InForm 3.1.1 Release Notes
InForm 3.1.1 Release Notes
1 Release Notes
Abstract
The information in this document is intended for use by HP customers, partners, and HP field representatives. These release
notes describe the features and fixes included in HP 3PAR InForm OS 3.1.1.
The information contained herein is subject to change without notice. The only warranties for HP products and services are set forth in the express
warranty statements accompanying such products and services. Nothing herein should be construed as constituting an additional warranty. HP shall
not be liable for technical or editorial errors or omissions contained herein.
Acknowledgments
Intel®, Itanium®, Pentium®, Intel Inside®, and the Intel Inside logo are trademarks or registered trademarks of Intel Corporation or its subsidiaries
in the United States and other countries.
Microsoft®, Windows®, Windows® XP, and Windows NT® are U.S. registered trademarks of Microsoft Corporation.
Java and Oracle are registered trademarks of Oracle and/or its affiliates.
http://www.hp.com/go/storagewarranty
Documentation
For the latest version of this document, go to http://www.hp.com/go/3par/, navigate to your product page, click Support for your product, and
then click Manuals.
Contents
1 Introduction...............................................................................................5
Audience.................................................................................................................................5
Related Documentation..............................................................................................................5
Open Source Software..............................................................................................................5
2 HP 3PAR InForm OS Release Notes...............................................................6
Online Upgrade Considerations.................................................................................................6
Supported Platforms..................................................................................................................6
End of Support Features............................................................................................................6
Components............................................................................................................................6
What’s New in the InForm OS...................................................................................................6
Multi-tenancy for Consolidation on Scalable, Resilient Infrastructure............................................6
HP P10000 3PAR Storage System Support..........................................................................6
Additional Multi-tenancy Enhancements.............................................................................7
Autonomic Management for Agility and Ease of Use................................................................7
HP 3PAR Peer Motion Software.........................................................................................7
Autonomic System Rebalance...........................................................................................8
Additional Enhancements to Dynamic Optimization.............................................................8
Configurable Priority for Snapshot Promote.........................................................................8
Additional Enhancements to Virtual Copy...........................................................................9
Expanded User Roles.......................................................................................................9
Additional Ease of Use Enhancements................................................................................9
Thin Technologies for Efficiency and Capacity Reduction...........................................................9
Fine-grained 16KB Space Reclaim.....................................................................................9
Remote Copy Bandwidth Optimization...............................................................................9
Additional Remote Copy Enhancements.............................................................................9
Subnet restrictions removed.................................................................................................10
Modifications to the InForm OS................................................................................................10
Limitations of the InForm OS.....................................................................................................15
Fibre Channel/SCSI Interface..............................................................................................15
Deleting Large Snapshot Volumes.........................................................................................15
LDAP authentication...........................................................................................................16
Remote Copy Link Speeds...................................................................................................16
Known Issues with the InForm OS..............................................................................................16
32-Bit Libraries for 64-Bit Red Hat Enterprise Linux 6.1 (Bug 63679)..........................................16
4Gb/s 2-Port adapter Issue.................................................................................................17
Downgrade Issues..............................................................................................................17
Issues Preventing Downgrades from InForm OS release 3.1.1 to InForm OS release 2.3.1.........17
Issues Preventing Downgrades from InForm OS release 3.1.1 to InForm OS release 2.2.4.........17
Issues Causing a Change to the InForm Configuration........................................................18
Running Recovery Manager and Remote Copy (from outside of Recovery Manager) at the Same
Time is not Supported.........................................................................................................18
Windows 2003 Pop-Up......................................................................................................19
Cage Loop Down Alert is not Auto-Recovered (Bug 40274)......................................................19
System Manager Could Fail During Rolling Upgrade if Remote Copy Groups Running (Bug
40867).............................................................................................................................19
No Explicit Way to Reset the Default Value for Tunelinks (Bug 43579).......................................19
Remote Copy Volumes go Stale Due to Lack of Adm Space (Bug 44963)..................................19
controlport Configuration Gives Warning after Being Taken Offline (Bug 44989).......................20
showtask for updatevv Displays Incorrect Completion Time (Bug 46328)....................................20
Tasks Become Active after a Failed Upgrade (Bug 46418).......................................................20
Remote Copy: Long Waits for Remote Copy Commands to Return Data.....................................20
Contents 3
IMC Crashes during Rolling Upgrade (Bug 46452)................................................................20
3 HP 3PAR InForm OS CLI Release Notes.......................................................21
Installation Notes for the InForm OS CLI.....................................................................................21
What’s New in the InForm OS CLI............................................................................................21
Removed and Deprecated Commands and Options....................................................................23
Modifications to the InForm CLI.................................................................................................25
Limitations of the InForm OS CLI...............................................................................................26
stat/hist commands could exit SSH sessions on large configuration systems...............................26
tunepd may not be able to run in a SSH session.....................................................................27
statvlun exits if there is insufficient memory to run an SSH session.............................................27
Maximum number of virtual copies per volume decreased.......................................................27
Known Issues with the InForm OS CLI........................................................................................27
4 Contents
1 Introduction
This document provides the release notes for the HP 3PAR InForm OS (Operating System) and
Command Line Interface (CLI).
You should read this document prior to installing the 3.1.1 version of this product.
NOTE: These release notes describe the features and fixes included in InForm OS 3.1.1. Refer
to the appropriate release notes for all prior releases.
Audience
The information in these release notes is intended for use by HP 3PAR customers, partners, and
HP 3PAR field representatives.
Related Documentation
The following documents also provide information related to HP 3PAR Storage Systems and the
InForm Operating System:
Using the InForm Command Line Interface (CLI) to configure HP 3PAR InForm OS 3.1.1 CLI Administrator’s Manual and
and administer HP 3PAR Storage Systems HP 3PAR InForm OS 3.1.1 Command Line Interface
Reference
Using the InForm Management Console graphical user HP 3PAR InForm Management Console Help Release 4.2
interface to configure and administer HP 3PAR Storage
Systems
HP 3PAR Storage System concepts and terminology HP 3PAR InForm OS 3.1.1 Concepts Guide
Using Remote Copy HP 3PAR Remote Copy 3.1.1 Software User's Guide
HP 3PAR Storage System hardware configurations, HP 3PAR E-Class/F-Class Storage System and Third-Party
component numbering and layout, and system cabling Rack Physical Planning Manual
HP 3PAR S-Class/T-Class Storage System Physical Planning
Manual
Upgrading HP P10000 3PAR Storage System hardware HP P10000 3PAR Storage System Upgrade Guide
components
Audience 5
2 HP 3PAR InForm OS Release Notes
Online Upgrade Considerations
This feature allows the upgrade of the InForm OS concurrently with I/O activity on the attached
hosts, provided certain conditions are met. For further information and pre-planning of Rolling
Upgrades, refer to the latest version of the HP 3PAR InForm OS Upgrade Pre-Planning Guide. To
obtain a copy of this documentation, go to http://www.hp.com/go/3par/, navigate to your
product page, click Support for your product, and then click Manuals.
Supported Platforms
The HP 3PAR InForm OS 3.1.1 Configuration Matrix contains information about supported hardware
and software platforms. To obtain a copy of this documentation, go to the Single Point of
Connectivity Knowledge for HP Storage Products (SPOCK) Web site at http://spock.corp.hp.com/
index.aspx.
Components
Component Version
OS 3.1.1.226
CLI 3.1.1.226
Resources added to system Data movement layer Data movement unit Sub/Whole volume
Autonomic system rebalance is supported both in InForm CLI (with the new tunesys command)
and InForm Management Console.
25066 The maximum size of a single The maximum size of a single Dynamic Optimization (tune) request has
Dynamic Optimization been limited to approximately 160TB. Tunes larger than 160TB are
request is approximately rejected. This can affect tunevv -cnt operations with a large number
160TB. of virtual volumes.
36594 Added new alert for If an RCFC port configures too many WWNs, an alert will be issued.
improperly-configured Remote
Copy Fibre Channel (RCFC)
ports
43412 Dynamic Optimization can be Dynamic optimization can be used to tune thin provisioned volumes back
used to tune a thin to their original CPG. This can improve volume layout without creating a
provisioned volume back to new CPG.
the same CPG.
44188 Port VLUNs are not exported A VLUN that is exported to a port will no longer be visible to hosts that
to domain hosts. are part of a domain, as that would violate domain security. They will
continue to be visible to hosts that are not part of a domain.
44597 Physical copy improvement Because there are two copies per node instead of cluster wide, vvcopy
on large clusters gives better throughput on large cluster.
45032 InForm OS kernel panics due If an iSCSI port on a storage system received a non-login-related iSCSI
to accessing an uninitialized PDU from an initiator before the login command sequence was completed,
pointer. the InForm OS kernel could panic due to accessing an uninitialized pointer.
This has been fixed by processing the PDU correctly and sending a login
reject per the iSCSI protocol.
45269 Pre-configured SNMP This condition has been fixed in InForm OS 3.1.1. There are no
community string exists on a pre-configured community strings on newly installed system.
newly installed system.
45348 The createhost command In previous releases, the createhost command for an iSCSI host did
for an iSCSI host does not not allow placing the host into a domain set; a separate movetodomain
allow placing the host into a step was required. It is now possible to provide a domain set for the
domain set. -domain option when creating an iSCSI host.
45951, SCSI-2 reservations remain on If an initiator logs out, gets disconnected or goes down while holding a
50858 the LUN after logging out or SCSI-2 reservation on a LUN on the InServ, the reservation used to remain
being disconnected. on the LUN and it had to be manually cleared for other initiators to access
the LUN. This has been corrected by the storage system clearing the SCSI-2
reservation on a LUN when the holder of the reservation logs out or is
disconnected from the array. The storage system also sets a
Power_on_Reset Unit Attention on the LUN to indicate to the initiators
accessing the LUN that the reservation has been cleared.
46473 Retransmit queue issues When a local host write is terminated with error and the remote copy
request has been scheduled for retransmission, it is possible that the remote
copy request is not removed from the retransmit queue before it is canceled.
This can result in retransmit queue having stale entries which could result
in controller nodes going offline at a much later stage. The most common
situation for local host write to be terminated with error is when a different
controller node goes offline. This has been fixed in InForm OS 3.1.1.
46695 System manager panic occurs In previous releases, when the storage system was in certain configurations,
on showvvpd. the CLI command showvvpd could cause a system manager panic. This
has been fixed.
46815 System manager panic occurs In previous releases, if the removehost -rvl command was used to
on removehost -rvl. remove a WWN from a host when the WWN was visible on multiple
ports and a volume was exported to the host, the next reference to the list
of VLUNs associated with that host (for example, a showvlun command)
would cause a system manager panic. This has been fixed.
47073 Unable to communicate with Incorrect iptables rules were being added due to a timing issue during
em_filter startup. This has been fixed.
47468 Failure during rapid Under certain circumstances, removing a VLUN and then immediately
re-exporting of VLUNs. creating a new VLUN to the same host and LUN number would fail with
the error message lumap should be lm_type_loop only. This has been
fixed.
48708 Virtual volumes incorrectly In previous releases, some virtual volume types that should not be exported
exported (for example, the target of a physical copy) could be exported if they were
the member of a virtual volume set. The checking around exporting a virutal
volume set has been extended in order to prevent this.
49407 LUN 0 is incorrectly selected In previous releases, the auto LUN-choosing functionality would not correctly
as a LUN for an HP/UX host. skip LUN 0 for HP/UX hosts, meaning that LUNs had to be manually
chosen for this platform. LUN 0 is now correctly skipped when choosing
an automatic LUN for an HP/UX host.
49487 SNMP configuration can In previous releases, a defect in the SNMP agent might have resulted in
sometimes be lost after a SNMP-related configuration data to be lost after a rolling upgrade. This
rolling upgrade. has been fixed.
50033 Removing a host does not When removing a host that had a virtual volume set exported to it the error
display correct error message would display UNKNOWN instead of the virtual volume set name.
message. This has been corrected so that the name of the virtual volume set is now
displayed.
50545 System manager panic upon Attempting to remove the snapshot CPG of a volume which had snapshots
removing volume CPG could cause a system manager panic. This has been fixed.
50858 SCSI 2 reservations are now When a host World Wide Name (WWN) disappears from a storage
released upon initiator system port, any SCSI 2 reservations held by that WWN will be released.
disappearance.
51186 Virtual volume exports Extending a virtual volume set that was exported as a matched set export
override existing exports. to a host could result in existing exports being overridden. The checking
when extending an exported virtual volume sets has been improved to
prevent this from happening and instead an appropriate error is issued.
51294 Remote copy miscompare Fixed a remote copy miscompare scenario when the DR commands are
scenario used incorrectly.
51523 R5 IDs incorrectly checked Corrected signed/unsigned 32–bit overflow issue so that R5 logical disks
are now correctly checked regardless of size.
51846 A Node down SNMP trap is Fixed an issue in which a Node down SNMP trap is not generated if the
not generated. node with the active Ethernet connection was forcefully removed.
51995 Double node failure condition If a compactcpg operation results in truncating logical disks and the
after a compactcpg current logical disk backup node fails during this operation, then the node
operation taking over as the logical disk backup (if there is one) will also fault,
resulting in a double node failure. This condition has been fixed by making
the kernel logical disk truncate code handle the case where the backup
node is newly assigned and not fully initialized by the system manager
node down processing.
52915 Node initialization process is On large systems, the system manager node initialization process to deliver
too lengthy. the physical disk, logical disk, and virtual volume information into the
kernel was taking too long. This was due to the creation of Linux device
entries for each of these devices. Because these are used only for internal
system access, pushing this task to a background thread fixes this issue.
53046 Node down issue with R6 There was a potential node down issue in the R6 CRC regenerate and
chunklet relocation. re-write processing when this action occurs as part of a R6 chunklet
relocation. This issue has been resolved.
53579 Link Power Down event affects The Link Power Down event needs to be cleared to avoid node being put
node rejoining. in degraded state after node rejoins.
53613 CPG Growth Failure Alert When a CPG fails to grow, an alert is issued. Previously, there was an
Clearing issue in which if the CPG could grow at a later time, the alert was not
cleared correctly. This issue has been resolved.
53676 Maximum number of exported It was possible to exceed the maximum number of exported LUNs allowed
LUNs exceeded (16383) by extending a previously-exported VV set. Checks have been
added to ensure that this is not allowed.
53686 Deleting large snapshots When deleting a very large snapshot, InForm OS may use up a large
causes memory allocation or amount of system memory which can cause memory allocation failures in
performance issues. InForm OS or cause performance issues. InForm OS is now enhanced to
enforce a limit on memory usage when deleting snapshots.
54303 Emulex-based 8Gb HBAs fill The Emulex-based 8Gb HBAs use a different fill word (ARB(FF) for Fibre
word compatibility Channel connections. The Fibre Channel switch port connecting the Emulex
8Gb HBA port must have the same fill word configured for compatibility.
Contact the switch vendor for help to configure the fill word. On Brocade
switches, the command to configure the fill word is portcfgfillword.
54726 I/O write credit pages in Fixed IO write credit pages merge code path, so that when LDIO flusher
inconsistent state flushes dirty cache pages to back end, it leaves the dirty mask in the log
entry for those credit pages in a consistent state.
55017 Cancelling inter-process A rare race condition existed in which cancelling inter-process messages
messages to avoid old message references were not being cleared. This has been
resolved.
55213 Improved SCSI-3 reservation The performance of SCSI-3 reservation handling has been improved,
handling performance preventing issues in which the storage system would be unable to keep
up with host requests.
55217 Invalid messages processed In rare cases, InForm OS could process invalid intra-cluster messages due
to a hardware timing issue on T and F-series storage systems. InFormOS
has been enhanced to detect this situation and re-read the data to ensure
only messages with correct contents are processed.
55218 Empty virtual volume set The error message when an empty virutal volume set was exported was
export error message not clear. Because exporting a virtual volume set isn't a logical action this
action is no longer allowed.
56082 LUN auto-choose Auto-choosing a LUN for a host that saw active exports using a host set
would sometimes incorrectly fail. The checks for host set exports have been
improved to resolve this issue.
56195 Defect in driver/firmware When the firmware in the 2-port 4GB Fibre Channel HBA in the E and
results in port failure. F-series storage systems received incorrectly formatted Fibre Channel
frames, the frame buffers were not released into the free buffer pool. When
a large number of these occurred, the port could run out of buffer credits
and fail. This was fixed by releasing these invalid frames correctly. This
problem was seen with a Cisco UCS server using the Palo interface card
connected to the storage system 4G port using a Brocade fabric when
SAN booting Windows 2008 R2. The Palo card driver/firmware had a
defect causing it to send incorrectly formatted Fibre Channel frames.
56573 RAID type displayed If zerold was mapped to a volume offset 0, then the Inform Management
incorrectly Console would display the incorrect RAID type. This has been fixed.
56754 InForm OS kernel panic due If an iSCSI port on a storage system received a non-login-related iSCSI
to accessing an un-initialized PDU from an initiator before the login command sequence was completed,
pointer. the InForm OS kernel could panic due to accessing an uninitialized pointer.
This has been fixed by processing the PDU correctly and sending a login
reject per the iSCSI protocol.
57546 System manager is unable to System manager does not start after a system or system manager restart
start up after panic, under the following scenario: running setvv -snp_cpg followed by
repeatedly panics after the removing the CPG which was being used as a snap CPG before running
initial panic. the command. Removal of the CPG could happen a few days later. Storage
systems with ESX clusters attached to them would have virtual host failures
because SCSI 2 reservations were not handled while system manager was
down. This has been fixed.
58215 The Link Power Down event The Link Power Down event is now properly cleared.
needs to be cleared to avoid
a node being placed in a
degraded state after the node
rejoins.
58269 InForm OS kernel incorrectly The InForm OS kernel could incorrectly identify SCSI-3 persistent reservation
identifies SCSI-3 events events from different nodes as duplicate events and omit some steps in
event validation and processing. As a consequence, each node's list of
registered SCSI initiators can be corrupted with un-initialized values. The
resulting inconsistency could trigger an assertion on all nodes when an
initiator registration is removed, causing cluster failure. A fix has been
implemented that extends the check for duplicate events to include the
node identifier and to ensure that initiator list fields are populated correctly
for duplicate events.
58586 Physical Disk (PD) scrub The initial design of the PD scrubber was to scrub the entire system every
performance issues seven days. Systems have gotten much larger and delays were added to
the prevent the scrubber from impacting system performance. These two
changes have resulted in many systems being scrubbed at a much slower
rate (every 60 days or more). This opens a larger window for double disk
failures. The previous design kept one chunklet per loop active and would
issue one ioctl to kernel every scan step. The new design issues one
ioctl for one chunklet of all the PDs on the node. Inside the kernel scan
steps are scheduled based on a self-implemented timer, so that the
scrubbing rate can be controlled, VERIFY operations can be spread out
evenly, and multiple PD scans can be processed at the same time.
58703 InForm OS kernel memory The InForm OS kernel could leak memory when processing some SCSI
leak commands, as memory buffers dynamically allocated for command
processing are not freed on command completion. This leak could affect
certain commands received on LUNs to which no virtual volume is exported.
If a sufficiently large number of these commands were received, available
memory was exhausted and the node would panic. A fix has been
implemented that ensures that all memory allocated for processing of these
commands is released on command completion.
58852 Recovery requires user During a power failure or multi-node panic recovery, there were some
intervention cases in which the system required user intervention in the form of issuing
a setsysmgr force_iderecovery command. These cases were a)
when all but one node had saved data without proper handshake with
the missing node and b) when the system was in the process of creating
a volume (base or snapshot) when the save occurred. A fix has been
implemented that eliminates the need for user intervention; the system will
now recover data automatically.
58886 Virtual volume with metadata For a virtual volume, if the metadata view of size is smaller than actual
issues does not fail properly size, the metadata checker will now fail the volume. Also while probing
for holes in snapspace due to reclaim, if metadata view cannot account
for all the holes, the metadata checker will now fail the volume.
58939 Recovery requires user During a power failure or multi-node panic recovery, if the storage system
intervention encountered problems paging in volume metadata, recovery could cause
all the nodes to panic. When the storage system restarted, it required user
intervention in the form of a setsysmgr force_iderecovery command
on order to proceed with recovery. A fix has been implemented in which,
if the metadata page has failed due to disk unavailability, the system will
now automatically proceeds with recovery, and does not require user
intervention.
59114 Disk drives unresponsive and If the firmware on the 4-port 4GB Fibre Channel HBA experienced a fatal
marked failed by storage error while I/O was ongoing to disk drives via the HBA, the firmware
system would not respond for several seconds while the transmit and receive lasers
were still active. This could lead to the disk drives becoming unresponsive
for up to a few minutes, resulting in the drives being marked failed by the
storage system. If disk cages were connected to multiple HBA ports, this
could result in multiple drives being 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 receives a fatal error, which
allows the drives to detect the link failure immediately and recover without
becoming unresponsive.
59274 System manager restart after A system manager restart after an online upgrade had a small chance of
online upgrade causes panics causing repeated system manager panics. This has been fixed.
59549 Improved SCSI-2 reservation The performance of SCSI-2 reservation handling has been improved to
handling performance allow more simultaneous incoming reservations.
59672 Old events appearing in After a master node restart or storage system restart, events from a long
showeventlog time period prior could reappear in showeventlog. This has been fixed.
60162 Remote Copy links down after After setting the port speed for Remote Copy links to 100Mbps from
changing port speed 1000Mbps the links failed due to negotiation issues. The links are
recovered by setting auto negotiation for those links. The ability to set the
speed to rates less than 1000Mbps will be removed in a future release.
60638 System manager copy When system manager attempted to create a copy of itself to handle a
creation failure results in showspace request and that copy creation failed, the event that was
repeated event filter panics. generated was not parsed correctly and resulted in repeated panics by
the event filter. Because this caused events to be queued in memory
indefinitely, an out of memory situation could result, especially in
configurations using remote copy. This has been fixed.
61125 Remote Port Indicator not Each 4GB Fibre Channel port on the storage system has 511 Remote Port
released properly Indicator (RPI) resources. One resource is used when an initiator logs in.
When the initiator logs out or gets disconnected, the RPI was not being
released. If the initiator logged back in with the same Fibre Channel D_ID,
the same RPI was being reused, but if it logged in with a different D_ID,
the original RPI was lost. After 511 such logins, the resource was exhausted
and all further logins were blocked. This was fixed by releasing the RPI
when an initiator logs out or is disconnected.
61212 Miscompare between primary A miscompare was occurring between primary and secondary volumes
and secondary volumes in the event of a node down. The miscompare happened when an I/O
was received that spanned two logical disks with different owners. The
I/O was split and passed to the logical disk owner nodes; however, if one
of these nodes failed , the backup node took over and completed the I/O.
This was where the remote copy miscompare was happening, although
the backup node reliably completed the I/O to the base volume on the
primary, it did not issue the remote copy request. This has been fixed.
61315 SCSI-2 reservations not being The handling of the Logical Unit Reset task management function has been
released corrected to ensure that SCSI-2 reservations on the addressed LUN are
released.
61808 Drive cage fails and becomes There was an error in the serial port in which characters were looped back
inaccessible after reset to the serial port which sometimes could cause a timeout, resulting in a
cage resetting and becoming inaccessible. A fix has been implemented
that prevents any characters from returning to the serial port and additional
checks have been put in place to prevent cages from becoming
inaccessible.
61903 Forked system manager System manager forked new processes to handle showspace and showvv
processes do not exit. commands. At times these processes would not exit after processing the
command. These processes consumed system resources and could lead
to problems such as filling /var/tmp_log and consuming all system memory.
A fix has been implemented that prevents the deadlock that caused the
issue.
62864 Alert issued when controller When a controller node unexpectedly goes offline, an alert with Message
node goes offline and rejoins Code 0x01a0007 or 0x1a0007 is generated with text indicating that a
the cluster BIOS log has been stored (saved) for that node. This alert does not get
auto-resolved when the node rejoins the cluster, but it can safely be deleted.
This alert will not be generated in a future release.
Downgrade Issues
The following are downgrade issues for InForm OS release 3.1.1.
Issues Preventing Downgrades from InForm OS release 3.1.1 to InForm OS release 2.3.1
These issues will prevent an HP 3PAR Storage System from downgrading from InForm OS release
3.1.1 to InForm OS release 2.3.1.
• Read/Write snapshots with zero detect policy - InForm OS 2.3.1 I/O code paths do not support
Read/Write snapshots on which zero detect or write-same-zero driven space optimization
has been performed. Before downgrading from InForm OS 3.1.1 to InForm OS 2.31, you
must remove all Read/Write snapshots that have zero detect policy enabled or that have
space optimization exception entries created by zero detect or write-same-zero.
• Users with Extended Roles - InForm OS releases prior to 3.1.1 do not support the users with
extended roles features. Before downgrading from InForm OS 3.1.1, you must remove all
users with extended roles.
• Read/Write Snapshots admitted to Remote Copy Groups - If read/write snapshots are admitted
to Remote Copy groups in InForm OS 3.1.1, downgrades to InForm OS 2.3.1 are not allowed.
Before downgrading from InForm OS 3.1.1 to InForm OS 2.3.1, ensure that all read/write
snapshots are removed from Remote Copy groups.
Issues Preventing Downgrades from InForm OS release 3.1.1 to InForm OS release 2.2.4
These issues will prevent an HP 3PAR Storage System from downgrading from InForm OS release
3.1.1 to InForm OS release 2.2.4.
• Dual SD Space - As of InForm OS 2.3.1, Thin Provisioned virtual volumes can have two Snap
Data (SD) volumes. This is not a supported feature in previous InForm releases. Prior to a
downgrade, dual SD TPVVs must be converted to single SD TPVVs by removing all of the
snapshots for those TPVVs.
• Raid Multi-Parity (MP) - InForm OS releases prior to 2.3.1 do not support Raid MP; therefore,
any Raid MP Logical Disks (LDs) must be tuned to a supported RAID type or removed prior to
downgrade.
• SD Reclaim Space - SD Reclamation is not supported by previous InForm OS releases. Prior
to a downgrade, SD volumes that has its space reclaimed must be removed. If that SD is used
by snapshots only, then all of its snapshots must be deleted so that SD can be freed. If that
SD space is used by the TPVV itself, then that TPVV would need to be copied over to a new
single SD TPVV. After the copy operation completed, the original TPVV can be removed.
• 256 Read/Write Snapshots - As of InForm OS 2.3.1, the number of read/write snapshots was
increased to 256 per virtual volume. Previous releases supported a smaller maximum number
of read/write snapshots per virtual volume. A downgrade will be prevented if any virtual
volume has more read/write snapshots than the maximum supported on the previous release.
When enough snapshots have been removed to meet this criterion the downgrade is allowed.
• Multiple Read/Write Snapshots per Read-Only Snapshot - As of InForm OS 2.3.1, the creation
of multiple read/write snapshots per read-only snapshot is permitted. Previous releases
Running Recovery Manager and Remote Copy (from outside of Recovery Manager)
at the Same Time is not Supported
To synchronize Recovery Manager snapshot data from the production site to secondary site,
Recovery Manager makes direct programmatic calls to the 3PAR Remote Copy product to perform
the background synchronization. When Recovery Manager is doing Application Consistent Remote
Syncs using Remote Copy in Asynchronous Periodic Mode, it is not required to setup the periodic
resync time for the Remote Copy groups. Setting up periodic resync time for Remote Copy groups
when recovery manager is active could disrupt the operation of Recovery Manager.
• Workaround:
Remote Copy can support the mixture of Recovery Manager snapshots and Remote Copy snapshots
(outside of Recovery Manager) if the following procedure is followed:
Inside the Recovery Manager Windows scheduler script:
1. Set the Remote Copy period to 0.
2. Verify Remote Copy SyncStatus as "Synced".
3. Execute Recovery Manager snapshot creation for Remote Copy.
4. Wait for SyncStatus to be completed and Recovery Manager to take snapshot of secondary
volumes.
5. Recovery Manger will monitor the SyncStatus to be "Synced" every 120 seconds as default
then issue snapshot creation of secondary volumes.
WARNING! Steps 2 and 4 are critical to ensure Recover Manager-created snapshots and
Remote Copy-created snapshots do not interfere with each other.
System Manager Could Fail During Rolling Upgrade if Remote Copy Groups Running
(Bug 40867)
When creating snapshots during the shutdown process there is a known race condition that may
result in a system manager crash. Therefore, the user is no longer permitted to manually take
snapshots during this process. However, this is not the case with snapshots that are taken when a
synchronous remote copy group is stopped.
During a rolling upgrade synchronous groups are normally not stopped. However, if there is a
target failure during the process any associated groups will automatically be stopped, and if the
group is synchronous a snapshot for each volume in the group will be taken. This leaves the system
open to hitting the race condition described above.
• Workaround - Stop all remote copy groups before starting a rolling upgrade.
No Explicit Way to Reset the Default Value for Tunelinks (Bug 43579)
The setrcopytargettunelinks command can be used to adjust the performance value for
the target's links using the <bandwidth> and <latency> specifiers. After using this command,
there is no explicit way for the user to reset the tuning parameters back to the default value.
• Workaround - Reset the parameter to the default value by reissuing tunelinks with
bandwidth=128000 (or higher) and latency=100.
Remote Copy Volumes go Stale Due to Lack of Adm Space (Bug 44963)
The default grow value of 64 GB might be insufficient for volumes which are in Remote Copy
groups. The reason is Remote Copy relies on taking snapshots during the initial sync phase and
controlport Configuration Gives Warning after Being Taken Offline (Bug 44989)
Using the -f option with controlport config, the system responded with a confirmation
request, or refused to force the mode change.
• Workaround - If the -f flag was used and the system still prompts for confirmation, respond
to the prompt. If there are other messages, they need to be resolved (e.g. take port offline),
or possibly reissue the command.
Remote Copy: Long Waits for Remote Copy Commands to Return Data
When using a 2-node secondary system for a 2-to-1 Remote Copy configuration, commands can
take quite some time to return on a busy system. For this reason, the secondary Remote Copy
system cannot be used as a standalone system. Commands such as showrcopy and showvv
have been observed to take between 2 and 8 minutes to return data. Remote Copy operations are
not impacted, however during the time of a DR restore, where multiple resyncs to different targets
could be taking place, commands can take a significant amount of time to execute.
Node Subsystem
Batteries showbattery -listcols New options -listcols and –showcols to display data based
showbattery -showcols on columns specified by the user
<column>[,<column>...]
Nodes shownode -listcols New options -listcols, -showcols, -fan and –mcu to support
shownode -showcols HP P10000 3PAR Storage Systems
shownode -fan
shownode -mcu
Ports histport –peer New option -peer to display the port histogram/statistics for peer
statport -peer ports
showport -peer New option -peer to list all the ports that have connection modes
set to peer
controlport config peer Added support to configure the port connection mode as peer for
the peer motion feature
System showsys -d The –d option now displays chunklet size in MB as either 256 or
1024 depending on system type
Copy Services
Virtual Copy promotesv -pri New option -pri to improve the performance of promote
operations
Physical Copy creategroupvvcopy -pri New option -pri to improve performance of the group copy
operation
createvvcopy -pri New option -pri to improve the performance of copy operation
Physical Disks admitpd -f New option -f to skip the confirmation when the storage system
is attempting to admit disks from other storage systems
Host Management
Hosts showhost —lesb New option -lesb to show the CRC error for the specified host(s)
Performance Management
Statistics statlink -detail New option -detail to display detailed queue statistics
information
Remote Copy
admitrcopyvv –pat New option –pat to specify that a specified pattern is treated as
dismissrcopyvv –pat glob_style pattern and that all remote copy volumes matching the
specified pattern will be admitted, dismissed, or set to the Remote
setrcopygroup -pat Copy group
Groups removercopygroup -pat New option -f to indicate that the command is forced. Specifies
-f that specified patterns are treated as glob-style patterns and all
Remote Copy groups matching the specified pattern will be
removed.
startrcopygroup -pat New option -pat to indicate that specified patterns are treated
stoprcopygroup -pat as glob-style patterns and all Remote Copy groups matching the
specified pattern will be started, stopped, or synced.
syncrcopy -pat
Links checkrclink New command performs a latency and throughput test between
two connected storage systems.
Security
Service
Sparing movepdtospare -vacate New option -vacate to make moves permanent and remove
source tags after relocation
Tasks settask New command to set the priority on the specified task
Volume Management
CPGs showvvcpg New command to show virtual volume space distribution among
CPGs
Volumes admitvv New command admits one or more virtual volumes to enable their
migration
setvv -wwn New -wwn option to modify the WWN of the volume. This option
is not allowed for the admitted volume before it is imported, or
while import process is taking place.
growvv -f
checkvv -offline New option -offline to specify that a virtual volume is offline
before validating the virtual volume administration information
removevv -nowait New option -nowait to indicate not to block until the virtual
volume is removed. Removal of the virtual volume is performed in
the background. Only snapshots can be removed with this option.
An attempt to remove a base virtual volume results in an error.
Batteries setbattery The -r and -s options are not supported on HP P10000 3PAR Storage
Systems.
CPGs createcpg The –sa, –sd, and -aw options to the createcpg command have
been deprecated and will be removed in a future release.
removecpg The –sa, –sd options to the removecpg command have been
deprecated and will be removed in a future release.
setcpg The –sa, –sd, and -aw options to the setcpg command have been
deprecated and will be removed in a future release.
Ports controlport • The –m option to the controlport rst command has been
deprecated and will be removed in a future release. Use
controlport config to set target (host) or initiator (disk) modes.
• The ct and cl2 subcommands have been deprecated and will be
removed in a future release.
Network setntp The command is deprecated and will be removed in a future release.
Configuration Please use the setnet command instead.
removercopytarget New –cleargroups option provides the ability to remove all groups
that have no other targets or dismiss a target from groups with additional
targets.
Templates createtemplate The -cpg, -aw, -al, -szs, and -pct options have been deprecated
and will be removed in a future release.
settemplate The -cpg, -aw, -al, -szs, and -pct options have been deprecated
and will be removed in a future release.
Volumes growaldvv –szs and • The –szs and –sza options to the growaldvv command have been
–sza options deprecated and will be removed in a future release.
• The following commands were deprecated in the 2.2.3 release and
will be removed in a future release:
◦ createaldvv
◦ createtpvv
◦ growaldvv
◦ growtpvv
◦ tunealdvv
◦ tunetpvv
Please use the createvv, growvv, and tunevv commands.
showvv The –cpg, -tpvv, -host, and –rcopygroup options to the showvv
command have been deprecated and will be removed in a future release.
Please use the showvv –p option..
N/A showtarget command The command functionality is new. It will display remote volumes in
showtarget but not in showpd. It will also display the SCSI inquiry and
mode page information.
35572 VLUN removal (removevlun When a VLUN with SCSI reservations is removed, the following warning
command) is issued and requires confirmation in order to proceed: Warning: The
VLUN has SCSI reservations. Are you sure you want to remove it?.
43690 The CLI client stops if the The CLI client will stop if the connection to the CLI server is lost. The CLI
connection to the CLI server client issues the following message if the connection is lost: cli% Server
is lost. connection broken, exiting ... Connection to s017 closed.
46372 The showcage -sfp In previous releases,showcage -sfp showed the state of an unqualified
command shows the state of SFP in the cage as OK instead of Degraded. This has been fixed in InForm
an unqualified SFP in the OS 3.1.1.
cage as OK; the state should
be Degraded.
46741 showcage command An event log will be created when the showcage command is run.
47270 Discrepancy between System manager does not send raw used MB and as a result CLI processes
showsys -space and raw total MB causing a discrepancy between the showsys –space and
showvv –r showvv –r.
49019 showhost command A new option -lesb has been added to the showhost command to show
the CRC error for the specified host(s).
showhost -lesb [<host_name>... | <pattern>... |
<host_set>...] -lesb
Shows the CRC error counts for the host/port.
49449 The vvcopy command Because the events are generated for vvcopy update events, it appears
appears to halt every five to halt. This has been fixed using another event type.
seconds.
51517 starttask permission The permission to run starttask is now granted to SUPER, EDIT and
extended to EDIT user. SERVICE roles only. Previously, only SUPER and SERVICE could run
starttask.
51621 showpd, showld –p and The Speed(K) column header has been changed to RPM. When using the
showpdvv commands -showcol and -listcol, options, use RPM to refer to that column.
53468 A warning message is issued The growvv, growavv, growaldvv and growtpvv commands will issue
if a virtual volume size grows a warning if the virtual volume size increases from less than 2TB to greater
from less than 2TB to greater than 2TB. The warning can be overridden by using the -f option.
than 2TB.
53689 showvv -cpg enhancement In previous releases, the showvv -cpg command listed only the base
volumes belonging to a CPG. This has been modified in InForm OS 3.1.1
so that both the snapshots and the base volumes belonging to a CPG are
listed.
54862 User password requirements The backslash symbol (\) is now allowed in the user password when the
user is created or the password is changed with the setpassword.
55215 The showspace -hist The command showspace -hist -cpg <cpg1> now only returns the
-cpg xxx command returns specific CPG information from the storage system back to the client.
excessive information,
resulting in memory issues.
55785 showvv -cpg command In previous releases, the showvv -cpg command only listed the base
enhancement volumes belonging to a CPG. This has been modified so that both the
snapshots and the base volumes belonging to a CPG are listed.
55804 showdate and stats In some cases, if a valid time zone was provided using the InForm OS CLI
commands time stamps command settz and a failure occurred during the propagation of the
mismatch time zone across appropriate directories in the storage system, it was
possible that showdate and stats commands time stamps did not match.
The piece of code handling the time zone propagation did not report errors
back to upper layers (which always assumed the time zone change was
successful). The code has been updated to issue a notification if the time
zone propagation fails.
57065 Updated root jail directory The SSH interface to the InForm OS CLI client code did not properly
access for Secure Shell (SSH) construct a changed root jail for the user signing on. This could allow an
users unprivileged SSH CLI user to access files outside the changed root. The
unprivileged user could exploit this flaw by examining world readable
files throughout the node file system. A fix has been implemented that
prevent CLI SSH users from accessing files outside of the chroot jail.
59159 sasl binding and base When binding is set to sasl, the base of the search is now set to the root
search setting to match of the directory tree if a common container is not found in the directory
multiple OUs hierarchy for multiple OUs.
59578 Insufficient iSCSI host I/O Due to insufficient throttling of iSCSI host I/O, hosts were able to push a
throttling results in controller very large number of I/O requests to the storage system that could result
node going offline. in a controller node going offline because it took too long to process those
requests to send heartbeat messages in a timely manner. This has been
fixed by sending back a SCSI QFULL error condition to a newly received
command on an iSCSI port when the number of pending commands on
the port has reached a predetermined limit.
59609 Needs to use the remote CLI If the storage system has a large configuration (for example, many virtual
client to run tunepd if the volumes, logical disks, and disks), tunepd could run out of memory in an
system configuration is too SSH session. In this case, it will return the following error message and
big exit: There is not enough memory to run this command on the system.
Please use the CLI remote client to run it. If this is the case, run the command
from the remote InForm OS CLI client.
60331 showportlesb reset The reset subcommand description has been updated to: “Specifies that
subcommand description internal counters are saved as the reference point from which deltas are
update calculated for the next sampling cycle”.
60356 setvv -snp_cpg on virtual Using setvv -snp_cpg on virtual volumes relying on shared logical
volumes with shared logical disks was causing a panic of the system manager. The behavior of setvv
disks causes system manager has been corrected. All virtual volumes sharing logical disks must be
panic specified as arguments; otherwise, the command will fail.
62573 Windows client does not On the Windows platform, the InForm OS CLI client can now be run
need to logoff/logon after immediately after installation without a user log off/on.
InForm OS CLI installation.
63602 Potential node panics during To avoid potential node panics during a Remote Copy resync operation,
resync operation the startrcopygroup -nosync <grp> command will now return a
failure if the RC group has already started.
44872 Passing a virtual volume set Use of createsv –i with virtual volume sets is not currently supported.
to the createsv –i
command results in an error.
44937 Misleading command output When using the createvvcopy command with virtual volume sets, if
when passing empty virtual both sets are empty, the output of the command erroneously references
volume sets to the the creategroupvvcopy command.
createvvcopy command. Workaround: Pass virtual volume sets that are not empty to the
createvvcopy command.
45413 Integer arguments to CLI The CLI shell is a TCL-based shell, as such it parses integers with leading
commands with leading 0 0 (zero) as octal numbers, just like many other languages and shells (bash,
(zeros) are treated as octal perl, etc). Consequently, integer arguments passed to the CLI commands
numbers. are also treated as octal numbers if they have a leading 0.
Workaround: If you do not wish for integers to be treated as octal integers,
do not use leading zeros.
46372 Drive cage displays do not The showcage -sfp command shows the sfp state as OK, Degraded,
correctly report degraded Failed or Unknown. For an unqualified SFP, the state should be Degraded
SFPs. but is shown as OK.
Workaround: If showcage –sfp shows an SFP as OK, try using
showcage –sfp –d <cage_name>. If the Qualified field shows No,
then it is an unqualified SFP and its state should be deemed as Degraded.
51356 Extended groups supersede The matching of a user’s extended group (basic_edit-map, create-map,
static groups 3PAR_AO-map, 3PAR_RM-map) supersedes static groups (super-map,
service-map, edit-map, browse-map).
Workaround: If there are multiple group matches, the first match determines
the user’s role.
57362 createsched does not createsched does not support [ and ] globbing characters.
support [ and ] globbing Workaround: These characters cannot be used when specifying a pattern
characters. with the createsched command.
59715 sasl-mechanism set to The combination of sasl-mechanism set to GSSAPI and binding set to
GSSAPI and binding set to simple does not function properly.
simple not functional. Workaround: Set sasl-mechanism to GSSAPI and binding to sasl
instead.
59796 HP 3PAR Peer Motion If CTRL+C is pressed while HP 3PAR Peer Motion Manager Software is
Manager Software exits processing a batch file, processing stops and the Migration Manager
during batch file processing. Wizard exists.
60105 servicecage syntax is The servicecage command's specifier 0|1 is required for many
incorrect. subcommands including staretfc and endfc in order for the command
to function properly.
60980 In HP 3PAR Peer Motion Pressing R in order to return to the previous menu does not refresh the
Manager Software, pressing screen.
R does not refresh the screen. Workaround: To refresh the screen, press Enter.
63624 showport command When an invalid option is entered, the showport command displays the
displays incorrect error message Expecting <NSP or pattern> got: -a instead of Invalid option: -a
message. and the command usage.
64033 Failed Battery Backup Unit If a BBU is marked permanently failed and is replaced, the permanent
(BBU) does not clear its status failure does not get cleared (for any fault and the BBU's status indicates
and issue a pending alert Failed as shown by the showbattery command and the associated
when replaced. pending alert does not resolve. To work around this issue, run the
nemoetool -z [-n <node#>] command on a root shell ( not the CLI
shell).