Pillar Axiom Replication Engine (MaxRep)
 

PLA-Support>Sun Systems>DISK>Axiom>SN-DK: MaxRep-2x
 

In this Document

Goal

Solution

Pre-Reversal Steps

Reversal Steps

Post Reversal Steps

References

Created from <SR 3-11746780301>

Applies to:

Goal

This document explains how to reverse the direction of a replication pair for a Protection Plan.

This document can assist customers that have asynchronous implementations with an FS1-2/Replication Engine at a primary or source site, and a FS1-2/Replication Engine at a secondary or destination site, and want to reverse the direction of a replication pair for a LUN at the primary site that is being replicated to a LUN at the destination site.

Once completed, the primary or source site and secondary or destination site roles will be reversed.

Solution

Pre-Reversal Steps

During the initial configuration of a Replication Engine, a retention LUN is created on the secondary site.  A retention LUN is a LUN created on the FS1-2 but mapped by the Replication Engine.  It holds the retention journal for the Replication Engine. The retention journal contains a list of time indexed replication events that allow rollback to any point in time. 

In order to have the ability to reverse the replication direction, a retention LUN is required to be mounted on both the primary (source) and secondary (destination) FS1-2 Systems.  ssh to the primary Replication Engine (see Document 2046703.1 FS System: Passwords Associated with the Oracle FS1-2 Flash Storage system) and run the mount command to see if a retention LUN is mounted:

If the retention LUN is not mounted, one may need to be created. 

1. Check to see if a retention LUN exists on the primary FS1-2:
   
   1. In the MaxRep GUI, select the Protect tab, then View Oracle Storage in the navigation tree on the left.
   2. Use the Filter option to select the primary FS1-2.
   3. Expand the UnMapped storage for a list of LUNs.
   4. If the retention LUN does not exist, proceed to step 2 to create, otherwise proceed to step 4 to mount it.
2. Use the retention LUN on the current secondary system as a model to create one for the primary FS1-2. 
3. Rediscover LUNs from the primary FS1-2's Replication Engine:
   1. From the Settings tab, select Manage Oracle Storage under Oracle Storage section of the navigation tree
   2. Select the Re-Discover option from the primary FS1-2
      
   3. Select OK in the pop-up confirmation window.
4. Using the MaxRep3 User Interface, map and mount the retention LUN from the primary FS1-2.
   1. In the MaxRep GUI, select the Settings tab, then Toolkit for MaxRep under Oracle Storage in the navigation tree on the left.
   2. Select the radio button next to Map and then Next>>.
   3. Select the primary FS1-2 for Oracle Storage and the primary Replication Engine for MaxRep.
   4. In the window below, expand the UnMapped LUNs and check the box next to the retention LUN and Next>>
   5. Create a mount point, check the format box and then Submit.
5. Repeat the mount command above to confirm it is now mounted.
   
   Note: Do NOT map the retention LUN from the Oracle FS System Manager GUI.

Reversal Steps

Below is the basic procedure to reverse direction of a replication pair.

1. Halt SAN Host I/O to the source LUN at the Primary Site, this includes making sure the San Host has properly released any SCSI reservations it may have on the LUN.
2. From the Replication Engine running the Control Service, perform a Rollback for a Protection Plan.
1. Create a Rollback Scenario that allows you to recover data at any historic point-in-time that has been stored in the retention period. The rollback process promotes the target LUN at the Destination Site as the new source LUN. You can then mount the LUN to a SAN Host for access.
1. In the MaxRep UI select the Recover tab then click "Create Rollback Scenario" under Data Validation and Backup in the navigation tree on the left.
2. Select the Protection Plan for the Rollback Scenario. Select the primary server and failover server for the Rollback in the Select Protection list, click "Next>>".
3. Verify the Rollback Options in the Pair Details listing and click "Next>>".
4. Verify the Rollback Plan and Rollback Scenario details and click Save.
2. Run a Rollback Scenario to recover a LUN from a specific point in time of a recovery period.
1. In the MaxRep UI navigate to Recover tab -> click "Manage Backup/Rollback Scenarios".
2. Select the appropriate Plan Name and under Recovery Scenario select Rollback from the pull down menus.
3. From the Recovery Scenarios table click "Run" for the protection plan that contains the desired scenario.
4. There are 2 Options for recovery using "Consistent Point Based Recovery" or "Time Based Recovery". Choose one of these Recovery Options.
   • Consistent Point Based recovery:
     
     NOTE: This option requires an application consistent agent and are using it to create bookmarks.
      
     
     1. Under Tag Details, click on Select.
     2. In the Execute Plan window, select the appropriate options based on your application:
        
     3. Select the appropriate Application Agent in the Search Results and then Next>>.
   • Time Based recovery:
     1. Under Tag Details, click on Select.
     2. Use the slide bar below the graph to select the Start Time and End Times
     3. Select Regenerate Graph
     4. Clicking inside that regenerated graph brings back the previous screen.  You should see data in the Recover Time cell:
        
         
     5. Click Next>> to continue
5. Confirm your selections and click "Run".
   
   NOTE: Once you click "Run" the Rollback of the target volume will stop replication, recovery, associated snapshots, and also will invalidate various policies associated with the replication pair. The LUN rollback is an irreversible process.
6. Select OK twice to continue and delete the retention logs.
7. Wait for the Rollback In Progress to complete, you may need to refresh the browser window.  Once complete the Execution Status will display "Active (Rollback Done)"
8. In the MaxRep UI click "Monitor" tab and select the Protection Plan. The protection status will show "Active (Pairs Delete Pending)", and the Status of the Protection Plan will show "Deletion pending". Wait for the process to finish, when it's done the Protection Status will change to "InActive" then proceed to Step 3.
3. Verify that the SAN Host at the target site can mount the destination LUN for data access.

Post Reversal Steps

1. Modify the Protection Plan and configure it so the target LUN is the source, and the source LUN is the target.
   1. In the MaxRep UI click "Protect" -> click "Manage Protection Plan". In the main page click "Modify".
   2. In the Modify Protection Options, choose "Click Here" to Modify Primary Server.
   3. Provide a Description and update the selection under Select Primary Oracle Storage to reflect the original target FS1-2.
   4. Expend the Unmapped LUNs and check the box next to the Primary LUN which was the original destination LUN.
   5. Select the Primary Replication Engine which is the MaxRep Host at the original target site.
   6. Click "Next>>"
   7. In the next screen, select the Secondary Oracle Storage which is the FS1-2 at the primary site.
   8. Select the Secondary LUNs which is the original source LUN.
   9. Select the Secondary Replication Engine which is the MaxRep Host at the original primary site.
   10. Click "Next>>"
   11. At the next screen select the appropriate Replication Options and click "Next>>"
   12. At the next screen configure the Retention Policy Settings and under section "Specify Retention Storage Path" choose the retention volume path at the original source site.
   13. Click "Next>>" and OK to confirm.
   14. At the next screen click "Save and Activate".
       At this point the Activation Status for the Protection Plan will display "Prepare Target Pending", once done the status will display "Active".
2. Once the Activation Status is "Active" click "Monitor" tab -> under "Protection Health" choose the Protection Plan.
3. In the main screen under the Protection Plan Status it will display "Resyncing (Step I)" and a "Resync progress" increases in %.
4. As the resync progresses the status will transition to "Resyncing (Step II)", then the protection plan Status will transition to "Differential Sync" and the protection plan will display an overall health of Green. At this point replication will occur in the opposite direction.
5. Under the Monitor tab, the Resync Required alert associated with this process can be deleted.

This solution has no attachment