Hitachi TrueCopy for IBM z/OS User and Reference Guide (T5211-96001, June 2007)

8-6 Troubleshooting
Hitachi Universal Storage Platform V TrueCopy for IBM z/OS User’s Guide
Table 8-4 provides troubleshooting instructions for the TCzA suspension
conditions caused by the offloading timer async option, the group timeout
options (copy pending and RCU ready), and recordset errors. Hardware
failures which affect the cache storage/shared memory of the MCU or RCU
may also cause the TCzA pairs to be suspended.
Table 8-4 Resolving TrueCopy for z/OS Async Suspension Conditions
Classification Causes of Suspension SIM Recovery procedure
MCU/RCU
hardware
Hardware redundancy has been lost due to some
blockade condition. As a result, MCU-RCU
communication, creating or receiving recordset,
or the staging or de-staging process could not
complete.
The pending recordset cannot be retained because
one side of cache storage or shared memory has
been blocked due to hardware failure.
MCU-Creating/sending recordset failed due to
unrecoverable hardware failure.
RCU-Reading/Settling recordset failed due to
unrecoverable hardware failure.
The drive parity group has been in the correction-
access status while the TCz pair was in pending state.
DB0x
DB1x
DB2x
According to SIM, remove the
hardware blockade or failure.
Re-establish failed pairs (Resume Pair).
MCU-RCU
communication
During the power-on-reset sequence, the MCU
could not communicate with the RCU within the
specified RCU ready timeout.
The RCU could not settle the pending recordset or
could not communicate with the MCU before the copy
pending timeout due to MCU not-ready or inoperative
facilities on the remote copy connections.
DB0x
DB1x
Remove the failed condition at the
RCU/MCU or on the remote copy
connection.
Re-establish failed pairs (Resume Pair).
RIO overload
The unrecoverable RIO (remote I/O) timeout
occurred due to overload of the RCU or the
communication facilities on the remote copy
connections.
No recordset could be sent within the specified copy
pending timeout.
The RCU could not settle the pending recordset
before the copy pending timeout due to overload of
the RIO or the RCU itself.
DB1x Release failed pairs (Delete Pair).
Reconsider the performance resources
necessary, and increase resources as
needed (cache amount, number of MCU-
RCU paths, etc.).
Re-establish failed pairs (Add Pair).
RIO failure
The RIO (remote I/O) could not complete due to
the failure at the RCU.
DB2x
According to SIM generated at the
RCU, remove the failure.
Re-establish failed pairs (Resume Pair).
MCU
planned outage
The TCzA pairs were temporarily suspended due
to a planned outage of the MCU.
DB8x
No recovery procedure is required.
The MCU will automatically remove
the suspension condition during the
next power-on-reset sequence.