Windows Server Troubleshooting - Cluster

Click here to start saving with ING DIRECT!

Home | Up | Methodology | Architecture | Tools | Memory | Processor | Registry | File System | Network | Active Directory | Contents

Get the Book

Major Topics
Task Manager
System Monitor
Computer Management
Device Manager
Start Options
Terminal Services
Blue Screen
Help Center
Fault Tolerance
More Information
Other Topics
More Detail

eXpert Genealogy

Memory from

2003-2006 Team Approach Limited
All rights reserved

MSCS Microsoft Cluster Service

With MSCS, each server is connected to

  • Public production network
  • Other cluster servers through a private network

In MSCS, servers have shared access to a SCSI or fiber channel bus. Each volume on the storage array will be the responsibility of one of the cluster servers. If that server fails, another server will take control of the storage volume. Failure detection is as follows.

Each server has two network cards
  • one connected to the public production network
  • one connected to the other cluster servers
Each server monitors the other servers with heartbeat packets every 1.2 seconds
  • A server is assumed to be offline after 2 misses heartbeats 2.4 seconds


Before assuming control of a volume, a server tries to communicate with the offline server through the public network.
  • The network card is checked with Plug and Play
  • The servers try to communicate over the public network to determine who takes control of the cluster volumes.
  • If the other server is down, the remaining server takes control of the offline server's disk drives
Normal operations return once the heartbeat resumes

Advances Server supports two-node clusters.
DataCenter Servers supports clusters up to four nodes

The following diagram shows how MSCS servers share a storage device and connect to both a public and private network.

 Public IP Network
    Cluster communications    


Private IP Network


     Shared SCSI bus    
     Shared RAID Array