MU3 Release Notes

ManualsBrandsHP ManualsComputer AccessoriesHP 3PAR Inform S800/4x400GB Magazine LTU

DescriptionItemBug ID

TYPE field. InForm OS is updated to reflect the new SCSI

specification.

The System Manager was failing when snapshot space was

removed from a Virtual Volume via the "setvv -snp_cpg ''"

Preventing System Manager failure

when removing snapshot space via

setvv

67897

command. This issue is resolved, and the command will now

either throw an error (if, for instance, there are still snapshots),

or work as expected.

A bug in Emulex LPe12004 FW prevents the port from responding

to ARB(FF) frames sent by the switch, and eventually the switch

issues a LR which causes the link to go down.

Emulex LPe12004 FW issue68007

In earlier versions, when a node-failure-analysis file is generated

by NEMOE, the file is automatically downloaded to the master

Sendhome NFA files and NEMOE core

dumps

68405

node and a service alert is generated. With this enhancement

now, we also notify the SP to send the NFA file home. This also

applies to NEMOE core dumps.

When an ATS request happens on a page with pending DCOW,

the ATS handling path does not resolve the pending DCOW

which causes data inconsistency.

Import of a snapshot in ESX fails with

invalid meta data when VAAI ATS is

enabled

69127

In the process of rolling an upgrade, the system is split into legacy

clusters and expedited clusters. It is possible to receive a skip

Handle incorrect packet sent during

rolling upgrade

70099

packet (special packet telling remote queue to wrap) from the

expedited cluster to the legacy cluster which the legacy cluster

does not understand. This issue is resolved now.

In earlier versions of InForm OS, manual clearing of SCSI

reservations could result in different reservation data being held

Ensure consistency when clearing SCSI

reservations

70115

on different controller nodes of the cluster. The handling of this

scenario is enhanced to ensure that the coherency of reservation

data is preserved during clearing of reservations.

This will prevent the System Manager termination when a CPG

has no space to grow when region mover is called via movereg

Fix for System Manager Termination

when using region mover

70119

function. Adaptive Optimization is the primary user of the

movereg function.

The recovery mechanism is changed from resetting the controller

node to resetting the HBA port for a rarely seen HBA firmware

error.

Change recovery mechanism for

firmware error from node reboot to port

reset

70195

If more than 16 host WWNs are registered for SCSI3 reservations

on a volume, online upgrades abort and offline upgrades fail to

start the system after upgrading.

Fixed aborted upgrades when more

than 16 WWNs register for SCSI3

reservations on a volume

70428

If power fails during node-down recovery, it is possible that Power

Failure recovery needs to recover log pages being processed

Power Failure recovery may need to

recover processing log pages from

more than 16 virtual volume families

70483

from more than 16 virtual volume families. This is resolved by

dynamically allocating slots for recovering log pages being

processed from various virtual volume families.

This STEC MACH8 FW relase consists of two FW version updates

beyond FW 2573: FW 2803 and FW 2806.

FW 2803 contains the following changes:

STEC MACH8 Firmware update70541

• Extended dynamic block selection

• Modified SMART attribute 0xB2

• Enhanced Read Verify command

• Enhanced NCQ error handling

• Check ECC errors when writing to the flash during shutdown.

• Resolved the incorrect status for FIFO overflow or underflow

• Added write buffer initialization

Modifications to the InForm OS 39