Release Note for systems built with IBM Spectrum Virtualize

This is the release note for the 8.4 release and details the issues resolved in all Program Temporary Fixes (PTFs) between 8.4.0.0 and 8.4.0.13. This document will be updated with additional information whenever a PTF is released.

Note: The v8.4.0.4 release is only provided, pre-installed, on new systems and will not be available on IBM Fix Central.

This document was last updated on 8th February 2024.

  1. New Features
  2. Known Issues and Restrictions
  3. Issues Resolved
    1. Security Issues Resolved
    2. APARs Resolved
  4. Useful Links

Note. Detailed build version numbers are included in the Update Matrices in the Useful Links section.


1. New Features

The following new features have been introduced in the 8.4.0.13 release: The following new features have been introduced in the 8.4.0.1 release: The following new features have been introduced in the 8.4.0 release:

2. Known Issues and Restrictions

Note: For clarity, the terms "node" and "canister" are used interchangeably.
Details Introduced

The three site orchestrator is not compatible with the new SSH security level 4.This will be resolved in a future release of the orchestrator.

8.4.0.13

Customers planning to upgrade to v8.4.0 or later should be aware that an update to OpenSSH has terminated support for all DSA keys. An update to OpenSSL has also terminated support for 1024-bit RSA keys used in SSL certificates.

Customers currently using DSA public keys for SSH access will need to generate new keys using alternative ciphers, such as RSA or ECDSA. If using RSA public keys for SSH access, it is recommended to use keys of 2048 bits or longer.

Customers currently using 1024-bit RSA keys in SSL certificates will need to generate new SSL certificates using 2048-bit RSA keys or ECDSA keys. This applies not only to the system certificate, but also to any SSL certificates used by external services such as LDAP servers.

 8.4.0.0

There is an existing limit on the number of files that can be returned by the CLI of approximately 780 entries. In many configurations this limit is of no concern. However, due to a problem with hot-spare node I/O stats files, 8-node clusters with many hardware upgrades or multiple spare nodes may see up to 900 I/O stats files. As a consequence the data collector for Storage Insights and Spectrum Control cannot list or download the required set of performance statistics data. The result is that there are many gaps in the performance data, leading to errors with the performance monitoring tools and a lack of performance history.

The workaround is to remove the files associated with spare nodes or previously/updated hardware using the cleardumps command (or to cleardumps the entire iostats directory).

This is a known issue that will be lifted in a future PTF. The fix can be tracked using APAR HU02403.

 8.4.0.0

Customers using Spectrum Control v5.3.2, or earlier, may notice some discrepancies between attributes displayed and related attributes on the Spectrum Virtualize GUI.

This issue will be resolved by a future release of Spectrum Control.

 8.3.1.0

Under some circumstances, the "Monitoring > System" GUI screen may not load completely.

This is a known issue that will be lifted in a future PTF.

 8.3.1.0

Customers using iSER attached hosts, with Mellanox 25G adapters, should be aware that IPv6 sessions will not failover, for example, during a cluster upgrade.

This is a known issue that may be lifted in a future PTF.

 8.3.0.0

Systems, with NPIV enabled, presenting storage to SUSE Linux Enterprise Server (SLES) or Red Hat Enterprise Linux (RHEL) hosts running the ibmvfc driver on IBM Power can experience path loss or read-only file system events.

This is cause by issues within the ibmvfc driver and VIOS code.

Refer to this troubleshooting page for more information.

 n/a
If an update stalls or fails then contact IBM Support for further assistance n/a
The following restrictions were valid but have now been lifted

For systems using remote copy without the 3-site functionality it is possible to trigger a resource leak when opening the remote copy panel in the GUI. After the remote copy panel has been opened multiple times this resource leak can reach the point where the GUI becomes unresponsive.

Restarting the GUI tomcat service will restore GUI access.

This issue has been resolved by APAR HU02319.

 8.4.0.1

Due to an issue in IP Replication, customers using this feature should not upgrade to v8.4.0.0 or later.

This issue has been resolved by APAR HU02340.

 8.4.0.0

Customers with SVC model SV1 systems that have iWARP 25G adapters, presenting storage to iSCSI hosts, should not upgrade those systems to v8.4.0.0 or later.

This issue has been resolved by APAR HU02301.

 8.4.0.0

Due to a performance issue, customers using Enhanced Callhome with censored mode should not upgrade to v8.4.0.0 or later.

This issue has been resolved by APAR HU02300.

 8.4.0.0

Configuration node warmstart will occur if mkhostcluster is run with -ignoreseedvolume and the ignored volumes have an id greater than 256.

This issue has been resolved by APARs HU02303 & HU02305 in PTF v8.4.0.2.

 8.3.1.0

Validation in the Upload Support Package feature will reject the new case number format in the PMR field.

This issue has been resolved by APAR HU02392.

 7.8.1.0

3. Issues Resolved

This release contains all of the fixes included in the 8.3.1.2 release, plus the following additional fixes.

A release may contain fixes for security issues, fixes for APARs or both. Consult both tables below to understand the complete set of fixes included in the release.

3.1 Security Issues Resolved

Security issues are documented using a reference number provided by "Common Vulnerabilities and Exposures" (CVE).
CVE Identifier Link for additional Information Resolved in
CVE-2023-22081 7114770 8.4.0.13
CVE-2023-22067 7114770 8.4.0.13
CVE-2023-5676 7114770 8.4.0.13
CVE-2023-46589 7114769 8.4.0.13
CVE-2023-45648 7114769 8.4.0.13
CVE-2023-42795 7114769 8.4.0.13
CVE-2024-21733 7114769 8.4.0.13
CVE-2023-50164 7114768 8.4.0.13
CVE-2023-2597 7065011 8.4.0.12
CVE-2023-34396 7065010 8.4.0.12
CVE-2022-21626 6858041 8.4.0.10
CVE-2022-1012 6858043 8.4.0.10
CVE-2021-45485 6858043 8.4.0.10
CVE-2021-45486 6858043 8.4.0.10
CVE-2022-43873 6858047 8.4.0.10
CVE-2022-42252 6858039 8.4.0.10
CVE-2022-43870 6858045 8.4.0.10
CVE-2022-0778 6622017 8.4.0.9
CVE-2021-35603 6622019 8.4.0.9
CVE-2021-35550 6622019 8.4.0.9
CVE-2018-25032 6622021 8.4.0.7
CVE-2022-25762 6622023 8.4.0.7
CVE-2021-38969 6584337 8.4.0.6
CVE-2021-42340 6541270 8.4.0.6
CVE-2021-29873 6497111 8.4.0.5
CVE-2020-10732 6497113 8.4.0.5
CVE-2020-10774 6497113 8.4.0.5
CVE-2021-33037 6497115 8.4.0.5
CVE-2020-2781 6445063 8.4.0.0
CVE-2020-13935 6445063 8.4.0.0
CVE-2020-14577 6445063 8.4.0.0
CVE-2020-14578 6445063 8.4.0.0
CVE-2020-14579 6445063 8.4.0.0

3.2 APARs Resolved

Show details for all APARs
APAR Affected Products Severity Description Resolved in Feature Tags
SVAPAR-94179 FS5100, FS5200, FS7200, FS9100, FS9200, V7000 HIPER Faulty hardware within or connected to the CPU can result in a reboot on the affected node. However it is possible for this to sometimes result in a reboot on the partner node (show details)
Symptom Loss of Access to Data
Environment All Flashsystems and V7000 Gen3, but not SVC
Trigger Node hardware fault
Workaround None
8.4.0.12 Reliability Availability Serviceability
SVAPAR-89694 All HIPER Kernel panics might occur on a subset of Spectrum Virtualize Hardware Platforms with a 10G Ethernet adapter running 8.4.0.10, 8.5.0.7 and 8.5.3.1 when taking a snap. For more details refer to this Flash (show details)
Symptom Loss of Access to Data
Environment V7000 Gen2 & Gen2+, SVC nodes (DH8 & SV1) and FS50xx systems with 10G Ethernet adapters are affected. Nodes with different node hardware, e.g. FS5200 are not affected even if they have a 10G Ethernet adapter installed
Trigger Taking a snap, livedump or when a node warmstarts
Workaround Do not take snaps or livedumps
8.4.0.11
SVAPAR-84116 All Critical The background delete processing for deduplicated volumes might not operate correctly if the preferred node for a deduplicated volume is changed while a delete is in progress. This can result in data loss which will be detected by the cluster when the data is next accessed (show details)
Symptom Data loss detected by the cluster when the data is accessed
Environment Systems running with deduplicated volumes on code levels up to 8.4.0.10, 8.5.0.0 through 8.5.0.7 and 8.5.1 through 8.5.4 are vulnerable to APAR SVAPAR-84116. The fix is available in 8.4.0.11 and 8.5.0.8
Trigger The background delete processing for deduplicated volumes might not operate correctly if the preferred node for a deduplicated volume is changed while a delete of another volume is in progress. This can result in data loss which will be detected by the cluster when the data is next accessed
Workaround Commands that change the preferred node of a deduplicated volume should not be run while another deduplicated volume is in a deleting state. These commands are: splitvdiskcopy and movevdisk
8.4.0.11 Data Reduction Pools, Deduplication
HU02420 All Critical During an array copyback it is possible for a memory leak to result in the progress stalling and an node warmstart of all nodes resulting in a temporary loss of access (show details)
Symptom Multiple node warmstarts
Environment Systems running < 8.4
Trigger Memory leak that causes resources used for the copyback to become depleted
Workaround None
8.4.0.10 RAID
HU02471 All Critical After starting a FlashCopy map with -restore in a graph with a GMCV secondary disk that was stopped with -access there can be a data integrity issue (show details)
Symptom Data Integrity Loss
Environment Systems using GMCV
Trigger Given a configuration where there is a GMCV secondary volume A that has a cascade of two FlashCopy maps to volumes B and C (A -> B -> C) The mapping B -> C has been started and GMCV is running Stop the GMCV relationship with -access Start FlashCopy map A -> B with -restore Any I/O to volume A will corrupt data on volume C
Workaround Wait for a GMCV backward map to complete before starting related FlashCopy maps with -restore
8.4.0.10 FlashCopy, Global Mirror with Change Volumes
HU02561 All Critical If there are a high number of FC mappings sharing the same target, the internal array that is used to track the FC mapping is mishandled, thereby causing it to overrun. This will cause a cluster wide warmstart to occur (show details)
Symptom Multiple node warmstarts
Environment System running v8.3.1,v8.4.0,v8.4.1,v8.4.2,v8.5.0
Trigger Cascaded flashcopy mappings and one of the flashcopy target volumes is the source of 255 flashcopy mappings
Workaround None
8.4.0.10 FlashCopy
HU02563 All Critical Improve dimm slot identification for memory errors (show details)
Symptom Single node warmstart
Environment System running v8.4.0,v8.4.1,v8.4.2,v8.5.0,v8.5.1,v8.5.2
Trigger Bad memory module
Workaround Work with IBM Support to replace the DIMM that reported the uncorrectable error and replace it
8.4.0.10 Reliability Availability Serviceability
IT41088 V5000, V5100, FS5000, FS5100, FS5200 Critical Systems with low memory that have a large number of RAID arrays that are resyncing can cause a system to run out of RAID rebuild control blocks (show details)
Symptom Loss of Access to Data
Environment Low memory systems such as 5015/5035
Trigger Systems with 64gb or less of cache with resync operations spread across multiple RAID arrays
Workaround None
8.4.0.10 RAID
SVAPAR-86139 All Critical Failover for VMware iSER hosts may pause I/O for more than 120 seconds (show details)
Symptom Loss of Access to Data
Environment Systems presenting volumes to VMware iSER hosts
Trigger None
Workaround None
8.4.0.10 Hosts
HU01782 All High Importance A node warmstart may occur due to a potentially bad SAS hardware component on the system such as a SAS cable, SAS expander or SAS HIC (show details)
Symptom Loss of access to data
Environment Systems that have a faulty SAS hardware component
Trigger Faulty SAS component
Workaround None
8.4.0.10 SAS
HU02010 All High Importance A single node warmstart may occur when a drive in a non-distributed RAID array is taken temporarily out-of-sync due to slow performance (show details)
Symptom Single node warmstart
Environment Systems with non-distributed RAID arrays
Trigger Slow drive
Workaround None
8.4.0.10 RAID
HU02088 All High Importance There can be multiple node warmstarts when no mailservers are configured (show details)
Symptom Multiple Node Warmstarts
Environment Systems running v8.1 or later
Trigger None
Workaround Configure a mailserver
8.4.0.10 System Monitoring
HU02439 All High Importance An IP partnership between a pre-v8.4.2 system and v8.4.2 or later system may be disconnected because of a keepalive timeout (show details)
Symptom Degraded Operation
Environment Systems running 8.3.1 or earlier in IP partnerships with systems running v8.4.2 or later
Trigger None
Workaround Ensure both systems are running v8.4.2 or later
8.4.0.10 IP Replication
HU02555 All High Importance A node may warmstart if the system is configured for remote authorization, but no remote authorization service, such as LDAP, has been configured (show details)
Symptom Single Node Warmstart
Environment Systems running 8.5.x onwards
Trigger Attempted login by a user that is not a locally configured user such as LDAP
Workaround Log in to the node with a known local user (i.e. superuser) then run 'svctask chauthservice -enable no -type ldap' to correct the auth inconsistency
8.4.0.10 LDAP
HU02562 All High Importance A node can warmstart when a 32 Gb Fibre Channel adapter receives an unexpected asynchronous event via internal mailbox commands. This is a transient failure caused during DMA operations (show details)
Symptom Loss of Access to Data
Environment Any system with a 32 Gb Fibre Channel adapter installed and running code level 8.4.0.4 or higher
Trigger A 32 Gb Fibre Channel adapter receives an unexpected asynchronous event via internal mailbox commands
Workaround None
8.4.0.10
HU02571 All High Importance In a Hyperswap cluster, a Tier 2 recovery may occur after manually shutting down both nodes that are in one IO group (show details)
Symptom Loss of access to data
Environment Hyperswap cluster where both nodes are in the same I/O group shutdown
Trigger If both nodes are in the same I/O group, and in a Hyperswap cluster, and have been shut down, then a T2 can occur when all remaining nodes assert
Workaround None
8.4.0.10 Distributed RAID, HyperSwap, RAID
HU02597 All High Importance A single node may warmstart to recover from the situation were different fibres update the completed count for the allocation extent in question (show details)
Symptom Node warmstart
Environment Any platform running either V8.4 or V8.5
Trigger None
Workaround None
8.4.0.10 Data Reduction Pools
IT41447 All High Importance When removing the DNS server configuration, a node may discover unexpected metadata and warmstart (show details)
Symptom Single Node Warmstart
Environment Systems running 8.5 or higher
Trigger Removal of DNS server configuration
Workaround None
8.4.0.10 Reliability Availability Serviceability
IT41835 All High Importance A T2 recovery may occur when a failed drive in the system is replaced with an unsupported drive type (show details)
Symptom Loss of Access to Data
Environment System that have drives reporting as UNSUPPORTED
Trigger A drive with a tech type as UNSUPPORTED will cause this T2 during drive replacement.
Workaround The system should recover automatically. To prevent the issue in the future, make sure system supported drive is used during replacement
8.4.0.10 Drives
SVAPAR-83290 FS5000 High Importance An issue with the Trusted Platform Module (TPM) in FlashSystem 50xx nodes may cause the TPM to become unresponsive. This can happen after a number of weeks of continuous runtime. (show details)
Symptom Single node warmstart
Environment FS50xx platforms running either V8.4.0,V8.4.1,V8.4.2,V8.5.0,V8.5.1,V8.5.2,V8.5.3
Trigger Unresponsive TPM
Workaround Reboot each node in turn. Wait 30 minutes between the two nodes in an I/O group, to allow hosts to failover. Check there are no volumes dependent on the second node before proceeding with the reboot. After all nodes have been rebooted, retry the configuration action, which should now complete successfully.
8.4.0.10
SVAPAR-84305 All High Importance A node may warmstart when attempting to run 'chsnmpserver -community' command without any additional parameter (show details)
Symptom Loss of Access to Data
Environment Any platform running V8.4.0
Trigger None
Workaround Use an additional parameter with the 'chsnmpserver -community' command
8.4.0.10 System Monitoring
SVAPAR-84331 All High Importance A node may warmstart when the 'lsnvmefabric -remotenqn' command is run (show details)
Symptom Single node warmstart
Environment Any system running NVMe
Trigger The warmstart can occur typically when the 'lsnvmefabric -remotenqn' command is run by a script or orchestration layer such as Redhat Openshift or Kubernetes, combined with the IBM CSI driver.
Workaround None
8.4.0.10 NVMe
SVAPAR-85396 FS5000, FS5100, FS5200, FS7200, FS9100, FS9200 High Importance Replacement Samsung NVME drives may show as unsupported, or they may fail during a firmware upgrade as unsupported, due to a VPD read problem (show details)
Symptom Loss of redundancy
Environment Systems running V8.5.3 with industry standard NVMe drives
Trigger Drive firmware update or drive replacements
Workaround Manually power cycling the slot of the failed drive often helps
8.4.0.10 Drives
SVAPAR-85980 All High Importance iSCSI response times may increase on some systems with 25Gb ethernet adapters, after upgrade to 8.4.0.9 or 8.5.x (show details)
Symptom Performance is impacted
Environment Any platform running V8.4 or V8.5
Trigger None
Workaround None
8.4.0.10 Performance, System Update
SVAPAR-86035 All High Importance Whilst completing a request, a DRP pool attempts to allocate additional metadata space, but there is no free space available. This causes the node to warmstart (show details)
Symptom Single node warmstart
Environment Any DRP pool that has run out of metadata space
Trigger Not enough metadata space available
Workaround Add additional space to the DRP pool
8.4.0.10 Data Reduction Pools
HU02367 All Suggested An issue with how RAID handles drive failures may lead to a node warmstart (show details)
Symptom Single Node Warmstart
Environment All systems
Trigger None
Workaround None
8.4.0.10 RAID
HU02372 V5000, FS9100, V7000, V5100, SVC Suggested Host SAS port 4 is missing from the GUI view on some systems. (show details)
Symptom None
Environment Any system that has SAS ports
Trigger None
Workaround Run the command lsportsas to view all ports
8.4.0.10 GUI, SAS
HU02391 All Suggested An issue with how websockets connections are handled can cause the GUI to become unresponsive requiring a restart of the Tomcat server (show details)
Symptom None
Environment All systems
Trigger None
Workaround None
8.4.0.10 Graphical User Interface
HU02443 All Suggested An inefficiency in the RAID code that processes requests to free memory can cause the request to timeout leading to a node warmstart (show details)
Symptom Single Node Warmstart
Environment Systems running v8.2.1 or later
Trigger None
Workaround None
8.4.0.10 RAID
HU02453 All Suggested It may not be possible to connect to GUI or CLI without a restart of the Tomcat server (show details)
Symptom None
Environment Systems running v8.4 or later
Trigger None
Workaround None
8.4.0.10 Command Line Interface, Graphical User Interface
HU02463 All Suggested LDAP user accounts can become locked out because of multiple failed login attempts (show details)
Symptom None
Environment Systems where LDAP accounts use one-time passwords
Trigger None
Workaround Use one or more of the following options: Use CLI instead of GUI. With CLI, after auth cache expires, do not issue any more commands after experiencing a CMMVC7069E error. Log out and log back in with new LDAP password; Max out authentication cache - "chldap -authcacheminutes <min>". Maximum value is 1440min = 24h. Setting it to 600 minutes can minimize probability of hitting the issue; Disable account lock-out on LDAP.
8.4.0.10 LDAP, Graphical User Interface
HU02559 All Suggested A GUI resource issue may cause an out-of-memory condition, leading to the CIMOM and GUI becoming unresponsive, or showing incomplete information (show details)
Symptom None
Environment Systems running v8.3.1,v8.4.0,v8.4.1,v8.4.2,v8.5.0
Trigger None
Workaround Restart the cimserver service or tomcat service
8.4.0.10 GUI, Graphical User Interface
HU02564 All Suggested The 'charraymember' command fails with a degraded DRAID array, even though the syntax of the command is correct (show details)
Symptom None
Environment All
Trigger Performing 'charraymember' command on a degraded DRAID array
Workaround Use the '-immediate' option when running the 'charraymember' command
8.4.0.10 Distributed RAID
HU02593 All Suggested NVMe drive is incorrectly reporting end of life due to flash degradation (show details)
Symptom Alert 1680 and event 090010 logged into the event log, indicating that the drive has reached end of life
Environment V8.2,V8.2.1,V8.3.0,V8.3.1
Trigger None
Workaround None
8.4.0.10 Drives
IT42403 All Suggested A limit is in place to prevent the use of 8TB drives or larger in RAID5 arrays due to the risk of data loss during an extended rebuild. This limit was intended for 8 TiB however it was implemented as 8TB. A 7.3TiB drive has a capacity of 8.02TB and as a result was incorrectly prevented from use in RAID5 (show details)
Symptom None
Environment System running v8.3.0,v8.3.1,v8.4.0,v8.4.1,v8.4.2,v8.5.0,v8.5.1
Trigger None
Workaround Upgrade to 8.5.0.6 or request an ifix
8.4.0.10 Distributed RAID, Drives, RAID
HU02475 All HIPER Power outage can cause reboots on nodes with 25Gb ethernet adapters, necessitating T3 recovery (show details)
Symptom Loss of Access to Data
Environment Any node that has a 25Gb ethernet adapter installed
Trigger Power outage occurs, causing both nodes to experience a kernel panic, meaning cluster information is lost
Workaround T3 recovery will be required
8.4.0.9 Reliability Availability Serviceability
HU02511 All High Importance Code version 8.5.0 includes a change in the driver setting for the 25Gb ethernet adapter. This change can cause port errors, which in turn can cause iSCSI path loss symptoms (show details)
Symptom Loss of Access to Data
Environment Systems running V8.5.0
Trigger None
Workaround None
8.4.0.9 Host Cluster, Hosts, SCSI Unmap, iSCSI
HU02532 All High Importance Nodes that are running 8.4.0.7 or 8.4.0.8, or upgrading to either of these levels may suffer asserts if NVME hosts are configured (show details)
Symptom Single node warmstart
Environment Nodes running either 8.4.0.7 or 8.4.0.8 with NVME hosts
Trigger Any NVMe hosts running on 8.4.0.7 or 8.4.0.8, or when upgrading to either of these code levels
Workaround Apply either 8.4.0.9 or upgrade to the latest 8.5 PTF
8.4.0.9 NVMe, System Update
HU02518 SVC, FS5000, V5000, V7000 Critical Certain hardware platforms running 8.4.0.7 have an issue with the Trusted Platform Module (TPM). This causes issues communicating with encryption keyservers and invalid SSL certificates (show details)
Symptom Loss of Access to Data
Environment Systems running v8.4.0.7 with any of the following hardware are affected: 214x-DH8, 214x-SV1, FS50xx, V50xxE and V7000 gen2/gen2+
Trigger 8.4.0.7
Workaround None
8.4.0.8 Encryption
HU02402 All Critical The remote support feature may use more memory than expected causing a temporary loss of access (show details)
Symptom Loss of Access to Data
Environment Systems running v8.4 or later
Trigger None
Workaround None
8.4.0.7 Support Remote Assist
HU02455 All Critical After converting a system from 3-site to 2-site a timing window issue can trigger a cluster tier 2 recovery (show details)
Symptom Loss of Access to Data
Environment Systems running v8.3 or later in a 3-site topology
Trigger None
Workaround None
8.4.0.7 3-Site
HU02482 All Critical Issue with 25Gb ethernet adapter card firmware can cause the node to warmstart should a specific signal be received from the iSer switch. It is possible for this signal to be propagated to all nodes resulting in a loss of access to data (show details)
Symptom Loss of Access to Data
Environment Systems with 25Gb ethernet adapters
Trigger None
Workaround None
8.4.0.7 iSCSI, Interoperability
IT33912 All Critical A multi-drive code download may fail resulting in a Tier 2 recovery (show details)
Symptom Loss of Access to Data
Environment Systems running 8.3 or later
Trigger None
Workaround None
8.4.0.7 Drives
IT41173 FS5200 Critical If the temperature sensor in an FS5200 system fails in a particular way, it is possible for drives to be powered off, causing a loss of access to data. This type of temperature sensor failure is very rare. (show details)
Symptom Loss of Access to Data
Environment FS5200
Trigger Temperature sensor failure
Workaround None
8.4.0.7 Reliability Availability Serviceability
HU02297 All High Importance Error handling for a failing backend controller can lead to multiple warmstarts (show details)
Symptom Multiple Node Warmstarts
Environment Systems attached to faulty backend controllers
Trigger None
Workaround None
8.4.0.7 Backend Storage
HU02339 All High Importance Multiple node warmstarts can occur if a system has direct Fibre Channel connections to an IBM i host, causing loss of access to data (show details)
Symptom None
Environment Systems running v8.4.0.4 and later, v8.4.2 and later, or v8.5 and later
Trigger Direct attachment to IBM i hosts
Workaround None
8.4.0.7 Hosts, Interoperability
HU02466 All High Importance An issue in the handling of drive failures can result in multiple node warmstarts (show details)
Symptom Multiple Node Warmstarts
Environment Systems running v8.3.0 or later
Trigger None
Workaround None
8.4.0.7 RAID
HU02479 All High Importance If an NVMe host cancels a large number of I/O requests, multiple node warmstarts might occur (show details)
Symptom Multiple Node Warmstarts
Environment Systems running 8.4 or later using NVMeFC
Trigger Cancelled I/O requests from an NVMe host
Workaround None
8.4.0.7 Hosts
HU02497 All High Importance A system with direct Fibre Channel connections to a host, or to another Spectrum Virtualize system, might experience multiple node warmstarts (show details)
Symptom Multiple Node Warmstarts
Environment Systems running v8.4.0.4 and later, v8.4.2 and later, or v8.5 with direct-attached Fibre Channel connections
Trigger Fibre Channel direct-attached hosts
Workaround Connect affected HBAs via a Fibre Channel switch
8.4.0.7 Hosts, Interoperability
HU01209 All Suggested It is possible for the Fibre Channel driver to be offered an unsupported length of data resulting in a node warmstart (show details)
Symptom Single Node Warmstart
Environment Systems using Fibre Channel connectivity
Trigger None
Workaround None
8.4.0.7 Storage Virtualisation
HU02171 All Suggested The timezone for Iceland is set incorrectly (show details)
Symptom None
Environment Systems using the Icelandic timezone
Trigger None
Workaround None
8.4.0.7 Support Data Collection
HU02174 All Suggested A timing window issue related to remote copy memory allocation can result in a node warmstart (show details)
Symptom Single Node Warmstart
Environment Systems using remote copy
Trigger None
Workaround None
8.4.0.7 Global Mirror, Global Mirror with Change Volumes, Metro Mirror
HU02386 FS9200, FS9100, FS7200, FS5100, V7000 Suggested Enclosure fault LED can remain on due to race condition when location LED state is changed (show details)
Symptom None
Environment Systems running v8.1 or later
Trigger None
Workaround None
8.4.0.7 System Monitoring
HU02450 SVC, FS9200, FS9100, FS7200, FS5100, FS5200, V7000, V5100 Suggested A defect in the frame switching functionality of 32Gbps HBA firmware can cause a node warmstart (show details)
Symptom Single Node Warmstart
Environment Systems running v8.3.1 or less using 32Gbps HBAs
Trigger None
Workaround None
8.4.0.7 Hosts
HU02452 SVC, FS9200, FS9100, FS7200, FS5100, FS5200, V7000, V5100 Suggested An issue in NVMe I/O write functionality can cause a single node warmstart (show details)
Symptom Single Node Warmstart
Environment Systems running v8.4 or later using NVMe
Trigger None
Workaround None
8.4.0.7 NVMe
HU02474 All Suggested An SFP failure can cause a node warmstart (show details)
Symptom Single Node Warmstart
Environment Systems running v8.4 or later
Trigger None
Workaround None
8.4.0.7 Reliability Availability Serviceability
IT33996 All Suggested An issue in RAID where unreserved resources fail to be freed up can result in a node warmstart (show details)
Symptom Single Node Warmstart
Environment Systems running v8.3.1 or later
Trigger None
Workaround None
8.4.0.7 RAID
IT40059 FS5200, FS7200, FS9200 Suggested Port to node metrics can appear inflated due to an issue in performance statistics aggregation (show details)
Symptom None
Environment Systems running 8.4 or later using Message Passing
Trigger None
Workaround None
8.4.0.7 Message Passing, System Monitoring
HU02335 All Suggested Cannot properly set the site for a host in a multi-site configuration (hyperswap or stretched) via the GUI (show details)
Symptom Configuration
Environment Systems running code older then 8.4.0.7
Trigger None
Workaround Upgrade to 8.4.0.7 or 8.4.1.0 or higher
8.4.0.7 3-Site, GUI, Graphical User Interface, HyperSwap
HU02296 All HIPER The zero page functionality can become corrupt causing a volume to be initialised with non-zero data (show details)
Symptom Data Integrity Loss
Environment Systems running v8.2.1 or later
Trigger None
Workaround None
8.4.0.6 Storage Virtualisation
HU02226 All Critical Due to an issue in DRP a node can repeatedly warmstart whilst rejoining a cluster (show details)
Symptom Offline Volumes
Environment Systems running v8.3.1 or later using Data Reduction Pools
Trigger None
Workaround None
8.4.0.6 Data Reduction Pools
HU02374 SVC, V5000, V7000 Critical Hosts with Emulex 16Gbps HBAs may become unable to communicate with a system with 8Gbps Fibre Channel ports, after the host HBA is upgraded to firmware version 12.8.364.11. This does not apply to systems with 16Gb or 32Gb Fibre Channel ports (show details)
Symptom Loss of Access to Data
Environment Systems with 8Gbps Fibre Channel ports
Trigger Host Emulex 16Gbps HBA is upgraded to firmware version 12.8.364.11
Workaround Do not upgrade host HBA to firmware version 12.8.364.11
8.4.0.6 Hosts
HU02409 All Critical If rmhost command, with -force, is executed for a MS Windows server then an issue in the iSCSI driver can cause the relevant target initiator to become unresponsive (show details)
Symptom Loss of Access to Data
Environment Systems with iSCSI connected hosts
Trigger Run rmhost -force CLI command for an iSCSI connected MS Windows host
Workaround Do not use -force when removing a host object
8.4.0.6 Hosts, iSCSI
HU02410 SVC Critical A timing window issue in the transition to a spare node can cause a cluster-wide Tier 2 recovery (show details)
Symptom Loss of Access to Data
Environment SVC systems running v8.1 or later with Hot Spare Nodes
Trigger None
Workaround None
8.4.0.6 Hot Spare Node
HU02423 All Critical Volume copies may be taken offline even though there is sufficient free capacity (show details)
Symptom Offline Volumes
Environment Systems using Data Reduction Pools
Trigger None
Workaround None
8.4.0.6 Data Reduction Pools
HU02428 All Critical Issuing a movevdisk CLI command immediately after removing an associated GMCV relationship can trigger a Tier 2 recovery (show details)
Symptom Loss of Access to Data
Environment Systems running v8.4 or later using GMCV
Trigger None
Workaround None
8.4.0.6 Command Line Interface, Global Mirror with Change Volumes
HU02434 All Critical An issue in the internal accounting of FlashCopy resources can lead to multiple node warmstarts taking a cluster offline (show details)
Symptom Loss of Access to Data
Environment Systems running v8.4 or later using large (>5TB) volumes
Trigger None
Workaround None
8.4.0.6 FlashCopy
HU02440 All Critical Using the migrateexts command when both source and target mdisks are unmanaged can trigger a Tier 2 recovery (show details)
Symptom Loss of Access to Data
Environment Systems running v8.4 or later
Trigger Issue the migrateexts command when both source and target mdisks are unmanaged
Workaround Ensure that the mdisks are managed before migrating extents
8.4.0.6 Command Line Interface, Storage Virtualisation
HU02442 All Critical Issuing a lspotentialarraysize CLI command with an invalid drive class can trigger a Tier 2 recovery (show details)
Symptom Loss of Access to Data
Environment Systems running v8.4 or later
Trigger Enter lspotentialarraysize CLI command with an invalid drive class parameter
Workaround Before entering a lspotentialarraysize CLI command ensure that the drive class parameter is valid
8.4.0.6 Command Line Interface
HU02343 All High Importance For Huawei Dorado V3 Series backend controllers it is possible that not all available target ports will be utilized. This would reduce the potential IO throughput and can cause high read/write backend queue time on the cluster impacting front end latency for hosts (show details)
Symptom Performance
Environment Systems using Huawei Dorado V3 Series backend controllers
Trigger None
Workaround None
8.4.0.6 Backend Storage
HU02438 All High Importance Certain conditions can provoke a cache behaviour that unbalances workload distribution across CPU cores leading to performance impact (show details)
Symptom Performance
Environment Systems running v8.3.1 or later
Trigger None
Workaround None
8.4.0.6 Cache
IT38015 All High Importance During RAID rebuild or copyback on systems with 16gb or less of memory, cache handling can lead to a deadlock which results in timeouts (show details)
Symptom Performance
Environment Systems with 16GB or less of memory
Trigger None
Workaround None
8.4.0.6 RAID
HU02263 All Suggested The pool properties dialog in the GUI displays thin-provisioning savings, compression savings and total savings. In Data Reduction Pools, the thin-provisioning savings displayed are actually the total savings instead of the thin-provisioning savings only (show details)
Symptom None
Environment Systems running v8.2.1 or later using Data Reduction Pools
Trigger None
Workaround Subtract compression savings from thin-provisioning savings to get the actual number
8.4.0.6 Data Reduction Pools
HU02382 FS9200, FS9100, FS7200, FS5100, V7000, V5100 Suggested A complex interaction of tasks, including drive firmware cleanup and syslog reconfiguration, can cause a 10 second delay when each node unpends (eg during an upgrade) (show details)
Symptom None
Environment Systems running v8.2.1 or later that have a remote syslog server configured
Trigger None
Workaround None
8.4.0.6 System Update
HU02383 FS9200, FS9100, FS7200, FS5100, V7000 Suggested An additional 20 second IO delay can occur when a system update commits (show details)
Symptom None
Environment Systems running v8.2.1 or later that have a remote syslog server configured
Trigger None
Workaround Remove remote syslog servers from the configuration to reduce the additional delay to 10 seconds. It is not possible to completely eliminate the delay using a workaround
8.4.0.6 System Update
HU02444 All Suggested Some security scanners can report unauthenticated targets against all the iSCSI IP addresses of a node (show details)
Symptom None
Environment Systems running v8.4.0 or later
Trigger None
Workaround None
8.4.0.6 Hosts, iSCSI
HU02418 All HIPER During a DRAID array rebuild data can be written to an incorrect location. For more details refer to this Flash (show details)
Symptom Data Integrity Loss
Environment Systems running v8.3.1 or later
Trigger None
Workaround None
8.4.0.5 Distributed RAID, RAID
HU02384 SVC HIPER An inter-node message queue can become stalled, leading to an I/O timeout warmstart, and temporary loss of access (show details)
Symptom Offline Volumes
Environment SVC systems using SV1 model nodes running v8.3.1 or later
Trigger None
Workaround None
8.4.0.4 Reliability Availability Serviceability
HU02400 All HIPER A problem in the virtualization component of the system can cause a migration IO to be submitted in an incorrect context resulting in a node warmstart. In some cases it is possible that this IO has been submitted to an incorrect location on the backend, which can cause data corruption of an isolated small area (show details)
Symptom Data Integrity Loss
Environment Systems running v8.2.1 or later
Trigger None
Workaround None
8.4.0.4 Storage Virtualisation
DT112601 All Critical Deleting image mode mounted source volume while migration is ongoing could trigger Tier 2 recovery (show details)
Symptom Loss of Access to Data
Environment Systems running v8.3.1 or later
Trigger Delete the source volume when migration progress is showing 0%
Workaround Wait for lsmigrate progress to report a non-zero progress value before issuing a volume delete
8.4.0.4 Storage Virtualization
HU02342 All Critical Occasionally when an offline drive returns to online state later than its peers in the same RAID array there can be multiple node warmstarts that send nodes into a service state (show details)
Symptom Loss of Access to Data
Environment Systems running v7.8 or later
Trigger None
Workaround None
8.4.0.4 RAID
HU02393 All Critical Automatic resize of compressed/thin volumes may fail causing warmstarts on both nodes in an I/O group (show details)
Symptom Loss of Access to Data
Environment Systems running v8.1 or later
Trigger None
Workaround None
8.4.0.4 Storage Virtualisation
HU02397 All Critical A Data Reduction Pool, with deduplication enabled, can retain some stale state after deletion and recreation. This has no immediate effect. However if later on a node goes offline this condition can cause the pool to be taken offline (show details)
Symptom Loss of Access to Data
Environment Systems running v8.1.3 or later using Data Reduction Pools
Trigger None
Workaround None
8.4.0.4 Data Reduction Pools
HU02401 All Critical EasyTier can move extents between identical mdisks until one runs out of space (show details)
Symptom Loss of Access to Data
Environment Systems running v8.1 or later using EasyTier
Trigger None
Workaround Disable EasyTier. Manually migrate extents between mdisks
8.4.0.4 EasyTier
HU02406 All Critical An interoperability issue between Cisco NX-OS firmware and the Spectrum Virtualize Fibre Channel driver can cause a node warmstart on NPIV failback (for example during an upgrade) with the potential for a loss of access. For more details refer to this Flash (show details)
Symptom Loss of Access to Data
Environment Systems using NPIV that are connected to Cisco SAN equipment running NX-OS 8.4(2c) or later
Trigger Initiate an NPIV failback operation by, for example, performing an upgrade
Workaround Disable NPIV (which will require any hot spare nodes to be removed first)
8.4.0.4 Interoperability
HU02414 All Critical Under specific sequence and timing of circumstances the garbage collection process can timeout and take a pool offline temporarily (show details)
Symptom Loss of Access to Data
Environment Systems running v8.3 or later using Data Reduction Pools
Trigger None
Workaround None
8.4.0.4 Data Reduction Pools
HU02345 All High Importance When connectivity to nodes in a local or remote cluster is lost, inflight IO can become stuck in an aborting state, consuming system resources and potentially adversely impacting performance (show details)
Symptom Performance
Environment Systems using remote copy
Trigger None
Workaround None
8.4.0.4 HyperSwap, Metro Mirror
HU02388 V5000, FS5000 High Importance GUI can hang randomly due to an out of memory issue after running any task (show details)
Symptom Degraded Operation
Environment Storwize V5000E and FlashSystem 5000 systems running v8.2.1 or later
Trigger None
Workaround None
8.4.0.4 Graphical User Interface
HU02422 All High Importance GUI performance can be degraded when displaying large numbers of volumes or other objects (show details)
Symptom Degraded Operation
Environment Systems running v8.3.1 or later
Trigger None
Workaround None
8.4.0.4 Graphical User Interface
HU02306 All Suggested An offline host port can still be shown as active in lsfabric and the associated host can be shown as online despite being offline (show details)
Symptom None
Environment Systems running v8.3 or later
Trigger None
Workaround None
8.4.0.4 Hosts
HU02405 FS5200 Suggested An issue in the zero detection of the new Message Passing (MP) functionality can cause thin volumes to allocate space when writing zeros (show details)
Symptom None
Environment FlashSystem 5200 systems
Trigger None
Workaround When writing zeros from a host, always submit IO to the preferred node
8.4.0.4 Message Passing
HU02426 All Suggested Where an email server accepts the STARTTLS command during the initial handshake if TLS v1.2 is disables or not supported then the system will be unable to send email alerts (show details)
Symptom None
Environment Systems running v8.4 or later connecting to a mail server that does not support/enable TLS v1.2
Trigger TLS v1.2 not supported or enabled on mail server
Workaround Enable TLS v1.2 on mail server if available
8.4.0.4 System Monitoring
IT37654 All Suggested When creating a new encrypted array the CMMVC8534E error (Node has insufficient entropy to generate key material) can appear preventing array creation (show details)
Symptom Configuration
Environment Systems using encryption
Trigger None
Workaround Power cycle the affected node
8.4.0.4 Encryption
HU02312 All HIPER Changing the preferred node for a volume when it is in a remote copy relationship can result in multiple node warmstarts. For more details refer to this Flash (show details)
Symptom Multiple Node Warmstarts
Environment Systems running v8.4 or later using remote copy
Trigger Change preferred node for a volume in a remote copy relationship
Workaround Remove any associated remote copy relationship before changing the preferred node of a volume
8.4.0.3 Global Mirror, Global Mirror with Change Volumes, Metro Mirror
HU02340 All HIPER High replication workloads can cause multiple warmstarts with a loss of access at the partner cluster (show details)
Symptom Loss of Access to Data
Environment Systems running v8.3.1 or later using IP Replication
Trigger None
Workaround None
8.4.0.3 IP Replication
HU02373 All Critical An incorrect compression flag in metadata can take a DRP offline (show details)
Symptom Loss of Access to Data
Environment Systems running v8.3.1 or later using Data Reduction Pools and Remote Copy
Trigger None
Workaround None
8.4.0.3 Data Reduction Pools
HU02319 All High Importance The GUI can become unresponsive (show details)
Symptom None
Environment Systems running v8.4.0.1, or later, using remote copy without 3-site replication
Trigger None
Workaround None
8.4.0.3 Graphical User Interface
HU02326 SVC High Importance Delays in passing messages between nodes in an I/O group can adversely impact write performance (show details)
Symptom Performance
Environment SVC systems running v8.2.1 or later
Trigger None
Workaround None
8.4.0.3 Performance
HU02360 All High Importance Cloud Callhome may stop working and provide no indication of this in the event log. For more details refer to this Flash (show details)
Symptom Degraded Operation
Environment Systems running v8.3 or later
Trigger None
Workaround None
8.4.0.3 System Monitoring
HU02362 SVC, FS9200, FS9100, FS7200, FS5100, FS5200, V7000, V5100 High Importance When the RAID scrub process encounters bad grains, the peak response time for reads and writes can be adversely impacted (show details)
Symptom Performance
Environment Systems running v8.2 or later using NVMe drives
Trigger None
Workaround None
8.4.0.3 RAID
HU02376 All High Importance FlashCopy maps may get stuck at 99% due to inconsistent metadata accounting between nodes (show details)
Symptom Degraded Operation
Environment Systems running v8.2.1 or later using FlashCopy
Trigger None
Workaround None
8.4.0.3 FlashCopy
HU02392 All High Importance Validation in the Upload Support Package feature will reject new case number formats in the PMR field (show details)
Symptom None
Environment All systems
Trigger None
Workaround At v8.4.0.0 or later use the CLI command satask supportupload -pmr pmr_number -filename fullpath/filename
8.4.0.3 Support Data Collection
HU02325 All Suggested Tier 2 and Tier 3 recoveries can fail due to node warmstarts (show details)
Symptom None
Environment Systems running v8.4 with remote (LDAP) users that were created at v8.3.1 or earlier
Trigger None
Workaround Recreate all remote users after upgrading to v8.4
8.4.0.3 Reliability Availability Serviceability
HU02331 All Suggested Due to a threshold issue an error code 3400 may appear too often in the event log (show details)
Symptom None
Environment Systems using compression cards
Trigger None
Workaround None
8.4.0.3 Compression
HU02332 & HU02336 All Suggested When an I/O is received, from a host, with invalid or inconsistent SCSI data but a good checksum it may cause a node warmstart (show details)
Symptom Single Node Warmstart
Environment Systems running v7.8 or later
Trigger None
Workaround None
8.4.0.3 Hosts
HU02366 All Suggested Slow internal resource reclamation by the RAID component can cause a node warmstart (show details)
Symptom Single Node Warmstart
Environment Systems running v8.2.1 or later
Trigger None
Workaround None
8.4.0.3 RAID
HU02375 All Suggested An issue in how the GUI handles volume data can adversely impact its responsiveness (show details)
Symptom Performance
Environment Systems running v8.3.1 or later with large numbers of volumes
Trigger None
Workaround None
8.4.0.3 Graphical User Interface
HU02381 All Suggested When the proxy server password is changed to one with more than 40 characters the config node will warmstart (show details)
Symptom Single Node Warmstart
Environment Systems using the system-wide web proxy server
Trigger Use chproxy CLI command to change password to one with >40 characters
Workaround Use a proxy password of <40 characters
8.4.0.3 Command Line Interface
HU02387 All Suggested When using the GUI the maximum Data Reduction Pools limitation incorrectly includes child pools (show details)
Symptom None
Environment Systems running v8.4 or later using Data Reduction Pools
Trigger Add a new Data Reduction Pool when 4 x parent and child pools already exist
Workaround Use CLI to create the new Data Reduction Pool
8.4.0.3 Data Reduction Pools
HU02425 All Suggested An issue in the handling of internal messages, when the system has a high IO workload to two or more different FlashCopy maps in the same dependency chain, can result in incorrect counters. The node will warmstart to clear this condition. (show details)
Symptom Single Node Warmstart
Environment Systems using FlashCopy
Trigger None
Workaround None
8.4.0.3 FlashCopy
HU02261 All HIPER A Data Reduction Pool may be taken offline when metadata is detected to hold an invalid compression flag. For more details refer to this Flash (show details)
Symptom Loss of Access to Data
Environment Systems running v8.1 or later using Data Reduction Pools
Trigger None
Workaround None
8.4.0.2 Data Reduction Pools
HU02277 All HIPER RAID parity scrubbing can become stalled causing an accumulation of media errors leading to multiple drive failures with the possibility of data integrity loss. For more details refer to this Flash (show details)
Symptom Data Integrity Loss
Environment Systems with model MZILS3T8HMLH read intensive SSDs at drive firmware MS24 are particularly susceptible to the data integrity (DI) issue. Other drive types may see multiple failures without DI issue
Trigger None
Workaround None
8.4.0.2 RAID
HU02310 All HIPER Where a FlashCopy mapping exists between two volumes in the same Data Reduction Pool and the same I/O group, and the target volume has deduplication enabled, then the target may contain invalid data (show details)
Symptom Data Integrity Loss
Environment Systems running v8.4 or later using Data Reduction Pools with FlashCopy or GMCV
Trigger None
Workaround None
8.4.0.2 Data Reduction Pools, FlashCopy, Global Mirror with Change Volumes
HU02313 FS9200, FS9100, FS7200, FS5100, V7000, V5100 HIPER When a FlashCore Module (FCM) fails there is a chance that this can trigger other FCMs in the same control enclosure to also fail. If enough additional drives fail, at the same time, this can take the array offline and cause a loss of access to data. For more details refer to this Flash (show details)
Symptom Loss of Access to Data
Environment Systems running v8.2 or later using Flash Core Modules
Trigger None
Workaround None
8.4.0.2 Drives
HU02338 All HIPER An issue in the setting up of reverse FlashCopy mappings can cause the background copy to finish prematurely providing an incomplete target image (show details)
Symptom Data Integrity Loss
Environment Systems using FlashCopy
Trigger None
Workaround None
8.4.0.2 FlashCopy
HU02282 All Critical After a code upgrade the config node may exhibit high write response times. In exceptionally rare circumstances an Mdisk group may be taken offline (show details)
Symptom Loss of Access to Data
Environment Systems running v8.2.1 or later
Trigger None
Workaround None
8.4.0.2 Cache
HU02315 All Critical Failover for VMware iSER hosts may pause I/O for more than 120 seconds (show details)
Symptom Loss of Access to Data
Environment Systems presenting volumes to VMware iSER hosts
Trigger None
Workaround None
8.4.0.2 Hosts
HU02321 All Critical Where nodes relying on RDMA clustering alone, if a node is removed, warmstarts or goes down for upgrade there may be a delay in internode communication leading to lease expiries (show details)
Symptom Loss of Access to Data
Environment Systems using iSER RDMA clustering with iSCSI hosts
Trigger None
Workaround When upgrading, add an alternative, non-iSER medium for node to node communications
8.4.0.2 iSCSI
HU02429 All Critical System can go offline shortly after changing the SMTP settings using the chemailserver command via the GUI (show details)
Symptom Loss of Access to Data
Environment Systems running v7.8.1 or later
Trigger None
Workaround None
8.4.0.2 System Monitoring
HU02201 & HU02221 All High Importance Shortly after upgrading drive firmware, specific drive models can fail due to Too many long IOs to drive for too long errors (show details)
Symptom Degraded Operation
Environment Systems with the following drive models:
  • ST300MM0009 (300GB) - B5B8
  • ST600MM0009 (600GB) - B5B8
  • ST900MM0009 (900GB) - B5B8
  • ST1200MM0009 (1200GB) - B5B8
  • ST1200MM0129 (1800GB) - B5C9
  • ST2400MM0129 (2400GB) - B5C9
  • ST300MP0006 (300GB) - B6AA
  • ST600MP0006 (600GB) - B6AA
  • ST900MP0146 (900GB) - B6CB
Trigger None
Workaround None
8.4.0.2 Drives
HU02227 FS7200, FS9100, FS9200, SVC, V5100, V7000 High Importance Certain I/O patterns can cause compression hardware to post errors. When those errors exceed a threshold the node can be taken offline (show details)
Symptom Loss of Redundancy
Environment Systems running v8.2 or later using compressed volumes
Trigger None
Workaround None
8.4.0.2 Compression
HU02300 All High Importance Use of Enhanced Callhome in censored mode may lead to adverse performance around 02:00 (2AM) (show details)
Symptom Performance
Environment Systems with Enhanced Callhome enabled in censored mode
Trigger None
Workaround None
8.4.0.2 System Monitoring
HU02301 SVC High Importance iSCSI hosts connected to iWARP 25G adapters may experience adverse performance impacts (show details)
Symptom Performance
Environment SVC model SV1 systems with iWARP 25G adapters
Trigger None
Workaround None
8.4.0.2 iSCSI
HU02304 FS9100, V7000, V5100 High Importance Some RAID operations for certain NVMe drives may cause adverse I/O performance (show details)
Symptom Performance
Environment Systems running v8.4 or later using NVMe drives
Trigger None
Workaround None
8.4.0.2 RAID
HU02311 All High Importance An issue in volume copy flushing may lead to higher than expected write cache delays (show details)
Symptom Performance
Environment Systems running v8.3.1 or later
Trigger None
Workaround None
8.4.0.2 Cache
HU02317 All High Importance A DRAID expansion can stall shortly after it is initiated (show details)
Symptom Degraded Operation
Environment Systems running v8.3.1 or later using DRAID
Trigger None
Workaround None
8.4.0.2 Distributed RAID
HU02095 All Suggested The effective_used_capacity field of lsarray/lsmdisk commands should be empty for RAID arrays which do not contain overprovisioned drives. However, sometimes this field can be zero even though it should be empty. This can cause incorrect provisioned capacity reporting in the GUI (show details)
Symptom None
Environment Systems with non-FCM arrays
Trigger None
Workaround None
8.4.0.2 Graphical User Interface
HU02280 All Suggested Spectrum Control or Storage Insights may be unable to collect stats after a Tier 2 recovery or system powerdown (show details)
Symptom None
Environment Systems running v8.3.1.2 or later
Trigger None
Workaround None
8.4.0.2 System Monitoring
HU02291 All Suggested Internal counters for upper cache stage/destage I/O rates and latencies are not collected and zeroes are usually displayed (show details)
Symptom None
Environment Systems running v8.4 or later
Trigger None
Workaround None
8.4.0.2 Cache, System Monitoring
HU02292 & HU02308 All Suggested The use of maximum replication delay within Global Mirror may occasionally cause a node warmstart (show details)
Symptom Single Node Warmstart
Environment Systems using Remote Copy
Trigger None
Workaround None
8.4.0.2 Global Mirror
HU02303 & HU02305 All Suggested Configuration node warmstart will occur if mkhostcluster is run with -ignoreseedvolume and the ignored volumes have an id greater than 256 (show details)
Symptom Single Node Warmstart
Environment Systems running v8.3, or later, where ignored volumes have an id greater than 256
Trigger Run mkhostcluster with its -ignoreseedvolume option
Workaround Do not use the -ignoreseedvolume option with the mkhostcluster command
8.4.0.2 Hosts
HU02419 All Suggested During creation of a drive FRU id the resulting unique number can contain a space character which can lead to CLI commands, that return this value, presenting it as a truncated string (show details)
Symptom None
Environment Systems running v8.3.1 or later
Trigger None
Workaround None
8.4.0.2 Command Line Interface, Drives
IT34949 All Suggested lsnodevpd may show DIMM information in the wrong positions (show details)
Symptom None
Environment Systems running v8.4 or later
Trigger None
Workaround None
8.4.0.2 Command Line Interface, Graphical User Interface
HU02186 FS5100, FS7200, FS9100, FS9200, V5100, V7000 HIPER NVMe drive pulls or firmware upgrades may lead to offline pools with the possibility of a small loss of data integrity. For more details refer to this Flash (show details)
Symptom Data Integrity Loss
Environment Systems running v8.2.1, or later with NVMe drives
Trigger None
Workaround None
8.4.0.0 RAID
HU02327 All HIPER Using addvdiskcopy in conjunction with expandvdisk with format may result in the original being overwritten, by the new copy, producing blank copies. For more details refer to this Flash (show details)
Symptom Data Integrity Loss
Environment Systems running v8.2.1 or later
Trigger Using addvdiskcopy and expandvdisk with format
Workaround Wait until the format is completed before adding a copy
8.4.0.0 Volume Mirroring
HU02058 All Critical Changing a remote copy relationship from GMCV to MM or GM can result in a Tier 2 recovery (show details)
Symptom Loss of Access to Data
Environment Systems using remote copy
Trigger None
Workaround None
8.4.0.0 Global Mirror, Global Mirror with Change Volumes, Metro Mirror
HU02092 All Critical The effectiveness of slow drain mitigation can become reduced causing fabric congestion to adversely impact all ports on an adapter (show details)
Symptom Loss of Access to Data
Environment Systems using 16Gb HBAs
Trigger None
Workaround None
8.4.0.0 Reliability Availability Serviceability
HU02172 All Critical The CLI command lsdependentvdisks -enclosure X causes node warmstarts if no nodes are online in that enclosure (show details)
Symptom Loss of Access to Data
Environment Systems where all nodes in an enclosure are offline
Trigger Run lsdependentvdisks for an enclosure where all nodes are offline
Workaround Do not run lsdependentvdisks for an enclosure where all nodes are offline
8.4.0.0 Command Line Interface
HU02184 All Critical When a 3PAR controller experiences a fault that prevents normal I/O processing it may issue a SCSI TARGET RESET command. This command is not supported and may cause multiple node asserts, possibly cluster-wide (show details)
Symptom Loss of Access to Data
Environment Systems with 3PAR backend controllers
Trigger None
Workaround None
8.4.0.0 Backend Storage
HU02196 & HU02253 All Critical A particular sequence of internode messaging delays can lead to a cluster wide lease expiry (show details)
Symptom Loss of Access to Data
Environment Systems running v7.8 or later
Trigger None
Workaround None
8.4.0.0 Reliability Availability Serviceability
HU02210 All Critical There is a very small timing window where a volume may be reported as offline, to a host, during its conversion from a regular volume to a HyperSwap volume (show details)
Symptom Offline Volumes
Environment Systems moving to a HyperSwap topology that have used Remote Copy
Trigger None
Workaround Individual volume copies of a HyperSwap volume can be converted between fully allocated, thin and compressed using the normal capacity savings conversion method by using addvdiskcopy. Then once the sync has completed the vdisk copy that is no longer needed can be deleted. This method avoids the need for removing a HyperSwap volume copy and hence the need to add a HyperSwap volume copy. NOTE: Please pay particular attention to the distinction between a vdisk copy and a volume copy.
8.4.0.0 HyperSwap
HU02213 SVC Critical A Hot Spare Node (HSN) timing window issue can, during an HSN activation or deactivation, cause the cluster to broadcast an invalid VPD update to other clusters on the SAN. This may trigger a Tier 2 recovery on the other cluster. For more details refer to this Flash (show details)
Symptom Loss of Access to Data
Environment SVC systems, with Hot Spare Nodes, using remote copy partnerships
Trigger None
Workaround Prior to an upgrade, or node hardware maintenance, remove the HSN
8.4.0.0 Hot Spare Node
HU02225 All Critical An issue in the Thin Provisioning feature can lead to multiple warmstarts with the possibility of a loss of access to data (show details)
Symptom Loss of Access to Data
Environment Systems running v8.3 or later
Trigger None
Workaround None
8.4.0.0 Thin Provisioning
HU02230 FS7200, FS9100, FS9200, V7000 Critical For IBM Flash Core Modules a change of state, from unused to candidate, can lead to a Tier 2 recovery (show details)
Symptom Loss of Access to Data
Environment Systems using IBM Flash Core Modules
Trigger None
Workaround None
8.4.0.0 Drives
HU02232 All Critical Forced removal of large volumes in FlashCopy mappings can cause multiple node warmstarts with the possibility of a loss of access (show details)
Symptom Loss of Access to Data
Environment Systems using FlashCopy
Trigger None
Workaround None
8.4.0.0 FlashCopy
HU02262 SVC, V7000, V5000 Critical Entering the CLI applydrivesoftware -cancel command may result in cluster-wide warmstarts (show details)
Symptom Loss of Access to Data
Environment Systems running v8.3 or later
Trigger None
Workaround None
8.4.0.0 Drives
HU02266 All Critical An issue in auto-expand can cause expansion to fail and the volume to be taken offline (show details)
Symptom Offline Volumes
Environment Systems running v8.2.1 or later using thin-provisioning
Trigger None
Workaround None
8.4.0.0 Thin Provisioning
HU02289 SVC, FS9200 Critical An issue with internal resource allocation in high-end systems, with 1000s of mirror copies, may cause multiple warmstarts with the possibility of a loss of access (show details)
Symptom Loss of Access to Data
Environment Systems with a high number of cores (e.g. SVC SV2 models & FS9200)
Trigger None
Workaround None
8.4.0.0 Volume Mirroring
HU02298 All Critical A high frequency of 1920 events and restarting of consistency groups may provoke a Tier 2 recovery (show details)
Symptom Loss of Access to Data
Environment Systems using remote copy
Trigger None
Workaround None
8.4.0.0 Global Mirror
HU02299 FS7200, FS9100, FS9200, V7000 Critical NVMe drives can become locked due to a missing encryption key condition (show details)
Symptom Loss of Access to Data
Environment Systems using NVMe drives
Trigger None
Workaround None
8.4.0.0 Drives
HU02314 FS9200, FS9100, FS7200, FS5100, V7000, V5100 Critical Due to a RAID issue when a bad block is detected on a NVMe drive there may be multiple node warmstarts with a possibility of a loss of access to data (show details)
Symptom Loss of Access to Data
Environment Systems running v8.2 or later using NVMe drives
Trigger None
Workaround None
8.4.0.0 Drives
HU02322 All Critical A deadlock condition in the Data Reduction Pool function may cause multiple node warmstarts and a temporary loss of access to data (show details)
Symptom Loss of Access to Data
Environment Systems running v8.3 or later using Data Reduction Pools
Trigger None
Workaround None
8.4.0.0 Data Reduction Pools
HU02323 All Critical Stalled I/O during DRAID expansion can cause node warmstarts and a temporary loss of access to data (show details)
Symptom Loss of Access to Data
Environment Systems running v8.3.1 or later using DRAID
Trigger None
Workaround None
8.4.0.0 Distributed RAID
HU02390 All Critical A memory handling issue in the REST API may cause an out-of-memory condition when listing a large number of volumes (show details)
Symptom None
Environment Systems running v8.1.3 or later
Trigger Use REST API to issue an lsvdisk when there are >10K volumes on the system or issue multiple lsvdisk commands concurrently for >2000 volumes
Workaround None
8.4.0.0 REST API
HU02467 All Critical When one node disappears from the cluster the surviving node can be unable to achieve quorum allegiance in a timely manner causing it to lease expire (show details)
Symptom Loss of Access to Data
Environment All systems
Trigger None
Workaround None
8.4.0.0 Quorum
HU02153 All High Importance Fabric or host issues can cause aborted IOs to block the port throttle queue leading to adverse performance that is cleared by a node warmstart (show details)
Symptom Performance
Environment Systems running v8.2.1, or later, with 8Gb FC adapters
Trigger None
Workaround None
8.4.0.0 Hosts
HU02156 All High Importance Global Mirror environments may experience more frequent 1920 events due to writedone message queuing (show details)
Symptom Performance
Environment Systems using Global Mirror
Trigger None
Workaround None
8.4.0.0 Global Mirror
HU02164 All High Importance An issue in Remote Copy may cause a loss of hardened data when a node is warmstarted (show details)
Symptom Loss of Redundancy
Environment Systems using remote copy
Trigger None
Workaround None
8.4.0.0 Global Mirror, Global Mirror with Change Volumes, Metro Mirror
HU02170 V7000, FS7200, FS9100, FS9200 High Importance During NVMe SSD firmware upgrade processes peak read latency may reach 10sec (show details)
Symptom Performance
Environment Systems with NVMe drives
Trigger None
Workaround None
8.4.0.0 RAID
HU02194 All High Importance Password reset via USB drive does not work as expected and user is not able to login to Management or Service assistant GUI with the new password (show details)
Symptom Degraded Operation
Environment Systems running v8.2 or later
Trigger None
Workaround None
8.4.0.0 Reliability Availability Serviceability
HU02250 All High Importance Duplicate volume names may cause multiple asserts (show details)
Symptom Multiple Node Warmstarts
Environment Systems running v7.8 or later
Trigger Create more than one volume with the same name
Workaround Use care when naming volumes
8.4.0.0 Storage Virtualisation
II14767 SVC High Importance An issue with how cache handles ownership of volumes across multiple sites can lead to cross-site destage, adversely impacting write latency. For more details refer to this Flash (show details)
Symptom Performance
Environment SVC systems in a Stretched Cluster configuration
Trigger None
Workaround None
8.4.0.0 Cache
IT33734 V5000 High Importance Lower cache partitions may fill up even though higher destage rates are available (show details)
Symptom Performance
Environment Systems running v8.2.1 or later
Trigger None
Workaround None
8.4.0.0 Cache
IT33868 SVC, V7000, FS9100 High Importance Non-FCM NVMe drives may exhibit high write response times with the Spectrum Protect Blueprint script (show details)
Symptom Performance
Environment Systems with non-FCM NVMe drives using the Spectrum Protect Blueprint script
Trigger None
Workaround Use Blueprint script v4.2b or later
8.4.0.0 Drives
IT36619 All High Importance After a node warmstart, system CPU utilisation may show an increase (show details)
Symptom Performance
Environment Systems running v8.3.1 or later
Trigger None
Workaround None
8.4.0.0 RAID
HU01238 All Suggested The mishandling of performance stats may occasionally result in some entries being overwritten (show details)
Symptom None
Environment All systems
Trigger None
Workaround None
8.4.0.0 System Monitoring
HU01977 All Suggested CLI commands can produce a return code of 1 even though execution was successful (show details)
Symptom None
Environment Systems running v8.1 or later
Trigger None
Workaround None
8.4.0.0 Command Line Interface
HU02139 FS9100, FS5100, V7000, V5100 Suggested When 32Gbps FC adapters are fitted the maximum supported ambient temperature is decreased leading to more threshold exceeded errors in the Event Log (show details)
Symptom None
Environment Systems running v8.2.1 or later with 32Gbps HBAs
Trigger Ambient air temperature above 30C
Workaround Check ambient temperature reported by lsenclosure. If it is 30C or above site the machine in a cooler location
8.4.0.0 System Monitoring
HU02142 All Suggested It is possible for a backend unmap process to become stalled, preventing system configuration changes from completing (show details)
Symptom Configuration
Environment Systems running v8.1.0, or later, using DRAID
Trigger None
Workaround None
8.4.0.0 Distributed RAID
HU02208 All Suggested An issue with the handling of files by quorum can lead to a node warmstart (show details)
Symptom Single Node Warmstart
Environment Systems running v8.2.1 or later
Trigger None
Workaround None
8.4.0.0 Quorum
HU02239 All Suggested A rare race condition in the Xcopy function can cause a single node warmstart (show details)
Symptom Single Node Warmstart
Environment Systems running v7.8 or later
Trigger None
Workaround None
8.4.0.0 Hosts
HU02241 All Suggested IP Replication can fail to create IP partnerships via the secondary cluster management IP (show details)
Symptom None
Environment All systems
Trigger None
Workaround Use primary management IP to run mkippartnership commands
8.4.0.0 IP Replication
HU02245 All Suggested First support data collection fails to upload successfully (show details)
Symptom None
Environment New systems
Trigger None
Workaround On new systems run a satask snap without the "-upload" option
8.4.0.0 Support Data Collection
HU02251 All Suggested A warmstart may occur when a node receives iSCSI host login/logout requests out of sequence (show details)
Symptom Single Node Warmstart
Environment Systems running 8.3.1 or later with iSCSI connected hosts
Trigger None
Workaround None
8.4.0.0 Hosts, iSCSI
HU02255 All Suggested A timing issue in the processing of login requests can cause a single node warmstart (show details)
Symptom Single Node Warmstart
Environment Systems running v8.3.1 or later
Trigger None
Workaround None
8.4.0.0 Command Line Interface, Graphical User Interface
HU02265 All Suggested Enhanced inventory can sometimes be missing from callhome data due to the lsfabric command timing out (show details)
Symptom None
Environment Systems running v8.2.1 or later
Trigger None
Workaround None
8.4.0.0 Support Data Collection
HU02267 All Suggested After upgrade it is possible for a node IP address to become duplicated with the cluster IP address and access to the config node to be lost as a consequence (show details)
Symptom Configuration
Environment Systems running v8.3 or later
Trigger None
Workaround None
8.4.0.0 Graphical User Interface, Command Line Interface
HU02334 All Suggested Node to node connectivity issues may trigger repeated logins/logouts resulting in a single node warmstart (show details)
Symptom Single Node Warmstart
Environment Systems running v8.2.1 or later
Trigger None
Workaround None
8.4.0.0 Reliability Availability Serviceability
HU02353 All Suggested The GUI will refuse to start a GMCV relationship if one of the change volumes has an ID of 0 (show details)
Symptom None
Environment Systems using GMCV
Trigger Using a volume with id 0 in a GMCV relationship
Workaround Use the CLI or a volume with a different id
8.4.0.0 Graphical User Interface, Global Mirror with Change Volumes
HU02358 All Suggested An issue in Remote Copy, that stalls a switch of direction, can cause I/O timeouts leading to a node warmstart (show details)
Symptom Single Node Warmstart
Environment Systems using Remote Copy
Trigger None
Workaround None
8.4.0.0 Global Mirror, Global Mirror with Change Volumes, Metro Mirror
HU02364 All Suggested False 989001 Managed Disk Group space warnings can be generated (show details)
Symptom None
Environment Systems running v8.3.1 or later
Trigger None
Workaround None
8.4.0.0 System Monitoring
HU02424 All Suggested Frequent GUI refreshing adversely impacts usability on some screens (show details)
Symptom None
Environment Systems running v8.3.1 or later
Trigger None
Workaround None
8.4.0.0 Graphical User Interface
IT32338 All Suggested Testing LDAP Authentication fails if username & password are supplied (show details)
Symptom None
Environment Systems running v8.2.1 or later
Trigger Test LDAP Authentication using username & password
Workaround Use remote authenication with LDAP
8.4.0.0 LDAP

4. Useful Links

Description Link
Support Websites
Update Matrices, including detailed build version
Support Information pages providing links to the following information:
  • Interoperability information
  • Product documentation
  • Limitations and restrictions, including maximum configuration limits
Spectrum Virtualize Family of Products Inter-System Metro Mirror and Global Mirror Compatibility Cross Reference
Software Upgrade Test Utility
Software Upgrade Planning