Computer Performance

 

Circular Logging in Disaster Recovery

Best Practice (Litmus Test)

Professionals:  Think of disaster recovery, they disable circular logging 99% of the time

Amateurs: Think Circular Logging is best practice for database logs


Why is Circular Logging not best practice?

During disaster recovery, if you have over-written the transaction logs then you can only restore as far as the last backup.  When all the logs are available, Exchange automatically replays the logs and recovers all the transactions.

What is Circular Logging?

Exchange, SQL and Active Directory databases all rely on transaction or write-ahead logs.  Events can be quickly written to the logs, then later 'committed' to the main database file.  Circular logging over-writes these transaction logs to save disk space.

 Circular Logging

  No Circular Logging

 

Log Numbers   Disk Usage

1 2 3 4 5                 25 MB

2 3 4 5 6                 25 MB

3 4 5 6 7                 25 MB

 

 

Log Numbers    Disk Usage

1 2 3 4 5                  25 MB

1 2 3 4 5 6               30 MB

1 2 3 4 5 6 7             35 MB

   

How Transaction Logging Works

When a new piece of mail arrives, a transaction is written to the log.  If the server is busy there will be a delay before the information is committed to the main EDB database file.  A separate pointer called a checkpoint file maintains a record of how much of the log has been written into the EDB database.

Disaster Recovery of Exchange Stores

Each Exchange Storage Group has one log for up to five individual stores.  The best practice of having multiple stores is to reduce the time needed to restore in disaster recovery conditions.

When disaster strikes and you have to restore an exchange store you need a full backup of the EDB database and all the logs since the the last backup.  So if you allow circular logging to over-write some of the logs, then you lose all data since the last backup.  However, if you disable circular logging, then you Exchange magically replays all the logs and brings back the Exchange store to how it was before the disaster.  This re-reading the logs is called a hard recovery and happens automatically.

Where do you check the circular logging setting?

  1. Open the Exchange Administrator, double-click Servers.
  2. Select the server which has the storage group you want to enable circular logging.
  3. Right-click the storage group, and then click Properties.
  4. On the General tab, click Enable circular logging, and then click Yes.

Why does such a potentially harmful setting exist?

The one time you may need circular logging is if your disk is full.  When you first install a database you always think 'no way will the disk ever get full' - experience teaches it will!


Download my Jumbo Litmus Test eBook $5.95

Litmus TestsOver 40 of Guy's litmus tests.  Have fun while you learn about Windows Server 2003.  Stacks of ideas to check your servers, networks and security.

Your eBook has printer friendly pages and lots more screen shots.

 

 


 

Sign up to my new Ezine and get a free Best Practice ebook. 

 

 *


Google

WebComputerperformance.co.uk

GFi Events Manager

Guy Recommends: GFi EventsManager

Here is a solution to monitor, manage and archive thousands of events that are generated by devices across your entire network.  Get your free evaluation copy of GFI EventsManager.

 

Home Copyright © 1999-2008 Computer Performance LTD All rights reserved

Please report a broken link, or an error.