Backing Up the Central Site

Backing up the central site is unique in the following ways:

  • It is the busiest site in the hierarchy, leaving less time available to back it up.

  • It has the largest amount of data that needs to be backed up, which increases the length of the backup operation.

As a result, you might attempt to reduce the risk of backlogs resulting from backup operations on the central site by planning to back up your central site less often. Because the central site is the most critical site in the hierarchy, this is not recommended.

Recovering the central site is unique in the following ways:

  • On the central site, many configuration changes can be constantly made, and it might be impossible to repeat all of them.

  • The central site does not have a parent site from which to obtain a copy of the site control file.

To resolve these issues, and to ensure that it is possible to recover all the data of the central site, follow these recommendations:

  • Back up the central site frequently.

  • Have a reference site for the central site, so that recovery tools can recover packages and advertisements created after the site is backed up.

  • Have a copy of the that is as recent as possible.

    • Between site backup cycles, back up the central site control file frequently. You can use system tools or batch files to automatically back up the site control file frequently. To back up site control files using the Hierarchy Maintenance tool (PreInst), see the "Recovery and Repair Tools" section later in this chapter.

    • Store the site control file backup at the designated reference site.

    • Store the site control file with a *.CT0 format so that during recovery it is sufficient to rename the file, and drop it on the recovering central site.

If you follow these recommendations, then you will be able to recover most of the configuration when recovering the central site. You will be able to recover configuration data and software distribution objects definition from the reference site. The administrative data that is lost if the central site fails is less critical, and is regenerated after the central site regains functionality.

For More Information

Did you find this information useful? Please send your suggestions and comments about the documentation to smsdocs@microsoft.com .