vSAN performance insights reports: "SSD congestion detected during resync activity”

Products

• VMware vSAN

Issue/Introduction

This article explains the vSAN performance diagnostics issue: "SSD congestion detected during resync activity." This alert typically triggers during heavy resync operations such as hardware failures, node/disk removals, or cluster rebalancing.

Symptoms:

• High VM disk latency during resync operations.

• vSAN disk group experiencing SSD congestion.

• High VM CPU usage.

• Increased I/O latency when adding a host to the cluster.

Environment

VMware vSAN 7.x, 8.x

Architecture: Original Storage Architecture (OSA)

Cause

SSD congestion occurs when the write ingest rate (from VMs or resync activity) exceeds the de-staging rate from the cache tier to the capacity tier. When the write buffer fills faster than it can be cleared, vSAN throttles I/O at the DOM client layer to prevent the cache from becoming completely exhausted.

Resolution

Use one of the following methods to address SSD congestion.

Option 1: Hardware/Configuration Adjustment (Permanent)

Consider increasing the size of the cache tier devices to handle larger write bursts.

• vSAN 8.x: Supports cache drives up to 1.6 TB. This is not enabled by default. Refer to the VMware vSAN Support for large cache tier capacity guide to adjust advanced configurations.

Option 2: Apply Resync Throttling (Software Workaround)

If hardware changes are not possible, you can manually limit the resync bandwidth at the disk group level.

Important: Do not use esxcli for this setting, as it is non-persistent and may be overwritten by vCenter. You must use the PowerShell script attached to this KB for the settings to persist.

Execution Steps:

1. Download the configure-resync-throttle.ps1 script attached to this article.
2. Launch PowerCLI on a machine with network access to vCenter.
3. Connect to vCenter:
   PowerShell
   Connect-VIServer <vCenter_IP_or_FQDN> -User [e-mail]
4. Run the script to apply the limit (Example: 100 MB/s limit for a specific cluster):
   .\configure-resync-throttle.ps1 -ResyncIopsLimit 100 -ClusterName "Your-Cluster-Name"