HP StorageWorks HSG60 and HSG80 Array Controller and Array Controller Software Troubleshooting Guide (EK-G80TS-SA. C01, March 2005)

Troubleshooting Information
49HSG60 and HSG80 Array Controller and Array Controller Software Troubleshooting Guide
DIMM or
cache
memory
controller
chip
failure.
Good. Data loss: Writeback data that
was not written to media when
failure occurred was not
recovered.
Cache policy: Controller A
supports write-through caching
only; Controller B supports
writeback caching.
■ Failover: In Transparent
Failover, all units fail over to
Controller B. In Multiple-bus
Failover with host-assist, only
those units that use writeback
caching, such as RAIDsets
and mirrorsets, fail over to
Controller B. All units with
lost data become inoperative
until they are cleared by
using the
CLEAR_ERRORS
unit-number
LOST_DATA
command.
Units that did not lose data
operate normally on
Controller B.
In single-controller
configurations, RAIDsets,
mirrorsets, and all units with lost
data become inoperative.
Although lost data errors can be
cleared on some units, RAIDsets
and mirrorsets remain
inoperative until the memory on
Cache A is repaired or replaced.
Data loss: Controller A recovers all
of writeback data from the
mirrored copy on Cache B.
Cache policy: Controller A
supports write-through caching
only; Controller B supports
writeback caching.
Failover: In Transparent Failover,
all units fail over to Controller B
and operate normally. In
Multiple-bus Failover with
host-assist, only those units that
use writeback caching, such as
RAIDsets and mirrorsets, fail over
to Controller B.
Table 8: Cache Policies—Cache Module Status (Continued)
Cache Module
Status
Cache Policy
Cache A Cache B Unmirrored Cache Mirrored Cache