Chapter 2. ClusterAutoscaler [autoscaling.openshift.io/v1]

balanceSimilarNodeGroups

boolean

BalanceSimilarNodeGroups enables/disables the --balance-similar-node-groups cluster-autoscaler feature. This feature will automatically identify node groups with the same instance type and the same set of labels and try to keep the respective sizes of those node groups balanced.

balancingIgnoredLabels

array (string)

BalancingIgnoredLabels sets "--balancing-ignore-label <label name>" flag on cluster-autoscaler for each listed label. This option specifies labels that cluster autoscaler should ignore when considering node group similarity. For example, if you have nodes with "topology.ebs.csi.aws.com/zone" label, you can add name of this label here to prevent cluster autoscaler from spliting nodes into different node groups based on its value.

ignoreDaemonsetsUtilization

boolean

Enables/Disables --ignore-daemonsets-utilization CA feature flag. Should CA ignore DaemonSet pods when calculating resource utilization for scaling down. false by default

logVerbosity

integer

Sets the autoscaler log level. Default value is 1, level 4 is recommended for DEBUGGING and level 6 will enable almost everything. This option has priority over log level set by the CLUSTER_AUTOSCALER_VERBOSITY environment variable.

maxNodeProvisionTime

string

Maximum time CA waits for node to be provisioned

maxPodGracePeriod

integer

Gives pods graceful termination time before scaling down

podPriorityThreshold

integer

To allow users to schedule "best-effort" pods, which shouldn’t trigger Cluster Autoscaler actions, but only run when there are spare resources available, More info: https://github.com/kubernetes/autoscaler/blob/master/cluster-autoscaler/FAQ.md#how-does-cluster-autoscaler-work-with-pod-priority-and-preemption

resourceLimits

object

Constraints of autoscaling resources

scaleDown

object

Configuration of scale down operation

skipNodesWithLocalStorage

boolean

Enables/Disables --skip-nodes-with-local-storage CA feature flag. If true cluster autoscaler will never delete nodes with pods with local storage, e.g. EmptyDir or HostPath. true by default at autoscaler

cores

object

Minimum and maximum number of cores in cluster, in the format <min>:<max>. Cluster autoscaler will not scale the cluster beyond these numbers.

gpus

array

Minimum and maximum number of different GPUs in cluster, in the format <gpu_type>:<min>:<max>. Cluster autoscaler will not scale the cluster beyond these numbers. Can be passed multiple times.

gpus[]

object

maxNodesTotal

integer

Maximum number of nodes in all node groups. Cluster autoscaler will not grow the cluster beyond this number.

memory

object

Minimum and maximum number of gigabytes of memory in cluster, in the format <min>:<max>. Cluster autoscaler will not scale the cluster beyond these numbers.

max

integer

min

integer

max

integer

min

integer

type

string

The type of GPU to associate with the minimum and maximum limits. This value is used by the Cluster Autoscaler to identify Nodes that will have GPU capacity by searching for it as a label value on the Node objects. For example, Nodes that carry the label key cluster-api/accelerator with the label value being the same as the Type field will be counted towards the resource limits by the Cluster Autoscaler.

max

integer

min

integer

delayAfterAdd

string

How long after scale up that scale down evaluation resumes

delayAfterDelete

string

How long after node deletion that scale down evaluation resumes, defaults to scan-interval

delayAfterFailure

string

How long after scale down failure that scale down evaluation resumes

enabled

boolean

Should CA scale down the cluster

unneededTime

string

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

utilizationThreshold

string

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

pretty

string

If 'true', then the output is pretty printed.

allowWatchBookmarks

boolean

allowWatchBookmarks requests watch events with type "BOOKMARK". Servers that do not implement bookmarks may ignore this flag and bookmarks are sent at the server’s discretion. Clients should not assume bookmarks are returned at any specific interval, nor may they assume the server will send any BOOKMARK event during a session. If this is not a watch, this field is ignored.

continue

string

The continue option should be set when retrieving more results from the server. Since this value is server defined, clients may only use the continue value from a previous query result with identical query parameters (except for the value of continue) and the server may reject a continue value it does not recognize. If the specified continue value is no longer valid whether due to expiration (generally five to fifteen minutes) or a configuration change on the server, the server will respond with a 410 ResourceExpired error together with a continue token. If the client needs a consistent list, it must restart their list without the continue field. Otherwise, the client may send another list request with the token received with the 410 error, the server will respond with a list starting from the next key, but from the latest snapshot, which is inconsistent from the previous list results - objects that are created, modified, or deleted after the first list request will be included in the response, as long as their keys are after the "next key".

This field is not supported when watch is true. Clients may start a watch from the last resourceVersion value returned by the server and not miss any modifications.

fieldSelector

string

A selector to restrict the list of returned objects by their fields. Defaults to everything.

labelSelector

string

A selector to restrict the list of returned objects by their labels. Defaults to everything.

limit

integer

limit is a maximum number of responses to return for a list call. If more items exist, the server will set the continue field on the list metadata to a value that can be used with the same initial query to retrieve the next set of results. Setting a limit may return fewer than the requested amount of items (up to zero items) in the event all requested objects are filtered out and clients should only use the presence of the continue field to determine whether more results are available. Servers may choose not to support the limit argument and will return all of the available results. If limit is specified and the continue field is empty, clients may assume that no more results are available. This field is not supported if watch is true.

The server guarantees that the objects returned when using continue will be identical to issuing a single list call without a limit - that is, no objects created, modified, or deleted after the first request is issued will be included in any subsequent continued requests. This is sometimes referred to as a consistent snapshot, and ensures that a client that is using limit to receive smaller chunks of a very large result can ensure they see all possible objects. If objects are updated during a chunked list the version of the object that was present at the time the first list result was calculated is returned.

resourceVersion

string

resourceVersion sets a constraint on what resource versions a request may be served from. See https://kubernetes.io/docs/reference/using-api/api-concepts/#resource-versions for details.

Defaults to unset

resourceVersionMatch

string

resourceVersionMatch determines how resourceVersion is applied to list calls. It is highly recommended that resourceVersionMatch be set for list calls where resourceVersion is set See https://kubernetes.io/docs/reference/using-api/api-concepts/#resource-versions for details.

Defaults to unset

sendInitialEvents

boolean

sendInitialEvents=true may be set together with watch=true. In that case, the watch stream will begin with synthetic events to produce the current state of objects in the collection. Once all such events have been sent, a synthetic "Bookmark" event will be sent. The bookmark will report the ResourceVersion (RV) corresponding to the set of objects, and be marked with "k8s.io/initial-events-end": "true" annotation. Afterwards, the watch stream will proceed as usual, sending watch events corresponding to changes (subsequent to the RV) to objects watched.

When sendInitialEvents option is set, we require resourceVersionMatch option to also be set. The semantic of the watch request is as following: - resourceVersionMatch = NotOlderThan is interpreted as "data at least as new as the provided resourceVersion`" and the bookmark event is send when the state is synced to a `resourceVersion at least as fresh as the one provided by the ListOptions. If resourceVersion is unset, this is interpreted as "consistent read" and the bookmark event is send when the state is synced at least to the moment when request started being processed. - resourceVersionMatch set to any other value or unset Invalid error is returned.

Defaults to true if resourceVersion="" or resourceVersion="0" (for backward compatibility reasons) and to false otherwise.

timeoutSeconds

integer

Timeout for the list/watch call. This limits the duration of the call, regardless of any activity or inactivity.

watch

boolean

Watch for changes to the described resources and return them as a stream of add, update, and remove notifications. Specify resourceVersion.

200 - OK

Status schema

401 - Unauthorized

Empty

allowWatchBookmarks

boolean

allowWatchBookmarks requests watch events with type "BOOKMARK". Servers that do not implement bookmarks may ignore this flag and bookmarks are sent at the server’s discretion. Clients should not assume bookmarks are returned at any specific interval, nor may they assume the server will send any BOOKMARK event during a session. If this is not a watch, this field is ignored.

continue

string

The continue option should be set when retrieving more results from the server. Since this value is server defined, clients may only use the continue value from a previous query result with identical query parameters (except for the value of continue) and the server may reject a continue value it does not recognize. If the specified continue value is no longer valid whether due to expiration (generally five to fifteen minutes) or a configuration change on the server, the server will respond with a 410 ResourceExpired error together with a continue token. If the client needs a consistent list, it must restart their list without the continue field. Otherwise, the client may send another list request with the token received with the 410 error, the server will respond with a list starting from the next key, but from the latest snapshot, which is inconsistent from the previous list results - objects that are created, modified, or deleted after the first list request will be included in the response, as long as their keys are after the "next key".

This field is not supported when watch is true. Clients may start a watch from the last resourceVersion value returned by the server and not miss any modifications.

fieldSelector

string

A selector to restrict the list of returned objects by their fields. Defaults to everything.

labelSelector

string

A selector to restrict the list of returned objects by their labels. Defaults to everything.

limit

integer

limit is a maximum number of responses to return for a list call. If more items exist, the server will set the continue field on the list metadata to a value that can be used with the same initial query to retrieve the next set of results. Setting a limit may return fewer than the requested amount of items (up to zero items) in the event all requested objects are filtered out and clients should only use the presence of the continue field to determine whether more results are available. Servers may choose not to support the limit argument and will return all of the available results. If limit is specified and the continue field is empty, clients may assume that no more results are available. This field is not supported if watch is true.

The server guarantees that the objects returned when using continue will be identical to issuing a single list call without a limit - that is, no objects created, modified, or deleted after the first request is issued will be included in any subsequent continued requests. This is sometimes referred to as a consistent snapshot, and ensures that a client that is using limit to receive smaller chunks of a very large result can ensure they see all possible objects. If objects are updated during a chunked list the version of the object that was present at the time the first list result was calculated is returned.

resourceVersion

string

resourceVersion sets a constraint on what resource versions a request may be served from. See https://kubernetes.io/docs/reference/using-api/api-concepts/#resource-versions for details.

Defaults to unset

resourceVersionMatch

string

resourceVersionMatch determines how resourceVersion is applied to list calls. It is highly recommended that resourceVersionMatch be set for list calls where resourceVersion is set See https://kubernetes.io/docs/reference/using-api/api-concepts/#resource-versions for details.

Defaults to unset

sendInitialEvents

boolean

sendInitialEvents=true may be set together with watch=true. In that case, the watch stream will begin with synthetic events to produce the current state of objects in the collection. Once all such events have been sent, a synthetic "Bookmark" event will be sent. The bookmark will report the ResourceVersion (RV) corresponding to the set of objects, and be marked with "k8s.io/initial-events-end": "true" annotation. Afterwards, the watch stream will proceed as usual, sending watch events corresponding to changes (subsequent to the RV) to objects watched.

When sendInitialEvents option is set, we require resourceVersionMatch option to also be set. The semantic of the watch request is as following: - resourceVersionMatch = NotOlderThan is interpreted as "data at least as new as the provided resourceVersion`" and the bookmark event is send when the state is synced to a `resourceVersion at least as fresh as the one provided by the ListOptions. If resourceVersion is unset, this is interpreted as "consistent read" and the bookmark event is send when the state is synced at least to the moment when request started being processed. - resourceVersionMatch set to any other value or unset Invalid error is returned.

Defaults to true if resourceVersion="" or resourceVersion="0" (for backward compatibility reasons) and to false otherwise.

timeoutSeconds

integer

Timeout for the list/watch call. This limits the duration of the call, regardless of any activity or inactivity.

watch

boolean

Watch for changes to the described resources and return them as a stream of add, update, and remove notifications. Specify resourceVersion.

200 - OK

ClusterAutoscalerList schema

401 - Unauthorized

Empty

dryRun

string

When present, indicates that modifications should not be persisted. An invalid or unrecognized dryRun directive will result in an error response and no further processing of the request. Valid values are: - All: all dry run stages will be processed

fieldManager

string

fieldManager is a name associated with the actor or entity that is making these changes. The value must be less than or 128 characters long, and only contain printable characters, as defined by https://golang.org/pkg/unicode/#IsPrint.

fieldValidation

string

fieldValidation instructs the server on how to handle objects in the request (POST/PUT/PATCH) containing unknown or duplicate fields. Valid values are: - Ignore: This will ignore any unknown fields that are silently dropped from the object, and will ignore all but the last duplicate field that the decoder encounters. This is the default behavior prior to v1.23. - Warn: This will send a warning via the standard warning response header for each unknown field that is dropped from the object, and for each duplicate field that is encountered. The request will still succeed if there are no other errors, and will only persist the last of any duplicate fields. This is the default in v1.23+ - Strict: This will fail the request with a BadRequest error if any unknown fields would be dropped from the object, or if any duplicate fields are present. The error returned from the server will contain all unknown and duplicate fields encountered.

body

ClusterAutoscaler schema

200 - OK

ClusterAutoscaler schema

201 - Created

ClusterAutoscaler schema

202 - Accepted

ClusterAutoscaler schema

401 - Unauthorized

Empty

name

string

name of the ClusterAutoscaler

pretty

string

If 'true', then the output is pretty printed.

dryRun

string

When present, indicates that modifications should not be persisted. An invalid or unrecognized dryRun directive will result in an error response and no further processing of the request. Valid values are: - All: all dry run stages will be processed

gracePeriodSeconds

integer

The duration in seconds before the object should be deleted. Value must be non-negative integer. The value zero indicates delete immediately. If this value is nil, the default grace period for the specified type will be used. Defaults to a per object value if not specified. zero means delete immediately.

orphanDependents

boolean

Deprecated: please use the PropagationPolicy, this field will be deprecated in 1.7. Should the dependent objects be orphaned. If true/false, the "orphan" finalizer will be added to/removed from the object’s finalizers list. Either this field or PropagationPolicy may be set, but not both.

propagationPolicy

string

Whether and how garbage collection will be performed. Either this field or OrphanDependents may be set, but not both. The default policy is decided by the existing finalizer set in the metadata.finalizers and the resource-specific default policy. Acceptable values are: 'Orphan' - orphan the dependents; 'Background' - allow the garbage collector to delete the dependents in the background; 'Foreground' - a cascading policy that deletes all dependents in the foreground.

body

DeleteOptions schema

200 - OK

Status schema

202 - Accepted

Status schema

401 - Unauthorized

Empty

resourceVersion

string

resourceVersion sets a constraint on what resource versions a request may be served from. See https://kubernetes.io/docs/reference/using-api/api-concepts/#resource-versions for details.

Defaults to unset

200 - OK

ClusterAutoscaler schema

401 - Unauthorized

Empty

dryRun

string

When present, indicates that modifications should not be persisted. An invalid or unrecognized dryRun directive will result in an error response and no further processing of the request. Valid values are: - All: all dry run stages will be processed

fieldManager

string

fieldManager is a name associated with the actor or entity that is making these changes. The value must be less than or 128 characters long, and only contain printable characters, as defined by https://golang.org/pkg/unicode/#IsPrint. This field is required for apply requests (application/apply-patch) but optional for non-apply patch types (JsonPatch, MergePatch, StrategicMergePatch).

fieldValidation

string

fieldValidation instructs the server on how to handle objects in the request (POST/PUT/PATCH) containing unknown or duplicate fields. Valid values are: - Ignore: This will ignore any unknown fields that are silently dropped from the object, and will ignore all but the last duplicate field that the decoder encounters. This is the default behavior prior to v1.23. - Warn: This will send a warning via the standard warning response header for each unknown field that is dropped from the object, and for each duplicate field that is encountered. The request will still succeed if there are no other errors, and will only persist the last of any duplicate fields. This is the default in v1.23+ - Strict: This will fail the request with a BadRequest error if any unknown fields would be dropped from the object, or if any duplicate fields are present. The error returned from the server will contain all unknown and duplicate fields encountered.

force

boolean

Force is going to "force" Apply requests. It means user will re-acquire conflicting fields owned by other people. Force flag must be unset for non-apply patch requests.

body

Patch schema

200 - OK

ClusterAutoscaler schema

401 - Unauthorized

Empty

dryRun

string

When present, indicates that modifications should not be persisted. An invalid or unrecognized dryRun directive will result in an error response and no further processing of the request. Valid values are: - All: all dry run stages will be processed

fieldManager

string

fieldManager is a name associated with the actor or entity that is making these changes. The value must be less than or 128 characters long, and only contain printable characters, as defined by https://golang.org/pkg/unicode/#IsPrint.

fieldValidation

string

fieldValidation instructs the server on how to handle objects in the request (POST/PUT/PATCH) containing unknown or duplicate fields. Valid values are: - Ignore: This will ignore any unknown fields that are silently dropped from the object, and will ignore all but the last duplicate field that the decoder encounters. This is the default behavior prior to v1.23. - Warn: This will send a warning via the standard warning response header for each unknown field that is dropped from the object, and for each duplicate field that is encountered. The request will still succeed if there are no other errors, and will only persist the last of any duplicate fields. This is the default in v1.23+ - Strict: This will fail the request with a BadRequest error if any unknown fields would be dropped from the object, or if any duplicate fields are present. The error returned from the server will contain all unknown and duplicate fields encountered.

body

ClusterAutoscaler schema

200 - OK

ClusterAutoscaler schema

201 - Created

ClusterAutoscaler schema

401 - Unauthorized

Empty

name

string

name of the ClusterAutoscaler

pretty

string

If 'true', then the output is pretty printed.

resourceVersion

string

resourceVersion sets a constraint on what resource versions a request may be served from. See https://kubernetes.io/docs/reference/using-api/api-concepts/#resource-versions for details.

Defaults to unset

200 - OK

ClusterAutoscaler schema

401 - Unauthorized

Empty

dryRun

string

When present, indicates that modifications should not be persisted. An invalid or unrecognized dryRun directive will result in an error response and no further processing of the request. Valid values are: - All: all dry run stages will be processed

fieldManager

string

fieldManager is a name associated with the actor or entity that is making these changes. The value must be less than or 128 characters long, and only contain printable characters, as defined by https://golang.org/pkg/unicode/#IsPrint. This field is required for apply requests (application/apply-patch) but optional for non-apply patch types (JsonPatch, MergePatch, StrategicMergePatch).

fieldValidation

string

fieldValidation instructs the server on how to handle objects in the request (POST/PUT/PATCH) containing unknown or duplicate fields. Valid values are: - Ignore: This will ignore any unknown fields that are silently dropped from the object, and will ignore all but the last duplicate field that the decoder encounters. This is the default behavior prior to v1.23. - Warn: This will send a warning via the standard warning response header for each unknown field that is dropped from the object, and for each duplicate field that is encountered. The request will still succeed if there are no other errors, and will only persist the last of any duplicate fields. This is the default in v1.23+ - Strict: This will fail the request with a BadRequest error if any unknown fields would be dropped from the object, or if any duplicate fields are present. The error returned from the server will contain all unknown and duplicate fields encountered.

force

boolean

Force is going to "force" Apply requests. It means user will re-acquire conflicting fields owned by other people. Force flag must be unset for non-apply patch requests.

body

Patch schema

200 - OK

ClusterAutoscaler schema

401 - Unauthorized

Empty

dryRun

string

When present, indicates that modifications should not be persisted. An invalid or unrecognized dryRun directive will result in an error response and no further processing of the request. Valid values are: - All: all dry run stages will be processed

fieldManager

string

fieldManager is a name associated with the actor or entity that is making these changes. The value must be less than or 128 characters long, and only contain printable characters, as defined by https://golang.org/pkg/unicode/#IsPrint.

fieldValidation

string

fieldValidation instructs the server on how to handle objects in the request (POST/PUT/PATCH) containing unknown or duplicate fields. Valid values are: - Ignore: This will ignore any unknown fields that are silently dropped from the object, and will ignore all but the last duplicate field that the decoder encounters. This is the default behavior prior to v1.23. - Warn: This will send a warning via the standard warning response header for each unknown field that is dropped from the object, and for each duplicate field that is encountered. The request will still succeed if there are no other errors, and will only persist the last of any duplicate fields. This is the default in v1.23+ - Strict: This will fail the request with a BadRequest error if any unknown fields would be dropped from the object, or if any duplicate fields are present. The error returned from the server will contain all unknown and duplicate fields encountered.

body

ClusterAutoscaler schema

200 - OK

ClusterAutoscaler schema

201 - Created

ClusterAutoscaler schema

401 - Unauthorized

Empty