HP 3PAR OS 3.1.2 MU5 Release Notes

DescriptionItemIssue ID
switches, show fcns database detail will display this
information. HP's SAN Visibility software will key off this information.
Earlier full VVs used to be in writethru during import, and this impacted
the VV performance as more regions are moved to local LDs over
Improve peer motion performance
for full VVs
65930
time. Now full VV and TPVV volumes are imported in the same way
and the local LDs are no longer in writethru.
The SNMP MIB definitions file used by HP 3PAR is provided on the
release CD as a courtesy for customers that wish to compile the SNMP
Renamed SNMP MIB definitions file66414
MIB. The file has been renamed so that the new name matches the
3PAR MIB name named defined in the file. The name has changed
from cd1/contents/snmp/3PAR-MIB.mib to ThreeParMIB.mib
in the same directory.
In HP 3PAR OS version 3.1.1, a check was added when selecting a
TOC after system reboot to prevent automatically selecting a TOC
more than a day old. this change is reverted.
TOCs more than a day old no longer
invalid choice for automatic
selection.
66436
Under certain conditions, if a node goes down, and an rw snapshot
with zero detect enabled has a partial page write in progress, the
Fixed partial page data inconsistency
issue for rw snapshots with zero tte
66491
rest of the page could get corrupted with data from the parent of the
rw snapshot instead of being set to zero. Fixed node down recovery
to handle this situation properly.
Read write snapshots can now directly be created from Read write
bases or other read-write snapshots. This process will not require an
explicit read-only snapshot creation in the middle.
Support for RW snapshots from RW
bases
66706
creategroupsv now supports creation of RW snapshots from RW bases
(or snapshots), instead of only from RO snapshots.
createsvgroup/createvvset: support
for RW snapshots from RW bases
66707
Fix for bug 64033 broke basic assumption that BBU perm. failure is
always cleared before I2C faults (if any). That caused NEMOE to
Fixed BBU events at insert; a
regression caused by 64033
66727
send same event twice when BBU monitor cleared any I2C event
leaving TPD in inconsistent state with one unresolved event. Added
fix to ignore I2C events for first 10 seconds and made sure that
NEMOE always sends two distinct events when an I2C error to BBU
is cleared.
An issue which may cause an HP 3PAR system to become unavailable
during the initial sync of volumes greater than 400GB has been fixed.
HP 3PAR system can become
unavailable during intial sync of
large volumes
66757
TPD packages have been enhanced in 3.1.1 and differ in structure
to previous patches. In instances where Node Rescue is being
Enhance Node Rescue to handle new
3.1.1 Patches
66787
performed with a patch installed on the HP 3PAR system the correct
image could not be formed The patch merging process has been
enhanced to correctly merge the patches with the base image to create
the correct image to be installed on the HP 3PAR system via Node
Rescue.
IMC will display correct raid type for the volumes. If a r6 type volume
is created, IMC with the fix in 3.1.2, it will display Raid 6 instead of
Raid 0.
IMC still showing RAID0 volumes67016
Code was doing an incorrect check on reductive operators for the
address range when the address being used was at the end of the
NODE0 C-S failed. PANIC: tpd:
Assertion point: file:
drvmain.c, line: 1101
67226
xcb bound area. In this case we should only be checking that the 64B
destination does not exceed the bounded range instead of using the
xcb length.
Upgraded 1G iSCSI firmware to version 3.0.1.67 in order to address
an incorrect bitmask check that could result in port reset due to loss
of response from the firmware.
Firmware could become unresponsive
and cause port resets
67338
During the migration of remote copy volumes using peer motion, the
need arose to create identical snapshots on the primary and secondary
Additional functionality within the
createsv command.
67384
12 HP 3PAR OS 3.1.2 GA Release Notes