Day-To-Day Tasks for Disaster Recovery
Planning for Reliability and Fault-Tolerance
The solution to total disaster recovery consists of the following phases:
Preparing the Hot-Site for the CommServeDisaster Recovery strategy includes the creation and maintenance of a secure hot-site or a disaster recovery site. The hot-site must contain a pre-staged CommServe (with a different host name and IP address), along with all the necessary software and applications. Install the SRM Server along with the CommServe, if applicable. This would help to quickly restore data in the event of a disaster. You can obtain a Dual IP License for the CommServe instead of obtaining an additional license for the CommServe in the disaster recovery site. The Dual IP License has two IP addresses; a primary IP address for a CommServe host computer and secondary IP address for the CommServe staged in the disaster recovery site. This provides the system with the capability to automatically identify the appropriate CommServe when the CommServe meta data is restored in the disaster recovery site. Contact your Software provider to obtain a copy of the Dual IP License before building the disaster recovery site. Creating a hot-site provides you quick recovery option in case of a disaster. However, you can also restore the data back to original machine — the machine with the primary IP Address. This can be done without any license change. When you prepare the hot-site for the CommServe, note that you cannot have Services running on both the CommServes (at the production site and the hot-site) at the same time. In other words, both the CommServes should not communicate with the client simultaneously. The diagram on the right provides a sample representation of a hot-site. |
![]() |
Sufficient resources to rebuild the CommCell, including MediaAgents and critical clients (clients and MediaAgents with different host names), with all the necessary software and applications installed.
Before you recreate your CommCell and clients at the Hot-site, you must set up the environment. Particular care must be taken to recreate your Active Directory at the Hot-site. If there is more than one Domain Controller in your Active Directory environment, roles may be divided among the servers; therefore, all of the servers must be recreated at the Hot-site, or the roles must be consolidated. Refer to the following Microsoft Knowledge Base articles for more information:
Do not assume that you can recreate or bring one Domain Controller from your primary environment and recreate your Active Directory.
A library and drive, compatible with the library and drives in the production environment.
While implementing a disaster recovery site, it would be necessary to re-create the CommServe and several other servers from the production environment.
Before you use your agent, review and understand the associated full system restore (or disaster recovery) procedure. The procedure for some agents may require that you plan specific actions or consider certain items before an emergency occurs.
Considerations for Windows Clients
If you are attempting to recover an upgraded Windows Server 2003 computer (Windows 2000 to Windows Server 2003), you must reinstall Windows Server 2003 to the WINNT directory (except in upgrade scenarios) due to Windows directory protocols. Refer to the Microsoft Knowledge Base for more information. Following is a slower but perhaps surer alternative to reinstalling Server 2003 to the WINNT directory:
Planning for this phase should focus on securing and maintaining the necessary data required in the event of disaster at the hot-site. The following data should be secured and maintained:
The CommServe in the disaster recovery site must be updated with the metadata at all times. The metadata can be restored using the CommServe Recovery Tool. (See Restore a Disaster Recovery Backup for step-by-step instructions.)
The Auxiliary Copy feature can be used to create backup copies of data, including Disaster Recovery backups.
Auxiliary copy operations can be scheduled to create copies for storage using either synchronous or selective copy features. Copies can be made to any media type. Auxiliary copy can be scheduled at regular intervals and media can be removed for off-site storage and/or transported to the DR site on a regular basis. Auxiliary copy can also be performed using a WAN to a tape or network share in the hot-site. See Auxiliary Copy for more information.
This report provides information on Disaster Recovery Backup jobs, including information about the media used by the various phases of the Disaster Recovery Backup.
This report provides information on the various clients and their associated storage policies.
This report provides information on the media associated with each storage policy, including secondary copies which are used to perform the auxiliary copy operation. This report also provides the back up time of the most recent backup available in the media, with the media location.
Media associated with Disaster Recovery Backups and Auxiliary copy operations of the backups can be exported at appropriate intervals, using the bulk tape export feature. This activity can also be automated by defining regular schedules for tape export.
CommServe and MediaAgent(s) can be implemented on a Microsoft Cluster Server to ensure that these servers provide the highest level of availability and reliability in case of a localized server failure.
The physical configuration of a cluster consists of active and passive servers. The active server hosts the services that comprise the virtual server. The passive server, or standby server, monitors the active server for failure, or interruption of service, and will take over hosting the virtual server in case of such events.
See Clustering for more information.
Secondary Storage Policy copies that use a different media and library can be created. Auxiliary copy operations can be scheduled to regularly synchronize the primary and secondary copies. If the primary copy become inoperative, perhaps due to a media, library or network malfunction, a synchronous copy can be promoted to become the primary copy. This allows you to continue operations as before and make repairs without interrupting back up and restore operations.