Cluster Server 8.0 Implementation Guide for Microsoft SQL Server - Windows
- Section I. Introducing Veritas InfoScale solutions for application high availability
- Understanding the InfoScale solutions for application high availability
- About the VCS agents for SQL Server
- How VCS monitors storage components
- How application availability is achieved in a physical environment
- How is application availability achieved in a VMware virtual environment
- Managing storage and installing the VCS agents
- Installing SQL Server
- Understanding the InfoScale solutions for application high availability
- Section II. Configuring SQL Server in a physical environment
- Overview
- Configuring the VCS cluster
- Configuring the SQL Server service group
- Configuring a SQL Server service group using the wizard
- Making SQL Server user-defined databases highly available
- Verifying the service group configuration
- Administering a SQL Server service group
- Configuring an MSDTC service group
- Configuring the standalone SQL Server
- Configuring an Active/Active cluster
- Configuring a disaster recovery setup
- Section III. Configuring SQL Server in a VMware environment
- Configuring application monitoring using the Veritas High Availability solution
- Administering application monitoring
- Administering application monitoring using the Veritas High Availability tab
- Administering application availability using Veritas High Availability dashboard
- Understanding the dashboard work area
- Section IV. Appendixes
- Appendix A. Troubleshooting
- Error and warning messages from VCS agent for SQL Server
- Troubleshooting application monitoring configuration issues
- Troubleshooting Veritas High Availability view issues
- Appendix B. Using the virtual MMC viewer
- Appendix A. Troubleshooting
How is application availability achieved in a VMware virtual environment
The Veritas High Availability solution for VMware employs Cluster Server (VCS) and its agent framework to monitor the state of applications and their dependent components running on the virtual machines that use non-shared storage. Specific agents are available to monitor the application, storage, and network components. Together, these agents monitor the overall health of the configured applications by running specific commands, tests, or scripts.
The storage configuration in the VMware virtual environment determines how VCS functions differently in a non-shared virtual environment. The non-shared storage configuration in the VMware virtual environment involves the VMware VMDK and RDM disks that reside on the shared datastore. This datastore is accessible to multiple virtual machines. However, the disks are attached to a single virtual machine at any given point of time. VCS provides a new storage agent "VMwareDisks" that communicates with the VMware ESX/ESXi hosts to perform the disk detach and attach operations to move the storage disk between the virtual machines, in a VCS cluster.
Note:
By default the VMwareDisks agent communicates with the ESX/ESXi host to perform the disk detach and attach operations. However, instead of the ESX/ESXi hosts you can choose to communicate with the vCenter Server to perform these operations.
See How the VMwareDisks agent communicates with the vCenter Server instead of the ESX/ESXi host.
In event of an application failure, the agents attempt to restart the application services and components for a configurable number of times. If the application fails to start, they initiate an application fail over to the failover target system. During the fail over, the VMwareDisks agent moves the storage disk to the failover target system, the network agents bring the network components online, and the application-specific agents then start the application services on the failover target system.
In case of a virtual machine fault, the VCS agents begin to fail over the application to the failover target system. The VMwareDisks agent sends a disk detach request to the ESX/ESXi host. After the detach operation is successful, the agent proceeds to attach the disks to the new failover target system.
In a scenario where the ESX/ESXi host itself faults, the VCS agents begin to fail over the application to the failover target system that resides on another host. The VMwareDisks agent communicates with the new ESX/ESXi host and initiates a disk detach operation on the faulted virtual machine. The agent then attaches the disk to the new failover target virtual machine.
In event of a failure in a site recovery configuration, the following tasks are performed for application monitoring continuity:
The virtual machines at the protected site are failed over to the recovery site.
The pre-online script defined in the form of a command in the SRM recovery plan applies the specified attribute values for the application components.
The status monitoring script retrieves the application status.
The network agents bring the network components online and the application-specific agents start the application services on the failover target system.
For details on the VCS configuration concepts and clustering topologies, refer to the Cluster Server Administrator's Guide.
For details on the application agents, refer to the application-specific agent guide.
For details on the storage agents, refer to the Cluster Server Bundled Agents Reference Guide.