How to set node selectors, tolerations, resources on the agent pods?

[emmahsax]

Hello. I am trying to install litmus with Terraform by calling the helm chart like the following:

resource "helm_release" "litmus" {
  chart      = "litmus"
  name       = "litmus"
  namespace  = "litmus"
  repository = "https://litmuschaos.github.io/litmus-helm/"
  version    = "3.1.0"
  values                     = [
    <<-EOT
        mongodb:
          auth:
            replicaSetKey: "..."
            rootPassword: "..."
          nodeSelector:
            "mywebsite.com/nodegroup": "nodegroupname"
          resources:
            "limits":
              "cpu": "512m"
              "memory": "512Mi"
            "requests":
              "cpu": "256m"
              "memory": "256Mi"
          tolerations:
            - "effect": "NoSchedule"
              "key": "mywebsite.com/nodegroup"
              "operator": "Equal"
              "value": "nodegroupname"

          arbiter:
            nodeSelector:
              "mywebsite.com/nodegroup": "nodegroupname"
            resources:
              "limits":
                "cpu": "256m"
                "memory": "256Mi"
              "requests":
                "cpu": "100m"
                "memory": "128Mi"
            tolerations:
              - "effect": "NoSchedule"
                "key": "mywebsite.com/nodegroup"
                "operator": "Equal"
                "value": "nodegroupname"

        portal:
          frontend:
            nodeSelector:
              "mywebsite.com/nodegroup": "nodegroupname"
            resources:
              "limits":
                "cpu": "256m"
                "memory": "256Mi"
              "requests":
                "cpu": "100m"
                "memory": "128Mi"
            tolerations:
              - "effect": "NoSchedule"
                "key": "mywebsite.com/nodegroup"
                "operator": "Equal"
                "value": "nodegroupname"

          server:
            authServer:
              resources:
                "limits":
                  "cpu": "256m"
                  "memory": "256Mi"
                "requests":
                  "cpu": "100m"
                  "memory": "128Mi"
            nodeSelector:
              "mywebsite.com/nodegroup": "nodegroupname"
            tolerations:
              - "effect": "NoSchedule"
                "key": "mywebsite.com/nodegroup"
                "operator": "Equal"
                "value": "nodegroupname"

        upgradeAgent:
          controlPlane:
            image:
              repository: litmuschaos/upgrade-agent-cp
              tag: 2.9.0
          nodeSelector:
            "mywebsite.com/nodegroup": "nodegroupname"
          resources:
            "limits":
              "cpu": "256m"
              "memory": "256Mi"
            "requests":
              "cpu": "100m"
              "memory": "128Mi"
          tolerations:
            - "effect": "NoSchedule"
              "key": "mywebsite.com/nodegroup"
              "operator": "Equal"
              "value": "nodegroupname"
    EOT,
  ]
}

But none of this configures the node selectors, tolerations, and resources on:

• chaos-exporter
• chaos-operator
• event-tracker
• subscriber
• workflow-controller

Since https://github.com/litmuschaos/litmus/pull/3527 was merged, I feel like we should be able to set them. But maybe the helm charts haven't been updated with this capability?

Any guidance anyone can provide would be greatly appreciated.

And yes, I did look at https://github.com/litmuschaos/litmus-helm/issues/163

[uditgaurav]

Hi @emmahsax , This should be fix by including this issue with the revamp of litmus-helm.