HP-UX 11i Mission Critical Operating Environment (new)

Enterprise Cluster Master Toolkit

The Enterprise Cluster Master Toolkit Version B.01.05 is a set of templates and scripts that allow you to configure ServiceGuard packages for the HP Domain Internet servers as well as for several third-party database management systems. The toolkit also includes other specialized tools for monitoring your mission critical environment.

This release of the Enterprise Cluster Master Toolkit is for use with HP-UX 11.0 and HP-UX 11i, and has all the features found in earlier versions in addition to new features and defect repairs.

This version includes a new toolkit that supports the use of Oracle 8i's Oracle Standby Database in Continental Cluster configurations.

HP-UX Workload Manager

HP-UX Workload Manager (WLM) Version A.01.00.02 provides goal-based workload management, enabling automatic resource allocation and application performance management.

For HP-UX 11i, HP-UX WLM uses the Process Resource Manager (PRM) C.01.08.2 to control resources. This coupling allows you to allocate resources using the shares model. In this model, a workload's resource allocation depends on its number of shares relative to the total number of shares assigned.

	NOTE: Process Resource Manager (see “HP-UX 11i Enterprise Operating Environment (new)”) and HP-UX Workload Manager both make use of the PRM API. Consequently, only one of these products should be used at a time.

Besides resource shares, HP-UX WLM's bundling with PRM provides:

This version of HP-UX WLM does not take advantage of PRM's hierarchical groups.

HP-UX WLM uses a configuration file specifying workloads and their prioritized service-level objectives (SLOs). HP-UX WLM automatically allocates CPU resources to the workloads based on priorities and current performance.

MC/ServiceGuard

Multi-Computer/ServiceGuard (MC/ServiceGuard) Version A.11.09 is a specialized facility for protecting mission critical applications from a wide variety of hardware and software failures.

	NOTE: When you purchase the Mission Critical OE, it is assumed that you want MC/ServiceGuard, NOT ServiceGuard OPS Edition. MC/ServiceGuard and ServiceGuard OPS Edition cannot coexist on the same system. Users of ServiceGuard OPS Edition are encouraged to purchase and install the Enterprise OE and then install ServiceGuard OPS Edition. For detailed installation information, see the ServiceGuard OPS Edition Version A.11.09 Release Notes for HP-UX 11i (B5161-90028).

This release of MC/ServiceGuard is for use with HP-UX 11.0 and HP-UX 11i, and has all the features found in earlier versions in addition to new functionality, defect repairs and support for new hardware configurations.

Auto-port aggregation is fully supported with 100BaseT network interface cards. The contents of MC/ServiceGuard releases A.11.01 through A.11.08 have been incorporated, including all A.11.08 patches.

New diagnostic error messages are written to the syslog file (/var/adm/syslog/syslog.log) when an attempt to obtain the cluster lock fails. Internal error codes are returned to facilitate troubleshooting.

MC/ServiceGuard supports the online replacement of network and I/O interface cards that is allowed by the HP-UX 11i Operating Environments.

Impact

Disk space required for MC/ServiceGuard is 36MB. The memory required is 6MB plus 700KB per package in the cluster plus 300KB per EMS resource in the cluster. This total amount is required on all cluster nodes, regardless of whether a given package or resource is on that node or not.

The Event Monitoring Service (EMS-CORE) file set is no longer included as part of the MC/ServiceGuard product, even thought EMS (B7609BA) is still a dependency and must be installed with MC/ServiceGuard.

	NOTE: For the 11i, MC/ServiceGuard does not provide Native Language Support. However, separate Japanese language versions of documentation are available as a part of product B3935BA with option ABJ and product B3936BJ.

MC/ServiceGuard NFS Toolkit

Multi-Computer/Service Guard Network File Server (MC/ServiceGuard NFS) Toolkit Version A.11.11 uses MC/ServiceGuard to set up highly available NFS servers. NFS servers are hosts that "export" their local directories and make them available for client hosts for mounting. On the NFS client, these mounted directories look like part of the client's local file system. MC/ServiceGuard NFS Toolkit is a set of configuration files and control scripts which can be customized to user specific needs. Post installation, these files can be found in the directory /opt/cmcluster/nfs.

However, before setting up MC/ServiceGuard NFS, you should set up the MC/ServiceGuard cluster. For instructions see Managing MC/ServiceGuard available on the HP-UX 11i Instant Information CD and on the web at:

http://docs.hp.com/

MC/ServiceGuard is used to create high availability clusters of HP 9000 servers. Computer systems with high availability clusters allow applications to continue in spite of a hardware or software failure. Such systems protect users from software failures as well as from failure of a system processing unit (SPU) or local area network (LAN) component. In the event that one component fails, the redundant component takes over and MC/ServiceGuard coordinates the transfer between components.

With MC/ServiceGuard NFS Toolkit, the NFS server package containing the exported file systems moves to a different node in the cluster in event of a failure. After MC/ServiceGuard starts the NFS package on the adoptive node, the NFS file systems are re-exported from the adoptive node with minimum disruption of service to users. The client side "hangs" until the NFS server package comes up on the adoptive node. When the service returns, the user can continue access to the file. You do not need to restart the client.

File locks are not preserved when a MC/ServiceGuard NFS package moves over to another node.

For use and configuration information, see Managing Highly Available NFS (B5125-90001) available on the HP-UX 11i Instant Information CD and on the web at:

http://docs.hp.com/