Chapter 4. Controlling pod placement onto nodes (scheduling)

4.1. Controlling pod placement using the scheduler

Pod scheduling is an internal process that determines placement of new pods onto nodes within the cluster.

The scheduler code has a clean separation that watches new pods as they get created and identifies the most suitable node to host them. It then creates bindings (pod to node bindings) for the pods using the master API.

Default pod scheduling
OpenShift Container Platform comes with a default scheduler that serves the needs of most users. The default scheduler uses both inherent and customization tools to determine the best fit for a pod.
Advanced pod scheduling

In situations where you might want more control over where new pods are placed, the OpenShift Container Platform advanced scheduling features allow you to configure a pod so that the pod is required or has a preference to run on a particular node or alongside a specific pod.

You can control pod placement by using the following scheduling features:

4.1.1. About the default scheduler

The default OpenShift Container Platform pod scheduler is responsible for determining the placement of new pods onto nodes within the cluster. It reads data from the pod and finds a node that is a good fit based on configured profiles. It is completely independent and exists as a standalone solution. It does not modify the pod; it creates a binding for the pod that ties the pod to the particular node.

4.1.1.1. Understanding default scheduling

The existing generic scheduler is the default platform-provided scheduler engine that selects a node to host the pod in a three-step operation:

Filters the nodes
The available nodes are filtered based on the constraints or requirements specified. This is done by running each node through the list of filter functions called predicates, or filters.
Prioritizes the filtered list of nodes
This is achieved by passing each node through a series of priority, or scoring, functions that assign it a score between 0 - 10, with 0 indicating a bad fit and 10 indicating a good fit to host the pod. The scheduler configuration can also take in a simple weight (positive numeric value) for each scoring function. The node score provided by each scoring function is multiplied by the weight (default weight for most scores is 1) and then combined by adding the scores for each node provided by all the scores. This weight attribute can be used by administrators to give higher importance to some scores.
Selects the best fit node
The nodes are sorted based on their scores and the node with the highest score is selected to host the pod. If multiple nodes have the same high score, then one of them is selected at random.

4.1.2. Scheduler use cases

One of the important use cases for scheduling within OpenShift Container Platform is to support flexible affinity and anti-affinity policies.

4.1.2.1. Infrastructure topological levels

Administrators can define multiple topological levels for their infrastructure (nodes) by specifying labels on nodes. For example: region=r1, zone=z1, rack=s1.

These label names have no particular meaning and administrators are free to name their infrastructure levels anything, such as city/building/room. Also, administrators can define any number of levels for their infrastructure topology, with three levels usually being adequate (such as: regions → zones → racks). Administrators can specify affinity and anti-affinity rules at each of these levels in any combination.

4.1.2.2. Affinity

Administrators should be able to configure the scheduler to specify affinity at any topological level, or even at multiple levels. Affinity at a particular level indicates that all pods that belong to the same service are scheduled onto nodes that belong to the same level. This handles any latency requirements of applications by allowing administrators to ensure that peer pods do not end up being too geographically separated. If no node is available within the same affinity group to host the pod, then the pod is not scheduled.

If you need greater control over where the pods are scheduled, see Controlling pod placement on nodes using node affinity rules and Placing pods relative to other pods using affinity and anti-affinity rules.

These advanced scheduling features allow administrators to specify which node a pod can be scheduled on and to force or reject scheduling relative to other pods.

4.1.2.3. Anti-affinity

Administrators should be able to configure the scheduler to specify anti-affinity at any topological level, or even at multiple levels. Anti-affinity (or 'spread') at a particular level indicates that all pods that belong to the same service are spread across nodes that belong to that level. This ensures that the application is well spread for high availability purposes. The scheduler tries to balance the service pods across all applicable nodes as evenly as possible.

If you need greater control over where the pods are scheduled, see Controlling pod placement on nodes using node affinity rules and Placing pods relative to other pods using affinity and anti-affinity rules.

These advanced scheduling features allow administrators to specify which node a pod can be scheduled on and to force or reject scheduling relative to other pods.

4.2. Scheduling pods using a scheduler profile

You can configure OpenShift Container Platform to use a scheduling profile to schedule pods onto nodes within the cluster.

4.2.1. About scheduler profiles

You can specify a scheduler profile to control how pods are scheduled onto nodes.

The following scheduler profiles are available:

LowNodeUtilization
This profile attempts to spread pods evenly across nodes to get low resource usage per node. This profile provides the default scheduler behavior.
HighNodeUtilization
This profile attempts to place as many pods as possible on to as few nodes as possible. This minimizes node count and has high resource usage per node.
NoScoring
This is a low-latency profile that strives for the quickest scheduling cycle by disabling all score plugins. This might sacrifice better scheduling decisions for faster ones.

4.2.2. Configuring a scheduler profile

You can configure the scheduler to use a scheduler profile.

Prerequisites

  • Access to the cluster as a user with the cluster-admin role.

Procedure

  1. Edit the Scheduler object: 

    $ oc edit scheduler cluster

  2. Specify the profile to use in the spec.profile field: 

    apiVersion: config.openshift.io/v1
kind: Scheduler
metadata:
  name: cluster
#...
spec:
  mastersSchedulable: false
  profile: HighNodeUtilization 1
#...
    1
    Set to LowNodeUtilization, HighNodeUtilization, or NoScoring.
  3. Save the file to apply the changes.

4.3. Placing pods relative to other pods using affinity and anti-affinity rules

Affinity is a property of pods that controls the nodes on which they prefer to be scheduled. Anti-affinity is a property of pods that prevents a pod from being scheduled on a node.

In OpenShift Container Platform, pod affinity and pod anti-affinity allow you to constrain which nodes your pod is eligible to be scheduled on based on the key-value labels on other pods.

4.3.1. Understanding pod affinity

Pod affinity and pod anti-affinity allow you to constrain which nodes your pod is eligible to be scheduled on based on the key/value labels on other pods.

  • Pod affinity can tell the scheduler to locate a new pod on the same node as other pods if the label selector on the new pod matches the label on the current pod.
  • Pod anti-affinity can prevent the scheduler from locating a new pod on the same node as pods with the same labels if the label selector on the new pod matches the label on the current pod.

For example, using affinity rules, you could spread or pack pods within a service or relative to pods in other services. Anti-affinity rules allow you to prevent pods of a particular service from scheduling on the same nodes as pods of another service that are known to interfere with the performance of the pods of the first service. Or, you could spread the pods of a service across nodes, availability zones, or availability sets to reduce correlated failures.

Note

A label selector might match pods with multiple pod deployments. Use unique combinations of labels when configuring anti-affinity rules to avoid matching pods.

There are two types of pod affinity rules: required and preferred.

Required rules must be met before a pod can be scheduled on a node. Preferred rules specify that, if the rule is met, the scheduler tries to enforce the rules, but does not guarantee enforcement.

Note

Depending on your pod priority and preemption settings, the scheduler might not be able to find an appropriate node for a pod without violating affinity requirements. If so, a pod might not be scheduled.

To prevent this situation, carefully configure pod affinity with equal-priority pods.

You configure pod affinity/anti-affinity through the Pod spec files. You can specify a required rule, a preferred rule, or both. If you specify both, the node must first meet the required rule, then attempts to meet the preferred rule.

The following example shows a Pod spec configured for pod affinity and anti-affinity.

In this example, the pod affinity rule indicates that the pod can schedule onto a node only if that node has at least one already-running pod with a label that has the key security and value S1. The pod anti-affinity rule says that the pod prefers to not schedule onto a node if that node is already running a pod with label having key security and value S2.

Sample Pod config file with pod affinity

apiVersion: v1
kind: Pod
metadata:
  name: with-pod-affinity
spec:
  affinity:
    podAffinity: 1
      requiredDuringSchedulingIgnoredDuringExecution: 2
      - labelSelector:
          matchExpressions:
          - key: security 3
            operator: In 4
            values:
            - S1 5
        topologyKey: failure-domain.beta.kubernetes.io/zone
  containers:
  - name: with-pod-affinity
    image: docker.io/ocpqe/hello-pod

1
Stanza to configure pod affinity.
2
Defines a required rule.
3 5
The key and value (label) that must be matched to apply the rule.
4
The operator represents the relationship between the label on the existing pod and the set of values in the matchExpression parameters in the specification for the new pod. Can be In, NotIn, Exists, or DoesNotExist.

Sample Pod config file with pod anti-affinity

apiVersion: v1
kind: Pod
metadata:
  name: with-pod-antiaffinity
spec:
  affinity:
    podAntiAffinity: 1
      preferredDuringSchedulingIgnoredDuringExecution: 2
      - weight: 100  3
        podAffinityTerm:
          labelSelector:
            matchExpressions:
            - key: security 4
              operator: In 5
              values:
              - S2
          topologyKey: kubernetes.io/hostname
  containers:
  - name: with-pod-affinity
    image: docker.io/ocpqe/hello-pod

1
Stanza to configure pod anti-affinity.
2
Defines a preferred rule.
3
Specifies a weight for a preferred rule. The node with the highest weight is preferred.
4
Description of the pod label that determines when the anti-affinity rule applies. Specify a key and value for the label.
5
The operator represents the relationship between the label on the existing pod and the set of values in the matchExpression parameters in the specification for the new pod. Can be In, NotIn, Exists, or DoesNotExist.
Note

If labels on a node change at runtime such that the affinity rules on a pod are no longer met, the pod continues to run on the node.

4.3.2. Configuring a pod affinity rule

The following steps demonstrate a simple two-pod configuration that creates pod with a label and a pod that uses affinity to allow scheduling with that pod.

Note

You cannot add an affinity directly to a scheduled pod.

Procedure

  1. Create a pod with a specific label in the pod spec:

    1. Create a YAML file with the following content: 

      apiVersion: v1
kind: Pod
metadata:
  name: security-s1
  labels:
    security: S1
spec:
  containers:
  - name: security-s1
    image: docker.io/ocpqe/hello-pod

    2. Create the pod. 

      $ oc create -f <pod-spec>.yaml

  2. When creating other pods, configure the following parameters to add the affinity:

    1. Create a YAML file with the following content: 

      apiVersion: v1
kind: Pod
metadata:
  name: security-s1-east
#...
spec
  affinity 1
    podAffinity:
      requiredDuringSchedulingIgnoredDuringExecution: 2
      - labelSelector:
          matchExpressions:
          - key: security 3
            values:
            - S1
            operator: In 4
        topologyKey: topology.kubernetes.io/zone 5
#...
      1
      Adds a pod affinity.
      2
      Configures the requiredDuringSchedulingIgnoredDuringExecution parameter or the preferredDuringSchedulingIgnoredDuringExecution parameter.
      3
      Specifies the key and values that must be met. If you want the new pod to be scheduled with the other pod, use the same key and values parameters as the label on the first pod.
      4
      Specifies an operator. The operator can be In, NotIn, Exists, or DoesNotExist. For example, use the operator In to require the label to be in the node.
      5
      Specify a topologyKey, which is a prepopulated Kubernetes label that the system uses to denote such a topology domain.

    2. Create the pod. 

      $ oc create -f <pod-spec>.yaml

4.3.3. Configuring a pod anti-affinity rule

The following steps demonstrate a simple two-pod configuration that creates pod with a label and a pod that uses an anti-affinity preferred rule to attempt to prevent scheduling with that pod.

Note

You cannot add an affinity directly to a scheduled pod.

Procedure

  1. Create a pod with a specific label in the pod spec:

    1. Create a YAML file with the following content: 

      apiVersion: v1
kind: Pod
metadata:
  name: security-s1
  labels:
    security: S1
spec:
  containers:
  - name: security-s1
    image: docker.io/ocpqe/hello-pod

    2. Create the pod. 

      $ oc create -f <pod-spec>.yaml

  2. When creating other pods, configure the following parameters:

    1. Create a YAML file with the following content: 

      apiVersion: v1
kind: Pod
metadata:
  name: security-s2-east
#...
spec
  affinity 1
    podAntiAffinity:
      preferredDuringSchedulingIgnoredDuringExecution: 2
      - weight: 100 3
        podAffinityTerm:
          labelSelector:
            matchExpressions:
            - key: security 4
              values:
              - S1
              operator: In 5
          topologyKey: kubernetes.io/hostname 6
#...
      1
      Adds a pod anti-affinity.
      2
      Configures the requiredDuringSchedulingIgnoredDuringExecution parameter or the preferredDuringSchedulingIgnoredDuringExecution parameter.
      3
      For a preferred rule, specifies a weight for the node, 1-100. The node that with highest weight is preferred.
      4
      Specifies the key and values that must be met. If you want the new pod to not be scheduled with the other pod, use the same key and values parameters as the label on the first pod.
      5
      Specifies an operator. The operator can be In, NotIn, Exists, or DoesNotExist. For example, use the operator In to require the label to be in the node.
      6
      Specifies a topologyKey, which is a prepopulated Kubernetes label that the system uses to denote such a topology domain.

    2. Create the pod. 

      $ oc create -f <pod-spec>.yaml

4.3.4. Sample pod affinity and anti-affinity rules

The following examples demonstrate pod affinity and pod anti-affinity.

4.3.4.1. Pod Affinity

The following example demonstrates pod affinity for pods with matching labels and label selectors.

  • The pod team4 has the label team:4. 

    apiVersion: v1
kind: Pod
metadata:
  name: team4
  labels:
     team: "4"
#...
spec:
  containers:
  - name: ocp
    image: docker.io/ocpqe/hello-pod
#...

  • The pod team4a has the label selector team:4 under podAffinity. 

    apiVersion: v1
kind: Pod
metadata:
  name: team4a
#...
spec:
  affinity:
    podAffinity:
      requiredDuringSchedulingIgnoredDuringExecution:
      - labelSelector:
          matchExpressions:
          - key: team
            operator: In
            values:
            - "4"
        topologyKey: kubernetes.io/hostname
  containers:
  - name: pod-affinity
    image: docker.io/ocpqe/hello-pod
#...
  • The team4a pod is scheduled on the same node as the team4 pod.

4.3.4.2. Pod Anti-affinity

The following example demonstrates pod anti-affinity for pods with matching labels and label selectors.

  • The pod pod-s1 has the label security:s1. 

    apiVersion: v1
kind: Pod
metadata:
  name: pod-s1
  labels:
    security: s1
#...
spec:
  containers:
  - name: ocp
    image: docker.io/ocpqe/hello-pod
#...

  • The pod pod-s2 has the label selector security:s1 under podAntiAffinity. 

    apiVersion: v1
kind: Pod
metadata:
  name: pod-s2
#...
spec:
  affinity:
    podAntiAffinity:
      requiredDuringSchedulingIgnoredDuringExecution:
      - labelSelector:
          matchExpressions:
          - key: security
            operator: In
            values:
            - s1
        topologyKey: kubernetes.io/hostname
  containers:
  - name: pod-antiaffinity
    image: docker.io/ocpqe/hello-pod
#...
  • The pod pod-s2 cannot be scheduled on the same node as pod-s1.

4.3.4.3. Pod Affinity with no Matching Labels

The following example demonstrates pod affinity for pods without matching labels and label selectors.

  • The pod pod-s1 has the label security:s1. 

    apiVersion: v1
kind: Pod
metadata:
  name: pod-s1
  labels:
    security: s1
#...
spec:
  containers:
  - name: ocp
    image: docker.io/ocpqe/hello-pod
#...

  • The pod pod-s2 has the label selector security:s2. 

    apiVersion: v1
kind: Pod
metadata:
  name: pod-s2
#...
spec:
  affinity:
    podAffinity:
      requiredDuringSchedulingIgnoredDuringExecution:
      - labelSelector:
          matchExpressions:
          - key: security
            operator: In
            values:
            - s2
        topologyKey: kubernetes.io/hostname
  containers:
  - name: pod-affinity
    image: docker.io/ocpqe/hello-pod
#...

  • The pod pod-s2 is not scheduled unless there is a node with a pod that has the security:s2 label. If there is no other pod with that label, the new pod remains in a pending state:

    Example output

    NAME      READY     STATUS    RESTARTS   AGE       IP        NODE
pod-s2    0/1       Pending   0          32s       <none>

4.4. Controlling pod placement on nodes using node affinity rules

Affinity is a property of pods that controls the nodes on which they prefer to be scheduled.

In OpenShift Container Platform node affinity is a set of rules used by the scheduler to determine where a pod can be placed. The rules are defined using custom labels on the nodes and label selectors specified in pods.

4.4.1. Understanding node affinity

Node affinity allows a pod to specify an affinity towards a group of nodes it can be placed on. The node does not have control over the placement.

For example, you could configure a pod to only run on a node with a specific CPU or in a specific availability zone.

There are two types of node affinity rules: required and preferred.

Required rules must be met before a pod can be scheduled on a node. Preferred rules specify that, if the rule is met, the scheduler tries to enforce the rules, but does not guarantee enforcement.

Note

If labels on a node change at runtime that results in an node affinity rule on a pod no longer being met, the pod continues to run on the node.

You configure node affinity through the Pod spec file. You can specify a required rule, a preferred rule, or both. If you specify both, the node must first meet the required rule, then attempts to meet the preferred rule.

The following example is a Pod spec with a rule that requires the pod be placed on a node with a label whose key is e2e-az-NorthSouth and whose value is either e2e-az-North or e2e-az-South:

Example pod configuration file with a node affinity required rule

apiVersion: v1
kind: Pod
metadata:
  name: with-node-affinity
spec:
  affinity:
    nodeAffinity: 1
      requiredDuringSchedulingIgnoredDuringExecution: 2
        nodeSelectorTerms:
        - matchExpressions:
          - key: e2e-az-NorthSouth 3
            operator: In 4
            values:
            - e2e-az-North 5
            - e2e-az-South 6
  containers:
  - name: with-node-affinity
    image: docker.io/ocpqe/hello-pod
#...

1
The stanza to configure node affinity.
2
Defines a required rule.
3 5 6
The key/value pair (label) that must be matched to apply the rule.
4
The operator represents the relationship between the label on the node and the set of values in the matchExpression parameters in the Pod spec. This value can be In, NotIn, Exists, or DoesNotExist, Lt, or Gt.

The following example is a node specification with a preferred rule that a node with a label whose key is e2e-az-EastWest and whose value is either e2e-az-East or e2e-az-West is preferred for the pod:

Example pod configuration file with a node affinity preferred rule

apiVersion: v1
kind: Pod
metadata:
  name: with-node-affinity
spec:
  affinity:
    nodeAffinity: 1
      preferredDuringSchedulingIgnoredDuringExecution: 2
      - weight: 1 3
        preference:
          matchExpressions:
          - key: e2e-az-EastWest 4
            operator: In 5
            values:
            - e2e-az-East 6
            - e2e-az-West 7
  containers:
  - name: with-node-affinity
    image: docker.io/ocpqe/hello-pod
#...

1
The stanza to configure node affinity.
2
Defines a preferred rule.
3
Specifies a weight for a preferred rule. The node with highest weight is preferred.
4 6 7
The key/value pair (label) that must be matched to apply the rule.
5
The operator represents the relationship between the label on the node and the set of values in the matchExpression parameters in the Pod spec. This value can be In, NotIn, Exists, or DoesNotExist, Lt, or Gt.

There is no explicit node anti-affinity concept, but using the NotIn or DoesNotExist operator replicates that behavior.

Note

If you are using node affinity and node selectors in the same pod configuration, note the following:

  • If you configure both nodeSelector and nodeAffinity, both conditions must be satisfied for the pod to be scheduled onto a candidate node.
  • If you specify multiple nodeSelectorTerms associated with nodeAffinity types, then the pod can be scheduled onto a node if one of the nodeSelectorTerms is satisfied.
  • If you specify multiple matchExpressions associated with nodeSelectorTerms, then the pod can be scheduled onto a node only if all matchExpressions are satisfied.

4.4.2. Configuring a required node affinity rule

Required rules must be met before a pod can be scheduled on a node.

Procedure

The following steps demonstrate a simple configuration that creates a node and a pod that the scheduler is required to place on the node.

  1. Add a label to a node using the oc label node command: 

    $ oc label node node1 e2e-az-name=e2e-az1
    Tip

    You can alternatively apply the following YAML to add the label: 

    kind: Node
apiVersion: v1
metadata:
  name: <node_name>
  labels:
    e2e-az-name: e2e-az1
#...

  2. Create a pod with a specific label in the pod spec:

    1. Create a YAML file with the following content:

      Note

      You cannot add an affinity directly to a scheduled pod.

      Example output

      apiVersion: v1
kind: Pod
metadata:
  name: s1
spec:
  affinity: 1
    nodeAffinity:
      requiredDuringSchedulingIgnoredDuringExecution: 2
        nodeSelectorTerms:
        - matchExpressions:
          - key: e2e-az-name 3
            values:
            - e2e-az1
            - e2e-az2
            operator: In 4
#...

      1
      Adds a pod affinity.
      2
      Configures the requiredDuringSchedulingIgnoredDuringExecution parameter.
      3
      Specifies the key and values that must be met. If you want the new pod to be scheduled on the node you edited, use the same key and values parameters as the label in the node.
      4
      Specifies an operator. The operator can be In, NotIn, Exists, or DoesNotExist. For example, use the operator In to require the label to be in the node.

    2. Create the pod: 

      $ oc create -f <file-name>.yaml

4.4.3. Configuring a preferred node affinity rule

Preferred rules specify that, if the rule is met, the scheduler tries to enforce the rules, but does not guarantee enforcement.

Procedure

The following steps demonstrate a simple configuration that creates a node and a pod that the scheduler tries to place on the node.

  1. Add a label to a node using the oc label node command: 

    $ oc label node node1 e2e-az-name=e2e-az3

  2. Create a pod with a specific label:

    1. Create a YAML file with the following content:

      Note

      You cannot add an affinity directly to a scheduled pod. 

      apiVersion: v1
kind: Pod
metadata:
  name: s1
spec:
  affinity: 1
    nodeAffinity:
      preferredDuringSchedulingIgnoredDuringExecution: 2
      - weight: 3
        preference:
          matchExpressions:
          - key: e2e-az-name 4
            values:
            - e2e-az3
            operator: In 5
#...
      1
      Adds a pod affinity.
      2
      Configures the preferredDuringSchedulingIgnoredDuringExecution parameter.
      3
      Specifies a weight for the node, as a number 1-100. The node with highest weight is preferred.
      4
      Specifies the key and values that must be met. If you want the new pod to be scheduled on the node you edited, use the same key and values parameters as the label in the node.
      5
      Specifies an operator. The operator can be In, NotIn, Exists, or DoesNotExist. For example, use the operator In to require the label to be in the node.

    2. Create the pod. 

      $ oc create -f <file-name>.yaml

4.4.4. Sample node affinity rules

The following examples demonstrate node affinity.

4.4.4.1. Node affinity with matching labels

The following example demonstrates node affinity for a node and pod with matching labels:

  • The Node1 node has the label zone:us: 

    $ oc label node node1 zone=us
    Tip

    You can alternatively apply the following YAML to add the label: 

    kind: Node
apiVersion: v1
metadata:
  name: <node_name>
  labels:
    zone: us
#...

  • The pod-s1 pod has the zone and us key/value pair under a required node affinity rule: 

    $ cat pod-s1.yaml

    Example output

    apiVersion: v1
kind: Pod
metadata:
  name: pod-s1
spec:
  containers:
    - image: "docker.io/ocpqe/hello-pod"
      name: hello-pod
  affinity:
    nodeAffinity:
      requiredDuringSchedulingIgnoredDuringExecution:
        nodeSelectorTerms:
          - matchExpressions:
            - key: "zone"
              operator: In
              values:
              - us
#...

  • The pod-s1 pod can be scheduled on Node1: 

    $ oc get pod -o wide

    Example output

    NAME     READY     STATUS       RESTARTS   AGE      IP      NODE
pod-s1   1/1       Running      0          4m       IP1     node1

4.4.4.2. Node affinity with no matching labels

The following example demonstrates node affinity for a node and pod without matching labels:

  • The Node1 node has the label zone:emea: 

    $ oc label node node1 zone=emea
    Tip

    You can alternatively apply the following YAML to add the label: 

    kind: Node
apiVersion: v1
metadata:
  name: <node_name>
  labels:
    zone: emea
#...

  • The pod-s1 pod has the zone and us key/value pair under a required node affinity rule: 

    $ cat pod-s1.yaml

    Example output

    apiVersion: v1
kind: Pod
metadata:
  name: pod-s1
spec:
  containers:
    - image: "docker.io/ocpqe/hello-pod"
      name: hello-pod
  affinity:
    nodeAffinity:
      requiredDuringSchedulingIgnoredDuringExecution:
        nodeSelectorTerms:
          - matchExpressions:
            - key: "zone"
              operator: In
              values:
              - us
#...

  • The pod-s1 pod cannot be scheduled on Node1: 

    $ oc describe pod pod-s1

    Example output

    ...

Events:
 FirstSeen LastSeen Count From              SubObjectPath  Type                Reason
 --------- -------- ----- ----              -------------  --------            ------
 1m        33s      8     default-scheduler Warning        FailedScheduling    No nodes are available that match all of the following predicates:: MatchNodeSelector (1).

4.4.5. Additional resources

4.5. Placing pods onto overcommited nodes

In an overcommited state, the sum of the container compute resource requests and limits exceeds the resources available on the system. Overcommitment might be desirable in development environments where a trade-off of guaranteed performance for capacity is acceptable.

Requests and limits enable administrators to allow and manage the overcommitment of resources on a node. The scheduler uses requests for scheduling your container and providing a minimum service guarantee. Limits constrain the amount of compute resource that may be consumed on your node.

4.5.1. Understanding overcommitment

Requests and limits enable administrators to allow and manage the overcommitment of resources on a node. The scheduler uses requests for scheduling your container and providing a minimum service guarantee. Limits constrain the amount of compute resource that may be consumed on your node.

OpenShift Container Platform administrators can control the level of overcommit and manage container density on nodes by configuring masters to override the ratio between request and limit set on developer containers. In conjunction with a per-project LimitRange object specifying limits and defaults, this adjusts the container limit and request to achieve the desired level of overcommit.

Note

That these overrides have no effect if no limits have been set on containers. Create a LimitRange object with default limits, per individual project, or in the project template, to ensure that the overrides apply.

After these overrides, the container limits and requests must still be validated by any LimitRange object in the project. It is possible, for example, for developers to specify a limit close to the minimum limit, and have the request then be overridden below the minimum limit, causing the pod to be forbidden. This unfortunate user experience should be addressed with future work, but for now, configure this capability and LimitRange objects with caution.

4.5.2. Understanding nodes overcommitment

In an overcommitted environment, it is important to properly configure your node to provide best system behavior.

When the node starts, it ensures that the kernel tunable flags for memory management are set properly. The kernel should never fail memory allocations unless it runs out of physical memory.

To ensure this behavior, OpenShift Container Platform configures the kernel to always overcommit memory by setting the vm.overcommit_memory parameter to 1, overriding the default operating system setting.

OpenShift Container Platform also configures the kernel not to panic when it runs out of memory by setting the vm.panic_on_oom parameter to 0. A setting of 0 instructs the kernel to call oom_killer in an Out of Memory (OOM) condition, which kills processes based on priority

You can view the current setting by running the following commands on your nodes: 

$ sysctl -a |grep commit

Example output

#...
vm.overcommit_memory = 0
#...

$ sysctl -a |grep panic

Example output

#...
vm.panic_on_oom = 0
#...

Note

The above flags should already be set on nodes, and no further action is required.

You can also perform the following configurations for each node:

  • Disable or enforce CPU limits using CPU CFS quotas
  • Reserve resources for system processes
  • Reserve memory across quality of service tiers

4.6. Controlling pod placement using node taints

Taints and tolerations allow the node to control which pods should (or should not) be scheduled on them.

4.6.1. Understanding taints and tolerations

A taint allows a node to refuse a pod to be scheduled unless that pod has a matching toleration.

You apply taints to a node through the Node specification (NodeSpec) and apply tolerations to a pod through the Pod specification (PodSpec). When you apply a taint a node, the scheduler cannot place a pod on that node unless the pod can tolerate the taint.

Example taint in a node specification

apiVersion: v1
kind: Node
metadata:
  name: my-node
#...
spec:
  taints:
  - effect: NoExecute
    key: key1
    value: value1
#...

Example toleration in a Pod spec

apiVersion: v1
kind: Pod
metadata:
  name: my-pod
#...
spec:
  tolerations:
  - key: "key1"
    operator: "Equal"
    value: "value1"
    effect: "NoExecute"
    tolerationSeconds: 3600
#...

Taints and tolerations consist of a key, value, and effect.

Table 4.1. Taint and toleration components

ParameterDescription

key

The key is any string, up to 253 characters. The key must begin with a letter or number, and may contain letters, numbers, hyphens, dots, and underscores.

value

The value is any string, up to 63 characters. The value must begin with a letter or number, and may contain letters, numbers, hyphens, dots, and underscores.

effect

The effect is one of the following:

NoSchedule [1]

  • New pods that do not match the taint are not scheduled onto that node.
  • Existing pods on the node remain.

PreferNoSchedule

  • New pods that do not match the taint might be scheduled onto that node, but the scheduler tries not to.
  • Existing pods on the node remain.

NoExecute

  • New pods that do not match the taint cannot be scheduled onto that node.
  • Existing pods on the node that do not have a matching toleration are removed.

operator

Equal

The key/value/effect parameters must match. This is the default.

Exists

The key/effect parameters must match. You must leave a blank value parameter, which matches any.

  1. If you add a NoSchedule taint to a control plane node, the node must have the node-role.kubernetes.io/master=:NoSchedule taint, which is added by default.

    For example: 

    apiVersion: v1
kind: Node
metadata:
  annotations:
    machine.openshift.io/machine: openshift-machine-api/ci-ln-62s7gtb-f76d1-v8jxv-master-0
    machineconfiguration.openshift.io/currentConfig: rendered-master-cdc1ab7da414629332cc4c3926e6e59c
  name: my-node
#...
spec:
  taints:
  - effect: NoSchedule
    key: node-role.kubernetes.io/master
#...

A toleration matches a taint:

  • If the operator parameter is set to Equal:

    • the key parameters are the same;
    • the value parameters are the same;
    • the effect parameters are the same.

  • If the operator parameter is set to Exists:

    • the key parameters are the same;
    • the effect parameters are the same.

The following taints are built into OpenShift Container Platform:

  • node.kubernetes.io/not-ready: The node is not ready. This corresponds to the node condition Ready=False.
  • node.kubernetes.io/unreachable: The node is unreachable from the node controller. This corresponds to the node condition Ready=Unknown.
  • node.kubernetes.io/memory-pressure: The node has memory pressure issues. This corresponds to the node condition MemoryPressure=True.
  • node.kubernetes.io/disk-pressure: The node has disk pressure issues. This corresponds to the node condition DiskPressure=True.
  • node.kubernetes.io/network-unavailable: The node network is unavailable.
  • node.kubernetes.io/unschedulable: The node is unschedulable.
  • node.cloudprovider.kubernetes.io/uninitialized: When the node controller is started with an external cloud provider, this taint is set on a node to mark it as unusable. After a controller from the cloud-controller-manager initializes this node, the kubelet removes this taint.

  • node.kubernetes.io/pid-pressure: The node has pid pressure. This corresponds to the node condition PIDPressure=True.

    Important

    OpenShift Container Platform does not set a default pid.available evictionHard.

4.6.1.1. Understanding how to use toleration seconds to delay pod evictions

You can specify how long a pod can remain bound to a node before being evicted by specifying the tolerationSeconds parameter in the Pod specification or MachineSet object. If a taint with the NoExecute effect is added to a node, a pod that does tolerate the taint, which has the tolerationSeconds parameter, the pod is not evicted until that time period expires.

Example output

apiVersion: v1
kind: Pod
metadata:
  name: my-pod
#...
spec:
  tolerations:
  - key: "key1"
    operator: "Equal"
    value: "value1"
    effect: "NoExecute"
    tolerationSeconds: 3600
#...

Here, if this pod is running but does not have a matching toleration, the pod stays bound to the node for 3,600 seconds and then be evicted. If the taint is removed before that time, the pod is not evicted.

4.6.1.2. Understanding how to use multiple taints

You can put multiple taints on the same node and multiple tolerations on the same pod. OpenShift Container Platform processes multiple taints and tolerations as follows:

  1. Process the taints for which the pod has a matching toleration.

  2. The remaining unmatched taints have the indicated effects on the pod:

    • If there is at least one unmatched taint with effect NoSchedule, OpenShift Container Platform cannot schedule a pod onto that node.
    • If there is no unmatched taint with effect NoSchedule but there is at least one unmatched taint with effect PreferNoSchedule, OpenShift Container Platform tries to not schedule the pod onto the node.

    • If there is at least one unmatched taint with effect NoExecute, OpenShift Container Platform evicts the pod from the node if it is already running on the node, or the pod is not scheduled onto the node if it is not yet running on the node.

      • Pods that do not tolerate the taint are evicted immediately.
      • Pods that tolerate the taint without specifying tolerationSeconds in their Pod specification remain bound forever.
      • Pods that tolerate the taint with a specified tolerationSeconds remain bound for the specified amount of time.

For example:

  • Add the following taints to the node: 

    $ oc adm taint nodes node1 key1=value1:NoSchedule
    $ oc adm taint nodes node1 key1=value1:NoExecute
    $ oc adm taint nodes node1 key2=value2:NoSchedule

  • The pod has the following tolerations: 

    apiVersion: v1
kind: Pod
metadata:
  name: my-pod
#...
spec:
  tolerations:
  - key: "key1"
    operator: "Equal"
    value: "value1"
    effect: "NoSchedule"
  - key: "key1"
    operator: "Equal"
    value: "value1"
    effect: "NoExecute"
#...

In this case, the pod cannot be scheduled onto the node, because there is no toleration matching the third taint. The pod continues running if it is already running on the node when the taint is added, because the third taint is the only one of the three that is not tolerated by the pod.

4.6.1.3. Understanding pod scheduling and node conditions (taint node by condition)

The Taint Nodes By Condition feature, which is enabled by default, automatically taints nodes that report conditions such as memory pressure and disk pressure. If a node reports a condition, a taint is added until the condition clears. The taints have the NoSchedule effect, which means no pod can be scheduled on the node unless the pod has a matching toleration.

The scheduler checks for these taints on nodes before scheduling pods. If the taint is present, the pod is scheduled on a different node. Because the scheduler checks for taints and not the actual node conditions, you configure the scheduler to ignore some of these node conditions by adding appropriate pod tolerations.

To ensure backward compatibility, the daemon set controller automatically adds the following tolerations to all daemons:

  • node.kubernetes.io/memory-pressure
  • node.kubernetes.io/disk-pressure
  • node.kubernetes.io/unschedulable (1.10 or later)
  • node.kubernetes.io/network-unavailable (host network only)

You can also add arbitrary tolerations to daemon sets.

Note

The control plane also adds the node.kubernetes.io/memory-pressure toleration on pods that have a QoS class. This is because Kubernetes manages pods in the Guaranteed or Burstable QoS classes. The new BestEffort pods do not get scheduled onto the affected node.

4.6.1.4. Understanding evicting pods by condition (taint-based evictions)

The Taint-Based Evictions feature, which is enabled by default, evicts pods from a node that experiences specific conditions, such as not-ready and unreachable. When a node experiences one of these conditions, OpenShift Container Platform automatically adds taints to the node, and starts evicting and rescheduling the pods on different nodes.

Taint Based Evictions have a NoExecute effect, where any pod that does not tolerate the taint is evicted immediately and any pod that does tolerate the taint will never be evicted, unless the pod uses the tolerationSeconds parameter.

The tolerationSeconds parameter allows you to specify how long a pod stays bound to a node that has a node condition. If the condition still exists after the tolerationSeconds period, the taint remains on the node and the pods with a matching toleration are evicted. If the condition clears before the tolerationSeconds period, pods with matching tolerations are not removed.

If you use the tolerationSeconds parameter with no value, pods are never evicted because of the not ready and unreachable node conditions.

Note

OpenShift Container Platform evicts pods in a rate-limited way to prevent massive pod evictions in scenarios such as the master becoming partitioned from the nodes.

By default, if more than 55% of nodes in a given zone are unhealthy, the node lifecycle controller changes that zone’s state to PartialDisruption and the rate of pod evictions is reduced. For small clusters (by default, 50 nodes or less) in this state, nodes in this zone are not tainted and evictions are stopped.

For more information, see Rate limits on eviction in the Kubernetes documentation.

OpenShift Container Platform automatically adds a toleration for node.kubernetes.io/not-ready and node.kubernetes.io/unreachable with tolerationSeconds=300, unless the Pod configuration specifies either toleration. 

apiVersion: v1
kind: Pod
metadata:
  name: my-pod
#...
spec:
  tolerations:
  - key: node.kubernetes.io/not-ready
    operator: Exists
    effect: NoExecute
    tolerationSeconds: 300 1
  - key: node.kubernetes.io/unreachable
    operator: Exists
    effect: NoExecute
    tolerationSeconds: 300
#...
1
These tolerations ensure that the default pod behavior is to remain bound for five minutes after one of these node conditions problems is detected.

You can configure these tolerations as needed. For example, if you have an application with a lot of local state, you might want to keep the pods bound to node for a longer time in the event of network partition, allowing for the partition to recover and avoiding pod eviction.

Pods spawned by a daemon set are created with NoExecute tolerations for the following taints with no tolerationSeconds:

  • node.kubernetes.io/unreachable
  • node.kubernetes.io/not-ready

As a result, daemon set pods are never evicted because of these node conditions.

4.6.1.5. Tolerating all taints

You can configure a pod to tolerate all taints by adding an operator: "Exists" toleration with no key and values parameters. Pods with this toleration are not removed from a node that has taints.

Pod spec for tolerating all taints

apiVersion: v1
kind: Pod
metadata:
  name: my-pod
#...
spec:
  tolerations:
  - operator: "Exists"
#...

4.6.2. Adding taints and tolerations

You add tolerations to pods and taints to nodes to allow the node to control which pods should or should not be scheduled on them. For existing pods and nodes, you should add the toleration to the pod first, then add the taint to the node to avoid pods being removed from the node before you can add the toleration.

Procedure

  1. Add a toleration to a pod by editing the Pod spec to include a tolerations stanza:

    Sample pod configuration file with an Equal operator

    apiVersion: v1
kind: Pod
metadata:
  name: my-pod
#...
spec:
  tolerations:
  - key: "key1" 1
    value: "value1"
    operator: "Equal"
    effect: "NoExecute"
    tolerationSeconds: 3600 2
#...

    1
    The toleration parameters, as described in the Taint and toleration components table.
    2
    The tolerationSeconds parameter specifies how long a pod can remain bound to a node before being evicted.

    For example:

    Sample pod configuration file with an Exists operator

    apiVersion: v1
kind: Pod
metadata:
  name: my-pod
#...
spec:
   tolerations:
    - key: "key1"
      operator: "Exists" 1
      effect: "NoExecute"
      tolerationSeconds: 3600
#...

    1
    The Exists operator does not take a value.

    This example places a taint on node1 that has key key1, value value1, and taint effect NoExecute.

  2. Add a taint to a node by using the following command with the parameters described in the Taint and toleration components table: 

    $ oc adm taint nodes <node_name> <key>=<value>:<effect>

    For example: 

    $ oc adm taint nodes node1 key1=value1:NoExecute

    This command places a taint on node1 that has key key1, value value1, and effect NoExecute.

    Note

    If you add a NoSchedule taint to a control plane node, the node must have the node-role.kubernetes.io/master=:NoSchedule taint, which is added by default.

    For example: 

    apiVersion: v1
kind: Node
metadata:
  annotations:
    machine.openshift.io/machine: openshift-machine-api/ci-ln-62s7gtb-f76d1-v8jxv-master-0
    machineconfiguration.openshift.io/currentConfig: rendered-master-cdc1ab7da414629332cc4c3926e6e59c
  name: my-node
#...
spec:
  taints:
  - effect: NoSchedule
    key: node-role.kubernetes.io/master
#...

    The tolerations on the pod match the taint on the node. A pod with either toleration can be scheduled onto node1.

4.6.2.1. Adding taints and tolerations using a machine set

You can add taints to nodes using a machine set. All nodes associated with the MachineSet object are updated with the taint. Tolerations respond to taints added by a machine set in the same manner as taints added directly to the nodes.

Procedure

  1. Add a toleration to a pod by editing the Pod spec to include a tolerations stanza:

    Sample pod configuration file with Equal operator

    apiVersion: v1
kind: Pod
metadata:
  name: my-pod
#...
spec:
  tolerations:
  - key: "key1" 1
    value: "value1"
    operator: "Equal"
    effect: "NoExecute"
    tolerationSeconds: 3600 2
#...

    1
    The toleration parameters, as described in the Taint and toleration components table.
    2
    The tolerationSeconds parameter specifies how long a pod is bound to a node before being evicted.

    For example:

    Sample pod configuration file with Exists operator

    apiVersion: v1
kind: Pod
metadata:
  name: my-pod
#...
spec:
  tolerations:
  - key: "key1"
    operator: "Exists"
    effect: "NoExecute"
    tolerationSeconds: 3600
#...

  2. Add the taint to the MachineSet object:

    1. Edit the MachineSet YAML for the nodes you want to taint or you can create a new MachineSet object: 

      $ oc edit machineset <machineset>

    2. Add the taint to the spec.template.spec section:

      Example taint in a machine set specification

      apiVersion: machine.openshift.io/v1beta1
kind: MachineSet
metadata:
  name: my-machineset
#...
spec:
#...
  template:
#...
    spec:
      taints:
      - effect: NoExecute
        key: key1
        value: value1
#...

      This example places a taint that has the key key1, value value1, and taint effect NoExecute on the nodes.

    3. Scale down the machine set to 0: 

      $ oc scale --replicas=0 machineset <machineset> -n openshift-machine-api
      Tip

      You can alternatively apply the following YAML to scale the machine set: 

      apiVersion: machine.openshift.io/v1beta1
kind: MachineSet
metadata:
  name: <machineset>
  namespace: openshift-machine-api
spec:
  replicas: 0

      Wait for the machines to be removed.

    4. Scale up the machine set as needed: 

      $ oc scale --replicas=2 machineset <machineset> -n openshift-machine-api

      Or: 

      $ oc edit machineset <machineset> -n openshift-machine-api

      Wait for the machines to start. The taint is added to the nodes associated with the MachineSet object.

4.6.2.2. Binding a user to a node using taints and tolerations

If you want to dedicate a set of nodes for exclusive use by a particular set of users, add a toleration to their pods. Then, add a corresponding taint to those nodes. The pods with the tolerations are allowed to use the tainted nodes or any other nodes in the cluster.

If you want ensure the pods are scheduled to only those tainted nodes, also add a label to the same set of nodes and add a node affinity to the pods so that the pods can only be scheduled onto nodes with that label.

Procedure

To configure a node so that users can use only that node:

  1. Add a corresponding taint to those nodes:

    For example: 

    $ oc adm taint nodes node1 dedicated=groupName:NoSchedule
    Tip

    You can alternatively apply the following YAML to add the taint: 

    kind: Node
apiVersion: v1
metadata:
  name: my-node
#...
spec:
  taints:
    - key: dedicated
      value: groupName
      effect: NoSchedule
#...
  2. Add a toleration to the pods by writing a custom admission controller.

4.6.2.3. Creating a project with a node selector and toleration

You can create a project that uses a node selector and toleration, which are set as annotations, to control the placement of pods onto specific nodes. Any subsequent resources created in the project are then scheduled on nodes that have a taint matching the toleration.

Prerequisites

  • A label for node selection has been added to one or more nodes by using a machine set or editing the node directly.
  • A taint has been added to one or more nodes by using a machine set or editing the node directly.

Procedure

  1. Create a Project resource definition, specifying a node selector and toleration in the metadata.annotations section:

    Example project.yaml file

    kind: Project
apiVersion: project.openshift.io/v1
metadata:
  name: <project_name> 1
  annotations:
    openshift.io/node-selector: '<label>' 2
    scheduler.alpha.kubernetes.io/defaultTolerations: >-
      [{"operator": "Exists", "effect": "NoSchedule", "key":
      "<key_name>"} 3
      ]

    1
    The project name.
    2
    The default node selector label.
    3
    The toleration parameters, as described in the Taint and toleration components table. This example uses the NoSchedule effect, which allows existing pods on the node to remain, and the Exists operator, which does not take a value.

  2. Use the oc apply command to create the project: 

    $ oc apply -f project.yaml

Any subsequent resources created in the <project_name> namespace should now be scheduled on the specified nodes.

Additional resources

4.6.2.4. Controlling nodes with special hardware using taints and tolerations

In a cluster where a small subset of nodes have specialized hardware, you can use taints and tolerations to keep pods that do not need the specialized hardware off of those nodes, leaving the nodes for pods that do need the specialized hardware. You can also require pods that need specialized hardware to use specific nodes.

You can achieve this by adding a toleration to pods that need the special hardware and tainting the nodes that have the specialized hardware.

Procedure

To ensure nodes with specialized hardware are reserved for specific pods:

  1. Add a toleration to pods that need the special hardware.

    For example: 

    apiVersion: v1
kind: Pod
metadata:
  name: my-pod
#...
spec:
  tolerations:
    - key: "disktype"
      value: "ssd"
      operator: "Equal"
      effect: "NoSchedule"
      tolerationSeconds: 3600
#...

  2. Taint the nodes that have the specialized hardware using one of the following commands: 

    $ oc adm taint nodes <node-name> disktype=ssd:NoSchedule

    Or: 

    $ oc adm taint nodes <node-name> disktype=ssd:PreferNoSchedule
    Tip

    You can alternatively apply the following YAML to add the taint: 

    kind: Node
apiVersion: v1
metadata:
  name: my_node
#...
spec:
  taints:
    - key: disktype
      value: ssd
      effect: PreferNoSchedule
#...

4.6.3. Removing taints and tolerations

You can remove taints from nodes and tolerations from pods as needed. You should add the toleration to the pod first, then add the taint to the node to avoid pods being removed from the node before you can add the toleration.

Procedure

To remove taints and tolerations:

  1. To remove a taint from a node: 

    $ oc adm taint nodes <node-name> <key>-

    For example: 

    $ oc adm taint nodes ip-10-0-132-248.ec2.internal key1-

    Example output

    node/ip-10-0-132-248.ec2.internal untainted

  2. To remove a toleration from a pod, edit the Pod spec to remove the toleration: 

    apiVersion: v1
kind: Pod
metadata:
  name: my-pod
#...
spec:
  tolerations:
  - key: "key2"
    operator: "Exists"
    effect: "NoExecute"
    tolerationSeconds: 3600
#...

4.7. Placing pods on specific nodes using node selectors

A node selector specifies a map of key/value pairs that are defined using custom labels on nodes and selectors specified in pods.

For the pod to be eligible to run on a node, the pod must have the same key/value node selector as the label on the node.

4.7.1. About node selectors

You can use node selectors on pods and labels on nodes to control where the pod is scheduled. With node selectors, OpenShift Container Platform schedules the pods on nodes that contain matching labels.

You can use a node selector to place specific pods on specific nodes, cluster-wide node selectors to place new pods on specific nodes anywhere in the cluster, and project node selectors to place new pods in a project on specific nodes.

For example, as a cluster administrator, you can create an infrastructure where application developers can deploy pods only onto the nodes closest to their geographical location by including a node selector in every pod they create. In this example, the cluster consists of five data centers spread across two regions. In the U.S., label the nodes as us-east, us-central, or us-west. In the Asia-Pacific region (APAC), label the nodes as apac-east or apac-west. The developers can add a node selector to the pods they create to ensure the pods get scheduled on those nodes.

A pod is not scheduled if the Pod object contains a node selector, but no node has a matching label.

Important

If you are using node selectors and node affinity in the same pod configuration, the following rules control pod placement onto nodes:

  • If you configure both nodeSelector and nodeAffinity, both conditions must be satisfied for the pod to be scheduled onto a candidate node.
  • If you specify multiple nodeSelectorTerms associated with nodeAffinity types, then the pod can be scheduled onto a node if one of the nodeSelectorTerms is satisfied.
  • If you specify multiple matchExpressions associated with nodeSelectorTerms, then the pod can be scheduled onto a node only if all matchExpressions are satisfied.
Node selectors on specific pods and nodes

You can control which node a specific pod is scheduled on by using node selectors and labels.

To use node selectors and labels, first label the node to avoid pods being descheduled, then add the node selector to the pod.

Note

You cannot add a node selector directly to an existing scheduled pod. You must label the object that controls the pod, such as deployment config.

For example, the following Node object has the region: east label:

Sample Node object with a label

kind: Node
apiVersion: v1
metadata:
  name: ip-10-0-131-14.ec2.internal
  selfLink: /api/v1/nodes/ip-10-0-131-14.ec2.internal
  uid: 7bc2580a-8b8e-11e9-8e01-021ab4174c74
  resourceVersion: '478704'
  creationTimestamp: '2019-06-10T14:46:08Z'
  labels:
    kubernetes.io/os: linux
    failure-domain.beta.kubernetes.io/zone: us-east-1a
    node.openshift.io/os_version: '4.5'
    node-role.kubernetes.io/worker: ''
    failure-domain.beta.kubernetes.io/region: us-east-1
    node.openshift.io/os_id: rhcos
    beta.kubernetes.io/instance-type: m4.large
    kubernetes.io/hostname: ip-10-0-131-14
    beta.kubernetes.io/arch: amd64
    region: east 1
    type: user-node
#...

1
Labels to match the pod node selector.

A pod has the type: user-node,region: east node selector:

Sample Pod object with node selectors

apiVersion: v1
kind: Pod
metadata:
  name: s1
#...
spec:
  nodeSelector: 1
    region: east
    type: user-node
#...

1
Node selectors to match the node label. The node must have a label for each node selector.

When you create the pod using the example pod spec, it can be scheduled on the example node.

Default cluster-wide node selectors

With default cluster-wide node selectors, when you create a pod in that cluster, OpenShift Container Platform adds the default node selectors to the pod and schedules the pod on nodes with matching labels.

For example, the following Scheduler object has the default cluster-wide region=east and type=user-node node selectors:

Example Scheduler Operator Custom Resource

apiVersion: config.openshift.io/v1
kind: Scheduler
metadata:
  name: cluster
#...
spec:
  defaultNodeSelector: type=user-node,region=east
#...

A node in that cluster has the type=user-node,region=east labels:

Example Node object

apiVersion: v1
kind: Node
metadata:
  name: ci-ln-qg1il3k-f76d1-hlmhl-worker-b-df2s4
#...
  labels:
    region: east
    type: user-node
#...

Example Pod object with a node selector

apiVersion: v1
kind: Pod
metadata:
  name: s1
#...
spec:
  nodeSelector:
    region: east
#...

When you create the pod using the example pod spec in the example cluster, the pod is created with the cluster-wide node selector and is scheduled on the labeled node:

Example pod list with the pod on the labeled node

NAME     READY   STATUS    RESTARTS   AGE   IP           NODE                                       NOMINATED NODE   READINESS GATES
pod-s1   1/1     Running   0          20s   10.131.2.6   ci-ln-qg1il3k-f76d1-hlmhl-worker-b-df2s4   <none>           <none>

Note

If the project where you create the pod has a project node selector, that selector takes preference over a cluster-wide node selector. Your pod is not created or scheduled if the pod does not have the project node selector.

Project node selectors

With project node selectors, when you create a pod in this project, OpenShift Container Platform adds the node selectors to the pod and schedules the pods on a node with matching labels. If there is a cluster-wide default node selector, a project node selector takes preference.

For example, the following project has the region=east node selector:

Example Namespace object

apiVersion: v1
kind: Namespace
metadata:
  name: east-region
  annotations:
    openshift.io/node-selector: "region=east"
#...

The following node has the type=user-node,region=east labels:

Example Node object

apiVersion: v1
kind: Node
metadata:
  name: ci-ln-qg1il3k-f76d1-hlmhl-worker-b-df2s4
#...
  labels:
    region: east
    type: user-node
#...

When you create the pod using the example pod spec in this example project, the pod is created with the project node selectors and is scheduled on the labeled node:

Example Pod object

apiVersion: v1
kind: Pod
metadata:
  namespace: east-region
#...
spec:
  nodeSelector:
    region: east
    type: user-node
#...

Example pod list with the pod on the labeled node

NAME     READY   STATUS    RESTARTS   AGE   IP           NODE                                       NOMINATED NODE   READINESS GATES
pod-s1   1/1     Running   0          20s   10.131.2.6   ci-ln-qg1il3k-f76d1-hlmhl-worker-b-df2s4   <none>           <none>

A pod in the project is not created or scheduled if the pod contains different node selectors. For example, if you deploy the following pod into the example project, it is not be created:

Example Pod object with an invalid node selector

apiVersion: v1
kind: Pod
metadata:
  name: west-region
#...
spec:
  nodeSelector:
    region: west
#...

4.7.2. Using node selectors to control pod placement

You can use node selectors on pods and labels on nodes to control where the pod is scheduled. With node selectors, OpenShift Container Platform schedules the pods on nodes that contain matching labels.

You add labels to a node, a machine set, or a machine config. Adding the label to the machine set ensures that if the node or machine goes down, new nodes have the label. Labels added to a node or machine config do not persist if the node or machine goes down.

To add node selectors to an existing pod, add a node selector to the controlling object for that pod, such as a ReplicaSet object, DaemonSet object, StatefulSet object, Deployment object, or DeploymentConfig object. Any existing pods under that controlling object are recreated on a node with a matching label. If you are creating a new pod, you can add the node selector directly to the pod spec. If the pod does not have a controlling object, you must delete the pod, edit the pod spec, and recreate the pod.

Note

You cannot add a node selector directly to an existing scheduled pod.

Prerequisites

To add a node selector to existing pods, determine the controlling object for that pod. For example, the router-default-66d5cf9464-m2g75 pod is controlled by the router-default-66d5cf9464 replica set: 

$ oc describe pod router-default-66d5cf9464-7pwkc

Example output

kind: Pod
apiVersion: v1
metadata:
#...
Name:               router-default-66d5cf9464-7pwkc
Namespace:          openshift-ingress
# ...
Controlled By:      ReplicaSet/router-default-66d5cf9464
# ...

The web console lists the controlling object under ownerReferences in the pod YAML: 

apiVersion: v1
kind: Pod
metadata:
  name: router-default-66d5cf9464-7pwkc
# ...
  ownerReferences:
    - apiVersion: apps/v1
      kind: ReplicaSet
      name: router-default-66d5cf9464
      uid: d81dd094-da26-11e9-a48a-128e7edf0312
      controller: true
      blockOwnerDeletion: true
# ...

Procedure

  1. Add labels to a node by using a machine set or editing the node directly:

    • Use a MachineSet object to add labels to nodes managed by the machine set when a node is created:

      1. Run the following command to add labels to a MachineSet object: 

        $ oc patch MachineSet <name> --type='json' -p='[{"op":"add","path":"/spec/template/spec/metadata/labels", "value":{"<key>"="<value>","<key>"="<value>"}}]'  -n openshift-machine-api

        For example: 

        $ oc patch MachineSet abc612-msrtw-worker-us-east-1c  --type='json' -p='[{"op":"add","path":"/spec/template/spec/metadata/labels", "value":{"type":"user-node","region":"east"}}]'  -n openshift-machine-api
        Tip

        You can alternatively apply the following YAML to add labels to a machine set: 

        apiVersion: machine.openshift.io/v1beta1
kind: MachineSet
metadata:
  name: xf2bd-infra-us-east-2a
  namespace: openshift-machine-api
spec:
  template:
    spec:
      metadata:
        labels:
          region: "east"
          type: "user-node"
#...

      2. Verify that the labels are added to the MachineSet object by using the oc edit command:

        For example: 

        $ oc edit MachineSet abc612-msrtw-worker-us-east-1c -n openshift-machine-api

        Example MachineSet object

        apiVersion: machine.openshift.io/v1beta1
kind: MachineSet

# ...

spec:
# ...
  template:
    metadata:
# ...
    spec:
      metadata:
        labels:
          region: east
          type: user-node
# ...

    • Add labels directly to a node:

      1. Edit the Node object for the node: 

        $ oc label nodes <name> <key>=<value>

        For example, to label a node: 

        $ oc label nodes ip-10-0-142-25.ec2.internal type=user-node region=east
        Tip

        You can alternatively apply the following YAML to add labels to a node: 

        kind: Node
apiVersion: v1
metadata:
  name: hello-node-6fbccf8d9
  labels:
    type: "user-node"
    region: "east"
#...

      2. Verify that the labels are added to the node: 

        $ oc get nodes -l type=user-node,region=east

        Example output

        NAME                          STATUS   ROLES    AGE   VERSION
ip-10-0-142-25.ec2.internal   Ready    worker   17m   v1.24.0

  2. Add the matching node selector to a pod:

    • To add a node selector to existing and future pods, add a node selector to the controlling object for the pods:

      Example ReplicaSet object with labels

      kind: ReplicaSet
apiVersion: apps/v1
metadata:
  name: hello-node-6fbccf8d9
# ...
spec:
# ...
  template:
    metadata:
      creationTimestamp: null
      labels:
        ingresscontroller.operator.openshift.io/deployment-ingresscontroller: default
        pod-template-hash: 66d5cf9464
    spec:
      nodeSelector:
        kubernetes.io/os: linux
        node-role.kubernetes.io/worker: ''
        type: user-node 1
#...

      1
      Add the node selector.

    • To add a node selector to a specific, new pod, add the selector to the Pod object directly:

      Example Pod object with a node selector

      apiVersion: v1
kind: Pod
metadata:
  name: hello-node-6fbccf8d9
#...
spec:
  nodeSelector:
    region: east
    type: user-node
#...

      Note

      You cannot add a node selector directly to an existing scheduled pod.

4.7.3. Creating default cluster-wide node selectors

You can use default cluster-wide node selectors on pods together with labels on nodes to constrain all pods created in a cluster to specific nodes.

With cluster-wide node selectors, when you create a pod in that cluster, OpenShift Container Platform adds the default node selectors to the pod and schedules the pod on nodes with matching labels.

You configure cluster-wide node selectors by editing the Scheduler Operator custom resource (CR). You add labels to a node, a machine set, or a machine config. Adding the label to the machine set ensures that if the node or machine goes down, new nodes have the label. Labels added to a node or machine config do not persist if the node or machine goes down.

Note

You can add additional key/value pairs to a pod. But you cannot add a different value for a default key.

Procedure

To add a default cluster-wide node selector:

  1. Edit the Scheduler Operator CR to add the default cluster-wide node selectors: 

    $ oc edit scheduler cluster

    Example Scheduler Operator CR with a node selector

    apiVersion: config.openshift.io/v1
kind: Scheduler
metadata:
  name: cluster
...
spec:
  defaultNodeSelector: type=user-node,region=east 1
  mastersSchedulable: false

    1
    Add a node selector with the appropriate <key>:<value> pairs.

    After making this change, wait for the pods in the openshift-kube-apiserver project to redeploy. This can take several minutes. The default cluster-wide node selector does not take effect until the pods redeploy.

  2. Add labels to a node by using a machine set or editing the node directly:

    • Use a machine set to add labels to nodes managed by the machine set when a node is created:

      1. Run the following command to add labels to a MachineSet object: 

        $ oc patch MachineSet <name> --type='json' -p='[{"op":"add","path":"/spec/template/spec/metadata/labels", "value":{"<key>"="<value>","<key>"="<value>"}}]'  -n openshift-machine-api 1
        1
        Add a <key>/<value> pair for each label.

        For example: 

        $ oc patch MachineSet ci-ln-l8nry52-f76d1-hl7m7-worker-c --type='json' -p='[{"op":"add","path":"/spec/template/spec/metadata/labels", "value":{"type":"user-node","region":"east"}}]'  -n openshift-machine-api
        Tip

        You can alternatively apply the following YAML to add labels to a machine set: 

        apiVersion: machine.openshift.io/v1beta1
kind: MachineSet
metadata:
  name: <machineset>
  namespace: openshift-machine-api
spec:
  template:
    spec:
      metadata:
        labels:
          region: "east"
          type: "user-node"

      2. Verify that the labels are added to the MachineSet object by using the oc edit command:

        For example: 

        $ oc edit MachineSet abc612-msrtw-worker-us-east-1c -n openshift-machine-api

        Example MachineSet object

        apiVersion: machine.openshift.io/v1beta1
kind: MachineSet
  ...
spec:
  ...
  template:
    metadata:
  ...
    spec:
      metadata:
        labels:
          region: east
          type: user-node
  ...

      3. Redeploy the nodes associated with that machine set by scaling down to 0 and scaling up the nodes:

        For example: 

        $ oc scale --replicas=0 MachineSet ci-ln-l8nry52-f76d1-hl7m7-worker-c -n openshift-machine-api
        $ oc scale --replicas=1 MachineSet ci-ln-l8nry52-f76d1-hl7m7-worker-c -n openshift-machine-api

      4. When the nodes are ready and available, verify that the label is added to the nodes by using the oc get command: 

        $ oc get nodes -l <key>=<value>

        For example: 

        $ oc get nodes -l type=user-node

        Example output

        NAME                                       STATUS   ROLES    AGE   VERSION
ci-ln-l8nry52-f76d1-hl7m7-worker-c-vmqzp   Ready    worker   61s   v1.24.0

    • Add labels directly to a node:

      1. Edit the Node object for the node: 

        $ oc label nodes <name> <key>=<value>

        For example, to label a node: 

        $ oc label nodes ci-ln-l8nry52-f76d1-hl7m7-worker-b-tgq49 type=user-node region=east
        Tip

        You can alternatively apply the following YAML to add labels to a node: 

        kind: Node
apiVersion: v1
metadata:
  name: <node_name>
  labels:
    type: "user-node"
    region: "east"

      2. Verify that the labels are added to the node using the oc get command: 

        $ oc get nodes -l <key>=<value>,<key>=<value>

        For example: 

        $ oc get nodes -l type=user-node,region=east

        Example output

        NAME                                       STATUS   ROLES    AGE   VERSION
ci-ln-l8nry52-f76d1-hl7m7-worker-b-tgq49   Ready    worker   17m   v1.24.0

4.7.4. Creating project-wide node selectors

You can use node selectors in a project together with labels on nodes to constrain all pods created in that project to the labeled nodes.

When you create a pod in this project, OpenShift Container Platform adds the node selectors to the pods in the project and schedules the pods on a node with matching labels in the project. If there is a cluster-wide default node selector, a project node selector takes preference.

You add node selectors to a project by editing the Namespace object to add the openshift.io/node-selector parameter. You add labels to a node, a machine set, or a machine config. Adding the label to the machine set ensures that if the node or machine goes down, new nodes have the label. Labels added to a node or machine config do not persist if the node or machine goes down.

A pod is not scheduled if the Pod object contains a node selector, but no project has a matching node selector. When you create a pod from that spec, you receive an error similar to the following message:

Example error message

Error from server (Forbidden): error when creating "pod.yaml": pods "pod-4" is forbidden: pod node label selector conflicts with its project node label selector

Note

You can add additional key/value pairs to a pod. But you cannot add a different value for a project key.

Procedure

To add a default project node selector:

  1. Create a namespace or edit an existing namespace to add the openshift.io/node-selector parameter: 

    $ oc edit namespace <name>

    Example output

    apiVersion: v1
kind: Namespace
metadata:
  annotations:
    openshift.io/node-selector: "type=user-node,region=east" 1
    openshift.io/description: ""
    openshift.io/display-name: ""
    openshift.io/requester: kube:admin
    openshift.io/sa.scc.mcs: s0:c30,c5
    openshift.io/sa.scc.supplemental-groups: 1000880000/10000
    openshift.io/sa.scc.uid-range: 1000880000/10000
  creationTimestamp: "2021-05-10T12:35:04Z"
  labels:
    kubernetes.io/metadata.name: demo
  name: demo
  resourceVersion: "145537"
  uid: 3f8786e3-1fcb-42e3-a0e3-e2ac54d15001
spec:
  finalizers:
  - kubernetes

    1
    Add the openshift.io/node-selector with the appropriate <key>:<value> pairs.

  2. Add labels to a node by using a machine set or editing the node directly:

    • Use a MachineSet object to add labels to nodes managed by the machine set when a node is created:

      1. Run the following command to add labels to a MachineSet object: 

        $ oc patch MachineSet <name> --type='json' -p='[{"op":"add","path":"/spec/template/spec/metadata/labels", "value":{"<key>"="<value>","<key>"="<value>"}}]'  -n openshift-machine-api

        For example: 

        $ oc patch MachineSet ci-ln-l8nry52-f76d1-hl7m7-worker-c --type='json' -p='[{"op":"add","path":"/spec/template/spec/metadata/labels", "value":{"type":"user-node","region":"east"}}]'  -n openshift-machine-api
        Tip

        You can alternatively apply the following YAML to add labels to a machine set: 

        apiVersion: machine.openshift.io/v1beta1
kind: MachineSet
metadata:
  name: <machineset>
  namespace: openshift-machine-api
spec:
  template:
    spec:
      metadata:
        labels:
          region: "east"
          type: "user-node"

      2. Verify that the labels are added to the MachineSet object by using the oc edit command:

        For example: 

        $ oc edit MachineSet ci-ln-l8nry52-f76d1-hl7m7-worker-c -n openshift-machine-api

        Example output

        apiVersion: machine.openshift.io/v1beta1
kind: MachineSet
metadata:
...
spec:
...
  template:
    metadata:
...
    spec:
      metadata:
        labels:
          region: east
          type: user-node

      3. Redeploy the nodes associated with that machine set:

        For example: 

        $ oc scale --replicas=0 MachineSet ci-ln-l8nry52-f76d1-hl7m7-worker-c -n openshift-machine-api
        $ oc scale --replicas=1 MachineSet ci-ln-l8nry52-f76d1-hl7m7-worker-c -n openshift-machine-api

      4. When the nodes are ready and available, verify that the label is added to the nodes by using the oc get command: 

        $ oc get nodes -l <key>=<value>

        For example: 

        $ oc get nodes -l type=user-node,region=east

        Example output

        NAME                                       STATUS   ROLES    AGE   VERSION
ci-ln-l8nry52-f76d1-hl7m7-worker-c-vmqzp   Ready    worker   61s   v1.24.0

    • Add labels directly to a node:

      1. Edit the Node object to add labels: 

        $ oc label <resource> <name> <key>=<value>

        For example, to label a node: 

        $ oc label nodes ci-ln-l8nry52-f76d1-hl7m7-worker-c-tgq49 type=user-node region=east
        Tip

        You can alternatively apply the following YAML to add labels to a node: 

        kind: Node
apiVersion: v1
metadata:
  name: <node_name>
  labels:
    type: "user-node"
    region: "east"

      2. Verify that the labels are added to the Node object using the oc get command: 

        $ oc get nodes -l <key>=<value>

        For example: 

        $ oc get nodes -l type=user-node,region=east

        Example output

        NAME                                       STATUS   ROLES    AGE   VERSION
ci-ln-l8nry52-f76d1-hl7m7-worker-b-tgq49   Ready    worker   17m   v1.24.0

Additional resources

4.8. Controlling pod placement by using pod topology spread constraints

You can use pod topology spread constraints to control the placement of your pods across nodes, zones, regions, or other user-defined topology domains.

4.8.1. About pod topology spread constraints

By using a pod topology spread constraint, you provide fine-grained control over the distribution of pods across failure domains to help achieve high availability and more efficient resource utilization.

OpenShift Container Platform administrators can label nodes to provide topology information, such as regions, zones, nodes, or other user-defined domains. After these labels are set on nodes, users can then define pod topology spread constraints to control the placement of pods across these topology domains.

You specify which pods to group together, which topology domains they are spread among, and the acceptable skew. Only pods within the same namespace are matched and grouped together when spreading due to a constraint.

4.8.2. Configuring pod topology spread constraints

The following steps demonstrate how to configure pod topology spread constraints to distribute pods that match the specified labels based on their zone.

You can specify multiple pod topology spread constraints, but you must ensure that they do not conflict with each other. All pod topology spread constraints must be satisfied for a pod to be placed.

Prerequisites

  • A cluster administrator has added the required labels to nodes.

Procedure

  1. Create a Pod spec and specify a pod topology spread constraint:

    Example pod-spec.yaml file

    apiVersion: v1
kind: Pod
metadata:
  name: my-pod
  labels:
    region: us-east
spec:
  topologySpreadConstraints:
  - maxSkew: 1 1
    topologyKey: topology.kubernetes.io/zone 2
    whenUnsatisfiable: DoNotSchedule 3
    labelSelector: 4
      matchLabels:
        region: us-east 5
  containers:
  - image: "docker.io/ocpqe/hello-pod"
    name: hello-pod

    1
    The maximum difference in number of pods between any two topology domains. The default is 1, and you cannot specify a value of 0.
    2
    The key of a node label. Nodes with this key and identical value are considered to be in the same topology.
    3
    How to handle a pod if it does not satisfy the spread constraint. The default is DoNotSchedule, which tells the scheduler not to schedule the pod. Set to ScheduleAnyway to still schedule the pod, but the scheduler prioritizes honoring the skew to not make the cluster more imbalanced.
    4
    Pods that match this label selector are counted and recognized as a group when spreading to satisfy the constraint. Be sure to specify a label selector, otherwise no pods can be matched.
    5
    Be sure that this Pod spec also sets its labels to match this label selector if you want it to be counted properly in the future.

  2. Create the pod: 

    $ oc create -f pod-spec.yaml

4.8.3. Example pod topology spread constraints

The following examples demonstrate pod topology spread constraint configurations.

4.8.3.1. Single pod topology spread constraint example

This example Pod spec defines one pod topology spread constraint. It matches on pods labeled region: us-east, distributes among zones, specifies a skew of 1, and does not schedule the pod if it does not meet these requirements. 

kind: Pod
apiVersion: v1
metadata:
  name: my-pod
  labels:
    region: us-east
spec:
  topologySpreadConstraints:
  - maxSkew: 1
    topologyKey: topology.kubernetes.io/zone
    whenUnsatisfiable: DoNotSchedule
    labelSelector:
      matchLabels:
        region: us-east
  containers:
  - image: "docker.io/ocpqe/hello-pod"
    name: hello-pod

4.8.3.2. Multiple pod topology spread constraints example

This example Pod spec defines two pod topology spread constraints. Both match on pods labeled region: us-east, specify a skew of 1, and do not schedule the pod if it does not meet these requirements.

The first constraint distributes pods based on a user-defined label node, and the second constraint distributes pods based on a user-defined label rack. Both constraints must be met for the pod to be scheduled. 

kind: Pod
apiVersion: v1
metadata:
  name: my-pod-2
  labels:
    region: us-east
spec:
  topologySpreadConstraints:
  - maxSkew: 1
    topologyKey: node
    whenUnsatisfiable: DoNotSchedule
    labelSelector:
      matchLabels:
        region: us-east
  - maxSkew: 1
    topologyKey: rack
    whenUnsatisfiable: DoNotSchedule
    labelSelector:
      matchLabels:
        region: us-east
  containers:
  - image: "docker.io/ocpqe/hello-pod"
    name: hello-pod

4.8.4. Additional resources

4.9. Evicting pods using the descheduler

While the scheduler is used to determine the most suitable node to host a new pod, the descheduler can be used to evict a running pod so that the pod can be rescheduled onto a more suitable node.

4.9.1. About the descheduler

You can use the descheduler to evict pods based on specific strategies so that the pods can be rescheduled onto more appropriate nodes.

You can benefit from descheduling running pods in situations such as the following:

  • Nodes are underutilized or overutilized.
  • Pod and node affinity requirements, such as taints or labels, have changed and the original scheduling decisions are no longer appropriate for certain nodes.
  • Node failure requires pods to be moved.
  • New nodes are added to clusters.
  • Pods have been restarted too many times.
Important

The descheduler does not schedule replacement of evicted pods. The scheduler automatically performs this task for the evicted pods.

When the descheduler decides to evict pods from a node, it employs the following general mechanism:

  • Pods in the openshift-* and kube-system namespaces are never evicted.
  • Critical pods with priorityClassName set to system-cluster-critical or system-node-critical are never evicted.
  • Static, mirrored, or stand-alone pods that are not part of a replication controller, replica set, deployment, or job are never evicted because these pods will not be recreated.
  • Pods associated with daemon sets are never evicted.
  • Pods with local storage are never evicted.
  • Best effort pods are evicted before burstable and guaranteed pods.
  • All types of pods with the descheduler.alpha.kubernetes.io/evict annotation are eligible for eviction. This annotation is used to override checks that prevent eviction, and the user can select which pod is evicted. Users should know how and if the pod will be recreated.
  • Pods subject to pod disruption budget (PDB) are not evicted if descheduling violates its pod disruption budget (PDB). The pods are evicted by using eviction subresource to handle PDB.

4.9.2. Descheduler profiles

The following descheduler profiles are available:

AffinityAndTaints

This profile evicts pods that violate inter-pod anti-affinity, node affinity, and node taints.

It enables the following strategies:

  • RemovePodsViolatingInterPodAntiAffinity: removes pods that are violating inter-pod anti-affinity.
  • RemovePodsViolatingNodeAffinity: removes pods that are violating node affinity.

  • RemovePodsViolatingNodeTaints: removes pods that are violating NoSchedule taints on nodes.

    Pods with a node affinity type of requiredDuringSchedulingIgnoredDuringExecution are removed.

TopologyAndDuplicates

This profile evicts pods in an effort to evenly spread similar pods, or pods of the same topology domain, among nodes.

It enables the following strategies:

  • RemovePodsViolatingTopologySpreadConstraint: finds unbalanced topology domains and tries to evict pods from larger ones when DoNotSchedule constraints are violated.
  • RemoveDuplicates: ensures that there is only one pod associated with a replica set, replication controller, deployment, or job running on same node. If there are more, those duplicate pods are evicted for better pod distribution in a cluster.
LifecycleAndUtilization

This profile evicts long-running pods and balances resource usage between nodes.

It enables the following strategies:

  • RemovePodsHavingTooManyRestarts: removes pods whose containers have been restarted too many times.

    Pods where the sum of restarts over all containers (including Init Containers) is more than 100.

  • LowNodeUtilization: finds nodes that are underutilized and evicts pods, if possible, from overutilized nodes in the hope that recreation of evicted pods will be scheduled on these underutilized nodes.

    A node is considered underutilized if its usage is below 20% for all thresholds (CPU, memory, and number of pods).

    A node is considered overutilized if its usage is above 50% for any of the thresholds (CPU, memory, and number of pods).

  • PodLifeTime: evicts pods that are too old.

    By default, pods that are older than 24 hours are removed. You can customize the pod lifetime value.

SoftTopologyAndDuplicates

This profile is the same as TopologyAndDuplicates, except that pods with soft topology constraints, such as whenUnsatisfiable: ScheduleAnyway, are also considered for eviction.

Note

Do not enable both SoftTopologyAndDuplicates and TopologyAndDuplicates. Enabling both results in a conflict.

EvictPodsWithLocalStorage
This profile allows pods with local storage to be eligible for eviction.
EvictPodsWithPVC
This profile allows pods with persistent volume claims to be eligible for eviction.

4.9.3. Installing the descheduler

The descheduler is not available by default. To enable the descheduler, you must install the Kube Descheduler Operator from OperatorHub and enable one or more descheduler profiles.

By default, the descheduler runs in predictive mode, which means that it only simulates pod evictions. You must change the mode to automatic for the descheduler to perform the pod evictions.

Important

If you have enabled hosted control planes in your cluster, set a custom priority threshold to lower the chance that pods in the hosted control plane namespaces are evicted. Set the priority threshold class name to hypershift-control-plane, because it has the lowest priority value (100000000) of the hosted control plane priority classes.

Prerequisites

  • Cluster administrator privileges.
  • Access to the OpenShift Container Platform web console.

Procedure

  1. Log in to the OpenShift Container Platform web console.

  2. Create the required namespace for the Kube Descheduler Operator.

    1. Navigate to Administration → Namespaces and click Create Namespace.
    2. Enter openshift-kube-descheduler-operator in the Name field, enter openshift.io/cluster-monitoring=true in the Labels field to enable descheduler metrics, and click Create.

  3. Install the Kube Descheduler Operator.

    1. Navigate to Operators → OperatorHub.
    2. Type Kube Descheduler Operator into the filter box.
    3. Select the Kube Descheduler Operator and click Install.
    4. On the Install Operator page, select A specific namespace on the cluster. Select openshift-kube-descheduler-operator from the drop-down menu.
    5. Adjust the values for the Update Channel and Approval Strategy to the desired values.
    6. Click Install.

  4. Create a descheduler instance.

    1. From the Operators → Installed Operators page, click the Kube Descheduler Operator.
    2. Select the Kube Descheduler tab and click Create KubeDescheduler.

    3. Edit the settings as necessary.

      1. To evict pods instead of simulating the evictions, change the Mode field to Automatic.

      2. Expand the Profiles section to select one or more profiles to enable. The AffinityAndTaints profile is enabled by default. Click Add Profile to select additional profiles.

        Note

        Do not enable both TopologyAndDuplicates and SoftTopologyAndDuplicates. Enabling both results in a conflict.

      3. Optional: Expand the Profile Customizations section to set optional configurations for the descheduler.

        • Set a custom pod lifetime value for the LifecycleAndUtilization profile. Use the podLifetime field to set a numerical value and a valid unit (s, m, or h). The default pod lifetime is 24 hours (24h).

        • Set a custom priority threshold to consider pods for eviction only if their priority is lower than a specified priority level. Use the thresholdPriority field to set a numerical priority threshold or use the thresholdPriorityClassName field to specify a certain priority class name.

          Note

          Do not specify both thresholdPriority and thresholdPriorityClassName for the descheduler.

        • Set specific namespaces to exclude or include from descheduler operations. Expand the namespaces field and add namespaces to the excluded or included list. You can only either set a list of namespaces to exclude or a list of namespaces to include. Note that protected namespaces (openshift-*, kube-system, hypershift) are excluded by default.

          Important

          The LowNodeUtilization strategy does not support namespace exclusion. If the LifecycleAndUtilization profile is set, which enables the LowNodeUtilization strategy, then no namespaces are excluded, even the protected namespaces. To avoid evictions from the protected namespaces while the LowNodeUtilization strategy is enabled, set the priority class name to system-cluster-critical or system-node-critical.

        • Experimental: Set thresholds for underutilization and overutilization for the LowNodeUtilization strategy. Use the devLowNodeUtilizationThresholds field to set one of the following values:

          • Low: 10% underutilized and 30% overutilized
          • Medium: 20% underutilized and 50% overutilized (Default)
          • High: 40% underutilized and 70% overutilized
          Note

          This setting is experimental and should not be used in a production environment.

      4. Optional: Use the Descheduling Interval Seconds field to change the number of seconds between descheduler runs. The default is 3600 seconds.
    4. Click Create.

You can also configure the profiles and settings for the descheduler later using the OpenShift CLI (oc). If you did not adjust the profiles when creating the descheduler instance from the web console, the AffinityAndTaints profile is enabled by default.

4.9.4. Configuring descheduler profiles

You can configure which profiles the descheduler uses to evict pods.

Prerequisites

  • Cluster administrator privileges

Procedure

  1. Edit the KubeDescheduler object: 

    $ oc edit kubedeschedulers.operator.openshift.io cluster -n openshift-kube-descheduler-operator

  2. Specify one or more profiles in the spec.profiles section. 

    apiVersion: operator.openshift.io/v1
kind: KubeDescheduler
metadata:
  name: cluster
  namespace: openshift-kube-descheduler-operator
spec:
  deschedulingIntervalSeconds: 3600
  logLevel: Normal
  managementState: Managed
  operatorLogLevel: Normal
  mode: Predictive                                     1
  profileCustomizations:
    namespaces:                                        2
      excluded:
      - my-namespace
    podLifetime: 48h                                   3
    thresholdPriorityClassName: my-priority-class-name 4
  profiles:                                            5
  - AffinityAndTaints
  - TopologyAndDuplicates                              6
  - LifecycleAndUtilization
  - EvictPodsWithLocalStorage
  - EvictPodsWithPVC
    1
    Optional: By default, the descheduler does not evict pods. To evict pods, set mode to Automatic.
    2
    Optional: Set a list of user-created namespaces to include or exclude from descheduler operations. Use excluded to set a list of namespaces to exclude or use included to set a list of namespaces to include. Note that protected namespaces (openshift-*, kube-system, hypershift) are excluded by default.
    Important

    The LowNodeUtilization strategy does not support namespace exclusion. If the LifecycleAndUtilization profile is set, which enables the LowNodeUtilization strategy, then no namespaces are excluded, even the protected namespaces. To avoid evictions from the protected namespaces while the LowNodeUtilization strategy is enabled, set the priority class name to system-cluster-critical or system-node-critical.

    3
    Optional: Enable a custom pod lifetime value for the LifecycleAndUtilization profile. Valid units are s, m, or h. The default pod lifetime is 24 hours.
    4
    Optional: Specify a priority threshold to consider pods for eviction only if their priority is lower than the specified level. Use the thresholdPriority field to set a numerical priority threshold (for example, 10000) or use the thresholdPriorityClassName field to specify a certain priority class name (for example, my-priority-class-name). If you specify a priority class name, it must already exist or the descheduler will throw an error. Do not set both thresholdPriority and thresholdPriorityClassName.
    5
    Add one or more profiles to enable. Available profiles: AffinityAndTaints, TopologyAndDuplicates, LifecycleAndUtilization, SoftTopologyAndDuplicates, EvictPodsWithLocalStorage, and EvictPodsWithPVC.
    6
    Do not enable both TopologyAndDuplicates and SoftTopologyAndDuplicates. Enabling both results in a conflict.

    You can enable multiple profiles; the order that the profiles are specified in is not important.

  3. Save the file to apply the changes.

4.9.5. Configuring the descheduler interval

You can configure the amount of time between descheduler runs. The default is 3600 seconds (one hour).

Prerequisites

  • Cluster administrator privileges

Procedure

  1. Edit the KubeDescheduler object: 

    $ oc edit kubedeschedulers.operator.openshift.io cluster -n openshift-kube-descheduler-operator

  2. Update the deschedulingIntervalSeconds field to the desired value: 

    apiVersion: operator.openshift.io/v1
kind: KubeDescheduler
metadata:
  name: cluster
  namespace: openshift-kube-descheduler-operator
spec:
  deschedulingIntervalSeconds: 3600 1
...
    1
    Set the number of seconds between descheduler runs. A value of 0 in this field runs the descheduler once and exits.
  3. Save the file to apply the changes.

4.9.6. Uninstalling the descheduler

You can remove the descheduler from your cluster by removing the descheduler instance and uninstalling the Kube Descheduler Operator. This procedure also cleans up the KubeDescheduler CRD and openshift-kube-descheduler-operator namespace.

Prerequisites

  • Cluster administrator privileges.
  • Access to the OpenShift Container Platform web console.

Procedure

  1. Log in to the OpenShift Container Platform web console.

  2. Delete the descheduler instance.

    1. From the Operators → Installed Operators page, click Kube Descheduler Operator.
    2. Select the Kube Descheduler tab.
    3. Click the Options menu kebab next to the cluster entry and select Delete KubeDescheduler.
    4. In the confirmation dialog, click Delete.

  3. Uninstall the Kube Descheduler Operator.

    1. Navigate to Operators → Installed Operators.
    2. Click the Options menu kebab next to the Kube Descheduler Operator entry and select Uninstall Operator.
    3. In the confirmation dialog, click Uninstall.

  4. Delete the openshift-kube-descheduler-operator namespace.

    1. Navigate to Administration → Namespaces.
    2. Enter openshift-kube-descheduler-operator into the filter box.
    3. Click the Options menu kebab next to the openshift-kube-descheduler-operator entry and select Delete Namespace.
    4. In the confirmation dialog, enter openshift-kube-descheduler-operator and click Delete.

  5. Delete the KubeDescheduler CRD.

    1. Navigate to Administration → Custom Resource Definitions.
    2. Enter KubeDescheduler into the filter box.
    3. Click the Options menu kebab next to the KubeDescheduler entry and select Delete CustomResourceDefinition.
    4. In the confirmation dialog, click Delete.

4.10. Secondary scheduler

4.10.1. Secondary scheduler overview

You can install the Secondary Scheduler Operator to run a custom secondary scheduler alongside the default scheduler to schedule pods.

4.10.1.1. About the Secondary Scheduler Operator

The Secondary Scheduler Operator for Red Hat OpenShift provides a way to deploy a custom secondary scheduler in OpenShift Container Platform. The secondary scheduler runs alongside the default scheduler to schedule pods. Pod configurations can specify which scheduler to use.

The custom scheduler must have the /bin/kube-scheduler binary and be based on the Kubernetes scheduling framework.

Important

You can use the Secondary Scheduler Operator to deploy a custom secondary scheduler in OpenShift Container Platform, but Red Hat does not directly support the functionality of the custom secondary scheduler.

The Secondary Scheduler Operator creates the default roles and role bindings required by the secondary scheduler. You can specify which scheduling plugins to enable or disable by configuring the KubeSchedulerConfiguration resource for the secondary scheduler.

4.10.2. Secondary Scheduler Operator for Red Hat OpenShift release notes

The Secondary Scheduler Operator for Red Hat OpenShift allows you to deploy a custom secondary scheduler in your OpenShift Container Platform cluster.

These release notes track the development of the Secondary Scheduler Operator for Red Hat OpenShift.

For more information, see About the Secondary Scheduler Operator.

4.10.2.1. Release notes for Secondary Scheduler Operator for Red Hat OpenShift 1.1.0

Issued: 2022-9-1

The following advisory is available for the Secondary Scheduler Operator for Red Hat OpenShift 1.1.0:

4.10.2.1.1. New features and enhancements
4.10.2.1.2. Known issues
  • Currently, you cannot deploy additional resources, such as config maps, CRDs, or RBAC policies through the Secondary Scheduler Operator. Any resources other than roles and role bindings that are required by your custom secondary scheduler must be applied externally. (BZ#2071684)

4.10.2.2. Release notes for Secondary Scheduler Operator for Red Hat OpenShift 1.0.1

Issued: 2022-07-28

The following advisory is available for the Secondary Scheduler Operator for Red Hat OpenShift 1.0.1:

4.10.2.2.1. New features and enhancements
  • The maximum OpenShift Container Platform version for Secondary Scheduler Operator for Red Hat OpenShift 1.0.1 is 4.11.
4.10.2.2.2. Bug fixes
  • Previously, the secondary scheduler deployment was not deleted after the secondary scheduler custom resource (CR) was deleted, which prevented the Secondary Scheduler Operator and operand from being fully uninstalled. The secondary scheduler deployment is now deleted when the secondary scheduler CR is deleted, so that the Secondary Scheduler Operator can now be fully uninstalled. (BZ#2100923)
4.10.2.2.3. Known issues
  • Currently, you cannot deploy additional resources, such as config maps, CRDs, or RBAC policies through the Secondary Scheduler Operator. Any resources other than roles and role bindings that are required by your custom secondary scheduler must be applied externally. (BZ#2071684)

4.10.2.3. Release notes for Secondary Scheduler Operator for Red Hat OpenShift 1.0.0

Issued: 2022-04-18

The following advisory is available for the Secondary Scheduler Operator for Red Hat OpenShift 1.0.0:

4.10.2.3.1. New features and enhancements
  • This is the initial release of the Secondary Scheduler Operator for Red Hat OpenShift.
4.10.2.3.2. Known issues
  • Currently, you cannot deploy additional resources, such as config maps, CRDs, or RBAC policies through the Secondary Scheduler Operator. Any resources other than roles and role bindings that are required by your custom secondary scheduler must be applied externally. (BZ#2071684)

4.10.3. Scheduling pods using a secondary scheduler

You can run a custom secondary scheduler in OpenShift Container Platform by installing the Secondary Scheduler Operator, deploying the secondary scheduler, and setting the secondary scheduler in the pod definition.

4.10.3.1. Installing the Secondary Scheduler Operator

You can use the web console to install the Secondary Scheduler Operator for Red Hat OpenShift.

Prerequisites

  • You have access to the cluster with cluster-admin privileges.
  • You have access to the OpenShift Container Platform web console.

Procedure

  1. Log in to the OpenShift Container Platform web console.

  2. Create the required namespace for the Secondary Scheduler Operator for Red Hat OpenShift.

    1. Navigate to Administration → Namespaces and click Create Namespace.
    2. Enter openshift-secondary-scheduler-operator in the Name field and click Create.

  3. Install the Secondary Scheduler Operator for Red Hat OpenShift.

    1. Navigate to Operators → OperatorHub.
    2. Enter Secondary Scheduler Operator for Red Hat OpenShift into the filter box.
    3. Select the Secondary Scheduler Operator for Red Hat OpenShift and click Install.

    4. On the Install Operator page:

      1. The Update channel is set to stable, which installs the latest stable release of the Secondary Scheduler Operator for Red Hat OpenShift.
      2. Select A specific namespace on the cluster and select openshift-secondary-scheduler-operator from the drop-down menu.

      3. Select an Update approval strategy.

        • The Automatic strategy allows Operator Lifecycle Manager (OLM) to automatically update the Operator when a new version is available.
        • The Manual strategy requires a user with appropriate credentials to approve the Operator update.
      4. Click Install.

Verification

  1. Navigate to Operators → Installed Operators.
  2. Verify that Secondary Scheduler Operator for Red Hat OpenShift is listed with a Status of Succeeded.

4.10.3.2. Deploying a secondary scheduler

After you have installed the Secondary Scheduler Operator, you can deploy a secondary scheduler.

Prerequisities

  • You have access to the cluster with cluster-admin privileges.
  • You have access to the OpenShift Container Platform web console.
  • The Secondary Scheduler Operator for Red Hat OpenShift is installed.

Procedure

  1. Log in to the OpenShift Container Platform web console.

  2. Create config map to hold the configuration for the secondary scheduler.

    1. Navigate to Workloads → ConfigMaps.
    2. Click Create ConfigMap.

    3. In the YAML editor, enter the config map definition that contains the necessary KubeSchedulerConfiguration configuration. For example: 

      apiVersion: v1
kind: ConfigMap
metadata:
  name: "secondary-scheduler-config"                  1
  namespace: "openshift-secondary-scheduler-operator" 2
data:
  "config.yaml": |
    apiVersion: kubescheduler.config.k8s.io/v1beta3
    kind: KubeSchedulerConfiguration                  3
    leaderElection:
      leaderElect: false
    profiles:
      - schedulerName: secondary-scheduler            4
        plugins:                                      5
          score:
            disabled:
              - name: NodeResourcesBalancedAllocation
              - name: NodeResourcesLeastAllocated
      1
      The name of the config map. This is used in the Scheduler Config field when creating the SecondaryScheduler CR.
      2
      The config map must be created in the openshift-secondary-scheduler-operator namespace.
      3
      The KubeSchedulerConfiguration resource for the secondary scheduler. For more information, see KubeSchedulerConfiguration in the Kubernetes API documentation.
      4
      The name of the secondary scheduler. Pods that set their spec.schedulerName field to this value are scheduled with this secondary scheduler.
      5
      The plugins to enable or disable for the secondary scheduler. For a list default scheduling plugins, see Scheduling plugins in the Kubernetes documentation.
    4. Click Create.

  3. Create the SecondaryScheduler CR:

    1. Navigate to Operators → Installed Operators.
    2. Select Secondary Scheduler Operator for Red Hat OpenShift.
    3. Select the Secondary Scheduler tab and click Create SecondaryScheduler.
    4. The Name field defaults to cluster; do not change this name.
    5. The Scheduler Config field defaults to secondary-scheduler-config. Ensure that this value matches the name of the config map created earlier in this procedure.

    6. In the Scheduler Image field, enter the image name for your custom scheduler.

      Important

      Red Hat does not directly support the functionality of your custom secondary scheduler.

    7. Click Create.

4.10.3.3. Scheduling a pod using the secondary scheduler

To schedule a pod using the secondary scheduler, set the schedulerName field in the pod definition.

Prerequisities

  • You have access to the cluster with cluster-admin privileges.
  • You have access to the OpenShift Container Platform web console.
  • The Secondary Scheduler Operator for Red Hat OpenShift is installed.
  • A secondary scheduler is configured.

Procedure

  1. Log in to the OpenShift Container Platform web console.
  2. Navigate to Workloads → Pods.
  3. Click Create Pod.

  4. In the YAML editor, enter the desired pod configuration and add the schedulerName field: 

    apiVersion: v1
kind: Pod
metadata:
  name: nginx
  namespace: default
spec:
  containers:
    - name: nginx
      image: nginx:1.14.2
      ports:
        - containerPort: 80
  schedulerName: secondary-scheduler 1
    1
    The schedulerName field must match the name that is defined in the config map when you configured the secondary scheduler.
  5. Click Create.

Verification

  1. Log in to the OpenShift CLI.

  2. Describe the pod using the following command: 

    $ oc describe pod nginx -n default

    Example output

    Name:         nginx
Namespace:    default
Priority:     0
Node:         ci-ln-t0w4r1k-72292-xkqs4-worker-b-xqkxp/10.0.128.3
...
Events:
  Type    Reason          Age   From                 Message
  ----    ------          ----  ----                 -------
  Normal  Scheduled       12s   secondary-scheduler  Successfully assigned default/nginx to ci-ln-t0w4r1k-72292-xkqs4-worker-b-xqkxp
...

  3. In the events table, find the event with a message similar to Successfully assigned <namespace>/<pod_name> to <node_name>.

  4. In the "From" column, verify that the event was generated from the secondary scheduler and not the default scheduler.

    Note

    You can also check the secondary-scheduler-* pod logs in the openshift-secondary-scheduler-namespace to verify that the pod was scheduled by the secondary scheduler.

4.10.4. Uninstalling the Secondary Scheduler Operator

You can remove the Secondary Scheduler Operator for Red Hat OpenShift from OpenShift Container Platform by uninstalling the Operator and removing its related resources.

4.10.4.1. Uninstalling the Secondary Scheduler Operator

You can uninstall the Secondary Scheduler Operator for Red Hat OpenShift by using the web console.

Prerequisites

  • You have access to the cluster with cluster-admin privileges.
  • You have access to the OpenShift Container Platform web console.
  • The Secondary Scheduler Operator for Red Hat OpenShift is installed.

Procedure

  1. Log in to the OpenShift Container Platform web console.

  2. Uninstall the Secondary Scheduler Operator for Red Hat OpenShift Operator.

    1. Navigate to Operators → Installed Operators.
    2. Click the Options menu kebab next to the Secondary Scheduler Operator entry and click Uninstall Operator.
    3. In the confirmation dialog, click Uninstall.

4.10.4.2. Removing Secondary Scheduler Operator resources

Optionally, after uninstalling the Secondary Scheduler Operator for Red Hat OpenShift, you can remove its related resources from your cluster.

Prerequisites

  • You have access to the cluster with cluster-admin privileges.
  • You have access to the OpenShift Container Platform web console.

Procedure

  1. Log in to the OpenShift Container Platform web console.

  2. Remove CRDs that were installed by the Secondary Scheduler Operator:

    1. Navigate to Administration → CustomResourceDefinitions.
    2. Enter SecondaryScheduler in the Name field to filter the CRDs.
    3. Click the Options menu kebab next to the SecondaryScheduler CRD and select Delete Custom Resource Definition:

  3. Remove the openshift-secondary-scheduler-operator namespace.

    1. Navigate to Administration → Namespaces.
    2. Click the Options menu kebab next to the openshift-secondary-scheduler-operator and select Delete Namespace.
    3. In the confirmation dialog, enter openshift-secondary-scheduler-operator in the field and click Delete.