Transferring Operations Back To Primary Sites - HP xp12000 User Manual

information transferred via host failover software or comparing contents of the S-VOL with other files
confirmed to be current (such as database log files). These S-VOLs should be recovered using files
confirmed to be current.
Remember that the Continuous Access XP Async time-stamp and C/T information indicates MCU service
processor time, not host system time.
NOTE: Actual data recovery must be done using recovery point data in the database redo log.
Table 29 Determining Continuous Access XP Async S-VOL consistency based on its pair status and
consistency status
Status of
S-VOL
PAIR
COPY
SMPL
PSUE-Group
PSUE-LU

Transferring operations back to primary sites

After applications are running at the secondary (remote) site, the next activity is to restore the primary
(main) site and transfer operations back to the primary site.
1. Bring up the host servers at the primary site, and ensure that all Continuous Access XP components are
fully operational.
2. At the primary site, delete all Continuous Access XP pairs at MCUs. The Delete Pair by Force option
must be Yes because previous S-VOLs are in the SMPL state at the secondary site.
3. At the main site, delete all Continuous Access XP Async consistency groups at MCUs.
4. At the main site, delete RCUs. Remember to connect with each MCU and each CU image to ensure that
all RCUs are deleted.
5. At the main site, configure MCU ports as needed. If you plan to use the same remote copy connections
to copy back, change existing RCPs to LCPs, or change existing initiator ports to ordinary target ports.
If you plan to use the same channel extenders, change the operating mode to the opposite direction.
6.
Boxes and nodes connected to MCUs must be set to channel-mode. Boxes and nodes connected to
RCUs must be set to device-mode.
7. At the remote site, configure appropriate ports to enable Continuous Access XP operations in the
reverse direction (change target ports to initiator ports for fibre, change LCPs to RCPs for serial). This
enables the original RCUs to send Continuous Access XP remote copy operations to the original MCUs
to bring the original P-VOLs up to date.
Usable for Description
recovery?
No These states do not usually occur during Continuous Access XP Asynchronous
disaster recovery because the RCU suspends all Continuous Access XP Async
S-VOLs when communications with the MCU is lost. Continuous Access XP Async
S-VOLs in these states should not be used for disaster recovery.
Simplex volumes cannot be distinguished from S-VOLs already deleted by the
pairsplit-S operation.
Yes Update sequence consistency across these S-VOLs is ensured at the point in time
indicated by the consistency time. These S-VOLs can be used for disaster recovery
at the secondary system.
Updates performed at the primary system after the indicated consistency time
were probably lost.
No Contents of this S-VOL might be behind other S-VOLs in the consistency group. If
this volume must be consistent with other volumes in the same group, this S-VOL
should not be used for disaster recovery.
The cause for this status is:
• The Continuous Access XP Async Error Level pair option for this pair is LU (not
Group).
• This pair was suspended before disaster or failure, at the beginning of the
rolling disaster, or during the initial copy operation.
Continuous Access XP user guide for the XP12000/XP10000 113