C H A P T E R 1 - Overview of Sun Cluster and Sun StorageTek Availability Suite Software

C H A P T E R 1

Overview of Sun Cluster and Sun StorageTek Availability Suite Software

The Sun Cluster and Sun StorageTek Availability Suite software combine to provide a highly available environment for cluster storage. The Remote Mirror software is a data replication application that provides access to data as part of business continuance and disaster recovery plans. The Point-in-Time Copy software is a point-in-time snapshot copy application that enables you to create copies of application or test data.

For a description of the terminology used in this guide, refer to the Glossary.

The topics in this chapter include:

Supported Software and Hardware

Using the Sun StorageTek Availability Suite Software in a Sun Cluster Environment

Exporting, Importing, and Joining Shadows in a Sun Cluster Operating Environment

VTOC Information

Supported Software and Hardware

TABLE 1-1 Supported Software and Hardware
Operating Environment Software	Solaris 10 Update 1 and higher.
Sun Cluster Software	Sun Cluster 3.1 Update 4 and Sun Cluster 3.2
Devices and Volume Manager Software	Solaris Raw Device `/dev/rdsk/c?t?d?s2` Solaris Volume Manager Sun Cluster Global Device Veritas Volume Manager (VxVM)
Sun StorageTek Availability Suite Software	Sun StorageTek Availability Suite Remote Mirror and Point-in-Time Copy software.
Supported Cluster Configuration	The Sun Cluster 3.0 Update 3 release, the Sun Cluster 3.1 initial release, and the Sun StorageTek Availability Suite software are supported in a two-node cluster environment only.
Hardware	If you plan to install the software from the product CD, a CD-ROM drive connected to the host server where the software is to be installed.
Disk space requirements	Disk space requirements for either SPARC or x86 platforms: 15 Mbytes Sun StorageTek Availability Suite core packages: 2.2 Mbytes Remote Mirror packages: 1.1 Mbytes Point-in-Time Copy packages: 0.6 Mbytes Availability Suite local configuration file: 5.4 Mbytes Availability Suite cluster configuration file: 5.4 Mbytes

Using the Sun StorageTek Availability Suite Software in a Sun Cluster Environment

To use Sun Cluster failover features with the Sun StorageTek Availability Suite software, your software environment requires the installation and configuration of Sun Cluster software on all nodes that are sharing storage in use by Availability Suite software.

As a failover data service, Availability Suite can be configured on some, or all nodes in a Sun Cluster Operating Environment. In this environment, the Sun StorageTek Availability Suite software is cluster aware. A software product is Sun Cluster aware if it can coexist within the Sun Cluster environment, failing over and failing back as the device group, or resource group containing a device group and logical host, fails over and fails back between configured nodes. As a Sun Cluster aware product, the Availability Suite data services and associated resource types can be configured as highly available, utilizing the High Availability framework that Sun Cluster provides.

Global and Local Use of the Sun StorageTek Availability Suite Software

Note - See Rules for the Remote Mirror Software.

The Sun StorageTek Availability Suite software can use volumes that are local or global devices. Global devices are those Sun StorageTek Availability Suite software or other volumes accessible from any cluster node and which will fail over under the control of the Sun Cluster framework. Local devices are volumes that are local to the individual node (host machine), not defined in a disk device or resource group, and not managed within a cluster file system. Local devices do not fail over and switch back.

To access lo cal devices, use the C local or -C local options as part of the sndradm commands, or the -C local option with iiadm commands. To access global devices, use the command options C tag and -C tag. (Typically, you do not need to specify the -C tag option as iiadm and sndradm automatically detect the disk device group.)

See Chapter 3 in this guide and the Sun StorageTek Availability Suite Administration Guides listed in Related Documentation.

Switching Over Global Devices Only

The scswitch(1M) command enables you to change all resource groups and device groups manually from the primary mastering node to the next preferred node. The Sun Cluster documentation describes how to perform these tasks.

Local devices do not fail over and switch back so do not configure them as part of your cluster. A file system mounted on a volume and designated as a local device must not be configured as a device to fail over and switch back in the Sun Cluster environment.

Volumes Eligible for Use

Note - When creating shadow volume sets, do not create shadow or bitmap volumes using partitions that include cylinder 0 because data loss might occur. See VTOC Information.

You can replicate the following critical volumes using the Remote Mirror software:

Database and database management system (DBMS) logs (the total database or online DBMS log)

Access control files

You can exclude volumes from replication if they can be reconstructed at the recovery site or if they seldom change:

Temporary volumes (such as those used in sort operations)

Spool files

Paging volumes

When selecting a volume to be used in the volume set (including the configuration location), ensure that volume does not contain disk label private areas (for example, slice 2 on a Solaris OS-formatted volume). The disk label region is contained in the first sectors of cylinder 0 of a disk.

The Point-in-Time Copy software supports all Sun-supported storage. It works independently of the underlying data reliability software (for example, RAID-1, RAID-5, or volume manager). Additionally, you can use it as a tool when migrating data to and from differing storage types.

Typical uses for the Point-in-Time Copy software include:

Backup of live application data

Load data warehouses and fast resynchronization of data warehouses at predefined intervals

Application development and test on a point-in-time snapshot of live data

Migrate data across different types of storage platforms and volumes

Hot back up of application data from frequent point-in-time snapshots

Exporting, Importing, and Joining Shadows in a Sun Cluster Operating Environment

Overview

The Sun StorageTek Availability Suite Point-in-Time Copy software is a volume-based snapshot facility that runs in the Solaris Operating System and allows creation and management of instant snapshot (point-in-time) copies, allowing repurposing of data for applications such as hot backup, decision support, or application development and testing.

This section provides an overview of exporting, importing, and joining shadows in a Sun Cluster OE. For detailed information, refer to Exporting, Importing, and Joining Shadows in a Sun Cluster OE

Requirements

Sun Cluster High Availability support for the Sun StorageTek Availability Suite Point-in-Time Copy software is achieved by using the Sun Cluster HAStoragePlus resource type. This resource type can be used to make the global devices or volume manager controlled volumes, and any file systems mounted on those volumes, highly available within a Sun Cluster Operating Environment. When this configured resource type includes volumes that are also configured with the Point-in-Time Copy software, the volumes are also highly available.

The Sun Cluster HAStoragePlus resource type is a failover resource type and as such is only active on one Sun Cluster node at a time. When one or more global devices or the file systems mounted on them are configured with the HAStoragePlus resource type, they become part of a Sun Cluster resource group and under Sun Cluster Resource Group Manager (RGM) control. The RGM causes a configured resource group to voluntarily or involuntarily failover or switchover to another configured Sun Cluster system. The global devices or volume manager controlled volumes and the file systems mounted on them are unmounted on the current Sun Cluster node and remounted on the failover Sun Cluster node under RGM control.

On a per Point-in-Time Copy set basis, each volume set requires a minimum of three raw global device partitions or volume manager controlled volumes (for example, Solaris Volume Manager or VERITAS Volume Manager).

Note - Sun Cluster DID devices are not supported as master, shadow, or bitmap volumes, due to disk data fencing functionality when Sun Cluster failure events are active. Sun Cluster Global devices that are symmetrical in name to the DID devices are supported.

Redundancy

To provide high availability to the data contained on a Point-in-Time Copy set when using global devices, it is assumed that the master, shadow, and bitmap volumes are on redundant storage, since there is no means today to provide for host-based data service redundancy on raw global devices. If controller-based redundancy is not available, then a Sun Cluster supported volume manager must be used. For performance reasons, it is recommended under all scenarios that bitmap volumes are NOT placed on RAID-5 volumes (either host-based or controller-based), since the nature of bitmap I/O processing within a Point-in-Time Copy volume set can be I/O intensive.

VTOC Information

The Solaris system administrator must be knowledgeable about the virtual table of contents (VTOC) that is created on raw devices by Solaris.

The creation and updating of a physical disk's VTOC is a standard function of Solaris. Software applications like Availability Suite, the growth of storage virtualization, and the appearance of SAN-based controllers have made it easy for an uninformed Solaris system administrator to inadvertently allow a VTOC to become altered. Altering the VTOC increases the possibility of data loss.

Remember these points about the VTOC:

A VTOC is a software generated virtual table of contents based on the geometry of a device and written to the first cylinder of that device by the Solaris format(1M) utility.

Various software components such as dd(1M), backup utilities, Point-in-Time Copy software, and Remote Mirror software can copy the VTOC of one volume to another volume if that volume includes cylinder 0 in its mapping.

If the VTOC of the source and destination volumes are not 100% identical, then there is a possibility of some type of data loss occurring.

This data loss may not be initially detectable, but can be detected later when other utilities are used, like fsck(1M).

When first configuring and validating volume replication, save copies of all affected device's VTOCs using the prtvtoc(1M) utility. The fmthard(1M) utility can be used to restore them later, if necessary.

When using volume managers like VxVM and SVM, copying between individual volumes created under these volume mangers is safe. VTOC issues are avoided because the VTOC is excluded from volumes created by these volume managers.

When formatting individual partitions on a raw device, for all partitions except the backup partition, make sure they do not map cylinder 0, which contains the VTOC. When using raw partitions as volumes, you are the volume manager and you need to exclude the VTOC from partitions that you configure.

When formatting the backup partition of a raw device, make sure that the physical geometries of the source and destination devices are identical.
(Partition 2, by default, maps all cylinders under the backup partition.) If identical device sizing is not possible, make sure that the source backup partition is smaller then the destination partition, and that the destination partition does not map cylinder 0.