Chapter 1

Getting Started with Solaris Volume Manager

The Solaris Volume Manager Administration Guide describes how to set up and maintain systems using Solaris Volume Manager to manage storage for high availability, flexibility, and reliability.

This chapter serves as a high-level guide to find information for certain Solaris Volume Manager tasks, such as setting up storage capacity. This chapter does not address all the tasks that you will need to use Solaris Volume Manager. Instead, it provides an easy way to find procedures describing how to perform common tasks associated with the following Solaris Volume Manager concepts:

Storage Capacity
Availability
I/O Performance
Administration
Troubleshooting

Caution - If you do not use Solaris Volume Manager correctly, you can destroy data. Solaris Volume Manager provides a powerful way to reliably manage your disks and data on them. However, you should always maintain backups of your data, particularly before you modify an active Solaris Volume Manager configuration.

Getting Started With Solaris Volume Manager

Solaris Volume Manager Roadmap--Storage Capacity

Table 1-1 Solaris Volume Manager Roadmap--Storage Capacity

Task	Description	For Instructions
Set up storage	Create storage that spans slices by creating a RAID 0 or a RAID 5 volume. The RAID 0 or RAID 5 volume can then be used for a file system or any application, such as a database that accesses the raw device	"How to Create a RAID 0 (Stripe) Volume" "How to Create a RAID 0 (Concatenation) Volume" "How to Create a RAID 1 Volume From Unused Slices" "How to Create a RAID 1 Volume From a File System" "How to Create a RAID 5 Volume"
Expand an existing file system	Increase the capacity of an existing file system by creating a RAID 0 (concatenation) volume, then adding additional slices.	"How to Expand Space for Existing Data"
Expand an existing RAID 0 (concatenation or stripe) volume	Expand an existing RAID 0 volume by concatenating additional slices to it.	"How to Expand an Existing RAID 0 Volume"
Expand a RAID 5 volume	Expand the capacity of a RAID 5 volume by concatenating additional slices to it.	"How to Expand a RAID 5 Volume"
Increase the size of a UFS file system on a expanded volume	Grow a file system by using the `growfs` command to expand the size of a UFS while it is mounted and without disrupting access to the data.	"How to Grow a File System"
Subdivide slices or logical volumes into smaller partitions, breaking the 8 slice hard partition limit	Subdivide logical volumes or slices by using soft partitions.	"How to Create a Soft Partition"
Create a file system	Create a file system on a RAID 0 (stripe or concatenation), RAID 1 (mirror), RAID 5, or transactional volume, or on a soft partition.	"Creating File Systems (Tasks)" in System Administration Guide: Basic Administration

Solaris Volume Manager Roadmap--Availability

Table 1-2 Solaris Volume Manager Roadmap--Availablity

Task	Description	For Instructions
Maximize data availability	Use Solaris Volume Manager's mirroring feature to maintain multiple copies of your data. You can create a RAID 1 volume from unused slices in preparation for data, or you can mirror an existing file system, including root (`/`) and `/usr`.	"How to Create a RAID 1 Volume From Unused Slices" "How to Create a RAID 1 Volume From a File System"
Add data availability with minimum hardware cost	Increase data availability with minimum of hardware by using Solaris Volume Manager's RAID 5 volumes.	"How to Create a RAID 5 Volume"
Increase data availability for an existing RAID 1 or RAID 5 volume	Increase data availability for a RAID 1 or a RAID 5 volume, by creating a hot spare pool then associate it with a mirror's submirrors, or a RAID 5 volume.	"Creating a Hot Spare Pool" "Associating a Hot Spare Pool With Volumes"
Increase file system availability after reboot	Increase overall file system availability after reboot, by adding UFS logging (transactional volume) to the system. Logging a file system reduces the amount of time that the `fsck` command has to run when the system reboots.	"About File System Logging"

Solaris Volume Manager Roadmap--I/O Performance

Table 1-3 Solaris Volume Manager Roadmap--I/O Performance

Task

Description

For Instructions

Tune RAID 1 volume read and write policies

Specify the read and write policies for a RAID 1 volume to improve performance for a given configuration.

"RAID 1 Volume Read and Write Policies"

"How to Change RAID 1 Volume Options"

Optimize device performance

Creating RAID 0 (stripe) volumes optimizes performance of devices that make up the stripe. The interlace value can be optimized for random or sequential access.

"Creating RAID 0 (Stripe) Volumes"

Maintain device performance within a RAID 0 (stripe)

Expands stripe or concatenation that has run out of space by concatenating a new component to it. A concatenation of stripes is better for performance than a concatenation of slices.

"Expanding Storage Space"

Solaris Volume Manager Roadmap--Administration

Table 1-4 Solaris Volume Manager Roadmap--Administration

Task	Description	For Instructions
Graphically administer your volume management configuration	Use the Solaris Management Console to administer your volume management configuration.	Online help from within Solaris Volume Manager (Enhanced Storage) node of the Solaris Management Console application
Graphically administer slices and file systems	Use the Solaris Management Console graphical user interface to administer your disks and file systems, performing such tasks as partitioning disks and constructing UFS file systems.	Online help from within the Solaris Management Console application
Optimize Solaris Volume Manager	Solaris Volume Manager performance is dependent on a well-designed configuration. Once created, the configuration needs monitoring and tuning.	"Solaris Volume Manager Configuration Guidelines" "Working with Configuration Files"
Plan for future expansion	Because file systems tend to run out of space, you can plan for future growth by putting a file system into a concatenation.	"Creating RAID 0 (Concatenation) Volumes" "Expanding Storage Space"

Solaris Volume Manager Roadmap--Troubleshooting

Table 1-5 Solaris Volume Manager Roadmap--Troubleshooting

Task

Description

For Instructions

Replace a failed slice

If a disk fails, you must replace the slices used in your Solaris Volume Manager configuration. In the case of RAID 0 volume, you have to use a new slice, delete and recreate the volume, then restore data from a backup. Slices in RAID 1 and RAID 5 volumes can be replaced and resynchronized without loss of data.

"Responding to RAID 1 Volume Component Failures"

"How to Replace a Component in a RAID 5 Volume"

Recover from boot problems

Special problems can arise when booting the system, due to a hardware problem or operator error.

"How to Recover From Improper /etc/vfstab Entries"

"How to Recover From Insufficient State Database Replicas"

"How to Recover From a Boot Device Failure"

Work with transactional volume problems

Problems with transactional volumes can occur on either the master or logging device, and they can either be caused by data or device problems. All transactional volumes sharing the same logging device must be fixed before they return to a usable state.

"How to Recover a Transactional Volume With a Panic"

"How to Recover a Transactional Volume With Hard Errors"