CODEWITHSUNDEEP
kuberneteskubernetes-deployment
Siraxis
Siraxis

Reputation: 41

Kubernetes Affinity Weight Issue

my use-case is to schedule a deployment only in fsn1 region nodes, unless there's an emergency and nodes of fsn1 go down, then only in that case I want to schedule as a fallback on hel1 nodes.

Unfortunately, with these rules, 5/26 total pods are still scheduled on hel1 nodes, even with weight: 1. All nodes have same resources and only accomodate this deployment, so it's not like other deployments occupy resources of fsn1 nodes.

affinity:
  nodeAffinity:
    preferredDuringSchedulingIgnoredDuringExecution:
      - weight: 100
        preference:
          matchExpressions:
            - key: "region"
              operator: "In"
              values:
                - "fsn1"
      - weight: 1
        preference:
          matchExpressions:
            - key: "region"
              operator: "In"
              values:
                - "hel1"

Is there another way? I want to avoid using a custom scheduler if possible.

Upvotes: 0

Views: 105

Answers (1)

Pranay Kumar Kyasala
Pranay Kumar Kyasala

Reputation: 367

Node affinity is conceptually similar to nodeSelector, allowing you to constrain which nodes your Pod can be scheduled on based on node labels. There are two types of node affinity:

  • requiredDuringSchedulingIgnoredDuringExecution: The scheduler can't schedule the Pod unless the rule is met. This functions like nodeSelector, but with a more expressive syntax.

  • preferredDuringSchedulingIgnoredDuringExecution: The scheduler tries to find a node that meets the rule. If a matching node is not available, the scheduler still schedules the Pod.

As mentioned in this other SO preferredDuringSchedulingIgnoredDuringExecution is not guaranteed. A scheduler will try to force a schedule but it is not guaranteed because of other factors.

Also go through this Document which might be helpful.

Upvotes: 1

Related Questions