hp matrix operating environment 7.0

HP Matrix Operating Environment 7.0Release Notes

AbstractThis document describes the HP Matrix Operating Environment 7.0 release content and provides information about limitations,and major and minor issues with suggested actions.

HP Part Number: T8671-91036Published: April 2013Edition: 3

© Copyright 2006–2013 Hewlett-Packard Development Company, L.P.

Confidential computer software. Valid license from HP required for possession, use or copying. Consistent with FAR 12.211 and 12.212, CommercialComputer Software, Computer Software Documentation, and Technical Data for Commercial Items are licensed to the U.S. Government undervendor's standard commercial license. The information contained herein is subject to change without notice. The only warranties for HP productsand services are set forth in the express warranty statements accompanying such products and services. Nothing herein should be construed asconstituting an additional warranty. HP shall not be liable for technical or editorial errors or omissions contained herein. UNIX is a registeredtrademark of The Open Group.

Acknowledgments

HP-UX Release 10.20 and later and HP-UX Release 11.00 and later (in both 32 and 64-bit configurations) on all HP 9000 computers are OpenGroup UNIX 95 branded products.

UNIX is a registered trademark of The Open Group.

Java is a US trademark of Sun Microsystems, Inc.

Intel and Itanium are trademarks or registered trademarks of Intel Corporation or its subsidiaries in the United States and other countries.

Microsoft and Windows are U.S. registered trademarks of Microsoft Corporation.

Oracle is a registered trademark of Oracle Corporation and/or its affiliates.

Revision history

Publication dateDocumentedition

Supported operating systemsSoftwareversion

Document partnumber

April 20133See the HP Insight Management 7.0 Support Matrix“For other languages and previous versions of thedocuments, go to the HP Support Center.”

7.0.0T8671-91036

May 20122See the HP Insight Management 7.0 Support Matrix7.0.0T8671-91036

March 20121See the HP Insight Management 7.0 Support Matrix7.0.0T8671-91036

http://www.hp.com/go/matrixoe/docs

https://h20566.www2.hp.com/portal/site/hpsc/public/psi/manualsResults/?sp4ts.oid=3725908

Contents

1 Description................................................................................................5Update recommendation...........................................................................................................5Supercedes..............................................................................................................................5Operating systems and devices..................................................................................................5Enhancements..........................................................................................................................5Fixes.......................................................................................................................................8Prerequisites.............................................................................................................................8Installation ..............................................................................................................................8

2 Issues and suggested actions.....................................................................15Matrix Operating Environment software.....................................................................................15Capacity Advisor....................................................................................................................16Infrastructure orchestration........................................................................................................16Insight managed system setup wizard........................................................................................26Instant Capacity Manager.......................................................................................................27Logical server management......................................................................................................27Process Resource Manager .....................................................................................................35Storage Provisioning Manager.................................................................................................36

3 Support and other resources......................................................................41Contacting HP........................................................................................................................41Related information.................................................................................................................43

Contents 3

1 DescriptionHP Matrix Operating Environment is advanced infrastructure management software that allowsyou to analyze and optimize both physical and virtual resources in the same way. Matrix OE givesyou a powerful toolkit to accelerate complex technology projects and simplify daily operations. Itenables you to plan capacity and power continuously, provision infrastructure consistently, andbalance across physical and virtual resources. It also allows you to ensure cost-effective highavailability for a larger set of applications.HP Matrix Operating Environment for ProLiant, HP Matrix Operating Environment for non-HPservers, and HP Matrix Operating Environment for HP-UX are integrated by design, deliveringcapabilities greater than the sum of its parts. Matrix OE is based on proven HP technologies, suchas HP Insight Control and HP Systems Insight Manager, the world’s most popular platformmanagement tool. Matrix OE takes advantage of HP Virtual Connect to bring the flexibility ofvirtualization to a physical environment.

• Matrix Operating Environment for ProLiant and Matrix Operating Environment for non-HPservers include capacity planning, advanced virtualization and logical server management,and Matrix OE infrastructure orchestration capabilities for automated design and provisioning.

• Matrix Operating Environment for ProLiant also includes Matrix OE recovery managementfor disaster recovery capabilities, Insight Control for advanced power management,performance management, and server deployment, and extends advanced virtualizationmanagement to HP server blades using HP Virtual Connect technology.

• Matrix Operating Environment for HP-UX includes capacity planning, advanced virtualizationmanagement, global workload management, and support for Integrity blades by HP IntegrityVirtual Machines Manager, HP-UX Workload Manager, and HP Insight Remote Support.

Update recommendationHP recommends that customers running 6.x and earlier versions upgrade to version 7.0.

SupercedesThe 7.0 version supersedes the 6.3.x released versions.

Operating systems and devicesSupported operating systems and devices are listed in the HP Insight Management Support Matrix:Matrix OE Information Library

EnhancementsMatrix OE 7.0 software provides new capabilities and expanded platform support.See the HP Insight Management Support Matrix at Matrix OE Information Library for moreinformation.A naming change from Insight Software to Insight Management is being made in this release tobetter describe and brand HP’s value-add server management products.

Insight Management product names

Previous nameNew name

HP Insight SoftwareHP Insight Management

HP Insight DynamicsHP Matrix Operating Environment

Update recommendation 5

www.hp.com/go/matrixoe/docs


Previous nameNew name

HP Insight OrchestrationHP Matrix Operating Environment infrastructureorchestration

HP Insight RecoveryHP Matrix Operating Environment recovery management

HP Virtualization ManagerHP Matrix Operating Environment visualization

New software on managed systems• Operating systems

Red Hat Enterprise Linux 6.2 (x86 and x64)◦◦ Red Hat Enterprise Linux 6.1 (x64)

◦ Red Hat Enterprise Linux 5.7 (x64)

◦ SUSE Linux Enterprise Server 10 SP4

• Hypervisors

Xen on RHEL 5.7◦◦ Xen on SLES 10 SP4

• Virtualization management software

Microsoft System Center Virtual Machine Manager 2008 R2 SP1◦◦ VMware vCenter Server 5.0

New servers that can be managed by Insight Management• HP ProLiant servers

HP ProLiant BL460c Gen8◦◦ HP ProLiant DL160 Gen8

◦ HP ProLiant DL360p Gen8

◦ HP ProLiant DL380p Gen8

◦ HP ProLiant ML350p Gen8

◦ HP ProLiant SL230s Gen8

◦ HP ProLiant SL250s Gen8

• HP Integrity servers

HP Integrity Superdome 2◦◦ HP Carrier-Grade rx2800 i2 Server DC Solution

6 Description

New features in Matrix Operating Environment• Matrix OE infrastructure orchestration (Windows CMS only)

Leading cloud functionality

◦ Unique bursting capabilities to Savvis

◦ With federated CMS (Central Management Server), scale to 10,000 HP ProLiant virtualmachines or 6,000 HP Integrity virtual machines

Increased infrastructure automation

◦ Auto-generation of storage pool entries from service template storage information

◦ Ability to use HP Storage Provisioning Manager (SPM) storage templates when definingservice template storage

◦ On-demand storage provisioning

◦ VMware VM Host/cluster awareness and provisioning; integrated with VMware vSphere5 Auto Deploy

◦ Support for VMware ESX resource pools and cloud capacity pools

◦ Enhanced support for physical networks with multiple VLANs (trunk network)

◦ Microsoft integration for System Center Virtual Machine Manager (SCVMM) and Hyper-Vtemplate deployment

Enhanced multi-tenancy support

◦ Organizations are managed in the new organization administrator portal

◦ Each organization has its own infrastructure orchestration environment

◦ VLAN separation for secure multi-tenancy

See the HP Matrix OE Infrastructure Orchestration User Guide at www.hp.com/go/matrixoe/docs for more information.

• Matrix OE recovery management (Windows CMS only)

Bi-directional failover of Recovery Group sets between two sites, allowing each set beactivated at chosen preferred site

◦

◦ Support for up to 500 disaster recovery-protected logical servers in a two-site configuration

◦ Support for 3PAR storage, allowing for metropolitan distances using synchronousreplication

◦ External storage adapter interface specification for customers to implement and integratewith the storage that is supported by Matrix OE but not yet integrated with Matrix OErecovery management

◦ Built-in retry capability that automatically restarts the failover job and recovers fromtransient issues

See the HP Matrix OE Recovery Management User Guide at www.hp.com/go/matrixoe/docsfor more information.

Enhancements 7




• Storage Provisioning Manager (SPM) (Windows CMS only)

On-demand storage provisioning – satisfy Matrix storage service requests by using eitherpre-provisioned volumes or by dynamically provisioning volumes to match the specifiedstorage requirements

◦

◦ Dynamic zoning – automatically zone Fibre Channel SANs (from supported vendors)based on service requirements

◦ Storage templates – provide administrative control to set policies and access controls forstorage provisioning

◦ New, enhanced user interface that includes:New summary screen with an advanced dashboard and statistics to monitor storageusage, oversubscription rations, and storage available to fulfill future service requests

–

– New interfaces for importing storage resources, creating storage templates, managingSPM security, monitoring service usage, and managing many other operations

– Improved responsiveness and ability to make real-time queries from the SPM catalog

See the HP Storage Provisioning Manager (SPM) User Guide at www.hp.com/go/matrixoe/docs for more information.

• Global Workload Manager (gWLM) for Integrity

◦ Managed iCAP integration and vPars v5 dynamic cores support for Superdome 2

FixesThe primary objectives of this release are to support of the latest HP hardware platforms, supportthe latest revisions of already supported operating systems and hypervisors, add new functionality,and fix software issues.Several software updates improved performance in large scale environments.

Prerequisites• Hardware, firmware, and software requirements for this release are listed in the HP Insight

Management 7.0 Support Matrix.• Installation and upgrade prerequisites are documented in the HP Insight Management 7.0

Preinstallation Worksheet, HP Insight Management 7.0 Installation and Configuration Guide,and HP Matrix Operating Environment Integrity CMS Installation and Configuration Guide.

These documents are available from the Matrix OE Information Library.

InstallationInstallation instructions are documented in the HP Insight Management 7.0 Installation andConfiguration Guide and HP Matrix Operating Environment Integrity CMS Installation andConfiguration Guide.These documents are available from the Matrix OE Information Library.

Newer version available for download

Matrix Operating Environment for HP-UXThe latest version of Matrix OE for Integrity is available from the following location:https://h20392.www2.hp.com/portal/swdepot/displayProductInfo.do?productNumber=vse

8 Description





https://h20392.www2.hp.com/portal/swdepot/displayProductInfo.do?productNumber=vse

Matrix Operating Environment for ProLiant and non-HP serversThe latest version of Matrix OE for ProLiant and non-HP servers is available from the followinglocation:https://h20392.www2.hp.com/portal/swdepot/displayProductInfo.do?productNumber=HPID

Accessing online help before licenses are installedYou can access the Matrix OE online help before you have installed licenses on the CMS byperforming the following steps.1. Select Help→For HP Systems Insight Manager from the top menu.2. In your browser address bar (if enabled), change the URL from the first line to the second line:

https://{CMS-FQDN}:50000/mxhelp/mxportal/en/help-welcome.htmlhttps://{CMS-FQDN}:50000/mxhelp/vsemgmt/en/licenses.html

where {CMS-FQDN} is the fully qualified domain name of the CMS.3. Press Enter to display the Matrix OE license management topic. From there, you can select

any of the help topics that apply to Matrix OE software. For help specific to an individualcomponent of Matrix OE, select the Related information topic and choose one of the links inthe section Help for related products.

Back up your existing CMS configuration before upgrading the softwareHP recommends making a backup of your pre-7.0 environment before upgrading the software.The upgrade process is not reversible. If, after you begin, you decide not to complete the upgradeto Matrix OE 7.0, a backup copy of your pre-7.0 environment is required to restore your CMS tothe pre-upgrade state.

IMPORTANT: After the upgrade process has successfully completed, you must also perform abackup of the upgraded 7.0 configuration to restore your environment to a known good state inthe event of a catastrophic failure. This is necessary to ensure that managed resources, if they werechanged by the 7.0 CMS, will continue to be manageable if the CMS ever needs to be restoredfrom backup.

See the Backing up and restoring HP Insight Software Central Management Server (Windows)white paper at Matrix OE Information Library for detailed backup and restore instructions.

Back up user-modified properties files before upgradeDuring an upgrade, Matrix OE properties files from an earlier release are replaced by newproperties files for the 7.0 release. If you made any modifications or customizations to the propertiesfiles, your changes are lost.Suggested actionBefore beginning an upgrade to Matrix OE 7.0, save a copy of user-modified properties files ornote any changes made to the files so that you can insert those modifications into the new versionsof the files.Following is a partial list of properties files that are commonly modified. Verify in your environmentif you have modified these files before beginning an upgrade, and if you have modified them,save the files to a location that is not under the ..\Program Files\HP\Virtual ServerEnvironment directory tree.C:\Program Files\HP\Virtual Server Environment\conf\esa.propertiesC:\Program Files\HP\Virtual Server Environment\conf\lsa\hp_lsa_service.confC:\Program Files\HP\Virtual Server Environment\conf\lsa\lsa.propertiesC:\Program Files\HP\Virtual Server Environment\conf\lsa\lsaclient.propertiesC:\Program Files\HP\Virtual Server Environment\conf\lsa\vc\vcController.propertiesC:\Program Files\HP\Virtual Server Environment\conf\lsa\vmm\vmmController.propertiesC:\Program Files\HP\Virtual Server Environment\conf\lsa\storage\storageController.properties

Installation 9

https://h20392.www2.hp.com/portal/swdepot/displayProductInfo.do?productNumber=HPID


After the upgrade has completed, but before the services are restarted or the CMS is rebooted,edit the new properties files and add any custom entries made to the previous version of the file.

Check logical server management upgrade log fileWhen Matrix OE logical server management is upgraded to 7.0, the pre-upgrade task completeseven if there are exceptions or missing objects.Suggested actionAfter the upgrade is complete, check for important upgrade messages at ../Program Files/HP/Virtual Server Environment/logs/vselog.lsm.log, and make any changes asnecessary.

Reconfigure deployment servers after upgrade from 6.0 and 6.1During an upgrade from Insight Dynamics 6.0 or 6.1 to Matrix OE 7.0, deployment serverconfiguration data is not migrated.Upgrades to Matrix OE 7.0 from Insight Dynamics 6.2 and 6.3 are not affected.Suggested actionAfter the upgrade is complete, reconfigure deployment servers (HP Server Automation, InsightControl server deployment, and Ignite-UX) before attempting to create a provisioning service ininfrastructure orchestration.

Changing the CMS passwordSystems Insight Manager includes CMS Reconfigure Tools that allow you to make commonreconfiguration changes to your CMS. One of these tools supports realigning the InsightManagement service account credentials with the CMS operating system password after theoperating system password is changed.In earlier releases, manual steps were required for the Insight Management components to recognizea CMS password change and properly operate, which are detailed in the Changing the InsightSoftware 6.0, 6.1 or 6.2 CMS password white paper. This white paper does not apply to MatrixOE 7.0.See the HP SIM 7.0 User Guide at www.hp.com/go/foundationmgmt/docs for information aboutusing the CMS Reconfigure Tools.

Enabling template access restrictionsHP Matrix Operating Environment infrastructure orchestration supports template access restrictionson a per user basis. This feature is disabled by default. If you do not modify the properties file, allusers will see all templates.To enable template access restrictions for users not in an organization, edit the C:\ProgramFiles\HP\Matrix infrastructure orchestration\conf\hpio.properties file,and set template.access.restricted=true. After access restrictions are enabled, onlyusers assigned to a template are able to see and use the template.

NOTE: Irrespective of this setting, all users in an organization have access to all templatesassigned to the organization. Template access controls for individual users in an organization arenot supported.

10 Description

www.hp.com/go/foundationmgmt/docs

Operations Orchestration Studio installation and upgrade issues

Pointing a remote Operations Orchestration Studio installation to Operations Orchestration Centralinstalled on a CMS

A “Failed Login” error message is displayed when you try to log in to Operations OrchestrationStudio 9.00 pointing to a remote repository (for example, Operations Orchestration Central 9.00on the CMS).To resolve this issue, perform the following steps:1. Make a backup of the rc_keystore file on the CMS.2. Copy the rc_keystore file installed on the CMS (C:\Program Files\HP\Operations

Orchestration\Central\conf\rc_keystore) to the same path on the remoteOperations Orchestration Studio system (C:\Program Files\HP\OperationsOrchestration\Studio\conf\rc_keystore).

3. Restart Operations Orchestration Studio.

Certificate key mismatch after Operations Orchestration Studio upgradeAfter upgrading Operations Orchestration Studio from version 7.51 to 9.0, Operations OrchestrationCentral and Studio have different certificate keys. Operations Orchestration Studio cannot accessthe URL https://localhost:16443.To resolve this issue, perform the following steps:1. Make a backup of the rc_keystore file on the CMS.2. Copy the rc_keystore file installed on the CMS (C:\Program Files\HP\Operations

Orchestration\Central\conf\rc_keystore) to the same path on the OperationsOrchestration Studio system (C:\Program Files\HP\Operations Orchestration\Studio\conf\rc_keystore).

3. Restart Operations Orchestration Studio.

Cannot download 64-bit Operations Orchestration Studio from infrastructure orchestration DesignerAn icon in the Workflow dialog box in infrastructure orchestration Designer allows you to downloadand install the 32-bit version of Operations Orchestration Studio. If you download and attempt toinstall the 32-bit version on a 64-bit CMS, the following error is displayed: “Studio version: 9.00does not match Central version: 2.0. Setup will exit.”If you have a 64-bit CMS, install the 64-bit version of Operations Orchestration Studio, which isavailable on the Insight Software DVD #2 in the /matrixio folder.

Red Hat Enterprise Linux 5.5 cannot be installed in VM guest on ESXi 4.1 host onBL465c G7 and BL685c G7

Red Hat Enterprise Linux 5.5 cannot be installed in a VM guest on an ESXi 4.1 host on BL465cG7 and BL685c G7 server blades.To resolve this issue, install RHEL 5.5 on the host system, install a RHEL 5.4 guest and then upgradethe guest to RHEL 5.5, ensuring that kernel-2.6.18-194.3.1.el5 or higher is used.For more information, see Red Hat’s Knowledgebase article at https://access.redhat.com/kb/docs/DOC-38013.

Managing Microsoft Windows servers without OpenSSH installedEven though the product documentation states OpenSSH is required, Matrix OE supports MicrosoftWindows environments not running OpenSSH. However, for Microsoft Windows systems withoutOpenSSH, there are several limitations with agent installation and troubleshooting.

Installation 11

https://access.redhat.com/kb/docs/DOC-38013

https://access.redhat.com/kb/docs/DOC-38013

For details on using Matrix OE with Microsoft Windows systems not running OpenSSH, see theHP Insight Dynamics 6.0 software without OpenSSH on HP ProLiant and Integrity Servers RunningMicrosoft Windows white paper, available at the following website:Matrix OE Information Library

Patch required to provision Hyper-V linked clone VM guests using HP ServerAutomation

If you are using HP Server Automation (SA) 9.0 or 9.10 on Red Hat Enterprise Linux 6 or SUSELinux Enterprise Server 10 platforms and you want to provision Hyper-V linked clone VM guests,install the following version-specific core patches and hotfixes for each SA version.For SA 9.0, download and install:• SA 9.05 core patch from http://support.openview.hp.com/selfsolve/patches

• “Linked Clone errors for RHEL6 and Above” hot fix from https://quixy.deu.hp.com/hotfix/index.php?HOTFIX_PROJECT=server_automation

For SA 9.1, download and install:• SA 9.11 core patch from http://support.openview.hp.com/selfsolve/patches

• “Linked Clone errors for RHEL6 and Above” hot fix from https://quixy.deu.hp.com/hotfix/index.php?HOTFIX_PROJECT=server_automation

Patch required for proper management of Microsoft Windows Server 2008 withHyper-V systems

A patch from Microsoft is required for managing Microsoft Windows Server 2008 with Hyper-Vsystems. Patches are installed on the Hyper-V hosts, and after you do so, you must install or updateHyper-V Integration Services from the updated host to each VM guest. Do this as a precautionbecause not all patches affect Integration Services on the guests.Microsoft Windows Server 2008 SP2 with Hyper-V requires you to install the following patch thatfixes the hang in host servers with many simultaneous virtual machine operations. The patch islocated at http://support.microsoft.com/kb/980081.See the Managing Microsoft Windows Server 2008 Hyper-V with HP Insight Software technologybrief for more information about other patches that might be required for later editions of MicrosoftWindows Server 2008 (R2):Matrix OE Information Library

VM kernel port name used for management traffic must be “Management Network”In VMware ESXi, the default name of the Virtual Machine (VM) kernel port used for managementtraffic is “Management Network”. Do not modify this name in the vCenter client user interface. Ifthis name is modified, the following will result.

• When trying to activate a logical server, the ESXi hosts appear as Rejected Target hosts withan error that the hosts are not part of vCenter, or that virtual machine management is unableto get vCenter information for the VM Host.

• The association between the host and vCenter does not appear in Systems Insight Manager.Suggested actionModify the Network Label of the VMKernel port used for management traffic to “ManagementNetwork” by performing the following steps.1. Log into the vCenter using the vCenter client.2. Select the Hosts and Clusters view from the menu View→Inventory.3. Select the ESXi server from the tree displayed in the left pane.

12 Description


http://support.openview.hp.com/selfsolve/patches

https://quixy.deu.hp.com/hotfix/d.php?P=server_automation&V=9.04&N=Linked+Clone+errors+for+RHEL6+and+Above


http://support.openview.hp.com/selfsolve/patches



http://support.microsoft.com/kb/980081


4. Select the Configuration tab from the right pane.5. From the Hardware section, select Networking.6. Click Properties for the Virtual Switch where the VMKernel port used for management traffic

is located. If there are multiple VMKernel ports located on multiple virtual switches andconfigured for management traffic, select the switch that has the VMKernel port that has theIP that is registered with HP Insight Control virtual machine management , or that will be usedfor registration with virtual machine management.

7. In the pop-up that displays the virtual switch properties, select the service console port andclick the Edit button.

8. Change the Network Label to Management Network and click OK. Close the virtual switchproperties pop-up window.

It can take a maximum of five minutes for the cache to be refreshed and the API to report the ESXihost.

Managing a Microsoft Hyper-V (or any MSCS) cluster when an alias is assigned byDHCP

In Windows Server 2008 Failover Clusters, if the cluster alias is assigned using DHCP, the clusteralias PTR (reverse lookup) record in DNS will be incorrect, and will cause the cluster to be incorrectlydiscovered in Systems Insight Manager.To correct this problem, delete the cluster objects from Systems Insight Manager, take the steps inthe Microsoft Knowledge Base article at http://support.microsoft.com/kb/969049 to modify thecluster DNS behavior, and then rediscover the cluster in Systems Insight Manager.

User experience and application performanceMatrix OE is based on web client server technologies. Your perception of performance can beaffected by many external sources, including the central management server’s health, networkcongestion and latency, and client side equipment and browser.Every effort to balance performance and functionality has been made to the default settings tocreate the best user experience under multiple browsers with different server environments. Changingthese default settings can result in adverse application responsiveness, which is more noticeablein some browsers; for example, there are known memory issues in Internet Explorer 7 and 8 thatcan be exposed in large scale environments. No matter which browser you choose to use, HPrecommends using the latest release and updates to improve your user experience and protectyour environment from the latest security threats.

Additional software dependenciesThe following are additional dependencies related to Matrix OE for Integrity software.• Instant Capacity Manager requires iCAP version 8.02 or higher installed on the HP-UX

managed systems.• Matrix OE for Integrity supports having multiple database instances hosted by Microsoft SQL

Server. Refer to the Microsoft SQL Server documentation for instructions on how to set up asystem to host multiple SQL Server database instances.

Installation 13

http://support.microsoft.com/kb/969049

2 Issues and suggested actionsThe issues and limitations in this release are listed in this chapter, organized by Matrix OperatingEnvironment feature or functional area. If a feature or functional area has no limitations or issues,that category will not appear. The following categories are used:Limitations Limitations of the implemented feature or functional areaMajor issues Issues that may significantly affect functionality and usabilityMinor issues Issues that may be noticeable but do not have a significant impact on

functionality or usability

Matrix Operating Environment software

Limitations

CMS identification must not be changedSeveral key attributes of the Central Management Server (CMS) operating system must not bemodified after Insight Management has been installed. The key attributes are:

• Hostname

• IP address associated with the hostname

• Full name or domain (including the fully qualified domain name)

• DNS entries corresponding to the hostname and domainIf any of these attributes are changed after Insight Management has been installed, many featuresof the solution will fail to function properly.Suggested actionVerify that the identified attributes are correct before installation. If any of these attributes requiresmodification after installation, you will need to install a new CMS.Optionally, if only the hostname or IP address of the CMS needs to be modified, you can considerusing the Data Migration Tool to migrate existing data onto a newly installed CMS. For moreinformation, see the Data migration of an existing Microsoft Windows CMS to a new InsightManagement system white paper available from the Matrix OE Information Library.

Configuring Matrix OE with a remote Microsoft SQL Server databaseBecause Windows Single Sign On is used by Matrix OE to connect to the SQL Server database,the credentials associated with the user account executing a command or service are used toconnect to the remote database. Therefore, user accounts used to execute Matrix OE software thataccesses the database must satisfy the following conditions:

• The user credentials specified for the CMS access and database access during the installationshould be the same, and be trusted domain credentials available to both the CMS and databasesystems.

• The user credentials specified during the installation for the database access should haveaccess permissions to the database.

If you run a CLI command as a user that is not a domain user or does not have database accesspermission on the remote database, and the CLI command requires database access, the commandwill fail.

Matrix Operating Environment software 15


Capacity Advisor

Limitations

Automatic power calibration of Integrity Linux systems running SUSE 11 operating system is notsupported

The Matrix OE software does not support the automatic power calibration of Integrity Linux systemsrunning the SUSE 11 operating system.Suggested actionManually calibrate Integrity Linux SUSE 11 systems for power.

Automatic power calibration requires power metering support in serverAuto-calibration of idle/max power with Capacity Advisor requires the power metering featuresof iLO in the managed server. For information about which servers support power metering, seeAppendix A in the HP Insight Management Support Matrix at Matrix OE Information Library.Suggested actionManual power calibration is recommended for servers that do not support iLO power metering.

Sub-OS workloads on Linux systems are not supportedOn x86 and Integrity Linux systems, Linux sub-OS workloads cannot be created or modeled, andsub-OS workload data cannot be collected.On Linux systems where the Utilization Provider was previously installed and sub-OS workloadswere defined, those sub-OS workloads will no longer be displayed in Matrix OE visualization.The definition of these sub-OS workloads on a Linux system will be removed from Matrix OEvisualization by the upgrade process because they are no longer used by Matrix OE GlobalWorkload Manager (gWLM), nor supported by Capacity Advisor or Matrix OE visualization.

Minor issues

Socket and core count inaccurate on Microsoft Windows 2003 and HP-UX 11 v2 (IA andPA) systemsWindows 2003 and HP-UX 11 v2 (IA and PA platforms) operating systems may not correctly reportthe number of processor sockets in a multi-core/hyperthreading environment.Suggested actionThe following patches are required to correct this issue.

• For Windows 2003, apply the KB932370 patch. For more information, see http://support.microsoft.com/kb/932370/.

• For HP-UX 11 v2, apply the PHKL_37803 patch. You can download this patch from http://www.hp.com/go/hpsc. Select Patch management from the Download options menu, thentype PHKL_37803 in the search box. After downloading, execute the command shPHKL_37803, and install the depot file that is generated.

No patch is necessary for Windows 2008 or HP-UX 11 v3.This information does not apply to HP-UX 11 v1.

Infrastructure orchestrationInfrastructure orchestration runs on an x86/x64 Windows CMS.

16 Issues and suggested actions


http://support.microsoft.com/kb/932370/

http://support.microsoft.com/kb/932370/

http://www.hp.com/go/hpsc


Limitations

Cross-technology logical servers are not supported by infrastructure orchestrationCross-technology logical servers are not supported by Matrix OE infrastructure orchestration. Onlylogical servers created or imported in Matrix OE logical server management, using Create→LogicalServer or Tools→Logical Servers→Import, can be moved from physical to virtual or virtual tophysical. A logical server created in infrastructure orchestration (by deploying a service template)cannot be managed as a cross-technology logical server. For example, if a physical logical serveris created in infrastructure orchestration, it cannot be moved to a virtual machine using logicalserver management. Similarly, if a virtual logical server is created in infrastructure orchestration,it cannot be moved to a server with Virtual Connect using logical server management.

Number of networks provisioned to a logical serverThe Matrix OE infrastructure orchestration Designer does not restrict the number of networks thatcan be connected to a logical server group. In practice, only four networks can be connected tovirtual logical servers, and 128 networks can be connected to physical logical servers designedto run on HP c-Class server blades. These practical limits will increase over time as Virtual Connectand hypervisors are enhanced.If these limits are exceeded in an infrastructure template, provisioning operations using that templatewill fail.Suggested actionLimit the number of network connections to a logical server group to four for virtual logical serversand 128 for physical logical servers.

Manually configure the SAN boot path for ProLiant DL/ML servers before provisioningThe job "Deploy ProLiant System Configuration (BL BFS),” contained in the Insight Control RapidDeployment software job folder deployed using infrastructure orchestration, cannot be used toconfigure ProLiant DL/ML target servers. The job configures ProLiant/Integrity BL target servers. Ifyou attempt to configure ProLiant DL/ML servers using this job, the request will fail.Suggested actionManually configure the SAN boot path before you use infrastructure orchestration to provision aProLiant DL/ML target server on SAN boot LUN, as follows.1. Create a server deployment job folder that does not contain the "Deploy ProLiant System

Configuration (BL BFS)" job.2. Change the Boot Controller Order (from ROM-Based Setup Utility) for the ProLiant DL/ML

server to select Fibre Channel port (HBA) as the first in the boot order.3. Use the HBA vendor's utility to configure the HBA's boot path to boot from the desired SAN

boot LUN.4. Power off the target server after the Boot Controller Order and the HBA's boot path have been

configured correctly.

Required HBA firmware and OS drivers to access multi-initiator N-Port ID Virtualization (NPIV) datadisks

The HP blade multi-initiator NPIV feature allows you to configure multiple WWNs for a single HBAconnected with a Virtual Connect Fibre Channel module. Although Matrix OE automates theconfiguration, certain versions of HBA firmware and drivers are required to access multi-initiatorNPIV data disks configured for any target BL server.Suggested actionThe HBA firmware and drivers shown for the following operating systems are required to supportmulti-initiator NPIV data disks.

Infrastructure orchestration 17

• QLogic HBA (BIOS version 2.07 or higher)RHEL 5.3 and higherSLES 11 and higherSLES 10 SP3 and higherWindows 2003/2008: QLogic driver 9.1.7.29 or higher

• Emulex HBA (BIOS version 3.11a3 or higher)RHEL 5.4 or higherSLES 10 SP3 or higherSLES 11 in-box driver not yet available. Requires out-of-box Emulex driver version: 8.2.8.32Windows 2003/2008: Emulex driver 5.00.17.06-2 or higher

Multipath I/O and Emulex FC HBA or Emulex FlexFabric Converged Network Adapter is notsupported on Windows Server 2008 and Windows Server 2008 R2

Using multipath I/O during OS installation is not supported with an Emulex FC HBA or an EmulexFlexFabric Converged Network Adapter (CNA) on any Matrix-supported server on Windows Server2008 and Windows Server 2008 R2.The following suggested actions enable you to install the OS single-path and then modify thestorage definitions to create multiple-path access to the boot volume. After a successful OSinstallation, these instructions will define an additional physical path to the boot volume by:

• Modifying the storage definition to add an additional server HBA definition (port)

• Modifying the boot volume definition to add a second physical path to its definitionSuggested action 1If you are creating a template using Matrix OE infrastructure orchestration designer:1. Do not select Redundant SAN paths to disk when configuring storage.

With this box unchecked, infrastructure orchestration will select a storage pool entry with asingle path, or will create a single-path storage pool entry during storage auto-generation.If you are using SPM with pre-existing volumes or with on-demand provisioning, SPM will fulfillthe storage request with a volume presented over a single path and will only zone one path.

2. (Optional) Modify the storage pool entry to add on additional path. SPM will make theappropriate adjustments.

Suggested action 2If you are defining a storage pool entry in Matrix OE logical server management:1. Define the storage pool entry, adding multiple ports, and marking all non-boot disks as

redundant. Keep the boot volume as a single path.2. Assign the storage pool entry to a logical server.3. Activate the logical server.4. Provision the OS, and install the multipath IO drivers as appropriate.5. Deactivate the logical server.6. Modify the storage pool entry to make the boot volume redundant.7. Reactivate the logical server.

Multi-initiator NPIV support on Integrity bladesMulti-initiator NPIV features used by Matrix OE and Virtual Connect Enterprise Manager aresupported on Integrity blades using the QLogic FC controller with the HP-UX 11i v3 Update 8(March 2011) and later release. Multi-initiator NPIV features are not supported on Integrity bladeswith Emulex FC controllers.


Suggested action 1To enable support for physical provisioning with NPIV enabled on HP-UX 11i v3 Integrity bladeswith QLogic FC controllers, set the attribute npiv.integrity.enabled=true in the C:\Program Files\HP\Matrix infrastructure orchestration\conf\hpio.propertiesfile.On earlier versions of HP-UX, or if the environment is mixed (QLogic and Emulex controllers), theattribute npiv.integrity.enabled=false (the default) should remain unchanged in thehpio.properties file.Suggested action 2On Integrity blades with Emulex controllers and/or versions of HP-UX earlier than 11i v3 Update8, create a single storage pool entry that combines a boot volume and the data volumes intendedto be allocated to a single physical server. You can do this in one of the following ways:

• Pre-provision any number of these composite storage pool entries to enable fully automatedallocation of SAN volumes to physical servers.

• Pre-provision storage pool entries with only boot volumes. An automated provisioning requestpauses after the operating system is deployed if the infrastructure service template specifiesdata disks. You can modify the storage pool entry allocated to the server to add data volumes,and then continue the request.

Required software versions for FlexFabric Integrity physical provisioning supportMatrix OE infrastructure orchestration requires the following minimum software versions to supportHP FlexFabric for Integrity physical provisioning:• Ignite-UX C.7.12.475 or higher

• HP-UX 11i v3 Update 8 (March 2011) or higher

Install Microsoft Sysprep tool for customization of Windows 2003 guests on Hyper-V after installationThe Microsoft System Preparation (Sysprep) tool, which is required for customization of Windows2003 guests on Microsoft Hyper-V, is not retained on the CMS after upgrading from previousversions of Insight Software. Without these tools, customization of Windows 2003 guests onHyper-V results in failures.Installation of the Sysprep tool is not required for customization of Windows 2008 and higher VMguests.Suggested actionTo enable customization of Windows 2003 guests on Hyper-V, install the Microsoft Sysprep toolon the CMS in the folder C:\Program Files\HP\Insight Control virtual machinemanagement\Sysprep\2003 after the installation or upgrade of Insight Management 7.0.

Cannot power on a virtual server during provisioning if VM hardware version is 4 and a templatecontains more than three NICs

A request to power on an ESX virtual server may fail if the ESX VM hardware version is set to 4and the virtual machine contains more than three NICs. Matrix OE infrastructure orchestrationprovisions virtual machines using ESX hardware version 4 by default, which allows a maximumof five PCI Controller devices (two SCSI controllers and three NICs) attached to a virtual machine.ESX hardware version 7 allows four SCSI controllers and ten NICs.Suggested actionUse an ESX native template with a hardware version of 7, or reduce the number of NICs in theinfrastructure orchestration template to three or fewer.


Physical disks with multiple partitions are unsupported unless all but the first partition are excludedIf your environment contains a physical disk with multiple partitions (volumes with the same diskidentifier) and you want to provision virtual machines on that disk, exclude all partitions exceptthe first partition on the disk. A virtual machine created on a physical disk with two or more partitionswithout this exclusion is unsupported, and the service could fail to deploy with the error “Importfailed for logical-server. VM Host datastore not found.”Suggested actionExclude all partitions except the first partition on the disk in the “VOLUMES TO EXCLUDE” sectionof the hpio.properties file.For example, if there are three volumes (C:, D:, and E:) on a physical disk, exclude the D: and E:volumes to provision a logical server on the virtual machine created on the C: volume of the physicaldisk.

Cluster shared volume does not appear after Microsoft Hyper-V VM provisioning with high availabilityenabled

If you use infrastructure orchestration to create and deploy a virtual provisioning template on aMicrosoft Hyper-V VM Host with the High Availability (HA) option enabled, and then view thevirtual machine in the Failover Cluster Manager on the Hyper-V host, the Cluster Shared Volume(CSV) is not displayed in the provisioned VM information.However, if you manually deploy the virtual machine using the Microsoft Hyper-V Manager on theHyper-V VM Host, the Failover Cluster Manager correctly displays the disk resource and the virtualmachine resources created.Suggested actionNo action is needed; the highly available virtual machine disk was created correctly on the CSV.

Cluster shared volume is not correctly recognized unless at least two VM Hosts managed by IOshare the volume

When only one VM Host in a Hyper-V cluster is managed by infrastructure orchestration, IO treatsa cluster shared volume (CSV) as a cluster disk, and allocates only one VM to the disk. (Normallymultiple VMs can be allocated to a CSV, but only one VM can be allocated to a cluster disk. )When this occurs, the following error is displayed: “Provisioning request has paused. Modifystorage volumes on VM Hosts to satisfy the logical disk requirements. An IO Administrator hasbeen notified.”Suggested actionAdd at least two VM Hosts that share the CSV to Matrix OE, and register them with Insight Controlvirtual machine management as VM Hosts. The second VM Host can be left in the Unassignedpool, or it can be used as a regular VM Host within IO.

Cannot provision a server group on Microsoft Hyper-V R1 with linked clone and high availabilityenabled

If you attempt to provision a template containing multiple High Availability (HA)-enabled linkedclones on a Hyper-V R1 server, or you attempt to add a server to a server group with a singleHA-enabled linked clone on a Hyper-V R1 cluster disk, the request pauses indefinitely for storageprovisioning. A linked clone group must be provisioned on a single datastore, but only one HAvirtual machine can be provisioned to a Hyper-V cluster disk.


Suggested action 11. Cancel the Create or Add Server request.2. Modify the template to remove the HA requirement, or

Modify the template to deselect linked clones and add a cluster disk for each virtual machine.

3. Resubmit the Create request, orFor the Add Server request, delete the existing service and recreate it with the updated template.Take care to migrate data from the existing service to the new service before deleting theexisting service.

Suggested action 2Upgrade to Microsoft Hyper-V R2, and specify a Cluster Shared Volume (CSV) instead of a clusterdisk in the template.

Microsoft SCVMM can be registered on one CMSAs part of the registration of Microsoft System Center Virtual Machine Manager (SCVMM), theweb service component is copied from the CMS to the SCVMM host, which enables communicationbetween the SCVMM server and the CMS. As a result, SCVMM can be registered on only oneCMS. This means that in a federated CMS configuration, each CMS requires a different SCVMMhost to manage Hyper-V hosts through SCVMM.

Browsing to user interface requires default Systems Insight Manager portHP Systems Insight Manager is installed with a default port of 50000 in the JBoss container.Browsing to the infrastructure orchestration user interfaces, for example, Designer, depends on thisport value.Suggested actionDo not modify the Systems Insight Manager default port of 50000 in the JBoss container.

Operations Orchestration Central and Studio contents are not localized to JapaneseIn new installations of Matrix OE 7.0, Operations Orchestration Central and Studio screens, menus,and buttons are localized to Japanese, but the contents (such as workflows and system properties)are not localized to Japanese when OO is installed in a Japanese environment.If you are upgrading to Matrix OE 7.0, Operations Orchestration Central and Studio display thelocalization behavior of the version that was originally installed. Therefore, if you are upgradingfrom Insight Dynamics 6.3 to Matrix OE 7.0, the Operations Orchestration Central and Studioscreens are displayed in English.

Major issues

Connection to the shared storage for a server may not occur when adding a server to a VM clusterserver group

Adding a server to a VM cluster server group when the shared storage pool entry does not havean available connection (“sharer”), prevents proper connection to the shared storage for that server.The storage pool entry does not automatically increase the number of available connections whena server is added. When there are no connections available, the newly created VM Host will beassigned a WWN for each HBA at initial activation time. If the storage pool entry is later modifiedto allow more sharers, the added WWNs will not match the WWNs originally assigned to theVM Host at initial activation. In this case, both the original WWNs and the newly added WWNswill be associated with the new server and be put into the VM Host’s VC profile at next activation.


Suggested action 1Before adding servers to a VM cluster server group, ensure the Maximum number of sharersis greater than or equal to the maximum number of servers (after the server is added) in the VMcluster server group for the storage pool entry. To change this value:1. Select the Matrix OE visualization logical server management Modify→Logical Server Storage

Pools screen and click Modify for the storage pool entry.2. Increase the value of Maximum Number of Sharers and click Update Maximum Sharers,

then save the storage pool entry.Suggested action 2If the server was added without an available connection:1. Select the Matrix OE visualization logical server management Modify→Logical Server Storage

Pools screen, and click Modify for the storage pool entry.2. Increase the value of Maximum Number of Sharers and click Update Maximum Sharers.3. In the Manage Server WWNs table at the bottom of the screen, check Ready for the newly

added server WWN and save the storage pool entry.4. Deactivate the logical server.5. Reactivate the logical server.The Virtual Connect profile will then contain the correct server WWN with the logical server's datastorage pool entry.

Minor issues

Delete service request leaves "clean-me" logical servers when used with Integrity serversif disk is not erased before delete service requestAfter successfully provisioning a physical Integrity server blade (using Ignite-UX, or through asecondary CMS in a federated CMS environment), there are manual steps you must perform toallow the disk to be erased during de-provisioning. Erasing the disk is part of a successfulde-provisioning (delete request) for an Integrity service.If the disk is not erased, the delete service request may fail and “clean me” logical servers willremain. In that case, a manual cleanup procedure is required. (See the Suggested action for thecleanup procedure.)Before you start a delete service request from infrastructure orchestration:1. Make sure that the Integrity service was successfully provisioned (the Integrity service is running).2. Discover the Integrity blade where the Integrity service is running using its IP address from

Systems Insight Manager. In a federated CMS environment, discover the Integrity blade onthe primary CMS using its IP address from Systems Insight Manager. The Integrity node'sidentification information should contain the UUID and UUID (logical). You can discover theblade in either of the following ways:a. Use the Systems Insight Manager menu selection Options→Discovery..., orb. Use the Systems Insight Manager command: mxnode -a <ip-address> -w

where <ip-address> is the IP address that was assigned to the Integrity blade afterprovisioning.Add the root login of the Integrity blade using the Systems Insight Manager menu selectionOptions→Security→Credentials→System Credentials....

3. Set up an SSH connection between the CMS and the Integrity blade by entering the command:mxagentconfig -a -n <ip-address> -u root -p <root password>

4. If the IP address was discovered on the primary CMS in a federated CMS environment (instep 2 above), manually delete the Systems Insight Manager node.Use the Systems Insight Manager command: mxnode -r <ip-address>


Suggested actionIf you did not perform the preceding steps before you started a delete request for the Integrityservice, the automatic delete service will not succeed completely. In that case, you must clean upthe logical server associated with the Integrity blade and manually erase the disk.1. Power on the server blade. (The blade was powered off when delete service request completed

or failed.)2. Perform steps 2 and 3 in the preceding section, if you did not do so before starting the delete

service request.3. To erase the disk, enter the following command from the CMS where the discovery was

executed (in step 2 in the preceding section):mxexec -t "Ignite Erase Disks" -n <ip-address>

4. If the IP address was discovered on the primary CMS in a federated CMS environment (instep 2 in the preceding section), manually delete the Systems Insight Manager node.Use the Systems Insight Manager command: mxnode -r <ip-address>

Service creation fails for physical requests when Windows Server OS is selectedMatrix OE infrastructure orchestration service creation fails when a physical template with WindowsServer OS is selected. This failure occurs when infrastructure orchestration times out waiting forInsight Control server deployment to complete installation of Insight Management Agents on thetarget server.Suggested actionWhen a request fails, infrastructure orchestration automatically retries the request and a new serveris provisioned with the requested service. The original server is moved to the maintenance pooland can be cleaned and moved back to a valid server pool. If the failure is observed frequently,the following steps can be followed to disable the installation of Insight Management WBEMproviders. Insight Management SNMP agents continue to provide management capabilities whenWBEM providers are not installed.1. Edit the install.cmd file, found under the <Insight Software installation>\rdp\

deployment server\hpfeatures\hpagents-ws\components folder.2. Remove the following two lines:

call :installsc wbem%PROCESSOR_ARCHITECTURE%if %cpresult% geq 1 set /a cpresults=cpresults+1

“Any Of” and “All Of” requirements in SPM are not displayed correctly in infrastructureorchestration designerUsing SPM, you can define complex requirements with nested “Any of” (or) and “All of” (and)constructs. These requirements are applied when provisioning storage in SPM, but infrastructureorchestration designer does not display these requirements correctly. The IO template architectmay be allowed to enter values when defining storage that will cause storage provisioning to fail.Suggested actionWhen using a specific SPM storage template in infrastructure orchestration designer, be aware ofthe storage template's Capacity, RAID, and Tag requirements. Do not specify values that conflictwith these requirements.

Provisioning a service may fail when storage volumes in a logical server group containdifferent tagsIn an IO template, different storage tags can be assigned to the individual volumes in a logicalserver group. When storage is auto-generated, each volume is provisioned based on a number


of defined requirements, including these tags. Within a logical server group, if the private datadisk(s) contain different tags than the boot disk, the service will fail to provision.Suggested actionUse the same storage tags for the boot disk and private data disks in a logical server group. Ifseparate tags are needed, separate storage pool entries can be manually defined for the bootdisk and private data disks with those separate tags.

Tags assigned to storage pools in SPM do not appear in infrastructure orchestration designerStorage tags that are assigned to volumes in SPM are correctly displayed in the tag list ininfrastructure orchestration designer. If a tag is assigned in SPM to a storage pool, but not a volume,it is not displayed.Suggested actionTo use storage pool tags to specify storage resources, manually enter the tags assigned to storagepools in the Additional Tags field in infrastructure orchestration designer.

Resources are not displayed if an organization is selected during startupWhen the Matrix OE infrastructure orchestration service is started or restarted, the startup periodlasts between 90 and 120 seconds in a small CMS, and longer in a large CMS. If an organizationis selected on the Organization tab during this startup period, the following information is neverdisplayed: Compute Servers for the service provider, Compute Servers for the organization, andAssigned Networks for the organization.Suggested actionRestart the Matrix OE infrastructure orchestration console by selecting Tools→Infrastructureorchestration, and wait until startup completes before selecting an organization on the Organizationtab.

IO service status is not updated when a VM Host is disconnected and reconnectedWhen a logical server is active on a VM Host and the VM Host is disconnected or powered off,the status of the infrastructure orchestration service incorrectly remains Up. When the VM Host isreconnected, the service incorrectly remains Down.Suggested actionAfter a VM Host is disconnected or powered off:1. From a command prompt on the CMS, run the following command to check the VM Host

service status and to refresh its status:cli getVmHostDetails –host HOST-IP-ADDRESS

The VM Host should display connected=false. (If connected=true is displayed, runthe command again until connected=false is displayed.)

2. Manually refresh server resources in Matrix OE visualization by selecting Tools→LogicalServers→Refresh....After this refresh, the status of the Matrix infrastructure orchestration service on the ConsoleServices tab will be Down and the status of the logical server will be Inoperable.

After a subsequent VM Host reconnection:1. Perform steps 1 and 2 above to check the VM Host service status and to refresh its status.Suggested actionDelete the IO service using the infrastructure orchestration console Services tab.


Set retry parameters for BL620c G7 and BL680c G7 bladesMatrix OE infrastructure orchestration may time out while waiting for the server to be recognizeddue to long server post times, particularly when using the BL620c G7 or the BL680c G7.Suggested actionTo avoid this issue on any server:1. Edit C:\Program Files\HP\Matrix infrastructure orchestration\conf\

hpio.properties and set the retry.count.ares.pxe parameter to a value greaterthan 30.

2. Restart the HP Matrix infrastructure orchestration service.

After upgrade, ioexec commands require installation pathIn Matrix OE 7.0, the infrastructure orchestration Windows environment default installation pathwas changed from C:\Program Files\HP\Insight Orchestration to C:\ProgramFiles\HP\Matrix infrastructure orchestration. In a new installation of Matrix OE7.0, the bin folder for IO is added to the DOS path, enabling you to run ioexec commandsfrom a command prompt in any folder on the CMS. However, after an upgrade from InsightDynamics 6.3 to Matrix OE 7.0, ioexec commands can be run only from the C:\ProgramFiles\HP\Matrix infrastructure orchestration\bin folder.Suggested action 1Manually enter C:\Program Files\HP\Matrix infrastructure orchestration\binon the command line before entering any ioexec commands.Suggested action 2Reboot the CMS to automatically update the IO bin folder to the DOS path.

Templates containing Unicode characters are not imported correctly using ioexecImporting an infrastructure orchestration service template using the ioexec import templatecommand causes any Unicode characters in the template to display improperly.Suggested actionImport the template using the Import button in infrastructure orchestration designer.

Enabling telnet for Internet Explorer 7 or laterMicrosoft removed support for telnet in Internet Explorer 7, as described in Release Notes forInternet Explorer 7.Suggested actionTo continue to use telnet to access deployed servers, perform the following steps.1. Copy the following registry file contents into a file called telnet.reg and save the file.

Windows Registry Editor Version 5.00[HKEY_LOCAL_MACHINE\SOFTWARE\Microsoft\Internet Explorer\Main\FeatureControl\FEATURE_DISABLE_TELNET_PROTOCOL] "iexplore.exe"=dword:00000000

2. Execute the file by double-clicking it, which enables IE execution of the telnet protocol on theclient.

3. Restart IE if it is already running.


http://msdn.microsoft.com/en-us/ie/aa740486.aspx

http://msdn.microsoft.com/en-us/ie/aa740486.aspx

Matrix OE infrastructure orchestration and Operations Orchestration issues when browseris set to JapaneseThe following issues may occur if your browser language is set to Japanese.

• If you specify Japanese characters for the service name of a provisioning task, the logicalserver is not provisioned. (See Suggested action 1.)

• If you create a new virtual provisioning template and configure mail settings, the “Approveor Reject” e-mail contains a garbled service name. (See Suggested action 2.)

• When you create a service, the “Unassigned” and “Maintenance” server pools are availablefor selection. However, these server pools are special pools that are used to hold servers notready to be used for provisioning, and are incorrectly displayed as choices in the list ofavailable server pools. Selecting these pools will cause the create service request to fail. (SeeSuggested action 3.)

Suggested action 1Specify a service name containing valid characters ([a-z] [A-Z] [0-9] [_] [-] [space]).Suggested action 2Use the infrastructure orchestration console to approve or reject requests that contain Japanesecharacters.Suggested action 3Select a server from a server pool other than “Unassigned” or “Maintenance” for use during acreate request.

Insight managed system setup wizard

Limitations

Privilege elevation configuration that requires a password by the privilege elevation tool is notsupported

Insight managed system setup wizard supports privilege elevation only on a CMS that has privilegeelevation configured so that no password is required by the privilege elevation tool (such as suand sudo). If you attempt to configure features when privilege elevation is turned on and requiresa password, you may see failures in the wizard's execution output.Suggested actionIf a feature fails to configure correctly because of privilege elevation password requirements,configure the feature outside of the managed system setup wizard.To turn off the requirement for a password, use the Systems Insight ManagerOptions→Security→Privilege Elevation menu, and uncheck the check box labeled “A passwordis required for this privilege elevation tool.” Note that systems managed by the CMS must bereconfigured to reflect this configuration change.

Patch required to register HP Integrity VM Hosts running Integrity VM 4.1 or higher and 11.31HA-OE or DC-OE

The Insight managed system setup wizard fails to register HP Integrity VM Hosts running IntegrityVM 4.1 or higher and either the 11.31 High Availability Operating Environment (HA-OE) or the11.31 Data Center Operating Environment (DC-OE).Suggested actionApply the patch PHSS_40784 to the HP-UX 11i v3 Integrity VM Host. The patch is needed onsystems where Serviceguard is installed, but the systems are not actively in a Serviceguard cluster.Serviceguard is included with the HP-UX 11i v3 HA-OE and DC-OE.


You can download this patch from http://www.hp.com/go/hpsc. Log on with your HP passport,and select Patch management from the Download options menu. Type PHSS_40784 in the searchbox.

HP Integrated Lights-Out 3 (iLO3) supportHP Integrated Lights-Out 3 (iLO3) is supported in the current release of Insight managed systemsetup wizard. However, iLO3 for ProLiant requires firmware version 1.05 or higher.Suggested actionUpdate iLO3 ProLiant firmware to version 1.05 or higher.

HP Integrity rx2620 is not supportedIn this release, using the Insight managed system setup wizard to configure an HP Integrity rx2620server is not supported.Suggested actionManually configure HP Integrity rx2620 servers.

Instant Capacity Manager

Limitations

Commands executed under privilege elevation may cause Instant Capacity Manager to monitorbut not control managed nodes

Some HP-UX security settings cause output to be generated every time a user is logged in. This canoccur, for example when sudo is used in Systems Insight Manager privilege elevation mode toexecute commands on HP-UX managed nodes, which can generate apparently extraneous outputto stderr. This output may be interpreted by some Instant Capacity (iCAP) Manager functionsexecuted under privilege elevation as error output from those commands.When this apparent error output occurs, the affected iCAP Manager functions fail. In this scenario,iCAP Manager can be used only to monitor, and not control, managed nodes configured in thisway.

NOTE: HP Integrity Superdome 2 systems obtain monitoring data by executing commands usingsudo. Monitoring data may not be available in this scenario.

Suggested actionConfigure HP-UX managed nodes so that no extraneous output is generated when the sudocommand is executed.For more information, see the following HP-UX documentation:

• HP-UX 11i Security Containment Administrator’s Guide

• HP-UX System Administrator's Guide: Configuration Management (HP-UX 11i v3)

• HP-UX Servers and Workstations – HP-UX Standard Mode Security Extensions (HP-UX 11i v2)

• the manpage man 4 security (particularly the discussion of DISPLAY_LAST_LOGIN)

Logical server managementLogical server management runs on an x86/x64 Windows CMS.

Instant Capacity Manager 27


http://h20000.www2.hp.com/bizsupport/TechSupport/DocumentIndex.jsp?contentType=SupportManual&lang=en&cc=us&docIndexId=64179&taskId=101&prodTypeId=18964&prodSeriesId=3367813

Limitations

Restrictions on using cross-technology logical serversCross-technology logical servers can be moved between ProLiant server blades with Virtual Connectto ESX and back, between Integrity server blades with Virtual Connect and Integrity VM and back,and between servers with Virtual Connect with unlike configurations.For more information, see the HP Matrix Operating Environment Logical Server Management UserGuide and the HP Matrix Operating Environment Recovery Management User Guide at the MatrixOE Information Library.

• Valid targets for activating or moving a cross-technology logical server:

ESX VM Hosts◦◦ HP Integrity VM Hosts (racks and server blades)

◦ ProLiant c-Class servers with Virtual Connect

◦ HP Integrity servers with Virtual Connect

• ESX storage requirements:

SAN volume presented to an ESX VM Host (RDM disk) that can be added to a VirtualConnect profile

◦

◦ One datastore shared on a SAN or iSCSI volume (to store the Insight Control virtualmachine management configuration file and RDM mapping file [.vmdk] on the ESX VMHost)Not supported: SPM (catalog) storage volumesIn this release, catalog storage pool entries are supported on servers with Virtual Connectlogical servers only. Catalog storage pool entries cannot be used with cross-technologylogical servers that may be activated on an ESX or Integrity VM Host.

• Integrity VM storage requirements:

◦ SAN-based logical server storage is supported, including storage pool entries. File basedstorage is not supported with Integrity VM cross-technology logical servers.

NOTE: SAN volumes used for Integrity VM cross-technology logical servers should notbe presented to the Integrity VM Host.

• ESX network requirements:

◦ Single or multiple networks (limited to 4 by ESX)Not supported: VLAN networks

• Integrity VM network requirements:

◦ Single or multiple networksNot supported: VLAN networks and Auto-Port Aggregation Link Aggregation configurations

Performing operations on a logical server running as a VM HostBefore performing any operations on a physical server running as a VM Host, make sure that alljobs related to controlling the VM guests running on the VM Host logical server have beencompleted. (A warning message is displayed only when you attempt to deactivate or delete virtualmachines registered with Insight Control virtual machine management.)




Refer to the following table for the appropriate steps to prevent VM guests and VM logical serversfrom becoming inoperable.

Steps to perform for the VM Host after performing thesteps on the VM guests

Steps to perform on VM guestsVM Host operation youwant to perform

1. Unregister the VM Host from Insight Control virtualmachine management. From HP SIM, selectConfigure→Virtual Machine→Unregister VirtualMachine Host…

2. Move the VM Host logical server. SelectTools→Logical Servers→Move.

3. Register the moved VM Host using Insight Controlvirtual machine management. From HP SIM, selectConfigure→Virtual Machine→Register VirtualMachine Host…

Power off or perform a gracefulshutdown of all VM guests

Move VC profile toanother blade

Shut down the VM Host logical server.Power off or perform a gracefulshutdown of all VM guests

Shut down VM Host

1. Perform a graceful shutdown of the VM Host.2. Power off the VC blade where the VM Host logical

server resides. Select Tools→LogicalServers→Power Off....


Power off the VC blade


2. Delete the VM Host logical server. SelectTools→Logical Servers→Delete.

• If you no longer need the VMguests, delete them. Otherwise,move all VM guests to a suitablehost.

• If you have VM guests that arenot logical servers, then use theappropriate tools to relocate theguests.

• If you have any inactive VMlogical servers, reactivate themand then move them to a suitablehost.

Delete the VM Hostlogical server

Deactivate the VM Host logical server. SelectTools→Logical Servers→Deactivate.


Deactivate or unassign theVC profile


2. Activate the VM Host logical server. SelectTools→Logical Servers→Activate.

3. Register the VM Host using Insight Control virtualmachine management. From HP SIM, selectConfigure→Virtual Machine→Register VirtualMachine Host…


Activate or reassign theVC profile to a differentblade

Moving a logical server from a Microsoft Hyper-V R2 host to a Hyper-V R1 host is not supportedA limitation from Microsoft prevents you from moving or reactivating a virtual machine logicalserver from a Microsoft Hyper-V R2 host to a Hyper-V R1 host. An error message is displayed ifyou attempt the operation.You can move or reactivate a virtual machine logical server from a Hyper-V R1 host to a Hyper-VR2 host.

Logical server management 29

NOTE: The Hyper-V host version is shown in the Platform column in the Available Target Hoststable on the activate or move logical server target screen. The host is a Hyper-V R1 host if the hostversion begins with 6.0 (for example, 6.0.6001 or 6.0.6002 in Windows 2008 Server – Hyper-VSP1 or Windows 2008 Server – Hyper-V SP2). The host is a Hyper-V R2 host if the host versionbegins 6.1 (for example, 6.1.7077 in Windows Server 2008 R2).

Suggested actionBe certain that you want to move or reactivate the logical server from Microsoft Hyper-V R1 to R2before you complete the operation. After the operation completes, you cannot move or reactivatethe logical server back to Hyper-V R1.

Multiple datastores are not supportedAll storage volumes for a virtual machine logical server must be part of the same datastore. (Toimport a virtual machine, the virtual machine must be using a single datastore.) File volumes arecreated on the same disk as the virtual machine's configuration file.

Virtual machine datastore size and configuration filename limitsWhen creating a virtual machine logical server (ESX, Hyper-V, or Integrity VM), the maximumdatastore size that you can specify is 2048 TB (2097151 GB). Additionally, the maximum numberof allowed characters for a virtual machine configuration filename is 255. If you specify a sizelarger than the maximum, a warning message is logged and the entry is ignored.VM technologies have individual configuration limits that may limit the VM disk size.

Logical server management supports virtual machines created only on the first partition of a physicaldisk

If a physical disk has multiple partitions (volumes with the same disk identifier), only the first volumeis recognized and supported by logical server management. For example, if a disk has volumesC:, D:, and E:, you can create and import a virtual machine logical server created only on the C:volume.

Importing a Microsoft Hyper-V virtual machine with configuration file in the default location is notsupported

A Hyper-V virtual machine whose configuration file is in the default location on the hard disk in\ProgramData\Microsoft\Windows\Hyper-V is not supported by logical server management.You cannot import a virtual machine guest as a logical server if the virtual machine's configurationfile is in this location. There is a limitation in Microsoft Hyper-V that allows Insight Control virtualmachine management to re-register the virtual machine if the configuration file is created in thisdirectory. To prevent issues with reactivation, logical server management checks the path andrejects any virtual machines with a configuration file in the default location.

Hyper-V cluster shared volumes created with GPT disk partitioning system have limited supportHyper-V VM Hosts configured with cluster shared volume (CSV) storage with a GUID partition table(GPT) disk partitioning system are supported with the following conditions:

• All hosts (cluster nodes) must contain the same LUN identifiers for the LUN

• LUN identifiers must be unique in the host if the LUNs are presented from different storagearrays

iSCSI disk requirements for use with Microsoft Hyper-V virtual machine logical serversTo enable selection of a specific iSCSI datastore as the storage definition during logical servercreation or modification, iSCSI disks associated with Microsoft Hyper-V hosts must be a clusterdisk or a shared cluster volume, online, and contain a volume created with a drive letter.


With this limitation, only MBR (master boot record) formatted iSCSI cluster disks are supported.

Microsoft Hyper-V virtual machine logical server boot disk is IDE and data disks are SCSIWhen you create a Hyper-V virtual machine logical server, the boot disk (where Windows or Linuxis installed) is always created using an IDE controller. Data disks are created using a SCSI controller.The IDE controller is required to allow the installation of the operating system using Deploy Toolsor using images without additional drivers. If a virtual machine is created outside of Matrix OEvisualization and imported as a logical server, it is possible that the boot disk controller type isSCSI. In this case, you must install the SCSI controller driver to allow installation of the operatingsystem.If a virtual machine is provisioned by Matrix OE infrastructure orchestration (using either a SCSIor an IDE controller) and the logical server is imported, the virtual machine has an operating systeminstalled so the SCSI driver is not needed.

One Microsoft Hyper-V target displayed when activating a logical server created on a cluster diskvolume

A limitation in Microsoft Hyper-V R1 prevents the hypervisor from providing logical servermanagement with required information about a Hyper-V cluster. The result is that only one targetis shown when you attempt to activate a virtual machine logical server configured to be createdon a cluster disk volume. In Hyper-V R1, a logical server that is high availability (HA)-enabled canbe configured only on a cluster disk volume.Suggested actionUpgrade to Hyper-V R2. In R2, Microsoft corrected the problem so that the activation of a virtualmachine logical server configured to be created on a cluster disk or a cluster shared volume listsall Hyper-V hosts in the cluster on the target screen.

Reactivating and moving logical servers created on ESX 4x VM Hosts to ESX 3x VM HostsThe ESX 4 virtual hardware version prevents virtual machine logical servers that were first activatedon an ESX 4x VM Host from being moved or reactivated on an ESX 3x VM Host.ESX 4x supports virtual hardware version 4 and version 7 VM guests, but ESX 3x supports onlyversion 4 VM guests.You can successfully move a virtual machine logical server first activated on ESX 3x to an ESX 4xVM Host. If you attempt to move a logical server first activated on ESX 4.x to an ESX 3.x VM Host,logical server management rejects the targets.

Different WWN and MAC range definitions for FCoE ports are not supportedIf the WWNs and MAC addresses are from different ranges (such as VCEM-defined or user-defined)when defining FCoE ports for the VC profile, the activation fails with the following error:Workflow execution error: Failed to set SAN storage in profile

Suggested actionBe sure that the WWNs and MAC addresses are either both from a VCEM-defined range or bothfrom a user-defined range.

Some Virtual Connect profile capabilities are not supported by logical server managementImporting a server with Virtual Connect that implements the following features is not supported byMatrix OE logical server management. If the Virtual Connect profile is using these features, theerror “The Server Profile cannot be imported because it contains the following capabilitiesunsupported by Logical Servers:” is displayed.Extended VLAN Multiple networks in the server profile have more than 28 Virtual Local Area

Networks (VLANs)


EXTENDEDFC Server profile has two or more FC Connections using the same connectionbay (with the same connectionBay attribute value), which identifies a serverprofile created for an HP Integrity BL860c)

EXTENDEDFCOE Server profile has more than one Fibre Channel over Ethernet (FCoE)connection associated with the same interconnect bay (IO bay)

ISCSI Server profile has at least one iSCSI connectionNAG Server profile has an associated Network Access Group

Applications that depend on IO hardware paths or legacy storage DSFs are not supported withIntegrity cross-technology logical servers

HP-UX applications that have a dependency on I/O hardware paths or Legacy Storage DeviceSpecial Files (DSFs) might fail if these applications are running on Integrity cross-technology logicalservers. This occurs because I/O hardware paths and Legacy Storage DSFs may change duringcross-technology move operations. After a cross-technology logical server move operation,Networking I/O Instance Numbers and Storage Agile DSFs will not cause application failure.

Major issues

A new Integrity VM logical server may be assigned MAC addresses and/or server WWNs belongingto previously unmanaged Integrity VM logical servers, resulting in duplicate MAC addresses and/orserver WWNs

If you unmanage an Integrity VM logical server, then create a new Integrity VM logical server,MAC addresses and/or WWNs from the unmanaged logical server may be assigned to the newlogical server resulting in duplicates. If this new logical server is then activated on the same IntegrityVM host as the host containing the previously unmanaged Integrity VM logical server, an activationerror occurs due to a conflict with a duplicate MAC address and/or server WWN assigned to theunmanaged Integrity VM logical server. Additionally, the previously unmanaged Integrity VMlogical server cannot be imported again until the newly created logical server that introduced theduplicate MAC addresses and/or server WWNs is corrected or deleted.Suggested actions• To prevent duplicate MACs and/or WWNs:

If possible, use the Delete action rather than the Unmanage action when removing anIntegrity VM logical server.

◦

◦ If possible, re-import any previously unmanaged Integrity VM logical servers.◦ If you must unmanage an Integrity VM logical server, you may reserve its WWNs and

MAC addresses to prevent them from being reassigned by using this procedure:1. Determine the WWNs of the unmanaged Integrity VM logical server by logging onto

the VM host and using the hpvmstatus command:hpvmstatus –P Name_of_Unmanaged_VM | grep npiv

Sample output:hba avio_stor 0 1 npiv /dev/fclp0-0x5001438002A30041,0x5001438002A30044

Make a note of the server WWNs, converting them to lsmutil format. In this example,0x5001438002A30041 and 0x5001438002A30044 become50:01:43:80:02:A3:00:41 and 50:01:43:80:02:A3:00:44.

2. If using Storage Pool Entries (SPEs), delete the SPE containing the WWNs of theunmanaged Integrity VM logical server.

3. Reserve the WWNs that belonged to the unmanaged Integrity VM logical server byusing the lsmutil –reserve –WWN WWN command (where WWN is the worldwidename that you made note of in Step 1. Issue this command once for each of theWWNs associated with the unmanaged Integrity VM logical server.


4. Create a placeholder logical server to prevent the MAC addresses from beingassigned to new logical servers:1. Create a new logical server template without specifying any storage entries in

the storage configuration.2. In the network selection page, click Add Network Entries.

NOTE: LSM shows the MAC addresses that were part of the unmanagedlogical server.

3. Click Next and then Finish to create the logical server template.This Integrity VM logical server template cannot be activated because it lacksstorage entries, but the new template prevents the MAC addresses of theunmanaged logical server from being used again.

You may now create and activate Integrity VM logical servers.

• To correct duplicate MAC and/or WWNs:1. Delete the newly created Integrity VM logical server (after backing up its data, if

necessary).2. Perform one of the procedures described above under To prevent duplicate MACs and/or

WWNs.You may now create and activate Integrity VM logical servers.

HP Logical Server Automation service may run out of memory on large scale systemsOn large scale systems, with large numbers of managed nodes and/or large numbers (1000+)of logical servers on a 64-bit CMS, the HP Logical Server Automation service may run out ofmemory and display out of memory errors in the /logs/hp_lsa_service.log or /logs/vselog.lsm.log file. You may see the following errors:java.lang.OutOfMemoryError: GC overhead limit exceededjava.lang.OutOfMemoryError: Java heap spacejava.lang.OutOfMemoryError: PermGen space

For more information, see “Configuring HP Insight Management for large environments” in the HPInsight Management 7.0 Installation and Configuration Guide at Matrix OE Information Library.Suggested action 1To correct the GC overhead limit and/or Java heap space errors, modify the configuration file toincrease memory allocation:1. Modify the configuration file (shown in the default location) at C:\Program Files\HP\

Virtual Server Environment\conf\hp_lsa_service.conf.2. Increase the value of wrapper.java.maxmemory=1024. HP suggests a value of 2048.3. Restart the HP Logical Server Automation service.Suggested action 2To correct the PermGen space error, modify the configuration file to increase permanent generationsize:1. Modify the configuration file (shown in the default location) at C:\Program Files\HP\

Virtual Server Environment\conf\hp_lsa_service.conf.2. Add the following line. (Change additional.2 to additional.3 if a line with

additional.2 already exists; change additional.2 to additional.4 if lines withadditional.2 and additional.3 already exist, and so on.)wrapper.java.additional.2=-XX:PermSize=32m -XX:MaxPermSize=128m

3. Restart the HP Logical Server Automation service.



Minor issues

Performing operations on logical servers outside of Matrix OE visualization may causeincorrect information to be displayedIf you use tools other than Matrix OE visualization to manage logical servers, modifications to theunderlying resources managed by logical server management are not immediately reflected in thelogical server.For example, if you modify the power state of a managed resource outside of logical servermanagement, the power state of the logical server is displayed incorrectly until logical servermanagement resets its data.Suggested actionWait several minutes for the logical server to reset its data, or use Tools→Logical Servers→Refresh...to manually refresh server resources.

Activating an Integrity VM virtual machine logical server with three or more vCPUs mayfail with entitlement error in VM HostIf you attempt to activate an Integrity VM logical server with three or more vCPUs on an IntegrityVM Host with 1596 Mhz processor speed or more, and the entitlement value used to create thevirtual machine on the VM Host falls below the minimum 5% value, the activation may fail withthe error “Value x.x for entitlement_min is below minimum, setting to 5.0.” This error is displayedon the VM Host.Suggested actionOn the Integrity VM Host, set the Integrity VM configuration variable GUESTMHZMIN to the nearestinteger value that is five percent of the host processor speed in cycles. The following commandscan be used to perform the GUESTMHZMIN variable setting on the Integrity VM Host:# typeset -i i j k# i=$(hpvmstatus -sM | cut -d: -f3)# j=i/20# k=j*20# if ((i>k)) ; then j=j+1 ; fi# ch_rc -a -p GUESTMHZMIN=$j /etc/rc.config.d/hpvmconf

Unmanaging an active physical logical server with shared storage pool entryIf you unmanage a physical logical server that is currently active on a server with Virtual Connect,and the logical server is configured with a shared storage pool entry, the logical server may beautomatically re-imported in a subsequent refresh unless the associated storage pool entry is deleted.If the storage pool entry is defined as part of the logical server and is not shared, the storage poolentry is automatically deleted when the logical server is unmanaged.Suggested actionImmediately after unmanaging a logical server, delete its associated storage pool entry, or modifythe storage pool entry to remove the unmanaged logical server’s WWNs.

Modifying a catalog storage pool entry and unmasking the volume removes candidatesBy default, the “Masked” checkbox is checked and grayed out when you create a SAN CatalogStorage Entry on the Create Storage Pool Entry screen. After the volume is activated in SPM, if yousubsequently modify the storage pool entry, the user interface allows you to deselect the “Masked”checkbox for newly added volumes. However, if you deselect “Masked” (thereby unmasking thevolume) and press Show Candidates, storage candidates are not displayed in the Candidatesdrop-down menu for this new volume, and this sequence renders the volume and entire storagepool entry unusable.


Suggested actionWhen modifying a storage pool entry, make sure that the corresponding “Masked” checkbox ischecked throughout the initial provisioning process (which includes modifying the storage poolentry and adding a new volume, getting and selecting candidates, and saving the storage poolentry to initiate activation of the new volume in SPM). If you unchecked “Masked” during thisprocess and the volume becomes inoperable, remove the volume and re-add it to the storage poolentry.After the volume has been successfully activated, you can modify the storage pool entry to unmaskthe volume, if desired.

Storage auto provisioning does not select a preconfigured data disk if the storage poolentry does not contain a boot diskDuring storage provisioning, if a storage pool entry contains a volume that matches the requirementsof an infrastructure orchestration service, the matching volume is used for the service.If the IO template contains a boot disk and a data disk, and the storage pool entry contains onlya boot disk, the boot disk is used and the data disk is automatically created. However, if the storagepool entry contains only a data disk, the volume is not used. Both the boot and data disk will becreated.Suggested actionAdd a boot disk to the storage pool entry containing the desired data disk, or allow both disks tobe automatically created.

Moving a managed virtual machine guest using Microsoft SCVMM may change BIOSserial number of logical serverOn rare occasions, the BIOS serial number of a VM guest managed by logical server managementis changed if a file copy move operation is performed on the VM guest using Microsoft SystemCenter Virtual Machine Manager (SCVMM). This may cause operations on the virtual machinelogical server to be unsuccessful, or the virtual machine logical server may become inoperable.Suggested actionUnmanage the logical server using the Virtualization Manager Delete→Unmanage Logical Server...menu selection, then import the virtual machine as a logical server using Tools→Import→LogicalServer....

Viewing or editing the logical server management properties file may cause failure of theHP Logical Server Automation serviceViewing or editing the C:\Program Files\HP\Virtual Server Environment\conf\lsa\lsa.properties file may write lock the file while the HP Logical Server Automation serviceis running. This may cause a serious failure of the service.Suggested actionUse a text editor that does not lock the file, such as Emacs, Notepad, or TextPad.

Process Resource Manager

Limitations

Creating and modifying configuration files with privilege elevation enabled is not supportedIf privilege elevation is enabled in Systems Insight Manager, you can monitor Process ResourceManager (PRM), but you may experience problems performing PRM operations. Specifically,creating and modifying PRM configuration files may report success even though the operationshang.

Process Resource Manager 35

Suggested actionDo not change PRM configuration files using the PRM user interface when privilege elevation isenabled.

Storage Provisioning ManagerStorage Provisioning Manager runs on an x86/x64 Windows CMS.

Limitations

Presenting an in-use storage volume to an unintended consumer may result in data lossPay special attention when importing pre-presented volumes into SPM because all storage volumesimported into SPM are made available to fulfill storage services requested by SPM consumers suchas Matrix Operating Environment. Import only storage volumes intended to fulfill storage requestsserviced by this instance of SPM. If you have multiple instances of SPM (either independent or ina federated configuration), you may wish to have each instance manage different storage arraypools or completely different arrays.When importing pre-presented storage volumes, ensure the volume is not already in use. Do notimport volumes that are in use, are intended for use by other systems or services, or that have beenimported into any other instance of SPM.

Storage volumes must be managed by one SPMLarge data center environments may make use of multiple CMS configurations (either independentor in a federated configuration). Any individual storage volume/LUN should be managed by onlyone SPM. This can be done by having different SPM instances on different CMSs manage differentstorage array pools, or completely different arrays.

Required version of Brocade Network Advisor and HP B-Series Network AdvisorBrocade Network Advisor 11.1.3 or HP B-Series Network Advisor 11.1.3 must be installed inorder to manage Brocade-based SAN environments.HP B-Series Network Advisor is available from http://h18006.www1.hp.com/products/storageworks/dc_fabricmgr/index.html. See the HP Storage Provisioning Manager (SPM) UserGuide at www.hp.com/go/matrixoe/docs for additional details.

SPM cannot support a mixture of SPM-created and administrator-created zonesWhen using the SPM automated zoning capability with HP B-series or Brocade SAN networks,storage administrators cannot present storage to the same initiator from two separate arrays if theyhave manually created zones via the Brocade Network Advisor that include that initiator.

SPMs managing the same Brocade network must have distinct zonesWhen using the SPM automated zoning capability with HP B-series or Brocade SAN networks,environments that make use of multiple CMS configurations (either independent or in a federatedconfiguration) may manage the same HP B-series or Brocade network. In this situation, however,the set of initiators managed by each instance of SPM must be distinct, therefore making correlatedzones created and managed by each instance of SPM also distinct.

Non-Brocade SAN environments must have unmanaged networks in SPMStorage administrators must provide SPM with information regarding SAN connectivity. In addition,for non-Brocade SAN environments, an unmanaged network must be created to communicate theSAN configuration to SPM as described in the HP Insight Management Installation and ConfigurationGuide at www.hp.com/go/matrixoe/docs.


http://h18006.www1.hp.com/products/storageworks/dc_fabricmgr/index.html

http://h18006.www1.hp.com/products/storageworks/dc_fabricmgr/index.html



SPM supports FC fabric repair but not FC fabric re-configurationWhen using the SPM automated zoning capability with HP B-series or Brocade SAN networks,SPM can recover from fabric repairs but not from fabric re-configuration. Fabric repairs are fabricchanges that keep Brocade Network Advisor’s persistent fabric name intact (such as switchreplacement, switch firmware upgrade). Re-configuration is changes to the fabric that break BrocadeNetwork Advisor’s persistent fabric name (such as all switches in the fabric taken down).

Importing a large number of volumes is time consuming and may result in failureDepending on your environment, importing 1,000 or 5,000 or more volumes at the same timeinto SPM can cause a service crash which requires a restart of the SPM service. Restrict simultaneousvolume imports into SPM to less than 5,000 for InForm OS 3.1.1 MU1 or later, and to less than1,000 for all other storage operating systems.

3PAR autonomic groups and virtual domains are not supported in SPMThe following 3PAR features are not supported in this release of SPM. See the HP StorageProvisioning Manager (SPM) User Guide at www.hp.com/go/matrixoe/docs for more information.

• Autonomic groupsSPM does not support 3PAR autonomic groups (host sets and virtual volume sets). Under certainconditions, SPM can coexist with autonomic groups, but it does not take advantage of thatfunctionality.

• Virtual domainsSPM does not support 3PAR virtual domains. Using virtual domains on 3PAR storage systemsmanaged by SPM will likely result in erratic behavior in SPM, particularly during volumeimport, volume refresh and goal binding.

SPM cannot import or resynchronize P9500 or XP storage array resources if they are locked byRWC

If resources from a P9500 or XP storage array, including pools and volumes, are locked by theRemote Web Console (RWC), SPM cannot import or re-sync those resources. (This does not affectSPM’s ability to perform mapping and masking during service activation.)Suggested actionEnsure that all web browsers running RWC are shut down during import or re-sync operations.See the P9000 Remote Web Console User Guide for more information.

Firmware for 3PAR storage systems3PAR storage systems must have InForm OS 3.1.1 MU1 or higher installed in order to work properlywith SPM. SPM will not allow 3PAR arrays with earlier versions to be imported into the SPM catalog.

Minor issues

SPM can report volume capacity in GiB or GBSPM and some storage devices, such as HP EVA and XP arrays, as well as Matrix OE visualizationreport volume capacity in standard GB (1 GB=1,000,000,000 bytes) while 3PAR reports GiB(1GiB=2^30=1,073,741,824 bytes).Suggested actionSPM can report volume capacity using GB or GiB. The default is GiB. Select GiB for consistencywith 3PAR. Select GB for consistency with other storage devices such as HP EVA and XP or MatrixOE visualization. For detailed information, see the HP Storage Provisioning Manager (SPM) UserGuide at www.hp.com/go/matrixoe/docs.

Storage Provisioning Manager 37


http://h20000.www2.hp.com/bc/docs/support/SupportManual/c02535251/c02535251.pdf?jumpid=reg_R1002_USEN


Restart is required if SPM loses connectivity to an XP storage array while it is attemptingto access the arrayUsing SPM to manage XP storage arrays requires the CMS to be connected to the FC SAN. If SPMloses connectivity to the array while it is attempting to access the storage array, either to gatherinformation for the catalog or perform configuration operations, the SPM catalog might be in aninconsistent state.Suggested actionAfter FC connectivity has been reestablished between the CMS and the XP storage array, restartthe “HP SPM Storage Module J2EE Service” and “HP SPM Storage Module Service” Windowsservices.

Importing a volume (LDEV) with exactly 256 presentation paths does not succeedFrom an XP storage array, attempting to import a volume (LDEV) with exactly 256 presentationpaths does not succeed. This is caused by a known firmware defect.Suggested actionChange the number of presentation paths to a number other than 256 and retry the import operation.

Improper host group deletion using an EVA multipathing configuration with SPMIn an EVA multipath configuration, it is possible to activate an SPM service that will present avolume to a host with multiple initiators. If that service is later deactivated, SPM does not properlydelete the host group from the array. As a result, there may be unused host groups on the EVAthat are visible through the Command View console. In general, this should not cause any issueswith use of that array unless the maximum number of EVA host groups are consumed.Suggested actionClean up an unused host group (called “host alias” for EVA arrays). From the Command Viewconsole:1. Verify the host alias that you need to clean up. (Note that unused host groups from SPM will

have no volumes presented to them.)2. Launch Command View in a Browser and log in.3. In the left panel, expand the array that has the unused host group.4. In the left panel, expand the Hosts folder in the Command View Browser.5. In the left panel, click the unused host alias in the list.6. In the right panel, click the Presentation tab to verify that no disks are presented to this host

alias. (Do not delete a host alias with an active presentation.)7. If this is the correct host alias, click the General tab in the right panel.8. In the right panel, click the Delete host button.9. In the popup window, click the OK button.After the unused host groups are deleted, log into SPM and perform a re-synch operation on thearray. This will ensure consistency between the array status and the SPM catalog.

SPM may mark an array offline when a volume is deleted during on-demand provisioningDuring on-demand provisioning, when a storage service is deactivated and the volume is deletedon the array, an error may occur that causes SPM to incorrectly mark the array as offline in itscatalog. While the array is marked offline, SPM will not perform any provisioning or reconfigurationoperations on the array even though the array is functioning properly.Suggested action 1Wait approximately 60 minutes for SPM to automatically resynchronize and restore the array toan online state in its catalog. After the array is online, provisioning and reconfiguration operationscan be performed.


Suggested action 2Manually log into SPM and browse to the offline array. Click Resync Array, which forces SPMsynchronization with the array to be performed immediately. If successful, the array will be markedonline.

Internet Explorer Enhanced Security Mode must be disabledIf Internet Explorer Enhanced Security Mode is enabled, Internet Explorer will not run any browserplugins. Attempting to access the SPM user interface in this mode results in a prompt to installSilverlight, even if Silverlight is already installed.Suggested actionDisable Internet Explorer Enhanced Security Mode or add the SPM user interface URL (the defaultis https://[machine-name]:8000) to your trusted sites list.

SPM 2.0 cannot process a request for a thin volume in an unmanaged arraySPM 2.0 cannot match a request for a thinly-provisioned volume (committed capacity is less thansubscribed capacity) in an unmanaged array. If the volumes in the unmanaged array’s XML fileare defined as none (default) or thin, SPM will not propose any thinly-provisioned volumes as acandidate for a volume service, and so an SPM user cannot provision a service using that volume.Suggested actionDefine all volumes in the unmanaged array’s XML file as thick volumes, regardless of whether theyare actually thin or thick. Adjust pool capacity defined in the same XML file to be large enough toaccommodate the additional capacity, and track actual capacities elsewhere. In addition, makesure the storage template does not require thinly-provisioned volumes.

Do no use “Must Not” requirements when defining the volume capacity in a storagetemplateThis functionality will be added in a future release.

Storage Provisioning Manager 39

3 Support and other resourcesContacting HP

Information to collect before contacting HPBe sure to have the following information available before you contact HP:

• Software product name

• Hardware product model number

• Operating system type and version

• Applicable error message

• Third-party hardware or software

• Technical support registration number (if applicable)

How to contact HPUse the following methods to contact HP technical support:

• See the Contact HP Worldwide website for contact options:http://www.hp.com/go/assistance

• Use the Contact hp link on the HP Support Center website:http://www.hp.com/go/hpsc

• In the United States, call 1-800-334-5144 to contact HP by telephone. This service is available24 hours a day, 7 days a week. For continuous quality improvement, conversations might berecorded or monitored.

Registering for software technical support and update serviceHP CloudSystem Matrix includes as standard, three or one year of 24 x 7 HP Software TechnicalSupport and Update Service and 24 x 7 four hour response HP Hardware Support Service. Thisservice provides access to HP technical resources for assistance in resolving software implementationor operations problems.The service also provides access to software updates and reference manuals in electronic form asthey are made available from HP. Customers who purchase an electronic license are eligible forelectronic updates.With this service, Insight Management software customers benefit from expedited problem resolutionas well as proactive notification and delivery of software updates. For more information about thisservice, see the following website:http://www.hp.com/services/insightRegistration for this service takes place following online redemption of the license certificate.

How to use your software technical support and update serviceAs HP releases updates to software, the latest versions of the software and documentation aremade available to you. The Software Updates and Licensing portal gives you access to software,documentation and license updates for products on your HP software support agreement.You can access this portal from the HP Support Center:http://www.hp.com/go/hpsc

Contacting HP 41

http://www.hp.com/go/assistance


http://www.hp.com/services/insight


After creating your profile and linking your support agreements to your profile, see the SoftwareUpdates and Licensing portal at http://www.hp.com/go/hpsoftwareupdatesupport to obtainsoftware, documentation, and license updates.

Warranty informationHP will replace defective delivery media for a period of 90 days from the date of purchase. Thiswarranty applies to all Insight Management software products.

HP authorized resellersFor the name of the nearest HP authorized reseller, see the following sources:

• In the United States, see the HP U.S. service locator web site:http://www.hp.com/service_locator

• In other locations, see the Contact HP worldwide web site:http://welcome.hp.com/country/us/en/wwcontact.html

Documentation feedbackHP welcomes your feedback. To make comments and suggestions about product documentation,send a message to:[email protected] the document title and manufacturing part number in your message. All submissions becomethe property of HP.

Security bulletin and alert policy for non-HP owned software componentsOpen source software (such as OpenSSL) or third-party software (such as Java) are sometimesincluded in HP products. HP discloses that the non-HP owned software components listed in theMatrix Operating Environment end user license agreement (EULA) are included with Matrix OE.To view the EULA, use a text editor to open the /opt/vse/src/README file on an HP-UX CMS,or the <installation-directory>\src\README file on a Windows CMS. (The defaultinstallation directory on a Windows CMS is C:\Program Files\HP\Virtual ServerEnvironment, but this directory can be changed during installation.)HP addresses security bulletins for the software components listed in the EULA with the same levelof support afforded HP products.HP is committed to reducing security defects and helping you mitigate the risks associated withsecurity defects when they do occur. HP has a well defined process when a security defect is foundthat culminates with the publication of a security bulletin. The security bulletin provides you with ahigh level description of the problem and explains how to mitigate the security defect.

Subscribing to security bulletinsTo receive security information (bulletins and alerts) from HP:1. Open a browser to the HP home page:

http://www.hp.com

2. Click the Support & Drivers tab.3. Click Sign up: driver, support, & security alerts, which appears under Additional Resources

in the right navigation pane.

42 Support and other resources

http://www.hp.com/go/hpsoftwareupdatesupport

http://www.hp.com/service_locator

http://welcome.hp.com/country/us/en/wwcontact.html

mailto:[email protected]

http://www.hp.com

4. Select Business & IT Professionals to open the Subscriber's Choice web page.5. Do one of the following:

• Sign in if you are a registered customer.

• Enter your e-mail address to sign-up now. Then, select the box next to Driver and Supportalerts and click Continue.

Related informationThe latest versions of manuals and white papers for Matrix Operating Environment and relatedproducts can be downloaded from the HP Web. Documents for Matrix OE software can be foundat Matrix OE Information Library.For more information about Matrix OE and related products and solutions, visit the following HPwebsites:• HP Insight Management: www.hp.com/go/insightmanagement

• HP Matrix Operating Environment: www.hp.com/go/matrixoe

Related information 43
