IBM Storwize V7000 ++++++++++++++++++++++++++++++++++++++++ ------------------------------------------------------------------------------- CONTENTS 1. Introduction 2. Available Education 3. Pre-Requisites 4. System Upgrade Information 5. Software Levels 6. Problems Resolved and New Features 7. Further Documentation 8. Known Issues and Restrictions in this Level 9. Maximum Configurations 10. Supported Hardware ------------------------------------------------------------------------------- 1. Introduction This document describes how to install version 7.3.0.7 of the IBM Storwize V7000. This release delivers fixes to resolve the APARs detailed in section 6. Please refer to the Recommended Software List and Supported Hardware List on the support website: http://www.ibm.com/storage/support/storwize/V7000 ------------------------------------------------------------------------------- 2. Available Education A course on Storwize V7000 Planning and Implementation is available. For further information or enrolment, contact your local IBM representative. Visit the support website for online learning materials and tutorials, IBM Redbooks and training information. ------------------------------------------------------------------------------- 3. Pre-Requisites Before installing this software level please check the following pre-requisites are met. Also please note the concurrent upgrade (upgrade with I/O running) restrictions. Performance of the system will be degraded during the software upgrade process while node canisters are taken offline to perform the upgrade and the write cache is flushed when each node is restarted. IBM recommends you perform the upgrade at a time of lower overall I/O activity and when you do not expect significant spikes in the system workload. Note that the cache flush mechanism is different when upgrading from releases prior to 7.3.0.0 to 7.3.0.0 or later. See section 8 for more details If you are performing a concurrent software upgrade, you must first ensure that all host paths are available and operating correctly. Check there are no unfixed errors in the event log or in the service GUI. Use normal service procedures to resolve any errors before proceeding. Before upgrading, IBM recommends using the software upgrade test utility "upgradetest". Refer to the following URL for further information: http://www.ibm.com/support/docview.wss?uid=ssg1S4000585 If you are running a version of Storwize V7000 software older than 6.4.0.0 then you will have to perform multiple upgrades to install the 7.3.0 software on your system. Please see the following web page for the full upgrade compatibility matrix. http://www.ibm.com/support/docview.wss?uid=ssg1S1003705 Note: Once an upgrade has completed, it is not possible to downgrade the software to an earlier level without deleting and manually rebuilding the system in its entirety, including restoring all volume data from an external backup. ------------------------------------------------------------------------------- 4. System Upgrade Information IMPORTANT: Systems with the 10 Gb/s Ethernet adapter will take longer to upgrade each node when upgrading from releases prior to V7.2.0 to V7.2.0 or later. This is due to an upgrade of the firmware on this adapter, and each node canister may take around 40 minutes to complete the upgrade. This additional delay will not affect systems upgrading from V7.2.0.x or V7.3.0.x to a later release. Global Mirror Relationships Must Be Stopped When Upgrading from releases prior to V7.2.0 to V7.2.0 or later http://www-01.ibm.com/support/docview.wss?uid=ssg1S1004472 ------------------------------------------------------------------------------- 5. Software Levels 2076 Software: 2076 Release Level 7.3.0.7 (97.5.1410080000) ------------------------------------------------------------------------------- 6. Problems Resolved and New Features New features in 7.3.0 * Compressed volumes can be shrunk in 7.3.0.0 or later, if supported by the host operating system * Support for 2076-524 control enclosure and 12 Gb/s expansion enclosures * Easy Tier V3, including: * Any two tiers in the same storage pool * Three tiers in the same storage pool * Rebalancing workload within a tier for optimised performance The rebalancing function works on storage pools with only one tier * New cache architecture New service features in 7.3.0 * Running the Software Upgrade Test Utility before applying this upgrade is mandatory This release contains all fixes included in the 7.2.0.6 release. APARs resolved in this release (7.3.0.7): Bash vulnerabilities CVE-2014-6271, CVE-2014-7169, CVE-2014-7186, CVE-2014-7187, CVE-2014-6277, CVE-2014-6278 Critical Fixes IT04654 Node assert when decompressing software-compressed data using compression accelerator hardware APARs resolved in previous 7.3.x releases: 7.3.0.6 Critical Fixes HU00497 Volumes offline due to incorrectly compressed data. Second fix for issue described in flash - www-01.ibm.com/support/docview.wss?rs=591&uid=ssg1S1004863 (HIPER) 7.3.0.5 This PTF addresses an issue affecting all earlier levels of 7.3.0 that may prevent the recovery procedure, in the unlikely event of a complete system failure, from completing successfully. Critical Fixes HU00277 Loss of access to data when adding a Global Mirror Change volume if the system is almost out of FlashCopy bitmap space (HIPER) HU00280 Multiple node asserts triggered by a Global Mirror disconnection HU00444 Node asserts due to overloaded compression engine HU00450 Manual upgrade with stopped Global Mirror relationships can not complete non-disruptively due to dependent volumes HU00464 Loss of access to data due to resource leak in V7.3.0 (HIPER) HU00496 SVC volumes offline and data unrecoverable due to issue described in flash - www-01.ibm.com/support/docview.wss?rs=591&uid=ssg1S1004863 (HIPER) HU00013 Nodes go into service state due to battery discharge fault HU00185 Cluster restart due to silent PSU failure High Importance Fixes HU00176 Node assert due to a IO deadlock when using the remote copy functions HU00274 Quiesce and resume of host IO when Global Mirror consistency group reaches consistent_synchronized HU00422 Node assert when using Global Mirror Change Volumes HU00443 Global Mirror Change Volumes stops replicating after an upgrade from 641 to 7.x HU00446 V7.3.0 cache does not make effective use of CPU resources Note: a restart is required to activate this fix if you upgrade from an earlier version of 7.3.0 HU00462 Node assert when using Easytier with a single tier in a pool on V7.3 IC90374 Node assert due to a IO deadlock when using the FlashCopy functions HU00069 Ierr generated in error on AC power loss HU00128 Node assert due to management firmware crash HU00170 Enclosure resets due to enclosure management firmware crash HU00185 PSU LED indicators latched due to enclosure firmware fault HU00237 Enclosure resets due to PSU battery communication problem IC91088 Upper node offline due to I2C bus 1 failure 7.3.0.4 Critical Fixes HU00354 Loss of access to data if upgrading directly from 6.4.x to 7.3.0.1, 7.3.0.2 or 7.3.0.3 with multiple access IO groups configured on any volume High Importance Fixes HU00432 Performance reduction and node asserts when running out of cache resources in 7.3.0.1, 7.3.0.2, or 7.3.0.3 7.3.0.3 OpenSSL vulnerability CVE-2014-0224 Critical Fixes HU00324 Compressed volumes offline after upgrading to 7.3.0.1 or 7.3.0.2 High Importance Fixes HU00296 Node assert when handling specific compression workloads HU00336 Single node assert when volumes go offline on systems running 7.3.0.0, 7.3.0.1 or 7.3.0.2 7.3.0.2 Critical Fixes HU00302 Multiple repeating node asserts if system has previously run a code release earlier than 6.4.0 and is upgraded to 7.3.0.1 without stepping through a 6.4.x release 7.3.0.1 Critical Fixes HU00014 Multiple node asserts if many volume-host mappings exist to a single host HU00133 Loss of access to data when an enclosure goes offline during software upgrade HU00195 Multiple node asserts when creating the first compressed volume in an IO group HU00247 Multiple node asserts when handling bad block errors on a RAID array HU00283 Multiple node asserts caused by invalid compressed volume metadata HU00287 Multiple node asserts when using hosts where the IQN is the same, but uses different capitalisation HU00298 Multiple node asserts when using IBM DS4000 using an incorrect host type IC92665 Multiple node asserts caused by iSCSI initiator using the same IQN as SVC or Storwize IT01250 Loss of access to data when node or node canister goes offline during drive update High Importance Fixes HU00017 Node assert after failed mkrcpartnership command HU00026 Node assert after all compressed volumes in an IO group are deleted HU00219 Node assert when stopping a FlashCopy map in a chain of FlashCopy mappings HU00253 Global Mirror with Change Volumes does not resume copying after an IO group goes offline at secondary cluster HU00257 Multiple node asserts when EMC RecoverPoint appliances restarted HU00291 Node assert caused by instability in IP replication connection HU00293 Node canister fails to boot after hard shutdown IC89562 Node assert when handling a large number of XCOPY commands IC89608 Node assert when a volume goes offline during an XCOPY command Suggested Fixes HU00183 GUI becomes non-responsive on larger configurations HU00232 Incorrect 1131 error in the event log HU00236 Performance degradation when changing the state of certain LEDs HU00241 Unresponsive GUI caused by locked IPC sockets HU00272 Arrays incorrectly reporting resync progress as 0% or 255% HU00288 GUI does not remember the most recently visited page HU00290 System incorrectly attempts to upgrade firmware during maintenance discharge HU00294 Event ID 981007 not always logged correctly HU00299 Array Rebuild can take longer than necessary IC92993 Fix Procedure for 1686 not replacing drive correctly IC94781 GUI Health status pod still showing red after offline node condition has been recovered ------------------------------------------------------------------------------- 7. Further Documentation The latest version of all Storwize V7000 documentation can be downloaded from the support website: http://www.ibm.com/storage/support/storwize/V7000 ------------------------------------------------------------------------------- 8. Known Issues and Restrictions in this Level Upgrades from releases prior to 7.3.0.0 to 7.3.0.0 or later will disable the cache when the upgrade starts and the cache will stay disabled until the upgrade completes, at which time it will be automatically re-enabled. If using IP replication, please review the set of restrictions published in the V7.3 Configuration Limits and Restrictions document. Intra-Cluster Global Mirror Not Supported Refer to the following flash for more information on this restriction: http://www.ibm.com/support/docview.wss?uid=ssg1S1003725 SAN Volume Controller or Storwize Family Inter-cluster Metro Mirror and Global Mirror Compatibility Cross Reference http://www.ibm.com/support/docview.wss?uid=ssg1S1003646 Use of a standards-compliant browser, such as Firefox, is recommended for using the management GUI. Use of Internet Explorer is supported but may result in slower performance of the GUI. If an upgrade stalls or fails then Contact IBM Support for Further Assistance http://www.ibm.com/support/docview.wss?uid=ssg1S1002894 ------------------------------------------------------------------------------- 9. Maximum Configurations The maximum configurations for 7.3.0 are documented in the 7.3 Configuration Limits and Restrictions document which is available from: http://www-01.ibm.com/support/docview.wss?rs=591&uid=ssg1S1004628 ------------------------------------------------------------------------------- 10. Supported Hardware This release of Storwize V7000 software is supported on the following node hardware types: Control enclosures: * 2076-112 * 2076-124 * 2076-312 * 2076-324 * 2076-524 Expansion enclosures: * 2076-212 * 2076-224 * 2076-12F * 2076-24F -------------------------------------------------------------------------------