RabbitMQ Warm Standby Backup

Warm Standby Replication Validation

Introduction

This testing is a showcase of the warm standby replication scenario.

Testing Scope and Result

VMware Tanzu RabbitMQ supports continuous schema definition and message replication to a remote cluster, which makes it easy to run a standby cluster for disaster recovery. This feature is not available in the open source RabbitMQ distribution.

To achieve Tanzu RabbitMQ ‘s warm standby replication capability, perform the following steps:

  1. Set up an upstream and downstream RabbitMQ cluster with required plugins and configurations by following the manifest files.
  2. Configure the upstream and downstream standby replication via standby offsite replication operator. The configuration files let the plugin know it should collect messages for all quorum queues in default vhost and connect RabbitMQ cluster downstream to RabbitMQ cluster at the endpoint — {EndPoint IP}:5552.
  3. We created 3 queues used by the replication and ran the rabbitmq-diagnostics inspect_standby_upstream_metrics on three nodes of RabbitMQ cluster to verify how the replication is working on the upstream side. The figure shows the replication is spread across the three nodes of upstream cluster:

  1. Verify the downstream replication is also working. The figure shows that the default vhost is responsible for replication on the downstream side.

Now we used PerfTool to publish 30,000 messages to a queue from upstream cluster and then replicated them to the downstream side. We could check the disk space used by the standby replication data. At this point, we supposed the upstream site went down and we initiated a recovery procedure to promote the downstream cluster to upstream. The figure shows that we promoted 30,000 messages on the downstream site.

 

 

 

 

 

Filter Tags

Modern Applications vSphere 7 vSphere with Tanzu Kubernetes Document Activity Path