click to visit StorageSearch.com home page
leading the way to the new storage frontier .....
the SSD Buyers Guide - click to see article
SSD buyers guide ....
storage reliability
storage reliability
are MLC SSDs safe in enterprise apps - recently updated   popular article
are MLC SSDs safe? ....
image shows Megabyte's hot air balloon - click to read the article SSD power down architectures and acharacteristics
SSD power loss ..
broken barrel image - click to see the SSD data recovery directory
SSD recovery ..
image shows Megabyte pouring clone juice onto his garden - click here for the backup software page
backup software ..

Developing a Disaster Recovery Procedure - an article by:- BakBone Software

BakBone Software
Editor's intro:- Buying a new tape library or backup software is not really a lot of use, if you don't invest some time in planning how you will deal with a major loss of data. For example, if your backup media is always left in your tape drive and someone breaks into your office at the weekend and removes your server and tape system, that's too late to start thinking about off-site copies. Or if you do keep off-site backups that may be the time you discover that some critical system files which are needed to bootstrap you back into business were never really backed up at all. This useful article from BakBone Software includes many general ideas which can help you plan, even if you're using someone else's software.
... Tape libraries
Data recovery
auto tiering SSDs
the Top 20 SSD OEMs
Cloud storage - with SSD twists
History of Disk to Disk Backup
Surviving Non-traditional Data Disasters
Preparedness

Preparing for a data loss disaster before it occurs can speed up the Disaster Recovery process tremendously. Will you be prepared when disaster strikes? As corporate data warehouses continue to quadruple annually, the dependency on computers exponentially increases the commercial risks associated with failure of any component of that storage system. Data corruption, viruses, hard disk failure, power failure, accidental or malicious data deletion, theft and natural disasters are all situations that necessitate attention for a meaningful disaster recovery policy.

Typically when a computer crashes, the first and foremost task is for IT to rebuild and reconfigure the system, restoring mission-critical applications without excessive down-time, all while the rest of the organization waits around for their data to be restored. One of the best places to start in developing a disaster recovery plan is in the daily business practices of the company. Critical business process recovery can involve the individual employee who not only participates in writing a disaster contingency plan, but shares ownership and responsibility of the overall process. Overall disaster risk assessment pertaining to monetary, customer, legal and regulatory exposures, as well as interdepartmental dependencies should be considered. Recovery and data freshness windows should also be considered when building a disaster recovery plan to ensure the recovery time objectives can be attained during a full-scale restore operation.

CPU performance and data set sizes have been doubling every 18 months for the last two decades. However, network topology and basic I/O technologies have not kept pace with the same rate of improvement .Over the last decade, overall network performance has improved by a factor of 1.5 every 18 months, while I/O bus performance by a factor of 1.3 every 18 months. During the same period , data and CPU performance has outstripped network performance by a factor of 10 and I/O bus performance by a factor of 30+. Improving the overall data rate and decreasing the volume of data to be moved, all while minimizing data server resources, is the primary objective while architecting a disaster recovery implementation in our 24*7 data access world .

Determine The Needs

For each operating system and application component of your architecture, you should consider the following . These questions should be addressed for each production environment that is supported by IT.
  • What are the possible failure scenarios?
  • What is the critical data?
  • How often should backups be performed?
  • When should a FULL backup versus an Incremental or Differential Backup be performed?
  • Should backups beperformed on-line while users are working, or off-line?
  • Will backups be done manually or via an automatic scheduling process?
  • How can you verify that a backup was successfully performed?
  • How will you determine if the backups are useable?
  • How long will you save backups before reusing the medium?
  • How much time will it take to restore from the last back up?
  • Where will your backup media be stored?
  • Do appropriate personnel have access to backup media for restores?
  • If the IT System Administrator is not available, is there a source for acquiring system passwords and procedures to back ups, and if necessary, perform restores?
  • Do you have a Disaster Recovery policy that incorporates all of the above recommendations?

Recommended Practices An effective backup strategy can only be developed by understanding the organization's data management goals and requirements, and effectively creating guidelines for daily operation . Here are a few common practices that should be considered when developing your Disaster Recovery contingency plans.
  • Develop your backup and restore procedures with appropriate resources and personnel.
  • Test the procedures, ensuring that you can recover data in a timely manner.
  • Ensure you have a designated backup and restore administrator.
  • Backup your entire system volume to prepare for the hopefully unlikely event of a catastrophic disk failure, providing the ability to restore the entire volume in a single operation .
  • For W2K users, backup your local directory services database (active directory) to prevent loss of user accounting and security information.
  • Keep backup logs available to enable faster, granular file restores.
  • Keep at least three copies of the backup media, with one off-site in a controlled environment.
  • Periodically perform trial restorations to verify that files are being backed up properly. Usually trial restorations will uncover any hidden hardware anomalies.
  • Ensure precautions are in place to prevent another system administrator from restoring yours or someone else's data onto the wrong server.

Additional Considerations

  • Keep detailed descriptions of your main system components, such as CPU, memory, video mode settings, drives, SCSI configuration, Network card configuration, tape drive, etc . You will want to include I/O, DMA, IRQ, and SCSI settings in as much detail as possible. This information will help you in the event of a severe hardware failure and you are forced to rebuild or replace the system.
  • Ensure that the following are readily available : O/S distribution media, patches, service paks, OEM specific drivers, NetVault distribution media. Al ways maintain a list of current patches installed on each of the corresponding system environments.
  • Maintain an updated profile of each system configuration including network settings (TCP/IP config), machine names, domain information, account information, disk configuration, disk partition layouts, partition sizes, partition file system format type, O/S patches installed.


backup
Backup software on
STORAGEsearch.com
Megabyte's uncle Spellerbyte was a wizard when it came to software. He'd given Megabyte a magic potion for copying critical data.
figure 1
Select the Data you want to recover (/ ,or c:\ ,or path you wa n t ) . If NT, also select the Registry.

Under the "Restore Options" tab choose the appropriate options. If the client is NT, and you are doing a complete restore of the system, select "Update a c t i ve registry " .
figure 2
Recovering From A Crash
In the event of data or system loss, you must first determine the cause of the problem . If data loss or data corruption occurred, you will need to repair the problem prior to performing a data restore. If the client has suffered a disaster or hardware failure, repairing or rebuilding the system will be necessary. Once the hardware has been repaired or rebuilt, you can continue with restoring the system .Restoring a system that has been damaged to this level or has had its disk(s) replaced, requires the following minimum steps: NetVault Client Restore:
  1. Reinstall the O/S
  2. Install necessary OEM drivers (if necessary)
  3. Install necessary O/S patches (if necessary )
  4. Install NetVault Client
  5. Install NetVault patches (if necessary )
  6. Restore data


NetVault Server Restore:

  1. Reinstall the O/S
  2. Install Necessary OEM drivers (if necessary )
  3. Install necessary OS patches (if necessary )
  4. Install NetVault Server
  5. Install NetVault patches (if necessary )
  6. Configure NetVault devices (Tapes/Libraries)
  7. Scanning in the tape that contains the NetVault database
  8. Restore NetVault database
  9. Restore data


Reinstalling the O/S: Restoring an Operating System is not a trivial task.

Typically backup and restore of individual components is not possible, as they must be saved and restored as a complete data set. You will also need to consider that backing up and restoring system data on a remote machine is not possible, and must be accomplished on an individual basis. In Windows environments, NEVER delete the system " hidden" files contained in the root directory, as a restore will be impossible. Other significant considerations are encrypted files and directories, replicated data restoration, cluster database info and other fault-tolerant options. Remember to take into consideration the disk layout, size and partition information to ensure that you have enough space to restore the data. It is also important to know that the partition should be the same format as the original, as NetVault will restore the file attributes of a file, and these attributes often are supported only on certain types of File system types (i.e. file ownership, permissions, and long file names are supported in NTFS, but only long file names are supported in FAT ) .

Here's a detailed example of an Operating System restoration for an NT environment.

As this example is NT specific, remember that there are other steps required for other non-NT Operating System environments using NetVault.
  • Restore WINS to last known state
  • Restore DHCP leases to last known state
  • Restore remote storage database information
  • Reissue orphaned Server Certificate Services Reinstall Windows Server Media Services
  • Restore Server Active Directory
  • If the machine is a network domain controller, restore SYSVOL info
  • Install OEM specific device drivers.
  • Install O/S specific patches, updates, service paks, etc.
    • Client:–
    • Install NetVault Client software & appropriate Plugins
    • Server:–
    • Install NetVault Server software & appropriate Plugins
    • Install NetVault patches/updates
    • Install NetVault Device Extension Pack
    • Restore NetVault Server database


Configure NetVault devices (Tapes/Libraries): Follow the step outlined in the NetVault User Guide to install and configure the tape library and/or tape device(s).
figure3
Scanning in the tape that contains the NetVault database: Once the devices are configured, you will need to scan in the tape that contains the NetVault database. Load the tape into the device, and right click on the tape. Select "scan" from the drop down menu .
figure 4
Once the scan is complete, you can view the contents of the tape via the restore screen in the NetVault GUI.
Restore the NetVault database: Restoring the NetVault database will be necessary to recover the original configuration, keys, logs, and the media database. In the Restore window of the NetVault GUI, select the appropriate client . Select the appropriate plugin, in this case the NetVault Database Plugin. Choose the appropriate " Save set ". Open the Saveset and select the NetVault database components. There are no "Restore Options" when restoring the NetVault database components. Continue to the " Target Client" and choose the appropriate client, select the appropriate schedule (typically immediate) and "Advanced Options" that you want, then submit the job for restore.
figure 5
Restore data: Use NetVault saveset by setting the appropriate " Target client", "Schedule", and any " Advanced Options", then submit the job for restore. Depending on the amount of data and the network connection speed between the client and server, restore time will vary. Once the restore is complete, reboot the system.

Conclusion

A Disaster Recovery Plan usually cannot be written by the IT Deparment alone and should not be created for a given computer or data center. Typically, effective Disaster Recovery plans are a long-term project that cannot be attempted without standard operating procedures, logbooks, data flow diagra m s, problem isolation procedures and a reliable tape backup rotation schedule. Formulation of a corporate-wide standardized Disaster Recovery plan will result in faster isolation of application bugs, fewer operational mistakes, fewer support personnel and requirements, and overall easier system and application maintenance. Utilizing NetVault as your network backup and restore software provides one of the fastest and simplest data insurance policies available today. ...BakBone Software profile

storage search banner

STORAGEsearch is published by ACSL