OpenShift docs are moving and will soon only be available at docs.redhat.com, the home of all Red Hat product documentation. Explore the new docs experience today.
  1. Documentation
    2. history bug_report picture_as_pdf
×

Cluster autoscaling

edit

Applying autoscaling to Red Hat OpenShift Service on AWS clusters involves configuring a cluster autoscaler and then configuring a machine autoscaler for at least one machine pool in your cluster.

You can configure the cluster autoscaler only in clusters where the machine API is operational.

Only one cluster autoscaler can be created per cluster.
edit

About the cluster autoscaler

The cluster autoscaler adjusts the size of an Red Hat OpenShift Service on AWS cluster to meet its current deployment needs. It uses declarative, Kubernetes-style arguments to provide infrastructure management that does not rely on objects of a specific cloud provider. The cluster autoscaler has a cluster scope, and is not associated with a particular namespace.

The cluster autoscaler increases the size of the cluster when there are pods that fail to schedule on any of the current worker nodes due to insufficient resources or when another node is necessary to meet deployment needs. The cluster autoscaler does not increase the cluster resources beyond the limits that you specify.

The cluster autoscaler computes the total memory and CPU on all nodes the cluster, even though it does not manage the control plane nodes. These values are not single-machine oriented. They are an aggregation of all the resources in the entire cluster. For example, if you set the maximum memory resource limit, the cluster autoscaler includes all the nodes in the cluster when calculating the current memory usage. That calculation is then used to determine if the cluster autoscaler has the capacity to add more worker resources.

Ensure that the maxNodesTotal value in the ClusterAutoscaler resource definition that you create is large enough to account for the total possible number of machines in your cluster. This value must encompass the number of control plane machines and the possible number of compute machines that you might scale to.

Automatic node removal

Every 10 seconds, the cluster autoscaler checks which nodes are unnecessary in the cluster and removes them. The cluster autoscaler considers a node for removal if the following conditions apply:

  • The node utilization is less than the node utilization level threshold for the cluster. The node utilization level is the sum of the requested resources divided by the allocated resources for the node. If you do not specify a value in the ClusterAutoscaler custom resource, the cluster autoscaler uses a default value of 0.5, which corresponds to 50% utilization.

  • The cluster autoscaler can move all pods running on the node to the other nodes. The Kubernetes scheduler is responsible for scheduling pods on the nodes.

  • The cluster autoscaler does not have scale down disabled annotation.

If the following types of pods are present on a node, the cluster autoscaler will not remove the node:

  • Pods with restrictive pod disruption budgets (PDBs).

  • Kube-system pods that do not run on the node by default.

  • Kube-system pods that do not have a PDB or have a PDB that is too restrictive.

  • Pods that are not backed by a controller object such as a deployment, replica set, or stateful set.

  • Pods with local storage.

  • Pods that cannot be moved elsewhere because of a lack of resources, incompatible node selectors or affinity, matching anti-affinity, and so on.

  • Unless they also have a "cluster-autoscaler.kubernetes.io/safe-to-evict": "true" annotation, pods that have a "cluster-autoscaler.kubernetes.io/safe-to-evict": "false" annotation.

For example, you set the maximum CPU limit to 64 cores and configure the cluster autoscaler to only create machines that have 8 cores each. If your cluster starts with 30 cores, the cluster autoscaler can add up to 4 more nodes with 32 cores, for a total of 62.

Limitations

If you configure the cluster autoscaler, additional usage restrictions apply:

  • Do not modify the nodes that are in autoscaled node groups directly. All nodes within the same node group have the same capacity and labels and run the same system pods.

  • Specify requests for your pods.

  • If you have to prevent pods from being deleted too quickly, configure appropriate PDBs.

  • Confirm that your cloud provider quota is large enough to support the maximum node pools that you configure.

  • Do not run additional node group autoscalers, especially the ones offered by your cloud provider.

The cluster autoscaler only adds nodes in autoscaled node groups if doing so would result in a schedulable pod. If the available node types cannot meet the requirements for a pod request, or if the node groups that could meet these requirements are at their maximum size, the cluster autoscaler cannot scale up.

Interaction with other scheduling features

The horizontal pod autoscaler (HPA) and the cluster autoscaler modify cluster resources in different ways. The HPA changes the deployment’s or replica set’s number of replicas based on the current CPU load. If the load increases, the HPA creates new replicas, regardless of the amount of resources available to the cluster. If there are not enough resources, the cluster autoscaler adds resources so that the HPA-created pods can run. If the load decreases, the HPA stops some replicas. If this action causes some nodes to be underutilized or completely empty, the cluster autoscaler deletes the unnecessary nodes.

The cluster autoscaler takes pod priorities into account. The Pod Priority and Preemption feature enables scheduling pods based on priorities if the cluster does not have enough resources, but the cluster autoscaler ensures that the cluster has resources to run all pods. To honor the intention of both features, the cluster autoscaler includes a priority cutoff function. You can use this cutoff to schedule "best-effort" pods, which do not cause the cluster autoscaler to increase resources but instead run only when spare resources are available.

Pods with priority lower than the cutoff value do not cause the cluster to scale up or prevent the cluster from scaling down. No new nodes are added to run the pods, and nodes running these pods might be deleted to free resources.

edit

Enable autoscaling during cluster creation with OpenShift Cluster Manager

You can use OpenShift Cluster Manager to autoscale during cluster creation.

Procedure

  1. During cluster creation, check the Enable autoscaling box. The Edit cluster autoscaling settings button becomes selectable.

    1. You can also choose the minimum or maximum amount of nodes to autoscale.

  2. Click Edit cluster autoscaling settings.

  3. Edit any settings you want and then click Close.

edit

Enable autoscaling after cluster creation with OpenShift Cluster Manager

You can use OpenShift Cluster Manager to autoscale after cluster creation.

Procedure

  1. In OpenShift Cluster Manager, click the name of the cluster you want to autoscale. The Overview page for the cluster has a Autoscaling item that indicates if it is enabled or disabled.

  2. Click the Machine Pools tab.

  3. Click the Edit cluster autoscaling button. The Edit cluster autoscaling settings window is shown.

  4. Click the Autoscale cluster toggle at the top of the window. All the settings are now editable.

  5. Edit any settings you want and then click Save.

  6. Click the x at the top right of the screen to close the settings window.

To revert all autoscaling settings to the defaults if they have been changed, click the Revert all to defaults button.

Cluster autoscaling settings using OpenShift Cluster Manager

The tables explain all the configurable UI settings when using cluster autoscaling with OpenShift Cluster Manager.

edit

General settings

Table 1. Configurable general settings for cluster autoscaling when using the OpenShift Cluster Manager
Setting Description Type or Range Default

log-verbosity

Sets the autoscaler log level. The default value is 1. Level 4 is recommended for debugging. Level 6 enables almost everything.

integer

1

skip-nodes-with-local-storage

If true, the cluster autoscaler never deletes nodes with pods with local storage, e.g. EmptyDir or HostPath.

boolean

true

max-pod-grace-period

Gives pods graceful termination time in seconds before scaling down.

integer

600

max-node-provision-time

Maximum time the cluster autoscaler waits for nodes to be provisioned.

string

15m

pod-priority-threshold

Allows users to schedule "best-effort" pods, which are not expected to trigger cluster autoscaler actions. These pods only run when spare resources are available.

integer

-10

ignore-daemonsets-utilization

Determines whether the cluster autoscaler ignores daemon set pods when calculating resource utilization for scaling down.

boolean

false

balance-similar-node-groups

If true, this setting automatically identifies node groups with the same instance type and the same set of labels and tries to keep the respective sizes of those node groups balanced.

boolean

false

balancing-ignored-labels

This option specifies labels that the cluster autoscaler should ignore when considering node group similarity. This option cannot contain spaces.

array (string)

Format should be a comma-separated list of labels.

Resource limits

Table 2. Configurable resource limit settings for cluster autoscaling when using the OpenShift Cluster Manager
Setting Description Type or Range Default

cores-total-min

Minimum number of cores in cluster. The cluster autoscaler does not scale the cluster less than this number.

object

0

cores-total-max

Maximum number of cores in cluster. The cluster autoscaler does not scale the cluster greater than this number.

object

180 * 64 (11520)

memory-total-min

Minimum number of gigabytes of memory in cluster. The cluster autoscaler does not scale the cluster less than this number.

object

0

memory-total-max

Maximum number of gigabytes of memory in cluster. The cluster autoscaler does not scale the cluster greater than this number.

object

180 * 64 * 20 (230400)

max-nodes-total

Maximum number of nodes in all node groups. Includes all nodes, not just automatically scaled nodes. The cluster autoscaler does not grow the cluster greater than this number.

integer

180

GPUs

Minimum and maximum number of different GPUs in cluster. The cluster autoscaler does not scale the cluster less than or greater than these numbers.

array

Format should be a comma-separated list of "<gpu_type>:<min>:<max>".

Scale down configuration

Table 3. Configurable scale down settings for cluster autoscaling when using the OpenShift Cluster Manager
Setting Description Type or Range Default

scale-down-enabled

Should the cluster autoscaler scale down the cluster.

boolean

true

scale-down-utilization-threshold

Node utilization level, defined as the sum of the requested resources divided by capacity, below which a node can be considered for scale down.

float

0.5

scale-down-unneeded-time

How long a node should be unneeded before it is eligible for scale down.

string

10m

scale-down-delay-after-add

How long after scale up that scale-down evaluation resumes.

string

10m

scale-down-delay-after-delete

How long after node deletion that scale-down evaluation resumes.

string

0s

scale-down-delay-after-failure

How long after scale down failure that scale-down evaluation resumes.

string

3m
edit

Enable autoscaling during cluster creation by using the interactive mode with the ROSA CLI

You can use the interactive mode of your terminal, if available, to set cluster-wide autoscaling behavior during cluster creation.

Interactive mode provides more information about available configurable parameters. Interactive mode also does basic checks and preflight validations, meaning that if a provided value is invalid, the terminal outputs a prompt for a valid input.

Procedure

  • During cluster creation, use the --enable-autoscaling and --interactive parameters to enable cluster autoscaling:

    Example
    $ rosa create cluster --cluster-name <cluster_name> --enable-autoscaling --interactive

If your cluster name is longer than 15 characters, it will contain an autogenerated domain prefix as a sub-domain for your provisioned cluster on *.openshiftapps.com.

To customize the subdomain, use the --domain-prefix flag. The domain prefix cannot be longer than 15 characters, must be unique, and cannot be changed after cluster creation.

When the following prompt appears, enter y to go through all available autoscaling options.

Example interactive prompt
? Configure cluster-autoscaler (optional): [? for help] (y/N) y <enter>
edit

Enable autoscaling after cluster creation by using the interactive mode with the ROSA CLI

You can use the interactive mode of your terminal, if available, to set cluster-wide autoscaling behavior after cluster creation.

Procedure

  • After you have created a cluster, type the following command:

    Example
    $ rosa create autoscaler --cluster=<mycluster> --interactive

    You can then set all available autoscaling parameters.

edit

Enable autoscaling during cluster creation with the ROSA CLI

You can use the ROSA CLI (rosa) to set cluster-wide autoscaling behavior during cluster creation. You can enable the autoscaler on the entire machine or just a cluster.

Procedure

  • During cluster creation, type --enable autoscaling after the cluster name to enable machine autoscaling:

If your cluster name is longer than 15 characters, it will contain an autogenerated domain prefix as a sub-domain for your provisioned cluster on *.openshiftapps.com.

To customize the subdomain, use the --domain-prefix flag. The domain prefix cannot be longer than 15 characters, must be unique, and cannot be changed after cluster creation.
Example
$ rosa create cluster --cluster-name <cluster_name> --enable-autoscaling

Set at least one parameter to enable cluster autoscaling by running the following command:

Example
$ rosa create cluster --cluster-name <cluster_name> --enable-autoscaling <parameter>
edit

Enable autoscaling after cluster creation with the ROSA CLI

The following procedure is only supported for ROSA Classic clusters. To enable autoscaling after cluster creation for ROSA with HCP clusters, see Enabling autoscaling nodes on a cluster.

You can use the ROSA CLI (rosa) to set cluster-wide autoscaling after cluster creation.

Procedure

  • After you have created a cluster, create the autoscaler:

    Example
    $ rosa create autoscaler --cluster=<mycluster>

    1. You can also create the autoscaler with specific parameters using the following command:

      Example
      $ rosa create autoscaler --cluster=<mycluster> <parameter>
Next steps
edit

Edit autoscaling after cluster creation with the ROSA CLI

You can edit any specific parameters of the cluster autoscaler after creating the autoscaler.

  • To edit the cluster autoscaler, run the following command:

    Example
    $ rosa edit autoscaler --cluster=<mycluster>

    1. To edit a specific parameter, run the following command:

      Example
      $ rosa edit autoscaler --cluster=<mycluster> <parameter>
edit

Delete autoscaling using the ROSA CLI

You can delete the cluster autoscaler if you no longer want to use it.

  • To delete the cluster autoscaler, run the following command:

    Example
    $ rosa delete autoscaler --cluster=<mycluster>
edit

Cluster autoscaling parameters using the ROSA CLI

You can add the following parameters to the cluster creation command to configure autoscaler parameters when using the ROSA CLI (rosa).

Table 4. Configurable autoscaler parameters available with the ROSA CLI (rosa)
Setting Description Type or Range Example/Instruction

--autoscaler-balance-similar-node-groups

Identify node groups with the same instance type and label set, and try to balance respective sizes of those node groups.

boolean

Add it to set to true, omit the option to set to false.

--autoscaler-skip-nodes-with-local-storage

If set, the cluster autoscaler does not delete nodes with pods that have local storage, for example, EmptyDir or HostPath.

boolean

Add it to set to true, omit the option to set to false.

--autoscaler-log-verbosity int

Autoscaler log level. Replace int in the command with the number you want to use.

integer

--autoscaler-log-verbosity 4

--autoscaler-max-pod-grace-period int

Gives pods graceful termination time before scaling down, measured in seconds. Replace int in the command with the number of seconds you want to use.

integer

--autoscaler-max-pod-grace-period 0

--autoscaler-pod-priority-threshold int

The priority that a pod must exceed to cause the cluster autoscaler to deploy additional nodes. Replace int in the command with the number you want to use, can be negative.

integer

--autoscaler-pod-priority-threshold -10

--autoscaler-gpu-limit stringArray

Minimum and maximum number of different GPUs in cluster. Cluster autoscaler does not scale the cluster less than or greater than these numbers. The format must be a comma-separated list of "<gpu_type>,<min>,<max>".

array

--autoscaler-gpu-limit nvidia.com/gpu,0,10 --autoscaler-gpu-limit amd.com/gpu,1,5

--autoscaler-ignore-daemonsets-utilization

If set, the cluster-autoscaler ignores daemon set pods when calculating resource utilization for scaling down.

boolean

Add it to set to true, omit the option to set to false.

--autoscaler-max-node-provision-time string

Maximum time that the cluster autoscaler waits for a node to be provisioned. Replace string in the command with an integer and time unit (ns,us,µs,ms,s,m,h).

string

--autoscaler-max-node-provision-time 35m

--autoscaler-balancing-ignored-labels strings

A comma-separated list of label keys that the cluster autoscaler should ignore when comparing node groups for similarity. Replace strings in the command with the relevant labels..

string

--autoscaler-balancing-ignored-labels topology.ebs.csi.aws.com/zone,alpha.eksctl.io/instance-id

--autoscaler-max-nodes-total int

Maximum amount of nodes in the cluster, including the autoscaled nodes. Replace int in the command with the number you want to use.

integer

--autoscaler-max-nodes-total 180

--autoscaler-min-cores int

Minimum number of cores to deploy in the cluster. Replace int in the command with the number you want to use.

integer

--autoscaler-min-cores 0

--autoscaler-max-cores int

Maximum number of cores to deploy in the cluster. Replace int in the command with the number you want to use.

integer

--autoscaler-max-cores 100

--autoscaler-min-memory int

Minimum amount of memory, in GiB, in the cluster. Replace int in the command with the number you want to use.

integer

--autoscaler-min-memory 0

--autoscaler-max-memory int

Maximum amount of memory, in GiB, in the cluster. Replace int in the command with the number you want to use.

integer

--autoscaler-max-memory 4096

--autoscaler-scale-down-enabled

If set, the cluster autoscaler should scale down the cluster.

boolean

Add it to set to true, omit the option to set to false.

--autoscaler-scale-down-unneeded-time string

How long a node should be unneeded before it is eligible for scale down. Replace string in the command with an integer and time unit (ns,us,µs,ms,s,m,h).

string

--autoscaler-scale-down-unneeded-time 1h

--autoscaler-scale-down-utilization-threshold float

Node utilization level, defined as sum of requested resources divided by capacity, below which a node can be considered for scale down. Value must be between 0 and 1.

float

--autoscaler-scale-down-utilization-threshold 0.5

--autoscaler-scale-down-delay-after-add string

How long after scale up that scale down evaluation resumes. Replace string in the command with an integer and time unit (ns,us,µs,ms,s,m,h).

string

--autoscaler-scale-down-delay-after-add 1h

--autoscaler-scale-down-delay-after-delete string

How long after node deletion that scale down evaluation resumes. Replace string in the command with an integer and time unit (ns,us,µs,ms,s,m,h).

string

--autoscaler-scale-down-delay-after-delete 1h

--autoscaler-scale-down-delay-after-failure string

How long after scale down failure that scale down evaluation resumes. Replace string in the command with an integer and time unit (ns,us,µs,ms,s,m,h).

string

--autoscaler-scale-down-delay-after-failure 1h