Description of the problem: spec.nodeSelector is missing in MCE instance created by MCH upon installing ACM on infra nodes Release version: Operator snapshot version: - ACM 2.5.0-DOWNSTREAM-2022-03-01-06-19-12 - MCE 2.0.0-DOWNANDBACK-2022-02-22-17-26-30 OCP version: ROSA 4.9.21 Browser Info: Steps to reproduce: 1. Install ACM and MCE on infra nodes Actual results: MCE pods are not resided on infra nodes Expected results: All ACM and MCE pods are resided on infra nodes Additional info: ACM sub + mch - oc get sub -n ocm NAME PACKAGE SOURCE CHANNEL advanced-cluster-management advanced-cluster-management brew-acm-iib release-2.5 oc get sub -n ocm advanced-cluster-management -oyaml | grep spec: -A13 spec: channel: release-2.5 config: nodeSelector: node-role.kubernetes.io/infra: "" tolerations: - effect: NoSchedule key: node-role.kubernetes.io/infra operator: Exists installPlanApproval: Manual name: advanced-cluster-management source: brew-acm-iib sourceNamespace: openshift-marketplace startingCSV: advanced-cluster-management.v2.5.0 oc get mch -n ocm multiclusterhub -oyaml apiVersion: operator.open-cluster-management.io/v1 kind: MultiClusterHub metadata: annotations: installer.open-cluster-management.io/mce-subscription-spec: '{"channel": "stable-2.0","installPlanApproval": "Automatic","name": "multicluster-engine","source": "brew-mce-iib","sourceNamespace": "openshift-marketplace","startingCSV": "multicluster-engine.v2.0.0"}' name: multiclusterhub namespace: ocm spec: availabilityConfig: High enableClusterBackup: true imagePullSecret: multiclusterhub-operator-pull-secret ingress: sslCiphers: - ECDHE-ECDSA-AES256-GCM-SHA384 - ECDHE-RSA-AES256-GCM-SHA384 - ECDHE-ECDSA-AES128-GCM-SHA256 - ECDHE-RSA-AES128-GCM-SHA256 nodeSelector: node-role.kubernetes.io/infra: "" separateCertificateManagement: false MCE sub + mce - created by mch oc get sub -n multicluster-engine multicluster-engine -oyaml | grep spec: -A13 spec: channel: stable-2.0 config: nodeSelector: node-role.kubernetes.io/infra: "" tolerations: - effect: NoSchedule key: node-role.kubernetes.io/infra operator: Exists installPlanApproval: Automatic name: multicluster-engine source: brew-mce-iib sourceNamespace: openshift-marketplace startingCSV: multicluster-engine.v2.0.0 oc get mce -n multicluster-engine multiclusterengine -oyaml | grep spec: -A8 spec: availabilityConfig: High imagePullSecret: multiclusterhub-operator-pull-secret targetNamespace: multicluster-engine tolerations: - effect: NoSchedule key: node-role.kubernetes.io/infra operator: Exists
G2Bsync 1060774809 comment thuyn-581 Mon, 07 Mar 2022 14:55:02 UTC G2Bsync - Validated on 2.5.0-DOWNSTREAM-2022-03-03-03-15-27 (Final S3).
G2Bsync 1057423242 comment zkayyali812 Wed, 02 Mar 2022 21:45:12 UTC /G2Bsync Hi @thuyn-581, Fix has been merged and validated in -`2.5.0-SNAPSHOT-2022-03-02-21-08-32` and later to ensure nodeSelector is set in MCE if set in MCH
Since the problem described in this bug report should be resolved in a recent advisory, it has been closed with a resolution of ERRATA. For information on the advisory (Important: Red Hat Advanced Cluster Management 2.5 security updates, images, and bug fixes), and where to find the updated files, follow the link below. If the solution does not work for you, open a new bug report. https://access.redhat.com/errata/RHSA-2022:4956
Based on the information provided, it appears that the issue you're facing is related to the placement of MCE (Multicluster Engine) pods on infra nodes after installing ACM (Advanced Cluster Management) on the infra nodes. The MCE pods are not residing on the expected infra nodes. To ensure that MCE pods are scheduled on infra nodes, you need to verify and ensure that the necessary configurations are set correctly. Here are a few steps you can take: Verify the nodeSelector and tolerations configuration in the MCE subscription and MCE Custom Resource (CR). These configurations define the requirements and preferences for pod scheduling. Check the MCE subscription (multicluster-engine) in the multicluster-engine namespace. Ensure that the nodeSelector field is correctly set to node-role.kubernetes.io/infra: "". This ensures that MCE pods are scheduled on nodes with the infra role. Similarly, check the MCE Custom Resource (multiclusterengine) in the multicluster-engine namespace. Verify that the tolerations field includes an entry for infra nodes: - key: node-role.kubernetes.io/infra. Confirm that the ACM subscription (advanced-cluster-management) in the ocm namespace also has the nodeSelector field correctly set to node-role.kubernetes.io/infra: "". Ensure that all the necessary labels and annotations required for node selection are correctly applied to the infra nodes. You can verify the labels and annotations on the infra nodes using the oc get nodes --show-labels command. If the configurations mentioned above are correct and the MCE pods are still not scheduled on the expected infra nodes, you may need to investigate further or seek assistance from the support channels specific to ACM and MCE, such as the ACM or Red Hat support teams. They can provide guidance and help troubleshoot the issue more effectively with access to the specific environment and configuration details. https://www.ballsportspro.com/