Log In
Or create an account ->
Imperial Library
Home
About
News
Upload
Forum
Help
Login/SignUp
Index
Backup & Recovery
Preface
I Wish I’d Had This Book
Only the Recovery Matters
Products Change
Backing Up Databases Is Not That Hard
Bare-Metal Recovery Is Not That Hard
How This Book Is Organized
Part I
Part II
Part III
Part IV
Part V
Part VI
What’s New in This Book
What’s Missing?
Speaking of BackupCentral.com
Conventions Used in This Book
How to Contact Us
Safari® Enabled
This Book Was a Team Effort
Contributors
Technical Editors
Horror Stories
Special Mention
I Don’t Know It All
How Can I Say Thanks?
1. Introduction
1. The Philosophy of Backup
Champagne Backup on a Beer Budget
Why Should I Read This Book?
Schadenfreude
You Never Want to Say These Words
You’re Curious About Open-Source Backup Products
You Want to Learn About Disk-Based Backup
Why Back Up?
What Will Lost Data Cost You?
What Will Downtime Cost You?
Wax On, Wax Off: Finding a Balance
Don’t Go Overboard
Get the Coverage That You Need
Why the Word “Volume” Instead of “Tape”?
2. Backing It All Up
Don’t Skip This Chapter!
The Impossible Job That No One Wants
Deciding Why You Are Backing Up
Deciding What to Back Up
Plan for the Worst
Take an Inventory
Are You Backing Up What You Think You’re Backing Up?
Back Up All or Part of the System?
Backing up only selected drives or filesystems
Backing up the entire system
Deciding When to Back Up
Backup Levels
Which Levels Do You Run and When?
Weekly schedule: All full/level 0 backups
Weekly schedule: Weekly full, daily level differentials/level 1s
Weekly schedule: Weekly full, daily leveled backups
Weekly schedule: Monthly full, daily Tower of Hanoi incrementals
“In the Middle of the Night...”
Deciding How to Back Up
Be Ready for Anything: 10 Types of Disasters
Automate Your Backup
Plan for Expansion
Don’t Forget Unix mtime, atime, and ctime
Backups change atime
The atime can be reset—with a penalty
Don’t Forget ACLs
Don’t Forget Mac OS Resource Forks
Keep It Simple, SA
Storing Your Backups
Storage in General
On-Site Storage
12,000 gold pieces
Off-Site Storage
Choosing a media vaulting vendor
Testing your chosen vendor
Electronic vaulting
Testing Your Backups
Test Everything!
Test Often
Monitoring Your Backups
You Can Always Make It Better
If It’s Not Baroque, Don’t Fix It
Following Proper Development Procedures
Unrelated Miscellanea
Protect Your Career
Self-preservation: Document, document, document
Strategy: Make backups an integral part of the installation process
Get the Money Your Backups Need
Be ready
Make a formal presentation
Good Luck
2. Open-Source Backup Utilities
3. Basic Backup and Recovery Utilities
An Overview
How Mac OS Filesystems Are Different
cpio
Why isn’t cpio more popular?
ditto
dd
dump and restore
ntbackup
rsync
System Restore
tar
Other Utilities
asr
pax
psync, rsyncx, hfstar, xtar, and hfspax
Backing Up and Restoring with ntbackup
Creating a Simple Backup Configuration
Executing Your Simple Backup
Restoring with ntbackup
Using System Restore in Windows
Creating Restore Points
Recovering Windows Using a Restore Point
Backing Up with the dump Utility
Syntax of the dump Command
The Options to the dump Command
Specifying a complete or incremental backup (0–9)
Specifying a blocking factor (b)
Updating the dumpdates file (u)
Notifying your backup operators (n)
Specifying density and size (d and s)
Do I have to use the s and d options?
Specifying a backup device file (f)
Displaying which filesystems need to be backed up (W and w)
Interesting options for Solaris’s ufsdump utility
What a dump Backup Looks Like
dump records an index on the volume
Using the index to create a table of contents
Restoring with the restore Utility
Is the Backup Volume Readable?
Blocking Factor
Byte-Order Differences
Different Versions of dump
Syntax of the restore Command
The Options to the restore Command
Determining the type of restore
Determining how the restore behaves
Creating a dump volume table of contents (t)
Performing a complete (recursive) filesystem restore (r)
Restoring files by name (x)
Restoring files interactively (i)
Restoring files to another location
Requesting verbose output (v)
Skipping files (s)
Specifying a blocking factor (b)
Specifying a backup drive or file (f)
Specifying no query during restore (y)
Limitations of dump and restore
Features to Check For
Backing Up and Restoring with the cpio Utility
The Syntax of cpio When Backing Up
The Options to the cpio Command
Specifying the output mode (o)
Restoring access times (a)
Specifying the ASCII format (c)
Requesting verbose output (v)
Specifying a blocking factor of 5,120 (B)
Specifying an I/O block size (C)
Specifying an output device or file (O)
Backing up to a remote device (piping to an rsh or ssh command)
Restoring with cpio
Different versions of cpio
Byte-order problems
Wrong header type
Strange block size
Full or partial restore, or table of contents only?
cpio’s Restore Options
Telling cpio Which Device to Use
Examples of a cpio Restore
Listing the files on a cpio volume
Doing an entire filesystem restore
Doing a pattern-match restore
Renaming files interactively
Other useful options
Restoring to a different directory
Using cpio’s Directory Copy Feature
Backing Up and Restoring with the tar Utility
The Syntax of tar When Backing Up
The Options to the tar Command
Listing files on standard input
Syntax of tar When Restoring
Restoring selected parts of the archive
Tricking tar into using wildcards during a restore
Changing ownership, permissions, and attributes during a restore
Some Other Neat Things About tar
Finding everything that’s under the directory
Using tar to move a directory
Restoring to an alternate location
Backing Up and Restoring with the dd Utility
Basic dd Options
Specifying the input file
Specifying the output file
Specifying the block size
Specifying the input and output block sizes separately
Specifying the number of records to read
Using dd to Copy a File or Raw Device
Using dd to Convert Data
Converting data to go into another command
Converting data that is in the wrong format
Using dd to Determine the Block Size of a Tape
Using dd to Figure out the Backup Format
Using rsync
Basic rsync Syntax
A few twists
rsync on Windows
rsync on Mac OS
Restoring with rsync
Backing Up and Restoring with the ditto Utility
Syntax of ditto When Backing Up
The Options to the ditto Command
Syntax of ditto when Restoring
Listing the files in a ditto archive
Comparing tar, cpio, and dump
Using ssh or rsh as a Conduit Between Systems
4. Amanda
Summary of Important Features
Client/Server Architecture Using Nonproprietary Tools
Amanda Security
Holding Disk
Backup Scheduling
Tape Management
Device Management
Configuring Amanda
Backing Up Clients via NFS or Samba
Backing Up Using NFS
Backing Up via Samba
Amanda Recovery
Community and Support Options
Future Plans
5. BackupPC
BackupPC Features
How BackupPC Works
Installation How-To
Security Versus Ease of Use
Basic Sizing
Installing BackupPC
Installation packages
Starting BackupPC
Using the CGI Interface
Configuration Files
Per-Client Configuration
The BackupPC Community
The Future of BackupPC
6. Bacula
Bacula Architecture
Bacula Components
Interaction Between Components
Authentication
Configuration
Bacula Features
An Example Configuration
Setting Up the Server
Initial Backup (Linux Client)
Initial Restore (Linux Client)
Windows Backup
Mac OS X Backup
Advanced Features
Bare-Metal Recovery
Backup Traffic and Storage Encryption
Python Script Support
Client Script Support
Autochanger Support
ANSI and IBM Tape Labels
File-Based Intrusion Detection
Future Directions
Pool Migration
Tracking Deleted/Renamed Files
Python-Based GUI Tool
Base Job Support
Client-Initiated Backups
Plug-in Support for File Daemons
7. Open-Source Near-CDP
rsync with Snapshots
An Example
Beyond the Example
Understanding Hard Links
Hard-Link Copies
A simple example script
Restoring from the Backup
Things to Consider
How large is each backup?
A brief word about mail formats
Other useful rsync flags
Backing up databases or other large files that keep changing
Backing up Windows systems
Large filesystems and rsync’s memory scaling
Atomicity and partial transfers
rsnapshot
Platform Support
When Not to Use rsnapshot
Setting Up rsnapshot
The rsnapshot Community
rdiff-backup
Advantages
Disadvantages
Quick Start
Windows, Mac OS X, and the Future
3. Commercial Backup
8. Commercial Backup Utilities
What to Look For
Full Support of Your Platforms
Should You Back Up Special Files?
Backup of Raw Partitions
Backup of Very Large Filesystems and Files
Aggressive Requirements
LAN-Free Backup
Server-Free (or Serverless) Backup
De-Duplication Backup Systems
Snapshots
Replication
Near-Continuous Data Protection Systems
Continuous Data Protection Systems
Remote Office Backup
Simultaneous Backup of Many Clients to One Drive
Disk-to-Disk-to-Tape Backup
Simultaneous Backup of One Client to Many Drives
Data Requiring Special Treatment
Network-Mounted Filesystems
Custom User Scripts
Databases
Storage Management Features
Archives
Backups make lousy archives
Satisfy electronic discovery requests
Other backup bugaboos
True archiving
Two types of archivers
Hierarchical Storage Management
Information Lifecycle Management
Reduction in Network Traffic
Keep Backup Traffic at the Subnet Level
Use Client-Side Compression
Incorporate Throttling
Storage Area Networks
Support of a Standard or Custom Backup Format
Standard Backup Formats
The dump utility
The tar, ditto, and cpio utilities
Custom Backup Formats
What happened to SIDF?
A Reality Check
Ease of Administration
Security
Ease of Recovery
Protection of the Backup Index
Robustness
Automation
Volume Verification
Cost
Vendor
Final Thoughts
9. Backup Hardware
Decision Factors
Reliability
Duty Cycle
Transfer Speed
Flexibility
Tape drives: Not so flexible
Optical drives: A little more flexible
Disk drives: Very flexible
Time-to-Data
Capacity
Removability
Cost
Summary
Using Backup Hardware
Compression
Density Versus Compression
How Often Should I Change My Media?
Cartridge Care
Drive Care
Nearline and Offline Storage
Tape Drives
Tape Drives Must Be Streamed
Compression Makes It Harder to Stream Drives
Variable Speed Tape Drives
Helical and Linear Tape Drives Are Different
Cartridges Versus Cassettes
Midrange Tape Drive Types
3480 (end-of-lifed)
3590
3592
TS1120
3570 drive (a.k.a. Magstar MP)
8 mm (8x0x) drives (end-of-lifed)
9840 drives
9940 drives
T10000 drives
AIT drive
DDS drive
DLT drives (end-of-lifed)
DLT-S drives (aka Super DLT)
DLT-V drives (aka Value DLT)
DTF drive
LMS NCTP drive
LTO drives
Mammoth drive (end-of-lifed)
MLR 1-3 drives
VXA
Optical Drives
Optical Recording Methods
Magneto-optical recording method
Phase change recording method
Dye polymer recording method
WORM recording methods
Optical Recording Formats
CD recording formats
DVD recording formats
Magneto-optical recording format
UDO recording format
Automated Backup Hardware
Disk Targets
Disk-As-Disk Targets
Advantages of disk-as-disk targets
Disadvantages of disk-as-disk targets
SAN disk-as-disk targets
NAS disk-as-disk targets
Disk-As-Tape: Virtual Tape Libraries
Advantages of VTLs
Disadvantages of VTLs
How do you eject virtual tapes?
Disk Features to Consider
Packaging
De-duplication
Replication
Content-awareness
Re-presentation
Stacking
Notification
Disk-As-Tape: Virtual Tape Cartridges
4. Bare-Metal Recovery
10. Solaris Bare-Metal Recovery
Using Flash Archive
Backup and Recovery Overview
Perform the backup
Perform the restore
Initial Considerations
System requirements
Frequency of backup
Back up to disk or tape?
Restore from tape or disk?
Interactive or noninteractive restore?
Other environmental constraints
Preparing for an Interactive Restore
Creating Flash Archive Images
Determining filesystems to back up
Using flar create
Creating a flash archive tape
Bare-Metal Recovery with Flash Archive
An interactive recovery example
Setup of a Noninteractive Restore
Noninteractive Setup Files
profile
The rules file
The sysidcfg file
Creating a Noninteractive Tape Image
Creating a Noninteractive Disk Image
Post-Recovery Procedures
Final Thoughts
11. Linux and Windows
How It Works
If Then GOTO
Choosing Backup Methods
Live or alternate boot?
Image level or filesystem level?
Complete disk or separate partitions?
Four backup options
The Steps in Theory
Step 1: Back Up Important Metadata
Linux, FreeBSD, and Solaris x86
Windows
Step 2: Back Up the OS with a Native Utility
Linux, FreeBSD, Solaris X86
Windows
Step 3: Boot the System from Alternate Media
Step 4: Restore the Boot Block Information
Step 5: Partition and Format the New Root Drive
Linux, FreeBSD, and Solaris X86
Windows
Step 6: Restore the OS to the New Root Drive
Assumptions
Alt-Boot Full Image Method
Create the Bare-Metal Backup
Back up the important metadata
Boot the system from alternate media
Back up the operating system with a native utility
Perform a Bare-Metal Recovery
Boot the system from alternate media
Restore the boot block information
Prepare the new root drive
Restore the operating system
Alt-Boot Partition Image Method
Create the Bare-Metal Backup
Back up the important metadata
Boot the system from alternate media
Back up the operating system with a native utility
Perform a Bare-Metal Recovery
Boot the system from alternate media
Restore the boot block and prepare the new root drive
Restore the operating system
Live Method
Create the Bare-Metal Backup
Back up the important metadata
Back up the operating system with a native utility
Perform a Bare-Metal Recovery
Boot the system from alternate media
Restore the boot block information and prepare the new root drive
Restore the operating system
Alt-Boot Filesystem Method
Create the Bare-Metal Backup
Back up the important metadata
Boot the system from alternate media
Back up the operating system with a native utility
Perform a Bare-Metal Recovery
Boot the system from alternate media
Restore the boot block and prepare the new root drive
Restore the operating system
Automate Bare-Metal Recovery with G4L
Advantages of G4L
Drawbacks of G4L
Setting Up G4L
Using G4L
Customizing G4L
Commercial Solutions
12. HP-UX Bare-Metal Recovery
System Recovery with Ignite-UX
Ignite-UX Overview
Network Services and Remote Boot Protocols
Differences Between HP Integrity and HP9000 Clients
Planning for Ignite-UX Archive Storage and Recovery
Considerations for the Remote Booting of Clients
Sizing the Recovery Archive
Configuring an Ignite-UX Network Server
Recovery Archive Management
Implementation Example
Command-Line Examples
Verifying Archive Contents
Troubleshooting Recovery Operations
System Cloning
Security
System Recovery and Disk Mirroring
13. AIX Bare-Metal Recovery
IBM’s mksysb and savevg Utilities
mksysb and savevg Format
Preparing to Use mksysb and savevg
Preparing for the restore
Backing Up with mksysb
mksysb Summary
Backing Up rootvg to Locally Attached Tape
Backing Up rootvg to a Remote Tape Drive
Backing Up to Disk
Making a Bootable DVD/CD from an Existing mksysb
Creating a CD/DVD Backup in One Step
Setting Up NIM
Setting Up a NIM Server
Adding a Client Definition to NIM
Setting a mksysb Definition for a Client
savevg Operations
Using savevg to Back Up a Volume Group
Verifying a mksysb or savevg Backup
Restoring an AIX System with mksysb
System Cloning
AIX 4.x Operating System
AIX 5.x Operating System
14. Mac OS X Bare-Metal Recovery
How It Works
Preparing for a Bare-Metal Recovery
Performing a Bare-Metal Recovery
A Sample Bare-Metal Recovery
Perform the Backup
Recover the System
5. Database Backup
15. Backing Up Databases
Can It Be Done?
Confusion: The Mysteries of Database Architecture
The Muck Stops Here: Databases in Plain English
What’s the Big Deal?
Database Structure
The Power User’s View: Logical Elements of a Database
Instance
Database
Table
Index
LOBs
Object
Row
Attribute
The DBA’s View: Physical Elements of a Database Environment
Page
Datafile
Extents
Tablespace
Partition
Master database
Transaction
Rollback log
Transaction log
Checkpoint
An Overview of a Page Change
ACID Compliance
What Can Happen to an RDBMS?
Backing Up an RDBMS
Physical and Logical Backups
Get Every Instance
Transaction Log Dumps Are Not Incremental Backups
Do It Yourself: Creating Your Own Backup Utility
Intermediary disk
Dedicated tape drive
Shell/batch scripts
Calling a Professional
DB2
Exchange
Informix
MySQL
Oracle
PostgreSQL
Sybase
SQL Server
Restoring an RDBMS
Loss of Any Nondata Disk
Loss of a Data Disk
Online Partial Restores
Documentation and Testing
Unique Database Requirements
16. Oracle Backup and Recovery
Two Backup Methods
rman
User-Managed Backups
Oracle Architecture
The Power User’s View
Instance
Database
Table
Index
Large object datatypes
Object
Row
Attribute
The DBA’s View
Blocks
Extents
Segment
Datafile
Tablespace
Partition
Control file
Transaction
Undo tablespace
Checkpoint
Flash recovery area
Redo log
Initialization parameters
Restore versus recover
Finding All Instances
Physical Backups Without rman
Cold Backup
Hot Backup
Debunking Hot-Backup Myths
Physical Backups with rman
Important New rman Features
Automating rman
Create a recovery catalog
Create persistent parameters
Create rman scripts
The database.inc.rman command file (level 1 backups)
Flashback
Other Commercial Backup Methods
Managing the Archived Redo Logs
Recovering Oracle
Using This Recovery Guide
Seriously, Think About rman
Step 1: Try Startup Mount
Step 2: Are All Control Files Missing?
Step 3: Replace Missing Control File
“But I don’t have a good control file!”
Step 4: Are All Datafiles and Redo Logs OK?
Step 5: Restore Damaged Datafiles or Redo Logs
Step 6: Is There a “Backup to Trace” of the Control File?
Step 7: Run the create controlfile Script
Step 8: Restore Control Files and Prepare the Database for Recovery
1) Restore control files from backup
2) Start up mount
Step 9: Recover the Database
Recover and open the database with rman
Attempt to recover database manually
Alter database open resetlogs
Step 10: Does “alter database open” Work?
Big shortcut for rman users
User-managed backups: Read on
Damaged datafile
Damaged log group
Damaged required tablespace
Damaged rollback segment
Before going any farther...
How media recovery works
Step 11: Are There Damaged Datafiles for Required Tablespaces?
Step 12: Restore All Datafiles in Required Tablespaces
Step 13: Damaged Nonrequired Datafile?
Step 14: Take Damaged Datafile Offline
Step 15: Were Any Datafiles Taken Offline?
Step 16: Restore and Recover Offline Datafiles
Restore the damaged datafiles
Datafile recovery
Tablespace recovery
Database recovery
Step 17: Is There a Damaged Online Log Group?
Step 18: Are Any Rollback Segments Unavailable?
Step 19: Recover Tablespace Containing Unavailable Rollback Segment
Step 20: Is the Current Online Log Damaged?
Step 21: Restore and Recover All Database Files from Backup
Step 22: Run alter database open resetlogs
Step 23: Is an Active Online Redo Log Damaged?
Step 24: Perform a Checkpoint
Step 25: Is an Inactive Online Redo Log Damaged?
Step 26: Drop/Add a Damaged, Inactive Log Group
You’re Done!
Logical Backups
Performing a Logical Backup
Recovering with a Logical Backup
A Broken Record
17. Sybase Backup and Recovery
Sybase Architecture
Overview of the Sybase Architecture
Sybase Command-Line Utilities
isql
bcp
dsedit
Required Environment Variables
The Power User’s View
Server
Engine
Database
Transaction
Table
System Table
Index
Stored Procedures
The DBA’s View
Page
Extent
Datafiles and Devices
Segment
Configuration File
Transaction Log
What Happens When Transaction Logs Fill Up?
Transaction log sizing
The interfaces File
The SYBASE.sh and SYBASE.csh Files
Backup Server
Dump Device
Hot and Cold Backups
Protecting Your Database
dbcc: The Database Consistency Checker
Standard/nightly dbcc checks
Reorgs
Update Statistics
Configuration Audits
Implement Mirroring and Disk Striping
How to Back Up Your Servers
Syntax of the dump statement
Backup striping and compression
Have a Run Book
Backup Automation Through Scripting
Backup Automation Basics
A simple update stats script
A sample transaction logfiles backup script
Schedule backups
Mailing crontab results
Logical Backups
Performing a logical backup
Performing a logical restore
Auditing using bcp
Physical Backups with a Storage Manager
Recovering Your Database
Recovering from a Disaster
Restoring from Backups
The online database command
Restoring to a specific time
Restoring from compressed backups
Common Sybase Procedures
Procedure 1: How to Start Sybase
Procedure 2: How to See Whether Your Server Is Alive
Procedure 3: How to Shut Down Your Server
The first thing to try—normal shutdown
The second thing to try—shutdown with nowait
The third thing to try—kill -15 on the dataserver
What you should NEVER do—kill -9 on the dataserver
On Windows
Procedure 4: How to Set Server Configuration Options
Procedure 5: How to Set Database-Level Options
Procedure 6: How to Run a Query
Sybase Recovery Procedure
Step 1: Can You Connect to Your Server Using isql?
Step 2: Run the Stored Procedure sp_who
Step 3: Blocked Processes
Step 4: Log Suspend
Step 5: You Can’t Connect Using isql
Step 6: Check the Sybase Server Error Log
Step 7: Check Whether Your Server Is Running
Step 8: Running Server but Can’t Connect Remotely
Step 9: Restart Your Server
Step 10: Startup Failure
Step 11: Contact Sybase Support Immediately
Step 12: Able to Get Shared Memory?
Step 13: Master Device Failure
Step 14: Disk Device Failure
Get a list of the databases that failed to load
Check your available free space
Get database recreation information
Drop the broken database
Recreate the database
Reload your database
Bring the database online
18. IBM DB2 Backup and Recovery
DB2 Architecture
The Power User’s View
Instance
Databases
Schemas
Tables
Views
Indexes
DB2 engine dispatch units
The DBA’s View
Connecting to a DB2 database
System catalog tables
Database partition
Containers
Tablespaces
Large objects (LOBs)
Transaction logs
Managing archive logs
The backup, restore, rollforward, and recover Commands
The backup Command
Backup levels
Backup path and filenaming convention
Discovering the history of your backup operations
Automatic maintenance
Using db2look
Recovery Types
Crash recovery
Version recovery
Rollforward recovery
The restore Command
The rollforward Command
The recover Command
Recovering Your Database
Performing an In-Place Version Recovery
Step 1: Gather your database backups
Step 2: Make sure the containers that existed during your backup are still around
Step 3: Issue the restore or recover database command
Step 4: Perform rollforward recovery
Step 5: Reorganize the data and collect statistics
Performing a Redirected Version Recovery
Step 1: Restore the database backup and specify the redirect option
Step 2: Define appropriate tablespace containers for the target database
Step 3: Continue the redirected restore operation
Step 4: Perform rollforward recovery
Step 5: Reorganize the data and collect statistics
Performing a Rollforward Recovery
Step 1: Gather your logfiles
Step 2: Determine the minimum PIT
Step 3: Issue the rollforward command
Step 4: Set constraints (if necessary)
Step 5: Perform a database backup (if necessary)
Step 6: Reorganize data and collect statistics
Reorganizing Data and Collecting Statistics
19. SQL Server
Overview of SQL Server
Connecting to and Administering SQL Server
SQL Server Authentication
User authentication
Service authentication
The Power User’s View
Instance
Databases
System databases
User databases
Viewing information about databases
Tables
System tables
Temporary tables
Index
Partitioned tables
Partitioned indexes
Stored Procedures
Memory Management
The DBA’s View
Database Files
Filegroups
Transaction Log
Monitoring logfile size with dbcc
Reducing the size of the physical log
Pages
Extents
Partitions
Table and Index Specifics
Snapshot Backups (2005)
Backups
Backup Devices
Logical and physical devices
Recovery Models
SQL Server 2005
SQL Server 2000
Backup Types
Full
Differential
Transaction log
Copy-only backup
Partial backups
File and filegroup
Backup/Restore of System Databases
Viewing Information About the Backup
Verify Backups
Backup Expiration Date
How to Back Up
Command-line backup with Transact-SQL
Transaction Log Backups
Log truncation
Master Database Backups
Scheduling a Backup
Logical (Table-Level) Backups
Restore and Recovery
Components of a Restore
Recovery Roadmap
Step 1: Check for obvious hardware errors or server problems
Step 2: Can you connect to the instance using a GUI or T-SQL?
Step 3: Can you connect to the master database?
Step 4: Can you connect to a specific, nonsystem database?
Step 5: Initial checks
Step 6: Are any datafiles missing?
Step 7: Is the transaction log full?
Step 8: Is it possible to repair the DB?
Step 9: Before you begin the restore process
Step 10: Restore under the simple recovery model
Step 11: Restore under the full or bulk logged recovery model
Database Restore
Command-line restore with Transact-SQL
Master Database Restore
20. Exchange
Exchange Architecture
Database Structure
Extensible Storage Engine
Stores
Storage Groups
Transaction Logfiles
Checkpoint Files
Reserve Logfiles
General Logfile Info
Circular Logging
Other Files
Single Instance Storage
Automatic Database Maintenance
Storage Limits
Backup
Backup Strategy
Backup Types
Normal
Copy
Incremental
Differential
Daily
Determining What to Back Up
Exchange-specific
Windows-specific
What not to back up when backing up Windows
Backup Methods
Online backups
Offline backups
Streaming backups
Shadow copy backups
Verifying backups
Using ntbackup to Back Up
Making a Basic Backup
Verifying the Backup
Restore
Repair or Restore?
Common Tasks for Repair or Restore
Exchange Repair
Repairing Exchange databases
Repairing by reinstalling
Exchange Restore
Overview
Restoring Exchange Mailbox or Public Folder Stores
Online database restore
Incremental and differential backups
Hard recovery versus soft recovery
Offline Database Restore
Recovery Storage Group
Example of RSG
When to use RSG: Dial-tone restore
Overlooked (and Often Easy) Restore Methods
Restoring deleted mailboxes
Restoring deleted items
Using ntbackup to Restore
Performing a basic restore
21. PostgreSQL
PostgreSQL Architecture
Clusters
Tablespace
Pagefile/Datafile
Startup Scripts
System Tables
Large Objects
Rollback Process
Write Ahead Log
Backup and Recovery
Using pg_dump with pg_restore
Backing up with pg_dump
Restoring with pg_restore
Using pg_dump with psql
Using pg_dumpall with psql
Point-in-Time Recovery
Creating a Backup to Use with Point-in-Time Recovery
Restoring from a Point-in-Time Backup
22. MySQL
MySQL Architecture
Shared Architectural Elements
The power user’s view
Instances
Startup scripts
Databases and tablespaces
Large objects
Binary log
MyISAM Storage Engine
InnoDB Storage Engine
Concerns about InnoDB
Transactions
Tablespace
Datafile
Rollback segment or log group
Transaction log
Other Storage Engines
MySQL Backup and Recovery Methodologies
SQL-Level Backup and Recovery
Backing up MyISAM tables
Backing up InnoDB tables
Repairing corrupted MyISAM tables
SQL-level MySQL restores
File-Level Backup and Recovery
Build your own file-level backup
File-level backup with mysqlhotcopy
Restoring from file-level backups
Using Point-in-Time Recovery
Directly applying binary logs
Applying binary logs via temporary SQL files
Applying binary logs using date ranges
MySQL Cluster Hot Backup and Recovery
Initial configuration
Performing a backup of MySQL cluster
Restoring a MySQL cluster
6. Potpourri
23. VMware and Miscellanea
Backing Up VMware Servers
VMware Architecture
VMware Backups
Back up virtual machines as physical machines
Back up suspended virtual machine files
Copy/export a running virtual machine using VMware’s tools (ESX only)
Using Bare-Metal Recovery to Migrate to VMware
Volatile Filesystems
Missing or Corrupted Files
Referential Integrity Problems
Corrupted or Unreadable Backup
Torture-Testing Backup Programs
Other warnings
Conclusions
Using Snapshots to Back Up a Volatile Filesystem
How do snapshots work?
Available snapshot software
Demystifying dump
Dumpster Diving
Pass I
Pass IIa
Pass IIb
Pass IIc
Pre-Pass III
Pass III
Pass IV
Post-Pass IV
Summary of dump steps
Answers to Our Questions
Question 1
Question 2
Question 3
Question 4
Question 5
A Final Analysis of dump
How Do I Read This Volume?
Prepare in Advance
Wrong Media Type
Bad or Dirty Drive or Tape
Different Drive Types
Wrong Compression Setting/Type
The Little Endian That Couldn’t
Block Size (Tape Volumes Only)
Determine the Blocking Factor
AIX and Its 512-Byte Block Size
Unknown Backup Format
Different Backup Format
Damaged Volume
Reading a “Flaky” Tape
Multiple Partitions on a Tape
If at First You Don’t Succeed...
Gigabit Ethernet
Disk Recovery Companies
Yesterday
Trust Me About the Backups
24. It’s All About Data Protection
Business Reasons for Data Protection
Mitigating Risk
Data availability
Internal/external security
Regulatory compliance
Reducing Costs
Downtime costs
Electronic discovery
Security breaches
Data classification
Improving Service Levels
Technical Reasons for Data Protection
Device Issues
Disk failures
Tape media wear, stolen/misplaced tapes
Networked storage risks
External Threats
Viruses
Worms
Trojan horses
Accidental deletion
Intentional deletion
Backup and Archive
What Needs to Be Backed Up?
What Needs to Be Archived?
Examples of Backup and Archive
Can Open-Source Backup Do the Job?
Very Active Filesystems
Very Large Filesystems
Filesystems with Too Many Files
Information Stored in Databases
Information Stored on Shared Storage
SAN-based filesystems
NAS-based filesystems
Disaster Recovery
Everything Starts with the Business
Define the Core Competency of the Organization
Prioritize the Business Functions Necessary to Continue the Core Competency
Correlate Each System to a Business Function, and Prioritize
Define RPO and RTO for Each Critical System
Create Consistency Groups
Determine for Each Critical System What to Protect from
Determine the Costs of an Outage
Plan for All Types of Disasters
Prepare for Cost Justification
Storage Security
Plain-Text Communication
Poor Authentication and Authorization Systems
Backup Flaws
Conclusion
Index
About the Author
Colophon
← Prev
Back
Next →
← Prev
Back
Next →