rules created in the existing Security Group are not automatically deletedservice.beta.kubernetes.io/scp-load-balancer-security-group-id
service.beta.kubernetes.io/scp-load-balancer-security-group-name
Cannot change the service classification (L4/L7) of the Load Balancerservice.beta.kubernetes.io/scp-load-balancer-layer-type
Cannot use L4 and L7 together within the same k8s Serviceservice.beta.kubernetes.io/scp-load-balancer-layer-type
Load Balancer cannot change subnetservice.beta.kubernetes.io/scp-load-balancer-subnet-id
Cannot change the Service IP of the Load Balancerservice.beta.kubernetes.io/scp-load-balancer-service-ip
LB Listener response timeout cannot be changed from enabled (1 - 120) to disabled (0)service.beta.kubernetes.io/scp-load-balancer-response-timeout
Cannot use TCP and UDP together on the same port number within the same k8s Service-
For L7 HTTP/HTTPS, the routing action applies the URL processing Default pattern ("/")
  • To add other URL patterns, you must add them directly in the Samsung Cloud Platform console
  • URL processing only supported and URL redirection is not supported
-
Table. Constraints when using Kubernetes annotations

3.4 - Considerations for Use

Managed Port Constraints

The following ports are used for SKE management and cannot be used for service use. In addition, if blocked by OS firewall, etc., node functions or some functions may not work normally.

PortDescription
UDP 4789calico-vxlan
TCP 5473calico-typha
TCP 10250kubelet
TCP 19100node-exporter
TCP 19400dcgm-exporter
Table. Managed Port List

kube-reserved resource constraints

kube-reserved is a feature that reserves resources for system daemons that do not run as pods on the node.

  • There are system daemons that do not run as pods, such as kubelet, container runtime, etc.
Reference

For more information on kube-reserved, please refer to the following document.

Kubernetes Engine reserves CPU and memory based on the following criteria.

CPU specificationMemory specification
  • First core’s 6%
  • Next core’s 1% (up to 2 cores)
  • Next 2 cores’ 0.5% (up to 4 cores)
  • Cores exceeding 4 cores’ 0.25%
  • First 4 GB memory’s 25%
  • Next 4 GB memory’s 20% (up to 8 GB)
  • Next 8 GB memory’s 10% (up to 16 GB)
  • Next 112 GB memory’s 6% (up to 128 GB)
  • Memory exceeding 128 GB’s 2%
Table. Resource reservation items based on CPU and memory
  • Example: For a Virtual Server with 16-core vCPU and 32G Memory, kube-reserved is calculated as follows.

    • CPU: (1 core × 0.06) + (1 core × 0.01) + (2 cores × 0.005) + (12 cores × 0.0025) = 0.11 core
    • Memory: (4 GB × 0.25) + (4 GB × 0.2) + (8 GB × 0.1) + (16 GB × 0.06) = 3.56 GB
  • Example: The resources reserved according to CPU size are as follows.

CPU specificationResource specification1Resource specification2Resource specification3Resource specification4
kube-reserved CPU70 m80 m90 m110 m
Table. Example of resources reserved according to CPU size
  • Example: The resources reserved according to the memory size are as follows.
Memory SpecificationResource Specification1Resource Specification2Resource Specification3Resource Specification4Resource Specification4Resource Specification4Resource Specification4
kube-reserved memory1 GB1.8 GB2.6 GB3.56 GB5.48 GB9.32 GB11.88 GB
Table. Example of resources reserved according to memory size

3.5 -

4 - API Reference

API Reference

5 - CLI Reference

CLI Reference

6 - Release Note

Kubernetes Engine

2025.10.23
FEATURE Added ServiceWatch log collection feature
  • Kubernetes Engine feature change
  • ServiceWatch provides log collection functionality.
2025.07.01
FEATURE Kubernetes version addition, public endpoint provision, private endpoint access control target addition, node pool Label/Taint, Block Storage CSI, kubectl login plugin addition
  • Kubernetes Engine feature change
  • Kubernetes v1.31 version is supported.
  • Provides the cluster’s public endpoint.
  • The private endpoint access control target of the cluster will include MNGC (Baremetal) products and DevOps Service products.
  • It provides Node Pool Label and Taint setting features.
  • Block Storage CSI and kubectl login plugin features are provided.
  • The kubeconfig vulnerability has been improved.
2025.04.28
FEATURE Private Endpoint Access Control, type: LB function added
  • Kubernetes Engine feature change
  • It provides private endpoint and access control features.
  • type: LoadBalancer provides the load balancer function.
2025.02.27
FEATURE Kubernetes version addition and Kubernetes version upgrade, Custom Image, GPU node creation functionality addition
  • Kubernetes Engine feature change
  • Kubernetes v1.30 version is supported.
  • Provides Kubernetes version upgrade functionality for clusters and node pools.
  • It provides Multi-Security Group feature.
  • Custom Image node and GPU node creation features are provided.
  • Samsung Cloud Platform common feature change
  • Account, IAM and Service Home, tags, etc. common CX changes have been reflected.
2024.10.01
NEW Kubernetes Engine Service Official Version Release
  • Launched Kubernetes Engine product that provides Container, a lightweight virtual computing, and Kubernetes cluster to manage it.
  • Container nodes are created and integrated management is possible through the cluster, allowing for the deployment of various Container applications.
2024.07.02
NEW Beta version release
  • Kuberntes Engine product Beta version has been released.