Skip to content

Multiple Availability Zones for Volumes

Volumes are dedicated by zone. Making a persistent volume (PV) highly available across multiple availability zones involves using distributed storage solutions and configurations that support replication across those zones.

Solution for Cloud Environments

For cloud providers like AWS, GCP, and Azure, you can choose a storage type or class that supports replication or regional persistent disks or multi-AZ deployment.

Solution for Openstack environments

Using Distributed Storage Systems (e.g., Ceph, GlusterFS, JuiceFS) to replicate PVs across zones and support failover between zones.

Setup Using Distributed Storage Systems (e.g., Ceph, GlusterFS, JuiceFS)

  • Install and configure distributed storage system.
  • Deploy a distributed storage system like Ceph, GlusterFS or JuiceFS that inherently provides high availability and replication across multiple nodes and availability zones.
  • Create StorageClass for distributed storage.
  • Define a StorageClass that points to the distributed storage solution and has settings for multi-AZ deployment.

Multi AZ configuration for StatefulSet CMON

You can use nodeSelector for the pods that use persistent volume and provide a fixed availability zone, so if the pods are re-scheduled to other nodes will land in the same availability zone as the volume. For example:

cmon:
  nodeSelector:
    zone: europe-west1-a

The deployment has the nodeSelector which defines in which zone will be deployed the pod.

In StorageClass, don't forget to configure:

volumeBindingMode: WaitForFirstConsumer

This is to delay volume binding and pv creation until Pod is scheduled. Distributed storage systems are an effective way to solve the high available Statefulsets.

Multi AZ configuration for multiple replicas of ccx

Use podAntiAffinity to tell the Kubernetes scheduler to deploy each replica in different nodes in different availability zones. A pod should be scheduled based on the topology of the underlying nodes. For example:

ccx:
  affinity:
    podAntiAffinity:
      requiredDuringSchedulingIgnoredDuringExecution:
        - podAffinityTerm:
            labelSelector:
              matchExpressions:
                - key: app
                  operator: In
                  values:
                    - ccx-deployer-service
            topologyKey: topology.kubernetes.io/zone
Back to top