大网数据广告活动
Hyper-V与虚拟化

Fails to connect to virtual server after failover

· 52CE 我爱测

Helps solve an issue where users in a different subnet are unable to connect to the virtual server after a failover from one cluster node to another.

This article provides a solution to an issue where users in a different subnet are unable to connect to the virtual server after a failover from one cluster node to another.

Original KB number:   244331

Symptoms

After a failover from one cluster node to another, users in a different subnet may be unable to connect to the virtual server.

Cause

Server Clusters and Failover Clustering perform a gratuitous Address Resolution Protocol (ARP) request when a failover occurs. However, some devices (such as switches) may not forward the gratuitous ARP request to other devices. This causes devices on the other side of the switch or router to have the incorrect MAC address for the virtual server that has failed over. Often, this situation corrects itself after a router or switch sees the failure and updates its ARP cache by performing a broadcast. Most routers and switches are configured not to forward ARP traffic between subnets to prevent ARP storms from occurring.

Resolution

Gratuitous ARP requests must be forwarded across networks so that all devices receive the updated MAC-to-IP address mappings. Contact your hardware manufacturer for information about how to change your switch or router's configuration so that gratuitous ARP requests are passed to all networks.

More information

In a cluster, each computer (or cluster node) has a network adapter attached to the corporate network, and each cluster node has its own IP address, network name (NetBIOS name), and MAC address. The virtual server has an IP address and network name, but uses the MAC address of the cluster node that is the current owner of the virtual server resources.

When a failover occurs, the cluster server of the node that is receiving IP resources sends a gratuitous ARP request so that all devices (computers, routers, and switches) are updated, and a new MAC address is assigned to an existing IP address. If a switch or router does not pass the updated MAC-to-IP address mappings, other network devices contain the old MAC address for the cluster node that is down.

For additional information, click the article numbers below to view the articles in the Microsoft Knowledge Base:

168567 Clustering Information on IP Address Failover

Source: Microsoft Learn / IT Pro troubleshooting by Microsoft and contributors. Documentation licensed CC BY 4.0. Code samples are covered by the MIT License. Converted from Markdown, with links normalized and embedded media omitted.

资料来源与适用范围

资料来源
Microsoft Learn / IT Pro troubleshooting
适用范围
- Supported versions of Windows Server · Source date 02/12/2026
内容语言
中文;命令、路径和错误标识保留技术原文
本地收录日期
2026-11-15
授权说明
CC BY 4.0 (documentation), MIT (code samples)

正文与所需资源由本站本地存储。查看来源及更新记录 ↗

请按适用版本和实际日志确认原因;本站没有对每一种系统或设备逐项执行文中的操作。