3PAR InForm OS 2.3.1 MU4 Release Notes (QL226-96261, September 2011)

Table 1 Modifications to the InForm OS (continued)
DescriptionItemBug ID
A new property ID has been added to the TPD_StorageHardwareIDCollection
class and a new association has been implemented between
TPD_AuthorizedPrivilege and TPD_StorageHardwareIDCollection.
The SPCAllowsNoTarget and SPCAllowsNoInitiators properties in
ProtocolControllerMaskingCapabilities did not properly represent
association with target and initiator ports.
SPCAllowsNoTarget
and
SPCAllowsNoInitiators
58796
The SPCAllowsNoTarget and SPCAllowsNoInitiators properties in
ProtocolControllerMaskingCapabilities have been set to TRUE indicating
that an SPC can have no associated target port or initiator port.
improperly set for
associated target and
initiator ports in the CIM
API.
During a power or multi-node failure recovery, the operator was required to issue a
setsysmgr force_iderecovery in the following situations.
Operator intervention is
required after a power
fail or multi-node failure
recovery.
58852
• When all but one node had saved data without initiating the proper handshaking
with the missing node.
• When the system was in the process of creating a volume (base or snapshot) at
the time when the save occurred.
Operator intervention is no longer required in these situations, and the system recovers
the data automatically.
In some cases, the size for a virtual volume and snapspace resulting from a reclaim
operation were not being properly evaluated and processed by the metadata checker.
Virtual volume size or
snapspace reclamation
58886
For a virtual volume, if the metadata view of size is smaller than the actual size, the
meta data checker now fails the volume. Also when checking snapspace due to a
provided by System
Manager is not properly
evaluated.
reclaim operation, if the metadata view cannot account for all the holes, the meta
data checker fails the volume.
During a powerfail (or multi-node panic) recovery, if the system encountered problems
paging in volume metadata, the recovery caused all the nodes to panic. In this case,
Operator intervention is
required after a power
58939
when the system came back on-line, it required operator intervention in the form of
a setsysmgr force_iderecovery command to proceed with the recovery.
fail or multi-node panic
recovery.
If the metadata page fails due to disk unavailability, the system now automatically
proceeds with recovery and operator intervention is no longer required.
If the firmware on the 4-port 4Gb fibre channel HBA (3PAR FC044X) received a
fatal error while IO was ongoing to disk drives via the HBA, the firmware would not
Multiple disk drives are
marked failed after a path
59114
respond for several seconds while the transmit and receive lasers were still active.goes down due to HBA
fwcore dumps. This could lead to the disk drives becoming unresponsive for several minutes resulting
in the drives being marked failed by the storage system. If disk cages were connected
to multiple ports of the HBA, this could result in multiple drives getting marked failed
leading to virtual volumes getting marked unavailable.
This has been resolved by having the firmware turn off the transmit and receive lasers
when it gets a fatal error thus allowing the drives to detect the link failure immediately
and recover without becoming unresponsive.
A System Manager restart occasionally ran into problems following an online
upgrade.
System Manager restart
encounters problems
59274
System Manager restarts are no longer affected by online upgrades.
following an online
upgrade.
When attempting to rebalance the load after a controller node rejoins a cluster, out
of resource conditions could cause the system to go off line.
The system goes off line
when a controller node
60141
Resources are now evaluated prior to initiating the load rebalancing process after
a controller node rejoins a cluster.
rejoins a cluster due to an
inefficiency in
rebalancing when
encountering insufficient
resources.
HP 3PAR InForm Operating System Release Notes 7