IBM FlashSystem A9000R 12.3.2.b RN
IBM FlashSystem A9000R 12.3.2.b RN
IBM FlashSystem A9000R 12.3.2.b RN
Version 12.3.2.b
Release Notes
IBM
Third Edition (January 2021)
This edition applies to IBM FlashSystem® A9000R version 12.3.2.b. Newer document editions may be issued for the
same product version in order to add missing information, update information, or amend typographical errors. The
edition is reset to 'First Edition' for every new product version.
© Copyright International Business Machines Corporation 2016, 2021.
US Government Users Restricted Rights – Use, duplication or disclosure restricted by GSA ADP Schedule Contract with
IBM Corp.
Contents
Overview...............................................................................................................1
Change log............................................................................................................ 9
Version 12.3.2.b (May 2020)....................................................................................................................... 9
Version 12.3.2.a-3 (May 2020)................................................................................................................. 11
Version 12.3.2.a-2 (December 2019)....................................................................................................... 11
Version 12.3.2.a-1 (September 2019)......................................................................................................12
Version 12.3.2.a (August 2019)................................................................................................................ 12
Version 12.3.2 (May 2019)........................................................................................................................ 15
Version 12.3.1 (January 2019)..................................................................................................................15
Version 12.3.0.a (November 2018)...........................................................................................................17
Version 12.3.0 (November 2018)..............................................................................................................17
Limitations.......................................................................................................... 25
Known issues...................................................................................................... 27
Notices................................................................................................................41
Trademarks................................................................................................................................................ 42
iii
iv
Overview
IBM FlashSystem A9000R is a high-end, all-flash storage system that delivers ultra-fast storage together
with mission-critical features, including data reduction by compression and inline deduplication, smart
scaling, distributed data, automatic load balancing, and a multitude of advanced enterprise-class features
and capabilities.
As a pre-integrated rack offering, IBM FlashSystem A9000R comprises grid controllers and flash
enclosures that are interconnected by integrated InfiniBand switches, forming a scale-out grid fabric that
delivers exceptional IOPS performance.
In its core architecture, IBM FlashSystem A9000R utilizes IBM FlashCore® technology together with IBM®
MicroLatency® modules, providing high density, low latency, and storage reliability. In addition, IBM
FlashSystem Enhanced Endurance Technology reduces flash disk wearout and ensures long-term
durability of the flash storage components, even under heavy workloads.
IBM FlashSystem A9000R is ideal for large enterprises that rely on fast, redundant, and high-capacity
data storage, offering high service levels for dynamic workloads and easy hyper-scaling, while supporting
multi-tenant environments, flexible consumption models, and robust cloud automation and integration
capabilities. IBM FlashSystem A9000R offers data protection through advanced remote mirroring,
including synchronous mirroring, asynchronous mirroring, and HyperSwap® high availability through
active-active pairing. It also helps protect data with encryption and a range of data security features.
For more detailed information about IBM FlashSystem A9000R, refer to its user documentation and
online information.
Note: The system e-license may automatically prompt in the management GUI upon the first post-
upgrade operation, if the e-license was updated since it was last accepted. In such a case, the new e-
license must be accepted before any GUI operation can be performed.
Note: Versions of IBM Hyper-Scale Manager earlier than 5.5.2 are not fully supported and must be
upgraded to 5.5.2 or later.
IBM FlashSystem A9000R can also be managed from its command-line interface (CLI) or through
representational state transfer (REST) application programming interfaces (APIs).
IBM FlashSystem A9000R customers are entitled to free use of IBM Hyper-Scale Manager, which can be
obtained at any time from the IBM Fix Central website (www.ibm.com/support/fixcentral).
Important: Quorum Witness cannot reside on the same set of FlashSystem A9000 and/or A9000R
systems of which it serves as part of a HyperSwap configuration.
Note: Newer versions of the Quorum Witness software may be released independently to better support
or enhance the HyperSwap operation. For more information, see the latest Quorum Witness release
notes and user guide.
IBM FlashSystem A9000R customers are entitled to free use of the IBM cloud storage solutions, which
can be obtained at any time from the IBM Fix Central website (www.ibm.com/support/fixcentral).
For more information about the available cloud storage solutions for IBM FlashSystem A9000R, see the
'Platform and application integration' section in the IBM FlashSystem A9000R space on IBM Knowledge
Center (www.ibm.com/support/knowledgecenter/STJKN5).
Note: New functional features of these versions are only briefly summarized. For a detailed summary of
the new functional features of a specific version, refer to the 'What's new' section of its release notes.
Ticket ID Description
SYS-307853 Fixed: In extremely rare cases, when utilizing large amounts of mirroring
pairs,snapshots, random write patterns, and heavy loads, internal resource
depletion may cause multiple system component failures which might lead to
loss of access.
Severity: High Impact
Affected versions: 12.0.0 or later
SYS-308559 Fixed: Common security issue. For more information about these security
issues, see CVE-2017-15906 on the Common Vulnerabilities and Exposures
(CVE) information website (cve.mitre.org).
Severity: High Impact
Affected versions: 12.2.1 or later
SYS-310434 Fixed: In rare cases, an automatic recovery of the internal interconnect network
will fail, causing possible loss of access.
Severity: High Impact
Affected versions: 12.0.0 or later
SYS-315321 Fixed: On extremely rare occasions, the internal background process in charge
of the data consistency verification might not complete it's scrubbing cycle.
Severity: High Impact
Affected versions: 12.1.0 or later
SYS-315429 Fixed: In very rare cases, a medium error from the flash enclosure may not be
properly handled by the FlashSystem A9000R code, resulting in loss of access.
Severity: High Impact
Affected versions: 12.0.0 or later
SYS-315539 Fixed: Using mirrored snapshot overwrite may cause internal system software
component failure, causing loss of access and, in rare cases, may also cause
data loss.
Severity: High Impact
Affected versions: 12.0.0 or later
SYS-315556 Fixed: In rare cases, error recovery of a flash enclosure during a loss of access
event may result in minor data loss.
Severity: High Impact
Affected versions: 12.0.0 or later
SYS-315671 Fixed: Common security issue. For more information about these security
issues, see the following on the Common Vulnerabilities and Exposures (CVE)
information website (cve.mitre.org): CVE-2018-0495, CVE-2018-0734,
CVE-2018-1122, CVE-2018-12404, CVE-2018-14647, CVE-2018-1547,
CVE-2018-15686, CVE-2018-16062, CVE-2018-16402, CVE-2018-16403,
CVE-2018-16842, CVE-2018-16866, CVE-2018-16888, CVE-2018-18310,
CVE-2018-18520, CVE-2018-18521, CVE-2018-3058, CVE-2018-3063,
CVE-2018-3066, CVE-2018-3081, CVE-2018-3282, CVE-2018-5407,
CVE-2019-11729, CVE-2019-11745, CVE-2019-1559, CVE-2019-15807,
CVE-2019-15807, CVE-2019-16089, CVE-2019-2503, CVE-2019-2529,
CVE-2019-2614, CVE-2019-2627, CVE-2019-3855, CVE-2019-3856,
CVE-2019-3857, CVE-2019-3858, CVE-2019-3861, CVE-2019-3862,
CVE-2019-3863, CVE-2019-5010, CVE-2019-6454, CVE-2019-7149,
CVE-2019-7150, CVE-2019-7664, CVE-2019-7665, CVE-2019-9740,
CVE-2019-9947, CVE-2019-9948
Severity: High Impact
Affected versions: 12.0.0 or later
SYS-315365 Fixed: In some cases, live vaulting, the automatic process of cache and
metadata saving to the vault devices, might not succeed during heavy system
loads. When this occurs, the
LIVE_VAULTING_PROCESS_CANCELLED_TOO_MANY_TIMES event is emitted.
Severity: Moderate
Affected versions: 12.1.0.e or later
SYS-315411 Fixed: In cases where the system fails to complete the live vaulting process, an
event is not emitted.
Severity: Moderate
Affected versions: 12.0.0 or later
SYS-315425 Fixed: Full Multi-site relation allowances are not able to be utilized. Depending
on the scenario, one of the following events may be emitted:
MAX_SYNC_MIRRORS_REACHED, MAX_ASYNC_MIRRORS_REACHED, or
MAX_MULTISITES_REACHED.
Severity: Moderate
Affected versions: 12.3.0 or later
SYS-314323 Fixed: When running the ldap_config_set command in order to set the
Storage Integration Administrator (SIA) role, the new SIA role does not take
effect, even though the command shows that it completed successfully.
Severity: Low
Affected versions: 12.3.2.a or later
Ticket ID Description
SYS-315562 Fixed: (Models 425 or U25 only) A MicroLatency module failure may result in I/O
latency between the flash enclosure and the rest of the system. As a result, the
host I/O cannot serve for a period of 3 - 4 minutes, with I/Os resuming
automatically after MicroLatency module failure is properly identified.
Severity: HIPER
Affected versions: 12.3.2.a or later
Ticket ID Description
SYS-315392 Fixed: The flash enclosure interface controllers may cause system outages due
to internal hardware failures.
Severity: High Impact
Affected versions: 12.0.0 or later
Change log 11
Version 12.3.2.a-1 (September 2019)
Software version 12.3.2.a-1 resolved the following issue.
Ticket ID Description
SYS-315292 Fixed: System models 415 that were encrypted when on system version prior to
12.2.0 and were shutdown due to a thermal issue or manual procedure might
not be able to be moved into a fully operational state without IBM support
involvement.
Severity: High Impact
Affected versions: 12.2.0 or later
Ticket ID Description
SYS-309772 Fixed: Common security issue. For more information about this security issue,
see (CVE-2017-17833) on the Common Vulnerabilities and Exposures (CVE)
information website (cve.mitre.org).
Severity: High Impact
Affected versions: 12.0.0 or later
SYS-315184 Fixed: In rare cases, in a Multi-site HA DR relation, a required failover will not
occur due to a problem with the primary system, following a disconnected
service reporting an incorrect peer health status.
Severity: High Impact
Affected versions: 12.1.0 or later
SYS-314004 Fixed: In extremely rare cases, when the system is undergoing heavy SCSI
unmap loads combined with excessive write operations, write performance will
be affected. The system resumes normal performance once the heavy unmap
operations complete.
Severity: High Impact
Affected versions: 12.0.0 or later
SYS-315195 Fixed: Common security issue. For more information about these security
issues, see CVE-2019-11477 , CVE-2019-11478, and CVE-2019-11479 on the
Common Vulnerabilities and Exposures (CVE) information website
(cve.mitre.org).
Severity: High Impact
Affected versions: 12.0.0 or later
SYS-310286 Fixed: If the storage system has reached its maximum storage capacity,
performance for user reads and remote mirroring might be degraded.
Severity: Moderate
Affected versions: 12.0.0 or later
SYS-310912 Fixed: When using the host_remove_port CLI command on a port designated
for a HyperSwap volume, the command should remove the port on both the
local and the remote systems. Instead, only the local system port is removed
and the virtual port on the remote system remains configured for the HyperSwap
volume. This can also expose the volume to SCSI reservation conflict until the
virtual port is manually removed.
Severity: Moderate
Affected versions: 12.1.0 or later
SYS-310952 Fixed: In rare cases, deleting an asynchronous mirror relation that is part of a
sync job batch immediately after start, causes other sync jobs to the same
remote target to get stuck. These sync job prolonged delays can be seen by
using the sync_job_list CLI command.
Severity: Moderate
Affected versions: 12.0.0 or later
SYS-311218 Fixed: When a volume has more than 10 user metadata items, the initial sync
occurring after the creation of an IBM Hyper-Scale Mobility or HyperSwap
relation may fail and management operations may not run on the system. This
can be seen when performing various CLI commands and receiving the
MCL_NETWORK_ERROR return code.
This does not impact any I/O operations.
Note: Note: In order to check how many metadata items are listed on a specific
volume, use the metadata_list CLI command.
Severity: Moderate
Affected versions: 12.0.0 or later
Change log 13
Ticket ID Description
SYS-314736 Fixed: In rare cases, in a Multi-site relation environment which includes many
mirroring relations, a target reconnection may cause some I/O delays.
Severity: Moderate
Affected versions: 12.1.0 or later
SYS-314991 Fixed: In rare cases following specific software failures on a secondary system,
when using synchronous mirroring and the volume is part of a consistency
group, management operations on the primary system will not run.
Severity: Moderate
Affected versions: 12.2.1 or later
SYS-315015 Fixed: In some cases, due to an issue in the internal communication, a faulty
FLASH_CANISTER_ETH_LINK_MISWIRE event is emitted. After approximately
one hour, this event is followed by the correcting
FLASH_CANISTER_ETH_WIRING_OK event.
Severity: Moderate
Affected versions: 12.0.0 or later
SYS-315029 Fixed: In a Multi-site HA DR configuration, a user will not be able to add a Multi-
site volume into a consistency group, if the following order of events have
occurred:
1. A standby replication relation was activated for a specific consistency group.
2. A role change occurred.
3. The new relation is not yet been activated for this consistency group.
Severity: Moderate
Affected versions: 12.3.0 or later
SYS-315146 Fixed: Flash enclosure daily snap collections may result in momentary latency
increase.
Severity: Moderate
Affected versions: 12.3.2.a
SYS-310343 Fixed: The status of the backup battery unit (BBU) in the flash enclosure might
be shown as failed while it is still operating normally. If this occurs, contact IBM
Support.
Severity: Service
Affected versions: 12.3.0 or later
SYS-314453 Fixed: In rare cases, latency spikes in system statistics might unnecessarily be
emitted. This is a false identification and can be ignore. This behavior has no
actual impact on system performance.
Severity: Service
Affected versions: 12.0.0 or later
Ticket ID Description
SYS-313334 Fixed: In rare conditions, a bug in the microcode might lead to undetected data
corruption (a range of 16 MB returning old data instead of zeros), on the
Secondary volume of any type of mirroring relationship: synchronous mirroring,
asynchronous mirroring, HyperSwap, or Multi-site HA DR.
Severity: HIPER
Affected versions: 12.0.0 or later
SYS-313522 Fixed: In rare cases, a CPU error might lead to a SCSI MEDIUM ERROR
condition.
Severity: High Impact
Affected versions: 12.2.1 or later
SYS-308548 Fixed: In a HyperSwap relation, when a remote system target is deleted and
then redefined, creating new HyperSwap relations on this target system is not
possible.
Severity: Moderate
Affected versions: 12.1.0 or later
SYS-310148 Fixed: In rare cases, due to a unique chain of events resulting in an internal
resource depletion, disconnected Fibre Channel hosts may not be able to
automatically reconnect.
Severity: Moderate
Affected versions: 12.0.0 or later
SYS-310990 Fixed: In rare cases where an internal process fails, the reported connectivity
for FC or iSCSI hosts might be wrong when running the
host_connectivity_list and fc_connectivity_list CLI commands.
Severity: Service
Affected versions: 12.1.0 or later
SYS-311467 Fixed: Statistics information is not kept for target ports in mirror relations.
Severity: Service
Affected versions: 12.0.0 or later
Change log 15
Ticket ID Description
SYS-310338 Fixed: In rare cases, for a system with encryption enabled:
1. During a replacement procedure of a malfunctioning MicroLatency module,
the procedure might fail and the new MicroLatency module might stay offline,
thus compromising the system redundancy.
2. During firmware upgrade of the flash enclosure, the procedure might halt and
the flash enclosure upgrade can not complete.
In both of the cases described above, a maintenance window might have to be
scheduled for a complete restart of the system.
Severity: HIPER
Affected versions: 12.2.0 or later
SYS-309744 Fixed: In rare cases, if the InfiniBand interconnect fabric of an A9000R system
becomes overloaded, an InfiniBand switch port might go offline for a short
period and then get back online. If at the same time, another issue such as a
port failure causes the grid controller to be connected to the InfiniBand fabric
via a single InfiniBand link, the grid controller might fail. If at the same time
additional grid controllers fail, hosts could lose access to the storage system.
Severity: High Impact
Affected versions: 12.0.0 or later
SYS-310091 Fixed: In rare cases, a flash RAID controller malfunction and a particular write
error might cause MicroLatency modules on the flash enclosure to fail, leading
to loss of access to the storage system.
Severity: High Impact
Affected versions: 12.2.0 or later
SYS-308309 Fixed: In rare cases, a replacement of a flash canister in the storage system
might cause performance degradation.
Severity: Moderate
Affected versions: 12.0.0 or later
SYS-311260 Fixed: If performance classes are used, the I/O bandwidth allocation that is
associated with some performance classes might be incorrectly calculated. This
could lead to incorrect performance limits for hosts or volumes.
Severity: Moderate
Affected versions: 12.0.0 or later
SYS-311325 Fixed: If the size of a written volume and its snapshots is larger than the total
volume size, the vol_list CLI command shows only the volume total size,
without the size of the volume snapshots.
Severity: Service
Affected versions: 12.0.0 or later
Ticket ID Description
SYS-309775 Fixed: In rare cases, after software upgrade on a system of model 425, the
system might experience growing latencies in I/O operations, until all hosts lose
access to the storage system.
Severity: HIPER
Affected versions: 12.3.0
SYS-310311 Fixed: In rare cases, systems that are using snapshots or asynchronous remote
mirroring might disconnect when a volume deletion operation is performed. In
such a case, the DATA_REDUCTION_TIER_BECOMING_OFFLINE event is
logged.
Severity: High Impact
Affected versions: 12.2.0 or later
Ticket ID Description
SYS-307098 Fixed: In a HyperSwap setup with an active SCSI reservation, if only a subset of
the host ports that are defined on the System holding the Primary volume, is
defined on the System which holds the Secondary volume, the latter system
might not conform to the SCSI reservation. As a result, the Secondary volume
might be read by another host; and after a HyperSwap failover, the new Primary
volume (former Secondary volume) might be written to by another host. This
violation of SCSI reservation will be undetected.
Severity: HIPER
Affected versions: 12.1.0 or later
SYS-305091 Fixed: In rare cases, using the io_pause CLI command might cause loss of
access to the storage system.
Severity: High Impact
Affected versions: 12.0.0 or later
Change log 17
Ticket ID Description
SYS-305121 Fixed: When using IBM Hyper-Scale Mobility and the volume migration is in
Proxy state, a broken link to the destination system might cause hosts to lose
access to one or more of the Proxy-state interface modules.
Severity: High Impact
Affected versions: 12.1.0 or later
SYS-306086 Fixed: When most of the physical memory of the storage system is occupied by
different operation processes, deleting a significant amount of obsolete data and
then restarting I/O operations before the data garbage collection process has
completed - might result in loss of access of hosts to the storage system.
Severity: High Impact
Affected versions: 12.0.0 or later
SYS-306139 Fixed: Before a hot upgrade, all primary HyperSwap volumes must be
deactivated in order to make the secondary volumes unavailable. Otherwise
local I/O operations during the hot upgrade will be blocked.
Severity: High Impact
Affected versions: 12.1.0 or later
SYS-306457 Fixed: In rare cases, the data redistribution after the addition of a grid element
might fail due to a data transfer error. If this occurs, hot upgrade of the storage
system software will not be possible.
Severity: High Impact
Affected versions: 12.0.3 or later
SYS-306775 Fixed: The iSCSI port driver might crash when invalid data length is specified in
a read or write I/O command, causing loss of access of all iSCSI hosts to the
storage system.
Severity: High Impact
Affected versions: 12.0.0 or later
SYS-307118 Fixed: In rare cases, a HyperSwap relation might get disconnected when trying
to synchronize SCSI registrations between the Primary and the Secondary
systems.
Severity: High Impact
Affected versions: 12.1.0 or later
SYS-307426 Fixed: In a HyperSwap relation, a host issuing a LUN reset task might cause
failure of all subsequent reservation operations, causing the HyperSwap relation
to break.
Severity: High Impact
Affected versions: 12.1.0 or later
SYS-307626 Fixed: In rare cases, a storage system might fail to isolate a failing component,
but wrongly detect that the component is successfully isolated. This might result
in hosts losing access to the storage system, or in data unavailability.
Severity: High Impact
Affected versions: 12.0.0.a or later
SYS-307715 Fixed: In rare cases, when hosts are connected to the storage system via iSCSI,
a Send Targets request (sent from a host) might cause a loss of access for all
hosts connected to the storage system.
Severity: High Impact
Affected versions: 12.2.0 or later
SYS-307743 Fixed: In rare cases, after the recovery of a failed software component, the
connectivity to a remote mirroring target might not be restored.
Severity: High Impact
Affected versions: 12.1.0 or later
SYS-307854 Fixed: In rare cases, after performing hot upgrade while using Fibre Channel
(FC) connectivity, the FC driver process might crash and then restart
automatically. In some system or host configurations, this might result in hosts
losing access to the storage system.
Severity: High Impact
Affected versions: 12.0.0 or later
SYS-307987 Fixed: In rare cases, when using asynchronous mirroring, specific write
sequences (writes of 0) combined with a mirror deactivation procedure may
result in loss of access upon re-activation. On even rarer occasions, loss of data
can occur.
Severity: High Impact
Affected versions: 12.0.0 or later
SYS-308098 Fixed: In rare cases, if the hot upgrade of a storage system that is using
encryption is canceled, an internal component failure might trigger an
emergency shutdown procedure, following by loss of access by hosts to the
storage system.
Severity: High Impact
Affected versions: 12.0.0 or later
Change log 19
Ticket ID Description
SYS-304905 Fixed: When performing a system wipe-out operation and then disabling the
system's data encryption (for example, for the purpose of moving the system
off-premise), the system performs several emergency shutdowns before the
encryption disabling is completed.
Severity: Moderate
Affected versions: 12.2.1 or later
SYS-306078 Fixed: In rare cases, I/O operations might be rejected due to an IOPS or
bandwidth limit of a QoS performance class, even though these I/O operations
are not subject to any QoS performance class limitation.
Severity: Moderate
Affected versions: 12.0.0 or later
SYS-306353 Fixed: Mirrored volumes that are not part of a consistency group (CG) cannot
benefit from the 12.2.0 software enhancements that allow configuring an RPO of
1 minute and mirroring interval of 20 seconds.
Severity: Moderate
Affected versions: 12.2.0 or later
SYS-306627 Fixed: In rare cases, after a hot upgrade of a Primary system in a HyperSwap
relationship over Fibre Channel (FC) connectivity, the FC driver process might
crash and then restart automatically, causing temporary HyperSwap
connectivity disconnection.
Severity: Moderate
Affected versions: 12.0.0 or later
SYS-307815 Fixed: In rare cases, a particular sequence of read and write I/O operations
might cause performance degradation.
Severity: Moderate
Affected versions: 12.2.0 or later
SYS-307088 Fixed: In rare cases, the size of the volume in the data migration source might
be incorrectly reported when using the dm_define or the dm_activate CLI
commands. If this occurs, data migration cannot be performed.
Severity: Moderate
Affected versions: 12.0.3.b or later
SYS-307675 Fixed: In rare cases, internal disconnections between a grid controller and the
flash enclosure might increase the latency. If this occurs, the
IO_TOOK_TOO_LONG event is logged.
Severity: Moderate
Affected versions: 12.2.1 or later
SYS-280653 Fixed: Some CLI commands might hang when storage pools are being cleaned
from snapshots.
Severity: Service
Affected versions: 12.0.0 or later
SYS-305901 Fixed: In rare cases, after a reboot or software upgrade of the storage system,
the output of the fan_list CLI command might list fans as failed, even though
the fans are fully operational.
Severity: Service
Affected versions: 12.0.0 or later
SYS-307419 Fixed: If the storage system stores secondary HyperSwap volumes, hot upgrade
of this storage system might fail if the HyperSwap relation is not disabled
beforehand.
Severity: Service
Affected versions: 12.2.1 or later
SYS-306886 Fixed: In the output of the ipinterface_list_ports CLI command, all iSCSI
ports might be listed with the Link Up status, even though some ports have no
connectivity.
Severity: Low
Affected versions: 12.0.0 or later
Change log 21
22 IBM FlashSystem A9000R: Release Notes
Fixes and features of earlier versions
IBM FlashSystem A9000R version 12.3.2.b includes all the features and fixes that were provided in
previous 12.0.x, 12.1.x, 12.2.x, 12.3.0.x, 12.3.1.x, 12.3.2, 12.3.2.a, 12.3.2.a-1, 12.3.2.a-2, and
12.3.2.a-3 versions.
To obtain information regarding previously introduced FlashSystem A9000R features or issues that were
resolved in previous versions, see FlashSystem A9000R on IBM Knowledge Center (www.ibm.com/
support/knowledgecenter/STJKN5).
Important:
• The issues listed below apply to version 12.3.2.b or earlier versions. As long as a newer version has
not yet been released, a newer release notes edition for version 12.3.2.b might be issued to provide a
more updated list of known issues and workarounds.
• When a newer version is released for general availability, the release notes of version 12.3.2.b will no
longer be updated. Accordingly, check the release notes of the newer version to learn whether any
newly discovered issues affect version 12.3.2.b or whether the newer version resolves any of the
issues listed below.
Affected
Ticket ID Severity versions Description
SYS-305091 High Impact 12.0.0 In rare cases, using the io_pause command might cause loss
or later of access to the storage system.
Note: Starting from version 12.2.1, this CLI command is
disabled until a fix for this issue becomes available.
Workaround: No workaround is currently available.
SYS-305121 High Impact 12.1.0 When using IBM Hyper-Scale Mobility and the volume migration
or later is in Proxy state, a broken link to the destination system might
cause hosts to lose access to one or more of the Proxy-state
interface modules.
Workaround: No workaround is currently available.
SYS-305998 High Impact 12.1.0 When the storage system enters an out-of-physical-space state
or later and the SYSTEM_OUT_OF_PHYSICAL_SPACE event is logged,
freeing up physical space will allow the storage system to return
to normal operation. However, in rare cases, the storage system
might enter the same state again, even if there is some free
space available.
Workaround: Contact IBM Support.
SYS-305999 High Impact 12.0.0 When an encrypted storage system is powered up, it may enter
or later the maintenance state due to an issue in the encryption key
management mechanism.
Workaround: Contact IBM Support.
SYS-316182 High Impact 12.3.2 - In extremely rare cases, in a system with any type of mirroring
12.3.2.b relation, a certain sequence of events, combined with a specific
write pattern of "zeros" to a deactivated mirror relation,
followed by a role change and specific time frame of the mirror
re-activation, an internal race condition might lead to
undetected data corruption on the primary system. This may
cause invalid data to be returned to the host as undetected data
corruption.
Workaround: To avoid this issue, upgrade to microcode version
12.3.2.c or later as soon as possible.
If your system is on an affected version and this specific
sequence of events is necessary, contact IBM Support.
SYS-316205 High Impact 12.3.0 - In extremely rare cases, when the write_same of "zeros" is
12.3.2.b used in a combination with the vol_copy command, an
incorrect handling of the written zeros on the backup cache may
occur. If the system later performs an automatic data
redistribution, this may result in one of the following:
• Undetected data corruption when reading the data.
• Multiple cache failures leading to loss of access during write
attempts to the data.
Workaround: To avoid this issue, upgrade to microcode version
12.3.2.c or later as soon as possible.
If your system is on an affected version, in order to prevent the
issue from occurring, contact IBM Support.
SYS-285786 Moderate 12.1.0 When unmapping a migrated volume (with IBM Hyper-Scale
or later Mobility) or a HyperSwap volume from a cluster, a delay of the
host I/O response might be experienced, ranging from 20 to
170 milliseconds.
Workaround: No workaround is currently available.
SYS-304905 Moderate 12.2.0 When performing a system wipe-out operation and then
or later disabling the system's data encryption (for example, for the
purpose of moving the system), the system performs several
emergency shutdowns before the encryption disabling is
completed.
Workaround: Manually restart the system and allow the
decryption process to complete.
SYS-314668 Moderate 12.3.2 IP filtering is not enforced by the IP_ACCESS CLI command
or later when the management ports are assigned with a VLAN ID.
Workaround: No workaround is currently available.
SYS-314708 Moderate 12.0.0 or When using the host_delete CLI command on one of the
later secondary system ports designated for a HyperSwap volume
only the secondary system port is removed and the port on the
primary system remains configured. This can expose the
volume to SCSI reservation conflict, causing unexpected access
restrictions until the port is removed.
Workaround: To avoid this issue, be sure to remove the port
from the primary system before deleting the host, using the
host_remove_port CLI command. If the port was not
removed prior to using the host_delete CLI command,
contact IBM support.
SYS-315170 Moderate 12.0.1 or In some cases when using a KeySecure key server, the
later encrypt_disable CLI command fails to complete and the
ENCRYPT_UNABLE_TO_UPDATE_KEY_DURING_DEACTIVATE_
ON_KEYSERVER event is emitted.
Workaround: If the key already exists, and the
ENCRYPT_UNABLE_TO_UPDATE_KEY_DURING_DEACTIVATE_
ON_KEYSERVER is emitted, go to the Life Cycle properties
dialog box of the key on the key server, and change the state
from Pre-Active to Active. Once the key state is Active, repeat
the encrypt_disable CLI command.
SYS-279815 Service 12.0.0 If a MicroLatency module is pulled out of the flash enclosure
or later when the enclosure's internal redundancy state is either
'Synching' or 'Degraded', the management GUI indicates that
the card has a spare, even if no spare is available.
Workaround: Use the flash_card_list CLI command to
display the correct status of spares.
Known issues 29
Affected
Ticket ID Severity versions Description
SYS-280653 Service 12.0.0 Some CLI commands might hang when storage pools are being
or later cleaned from snapshots.
Workaround: In such a case, wait until the command is carried
out.
SYS-282223 Service 12.0.0 Data migration over Fibre Channel (FC) does not work if the
or later number of LUNs per host exceeds 255.
Workaround: To resolve this issue, use 255 or less LUNs per
host.
SYS-284723 Service 12.0.0 Enabling encryption of a flash enclosure that contains a failed
or later flash canister could lead to partial encryption of the flash
enclosure, or to complete encryption failure.
Workaround: To avoid this issue, before enabling encryption,
verify that no flash enclosure contains a failed canister.
Otherwise, contact IBM Support.
SYS-284298 Service 12.0.0 When setting up LDAP authentication, entering the storage
or later system password is required, but this requirement is not
enforced. This could lead to LDAP authentication failure if the
password is not provided.
Workaround: To avoid this issue, always make sure to provide
the storage system password in order to allow access to the
storage system.
SYS-284512 Service 12.1.0 An attempt to deactivate a Quorum Witness might render the
or later Quorum Witness unresponsive, and remain in the Deactivating
state without timing out (see also SYS-285120 below).
Workaround: To resolve this issue, proceed as follows in IBM
Hyper-Scale Manager:
1. Remove the current Quorum Witness configuration.
This action disassociates the Quorum Witness from the
connected system(s).
2. Define a new Quorum Witness and make sure that the
correct port is used.
3. Activate the newly defined Quorum Witness.
4. Connect the system(s), that were previously using the
removed Quorum Witness, to the newly defined Quorum
Witness.
SYS-286013 Service 12.0.0 When an IBM Support technician is removing a canister from
or later the flash enclosure for maintenance purposes, multiple events
of
POD_IB_LINK_DETECTION_LINK_PERSISTENTLY_DISCONN
ECTED might be unnecessarily emitted.
These events can be safely ignored.
SYS-287068 Service 12.1.0 In rare cases, the ha_create command might fail.
or later
Workaround: Retry the operation.
SYS-287305 Service 12.1.0 If there are more than 200 volumes in a HyperSwap relation
or later and a failover takes place, some of the
HA_AUTOMATIC_FAILOVER_SUCCESFUL events might be lost.
This does not affect the actual failover.
Workaround: No workaround is currently available.
SYS-287279 Service 12.1.0 When removing a target port and adding it back within less than
or later 3 seconds, the high availability state for HyperSwap volumes is
insufficient, and a
TARGET_CONNECTIVITY_HA_INSUFFICIENT event is issued.
Workaround: No workaround is currently available.
Known issues 31
Affected
Ticket ID Severity versions Description
SYS-287533 Service 12.1.0 or Previously listed as SYS-310649.
later
In rare cases, when adding a volume to a consistency group
(CG) with the cg_add_vol command, the operation might fail
with the event MIRROR_LAST_SYNC_TIMES_DIFFER recorded.
Workaround: To avoid this issue, retry the cg_add_vol
command after the mirroring synchronization is completed.
SYS-305433 Service 12.1.0 When using IBM Hyper-Scale Mobility and aborting (with
or later olvm_abort) a migration process that is currently in Proxy
state, the remote volume remains in locked read-only state.
Workaround: Use the vol_unlock command to unlock the
volume state.
SYS-305450 Service 12.0.0 Manually created snapshots that have a reserved name prefix
or later prevent the automatic creation of additional snapshots.
Workaround: Delete and do not manually create snapshots that
have a reserved name prefix.
SYS-305581 Service 12.0.0 The switch_mgmt_ip_list CLI command does not work
or later when using the additional switch argument for displaying the
management IP address of a specific InfiniBand switch.
Workaround: Use the switch_mgmt_ip_list command
without the switch argument. This will display the
management IP addresses of all InfiniBand switches.
SYS-313056 Service 12.1.0 When using HyperSwap and a volume fails to be synchronized,
or later the VOLUME_IS_NOT_CONSISTENT_OLVM_DESTINATION
event is logged instead of the
VOLUME_IS_NOT_CONSISTENT_HA_DESTINATION event.
Workaround: No workaround is currently available.
SYS-310909 Service 12.2.1 In rare cases, the Fibre Channel (FC) driver might crash and
or later then immediately be restarted automatically. In such a case,
the FC_LINK_IS_NOW_UP event is logged.
Workaround: No workaround is currently available.
SYS-276870 Low 12.0.0 When a grid controller is replaced, or whenever the system is
or later undergoing a software initiation process, data reduction saving
statistics could be temporarily inaccurate. The statistics
accuracy is regained as data is being accessed and processed
again.
This issue has a noticeable effect only when 1 TB or less has
been written to the system.
Workaround: No workaround is currently available.
SYS-306320 Low 12.0.0 In rare cases, the process of enabling encryption may not
or later complete successfully due to a timeout. If this occurs, the
system emits the ENCRYPT_ENABLE_NOT_COMPLETED event
with all the relevant details.
Workaround: Retry activating encryption.
SYS-306814 Low 12.0.1 By design, an attempt to set up IBM Hyper-Scale Mobility, when
or later a volume with the same name as the source volume already
exists at the destination system, fails due to an illegal volume
name. However, the emitted return code
(REMOTE_VOLUME_NOT_APPLICABLE_FOR_OLVM) provides a
wrong reason for the failure.
Workaround: No workaround is currently available.
SYS-313010 Low 12.0.0 When iSCSI connectivity is used with a host, the
or later HOST_NO_MULTIPATH_ONLY_ONE_PORTevent might be
logged, even though the host is connected through more than
one iSCSI path.
Workaround: No workaround is currently available.
Known issues 33
34 IBM FlashSystem A9000R: Release Notes
Known interoperability issues
This section describes known issues or limitations that could arise when using IBM FlashSystem A9000R
in conjunction with other components in the production environment, such as hardware switches, host
bus adapters (HBAs), other IBM products, or different software platforms such as operating systems,
clustering solutions, virtual machine servers, multipathing solutions, and other third-party components.
Affected
Ticket ID Severity versions Description
SYS-286719 Service 12.1.0 Failover of a Windows Server 2008 or 2008 R2 cluster group
or later results in a stop error on the new active node. This could occur
when using these Windows Server versions with HyperSwap
volumes. No hotfix is currently available.
Workaround: Refer to IBM Support guidelines for Windows
Server 2008 and 2008 R2 as described in IBM Technote
#S1010299.
SYS-286992 Service 12.1.0 The HyperSwap feature is not compatible with VMware ESXi 6.0
or later Update 2 or earlier.
Workaround: Use ESXi 6.0 Update 3 or later.
SYS-287993 Service 12.1.0 Veritas Dynamic Multi-Pathing (DMP) in any configuration does
or later not support HyperSwap volumes.
Workaround: No workaround is currently available.
If you need help, service, technical assistance, or want more information about IBM products, you can
find various sources to assist you. You can view the following websites to get information about IBM
products and services and to find the latest technical information and support.
• IBM website (ibm.com)
• IBM Support Portal website (ibm.com/storage/support)
• IBM Directory of Worldwide Contacts website (ibm.com/planetwide)
• IBM service requests and PMRs (ibm.com/support/servicerequest/Home.action)
Use the Directory of Worldwide Contacts to find the appropriate phone number for initiating voice call
support. Voice calls arrive to Level 1 or Front Line Support.
For license inquiries regarding double-byte character set (DBCS) information, contact the IBM Intellectual
Property Department in your country or send inquiries, in writing, to:
Any references in this information to non-IBM Web sites are provided for convenience only and do not in
any manner serve as an endorsement of those Web sites. The materials at those Web sites are not part of
the materials for this IBM product and use of those Web sites is at your own risk.
IBM may use or distribute any of the information you supply in any way it believes appropriate without
incurring any obligation to you.
Licensees of this program who wish to have information about it for the purpose of enabling: (i) the
exchange of information between independently created programs and other programs (including this
one) and (ii) the mutual use of the information which has been exchanged, should contact:
Such information may be available, subject to appropriate terms and conditions, including in some cases,
payment of a fee.
The licensed program described in this document and all licensed material available for it are provided by
IBM under terms of the IBM Customer Agreement, IBM International Program License Agreement or any
equivalent agreement between us.
The performance data discussed herein is presented as derived under specific operating conditions.
Actual results may vary.
Information concerning non-IBM products was obtained from the suppliers of those products, their
published announcements or other publicly available sources. IBM has not tested those products and
cannot confirm the accuracy of performance, compatibility or any other claims related to non-IBM
products. Questions on the capabilities of non-IBM products should be addressed to the suppliers of
those products.
All statements regarding IBM's future direction or intent are subject to change or withdrawal without
notice, and represent goals and objectives only.
Trademarks
IBM, IBM FlashSystem, the IBM logo, and ibm.com are trademarks or registered trademarks of
International Business Machines Corp., registered in many jurisdictions worldwide.
Other product and service names might be trademarks of IBM or other companies. A current list of IBM
trademarks is available on the Copyright and trademark information website (ibm.com/legal/us/en/
copytrade.shtml).
HP, HP-UX, and HP UNIX are trademarks or registered trademarks of Hewlett Packard Company in the
United States, other countries, or both.
Linux is a trademark of Linus Torvalds in the United States, other countries, or both.
Microsoft, Windows, Windows Server, and the Windows logo are trademarks or registered trademarks of
Microsoft Corporation in the United States, other countries, or both.
Oracle and Solaris are trademarks or registered trademarks of Oracle and/or its affiliates.
UNIX is a registered trademark of The Open Group in the United States and other countries.
VMware, ESX, ESXi, vSphere, vCenter, and vCloud are trademarks or registered trademarks of VMware
Corporation in the United States, other countries, or both.
Other product and service names might be trademarks of IBM or other companies.
Printed in USA