Environmental disasters such as floods, fires, and tornadoes are an ever-present threat to data. Statistics show that 50% of businesses that lose their data due to disaster go out of business; of those, 90% go out of business within 24 months. A successful disaster recovery plan not only requires a capable storage platform but also a strictly-adhered-to storage management plan. An example of such a plan is presented below:
1. Archive each file to at least two exclusive sets of media.
2. One set is for near- or off-line storage, the second is for off-site disaster recovery.
3. Once an off-site-destined media becomes full, remove it from the site and transfer it to the off-site storage location.
4. Create a samfsdump of the entire file system and send it along with the media mentioned in step 3 for off-site storage.
5. Depending upon the selected storage media, the media will have to be copied fom time to time to prevent loss due to media failure. This is mostly required for magnetic tape which has a considerably shorter expected storage life than optical disk.
The loss of a single archive copy, due to operational error, is the most likely cause of lost data. Without proper data management and planning, the loss is irrocoverable. After the loss of either a single or small number of archive media, the following procedure can be used:
1. Retrieve the off-site archive media for the files affected. An archive_audit will generate a list of all files resident on the lost media.
2. Delete the archive entry from each file that has been archived to the lost media.
3. Stage each affected file to on-line storage from the off-site archive copies.
4. Allow the migrator to automatically re-archive the files onto new archive media, thus creating a new primary archive copy.
5. Return the off-site archive media to "safe" off-site storage.
In the event of a total disaster (e.g., fire), the following procedure can be used to recover data:
1. Replace damaged equipment or use an off-site SAM-FS system for remote data operations.
2. Retrieve off-site archive media from its remote storage location.
3. Restore the last complete samfsdump file.
4. Delete the primary archive entry for each file in the system.
5. Stage each file to on-line storage.
6. Allow the migrator to automatically re-archive the files on to new archive media, creating a new primary archive copy for each file in the system.
7. Return the off-site archive media to "safe" off-site storage.
The above disaster planning and recovery procedures emphasize the importance of an archiving system, rather than a backup system. To ensure protection against disasters such as flooding and tornadoes it is essential that at least two copies of each archived file be made, and that one of those copies be stored in a somewhat distant "safe" storage area.
For additional information on implementing a disaster recovery program refer to Chapter 5 of the LSC Site Preparation manual (Document No. LS0000XP.) or call us at (612) 482-4535.
(C)1994, LSC, Inc. All rights reserved.
Storage and Archiving Manager (SAM-FS) is a trademark of LSC, Inc. All
other trademarks are the property of their respective owners.
Return to LSC's Home Page.