In the Specify Transaction Log Backup Disk Directory screen, select the Use this directory check box and navigate to the primary server's transaction-log folder. Make sure the specified Remove files older than duration is sufficient to keep the log files on the primary server until you can back them up as part of your regular backup strategy. If you specify too short a duration and if the copy job on the secondary server fails, the primary server's SQL Agent job will delete transaction-log files before SQL Server can copy them to the secondary server, and log shipping will fail. Specify the default .trn extension for the transaction-log files, which the plan will name according to the format dbname_tlog_yyyymmddhhmm.trn.
The next screen, Specify the Transaction Log Share, appears only when you've specified that the plan will install log shipping. On this screen, you must identify the file share on the primary server. You can use the ellipsis (...) button to navigate to the file share.
The Specify the Log Shipping Destinations screen lets you add your secondary server or servers, one at a time. Click Add to open the Add Destination Database dialog box, which Figure 3 shows and in which you enter all the secondary-server information. The Server Name text box shows the secondary-server name you registered in Enterprise Manager as part of your preparation steps. In the Directory text box, enter the name of the secondary-server directory that will receive copies of the source database's transaction logs. This name is a local pathname, not a file share. You can choose the option to Create and initialize new database on the secondary server, in which case SQL Server will copy and restore the initial database backup. Or you can choose Use existing database (No initialization) if you've already performed the database restore. If your database is large, you might want to make your own backup and restore it during off-hours.
The destination database must be in a nonrecovered state for SQL Server to restore transaction logs to it. You have two options regarding the database load state: No recovery mode or Standby mode. No recovery mode means that users can't query the database; transaction-log restores will be the only activity that occurs on it. Standby mode leaves the database in a read-only state so that you can query it when you aren't restoring transaction logs. The Add Destination Database screen also presents an option to Terminate users in database (Recommended) during the database restore (if you're using an existing database) or during transaction-log restores. During a database or transaction-log restore, the restore process is the only "user" permitted in the database. Therefore, Microsoft recommends that you use this option because the presence of other users in the database will delay the restore.
You might also want to select the Allow database to assume primary role option, which lets the destination database become a new log shipping source database and thus permits a possible future role reversal between the primary and secondary servers. When you select this option, specify the secondary server's transaction-log file share as the location for transaction-log backups from the new source database.
On the next screen, Initialize the Destination Databases, you can choose to use a recent backup or to make a new one. For large databases, using an existing backup might be more convenient. However, all the transaction logs that occurred since that backup must reside in the primary server's transaction-log file share, in the correct naming style, so that the wizard can copy and restore them to the secondary server. For smaller databases, letting the wizard generate the backup is much easier.
On the Log Shipping Schedules screen, you set the frequency of transaction-log backups for the source database and of the SQL Agent copy and load jobs that the wizard creates on the secondary server. Log shipping frequencies can be as granular as once per minute, but once every 5 minutes is a more common choice for larger databases.
The Log Shipping Thresholds screen presents options to set the acceptable latency times for transaction-logbackup and copy and restore jobs. When the threshold times are exceeded, the Log Shipping Monitor dialog box on the monitor server raises an alert.
Speaking of the monitor server, the next screen, Specify Log Shipping Monitor Server Information, lets you specify which server to use as your monitor server. Be careful: You might be tempted to simply accept the default, but that default is often the primary server. Typically, you don't want to use the primary or secondary server as a monitor server because you won't be able to determine the current state of log shipping if either of those servers becomes unavailable.
After a few reporting screens, the wizard's final screen lets you choose a plan name. I suggest that for clarity, you put the phrase log shipping somewhere in this name. You might not want to include the server names in the plan name if you want to perform a future server-role change. Click Finish. Now, the wizard automatically sets up log shipping from your primary to your secondary server and also sets up the Log Shipping Monitor on the monitor server.
Prev. page
1
[2]
3
4
next page 
