Dell 4200 manual Setting Up the Quorum Resource, Using the ftdisk Driver

Setting Up the Quorum Resource

A quorum resource is typically a hard-disk drive in the shared storage subsystem that serves the following two purposes in a cluster system:

•Acts as an arbiter between the two nodes to ensure that the specific data necessary for system recovery is maintained consistently across the nodes

•Logs the recovery data sent by the cluster nodes

Only one cluster node can control the quorum resource at one time, and it is that node that remains running when the two nodes are unable to communicate with each other. Once the two nodes are unable to communicate, the Cluster Service automatically shuts down the node that does not own the quorum resource.

With one of the cluster nodes down, changes to the cluster configuration database are logged to the quorum disk. The purpose of this logging is to ensure that the node that gains control of the quorum disk has access to an up-to-date version of the cluster configuration database.

Because the quorum disk plays a crucial role in the oper- ation of the cluster, the loss of a quorum disk causes the failure of the Cluster Server. To prevent this type of fail- ure, the quorum resource should be set up on a redundant array of hard-disk drives in the shared storage subsystem.

Using the ftdisk Driver

Microsoft Cluster Server does not support use of the Windows NT software-based, fault-tolerance driver ftdisk with any of the hard-disk drives in the shared stor- age subsystem. However, ftdisk can be used with the internal drives of the cluster nodes.

Cluster RAID Controller

Functionality

The following subsections describe functional variances of standard and cluster-enabled PowerEdge Expandable RAID Controllers operating in a PowerEdge Cluster.

Rebuild Function Does Not Complete After Reboot or Power Loss

If the cluster node is rebooted or power to the node is lost while a PowerEdge Expandable RAID Controller is rebuilding a hard-disk drive, the RAID controller termi- nates the rebuild operation and marks the drive as failed. This also occurs if the rebuild is performed from the RAID controller basic input/output system (BIOS) con- figuration utility and the user exits the utility before the rebuild completes. This occurs with all versions of the PowerEdge Expandable RAID Controller firmware on both standard and cluster-enabled RAID controllers.

If the rebuild fails to complete due to a system restart, the rebuild must be reinitiated using the RAID Controller BIOS configuration utility or using the PowerEdge RAID Console program running in the Windows NT operating system.

Rebuild Rate Not Adjustable on Cluster-Enabled RAID Controller

If a hard-disk drive fails in a redundant array, you can recover the lost data by rebuilding the drive. The rate of data reconstruction is called the rebuild rate. The rebuild rate cannot be adjusted in a cluster-enabled RAID con- troller as it can in a standard RAID controller. The cluster-enabled RAID controller rebuilds drive informa- tion at a default rate.

Using the Maximize Feature in PowerEdge RAID Console

The Maximize feature of the PowerEdge RAID Console has the following functional limitations when running in the PowerEdge Cluster:

•The Maximize icon at the upper right corner of the PowerEdge RAID Console is disabled when you open the program in the PowerEdge Cluster.

•Whenever the PowerEdge RAID Console is mini- mized to the task bar, the right-click option to maximize the application is not available.

•Whenever the PowerEdge RAID Console is mini- mized to the task bar and you minimize another application, the PowerEdge RAID Console maxi- mizes itself automatically.

3-8 Dell PowerEdge Cluster (PowerEdge 4200) Installation and Troubleshooting Guide