Write Protect Mode

Use this page on the TS7700 Management Interface to view Write Protect Mode settings in a cluster.

Write Protect Mode is used primarily for disaster recovery (DR) testing in a scenario where a recovery host that is attached to a DR (non-production) cluster must access and validate replicated production data within the DR location but without the risk of modifying it. If Write Protect Mode is enabled with default settings on a cluster, any host command fails if it attempts to modify volume data or volume attributes through logical devices that are associated with that cluster. However, any host command that is issued to a peer cluster that is not write protected can continue with full read and write access to all volumes in the grid configuration. To allow a test system to create content during a DR test, virtual volume categories can be excluded from Write Protect Mode. Up to 128 categories can be identified and set to be included or excluded from Write Protect Mode by using the Category Write Protect Properties table. You can configure these categories before you enable the Write Protect Mode; configuration information is retained for future tests. You can enable Write Protect Mode on multiple clusters within a grid configuration.

Write Protect Mode settings
Write Protect Mode does not affect standard grid functions such as virtual volume replication or virtual volume ownership transfer.
Table 1. Write Protect Mode settings for the active cluster
Setting Description
Write Protect State The status of Write Protect Mode on the active cluster. Possible values include:
Disabled
Write Protect Mode is disabled. No Write Protect settings are in effect.
Enabled
Write Protect Mode is enabled. Any host command to modify volume data or attributes by using this cluster's devices fails, subject to any defined category exclusions.
Write Protect for Flash Copy enabled
Write Protect Mode is enabled by the host. The Write Protect for Flash Copy function was enabled through the LI REQ z/OS command and a DR test is likely in progress.
Important: Write Protect Mode settings cannot be modified while LI REQ-initiated write protection for Flash Copy is enabled. You must disable Write Protect for Flash Copy using the LI REQ host command before you can attempt to change any Write Protect Mode settings.
DR family
The name of the disaster recovery (DR) family the Write Protect for Flash Copy was initiated against.
Flash time
The date and time at which flash copy was enabled by the host. This mimics the time at which a real disaster occurs.
Disable Write Protect Mode Select this option to disable Write Protect Mode for the cluster.
Enable Write Protect Mode Select this option to enable Write Protect Mode for devices that are associated with this cluster. When enabled, any host command fails if it attempts to modify volume data or volume attributes through logical devices that are associated with this cluster, subject to any defined category exclusions. After Write Protect Mode is enabled through the TS7700 Management Interface, it persists through any outage and can be disabled only through the TS7700 Management Interface.
Note: You can enable Write Protect Mode on a cluster only if it is online and no Write Protect Flash Copy is in progress.
Ignore fast ready characteristics of write protected categories Check this box to permit write protected volumes that were returned to a scratch or a fast ready category to be recognized as private volumes.

When this box is checked, DR test hosts can mount production volumes as private volumes even if the production environment has already returned them to scratch. However, peer clusters, such as production clusters, continue to view these volumes as scratch volumes. This setting does not override the Fast Ready characteristics of excluded categories.

Notelist:
  1. When the production environment returns volumes to scratch that are still needed for read operations on a DR test host you must ensure that the data is available to be read by the DR hosts. To meet this requirement, scratch categories in the production environment can use delete expire hold for a period of time that is long enough to cover the DR test. Otherwise, even if selective write protect is being used on the DR clusters, the data is no longer available to the DR hosts if the production volume is returned to scratch and deleted or reused. One alternative is to avoid running return to scratch processing from the production environment during the DR test. Another alternative in a DR environment that is composed only of TS7700s is to use Flash Copy for DR testing, which is immune to production changes, including any scratch processing, deletion, and reuse.
  2. When this box is checked, if the production host deletes a volume, that volume is also deleted on the DR cluster.
Category Write Protect Properties
Use the Category Write Protect Properties table to add, modify, or delete categories to be selectively excluded from Write Protect Mode. Disaster recovery test hosts or locally connected production partitions can continue to read and write to local volumes while their volume categories are excluded from write protect. These hosts must use a set of categories different from those primary production categories that are write protected.
Note: Categories configured and displayed on this table are not replicated to other clusters in the grid
When Write Protect Mode is enabled, any categories added to this table must display a value of Yes in the Excluded from Write Protect field before the volumes in that category can be modified by an accessing host.
Table 2. Information displayed on the Categories Write Protect Properties table
Column name Definition
Category The identifier for a defined category. This name is an alphanumeric hexadecimal value. Possible values include:
  • The range between 0x0001 and 0xFEFF
  • 0xFFFF
The value 0x0000 cannot be used.
Restriction: Values entered do not include the 0x prefix, although this prefix is displayed on the cluster summary page. Values entered are padded up to four places. Letters used in the category value must be capitalized. This value cannot be modified once created.
Excluded from Write Protect Whether the category is excluded from Write Protect Mode. Possible values include:
Yes
When Write Protect is enabled, volumes in this category can be modified by a host.
No
When Write Protect is enabled, volumes in this category cannot be modified by a host.
Note: Volumes in the insert category (0xFF00) are always excluded from Write Protect. This is so that any volume in the insert category can later be moved to a write protect-excluded category.
Description A description of the category and its purpose. This value consists of 0-63 Unicode characters.
Use the drop-down menu on the Category Write Protect Properties table to add a category or modify or delete an existing category.
Note: You can add up to 128 categories per cluster when all clusters in the grid operate at microcode level 8.50.0.x or later.
Click Submit Changes to save any changes to Write Protect Mode settings.
DR Testing Notes:
  1. It is always recommended to use the Write Protect Mode during a DR test to prevent any accidental DR host-initiated changes to your production content.
  2. During a DR test, it is recommended to avoid housekeeping—or return to scratch processing—within the DR test host configuration unless the process specifically targets volumes only within the DR host test range. Otherwise, even with the Selective Write Protect function enabled, the DR host can attempt to return production volumes to scratch. This can occur because the tape management system snapshot that is used for the DR test can interpret the volumes as expired and ready for processing. Never assume the return to scratch process acts only on DR test volumes. If Write Protect Mode is enabled prior to DR testing and return to scratch processing is run on the DR host, then the Selective Write Protect function prevents the return to scratch from occurring on protected categories. Further, options in the tape management system can be used to limit which volumes within the DR host are returned to scratch. For example, in the DFSMSrmm tape management system, the VOLUMES or VOLUMERANGES options can be used on the EXPROC command to limit volumes returned to scratch. When tape management and write protect safeguards are used, protection against data loss occurs both on the TS7700 and at the host.
  3. It is not necessary to disable return to scratch processing within the production environment as long as the following category settings are used:
    • Delete expire hold option is set to prevent the reuse of production volumes before they are verified within the DR test application.
    • Expire hold is not required when the Flash Copy for DR testing mechanism is used and no TS7700 Tape Attach is present within the DR location, since the flash snapshot is unchanged by any scratch and reuse of volumes in the production environment.
    • Ignore fast ready characteristics option is set to ensure scratched volumes are viewed as private when accessed through the DR test devices.
  4. Write protection extends only to host commands issued to logical devices within a write-protected cluster. Volume ownership and data location are independent of write protection. For example, production access to non-write-protected devices within a production cluster can still alter content through a remote mount to a DR cluster in the Write Protect State.
  5. You can leave a cluster in the Write Protect State indefinitely. Doing so can prevent unexpected host access in a DR location from modifying production content. However, write protect must be disabled in the event of a true failover to the secondary location.
  6. Refer to the TS7700 Redbook at the IBM Support Portal link in the Related information section for additional information. Pertinent chapters are Disaster Recovery Testing Considerations and Disaster Recovery Testing Detailed Procedure.