Manage Node Pools
Last updated
Last updated
Prerequisite: Only contract owners, administrators, and users having Create Kubernetes Clusters permission can create node pools. Other user types have read-only access.
You can create a cluster using the Kubernetes Manager in DCD for Public Node Pools.
In the DCD, go to Containers > Managed Kubernetes.
Select a cluster from the list and go to the Node pools in Cluster tab.
Select + Create node pool.
In Create Kubernetes node pool, configure your node pools.
In Pool Settings, provide the following information:
Pool Name: Enter a name that aligns with the Kubernetes naming convention.
Data Center: Select an option from the drop-down list. Your node pool will be included in the selected data center. If you do not have a data center, you must first create one.
Node pool version: Select an appropriate version from the drop-down list.
Node count: Select the number of nodes in the node count.
Autoscale: Select the checkbox to enable autoscale and provide a minimum and maximum number of the total nodes.
Attached private LANs: Select + and choose a private LAN from the drop-down list.
Reserved IPs: Select + and choose a reserved IP address from the drop-down list.
In the Node Pool Template, provide the following information:
CPU: Select an option from the drop-down list.
Cores: Select the number of cores.
RAM: Select the size of your RAM.
Availability Zone: Select a zone from the drop-down list.
Storage Type: Select a type of storage from the drop-down list.
Storage Size: Select the storage size for your storage.
Note: Make sure to use the following naming convention for the Kubernetes cluster:
Can be a maximum of 63 characters in length.
Begins and ends with an alphanumeric character ([a-z0-9A-Z]).
Must not contain spaces or any other white-space characters.
Can contain dashes (-), underscores (_), and dots (.) in between.
Select Create node pool.
Result: A node pool is successfully created and can be used once it reaches the active state.
When a node fails or becomes unresponsive you can rebuild that node. This will create a new node with an identical configuration that will replace the failed node.
Prerequisite: Make sure your node is active.
Select a cluster from the list and go to the Node pools in Cluster tab.
Select the node pool that contains the failed node.
Select Rebuild.
Confirm your selection by selecting OK.
Result:
Managed Kubernetes starts a process that is based on the Node Template. The template creates and configures a new node. Once the status is updated to ACTIVE, then it migrates all the pods from the faulty node to the new node.
The faulty node is deleted once it is empty.
While this operation occurs, the node pool will have an extra billable active node.
The node pool is successfully rebuilt.
Avoid accessing the target data center while Managed Kubernetes is provisioning nodes, as concurrent manual interaction can cause undesirable results.
