Intermittent RPO violations in VMware Cloud Director Availability

Products

• VMware Live Recovery

Issue/Introduction

• RPO violations are seen for various replications in the VMware Cloud Director Availability RPO compliance report.
• The violations are "little" and typically last 1 or 2 minutes.

Environment

VMware Cloud Director Availability 4.7.x

Cause

vCDA, by design, will initiate synchronizations for replications based on the anticipated throughput and average data change calculated from previous instances.

Naturally, the potential transfer rate and the amount of data to transfer will fluctuate. For example, if there were a higher than average number of files posted to a file server during one instance, the synchronization will take longer than expected and the larger delta may push the RPO out of compliance for a short time. This type of variance is expected and should not be a concern if the RPO is recovering in a reasonable amount of time.

Additionally, there are simple operational activities that can cause RPO violations. For example, if the source VM is powered off, synchronization will not be initiated automatically.

Resolution

This behavior is expected under normal operations. If RPO violations are lengthy or persistent, additional investigation would be required.