You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
The cluster autoscaler adjusts the size of an {product-title} 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.
16
+
ifdef::openshift-rosa-hcp[]
17
+
In {product-title}, the Cluster Autoscaler is fully managed, which means it is hosted along with the control plane.
18
+
endif::openshift-rosa-hcp[]
19
+
14
20
15
21
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.
16
22
23
+
ifdef::openshift-rosa-hcp[]
24
+
The cluster autoscaler computes the total memory, CPU, and GPU only on the nodes that belong to autoscaling machine pools. All of the machine pool nodes that are not autoscaling are excluded from this aggregation. For example, if you set the `maxNodesTotal` to `50` on a {product-title} cluster with three machine pools in which a single machine pool is not autoscaling, the cluster autoscaler restricts the total nodes to `50` in only those two machine pools that are autoscaling. The single manually scaling machine pool can have additional nodes, making the overall cluster nodes total more than `50`.
25
+
endif::openshift-rosa-hcp[]
26
+
27
+
ifndef::openshift-rosa-hcp[]
17
28
The cluster autoscaler computes the total
18
29
ifndef::openshift-dedicated,openshift-rosa[]
19
30
memory, CPU, and GPU
20
-
endif::[]
31
+
endif::openshift-dedicated,openshift-rosa[]
21
32
ifdef::openshift-dedicated,openshift-rosa[]
22
33
memory and CPU
23
-
endif::[]
34
+
endif::openshift-dedicated,openshift-rosa[]
24
35
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.
36
+
endif::openshift-rosa-hcp[]
25
37
38
+
ifndef::openshift-rosa-hcp[]
26
39
[IMPORTANT]
27
40
====
28
41
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.
= Enable autoscaling after cluster creation with the ROSA CLI
8
8
9
+
////
9
10
[IMPORTANT]
10
11
====
11
12
The following procedure is only supported for ROSA Classic clusters. To enable autoscaling after cluster creation for ROSA with HCP clusters, see link:https://docs.openshift.com/rosa/rosa_cluster_admin/rosa_nodes/rosa-nodes-about-autoscaling-nodes.html#nodes-enabling-autoscaling-nodes[Enabling autoscaling nodes on a cluster].
12
13
====
14
+
////
13
15
14
16
You can use the ROSA CLI (`rosa`) to set cluster-wide autoscaling after cluster creation.
15
17
16
18
.Procedure
17
-
18
-
- After you have created a cluster, create the autoscaler:
19
+
20
+
* After you have created a cluster, create the autoscaler:
19
21
+
20
22
.Example
21
23
[source,terminal]
22
24
----
23
25
$ rosa create autoscaler --cluster=<mycluster>
24
26
----
25
27
+
26
-
.. You can also create the autoscaler with specific parameters using the following command:
28
+
** You can also create the autoscaler with specific parameters using the following command:
Copy file name to clipboardExpand all lines: modules/rosa-cluster-autoscaler-cli-settings.adoc
+16Lines changed: 16 additions & 0 deletions
Display the source diff
Display the rich diff
Original file line number
Diff line number
Diff line change
@@ -17,6 +17,8 @@ You can add the following parameters to the cluster creation command to configur
17
17
|Type or Range
18
18
|Example/Instruction
19
19
20
+
ifdef::openshift-rosa[]
21
+
//Classic only
20
22
|`--autoscaler-balance-similar-node-groups`
21
23
|Identify node groups with the same instance type and label set, and try to balance respective sizes of those node groups.
22
24
|`boolean`
@@ -31,17 +33,22 @@ You can add the following parameters to the cluster creation command to configur
31
33
|Autoscaler log level. Replace _int_ in the command with the number you want to use.
32
34
|`integer`
33
35
|`--autoscaler-log-verbosity 4`
36
+
endif::openshift-rosa[]
34
37
38
+
//Both Classic and HCP
35
39
|`--autoscaler-max-pod-grace-period _int_`
36
40
|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.
37
41
|`integer`
38
42
|`--autoscaler-max-pod-grace-period 0`
39
43
44
+
//Both Classic and HCP
40
45
|`--autoscaler-pod-priority-threshold _int_`
41
46
|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.
42
47
|`integer`
43
48
|`--autoscaler-pod-priority-threshold -10`
44
49
50
+
ifdef::openshift-rosa[]
51
+
//Classic only
45
52
|`--autoscaler-gpu-limit _stringArray_`
46
53
|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>".
47
54
|`array`
@@ -51,22 +58,30 @@ You can add the following parameters to the cluster creation command to configur
51
58
|If set, the cluster-autoscaler ignores daemon set pods when calculating resource utilization for scaling down.
52
59
|`boolean`
53
60
|Add it to set to true, omit the option to set to false.
61
+
endif::openshift-rosa[]
54
62
63
+
//Both Classic and HCP
55
64
|`--autoscaler-max-node-provision-time _string_`
56
65
|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).
|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..
|Maximum amount of nodes in the cluster, including the autoscaled nodes. Replace _int_ in the command with the number you want to use.
67
80
|`integer`
68
81
|`--autoscaler-max-nodes-total 180`
69
82
83
+
ifdef::openshift-rosa[]
84
+
//Classic only
70
85
|`--autoscaler-min-cores _int_`
71
86
|Minimum number of cores to deploy in the cluster. Replace _int_ in the command with the number you want to use.
72
87
|`integer`
@@ -116,5 +131,6 @@ You can add the following parameters to the cluster creation command to configur
116
131
|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).
0 commit comments