Searching for results...

No results

Your search did not match anything from Mirantis documentation.
Check your spelling or try different keywords.

An error occurred

An error occurred while using the search.
Try your search again or contact us to let us know about it.

Newer documentation is now live.You are currently reading an older version.

Specify placement of Ceph cluster daemons

Warning

This procedure is valid for MOSK clusters that use the MiraCeph custom resource (CR), which is available since MOSK 25.2 to replace the unsupported KaaSCephCluster resource. And MiraCeph will be automatically migrated to CephDeployment in MOSK 26.1. For details, see Deprecation Notes: KaaSCephCluster API on management clusters.

For the equivalent procedure with the unsupported KaaSCephCluster CR, refer to the following section:

Specify placement of Ceph cluster daemons

If you need to configure the placement of Rook daemons on nodes, you can add extra values in the Cluster providerSpec section of the ceph-controller Helm release.

The procedures in this section describe how to specify the placement of rook-ceph-operator, rook-discover, and csi-rbdplugin.

To specify rook-ceph-operator placement:

  1. On the management cluster, edit the Cluster resource of the target MOSK cluster:

    kubectl -n <moskClusterProjectName> edit cluster

  2. Add the following parameters to the ceph-controller Helm release values:

    spec:
  providerSpec:
    value:
      helmReleases:
      - name: ceph-controller
        values:
          rookOperatorPlacement:
            affinity: <rookOperatorAffinity>
            nodeSelector: <rookOperatorNodeSelector>
            tolerations: <rookOperatorTolerations>

    • <rookOperatorAffinity> is a key-value mapping that contains a valid Kubernetes affinity specification

    • <rookOperatorNodeSelector> is a key-value mapping that contains a valid Kubernetes nodeSelector specification

    • <rookOperatorTolerations> is a list that contains valid Kubernetes toleration items

  3. Wait for some time and verify on the MOSK cluster that the changes have applied:

    kubectl -n rook-ceph get deploy rook-ceph-operator -o yaml

To specify rook-discover and csi-rbdplugin placement simultaneously:

  1. On the management cluster, edit the required Cluster resource:

    kubectl -n <moskClusterProjectName> edit cluster

  2. Add the following parameters to the ceph-controller Helm release values:

    spec:
  providerSpec:
    value:
      helmReleases:
      - name: ceph-controller
        values:
          rookExtraConfig:
            extraDaemonsetLabels: <labelSelector>

    Substitute <labelSelector> with a valid Kubernetes label selector expression to place the rook-discover and csi-rbdplugin DaemonSet pods.

  3. Wait for some time and verify on the MOSK cluster that the changes have applied:

    kubectl -n rook-ceph get ds rook-discover -o yaml
kubectl -n rook-ceph get ds csi-rbdplugin -o yaml

To specify rook-discover and csi-rbdplugin placement separately:

  1. On the management cluster, edit the required Cluster resource:

    kubectl -n <moskClusterProjectName> edit cluster

  2. If required, add the following parameters to the ceph-controller Helm release values:

    spec:
  providerSpec:
    value:
      helmReleases:
      - name: ceph-controller
        values:
          hyperconverge:
            nodeAffinity:
              csiplugin: <labelSelector1>
              rookDiscover: <labelSelector2>

    Substitute <labelSelectorX> with a valid Kubernetes label selector expression to place the rook-discover and csi-rbdplugin DaemonSet pods. For example, "role=storage-node; discover=true".

  3. Wait for some time and verify on the MOSK cluster that the changes have applied:

    kubectl -n rook-ceph get ds rook-discover -o yaml
kubectl -n rook-ceph get ds csi-rbdplugin -o yaml