Persistent Volume Claim requirements

You can use Kubernetes to manage several types of storage that are used by containers. For BMC Helix, cluster nodes must have access to block storage that is made available to Kubernetes as a PersistentVolume (PV). Claims of reserved memory are then allocated to containers as a PersistentVolumeClaim (PVC).

You can configure PersistentVolume and PersistentVolumeClaim in several access modes. The access mode dictates how many nodes or pods can mount and interact with the volume. The following access modes are available:

  • ReadWriteOnce: The volume can be mounted as read-write only by a single node. Multiple pods can use this volume if they are running on the node.
  • ReadWriteOncePod: The volume can be mounted as read-write by more than one node. Only a single Pod can make use of the volume or claim.
  • ReadOnlyMany: The volume can be mounted as read-only by more than one node.
  • ReadWriteMany: The volume can be mounted as read-write by more than one node.

BMC Helix Innovation Suite and service management application pods do not use PVC directly for any of the services installed in the BMC Helix Innovation Suite namespace. However, the pods leverage PVC from the BMC Helix Platform namespace for the following services:

  • OpenSearch—The log analytics OpenSearch instance, opensearch-logs-data, uses PVC for storing FTS data.
  • EFK—Elasticsearch use PVC for streaming logs.


Important

If you are performing a combined deployment of BMC Helix Service Management and BMC Helix IT Operations Management, see PersistentVolumeClaim (PVC) requirements in BMC Helix IT Operations Management deployment documentation.

PVC requirements for BMC Helix Platform

Service

Storage Read Write type

Storage (PVC)

Elastic - events

ReadWriteOnce

Block storage

Kafka/Zookeeper

Postgres

MinIO

Redis

VictoriaMetrics

VictoriaMetrics Aggregator

AIOps (ML Model)

Important

If you are using VMWare Tanzu to manage your Kubernetes cluster, note the following points:

  • The VMWare NFS storage class provider does not support the ReadWriteMany mode. Refer to the following link to know about the NFS storage class provider you can use instead of the out-of-the-box provider:
    https://core.vmware.com/blog/using-readwritemany-volumes-tkg-clusters

  • The default PVC has root only permissions. Ensure that you are mapping with a user who can read or write to the default PVC share, or enable no_root_squash option. For information about this, see https://serverfault.com/questions/1089557/what-does-the-no-all-squash-option-do-in-nfs-exports.

    Expand to see the sample installation
    [root@control1 nfs-tanzu-rwm-support]# more helm-add-nfs.sh
    helm repo add nfs-subdir-external-provisioner https://kubernetes-sigs.github.io/nfs-subdir-external-provisioner
    helm repo update
    kubectl create namespace infra
    helm install nfs-subdir-external-provisioner --namespace infra nfs-subdir-external-provisioner/nfs-subdir-external-prov
    isioner -f /root/_ade/nfs-tanzu-rwm-support/values.yaml
    [root@control1 nfs-tanzu-rwm-support]#

    For more information, see https://kubernetes-sigs.github.io/nfs-subdir-external-provisioner.

Where to go from here

Installing

 

Tip: For faster searching, add an asterisk to the end of your partial query. Example: cert*