Chat now with support
Chat with Support

Welcome, Quadrotech customers to Quest Support Portal click here for for frequently asked questions regarding servicing your supported assets.

NetVault Plug-in for Exchange 12.1 - User Guide

Introducing NetVault Backup Plug-in for Exchange Defining an Exchange data protection strategy Planning your Exchange Server deployment Installing and removing the plug-in Configuring the plug-in Backing up data Restoring data Troubleshooting

Protecting Exchange Server databases

A database is the finest granularity of storage organization within the Exchange Server schema. Mailboxes are assigned to specific databases, which allows segregation of data by mailbox for security or scalability purposes. Occasionally, critical or high-volume mailboxes are segregated to a separate database to improve performance or to perform more frequent backups.

Databases are used to contain mailboxes, messages, folder stores, and various other data objects supported by the Exchange Server. While they can contain a large variety of data objects, databases are typically one of two types: mail stores and public folder stores. In Exchange Server terminology, a store is the same as a database.

Microsoft also differentiates editions of the Exchange Server product by increasing support for concurrent databases. The Enterprise Edition is defined as supporting more concurrent databases than the Standard Edition.

Protecting the Exchange Server transaction log

All changes made to the Exchange Server database are first committed to transaction log files. Any time a user modifies data stored in a mailbox or data is added to the mailbox, that change is written to a transaction log file before it is written to the Exchange Server database.

Reviewing the Volume Shadow Copy Service (VSS) backup method

Microsoft supports the ability to create snapshots of Exchange data using VSS. Microsoft provides Exchange-specific VSS writers that coordinate with the Exchange Services, operating on behalf of the Plug‑in for Exchange, to prepare the Mailbox Database files for backups and freeze the input/output (I/O) activity due to Exchange transactions before backing it up, and then to unfreeze and truncate logs after the backup is complete.

Using VSS, you can:

NOTE: The Backup Files to Storage option is supported with any disk-based storage. To use the Retain Snapshot as Persistent and Discard After options, the data that you back up must reside on a NetVault Backup-supported disk array. Also, for persistent snapshots, only the metadata is copied to the target.

Managing transaction log files

When defining an Exchange Server backup strategy, transaction log file management is a primary consideration.

Transaction log file truncation is used to “clean” the Exchange Server Mailbox Database, thus improving performance, and reducing the disk space requirements and the time required to restore a database.

Quest recommends that backups that perform transaction log file truncation be performed regularly. Often, a weekly or semimonthly backup that performs transaction log file truncation is recommended. The optimum frequency might vary considerably, depending on the use and configuration of each Exchange Server.

If you use a Backup Type that supports transaction log file truncation, such as the Full and Incremental Backup types, the truncation of transaction log files is performed by the Exchange Server after the plug-in informs it that the backup completed successfully. Also, when the truncation occurs depends on Exchange Server and whether it still needs the logs for more purposes, such as replication; therefore, truncation might not occur immediately after a successful backup is completed.

In a DAG environment, transaction log file truncation is also determined by the replay lag time and truncation lag time properties of the database. The properties are configurable. Replay lag time defines, in minutes, the amount of time to delay log replay for a database copy. Truncation lag time defines, in minutes, the amount of time to delay log deletion for the database copy, after the log file has been replayed into the database copy.

For Exchange Server to truncate a transaction log file, the following criteria must be met:

In a DAG environment, each database copy keeps transaction log files until all the database copies have confirmed that the transaction log file has been replayed. If one or more passive copies of the database is suspended or offline, log truncation does not occur, resulting in transaction log file buildup and the consumption of disk space.

To reduce the effects of transaction log file buildup due to suspended or offline database copies, Exchange Server 2013 Service Pack 1 introduced loose truncation. With loose truncation, each database copy tracks its own available disk space and applies loose truncation when disk space gets significantly low. With loose truncation applied, each passive database copy independently truncates its own transaction log files. For the active database copy, truncation ignores the passive database copy that is farthest behind replaying logs.

Loose truncation is disabled by default. To enable loose truncation, you must edit the Windows registry on each Exchange Server DAG node. Before enabling loose truncation, make sure that it would benefit your data protection goals. For more information about enabling loose truncation, see your Exchange Server documentation.

Full Backups back up all files for a database, regardless of the type of file. Transaction Logs-Only Backups back up only the transaction log files for a database.

Full Backups allow all database files to be backed up, providing standalone restore capabilities. Depending on the size of the database, Full Backups can be demanding in terms of storage requirements, and time required to complete the backup. For large databases, storage and time requirements might be a significant consideration.

Transaction Logs-Only Backups are lighter-weight backups that are intended to capture new activity since the last Full Backup was performed. This type of backup can radically reduce the backup time and storage requirements for large databases, but it also introduces dependencies on one or more prior backups to perform a complete restore.

Related Documents