HP StorageWorks XP Remote Web Console Software - Manual (Page 150)

150 Hitachi TrueCopy z/OS for HP XP12000/XP10000 and SVS 200 storage systems
To transfer operations to the main site:
1. Bring up the host system at the main (primary) site and ensure that all TC390 components are fully
operational.
2. At the main site, delete all TC390 pairs at the MCUs. The Delete Pair by Force option must be used
because the previous R-VOLs are now in the simplex state at the remote site. Use the Delete All Pairs
option to delete all TC390 sync pairs in each CU image. Use the Delete-Group option to delete all
TC390A pairs in each group. Make sure to connect with all MCUs and all CU images to delete all
TC390 pairs.
3. At the main site, delete all TC390A consistency groups at the MCUs.
4. At the main site, delete the RCUs. Remember to connect with each MCU and each CU image to verify
that all RCUs have been deleted.
5. At the main site, configure the MCU ports as needed. If you plan to use the same remote copy
connections to copy back, change the existing RCPs to LCPs and change the existing initiator ports to
RCU target ports. If system option mode 114 is enabled on all MCUs and RCUs, the ports will
reconfigure automatically if you use the CESTPATH command to add the pairs at the secondary site.
6. If you plan to use the same channel extenders, change the operating mode to the opposite direction.
The boxes/nodes connected to the MCUs must be set to channel-mode and the boxes/nodes
connected to the RCUs must be set to device-mode.
7. At the remote site, configure the RCU ports to enable TC390 operations in the reverse direction (change
LCPs to RCPs for serial, change RCU target ports to initiator ports for fibre). This enables the original
RCUs to send TC390 remote copy operations to the original MCUs to bring the original M-VOLs up to
date. If system option mode 114 is enabled on all MCUs and RCUs, the ports will reconfigure
automatically if you use the TSO CESTPATH command to add the pairs at the remote site.
CAUTION: Before issuing the CESTPATH command, verify that the relevant paths are offline from
the host(s) (for example, configure the Chipid offline, deactivate the LPAR, or block the port in the
ESCD). If any active logical paths still exist, the add path operation will fail because the port mode
(LCP/RCP) cannot be changed.
8. At the remote site, establish the same TC390A groups and TC390 pairs in the reverse direction to
synchronize the original M-VOLs with the R-VOLs. Make sure to use the Entire TC390 initial copy option.
The following table shows the correct timer types for performing TC390A operations in the reverse
direction.
Resuming normal operations at the main site
After the TC390 pairs have been established in the reverse direction, you are ready to resume normal
operations at the main site. Remember that the TC390 terminology is now reversed: the original RCUs and
R-VOLs (remote site) are now the MCUs and M-VOLs, and the original MCUs and M-VOLs (main site) are
now the RCUs and R-VOLs.
To resume normal operations at the main site:
1. At the remote site, verify that all TC390 pairs are in the duplex state. This indicates that the TC390
initial copy operations are complete.
Table 28 Selecting the correct timer type for TrueCopy Async disaster recovery
Original Configuration Timer Type for Copy Back
MCU-RCU Configuration Timer Type I/O Time-Stamp Function at
Secondary System?
Yes
No
n-to-1 (n>1) System None None
1-to-1 System System None
Local Local Local