API Reference

Update cluster policies

patch
https://api.cast.ai/cluster-autoscaler/v2/clusters//policies

Updates cluster's V2 policies configuration.

Recent Requests
Log in to see full request history
TimeStatusUser Agent
Retrieving recent requests…
LoadingLoading…
Path Params
string
required

The ID of the cluster.

Body Params
boolean
Defaults to false

Enable/disable all policies (global master switch).

unschedulablePods
object

UnschedulablePodsPolicy defines autoscaler's behavior when unschedulable pods are detected. V2 Simplification:

  • Removed: headroom, headroom_spot (deprecated)
  • Removed: node_constraints, custom_instances_enabled (managed via Node Template API)
  • Removed: disk_gib_to_cpu_ratio (deprecated)
  • Kept: pod_pinner (all fields from V1)
  • Added: partial_template_matching_enabled (moved from top-level in V1)
clusterLimits
object

ClusterLimitsPolicy defines minimum and maximum amount of CPU the cluster can have.

nodeDownscaler
object

NodeDownscalerPolicy defines policies for removing nodes based on the configured conditions. V2 Simplification:

  • Evictor removed (managed via standalone Evictor API)
  • Empty nodes settings are flattened (no nested message)
  • Policy is always present in responses (non-optional)
boolean
Defaults to false

Run the node autoscaler in scoped mode. Only pods with the appropriate toleration will trigger node scale-up decisions, and only nodes provisioned by the autoscaler will be eligible for downscaling.

string

Policy version for optimistic locking. When updating: provide the version you're updating from. When creating: omit or set to 0. In responses: always populated with current version.

Responses

Updated 26 days ago

Language
Credentials
URL
LoadingLoading…
Response
Click Try It! to start a request and see the response here! Or choose an example:
application/json

Updated 26 days ago