Release Note for systems built with IBM Spectrum Virtualize

This is the release note for the 7.7.1 release and details the issues resolved in all Program Temporary Fixes (PTFs) between 7.7.1.0 and 7.7.1.9. This document will be updated with additional information whenever a PTF is released.

This document was last updated on 18 November 2020.

  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 7.7.1 release:

2. Known Issues and Restrictions

Note: For clarity, the term "node" will be used to refer to a SAN Volume Controller node or Storwize system node canister.
Details Introduced

In the GUI, when filtering volumes by host, if there are more than 50 host objects, then the host list will not include the hosts’ names.

This issue will be fixed in a future PTF.

 7.7.1.7

Systems with encrypted managed disks cannot be upgraded to v7.7.1.5.

This is a temporary restriction that is policed by the software upgrade test utility. Please check this release note regularly for updates.

 7.7.1.5

There is a small possibility that during an upgrade, a 2145-SV1 node may fail to reboot correctly.

If any 2145-SV1 nodes fail to come back online after 30 minutes, please check whether the power LED on the front of the machine is lit. If the power LED is not lit then the fix is to remove the input power from both PSUs, wait 10 seconds and then re-insert the power and turn the node on. The software upgrade will then continue as normal.

The new BIOS that is being installed will fix the issue that causes the reboot to fail for future upgrades.

 7.7.1.1

When a system first upgrades from pre-7.7.0 to 7.7.0 or later, the in-memory audit log will be cleared. This means the catauditlog CLI command, and the GUI, will not show any commands that were issued before the upgrade.

The upgrade test utility will write the existing auditlog to a file in /dumps/audit on the config node. If the auditlog is needed at a later point, this file can be copied from the config node.

Subsequent upgrades will not affect the contents of the auditlog.

 7.7.0.0

Priority Flow Control for iSCSI is only supported on Brocade VDX 10GbE switches.

This is a temporary restriction that will be lifted in a future v7.7 PTF.

 7.7.0.0

It is not possible to replace the mid-plane in a SVC 12F SAS expansion enclosure.

If a SVC 12F mid-plane must be replaced then a new enclosure will be provided.

This is a temporary restriction that will be lifted in a future v7.7 PTF.

 7.7.0.0
V3500 and V3700 systems with 4GB of memory cannot be upgraded to v7.6 or later.

It should be noted that V3500 systems with 4GB of memory will need to upgrade to v7.5.0.4 or a later v7.5.0 PTF before attempting to increase memory to 8GB.

 7.6.0.0
Host Disconnects Using VMware vSphere 5.5.0 Update 2 and vSphere 6.0

Refer to this flash for more information

 n/a
If an update stalls or fails then Contact IBM Support for Further Assistance n/a

3. Issues Resolved

This release contains all of the fixes included in the 7.7.0.1 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-2016-10708 ibm10717661 7.7.1.9
CVE-2016-10142 ibm10717931 7.7.1.9
CVE-2017-11176 ibm10717931 7.7.1.9
CVE-2018-1433 ssg1S1012263 7.7.1.9
CVE-2018-1434 ssg1S1012263 7.7.1.9
CVE-2018-1438 ssg1S1012263 7.7.1.9
CVE-2018-1461 ssg1S1012263 7.7.1.9
CVE-2018-1462 ssg1S1012263 7.7.1.9
CVE-2018-1463 ssg1S1012263 7.7.1.9
CVE-2018-1464 ssg1S1012263 7.7.1.9
CVE-2018-1465 ssg1S1012263 7.7.1.9
CVE-2018-1466 ssg1S1012263 7.7.1.9
CVE-2016-6210 ssg1S1012276 7.7.1.9
CVE-2016-6515 ssg1S1012276 7.7.1.9
CVE-2013-4312 ssg1S1012277 7.7.1.9
CVE-2015-8374 ssg1S1012277 7.7.1.9
CVE-2015-8543 ssg1S1012277 7.7.1.9
CVE-2015-8746 ssg1S1012277 7.7.1.9
CVE-2015-8812 ssg1S1012277 7.7.1.9
CVE-2015-8844 ssg1S1012277 7.7.1.9
CVE-2015-8845 ssg1S1012277 7.7.1.9
CVE-2015-8956 ssg1S1012277 7.7.1.9
CVE-2016-2053 ssg1S1012277 7.7.1.9
CVE-2016-2069 ssg1S1012277 7.7.1.9
CVE-2016-2384 ssg1S1012277 7.7.1.9
CVE-2016-2847 ssg1S1012277 7.7.1.9
CVE-2016-3070 ssg1S1012277 7.7.1.9
CVE-2016-3156 ssg1S1012277 7.7.1.9
CVE-2016-3699 ssg1S1012277 7.7.1.9
CVE-2016-4569 ssg1S1012277 7.7.1.9
CVE-2016-4578 ssg1S1012277 7.7.1.9
CVE-2016-4581 ssg1S1012277 7.7.1.9
CVE-2016-4794 ssg1S1012277 7.7.1.9
CVE-2016-5412 ssg1S1012277 7.7.1.9
CVE-2016-5828 ssg1S1012277 7.7.1.9
CVE-2016-5829 ssg1S1012277 7.7.1.9
CVE-2016-6136 ssg1S1012277 7.7.1.9
CVE-2016-6198 ssg1S1012277 7.7.1.9
CVE-2016-6327 ssg1S1012277 7.7.1.9
CVE-2016-6480 ssg1S1012277 7.7.1.9
CVE-2016-6828 ssg1S1012277 7.7.1.9
CVE-2016-7117 ssg1S1012277 7.7.1.9
CVE-2016-10229 ssg1S1012277 7.7.1.9
CVE-2016-0634 ssg1S1012278 7.7.1.9
CVE-2017-5647 ssg1S1010892 7.7.1.7
CVE-2017-5638 ssg1S1010113 7.7.1.6
CVE-2016-6796 ssg1S1010114 7.7.1.6
CVE-2016-6816 ssg1S1010114 7.7.1.6
CVE-2016-6817 ssg1S1010114 7.7.1.6
CVE-2016-2177 ssg1S1010115 7.7.1.6
CVE-2016-2178 ssg1S1010115 7.7.1.6
CVE-2016-2183 ssg1S1010115 7.7.1.6
CVE-2016-6302 ssg1S1010115 7.7.1.6
CVE-2016-6304 ssg1S1010115 7.7.1.6
CVE-2016-6306 ssg1S1010115 7.7.1.6
CVE-2016-5696 ssg1S1010116 7.7.1.6
CVE-2016-2834 ssg1S1010117 7.7.1.6
CVE-2016-5285 ssg1S1010117 7.7.1.6
CVE-2016-8635 ssg1S1010117 7.7.1.6
CVE-2016-2183 ssg1S1010205 7.7.1.6
CVE-2016-5546 ssg1S1010205 7.7.1.6
CVE-2016-5547 ssg1S1010205 7.7.1.6
CVE-2016-5548 ssg1S1010205 7.7.1.6
CVE-2016-5549 ssg1S1010205 7.7.1.6
CVE-2016-5385 ssg1S1009581 7.7.1.3
CVE-2016-5386 ssg1S1009581 7.7.1.3
CVE-2016-5387 ssg1S1009581 7.7.1.3
CVE-2016-5388 ssg1S1009581 7.7.1.3
CVE-2016-3092 ssg1S1009284 7.7.1.2
CVE-2016-4430 ssg1S1009282 7.7.1.0
CVE-2016-4431 ssg1S1009282 7.7.1.0
CVE-2016-4433 ssg1S1009282 7.7.1.0
CVE-2016-4436 ssg1S1009282 7.7.1.0
CVE-2016-4461 ssg1S1010883 7.7.1.0

3.2 APARs Resolved

Show details for all APARs
APAR Affected Products Severity Description Resolved in Feature Tags
HU01866 SVC HIPER A faulty PSU sensor, in a node, can fill the sel log causing the service processor (BMC) to disable logging. If a snap is subsequently taken, from the node, a timeout will occur and it will be taken offline. It is possible for this to affect both nodes in an I/O group (show details)
Symptom Loss of Access to Data
Environment SVC systems using SV1 model nodes
Trigger None
Workaround None
7.7.1.9 System Monitoring
HU01767 All Critical Reads of 4K/8K from an array can under exceptional circumstances return invalid data. For more details refer to this Flash (show details)
Symptom Loss of Access to Data
Environment Systems running v7.8.0 or earlier
Trigger None
Workaround None
7.7.1.9 RAID, Thin Provisioning
HU01771 SVC, V7000 High Importance An issue with the CMOS battery in a node can cause an unexpectedly large log file to be generated by the BMC. At log collection the node may be taken offline (show details)
Symptom Loss of Redundancy
Environment SVC & V7000 systems running v7.8 or later
Trigger Node CMOS battery issue
Workaround None
7.7.1.9 System Monitoring
HU01445 SVC, V7000 Suggested Systems with heavily used RAID-1 or RAID-10 arrays may experience a node warmstart (show details)
Symptom Single Node Warmstart
Environment SVC & V7000 systems running v7.7 or later
Trigger None
Workaround None
7.7.1.9
HU01624 All Suggested GUI response can become very slow in systems with a large number of compressed and uncompressed volume (show details)
Symptom None
Environment Systems that have a large number of volumes (1000+) including some that are compressed
Trigger None
Workaround None
7.7.1.9 Graphical User Interface
HU01628 All Suggested In the GUI on the Volumes page whilst using the filter function some volumes entries may not be displayed until the page has completed loading (show details)
Symptom None
Environment All systems
Trigger Enter a filter string on the Volumes page
Workaround None
7.7.1.9 Graphical User Interface
HU01664 All Suggested A timing window issue during an upgrade can cause the node restarting to warmstart stalling the upgrade (show details)
Symptom Single Node Warmstart
Environment Systems running v7.8 or later
Trigger None
Workaround None
7.7.1.9 System Update
HU01687 All Suggested For volumes by host, ports by host and volumes by pool pages in the GUI when the number of items is greater than 50 then the item name will not be displayed (show details)
Symptom None
Environment Systems running v7.7 or v7.8
Trigger More than 50 hosts/pools
Workaround Use CLI
7.7.1.9 Graphical User Interface
HU01698 All Suggested A node warmstart may occur when deleting a compressed volume if a host has written to the volume minutes before the volume is deleted (show details)
Symptom Single Node Warmstart
Environment Systems with compressed volumes
Trigger Write I/O received by compression code for a volume that has been deleted
Workaround Remove volume host mappings and leave an interval of time (90mins) before removing the volume
7.7.1.9 Compression
HU01730 SVC Suggested When running the DMP for a 1046 error the picture may not indicate the correct position of the failed adapter (show details)
Symptom None
Environment DH8 systems
Trigger Run DMP for 1046 error
Workaround Be aware of which adapter is failed and do not use the picture in the GUI
7.7.1.9 GUI Fix Procedure
HU01763 SVC Suggested A single node warmstart may occur on a DH8 config node when inventory email is created. The issue only occurs if this coincides with a very high rate of CLI commands and high I/O workload on the config node (show details)
Symptom Single Node Warmstart
Environment DH8 systems running v7.7.1 or later
Trigger High CLI workload and I/O workload at same time as inventory email
Workaround None
7.7.1.9 Command Line Interface, System Monitoring
HU01706 All HIPER Areas of volumes written with all-zero data may contain non-zero data. For more details refer to this Flash (show details)
Symptom Incorrect data read from volume
Environment Systems running 7.7.1.7 or 7.8.1.3
Trigger See Flash
Workaround None
7.7.1.8
HU00744 (reverted) All Suggested This APAR has been reverted in light of issues with the fix. This APAR will be re-applied in a future PTF 7.7.1.8 Cache
HU01239 & HU01255 & HU01586 All HIPER The presence of a faulty SAN component can delay lease messages between nodes leading to a cluster-wide lease expiry and consequential loss of access (show details)
Symptom Loss of Access to Data
Environment Systems with 16Gb HBAs
Trigger Faulty SAN hardware (adapter, SFP, switch)
Workaround None
7.7.1.7 Reliability Availability Serviceability
HU01505 All HIPER A non-redundant drive experiencing many errors can be taken offline obstructing rebuild activity (show details)
Symptom Loss of Access to Data
Environment All systems
Trigger Drive with many errors
Workaround Bring drive online
7.7.1.7 Backend Storage, RAID
HU01646 All HIPER A new failure mechanism in the 16Gb HBA driver can under certain circumstances lead to a lease expiry of the entire cluster (show details)
Symptom Loss of Access to Data
Environment Systems with 16Gb HBAs
Trigger Faulty SAN hardware (adapter; SFP; switch)
Workaround None
7.7.1.7 Reliability Availability Serviceability
HU01267 All Critical An unusual interaction between Remote Copy and FlashCopy can lead to both nodes in an I/O group warmstarting (show details)
Symptom Loss of Access to Data
Environment Systems using GMCV
Trigger There may be a number of different scenarios that trigger this issue. Here is one: With GMCV running; Stop GMCV with -access (i.e. svctask stoprcconsistgrp -access); Secondary site becomes primary site and vice versa; Start a FlashCopy on the now primary site; Stop FlashCopy restore before it completes (with -force); Current primary becomes exposed to the issue; Reinstate GMCV (primary becomes secondary again); Stop GMCV with -access again; Both nodes warmstart.
Workaround None
7.7.1.7 Global Mirror with Change Volumes
HU01416 All Critical ISL configuration activity may cause a cluster-wide lease expiry (show details)
Symptom Loss of Access to Data
Environment All systems
Trigger None
Workaround None
7.7.1.7 Reliability Availability Serviceability
HU01490 All Critical When attempting to add/remove multiple IQNs to/from a host the tables that record host-wwpn mappings can become inconsistent resulting in repeated node warmstarts across I/O groups (show details)
Symptom Loss of Access to Data
Environment Systems with iSCSI connected hosts
Trigger
  • A addhostport command with iqn2 and iqn1 (where iqn1 is already recorded) is entered;
  • This command attempts to add iqn2 but determines that iqn1 is a duplicate and the CLI command fails;
  • Later whenever a login request from iqn2 is received internal checking detects an inconsistency and warmstarts the node
Workaround Do not use multiple IQNs in iSCSI add/remove commands
7.7.1.7 iSCSI
HU01519 V7000 Critical One PSU may silently fail leading to the possibility of a dual node reboot (show details)
Symptom Loss of Access to Data
Environment Storwize V7000 Gen 1 systems
Trigger None
Workaround None
7.7.1.7 Reliability Availability Serviceability
HU01528 SVC Critical Both nodes may warmstart due to Sendmail throttling (show details)
Symptom Loss of Access to Data
Environment V9000 systems
Trigger None
Workaround None
7.7.1.7
HU01549 All Critical During a system upgrade HyperV-clustered hosts may experience a loss of access to any iSCSI connected volumes (show details)
Symptom Loss of Access to Data
Environment Systems running v7.5 or earlier with iSCSI connected HyperV clustered hosts
Trigger Upgrade to v7.6 or later
Workaround None
7.7.1.7 System Update, iSCSI
HU01572 All Critical SCSI 3 commands from unconfigured WWPNs may result in multiple warmstarts leading to a loss of access (show details)
Symptom Loss of Access to Data
Environment Systems with iSCSI connected hosts
Trigger None
Workaround None
7.7.1.7 iSCSI
HU01635 All Critical A slow memory leak in the host layer can lead to an out-of-memory condition resulting in offline volumes or performance degradation (show details)
Symptom Offline Volumes
Environment Systems running v7.6 or later
Trigger None
Workaround None
7.7.1.7 Hosts, Performance
IT20627 All Critical When Read-Intensive drives are used as quorum disks a drive outage can occur. Under some circumstances this can lead to a loss of access (show details)
Symptom Loss of Access to Data
Environment All systems
Trigger Using Read-Intensive drives as quorum disks
Workaround None
7.7.1.7 Quorum
HU00762 All High Importance Due to an issue in the cache component, nodes within an I/O group are not able to form a caching-pair and are serving I/O through a single node (show details)
Symptom Loss of Redundancy
Environment Systems running v7.3 or later
Trigger None
Workaround None
7.7.1.7 Cache
HU01428 V7000, V5000, V3700, V3500 High Importance Scheduling issue adversely affects performance resulting in node warmstarts (show details)
Symptom Multiple Node Warmstarts
Environment Storwize systems running v7.7.1 or later
Trigger None
Workaround None
7.7.1.7 Reliability Availability Serviceability
HU01477 V7000, V5000, V3700, V3500 High Importance Due to the way enclosure data is read it is possible for a firmware mismatch between nodes to occur during an upgrade (show details)
Symptom Loss of Redundancy
Environment Storwize systems running v7.8 or later
Trigger Upgrade from v7.8.0 to v7.8.1
Workaround Check for dependent volumes. If there are none then manually load enclosure firmware from service mode
7.7.1.7 System Update
HU01488 V7000, V5000 High Importance SAS transport errors on an enclosure slot have the potential to affect an adjacent slot leading to double drive failures (show details)
Symptom Loss of Redundancy
Environment Storwize V5000 and V7000 systems running v7.7 or later
Trigger None
Workaround None
7.7.1.7 Drives
HU01506 All High Importance Creating a volume copy with the -autodelete option can cause a timer scheduling issue leading to node warmstarts (show details)
Symptom Multiple Node Warmstarts
Environment All systems
Trigger None
Workaround Do not use the -autodelete option when creating a volume copy
7.7.1.7 Volume Mirroring
HU01569 SVC High Importance When compression utilisation is high the config node may exhibit longer I/O response times than non-config nodes (show details)
Symptom Performance
Environment SVC systems using compression
Trigger High compression workloads
Workaround None
7.7.1.7 Compression
HU01579 All High Importance In systems where all drives are of type HUSMM80xx0ASS20 it will not be possible to assign a quorum drive (show details)
Symptom Loss of Redundancy
Environment Systems with drives of type HUSMM80xx0ASS20
Trigger Attempt to assign drive type as quorum
Workaround Manually assign a different drive type as quorum
7.7.1.7 Drives, Quorum
HU01609 & IT15343 All High Importance When the system is busy, the compression component may be paged out of memory resulting in latency that can lead to warmstarts (show details)
Symptom Multiple Node Warmstarts
Environment Systems using compression
Trigger None
Workaround Reduce compression workload
7.7.1.7 Compression
HU01614 All High Importance After a node is upgraded hosts defined as TPGS may have paths set to inactive (show details)
Symptom Loss of Redundancy
Environment Systems running v7.6 or earlier with host type TPGS
Trigger Upgrade to v7.7 or later
Workaround None
7.7.1.7 Hosts
HU01636 V5000, V3700, V3500 High Importance A connectivity issue with certain host SAS HBAs can prevent hosts from establishing stable communication with the storage controller (show details)
Symptom Performance
Environment Systems presenting storage to hosts with N2225 adapters
Trigger Host with N2225 adapters running Windows Server 2012R2
Workaround None
7.7.1.7 Hosts
HU01638 All High Importance When upgrading to v7.6 or later, if there is another cluster in the same zone which is at v5.1 or earlier then nodes will warmstart and the upgrade will fail (show details)
Symptom Multiple Node Warmstarts
Environment Systems running v7.5 or earlier
Trigger Upgrade to v7.6 or later with a cluster in the same zone running v6.1 or earlier
Workaround Unzone any cluster running v5.1 or earlier from the cluster being upgraded
7.7.1.7 System Update
IT17564 All High Importance All nodes in an I/O group may warmstart when a DRAID array experiences drive failures (show details)
Symptom Loss of Access to Data
Environment Systems with DRAID arrays
Trigger Failing drives
Workaround None
7.7.1.7 Distributed RAID
IT19726 SVC High Importance Warmstarts may occur when the attached SAN fabric is congested and HBA transmit paths become stalled preventing the HBA firmware from generating the completion for a FC command (show details)
Symptom Multiple Node Warmstarts
Environment SVC systems
Trigger SAN congestion
Workaround None
7.7.1.7 Hosts
IT21383 SVC, V7000, V5000 High Importance Heavy I/O may provoke inconsistencies in resource allocation leading to node warmstarts (show details)
Symptom Multiple Node Warmstarts
Environment Systems attached to slow backend controllers
Trigger None
Workaround None
7.7.1.7 Reliability Availability Serviceability
IT22376 V5000 High Importance Upgrade of V5000 Gen 2 systems, with 16GB node canisters, can become stalled with multiple warmstarts on first node to be upgraded (show details)
Symptom Loss of Redundancy
Environment Storwize V5000 Gen 2 systems with 16GB memory in each node canister
Trigger Upgrade to v7.6.1 or v7.7.1
Workaround None
7.7.1.7 System Update
HU00744 (reverted in 7.7.1.8) All Suggested Single node warmstart due to an accounting issue within the cache component (show details)
Symptom Single Node Warmstart
Environment Systems running v7.3 or later
Trigger None
Workaround None
7.7.1.7 Cache
HU00763 & HU01237 V7000, V5000, V3700, V3500 Suggested A node warmstart may occur when a quorum disk is accessed at the same time as the login to that disk is closed (show details)
Symptom Single Node Warmstart
Environment Storwize systems
Trigger None
Workaround None
7.7.1.7 Quorum
HU01098 All Suggested Some older backend controller code levels do not support C2 commands resulting in 1370 entries in the Event Log for every detectmdisk (show details)
Symptom None
Environment Systems running v7.6 or later attached to backend controllers running older code levels
Trigger Issue CLI command detectmdisk
Workaround None
7.7.1.7 Backend Storage
HU01228 All Suggested Automatic T3 recovery may fail due to the handling of quorum registration generating duplicate entries (show details)
Symptom None
Environment All systems
Trigger None
Workaround None
7.7.1.7 Reliability Availability Serviceability
HU01229 V7000, V5000, V3700, V3500 Suggested The DMP for a 3105 event does not identify the correct problem canister (show details)
Symptom None
Environment Storwize systems
Trigger None
Workaround Problem canister identified in Event log
7.7.1.7 GUI Fix Procedure
HU01332 All Suggested Performance monitor and Spectrum Control show zero CPU utilisation for compression (show details)
Symptom None
Environment Systems running v7.6 or later
Trigger None
Workaround None
7.7.1.7 System Monitoring
HU01385 All Suggested A warmstart may occur if a rmvolumecopy or rmrcrelationship command are issued on a volume while I/O is being forwarded to the associated copy (show details)
Symptom Single Node Warmstart
Environment Systems running v7.6 or v7.7 that are using Hyperswap
Trigger Issue an rmvolumecopy or rmrcrelationship command whilst hosts are still actively using the Hyperswap volume
Workaround Do not remove a Hyperswap volume or relationship whilst hosts are still mapped to it
7.7.1.7 HyperSwap
HU01391 & HU01581 V7000, V5000, V3700, V3500 Suggested Storwize systems may experience a warmstart due to an uncorrectable error in the SAS firmware (show details)
Symptom Single Node Warmstart
Environment Storwize systems
Trigger None
Workaround None
7.7.1.7 Drives
HU01430 V7000, V5000, V3700, V3500 Suggested Memory resource shortages in systems with 8GB of RAM can lead to node warmstarts (show details)
Symptom Single Node Warmstart
Environment Storwize Gen 1 systems running v7.6 or later
Trigger None
Workaround None
7.7.1.7
HU01457 V7000 Suggested In a hybrid V7000 cluster where one I/O group supports 10k volumes and another does not some operations on volumes may incorrectly be denied in the GUI (show details)
Symptom None
Environment Systems running v7.7.1 or later
Trigger None
Workaround Performed required actions in the CLI
7.7.1.7 Graphical User Interface
HU01466 SVC, V7000, V5000 Suggested Stretched cluster and HyperSwap I/O routing does not work properly due to incorrect ALUA data (show details)
Symptom None
Environment Systems running v7.8 or later using HyperSwap
Trigger Move a host to an alternate I/O group
Workaround Use chhost -site to set the correct site
7.7.1.7 Hosts, HyperSwap
HU01467 All Suggested Failures in the handling of performance statistics files may lead to missing samples in Spectrum Control and other tools (show details)
Symptom None
Environment All systems
Trigger None
Workaround Increase the sampling interval
7.7.1.7 System Monitoring
HU01469 V3700, V3500 Suggested Resource exhaustion in the iSCSI component can result in a node warmstart (show details)
Symptom Single Node Warmstart
Environment Storwize V3k systems using iSCSI
Trigger None
Workaround None
7.7.1.7 iSCSI
HU01484 All Suggested During a RAID array rebuild there may be node warmstarts (show details)
Symptom Single Node Warmstart
Environment All systems
Trigger None
Workaround None
7.7.1.7 RAID
HU01566 SVC Suggested After upgrading, numerous 1370 errors are seen in the Event Log (show details)
Symptom None
Environment SVC CG8 systems containing old SSD drives (type Z16IZD2B-146)
Trigger System Upgrade
Workaround Ignore 1370 errors
7.7.1.7 System Update
HU01582 All Suggested A compression issue in IP replication can result in a node warmstart (show details)
Symptom Single Node Warmstart
Environment Systems running v7.7 or later using IP Replication
Trigger None
Workaround None
7.7.1.7 IP Replication
HU01474 SVC, V7000 HIPER Host writes to a read-only secondary volume trigger I/O timeout warmstarts (show details)
Symptom Multiple Node Warmstarts
Environment SVC SV1 & V7000 Gen 2+ systems running v7.7 or later using Remote Copy
Trigger Host write I/O to a read-only secondary volume
Workaround Do not attempt to issue host writes to a secondary volume while remote copy is active
7.7.1.6 Global Mirror, Global Mirror with Change Volumes, Metro Mirror
HU01479 All HIPER The handling of drive reseats can sometimes allow I/O to occur before the drive has been correctly failed resulting in offline MDisks (show details)
Symptom Loss of Access to Data
Environment Systems running v7.6 or later using DRAID
Trigger Reseat a failed drive
Workaround Rather than reseating the drive use the CLI or GUI to fail then unfail it
7.7.1.6 Distributed RAID
HU01483 All HIPER mkdistributedarray command may get stuck in the prepare state. Any interaction with the volumes in that array will result in multiple warmstarts (show details)
Symptom Loss of Access to Data
Environment Systems running v7.7.1 or later
Trigger mkdistributedarray command gets stuck in the prepare state; run rmarray or rmvdisk commands or generate I/O to volumes created on that array
Workaround None
7.7.1.6 Distributed RAID
HU01500 All HIPER Node warmstarts can occur when the iSCSI Ethernet MTU is changed (show details)
Symptom Loss of Access to Data
Environment Systems running v7.7.1 with iSCSI connected hosts
Trigger Change Ethernet MTU for iSCSI
Workaround None
7.7.1.6 iSCSI
HU01225 & HU01330 & HU01412 All Critical Node warmstarts due to inconsistencies arising from the way cache interacts with compression (show details)
Symptom Loss of Access to Data
Environment Systems running v7.3 or later with compressed volumes
Trigger None
Workaround None
7.7.1.6 Cache, Compression
HU01371 SVC, V7000, V5000 High Importance A remote copy command related to HyperSwap may hang resulting in a warmstart of the config node (show details)
Symptom Loss of Redundancy
Environment Systems running v7.5 or later using HyperSwap
Trigger None
Workaround None
7.7.1.6 HyperSwap
HU01480 All High Importance Under some circumstances the config node does not fail over properly when using IPv6 adversely affecting management access via GUI and CLI (show details)
Symptom Configuration
Environment Systems using IPv6 cluster addresses
Trigger None
Workaround None
7.7.1.6 Command Line Interface, Graphical User Interface
HU01473 All Suggested EasyTier migrates an excessive number of cold extents to an overloaded nearline array (show details)
Symptom Performance
Environment Systems running v7.6.1 or later using EasyTier
Trigger None
Workaround None
7.7.1.6 EasyTier
HU01487 All Suggested Small increase in read response time for source volumes with additional FlashCopy maps (show details)
Symptom Performance
Environment Systems running v7.5 or later using Global Mirror with Change Volumes
Trigger Add FlashCopy mappings to GMCV source volumes
Workaround Avoid using incremental FlashCopy
7.7.1.6 FlashCopy, Global Mirror with Change Volumes
HU01498 All Suggested GUI may be exposed to CVE-2017-5638 (see Section 3.1) 7.7.1.6 Security
IT18752 All Suggested When the config node processes an lsdependentvdisks command, issued via the GUI, that has a large number of objects in its parameters, it may warmstart (show details)
Symptom Single Node Warmstart
Environment Systems running v7.7 or later
Trigger Issue lsdependentvdisks command, via the GUI, for a large number of objects
Workaround Issue those commands, via the CLI
7.7.1.6 Graphical User Interface
HU01193 All HIPER A drive failure whilst an array rebuild is in progress can lead to both nodes in an I/O group warmstarting (show details)
Symptom Loss of Access to Data
Environment Systems using DRAID
Trigger Drive failure
Workaround None
7.7.1.5 Distributed RAID
HU01382 All HIPER Mishandling of extent migration following a rmarray command can lead to multiple simultaneous node warmstarts with a loss of access (show details)
Symptom Loss of Access to Data
Environment Systems running v7.7.1 or later using DRAID
Trigger Executing a rmarray command
Workaround Avoid using rmarray command
7.7.1.5 Distributed RAID
HU01340 All Critical A port translation issue between v7.5 or earlier and v7.7.0 or later requires a Tier 2 recovery to complete an upgrade (show details)
Symptom Loss of Access to Data
Environment Systems running v7.5 or earlier
Trigger Upgrade to v7.7.x
Workaround None
7.7.1.5 System Update
HU01392 All Critical Under certain rare conditions FC mappings not in a consistency group can be added to a special internal consistency group resulting in a Tier 2 recovery (show details)
Symptom Loss of Access to Data
Environment Systems running v7.7.1 or later
Trigger None
Workaround None
7.7.1.5 FlashCopy
HU01410 SVC Critical An issue in the handling of FlashCopy map preparation can cause both nodes in an I/O group to be put into service state (show details)
Symptom Loss of Access to Data
Environment SVC systems using FlashCopy
Trigger None
Workaround None
7.7.1.5 FlashCopy
HU01223 All High Importance The handling of a rebooted nodes return to the cluster can occasionally become delayed resulting in a stoppage of inter-cluster relationships (show details)
Symptom Loss of Redundancy
Environment Systems running v7.3 or later using MetroMirror
Trigger None
Workaround None
7.7.1.5 Metro Mirror
HU01254 SVC High Importance A fluctuation of input AC power can cause a 584 error on a node (show details)
Symptom Loss of Redundancy
Environment SVC CF8/CG8 systems
Trigger None
Workaround None
7.7.1.5 Reliability Availability Serviceability
HU01402 V7000 High Importance Nodes can power down unexpectedly as they are unable to determine from their partner whether power is available (show details)
Symptom Loss of Redundancy
Environment V7000 Gen 1 systems running v7.6 or later
Trigger None
Workaround None
7.7.1.5 Reliability Availability Serviceability
HU01409 All High Importance Cisco Nexus 3000 switches at v5.0(3) have a defect which prevents a config node IP address changing in the event of a fail over (show details)
Symptom Loss of Redundancy
Environment Systems connected to Cisco Nexus 3000 switches
Trigger Config node fail over
Workaround None
7.7.1.5 Reliability Availability Serviceability
IT14917 All High Importance Node warmstarts due to a timing window in the cache component. For more details refer to this Flash (show details)
Symptom Multiple Node Warmstarts
Environment Systems running v7.4 or later
Trigger None
Workaround None
7.7.1.5 Cache
HU00831 All Suggested Single node warmstart due to hung I/O caused by cache deadlock (show details)
Symptom Single Node Warmstart
Environment Systems running v7.2 or later
Trigger Hosts sending many large block IOs that use many credits per I/O
Workaround None
7.7.1.5 Cache
HU01022 SVC, V7000 Suggested Fibre channel adapter encountered a bit parity error resulting in a node warmstart (show details)
Symptom Single Node Warmstart
Environment Systems running v7.6 or later using 16Gb HBAs
Trigger None
Workaround None
7.7.1.5 Hosts
HU01269 All Suggested A rare timing conflict between two process may lead to a node warmstart (show details)
Symptom Single Node Warmstart
Environment All systems
Trigger None
Workaround None
7.7.1.5
HU01399 All Suggested For certain config nodes the CLI Help commands may not work (show details)
Symptom None
Environment All systems
Trigger None
Workaround Use Knowledge Center
7.7.1.5 Command Line Interface
HU01432 All Suggested Node warmstart due to an accounting issue within the cache component (show details)
Symptom Single Node Warmstart
Environment Systems running v7.3 or later
Trigger None
Workaround None
7.7.1.5 Cache
IT17302 V5000, V3700, V3500 Suggested Unexpected 45034 1042 entries in the Event Log (show details)
Symptom None
Environment Systems running v7.7 or later
Trigger None
Workaround None
7.7.1.5 System Monitoring
IT18086 All Suggested When a volume is moved between I/O groups a node may warmstart (show details)
Symptom None
Environment Systems running v7.7.1 or later
Trigger No warmstarts since upgrade. Move a volume between I/O groups
Workaround None
7.7.1.5
HU01379 All HIPER Resource leak in the handling of Read Intensive drives leads to offline volumes (show details)
Symptom Offline Volumes
Environment Systems running v7.7.1 or later using read intensive drives
Trigger None
Workaround None
7.7.1.4 Drives
HU01783 All Critical Replacing a failed drive in a DRAID array, with a smaller drive, may result in multiple Tier 2 recoveries putting all nodes in service state with error 564 and/or 550 (show details)
Symptom Loss of Access to Data
Environment Systems using DRAID
Trigger Replacing a drive with one of less capacity
Workaround Ensure replacement drives are the same capacity
7.7.1.4 Distributed RAID
HU01347 All High Importance During an upgrade to v7.7.1 a deadlock in node communications can occur leading to a timeout and node warmstarts (show details)
Symptom Multiple Node Warmstarts
Environment Systems running v7.7.0 or earlier using SE volumes
Trigger Upgrade to v7.7.1
Workaround None
7.7.1.4 Thin Provisioning
HU01381 All High Importance A rare timing issue in FlashCopy may lead to a node warmstarting repeatedly and then entering a service state (show details)
Symptom Multiple Node Warmstarts
Environment Systems running v7.2 or later using FlashCopy
Trigger None
Workaround None
7.7.1.4 FlashCopy
HU01247 All Suggested When a FlashCopy consistency group is stopped more than once in rapid succession a node warmstart may result (show details)
Symptom Single Node Warmstart
Environment Systems using FlashCopy
Trigger Stop the same FlashCopy consistency group twice in rapid succession using -force on second attempt
Workaround Avoid stopping the same flash copy consistency group more than once
7.7.1.4 FlashCopy
HU01323 All Suggested Systems using Volume Mirroring that upgrade to v7.7.1.x and have a storage pool go offline may experience a node warmstart (show details)
Symptom Single Node Warmstart
Environment Systems running v7.7.1 or later using volume mirroring
Trigger 1. Mirrored volumes are created at a pre-7.7.1 code level; 2. The system is upgraded to 7.7.1.1 or later; 3. A storage pool goes offline before the nodes next restart.
Workaround None
7.7.1.4 Volume Mirroring
HU01374 All Suggested Where an issue with Global Mirror causes excessive I/O delay, a timeout may not function resulting in a node warmstart (show details)
Symptom Single Node Warmstart
Environment Systems running v7.7.0 or later using Global Mirror
Trigger None
Workaround None
7.7.1.4 Global Mirror
HU01226 All High Importance Changing max replication delay from the default to a small non-zero number can cause hung IOs leading to multiple node warmstarts and a loss of access (show details)
Symptom Loss of Access to Data
Environment Systems running v7.6 or later using Global Mirror
Trigger Changing max replication delay to a small non-zero number
Workaround Do not change max replication delay to below 30 seconds
7.7.1.3 Global Mirror
HU01257 All High Importance Large (>1MB) write IOs to volumes can lead to a hung I/O condition resulting in node warmstarts (show details)
Symptom Multiple Node Warmstarts
Environment Systems running v7.7.1 or later
Trigger Large (>1MB) write IOs
Workaround Try to keep maximum I/O write transfer size to volumes below 1MB
7.7.1.3
HU01386 All High Importance Where latency between sites is greater than 1ms host write latency can be adversely impacted. This is can be more likely in the presence of large I/O transfer sizes or high IOPS (show details)
Symptom Performance
Environment Systems running v7.5 or later using HyperSwap with no Remote Copy partnerships
Trigger Inter-site latency is greater than 1ms
Workaround Keep inter-site latency below 1ms. Use Metro Mirror or Global Mirror replication
7.7.1.3 HyperSwap
HU01017 All Suggested The result of CLI commands are sometimes not promptly presented in the GUI (show details)
Symptom None
Environment All systems
Trigger None
Workaround In the GUI, navigate to "GUI Preferences" ("GUI Preferences->General" in v7.6 or later) and refresh the GUI cache.
7.7.1.3 Graphical User Interface
HU01227 All Suggested High volumes of events may cause the email notifications to become stalled (show details)
Symptom None
Environment Systems running v7.6 or later
Trigger More than 15 events per second
Workaround None
7.7.1.3 System Monitoring
HU01234 All Suggested After upgrade to 7.6 or later iSCSI hosts may incorrectly be shown as offline in the CLI (show details)
Symptom None
Environment Systems with iSCSI connected hosts
Trigger Upgrade to v7.6 or later from v7.5 or earlier
Workaround None
7.7.1.3 iSCSI
HU01251 V7000, V5000, V3700, V3500 Suggested When following the DMP for a 1685 event, if the option for drive reseat has already been attempted is selected, the process to replace a drive is not started (show details)
Symptom None
Environment Storwize systems running v7.3 or later
Trigger DMP for a 1685 event is run. Select "drive reseat has already been attempted" option
Workaround Manually replace drive
7.7.1.3 GUI Fix Procedure
HU01292 All Suggested Under some circumstances the re-calculation of grains to clean can take too long after a FlashCopy done event has been sent resulting in a node warmstart (show details)
Symptom Single Node Warmstart
Environment Systems running v7.7.0 or later
Trigger None
Workaround None
7.7.1.3 FlashCopy
IT17102 All Suggested Where the maximum number of I/O requests for a FC port has been exceeded, if a SCSI command, with an unsupported opcode, is received from a host then the node may warmstart (show details)
Symptom Single Node Warmstart
Environment Systems running v7.5 or later
Trigger None
Workaround None
7.7.1.3
HU01272 All HIPER Replacing a drive in a system with a DRAID array can result in T2 recovery warmstarts. For more details refer to this Flash (show details)
Symptom Loss of Access to Data
Environment Systems running v7.7.1.0 or later using DRAID
Trigger Replacing a drive
Workaround Change the failed drive to unused prior to removal by using "chdrive -use unused $driveid" before it is removed from the system; Physically replace the drive; Manually add the drive to the array using "charraymember -newdrive $driveid"; Remove the ghost drive by using "chdrive -use unused $ghostdriveid".
7.7.1.2 Distributed RAID
HU01118 V7000, V5000, V3700, V3500 HIPER Due to a firmware issue both nodes in a V7000 Gen 2 may be powered off (show details)
Symptom Loss of Access to Data
Environment Storwize V3500, V3700, V5000 & V7000 Gen 2 systems running v7.6 or later
Trigger None
Workaround None
7.7.1.1
HU01208 All HIPER After upgrading to v7.7 or later from v7.5 or earlier and then creating a DRAID array, with a node reset, the system may encounter repeated node warmstarts which will require a Tier 3 recovery (show details)
Symptom Loss of Access to Data
Environment Systems running v7.5.0.0 or earlier
Trigger Code upgrade, create DRAID array and then reset nodes
Workaround None
7.7.1.1 Distributed RAID
HU00271 All High Importance An extremely rare timing window condition in the way GM handles write sequencing may cause multiple node warmstarts (show details)
Symptom Multiple Node Warmstarts
Environment Systems using GM
Trigger None
Workaround None
7.7.1.1 Global Mirror
HU00734 All High Importance Multiple node warmstarts due to deadlock condition during RAID group rebuild (show details)
Symptom Multiple Node Warmstarts
Environment All
Trigger Array rebuilds
Workaround Hold one node in service state till the array rebuild finishes
7.7.1.1
HU01109 SVC, V7000, V5000 High Importance Multiple nodes can experience a lease expiry when a FC port is having communications issues (show details)
Symptom Loss of Access to Data
Environment Systems running v7.6 or later using 16Gb HBAs
Trigger None
Workaround None
7.7.1.1 Reliability Availability Serviceability
HU01140 All High Importance EasyTier may unbalance the workloads on MDisks using specific Nearline SAS drives due to incorrect thresholds for their performance (show details)
Symptom Performance
Environment Systems running v7.3 or later using EasyTier
Trigger None
Workaround Add Enterprise-class drives to the MDisk or MDisk group that is experiencing unbalanced workloads
7.7.1.1 EasyTier
HU01141 All High Importance Node warmstart (possibly due to a network problem) when a CLI mkippartnership is issued. This may lead to loss of the config node requiring a Tier 2 recovery (show details)
Symptom Multiple Node Warmstarts
Environment Systems running v7.5 or later using IP Replication
Trigger Enter mkippartnership CLI command
Workaround Ensure partner cluster IP can be pinged before issuing a mkippartnership CLI command
7.7.1.1 IP Replication
HU01182 SVC, V7000, V5000 High Importance Node warmstarts due to 16Gb HBA firmware receiving an invalid SCSI TUR command (show details)
Symptom Multiple Node Warmstarts
Environment Systems running v7.4 or later using 16Gb HBAs
Trigger 16G FC HBA receives a SCSI TUR command with Total XFER LEN > 0
Workaround None
7.7.1.1
HU01184 All High Importance When removing multiple MDisks node warmstarts may occur (show details)
Symptom Multiple Node Warmstarts
Environment Systems running v7.6 or later
Trigger Multiple rmmdisk commands issued in rapid succession
Workaround Remove MDisks one at a time and let migration complete before proceeding to next MDisk removal
7.7.1.1
HU01185 All High Importance iSCSI target closes connection when there is a mismatch in sequence number (show details)
Symptom Loss of Access to Data
Environment Systems with iSCSI connected hosts
Trigger Out of sequence I/O
Workaround None
7.7.1.1 iSCSI
HU01210 SVC High Importance A small number of systems have broken, or disabled, TPMs. For these systems the generation of a new master key may fail preventing the system joining a cluster (show details)
Symptom Loss of Redundancy
Environment CG8 systems running v7.6.1 or later
Trigger Broken or disabled TPM
Workaround None
7.7.1.1
HU01221 SVC, V7000, V5000 High Importance Node warmstarts due to an issue with the state machine transition in 16Gb HBA firmware (show details)
Symptom Multiple Node Warmstarts
Environment Systems running v7.6 or later using 16Gb HBAs
Trigger None
Workaround None
7.7.1.1
HU01250 All High Importance When using lsvdisklba to find a bad block on a compressed volume, the volume can go offline (show details)
Symptom Offline Volumes
Environment Systems running v7.3 or later that are using compressed volumes
Trigger Using lsvdisklba
Workaround None
7.7.1.1 Compression
HU01516 All High Importance When node configuration data exceeds 8K in size some user defined settings may not be stored permanently resulting in node warmstarts (show details)
Symptom Loss of Redundancy
Environment Systems with large, complex configurations
Trigger A high number changes to node configuration
Workaround Minimise node configuration changes
7.7.1.1 Reliability Availability Serviceability
IT16148 All High Importance When accelerate mode is enabled due to the way promote/swap plans are prioritized over demote EasyTier is only demoting 1 extent every 5 minutes (show details)
Symptom Performance
Environment Systems running v7.5 or later using EasyTier
Trigger Over use of accelerate mode
Workaround Only use accelerate mode when migrating data between pools
7.7.1.1 EasyTier
IT16337 SVC, V7000, V5000 High Importance Hardware offloading in 16G FC adapters has introduced a deadlock condition that causes many driver commands to time out leading to a node warmstart. For more details refer to this Flash (show details)
Symptom Multiple Node Warmstarts
Environment Systems running v7.4 or later using 16Gb HBAs
Trigger None
Workaround None
7.7.1.1
HU01024 V7000, V5000, V3700, V3500 Suggested A single node warmstart may occur when the SAS firmwares ECC checking detects a single bit error. The warmstart clears the error condition in the SAS chip (show details)
Symptom Single Node Warmstart
Environment Systems running v7.4 or later
Trigger None
Workaround None
7.7.1.1
HU01050 All Suggested DRAID rebuild incorrectly reports event code 988300 (show details)
Symptom None
Environment Systems running v7.6 or later
Trigger None
Workaround None
7.7.1.1 Distributed RAID
HU01063 SVC, V7000, V5000 Suggested 3PAR controllers do not support OTUR commands resulting in device port exclusions (show details)
Symptom None
Environment Systems virtualising 3PAR storage
Trigger None
Workaround None
7.7.1.1 Backend Storage
HU01155 All Suggested When a lsvdisklba or lsmdisklba command is invoked, for an MDisk with a back end issue, a node warmstart may occur (show details)
Symptom Single Node Warmstart
Environment Systems running v7.7.0 or earlier using compression
Trigger None
Workaround None
7.7.1.1 Compression
HU01187 All Suggested Circumstances can arise where more than one array rebuild operation can share the same CPU core resulting in extended completion times (show details)
Symptom Performance
Environment Systems running v7.4 or later
Trigger None
Workaround Avoid R5 array configurations
7.7.1.1
HU01194 All Suggested A single node warmstart may occur if CLI commands are received from the VASA provider in very rapid succession. This is caused by a deadlock condition which prevents the subsequent CLI command from completing (show details)
Symptom Single Node Warmstart
Environment Systems running v7.6 or later that are using vVols
Trigger None
Workaround None
7.7.1.1 vVols
HU01198 All Suggested Running the Comprestimator svctask analyzevdiskbysystem command may cause the config node to warmstart (show details)
Symptom Single Node Warmstart
Environment Systems running v7.6 or later using Comprestimator
Trigger Run svctask analyzevdiskbysystem
Workaround Avoid using svctask analyzevdiskbysystem
7.7.1.1 Comprestimator
HU01214 All Suggested GUI and snap missing EasyTier heatmap information (show details)
Symptom None
Environment Systems running v7.6 or later
Trigger None
Workaround Download the files individually via CLI
7.7.1.1 Support Data Collection
HU01219 SVC, V7000, V5000 Suggested Single node warmstart due to an issue in the handling of ECC errors within 16G HBA firmware (show details)
Symptom Single Node Warmstart
Environment Systems running v7.6 or later using 16Gb HBAs
Trigger None
Workaround None
7.7.1.1
HU01244 All Suggested When a node is transitioning from offline to online it is possible for excessive CPU time to be used on another node in the cluster which may lead to a single node warmstart (show details)
Symptom Single Node Warmstart
Environment Systems running v7.3 or later
Trigger None
Workaround None
7.7.1.1
HU01258 SVC Suggested A compressed volume copy will result in an unexpected 1862 message when site/node fails over in a stretched cluster configuration (show details)
Symptom None
Environment SVC systems running v7.4 or later in a stretched cluster configuration
Trigger Site/node failover
Workaround None
7.7.1.1 Compression

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
Supported Drive Types and Firmware Levels
SAN Volume Controller and Storwize Family Inter-cluster Metro Mirror and Global Mirror Compatibility Cross Reference
Software Upgrade Test Utility
Software Upgrade Planning