Frequently asked questions - SQL Server 2000 - failover clustering

This article covers some of the most frequently asked questions about failover clustering in SQL Server 2000 Enterprise Edition.

For additional frequently asked questions about failover clustering, see "Failover Clustering FAQ" in SQL Server 2000 Books Online. Most SQL Server 2000 clustering issues are now covered in SQL Server Books Online.

Questions and Answers

1. Question: Is the use of software fault-tolerant disk sets for cluster storage supported?
   
   Answer: No. For additional information, click the following article number to view the article in the Microsoft Knowledge Base:
   171052

   (http://support.microsoft.com/kb/171052/ )

   Software FT sets are not supported in Microsoft Cluster Server
2. Question: Does SQL Server 2000 clustering support load balancing?
   
   Answer: SQL Server 2000 clustering does not provide load balancing; it provides failover support. To achieve load balancing, you need software that balances the load between clusters, not between servers within a cluster.
3. Question: Does SQL Server 2000 Full-Text Search support clustering?
   
   Answer: Yes.
4. Question: Is it necessary to configure MSDTC as a clustered resource for SQL Server clustering to work?
   
   Answer: No. MSDTC is required only if you need the ability to perform distributed transactions on your cluster. If you intend to use linked servers or perform replication (snapshot/transactional with immediate/queued updating subscribers), MSDTC is required. For more information, see "Distributed Transactions architecture" in SQL Server 2000 Books Online.
5. Question: How do I upgrade Microsoft SQL Server 6.5 or 7.0 to SQL Server 2000 when the version 6.5 or 7.0 server is part of a cluster?
   
   Answer: If your SQL Server 6.5 or 7.0 server is part of a cluster, you must perform the following steps to upgrade the server to SQL Server 2000:
   
   
   1. Remove clustering of SQL Server 6.5 or 7.0 on both nodes.
   2. Cluster a group for use with Microsoft Distributed Transaction Coordinator (MS DTC) as described in the "Failover Clustering Dependencies" section of SQL Server 2000 Books Online.
   3. Install SQL Server 2000 on the primary node in the clustered configuration.
   For more information, see the following topics in the SQL Server Books Online:
   • How to upgrade from a SQL Server 6.5 active/passive failover cluster (Setup)
   • How to upgrade from a SQL Server 6.5 active/active failover cluster (Setup)
   • How to upgrade from a SQL Server 7.0 active/passive failover cluster (Setup)
   • How to upgrade from a SQL Server 7.0 active/active failover cluster (Setup)
6. Question: What hardware is required to run SQL Server 2000 in a clustering environment?
   
   Answer: Refer to the Microsoft Cluster Server Administrator's Guide for a list of supported hardware configurations and hardware configuration information.
   
   Note Search by using the word "cluster" because individual components cannot be combined to create a supported system. For additional information, click the following article number to view the article in the Microsoft Knowledge Base:
   309395

   (http://support.microsoft.com/kb/309395/ )

   The Microsoft support policy for server clusters, the Hardware Compatibility List, and the Windows Server Catalog
7. Question: Can SQL Server 6.5 or 7.0 be installed on one node of a cluster and SQL Server 2000 be installed on the other node?
   
   Answer: No. For two SQL Server servers to participate in a cluster, they must be the same version.
8. Question: Where do I place the SQL Server 2000 files to provide failover support?
   
   Answer: SQL Server setup installs a new instance of SQL Server binaries on the local disk of each computer in the cluster and installs the system databases on the specified shared cluster disk. The binaries are installed in exactly the same path on each cluster node, so you must ensure that each node has a local drive letter in common with all of the other nodes in the cluster.
9. Question: How do I shut down SQL Server from the command line without the Cluster Service interpreting the shutdown as a failure?
   
   Answer: The proper way to shut down SQL Server from the command line is to use the Cluster.exe application that comes with Microsoft Windows NT Enterprise Edition, Windows 2000 Advanced Server, or Microsoft Windows 2000 Data Center. Cluster.exe is installed as part of the normal Cluster setup, which can also be run on a Microsoft Windows NT Workstation, Microsoft Windows 2000 Professional, or Microsoft Windows 2000 Server member server computer to install just the Cluster Administrator and the other administrative applications. The basic syntax for this command is as follows.

   cluster [cluster name] RESOURCE [resource name] /option
   						

   In the preceding command syntax, the /option switch controls this functionality.
   
   The specific options to be used are "/online" and "/offline". These two options are equivalent to the commands "net start mssqlserver" (the method to start SQL Server from the command line) and "net stop mssqlserver" (the method to shut down SQL Server from the command line) for a non-virtualized server respectively. You can perform this procedure on the Generic Service, the SQL Server Agent, and the SQL Server resources.
   
   Following are some examples of how to use this command syntax: To take the SQL Server 2000 resource offline if the Cluster Name is "SQLCluster" and the resource is named "VirtualSQL" (where 'VirtualSQL' is the name of the SQL Server 2000 resource, not the virtual network name resource):

   cluster "SQLCluster" resource "VirtualSQL" /offline
   					

   To bring the SQL Server 2000 resource back online:

   cluster "SQLCluster" resource "VirtualSQL" /online
   					

10. Question: Can a clustered instance of a SQL Server 2000 virtual server and a stand-alone instance of SQL Server 2000 coexist on the same computer?
    
    Answer: Yes. Several SQL Server 2000 virtual servers and stand-alone instances can coexist on the same computer. However, only the SQL Server 2000 virtual servers will fail over to other nodes in the cluster. The issues to consider are the same as those when multiple instances of SQL Server 2000 reside on the same computer. Consider the following issues:
    • A SQL Server instance name must be unique on any particular node. This includes the default instance. Even if the node does not currently own an instance, the node has registry keys and directories whose names include the instance name if it is a configured owner of an instance.
    • Also, resources on the computer are shared among the instances that currently are active on the computer. Make sure that you understand these resources.

REFERENCES

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