Hewlett Packard Enterprise L 17.02 NonStop Software Software Installation and Upgrade Guide
Pubblicità
Pubblicità
L17.02 Software Installation and
Upgrade Guide
Part Number: 875812-001
Published: March 2017
Edition: L17.02 RVU
© Copyright 2017
Legal Notice
Confidential computer software. Valid license from HPE required for possession, use or copying. Consistent with FAR 12.211 and 12.212,
Commercial Computer Software, Computer Software Documentation, and Technical Data for Commercial Items are licensed to the U.S. Government under vendor’s standard commercial license.
The information contained herein is subject to change without notice. The only warranties for HPE products and services are set forth in the express warranty statements accompanying such products and services. Nothing herein should be construed as constituting an additional warranty. HPE shall not be liable for technical or editorial errors or omissions contained herein.
Export of the information contained in this publication may require authorization from the U.S. Department of Commerce.
Acknowledgements
Microsoft ®, Windows®, Windows NT®, Windows® XP, and Windows Vista® are U.S. registered trademarks of Microsoft Corporation. Intel® and
Intel ® Itanium® are trademarks of Intel Corporation in the U.S. and other countries. Java is a registered trademark of Oracle and/or its affiliates.
Java is a registered trademark of Oracle and/or its affiliates.
Contents
About This Document...........................................................................................10
1 Overview of Installing This RVU........................................................................13
2 Managing Firmware on NonStop X Systems....................................................21
3 New Products and Features for This RVU........................................................23
Contents 3
4 Upgrade Checklist.............................................................................................33
5 Preparing to Upgrade the System Software (Preinstallation Tasks)..................37
4 Contents
6 Installing a SUT for this RVU Using NonStop Software Essentials 5.0.............47
7 Running ZPHIRNM ...........................................................................................59
8 Loading this RVU...............................................................................................65
Run the Prepare for Down System CLIM Management Action in the OSM Service
9 Resuming Normal Operations...........................................................................78
Contents 5
Step 1: Copy the Sample TCP/IP Files Into $SYSTEM.ZTCPIP (Optional).............................81
10 Backing Out the DSM/SCM Revision..............................................................85
11 Running ZPHIRNM to Reactivate the Previous Software Revision.................90
12 Falling Back on the CLIMs, Loading the Previous RVU, and Resuming Normal
Operations............................................................................................................92
6 Contents
13 Support and other resources.........................................................................108
A Using DSM/SCM With NonStop Software Essentials to Install SPRs............110
B Copying the DSM/SCM, NonStop Software Essentials, and SWID SPRs from the SUT Using NonStop Software Essentials....................................................117
Determine the Product Version of DSM/SCM on the Host System and the SUT........................118
Copy (RESTORE) the DSM/SCM Server, NonStop Software Essentials Server, and SWID Product
Receive the H03 DSM/SCM, NonStop Software Essentials Server, and SWID SPRs ...................118
Create a New Software Revision for DSM/SCM, NonStop Software Essentials, and SWID SPRs
Verify that the SUT or Scout Version of the NonStop Software Essentials Server is Installed....122
Contents 7
C Recovering from a System Load Failure........................................................124
Index...................................................................................................................125
8 Contents
Tables
About This Document
CAUTION:
Any person who performs the procedures and actions documented in this guide is required to have direct and authorized access to a NonStop X system’s NonStop system console in order to view the important procedural information contained in OSM online help files and to use OSM tools to perform procedures identified in this guide.
This guide describes:
•
Upgrading the initial L17.02 RVU through the use of SUT refreshes, if necessary.
•
Falling back to a previous SYSnn.
Supported RVUs
This manual supports the L17.02 RVU and all subsequent L-series RVUs unless otherwise noted.
Intended Audience
This guide is written for operations staff and support personnel and requires a working knowledge of Windows Server 2008 or Windows Server 2012, OSM client and server components, system console software, NonStop Software Essentials, and DSM/SCM.
New and Changed Information in This Edition
This is a new edition. New and changed information includes:
•
Added references to the Virtualized NonStop Deployment and Configuration Guide as needed for certain L17.02 content.
•
Summarized the features for L17.02.
•
Updated procedures throughout the guide as needed for L17.02.
What This Guide Does Not Cover
•
Changes from G-, H-, and J-series RVUs.
•
System migration planning. System-wide issues involved in migrating from an earlier L-series
RVU to the L17.02 RVU are covered in:
◦
The L17.02 Release Version Update Compendium.
◦
The Interactive Upgrade Guide 2 (IUG2)
•
Installing an L-series RVU from a host system to a remote or separate network-connected target system using an Expand connection or tape. See the Expand L-Series Technical
Update and NonStop Software Essentials User's Guide for information about setting up target systems.
•
Installing and configuring OSS.
•
Upgrading NonStop System Console workstation applications. To install or upgrade HPE and third-party system console client software, such as OSM client-based components, see the NonStop System Console Installer Guide.
•
Installing and configuring OSM server components. To install and configure OSM server components, and start using client-based components, see the OSM Configuration Guide.
•
Installing, configuring, and migrating to newer versions of NonStop SQL/MX. The SQL/MX
Installation and Management Guide provides step-by-step instructions for installing and configuring specific versions of NonStop SQL/MX.
10
•
Installing and configuring NonStop X systems and their hardware components. See the
NonStop X NS7 Planning Guide or the NonStop X NS3 Planning Guide, or contact your
HPE authorized service provider.
•
Configuring and deploying HPE Virtualized NonStop (vNS) and any hardware components.
See the Virtualized NonStop Deployment and Configuration Guide for detailed information.
Notation Conventions
General Syntax Notation
This list summarizes the notation conventions for syntax presentation in this manual.
UPPERCASE LETTERS
Uppercase letters indicate keywords and reserved words. Type these items exactly as shown.
Items not enclosed in brackets are required. For example:
MAXATTACH
Italic Letters
Italic letters, regardless of font, indicate variable items that you supply. Items not enclosed in brackets are required. For example:
file-name
Computer Type
Computer type letters indicate:
•
C and Open System Services (OSS) keywords, commands, and reserved words. Type these items exactly as shown. Items not enclosed in brackets are required. For example:
Use the cextdecs.h header file.
•
Text displayed by the computer. For example:
Last Logon: 14 Oct 2015, 08:02:23
•
A listing of computer code. For example if (listen(sock, 1) < 0)
{ perror("Listen Error"); exit(-1);
}
Bold Text
Bold text in an example indicates user input typed at the terminal. For example:
ENTER RUN CODE
?
123
CODE RECEIVED: 123.00
The user must press the Return key after typing the input.
Related Information
NOTE:
Many of the documents referenced in this guide are located in the NonStop Technical
Library (NTL) at http://hpe.com/info/nonstop-docs.
Notation Conventions 11
Publishing History
Part Number
875811-001
Product Version
N.A
Publication Date
March 2017
HPE Encourages Your Comments
HPE encourages your comments concerning this document. We are committed to providing documentation that meets your needs. Send any errors found, suggestions for improvement, or compliments to [email protected].
Include the document title, part number, and any comment, error found, or suggestion for improvement you have concerning this document.
12
1 Overview of Installing This RVU
CAUTION:
Any person who performs the procedures and actions documented in this guide is required to have direct and authorized access to a NonStop X system’s NonStop system console in order to view the important procedural information contained in OSM online help files and to use OSM tools to perform the procedures described in this guide.
For information on installing SPRs, see
“Using DSM/SCM With NonStop Software Essentials to
SUT on DVD
The L17.02 SUT is provided on DVD for NonStop X systems. For instructions on installing the
SUT from the DVD, see the IPSetup User’s Guide.
You can also obtain the SUT from Scout.
SUT on QCOW2 for Virtualized NonStop systems
Customers using vNS on L17.02 and later receive an initialized customized SUT for this product in GCOW2 format. For more information, see the Virtualized NonStop Deployment and
Configuration Guide.
CLIM Software for this RVU
CLIM software for NonStop X systems is provided on a separate DVD and in Scout. In order to install CLIM software onto the CLIMs in your NonStop X system, you must first install the software on the NonStop system console pair as described in the Cluster I/O Protocols (CIP) Configuration
and Management Manual.
For information about configuring Virtualized CLIMs using this QCOW2 image, see the Virtualized
NonStop Deployment and Configuration Guide.
CLIM software compatibility across RVUs is documented in the CLuster I/O Module (CLIM)
Software Compatibility Reference.
Disk Space Requirements
For NonStop X systems, the maximum SUT with all standard products, plus all optional products available to customers, contains approximately 20 gigabytes (GB) of data, depending on optional products (however, actual disk space required can be much greater).
Planning the Installation of an L-Series RVU
Several resources are available to help you plan for installing an L-series RVU:
•
Product softdocs and the README file.
•
The L17.02 Release Version Update Compendium provides highlights of the new products and features for this RVU. It also discusses the effect these changes have on installation and configuration, operations and management, data files, applications, and networks.
•
Scout for NonStop Servers is a Web-based SPR analysis and delivery tool. You can access
Scout through HPE NonStop eServices Portal at: https://h22204.www2.hpe.com/NEP/Logon/Home
Online help for using Scout is available on the Scout Web site.
•
Use Scout to view and print softdocs and release documents on the SUT after you receive new software into the archive. For more information, see the NonStop Software Essentials
User's Guide or Scout.
SUT on DVD 13
NonStop System Console Requirements
A NonStop system console (NSC) is an HPE-supplied personal computer used to run OSM and other maintenance and diagnostic software for NonStop X systems. For information on currently supported hardware models and other requirements, as well as the latest available console software, see the NonStop System Console Installer Guide.
For information about the vNSC used on Virtualized NonStop, see the Virtualized NonStop
Deployment and Configuration Guide.
Console Requirements for Dedicated Service LANs
NOTE:
If you are using vNS, see the Virtualized NonStop Deployment and Configuration Guide.
When a dedicated service LAN is present, make sure that:
•
One pair of NonStop system consoles (primary and backup) serves the entire dedicated service LAN. All other consoles in LAN systems must be powered OFF.
•
Both consoles must be running Windows Server 2008 or Windows Server 2012 and should have the latest software installed from the latest NSC DVD update applicable to your system.
For more information, ask your authorized service provider to refer to the NonStop Dedicated
Service LAN Installation and Configuration Guide for the NonStop System.
Updating System Consoles (Optional)
IMPORTANT:
HPE recommends that you always update the software on a console pair, including the OSM Console Tools, to the versions provided on the latest NonStop System Console
Installer DVD.
NonStop System Console Installer DVD Update 27 (or later) contains important hotfixes for using
OSM server software for consoles running Windows Server 2012.
Insight Remote Support for NonStop, which replaces Insight Remote Support Advanced, is not a SIM plug-in, but has repercussions for HPE SIM versions being run on any consoles that contain them. For more information, see
“Insight Remote Support for NonStop” (page 23)
.
NonStop Software Essentials and Console Requirements
NonStop Software Essentials is the software installation and management tool for the NonStop
X system. It works with DSM/SCM server software and provides the replacement for the DSM/SCM
Planner Interface and CNFGINFO, which are not supported on the 64-bit NonStop system consoles required for NonStop X systems.
You can use either the NonStop Software Essentials 5.0 client or a previous version to migrate to a new L-series RVU. Starting with the NonStop Software Essentials 5.0 client, NonStop
Software Essentials is no longer a plug-in to HPE SIM (Systems Insight Manager). This makes it easier and more feasible to install NonStop Software Essentials on the NonStop system console instead of on a separate Central Management Server (CMS), as was previously recommended due to the size and memory requirements for installing and running HPE SIM.
The NonStop Software Essentials Windows client (T0895) can be installed from the Insight
Control for NonStop DVD that ships with important hotfixes for using OSM server software for consoles running Windows Server 2012..
These considerations apply when choosing which version of NonStop Software Essentials to use:
•
You can use the NonStop Software Essentials 5.0 client with an older NonStop Software
Essentials server (T0894) for the purposes of RVU migration, although you will receive a
14 Overview of Installing This RVU
warning that not all functionality is available without the matching 5.0 server SPR (T0894
H02^AAH) that is available on the L17.02 SUT.
•
Likewise, you can use an existing pre-5.0 NonStop Software Essentials client with the new
5.0 server; however, the older client interfaces will continue to require a supported version of HPE SIM.
•
The NonStop Software Essentials 5.0 client can co-exist on the same console as a previous
HPE SIM plug-in version of NonStop Software Essentials. You can use either version.
This guide shows you how to upgrade the RVU using NonStop Software Essentials. You can use either the 5.0 client or an earlier HPE SIM plug-in version to perform the migration. For more information on installing and getting started with NonStop Software Essentials server and client components, see the NonStop Software Essentials Installation and Quick Start Guide.
Software Product
NonStop Software Essentials Server
NonStop Software Essentials 5.0 Client (Windows)
NonStop Software Essentials Client (Linux)
T-Number
T0894
T0895
T0996
Supported on L17.02?
Yes
Yes
No
NOTE:
NonStop Software Essentials requires that a server component be installed on the
NonStop X system. That server component is provided on the SUT for L15.02 and later L-series
RVUs, but can also be downloaded from Scout. See the Nonstop Software Essentials Installation
and Quick Start Guide.
Using OSM With This RVU
HPE recommends that you always use the latest versions of the OSM server and client software, along with other console software.
OSM Server-Based Component
The server-based OSM SPR released with this RVU is T0682L02^BAI. This version may be superceded by a later version.
NOTE:
HPE recommends that you always update console software to the versions available on the latest NSC DVD update applicable to your system.
OSM server-based components, which include the OSM Service Connection and OSM Event
Viewer applications, are both contained in T0682. For more information about this server version, see the L17.02 Release Version Update Compendium or the OSM Service Connection User’s
Guide, which is also available as online help with the OSM Service Connection. For detailed information about the OSM server-based component, including configuration options, see the
OSM Configuration Guide.
OSM Client-Based Components for Virtualized NonStop
vNS uses these OSM client-based components:
OSM Service Connection (T0862 L02 BAI), which supports:
•
Virtualized NonStop
•
Java 8, Update 11. For more information, see the OSM Configuration Guide.
OSM Console Tools (T0634 H06 ABT), which supports:
•
System Startup Tool for booting the vNS.
•
System Configuration Tool for configuring the vNS.
Using OSM With This RVU 15
•
CLIM Management Tool using SSH instead of FTP to perform Virtual CLIM updates, Virtual
CLIM software patch installation, and INS software updates.
•
Java 8, Update 111.
For more information about OSM components used with vNS, see the Virtualized NonStop
Deployment and Configuration Guide.
OSM Client-Based Components for NonStop X Systems
The OSM client-based components used for NonStop X systems, all part of the OSM Console
Tools (T0634), include:
•
The following OSM tools replace Low-Level Link functionality, which does not support
NonStop X systems:
OSM Tools
CLIM Management Tool
HSS and HCA Firmware
Management Tool
System Configuration Tool
System Startup Tool
Functions
Used to perform CLIM software updates and down system CLIM firmware updates. For detailed information about using the CLIM Management Tool to update CLIM software and firmware, see the NonStop Cluster I/O Protocols (CIP)
Configuration and Management Manual and the online help available within the tool.
Used exclusively for NonStop X systems to perform updates to HSS firmware and the InfiniBand HCA firmware required for processors. For more information, see the NonStop System Console Installer Guide (for downloading the latest firmware) and the online help available within the tool (for performing copy and update actions).
Used exclusively for NonStop X systems to modify system settings such as
System name, Expand node number, and InfiniBand network number; change
Maintenance Entity Unit; configure Processor Blade; configure CLIM; set/remove
System Maintenance Entity User IDs; and display/save InfiniBand network connectivity information. For more information, see the online help available in the tool.
Used exclusively for NonStop X systems to load the system; check processor status and perform processor actions such as Halt, Hard Reset, Prime for Load,
Reset, Load, Enable or Disable Freeze, and Dump to ME Unit.
•
The OSM Console Tools (T0634) also includes these tools that can be used for NonStop X systems:
◦
OSM System Inventory Tool
◦
OSM Certificate Tool (for Onboard Administrators)
◦
NonStop Maintenance LAN DHCP DNS Configuration Wizard
◦
OSM Quick Start Tool
◦
Apache OpenOffice
NOTE:
HPE recommends that you always update console software to the versions on the latest NSC DVD update applicable to your system (for example, NonStop System Console
Installer DVD Update 27 for L17.02).
16 Overview of Installing This RVU
Using DSM/SCM and NonStop Software Essentials With This RVU
CAUTION:
Before you install the new SUT, you must copy the SUT version of the DSM/SCM server to your host system. When you copy the DSM/SCM server SPR, you must also upgrade the SWID product (T9298H03) at the same time to avoid a fingerprint or mismatched file problem.
For information about upgrading DSM/SCM paired with NonStop Service Essentials, see
DSM/SCM With NonStop Software Essentials to Install SPRs” (page 110)
for the detailed instructions.
NonStop Software Essentials and This RVU
NonStop Software Essentials is the software installation and management interface for NonStop
X systems. See
“NonStop Software Essentials and Console Requirements” (page 14)
for important
NonStop Software Essentials version information.
DSM/SCM and this RVU
NOTE:
For information on upgrading DSM/SCM paired with NonStop Software Essentials, see
“Copying the DSM/SCM, NonStop Software Essentials, and SWID SPRs from the SUT Using
NonStop Software Essentials” (page 117)
for the detailed instructions.
Also be aware of the following:
•
By default, the DSM/SCM H03 PVU is configured to manage OSS files. Check that OSS is configured and started if DSM/SCM is set to manage OSS files. If not, the Apply can fail.
•
Creating a SIT is not supported in L-series RVUs.
DSM/SCM and Public DLLs
DSM/SCM manages and controls the distribution, installation, and activation of public DLLs in
L-series RVUs the same way as it does for H-series and J-series RVUs. Public DLLs are installed on L-series target systems in their own subvolume.
DSM/SCM and OSS
When performing a Build/Apply, ensure that the setting of the Manage OSS Files option matches your target system. If the target system is not to use OSS, you must uncheck the OSS files option box before beginning a Build request. If the box is left checked, OSS is included by the Build request. The OSS root fileset needs to be configured and started before files can be copied into the OSS file system by DSM/SCM. For OSS users who request HPE to install OSS, it is assumed that OSS is installed and configured. However, the OSS name server ($ZPNS) must be in the started state; otherwise, the Build/Apply fails.
Upgrading the NonStop Software Essentials Client
For information about the manageability tools used with vNS, see the Virtualized NonStop
Deployment and Configuration Guide.
You can upgrade your NonStop Software Essentials 5.0 client from the Insight Control on the on the latest NSC DVD update applicable to your system. For more information, see the NonStop
Using DSM/SCM and NonStop Software Essentials With This RVU 17
System Console Installer Guide. You can also download the NonStop Software Essentials client component from Scout.
Software Product
NonStop Software Essentials Server
NonStop Software Essentials Client (Windows)
NonStop Software Essentials Client (Linux)
T-Number
T0894
T0895
T0996
Supported for L17.02?
Yes
Yes
No
For instructions on installing NonStop Software Essentials, see the NonStop Software Essentials
Installation and Quick Start Guide.
Summary of Software Installation Tasks
To install software using DSM/SCM, perform these tasks:
1.
Receive new software into the DSM/SCM Archive.
When software is received, the files from each product are stored in the DSM/SCM Archive, a holding area for your software. File attributes are stored in the DSM/SCM host database.
Each time a new version of a product is received, only changed files are placed in the Archive, and file attributes are updated in the host database.
2.
Create a new software revision.
A software revision is a list of products stored in the DSM/SCM Archive that will be built into a new software revision. A software revision can include a SUT, SPRs, and third-party software. When you create a new software revision, DSM/SCM selects the products in the
SUT that are licensed for your site and replaces previous versions of the products.
3.
Build and apply the new software configuration revision. The Build process creates an activation package containing all the products listed in the new software revision. During the
Apply process, the activation package is transferred to the target system, and the new software is placed on the target system’s subvolumes. When the Apply request is finished, a snapshot is automatically created and returned to the host, if Expand-connected.
4.
Activate the new operating system image.
During the Build/Apply phase, DSM/SCM places product files in their TSVs with temporary
(fabricated) file names so that the new files do not conflict with the file names of currently running programs. You run the ZPHIRNM program on the target system to:
•
Rename the files that currently have actual file names to fabricated names. These files become obsolete when the new files are renamed.
•
Rename the new files, which have fabricated names, to their actual names.
After ZPHIRNM has completed renaming all OSS and Guardian files, it will run ZMODGP if:
◦
One or more of the ZcccGP files in the ZCNFDEFS TSV are new or changed (as compared to the last configuration activated).
◦
If no ZcccGP files change, ZMODGP is not run.
◦
ZMODGP is a TACL script that makes changes to the generic-process definitions in the system configuration database. When you use the ZMODGP script, you do not need to hard code new generic process definitions directly into the $ZCDB process (T6586, Configuration Services).
18 Overview of Installing This RVU
Check or Alter the CONFTEXT File Before Installing This RVU SUT
NOTE:
Tape_Label_Processing is controlled by the Storage Subsystem ($ZZSTO) ALTER
SUBSYS command. The CONFTEXT file directive TAPE_LABEL_PROCESSING gets overridden by the Storage Subsystem settings. See the SCF Reference Manual for the Storage Subsystem for more information about the ALTER SUBSYS command.
The processor type (SYSTEM_PROCESSOR_TYPE) is specified in the CONFTEXT file. It currently has only one value:
System
NonStop X system
SYSTEM_PROCESSOR_TYPE
NSX-E
Check the CONFTEXT file to ensure that the following entries are present. They are needed for the implicit DLLs:
SYSTEM_IMPLICIT_INIT_FILES
SYSTEM_IMPLICIT_MCP_FILES
SYSTEM_IMPLICIT_LIL_FILES
TANDEM^IMPLICIT^INIT^FILES;
TANDEM^IMPLICIT^MCP^FILES;
TANDEM^IMPLICIT^LIL^FILES;
This example shows a CONFTEXT file:
SYSTEM_PROCESSOR_TYPE
SYSTEM_VOLUME_SUBVOL
SYSTEM_LIBRARY_CODE_FILES
SYSTEM_PROCESS_CODE_FILES
SYSTEM_PROCESS_LIBRARY_FILES
STANDARD_MICROCODE
MICROCODE_FILES
SYSTEM_IMPLICIT_INIT_FILES
SYSTEM_IMPLICIT_MCP_FILES
SYSTEM_IMPLICIT_LIL_FILES
FILES_TO_COPY_TO_NEW_SYSTEM
FORMATTER_TEMPLATE_FILES
INITIAL_COMINT_INFILE
NSE-X;
$SYSTEM.SYS00;
TANDEM^LIBRARY^CODE^FILES;
TANDEM^PROCESS^CODE^FILES;
TANDEM^PROCESS^LIBRARY^FILES;
TANDEM^STANDARD^MICROCODE;
TANDEM^MICROCODE^FILES;
TANDEM^IMPLICIT^INIT^FILES;
TANDEM^IMPLICIT^MCP^FILES;
TANDEM^IMPLICIT^LIL^FILES;
TANDEM^FILES^TO^COPY;
TANDEM^FORMATTER^TEMPLATE^FILES;
$DSMSCM.SAVE.CIIN;
Installing CLIM Software
You must upgrade the CLIM software (and firmware) with the CLIM Management Tool before you can load the new SUT. For instructions, see
“5. Upgrade the CLIM Software and Firmware”
and the OSM online help for the CLIM Management Tool.
Backing Out the DSM/SCM Revision
If you encounter a problem with a SUT refresh that cannot be resolved, you can back out the
DSM/SCM revision.
Table 1 Tasks and Tools for Backing Out to the Previous SYSnn
To Perform This Task...
Use This Tool...
Documented in This Chapter...
1. Submit a backout request.
DSM/SCM Target Interface (ZPHITI)
“Backing Out the DSM/SCM Revision ”
2. Reactivate the previous software revision.
DSM/SCM ZPHIRNM program
Chapter 11: “Running ZPHIRNM to
Reactivate the Previous Software
Installing CLIM Software 19
Table 1 Tasks and Tools for Backing Out to the Previous SYSnn (continued)
To Perform This Task...
3. As needed, perform CLIM downgrades.
Use This Tool...
CLIM Management Tool
Documented in This Chapter...
“Falling Back on the CLIMs, Loading the
Previous RVU, and Resuming Normal
4. Load the NonStop operating system after performing the previous step.
System Startup Tool
Managing Firmware
When you install a new RVU, you must ensure you are running the latest firmware and upgrade those products, if necessary. The latest firmware versions and upgrade procedures for L-series
RVUs are identified in
“Managing Firmware on NonStop X Systems” (page 21)
and the NonStop
Firmware Matrices.
Installation, Migration, and Fallback
Some products associated with a new RVU have special considerations for installation, migration, and fallback, which you should be aware of before installing the new RVU. These considerations are described in
“New Products and Features for This RVU” (page 23)
or the Release Version
Update Compendium for that RVU.
Perform an RVU Upgrade on a Node
For instructions on how to perform an RVU upgrade on a node connected to an InfiniBand cluster, see the NonStop X Cluster Solution Manual.
20 Overview of Installing This RVU
2 Managing Firmware on NonStop X Systems
IMPORTANT:
This chapter contains information for NonStop X systems.
NOTE:
HPE authorized service providers will update the firmware when they install any FRUs in a NonStop X system. Firmware, unless otherwise noted, is considered customer-updateable.
You must manage all firmware within your system and ensure compatible versions are installed.
Failure to do so can result in a wide range of system problems, from intermittent faults to partial or complete system outages.
To identify the latest supported versions of firmware for your L-series RVU, see the NonStop
Firmware Matrices.
OSM displays firmware version information, including file names and a Compare State attribute, which compares the current version against the default or available version.
Updating the InfiniBand Switch Firmware on NonStop X Systems
Whether your Blade IB switches, IO Expansion IB switches, or IB Cluster switches were installed in the NonStop X system at the factory or have been replaced at the customer site, always check the firmware version on these switches and confirm that it is the latest supported firmware version as described in the NonStop Firmware Matrices. If the firmware is downrev, update it as described in the NonStop Firmware Matrices.
Managing OA Firmware for the c7000 Enclosure
Check the NonStop Firmware Matrices document for the minimum supported Onboard
Administrator (OA) firmware for this release.
Managing Blade Firmware
Firmware updates can be performed for:
NOTE:
This firmware is updated from the Blade iLO console. Firmware updates can take place on a running system. However, for the BIOS to take effect, the blade needs to be reset. Blade outage is not required to perform Blade iL0 updates.
•
Blade BIOS
•
Blade iLO
NOTE:
For minimum Blade BIOS and Blade ILO firmware levels for this RVU, see the NonStop
Firmware Matrices.
Managing CLIM Firmware and Software
It is very important that your CLIM firmware is compatible with your system. Refer to the NonStop
Firmware Matrices, which is located in the Service Information section of the Support and Service collection of the NonStop Technical Library at http://hpe.com/info/nonstop-docs, and ensure you have compatible firmware levels for all CLIM components.
The OSM Service Connection can be used to update firmware on individual CLIMs. The CLIM
Management Tool provides a fast and efficient method for updating CLIM software and firmware for all CLIMs on the system at the same time, while the system is up or down.
For information on installing the CLIM Management Tool, see the NonStop System Console
Installer Guide.
Updating the InfiniBand Switch Firmware on NonStop X Systems 21
For instructions on using the CLIM Management Tool to upgrade CLIM firmware when a system is halted or when it is running, see the Cluster I/O Protocols (CIP) Configuration and Management
Manual.
For information on performing CLIM software updates, see the OSM online help for the CLIM
Management Tool and OSM Service Connection, plus:
•
CLuster I/O Module (CLIM) Software Compatibility Reference
•
Cluster I/O Protocols (CIP) Configuration and Management Manual
22 Managing Firmware on NonStop X Systems
3 New Products and Features for This RVU
This chapter briefly describes the new products, features, and enhancements released with each
L-series RVU. New products and features in this chapter are categorized according to the RVU at which they were first introduced.
For the latest information about migration or fallback information associated with these new products and features, see the L17.02 Release Version Update Compendium and any associated
Readme files and softdocs .
L15.08
Manageability Product Enhancements
Insight Remote Support for NonStop
HPE Insight Remote Support for NonStop is available as the go-forward remote support solution for NonStop systems. It replaces Insight Remote Support Advanced, which was an HPE SIM plug-in. Insight Remote Support for NonStop is not a SIM plug-in, making it easier and more practical to run on NonStop system consoles. It is supported on NonStop system consoles running
Windows Server 2008 and Windows Server 2012. For more information, see Insight Remote
Support for NonStop located in the HPE Integrity NonStop Service Information collection of NTL.
NOTE:
While Insight Remote Support for NonStop does not require HPE SIM, if HPE SIM is already installed on the console, it must be a compatible version or the Insight Remote Support for NonStop installer will remove it without warning. This would impact your ability to use plug-ins such as NonStop Software Essentials from the console until HPE SIM is re-installed. To ensure you have a compatible version of HPE SIM before installing Insight Remote Support for NonStop, see the NonStop System Console Installer Guide.
NonStop Software Essentials 5.0
The NonStop Software Essentials 5.0 client that is provided on the latest NSC DVD update applicable to your system is not an HPE SIM (Systems Insight Manager) plug-in, unlike earlier
NonStop Software Essentials clients.
NOTE:
Falling back to NonStop Software Essentials 4.0 client requires HPE SIM to be installed on a qualified system console or CMS. For more information, see the NonStop Software Essentials
Installation and Quick Start Guide for the NonStop Software Essentials 4.0 release.]
For more information about NonStop Software Essentials 5.0, see
“NonStop Software Essentials and Console Requirements” (page 14)
and
“NonStop Software Essentials and This RVU”
NonStop SSH
This release of NonStop SSH supports:
•
A new set of SLOWDOWN parameters that allow users to reduce CPU use of SSH processes.
•
A new USER attribute REQUIRED-AUTHENTICATIONS for configuring multiple required user authentication methods per loin.
For more information, see the NonStop SSH Reference Manual.
NonStop System Console Changes
NonStop X systems now support NonStop system consoles running Windows Server 2012 or
Windows Server 2008.
L15.08
23
The latest NSC DVD update applicable to your system is provided and used to upgrade existing
NSC software. It includes upgrades to several HPE and third-party software products. These include Internet Explorer 11 (now supported by OSM in compatibility mode) and Windows Server
2012 Service Pack 1. For more information, see the NonStop System Console Installer Guide.
OSM Changes
The OSM Service Connection for L15.08 (T0682L02^BAA) includes these changes:
•
OSM supports the NonStop X Cluster Solution. For more information, see the NonStop X
Cluster Solution Manual. The OSM Service Connection monitors internal InfiniBand switches for problems such as temperature out of range or minimum software not installed and generates alarms if problems are detected.
•
Java 8, which OSM will install for you, is required. For more information, see the OSM Service
Connection online help and the OSM Configuration Guide .
•
Power-Fail Support for site UPS. OSM Service Connection actions allow you to configure power rails as “Site UPS.” When a power failure is detected, OSM disables Write Cache
Enabled (WCE) disk volumes to protect against data loss, re-enabling it when power is restored. For more information, see the OSM Service Connection online help or the OSM
Configuration Guide.
•
The OSM Service Connection and OSM Event Viewer now default to using an SSL
MINIVERSION of 3.1 (TLS 1.0) instead of 3.0 (SSL 3.0).
L16.05
This section provides information about new products and enhancements that can affect the software installation and upgrade steps for L16.05. See the L16.05 Release Version Update
Compendium for more information about these and other L16.05 products and feature enhancements and their migration and fallback considerations.
ACL System Serviceability Enhancements
In L-series RVUs prior to L16.05, only NonStop users in the super group can perform CLIM configuration and management operations. With the L16.05 RVU, NonStop CLIM Management command line tools (T0834) have been enhanced to allow non-super group users to perform those actions, provided they are members of the Safeguard SECURITY-CLIM-ADMIN group.
For more information, see the Cluster I/O Protocols (CIP) Configuration and Management Manual
For fallback and migration considerations, see the L16.05 Release Version Update Compendium.
Boost C++ Libraries
Boost is a collection of C++ class libraries. NonStop supports Boost version 1.56. NonStop support for Boost consists of providing Boost header files configured for NonStop along with pre-built (as archive) binary files.
For more information, see the C/C++ Programmer’s Guide. For migration considerations, see the L16.05 Release Version Update Compendium.
Expand Large Message
The NonStop i (L-series) Expand-over-IP product has a new feature, Expand Large Message.
This feature increases the maximum message request data size and reply data size from 60
Kbytes to 2 MB for message system transfers across Expand-over-IP.
For more information, see the Expand Configuration and Management Manual (H06.21+, J06.10+).
This feature is only available on this and subsequent RVUs.
24 New Products and Features for This RVU
Gen9 CLIM
The Gen9 CLIM is the latest generation of CLIM and the successor to the Gen8 CLIM. It supports all of the functionality of previous CLIM models. It also supports several features, including using a hotfix update (as do Gen8 CLIMs) to install security updates to the CLIM without having to install a new CLIM DVD, and the ability to sanitize the CLIM hard disk when reimaging it for increased security. For more information about the hotfix, see the OSM CLIM Management Tool.
For more information, see the latest edition of the CLuster I/O Module (CLIM) Installation and
Configuration Guide (H06.16, J06.04+). For migration and fallback considerations, see the L16.05
Release Version Update Compendium.
Measure Application Profiling
Measure Application Profiling introduces several new enhancements for the L16.05 RVU.
For more information, see the Measure Reference Manual. For migration considerations, see the L16.05 Release Version Update Compendium.
NonStop Application Direct Interface (NSADI)
NonStop Application Direct Interface (NSADI) is a new product for the L16.05 RVU that supports low-latency InfiniBand-based communication between applications running on NonStop X systems and Linux servers using Remote Direct Memory Access (RDMA).
The API layer provides an InfiniBand-based interface for the applications running in a 64-bit OSS user environment using the PUT model. Linux applications may use any of the RDMA interfaces provided by the OS vendor (L16.05 supports connectivity to Linux servers running RHEL).
For information about this feature, see the NonStop Application Direct Interface Reference Manual.
NonStop cF Secure Tape
The L16.05 RVU offers the HPE NonStop cF Secure Tape product. This is a software-based encryption solution for securing NonStop backup tapes. It supports physical tape devices as well as virtual tape devices. This product is distributed on T0967.
For more information, see the NonStop cF Secure Tape Reference Manual.
NonStop cF Data Security
The L16.05 RVU offers the HPE NonStop cF Data Security (cFDS) product. Using cFDS in combination with HPE SecureData, you can protect sensitive data on NonStop systems using the HPE Format Preserving Encryption (HPE FPE) and HPE Secure Stateless Tokenization
(SST) technologies. cFDS and NonStop XYGATE Data Protection offer similar functionalities.
For more information, see the NonStop cf Data Security Reference Manual.
NonStop Development Environment for Eclipse 6.0
The NSDEE Server (T0961) adds support for launching debug sessions in NSDEE on Windows clients with IPv6 addresses. The Inspect Subsystem (T9226) adds support for launching debug sessions for terminals connected to clients with IPv6 addresses. The Native Inspect Debugger
(T1237) and TNS/X Native Inspect Debugger (T0903) add support for new NSDEE 6.0 features in the machine interface.
NOTE:
You may use NSDEE 6.0 with L16.05 or earlier L-series RVUs. However, some NSDEE
6.0 debugging features require Native Inspect SPR T1237H01^ABJ or T0903L01^AAD on the
NonStop hose. Note also that Program Initiated debug launch is only supported on L15.08 and later L-series RVUs.
L16.05
25
For more information about NSDEE 6.0, see the NonStop Development Environment for Eclipse
6.0 User Guide, the NonStop Development Environment for Eclipse 6.0 Debugging Supplement, or the online help delivered with the product.
For migration considerations, see the L16.05 Release Version Update Compendium.
NonStop Server for Java 8 and DDL2Bean
The L16.05 RVU provides these new features and enhancements:
•
All the features of Oracle JDK 8.0 that are applicable to NonStop are made available in
NonStop Server for Java 8.0.
•
G1GC has been made a generally available feature of NonStop Server for Java 8.0.
•
The fastExec option defaults to true in NonStop Server for Java 8.0.
•
NSJ infrastructure introduces a new tool, DDL2Bean, which is used to represent and use
DDL in various formats for cross-language interoperability.
For more information, see the NonStop Server for Java 8.0 Programmer’s Reference.
NonStop SSH
For L16.05, NonStop SSH has been enhanced to encrypt passwords in the SSH database with a stronger encryption algorithm. SSH now allows different ways to map file names with an extension to Guardian file names during SFTP file transfers to a NonStop system. SSH now supports DEVICE TYPE ENQUIRY system messages for the basic terminals provided by the
SSH2 process when a client does not request a PTY allocation for an SSH terminal session.
For more information, see the NonStop SSH Reference Manual. For migration and fallback considerations, see the L16.05 Release Version Update Compendium.
NonStop SSH Library
The L16.05 RVU contains the HPE NonStop cF SSH-Lib secure library. This is an add-on library to NonStop SSH which provides a programmatic interface for controlling the NonStop SSH client from within a NonStop application. This allows customers to easily extend NonStop applications to be able to communicate over an SSH connection to a peer system. The library enables users to manage a subsystem, monitor status, or control resources on a remote host.
For more information, see the NonStop cF SSH Library Reference Manual.
NonStop SSL
For L16.05, NonStop SSL has been enhanced as follows:
•
RemoteProxy 5.n has been enhanced to provide a more intuitive GUI (Graphical User
Interface) and extended to provide support for stronger ciphers and full TLS 1.2 support.
•
To ease configuration of certificates, the new parameter CERTSTORE allows users to store certificates in one subvolume without having to specifically configure them.
•
OpenSSL’s base version is updated to 1.01s. To strengthen security, SSLv2.0 is no longer supported.
For more information, see the NonStop SSL Reference Manual. For migration considerations, see the L16.05 Release Version Update Compendium.
NonStop SSL Libraries
The L16.05 RVU contains the recently launched HPE NonStop cF SSL-Lib and the HPE NonStop cF SSL-AT secure libraries. These SSL-based libraries share the TLS/SSL engine with HPE
NonStop SSL (T0910).
26 New Products and Features for This RVU
You can use NonStop cF SSL-Lib by make minor modifications to application source code. This product provides an option for applications to see the remote IP address and port number of the connection, unlike the SSL proxy-based approach where applications only see the loop-back address. Tighter integration into the application results in less processing overhead and thus increased performance.
For more information, see the NonStop cF SSL AT Library Reference Manual.
NonStop System Console Changes
The latest NSC DVD update applicable to your system is available to upgrade existing NSC software. It includes upgrades to several HPE and third-party software products. For more information, see the NonStop System Console Installer Guide and the L16.05 Release Version
Update Compendium.
NonStop XYGATE Data Protection
The L16.05 RVU contains the HPE NonStop XYGATE Data Protection (XDP). Using XDP in combination with HPE SecureData, you can protect sensitive data on NonStop systems using the HPE Format Preserving Encryption (HPE FPE) and HPe Secure Stateless Tokenization
(SST) technologies. XDP makes it possible to implement this data-centric security without modifying the application code on NonStop. XDP is distributed in T1364.
For more information, see the XYGATE Data Protection Reference Manual. For fallback considerations, see the L16.05 Release Version Update Compendium.
Open Control Block (OCB) Limits Increase
With the L16.05 RVU, L-series systems are now allowed up to 524,282 (512K-6) opens per DP2 volume. This new limit is supported locally and remotely by the #FILEGETLOCKINFO and
#OPENINFO
TACL functions and the FILE_GETLOCKINFO_ and FILE_GETOPENINFO_ Guardian
APIs, but is not fully supported by the #LOCKINFO TACL function and the superceded LOCKINFO and OPENINFO APIs. Previously, file system information interfaces supported no more than
65,279 opens per DP2 volume (locally or remote).
Customers who take advantage of this new feature may be impacted, depending on their application and TACL script usage of the Guardian APIs and TACL macros, which provide file information on opens and locks.
For more information, see the Guardian Procedure Calls Reference Manual. For migration considerations, see the L16.05 Release Version Update Compendium.
OSM Enhancements for L16.05
OSM Service Connection T0682L02^BAE includes:
•
Support for Gen9 CLIMs, including a new Configure SNMP Credentials action and Software
Patch Version attribute on the CLIM object.
•
The OSM CLIM Management Tool has been updated to support Gen9 CLIMs.
•
Support for Java 8, Update 60. For more information, see the OSM Configuration Guide.
•
To more easily identify the running firmware version, a Blade Ethernet Switch object has been added under the c7000 Enclosure object for NonStop BladeSystems.
•
A Start Monitoring NSC action, located on the Monitored Service LAN Devices object, to enable OSM monitoring of NonStop System Consoles on the service LAN.
•
ACL (Access Control List) changes to allow designated non-super users to perform configuration changes on CLIMs. For more information, see the Cluster I/O Protocols (CIP)
Configuration and Management Manual.
•
The OA Certificate tool now supports 2048 bit certificates.
L16.05
27
OSS Core Utilities
For L16.05, the Open System Services (OSS) Core Utilities product (T1202H01^AAG) contains two new packages. For more information and for migration and fallback considerations, see the
L16.05 Release Version Update Compendium.
OSS Directory Traversal Performance Improvements
For L16.05, infrastructure has been added to improve the performance of common OSS directory traversal operations. For more information, see the Open System Services Programmer's Guide.
For migration and fallback considerations, see the L16.05 Release Version Update Compendium.
OSS User and Group Management Utilities
The Open System Services (OSS) System Utilities product (T8628H01^ABD) introduces five new utilities:
• groupadd
• groupmod
• groupdel
• listusers
• passwe
For information about these utilities, see the Open System Services Management and Operations
Guide. For fallback considerations, see the L16.05 Release Version Update Compendium.
OSS Utilities Changes
Effective with SPRs T8626H04^ADX and T8628H01^ABD, over twenty utilities are moved from
T8626 to T8628. However, their version information (VPROCs) now reflects T8628.
For a list of these utilities and for migration and fallback considerations, see the L16.05 Release
Version Update Compendium.
Pathway TS/MP Release 2.6
For L16.05, Pathway TS/MP Release 2.5 has been updated with these features:
•
GET-LINK — Command to acquire links in advance. Acquires links to a serverclass in advance even then there is no incoming Pathsend request.
•
SCLONGPROCESSNAME – For 5-character system-generated server process names,
PATHWAY parameter allows server process names to be created with five characters
(excluding $) instead of four characters.
•
STOP SERVER command with FORCED option. This forcefully stops the server processes
(including associative server process started by this PATHMON) without waiting for them to perform cleanup or to terminate themselves.
•
Prevent PATHTCP/2/3/4 from dialing out during PATHTCP startup. PATHTCP2/3/4 is prevented from dialing out when it does not receive a valid startup message from PATHMON within ten minutes.
For more information, see the HPE NonStop TS/MP 2.5 Release Supplement and the TS/MP
2.5 System Management Manual.
SQL/MX Release 3.4
L16.05 supports SQL/MX 3.4, which provides several enhancements, including availability improvement for utilities, new security features, cloud enablement, and manageability feature improvements.
28 New Products and Features for This RVU
For detailed information about SQL/MX Release 3.4, including migration and connectivity issues, see the SQL/MX 3.4 Installation and Management Guide and SQL/MX 3.4 Reference Manual.
Telco CLIM Auto Failover
This enhancement provides infrastructure to INS to support the ability to failover Telco CLIM services to another Telco CLIM in a manner similar to the failover capabilities of the IP CLIM.
For configuration information, see the Cluster I/O Protocols (CIP) Configuration and Management
Manual and the INS documentation.
TNS/X C/C++ Compiler Support for 2011 C
With the L16.05 RVU, new SSE intrinsic functions are added to improve programming with AES intrinsics and the TNS/X C/C++ Compiler now supports the 2011 C (C11) standard.
For more information, see the C/C++ Programmer’s Guide. For migration and fallback considerations, see the L16.05 Release Version Update Compendium.
L17.02
This section provides information about new products and enhancements that can affect the software installation and upgrade steps for L17.02. See the L17.02 Release Version Update
Compendium for more information about these and other L17.02 products and feature enhancements and their migration and fallback considerations.
C/C++ Compiler
Starting with L17.02, the TNS/X C/C++ compiler supports C++11 (the ISO/IEC 14882:2011 standard), which is designated VERSION4.
For more information about C/C++ compile changes, including migration and fallback considerations, see the L17.02 Release Version Update Compendium.
CLIM Multi-Tenancy
Starting with L17.02, NonStop introduces support for multi-tenancy on NonStop X Gen9 IP and
Telco CLIMs. Multi-tenancy is the partitioning of a physical network into a number of virtual, or tenant, networks.
For more information about CLIM multi-tenancy, see the Cluster I/O Protocols (CIP) Configuration
and Management J- and L-Series Technical Update.
iTP Secure Webserver System 7.6
iTP Webserver is enhanced to store all the passwords in encrypted format only
For more information about iTP Secure Webserver System 7.6 changes, including migration and fallback considerations, see the L17.02 Release Version Update Compendium.
Measure
For L17.02, Measure has been enhanced to:
•
Improve large data files analysis performance with the user of 64-bit addressable memory in MEASFH.
•
Improve OSS name translation accuracy by resolving MIDs during measurement.
For more information, see the Measure Reference Manual.
L17.02
29
NonStop Server for Java 8.0
For ported binaries on HPE Integrity NonStop Servers, NonStop SE has been enhanced as follows:
•
Precompilation of methods (compilealso) provides an option to compile Java methods on their first invocation instead of compiling them after the invocation count reaches the hotspot count.
•
The installation method for both NSJ 8 32-bit and NSJ 8 64-bit is modified.
•
New tools including jmap, jinfo, jstack, and jstatd.
For more information about NonStop Server for Java 8.0, see the NonStop Server for Java 8.0
Tools Reference Pages. For migration and fallback considerations, see the L17.02 Release
Version Update Compendium.
NonStop SSH
For L17.02, NonStop SSH includes these enhancements:
•
Add support for RSA key lengths of 3072 bits and 4096 bits for host and user. Use of these two larger key sizes may impact message processing performance.
•
Change default algorithm for new host keys from DSA to RSA and default key length from
1024 to 2048 bits.
•
Add support for key exchange algorithm diffie-hellman-group-exchange-sha256 set by parameter ALLOWEDKEYEXCHANGEALGORITHMS.
•
Restrict maximum number of user authentications for incoming connections via new parameter
MAXAUTHTRIES with a default of no restriction.
•
Convert SSH database to use a different value of CUSTOMER, allowing merging of SSH databases encrypted with different CUSTOMER values.
For more information, see the HPE NonStop SSH Reference Manual.
NonStop SSL
For L17.02, six cipher suites have been removed from the list of default cipher suites. For a summary of these removed cipher suites and SSL migration and fallback considerations, see the L17.02 Release Version Update Compendium.
Open System Services (OSS)
For L17.02, OSS changes include:
•
Enhanced security for OSS files: Owned ACLs have been introduced to allow centralized management of OSS file permissions and to eliminate the security risk of inadvertent file modification by file owners.
•
New open source software and features: The superuser do utility(commonly referred to as sudo
) has been added to the OSS Core Utilities package T1202.
For more information about OSS changes, including migration and fallback considerations, see the L17.02 Release Version Update Compendium.
OSM Changes
For a summary of OSM changes for L17.02 on vNS, see
30 New Products and Features for This RVU
Safeguard
For L17.02, Safeguard has been enhanced to:
•
Protect SQL/MP objects through DISKFILE and DISKFILE-PATTERN Access Control List
(ACL). For more information, see the Safeguard Reference Manual.
•
Include the details of security group membership of the user/alias in the output of INFO
USER/ALIAS commands. For more information, see the Safeguard Reference Manual and
Safeguard Management Programming Manual.
•
Support local and personal customization of SAFECOM sessions through the user of
SAFELOCL and SAFECSTM files respectively. For more information, see the Safeguard
Reference Manual and the Safeguard Administrator’s Manual.
For more information about OSS changes, including migration and fallback considerations, see the L17.02 Release Version Update Compendium.
SQL/MX 3.5
Starting with L17.02, SQL/MX 3.5 is provided to support these features:
•
Database Compatibility Features: SQL/MX 3.5 supports additional language, SQL, and datatype constructs which make application migration easier. For more information, see the
SQL/MX 3.5 Release Technical Update.
•
Database Services and Multi-tenancy: The mxdbs command line utility provides database services for adding or deleting a user database in multi-tenant environments. For more information, see the SQL/MX Database Services Manual.
•
MXOAS: Enhanced to support multi-tenant environments. For more information, see the
SQL/MX 3.5 Release Technical Update.
•
User Management: SQL metadata is extended to represent users, and groups of users, known to SQL. For more information, see the SQL/MX 3.5 Release Technical Update.
•
Schema Level Privileges: Schema level privileges allow the schema owner to grant or revoke privileges for an entire schema to a user or group of users. For more information, see the
SQL/MX 3.5 Release Technical Update.
•
PL/MX and User-Defined Functions: PL/MX is a procedural language that is similar in many respects to Oracle PL/SQL and ANSI SQL/PSM. It is the implementation language for user-defined functions (UDFs) in NonStop SQL/MX. For more information, see the Procedural
Language for NonStop SQL/MX (PL/MX) Reference Manual.
•
MXCS AF_UNIX Protocol: ODBC/MX supports AF_UNIX protocol for communication between
OSS ODBC/MX driver and ODBC/MX Server to ensure that database connections are not impacted with CLIM failures. For more information, see the SQL/MX 3.5 Release Technical
Update.
•
The SQL/MX Database Manager (MXDM): MXDM has been enhanced to support new datatypes, PL/MX functions, creating and dropping store procedures in Java, changing ownership of SQL/MX objects, support for Materialized Views, configuration of Workload
Management Services, and support for muti-tenant isolation. For more information about these features, see the MXDM User Guide for SQL/MX Release 3.5
For more information about SQL/MX 3.5, see the NonStop Server for Java 8.0 Tools Reference
Pages. For migration and fallback considerations, see the L17.02 Release Version Update
Compendium.
L17.02
31
TMF64 Auxiliary Trail
For L17.02, TMF 64 has increased the maximum audit trail limit from 16 to 64. For more information, see the TMF Reference Manual, TMF Application Programming Manual, TMF
Management Programming Manual, and L17.02 Release Version Update Compendium.
Virtualized NonStop
L17.02 supports the new HPE Virtualized NonStop.
For detailed information, see the Virtualized NonStop Deployment and Configuration Guide.
For more information, see the Virtualized NonStop Deployment and Configuration Guide.
32 New Products and Features for This RVU
4 Upgrade Checklist
NOTE:
For vNS, see the Virtualized NonStop Deployment and Configuration Guide.
Use the upgrade checklist in
to track your progress as you perform the tasks required to upgrade to this RVU and the checklists in
and
to record the information needed when performing these tasks.
CAUTION:
The upgrade steps must be performed in the order specified in this chapter.
NOTE:
When you perform these upgrade checklist tasks, see the NonStop System Console
(NSC) Installer Guide for more information. For information about using the CLIM Management
Tool, see the OSM online help for the CLIM Management Tool, the Cluster I/O Protocols (CIP)
Configuration and Management Manual, the CLuster I/O Module (CLIM) Software Compatibility
Reference, and the NonStop Firmware Matrices.
NOTE:
If you are migrating to this RVU from an earlier L-series RVU, use the OSM CLIM
Management Tool and the Prepare for Down System CLIM action from the OSM Service
Connection to upgrade your CLIM firmware and software in parallel.
Required Information Checklists
Gather the information in
before you start. Obtain firmware levels from the CLuster I/O
Module (CLIM) Software Compatibility Reference.
Table 2 Required Information Checklist, Before You Start
Type of Information
Volume where SYSnn subvolumes are located (usually
$SYSTEM)
Current SYSnn (before upgrading to this RVU) c7000 enclosure IB switch firmware
OA firmware level
CLIM software
CLIM firmware
InfiniBand switch firmware
DSM/SCM product version (server only)
Processor type
Value
Record the information in
while you perform the upgrade.
Table 3 Information to Gather During the Upgrade (Checklist)
Value Type of Information
SYSnn created during this RVU Build/Apply
Volume where ZDSMSCM subvolume is located (by default,
$DSMSCM)
Volume where audit trails are located (by default, $AUDIT)
Name of saved configuration file
Required Information Checklists 33
Upgrade Checklist
Go through this list in order and check off each task in the Done column after it is complete.
Table 4 Upgrade Checklist for NonStop X Systems
Task Description
1. Perform all the necessary pre-installation tasks before updating the system console and receiving the SUT.
Reference
“Preparing to Upgrade the System Software
(Preinstallation Tasks)” (page 37)
2. Before you receive and install the
SUT, first install the DSM/SCM server and NSE 5.0 client product
SPR and SWID SPR from the SUT to your host system.
“DSM/SCM and this RVU” (page 17)
Appendix B: “Copying the DSM/SCM, NonStop Software
Essentials, and SWID SPRs from the SUT Using
3. Back up important files.
4. Record the current SYSnn.
“Back Up Important Files” (page 37)
“Record the Current SYSnn” (page 38)
and
Done?
5. Save current system configuration file.
“Save the Current Configuration File (CONFxxyy)”
6. Install required or recommended fallback SPRs and node interoperability SPRs, if necessary.
“Using DSM/SCM With NonStop Software Essentials to
7. Upgrade system console software.
???
8. If you are migrating a NonStop X
NonStop X Core Licensing Guide
system to this RVU to use six cores NonStop X NS7 Planning Guide instead of the four cores that were used with the previously-installed
L-series RVU, first update to this
RVU. Then use CLICVAL and the
OSM action “Read Core License” to validate and enable the 6-core license file.
9. Install the latest Halted State
Services firmware on the NonStop system console, if necessary.
Refer to the NSC DVD for the HSS installers.
10. Verify system operations.
11. Install CLIM software on the
NonStop system console.
“Verify System Processes” (page 38)
• Cluster I/O Protocols (CIP) Configuration and
Management Manual (installation procedures)
• CLuster I/O Module (CLIM) Software Compatibility
Reference (CLIM software and firmware compatibility information)
12.Receive this RVU SUT into the
DSM/SCM software archive.
13. Create this RVU software configuration revision.
“Receive New Software Into the Archive” (page 47)
“Create a New Software Revision” (page 50)
14. Using NonStop Software
Essentials, build and apply this RVU software revision.
“Build and Apply the New Software Revision” (page 52)
15. Print DSM/SCM Operator
Instructions.
“Print Operator Instructions From the Target Interface
34 Upgrade Checklist
Table 4 Upgrade Checklist for NonStop X Systems (continued)
Task Description
16. Run the Prepare for Down
System CLIM Management action in the OSM Service Connection.
Reference
Refer to:
•
“Managing CLIM Firmware and Software” (page 21)
• CLIM Management Tool online help
• OSM Service Connection online help
Done?
17. Run ZPHIRNM.
18. If OSS is being managed by
DSM/SCM, stop the OSS subsystem.
OSS Management and Operations Guide
19. Halt the processors. If the node being upgraded is part of an existing
IB cluster, advise your HPE authorized service provider to perform the necessary steps.
“2. Halt the Processors” (page 69)
20. (Optional) Update system firmware.
“3. (Conditional): Update the System Firmware”
NonStop Firmware Matrices
21. Update the Halted State
Services firmware using the HSS and HCA Firmware Management
Tool, if necessary. This will automatically perform a hard reset on the processors.
“4. (Conditional): Update the HSS Firmware” (page 70)
NonStop Firmware Matrices
22. Update c7000 Enclosure firmware, if necessary.
This update is done through a Field Change Order
(FCO).
23. Upgrade the CLIM software and/or firmware on all CLIMs using the CLIM Management Tool.
Refer to:
•
“Managing CLIM Firmware and Software” (page 21)
• CLIM Management Tool online help
24. Load the system with this new
RVU.
25. Verify the system's core license file.
Use the OSM Read Core License action to verify that the core license is valid.
IMPORTANT: If your license file is invalid or missing, you must correct this problem. For more information, refer to the NonStop X Core Licensing Guide.
26. Updating the remaining firmware on the server blades, if necessary.
This update is done through a Field Change Order
(FCO).
“Update NonStop TCP/IP Configuration Files” (page 81)
27. Recover TCP/IP files, if necessary.
28. Check processor status using the System Startup Tool.
“Check Processor Status Using the OSM System Startup
29. Start system software.
30. Configure and start OSS, if necessary.
31. Configure and start SMF, if necessary.
“Start System Applications” (page 78)
“Configure SMF (Optional)” (page 83)
Upgrade Checklist 35
Table 4 Upgrade Checklist for NonStop X Systems (continued)
Task Description
32. Configure and start RDF, if necessary.
“Set Up iTP WebServer” (page 84)
Reference
“Configure the Remote Database Facility” (page 84)
34. Re-deploy NonStop SOAP 4, if installed on the previous RVU.
“Re-deploy NonStop SOAP 4” (page 84)
35. Start applications.
Done?
36 Upgrade Checklist
5 Preparing to Upgrade the System Software
(Preinstallation Tasks)
NOTE:
For vNS, see the Virtualized NonStop Deployment and Configuration Guide.
Before updating the system console and receiving the SUT, perform these tasks:
Back Up Important Files
Before you upgrade, create tape backups of critical host applications and files.
Back Up NonStop SQL/MP
HPE recommends that you always have a current backup copy of the
$SYSTEM.SYSTEM.ZZSQLCI2 file in case you need to initialize SQL later if you have problems with your new RVU during or after installation.
1.
Save a copy of the existing ZZSQLCI2 file on the $SYSTEM disk. For example:
> FUP DUP $SYSTEM.SYSTEM.ZZSQLCI2,$SYSTEM.BKUPZSQL.*, SAVEALL
2.
Enter:
> SQLCI INITIALIZE SQL; EXIT;
3.
Copy the ZZSQLCI2 file saved previously back to the $SYSTEM disk. For example:
> FUP DUP $SYSTEM.BKUPZSQL.ZZSQLCI2, $SYSTEM.SYSTEM.*, SAVEALL
Back Up NonStop SQL/MX files
See the latest edition of the NonStop SQL/MX Release 3.x Installation and Upgrade Guide for information about backing up SQL/MX files.
Back Up Spooler Control files
Before executing the SPOOLCOM;SPOOLER,DRAIN command:
•
Use the SPOOLCOM;COLLECT $collector-process, LISTOPENS command to identify processes that currently have jobs open.
•
Use the SPOOLCOM;JOB (STATE PRINT) command to determine which jobs are currently printing.
Take the appropriate action to ensure that these jobs either terminate when finished or are stopped.
CAUTION:
To maintain the integrity of the spooler environment, wait until the Spooler has finished draining rather than stop any spooler processes by using the TACL STOP command.
The control file can be corrupted during the next warm start or rebuild.
An example of backing up the spooler control files:
BACKUP/OUT SPLBKUP/$TAPE,($SYSTEM.SPOOLER.*,$DATA01.SPOOLER.DATA),LISTALL
Save TMF Configuration Files
Use the OBEYFORM option available in TMFCOM INFO commands to save current configuration information in command files.
Back Up Important Files 37
Back Up CLIM Configuration Files
You back up the CLIM configuration files for each CLIM using the TACL command CLIMBKUP.
NOTE:
If you need to fall back to an earlier L-series RVU, you must restore these saved configuration files because the configuration file version must match the NonStop X system RVU.
For upgrades, you can specify a unique filename in the same command. For example,
>$SYSTEM NCLIM000 21> CLIMBKUP.NCLIM000.$system.nclim000.bk0522
Record the Current SYSnn
Identify and record the current SYSnn subvolume on your system. You will have to specify this
SYSnn subvolume if you need to fall back from this RVU.
At a TACL prompt, issue the status command to determine the current SYSnn subvolume. For example:
3> status 0,0
System \ELI
Process Pri PFR %WT Userid Program file Hometerm
0,0 201 P R 000 255,255 $SYSTEM.SYS01.OSIMAGE.$YMIOP.#CLCI
Swap File Name: $SYSTEM.#0
Save the Current Configuration File (CONFxxyy)
A saved system configuration file contains a copy of the current configuration database CONFIG file, saved for future use. For fallback purposes, save a copy of the current CONFIG file by using a unique file name. Do not omit this step.
You can use the saved configuration file to return to an earlier, stable configuration. If you need to fall back to the previous RVU, you must load the system by using a saved configuration file and the SYSnn containing the previous operating system.
To save the current system configuration database CONFIG file, use the SCF SAVE command:
1.
At a TACL prompt, start SCF:
12> SCF
2.
List the saved configuration files. At the SCF prompt, enter:
-> FUP INFO $SYSTEM.ZSYSCONF.CONF*
3.
Save the current configuration by using a unique file name. At the prompt, enter:
-> SAVE CONFIGURATION xx.yy where xx.yy is a number from 0.0 through 99.99.
This example saves the current system configuration to the file $SYSTEM.ZSYSCONF.CONF1101:
-> SAVE CONFIGURATION 11.1
The configuration file $SYSTEM.ZSYSCONF.CONF1101 has been created.
Verify System Processes
Check Processor Status Using the OSM System Startup Tool
To check processor status, use the System Startup Tool.
1.
Start the OSM System Startup Tool.
The Processor Status information for the NonStop X system is displayed.
38 Preparing to Upgrade the System Software (Preinstallation Tasks)
2.
The Processor Status for all the processors should be Executing NonStop OS .
Check System Components
To check the status of system components:
1.
Log on to the OSM Service Connection.
See the OSM Configuration Guide if you have any problems accessing OSM. After the browser session begins, you can close the original IE window.
Verify System Processes 39
Under “OSM Service Connection,” you can select from any one of the displayed system components to view more information about that component or group of components. In this example, ME Unit MEU01 has been selected.
The system name box in the Physical view indicates the overall health of a system by the color of the system icon. Green indicates no problems reported by OSM. Yellow indicates a possible system problem.
2.
Double-click to expand the object further to identify components that require attention (yellow or a yellow arrow) or immediate service (red).
CAUTION:
If a red triangle appears over an object, or if any component is highlighted in red, contact your HPE authorized service provider.
40 Preparing to Upgrade the System Software (Preinstallation Tasks)
3.
Double-click the Enclosure object for the c7000 enclosure.
4.
For each Logical Processor in the Enclosure object:
a.
Select the Logical Processor.
b.
In the Attributes tab, verify the values of the following attributes:
•
Halt Flag should have a value of False
•
Processor State (from ME) should have a value of Running NSK
•
Logical Processor State should have a value of Up
NOTE:
If the Halt Code or Halt Flag attributes do not match these values, see the
Processor Halt Codes Manual.
Verify the State of the InfiniBand Network Fabric
To verify that the paths and connections within the InfiniBand network are properly functional:
1.
Log on to the OSM Service Connection.
2.
In the tree pane, expand the System object to check for alarms on both Internal InfiniBand
Fabric objects.
3.
Run the InfiniBand Path Test on both Internal InfiniBand Fabric objects.
4.
Diagnose and resolve any problems that are causing alarms or are causing the path test to fail. If necessary, contact your HPE authorized service provider before continuing with the migration.
For more information, see
“Managing Firmware on NonStop X Systems” (page 21)
.
Verify Firmware
Use the OSM Service Connection to verify that all updateable firmware is at the proper revision level for your current RVU. For more information on updating firmware, see
Firmware Matrices.
Verify System Processes 41
Verify Critical System Processes
From a TACL prompt, log on to the system using the super group ID (255,255). Enter the LISTDEV command:
SCF — T9082L01 – (18SEP15) (31OCT15) – 11/22/2015 10:09:59 System \CNSKQA4
(C) Copyright 1986-2014 Hewlett–Packard Development Company L.P.
LDev Name
0 $0
1 $NCP
3 $YMIOP
5 $Z0
6 $SYSTEM
7 $ZOPR
48 $ZZKRN
50 $ZZWAN
51 $ZZSTO
69 $ZZCIP
81 $ZZWAN
87 $ZTC3
88 $ZTC2
89 $ZTC1
90 $ZTC0
96 $ZPTY
97 $ZPMON
101 $ZNET
114 $ZM03
115 $ZM02
116 $ZM01
117 $ZM00
118 $ZLOG
121 $ZFAB
122 $ZEXP
136 $ZCM03
137 $ZCM02
138 $ZCM01
139 $ZCM00
140 $ZCLA
142 $ZACS
143 $Y0SQA3
PPID
0,3
0,14
0,5
BPID
1,3
1,12
1,5
Type
(62,6 )
( 6,4 )
RSize Pri Program
( 1,0 ) 102 201 \CNSQA4.$SYSTEM.SYS00.OPCOLL
1 199 \CNSQA4.$SYSTEM.SYS00.NCPOBJ
80 205 \CNSQA4.$SYSTEM.SYS00.TMIOP
0,7 1,7 ( 1,2 ) 102 200 \CNSQA4.$SYSTEM.SYS00.OCDIST
0,257 1,257 ( 3,51) 4096 220 \CNSQA4.$SYSTEM.SYS00.TSYSDP2
0,8 1,8 ( 1,0 ) 102 201 \CNSQA4.$SYSTEM.SYS00.OAUX
0,341 1,515 (66,0 ) 4096 180 \CNSQA4.$SYSTEM.SYS00.OZKRN
0,339 1,397 (50,3 ) 132 180 \CNSQA4.$SYSTEM.SYS00.WANMGR
0,340 1,516 (65,0 ) 4096 180 \CNSQA4.$SYSTEM.SYS00.TZSTO
0,439 1,377 (74,0 ) 132 200 \CNSQA4.$SYSTEM.SYS00.CIPMAN
0, 299 1,373 (50,3 ) 132 180 \CNSQA4.$SYSTEM.SYS00.WANMGR
0,1163 3,1189 (48,0 ) 57344 201 \CNSQA4.$SYSTEM.SYS00.CIPSAM
0,441 1,157 (48,0 ) 57344 201 \CNSQA4.$SYSTEM.SYS00.CIPSAM
2,442 1,158 (48,0 ) 57344 201 \CNSQA4.$SYSTEM.SYS00.CIPSAM
0,440 1,519 (48,0 ) 57344 201 \CNSQA4.$SYSTEM.SYS00.STN
0,1508
0,443
(46,0 )
(24,0 )
6144 150 \CNSQA4.$SYSTEM.SYS00.SCP
4096 146 \CNSQA4.$SYSTEM.SYS00.OSSMON
0,18
2,329
2,326
1,15 (50,63) 132 201 \CNSQA4.$SYSTEM.SYS00.SCP
(45,0 ) 132 201 \CNSQA4.$SYSTEM.SYS00.QIOMON
(45,0 ) 132 201 \CNSQA4.$SYSTEM.SYS00.QIOMON
1,331
0,343
(45,0 ) 4024 150 \CNSQA4.$SYSTEM.SYS00.QIOMON
( 1,0 ) 57344 201 \CNSQA4.$SYSTEM.SYS00.QIOMON
0,436 1,533 (66,0 ) 4024 150 \CNSQA4.$SYSTEM.SYS00.EMSACO
0,341 (63,30) 132 175 \CNSQA4.$SYSTEM.SYS00.OZKRN
0,15
3,367
1,22 (74,1 ) 57344 201 \CNSQA4.$SYSTEM.SYS00.OZEXP
(74,1 ) 57344 201 \CNSQA4.$SYSTEM.SYS00.CIPMON
2,385
1,418
0,448
0,437
(74,1 ) 57344 201 \CNSQA4.$SYSTEM.SYS00.CIPMON
(74,1 ) 57344 201 \CNSQA4.$SYSTEM.SYS00.CIPMON
(74,1 ) 57344 201 \CNSQA4.$SYSTEM.SYS00.CIPMON
( 1,0 ) 4024 150 \CNSQA4.$SYSTEM.SYS00.EMSACO
1,30 0,107 (73,0 ) 3900 148 \CNSQA4.$SYSTEM.SYS00.BC
0,365 1,376 (63,0 ) 3 199 \CNSQA4.$SYSTEM.SYS00.LH0BJ
Verify that Disk Drive Paths Are in the Started State
1.
Enter the SCF STATUS DISK command:
> SCF STATUS DISK $*,SEL NOT STARTED
2.
Verify that the STATUS DISK display does not show any disk drive primary, backup, mirror, and mirror-backup paths that are not in the STARTED state:
$SYSTEM TOOLS 48> SCF STATUS DISK $*
SCF - T9082H01 - (01MAY08) (15MAR08) - 09/06/2015 10:55:16 System \ATLAS
(C) 1986 Tandem (C) 2015 Hewlett Packard Development Company, L.P.
STORAGE - Status DISK \ATLAS.$SYSTEM
LDev Primary
6 *STARTED
Backup
STARTED
Mirror
STOPPED
STORAGE - Status DISK \ATLAS.$AD01
LDev Primary Backup Mirror
152 *STARTED STARTED STOPPED
STORAGE - Status DISK \ATLAS.$ADSAV
MirrorBackup Primary Backup
STOPPED
PID
0,257
PID
1,257
MirrorBackup
STOPPED
Primary
PID
0,333
Backup
PID
1,288
42 Preparing to Upgrade the System Software (Preinstallation Tasks)
LDev Primary
149 *STARTED
Backup
STARTED
Mirror
STOPPED
MirrorBackup Primary Backup
STOPPED
PID
0,330
PID
1,291
Total Errors = 0
$SYSTEM TOOLS 49>
Total Warnings = 0
Verify that Tape Drives Are in the Started State (If Necessary)
You can use SCF commands or OSM to obtain status information about a tape drive. Check that the tape drive is online before continuing with installing SPRs, RVUs, or RVURs. OSM does not test tape libraries.
1.
Enter the SCF STATUS TAPE command:
> SCF STATUS TAPE $*
2.
Verify that the STATUS TAPE display includes a least one tape drive configured and running, as shown:
STORAGE - Status TAPE $T0150
LDev
71
State
STARTED
Sub State
UP
Primary Backup
PID PID
0,282 1,273
Device Status
READY
3.
Use the SCF STATUS TAPE, DETAIL command to display the current state of the tape drive.
4.
If the status of the tape drive is stopped or not ready, start the tape drive and put the tape drive online.
Verify That the CLIMs Are in the Started State
1.
Enter the SCF STATUS CLIM command:
-> STATUS CLIM $ZZCIP.*
2.
Verify that the state of all configured CLIMs is STARTED:
CIP Status CLIM \MYSYS.$ZZCIP.*
Name Config State
NCLIM000 Yes STARTED
NCLIM001 Yes
NCLIM002 Yes
STARTED
STARTED
NCLIM003 Yes
NCLIM005 Yes
SCLIM000 Yes
SCLIM001 Yes
STARTED
STARTED (Y)
STARTED
STARTED
Trace
OFF
OFF
OFF
OFF
OFF
OFF
OFF
Verify That the CIP Providers Are in the Started State
1.
Enter the SCF STATUS PROV command:
-> STATUS PROV $ZZCIP.*
2.
Verify that the status of all providers is STARTED:
CIP Status PROVIDER
Name
ZTCO
ZTC1
ZTC2
Status
STARTED
STARTED
STARTED
TPStatus
Started
Started
Started
Trace
OFF
OFF
OFF
Verify System Processes 43
ZTC3
ZTC5
ZTCPO
ZTCP1
STARTED
STARTED
STARTED
STARTED
Started
Started
Started
Started
OFF
OFF
OFF
OFF
Check the Status of the Spooler Collector
Check that the spooler collector is active and not 100 percent full.
At a TACL prompt, start SPOOLCOM and issue a COLLECT, STATUS command. For example:
\CSCALE.$SYSTEM,STARTUP 13> spoolcom $spls; collect
COLLECT STATE
$S ACTIVE
And this:
FLAGS CPU
0 , 1
PRI
160
UNIT
16
DATA FILE
$MAINT.SPL.DATAFILE
%FULL
0
\CSCALE.$SYSTEM,STARTUP 13> spoolcom $spls; collect;status detail
COLLECTOR: $S
STATE: ACTIVE
LAST ERROR: NONE
PROGRAM FILE: $SYSTEM.SYSTEM.CSPOOL
CPU: 0
BACKUP: 1
PRIORITY: 160
DATA FILE: $MAINT.SPL.DATAFILE
PAGESIZE: 60
UNIT SIZE: 16
$S
ALLOCATED UNITS: 4
TOTAL UNITS: 2071
PERCENT FULL: 0
COLLECTOR: $S STATE
ACTIVE 0 , 1
FLAGS CPU
160 16
PRI UNIT DATA FILE
$MAINT.SPL.DATAFILE
0
%FULL
Check for Sufficient Swap File Space
HPE recommends that you never allow the $SYSTEM disk to become full. There are a number of reasons, but for kernel-managed swap files (KMSF), the processor that the system is brought up on and loaded from requires at a minimum 512 MB of free $SYSTEM disk space. This default swap file will automatically be created should one not exist. In addition, if no user configured swap files have been created, each processor that is loaded using the RELOAD utility will also require 512 MB of free $SYSTEM disk space and swap files will be automatically created should they not already exist.
Always ensure that there are enough user-configured kernel-managed swap files of sufficient size—preferably not off of $SYSTEM—so that all system and application software processes can run. The optimal amount of swap space will vary depending on the activity preformed in each processor. In addition, one must use a super-group user (255,n) in order to manage the KMSF subsystem. Adding, altering, stopping and deleting swap files will prevent other super-group users from making similar changes to the configuration of the KMSF subsystem.
The best practice is to monitor swap file usage. See the Kernel-Managed Swap Facility (KMSF)
Manual for information on configuring the KMSF subsystem.
To check the size of the configured swap files for each processor:
1.
At a TACL prompt, start NSKCOM:
$SYSTEM STARTUP 5> nskcom
NSKCOM - T5838H01 BASE (01OCT07) - Oct 15 2015
Copyright 1995 Compaq Computer Corporation
$SYSTEM.SYSTEM.ZSYSCFG
KMS.SWAPFILE = 0 $SYSTEM.ZSYSSWAP.SWAP00
KMS.SWAPFILE = 1 $SYSTEM.ZSYSSWAP.SWAP01
44 Preparing to Upgrade the System Software (Preinstallation Tasks)
KMS.SWAPFILE = 2 $SYSTEM.ZSYSSWAP.SWAP02
KMS.SWAPFILE = 3 $SYSTEM.ZSYSSWAP.SWAP03
2.
Issue the NSK STATUS SWAPFILE * command to display the peak pages reserved so that you can compare that information to the total amount of swap space available.
$SYSTEM TOOLS 50> nskcom STATUS SWAPFILE *
Status of $SYSTEM.ZSYSSWAP.SWAP00
CPU 0
CPU Pages: Total 102305* Reserved 14936 Available 87369
Peak CPU Pages ever reserved 101855
Threshold 86959 CPU pages
* Default swapfile extents may not be fully allocated.
Status of $SYSTEM.ZSYSSWAP.SWAP01
CPU 1
CPU Pages: Total 102305* Reserved 12213 Available 90092
Peak CPU Pages ever reserved 15239
Threshold 86959 CPU pages
* Default swapfile extents may not be fully allocated.
Status of $SYSTEM.ZSYSSWAP.SWAP02
CPU 2
CPU Pages: Total 102305* Reserved 6361 Available 95944
Peak CPU Pages ever reserved 7280
Threshold 86959 CPU pages
* Default swapfile extents may not be fully allocated.
Status of $SYSTEM.ZSYSSWAP.SWAP03
CPU 3
CPU Pages: Total 102305* Reserved 5242 Available 97063
Peak CPU Pages ever reserved 6863
Threshold 86959 CPU pages
* Default swapfile extents may not be fully allocated..
3.
Stop the NSKCOM utility.
exit
$SYSTEM STARTUP 6>
Back Up System Information
To ensure that you always have a backup of your $SYSTEM, consider saving a $SYSTEM disk before you perform the SUT upgrade process. You can do this by running $SYSTEM un-mirrored during the upgrade process and reviving it after you are confident that the RVU upgrade completed successfully. You can also remove and store $SYSTEM-m and use a third disk to re-mirror
$SYSTEM to provide constant fault tolerance. In the unlikely event that your RVU upgrade fails, you can abandon your existing $SYSTEM and load exclusively on the stored $SYSTEM disk.
To safely remove a running $SYSTEM mirror drive, perform these steps:
CSCALE.$SYSTEM.STARTUP 2> scf
SCF - T9082L01 - (18FEB15) (31OCT14) - 06/24/2015 15:49:07 System \CSCALE
(C) Copyright 1986-2014 Hewlett-Packard Development Company L.P.
(Invoking \CSCALE.$MAINT.ZQATOOLS.TIPCFSCF)
1-> status $system
STORAGE - Status DISK \CSCALE.$SYSTEM
LDev Primary Backup Mirror
6 *STARTED STARTED
MirrorBackup Primary Backup
PID PID
STARTED *STARTED 0,257 1,257
2-> info $system,bad
Back Up System Information 45
STORAGE - Bad Sector Info Magnetic DISK \CSCALE.$SYSTEM
Bad Sectors Information $SYSTEM Primary:
No bad sectors found.
Bad Sectors Information $SYSTEM Mirror:
No bad sectors found.
3-> control $system, refresh
4-> stop $system-m
STORAGE W01008 DISK \CSCALE.$SYSTEM-M contains 255 open files.
Are you sure you want to STOP DISK $SYSTEM (Y/[N])y
5-> stop $system-mb
STORAGE W01008 DISK \CSCALE.$SYSTEM-MB contains 256 open files.
Are you sure you want to STOP DISK $SYSTEM (Y/[N])y
6-> status $system
STORAGE - Status DISK \CSCALE.$SYSTEM
LDev Primary Backup Mirror
6 *STARTED STARTED STOPPED
MirrorBackup Primary Backup
PID PID
STOPPED 0,257 1,257
Upon completion of these commands, you can now safely remove this disk.
Make Sure the Core License File is Compatible With this RVU
After you upgrade a NonStop X system from an earlier L-series RVU to this RVU, use the CLICVAL tool and the OSM action “Read Core License” to validate and enable the core license file from the new RVU, particularly when a license file for more cores is required. For more information, see the NonStop X Core Licensing Guide and the NonStop X NS7 Planning Guide.
For information about core licensing on vNS, see the Virtualized NonStop Deployment and
Configuration Guide.
If Necessary, Manually Install the NonStop Software Essentials 5.0 Server
Product
Use NonStop Software Essentials to upgrade your RVU.
See the NonStop Software Essentials Installation and Quick Start Guide for all of the NonStop
Software Essentials server installation procedures, including the manual procedure. See the
NonStop Software Essentials User’s Guide for detailed information about the product.
46 Preparing to Upgrade the System Software (Preinstallation Tasks)
6 Installing a SUT for this RVU Using NonStop Software
Essentials 5.0
NOTE:
For vNS, see the Virtualized NonStop Deployment and Configuration Guide.
NOTE:
If you do not have NonStop Software Essentials installed, see the NonStop Software
Essentials Installation and Quick Start Guide.
Alerts for Installing a SUT
Do not use this procedure to upgrade more than one system at a time or to install SPRs. For
SPRs, see
“Using DSM/SCM With NonStop Software Essentials to Install SPRs” (page 110) .
Before you install a SUT or RVU, these requirements must be met:
•
If OSS files are to be managed by DSM/SCM, the target volume must first be loaded before
OSS files are installed so that it is the active $SYSTEM. Then check that the OSS Root
Name Server $ZPNS is started and running on that active $SYSTEM. See the Open System
Services Management and Operations Guide or, if using EasySetup, see the Open System
Services Installation Guide.
•
Save the configuration database. See
“Save the Current Configuration File (CONFxxyy)”
•
Verify your system operations. See
“Verify System Processes” (page 38) .
•
If necessary, update the system console. See
“NonStop System Console Requirements”
•
Obtain a user name and password authorized to use NonStop Software Essentials. See the
NonStop Software Essentials User’s Guide for more information about configuring which users are allowed to use DSM/SCM and which tasks they are permitted to perform.
NOTE:
Safeguard aliases are case-sensitive, and they must be entered exactly when you log onto NonStop Software Essentials.
You can add a Safeguard alias anytime Safeguard is running.
Before using an existing alias for NonStop Software Essentials, check that Safeguard is started. Aliases are available only while Safeguard is running.
•
Pause the hometerm configured for use by DSM/SCM Pathway servers and other processes when DSM/SCM is running. The default hometerm is $ZHOME. See the NonStop Software
Essentials User’s Guide for more information about the hometerm used by DSM/SCM
Pathway servers.
•
For those OSS products that are not managed by DSM/SCM, you must install them manually using PINSTALL or COPYOSS. Reference those product installation procedures for the proper method to use.
Back Up Important Files
•
If you have not already done so, you should back up important files (such as, TMF, NonStop
SQL/MP, NonStop SQL/MX, CLIM, and so forth). See
“Back Up Important Files” (page 37) .
•
Drain the Spooler and back up spooler control files.
Receive New Software Into the Archive
1.
Start TMF and DSM/SCM if they are not already running. For example:
Alerts for Installing a SUT 47
> TMFCOM START TMF
> VOLUME $dsmscm_vol.ZDSMSCM
> RUN STARTSCM startscm
NOTICE: * If the hometerm chosen during INITENV, $zhome,
* is not currently in a paused state, STARTSCM and DSM/SCM
* server processes may hang indefinitely.
Be sure this
* hometerm has been paused!
PATHWAY CONTROL FILE DATED: 18 OCT 2015, 09:56:08
$Z038: WARN - *1079* SERVER CB-001, NO STATIC SERVERS DEFINED
$Z038: WARN - *1079* SERVER IS-001, NO STATIC SERVERS DEFINED
$Z038: WARN - *1079* SERVER MISERVER, NO STATIC SERVERS DEFINED
$Z038: SERVER OM-001, STARTED
$Z038: SERVER SM-001, STARTED
$Z038: SERVER SM-T-001, STARTED
$Z038: SERVER T9154-HELP-SERV, STARTED
$Z038: WARN - *1079* SERVER TA-001, NO STATIC SERVERS DEFINED
$Z038: WARN - *1079* SERVER TISERVER, NO STATIC SERVERS DEFINED
$Z038: WARN - *1079* SERVER US-001, NO STATIC SERVERS DEFINED
$Z038: TCP TCP-H, STARTED
$Z038: TCP TCP-T, STARTED
2.
Open a NonStop Software Essentials window in a browser. For instructions, see the NonStop
Software Essentials Installation and Quick Start Guide.
3.
When the Logon appears, enter the user ID or alias and password and click Logon.The
NonStop Software Essentials home page is displayed.
4.
Click Receive Software under the Software Installation group on the home page.
The Receive Software Request page appears. For example:
5.
In the Receive options group box, do one of, depending on your system environment:
•
Select Receive only changed files (default)
•
Select Receive all files
6.
Under Software name, select Provided by planner and enter a name for the RVU.
7.
Click the Input Source tab.
8.
Select Tape drive.
9.
From the list, select the tape drive that contains the SUT.
10. Click the Scheduling tab. Make the appropriate selections.
48 Installing a SUT for this RVU Using NonStop Software Essentials 5.0
11. Click Submit. The request status page appears and is updated automatically.
A Status window appears and is updated automatically at regular intervals.
12. While waiting for the request to finish, check:
•
The status of the request in the Status window. For example:
•
EMS event messages to see if a cartridge on a non-autoloading drive of the SUT needs to be mounted. EMS messages are displayed in the EMS Events window displayed by
NonStop Software Essentials when you click the system name in the System column in the table from the Status page. See the above screen shot.
•
You can also check the Requests window to ensure that the request is finished. (Select
Requests>Show Requests.) For example:
13. Expand the SUTs node in the navigation tree and monitor the request status page. When the request completes, the new SUT will appear at the top of the list of SUTs.
14. Keep NonStop Software Essentials open for the next task.
Receive New Software Into the Archive 49
Create a New Software Revision
When you create a new software revision, NonStop Software Essentials selects the products in the SUT that are licensed for your site and replaces any previous versions of the products.
The following steps assume that you are creating a new software revision that manages OSS files. If you are not, go to step
after you have loaded the Configuration Revisions page in
NonStop Software Essentials. These instructions also assume that you have installed OSS. To configure OSS for the first time, see the Open System Services Installation Guide. For details on modifying a DSM/SCM target, see the NonStop Software Essentials User’s Guide.
1.
If OSS files are to be managed by DSM/SCM, the target volume must be system loaded before OSS files are installed so that it is the active $SYSTEM. Then check that the OSS
Root Name Server $ZPNS is started and running on that active $SYSTEM.
\CNSKQA4.$SYSTEM.STARTUP 3> STATUS $ZPNS
System \CNSKQA4
Process
$ZPNS
$ZPNS
Pri PFR %WT Userid Program file
0,491 199 P 015 255,255 $SYSTEM.SYS01.NS
Swap File Name: $SYSTEM.#0
B 1,456 199 P 015 255,255 $SYSTEM.SYS01.NS
Swap File Name: $SYSTEM.#0
Hometerm
$ZHOME
$ZHOME
2.
In NonStop Software Essentials, load the Configuration Revisions page and then select the target configuration.
3.
Click the Modify button in the Target Maintenance group box.
a.
Check that the Manage OSS Files option is selected.
b.
Click the Output tab to view the default output values.
c.
Enter the appropriate values in the data fields.
NOTE:
The alternate name for the $SYSTEM drive defaults to $FC1 for fibre channel drives or $SCSI for SCSI drives. If this default has been changed, then the target volume specified in NonStop Software Essentials for new subvolumes must reflect this change.
50 Installing a SUT for this RVU Using NonStop Software Essentials 5.0
4.
Click OK to return to the Configuration Revisions page.
5.
Click the Create or Edit in the New Revision group box.
The new software revision page loads into the work panel and this page lists the products in the last software revision that was built for the target system. It might take several minutes.
a.
In the New software revision dialog box, click Edit Revision Notes.
b.
A Revision Notes text box containing existing revision notes appears.
c.
Add a description describing the new software revision (such as L17.02 SUT) and click
OK.
d.
Click on SUTs in the source tree to get the list of SUTs loaded into the source table.
Select the SUT just received into the archive and click Copy to.
6.
Click Continue when the Confirmation message appears:
If there are any discrepancies between the set of products included in the previous configuration and the set in your new configuration, the popup shown below is displayed.
(These products are removed by default from the system by DSM/SCM during the Build/Apply of the new configuration.)
Create a New Software Revision 51
The dialog box lists products installed on the previous revision that are not available on the new SUT. To retain these products from the current revision into the new revision:
•
To include one or more products in the configuration to be built, select the products and click Continue.
•
To include all products, click the check box in the left column header to select all the products.
The New software revision list now displays the new revision level. For example:
7.
Keep NonStop Software Essentials open for the next task.
Build and Apply the New Software Revision
The Build process creates an activation package containing all the products listed in the new software revision.
52 Installing a SUT for this RVU Using NonStop Software Essentials 5.0
To avoid disrupting other system activities, schedule the Build to run at night and the Apply to run immediately after the build. See the NonStop Software Essentials User’s Guide for information about scheduling the Build/Apply to run at different times.
To build and apply the new software revision:
1.
In the New software revision page, click Build/Apply to display the options for your
Build/Apply.
It might take several minutes for DSM/SCM to assemble the build information. During this time, the Processing New Revision progress bar is displayed.
When the Build information is assembled, the Build/Apply Request page is displayed. It displays default options and assigns a name to the Build/Apply request. For example:
2.
Under Perform steps, select Build, apply & process snapshot.
3.
In the Activation SYSnn: box, enter a unique number to create a new SYSnn subvolume.
The value nn must be a two-digit octal number in the range %00...%77.
You can issue a FILEINFO command at a TACL prompt to view the SYSnn subvolumes that are currently on your system.
> fileinfo $system.*.osimage
4.
Make a note of the new SYSnn. You will specify this SYSnn subvolume when:
•
Using the VPROC command to verify the location and version of your firmware
•
Specifying the location of firmware for a new RVU
•
Running ZPHIRNM
•
Loading a SUT for this RVU
5.
Edit Planner instructions to operator text to provide the correct (new) SYSnn number.
6.
Select the Scheduling, Output, SYSGEN, TSV Locations, and File Placement tabs and configure as appropriate for your system. See the NonStop Software Essentials User’s Guide or the online help for more information about Build/Apply Request options.
Build and Apply the New Software Revision 53
7.
(Optional) Confirm that the Spooler is active and not 100 percent full before executing the
Build/Apply request. At a TACL prompt, start SPOOLCOM and issue a COLLECT, STATUS command. For example:
12> spoolcom $spls; collect, status
COLLECT STATE
$S ACTIVE
FLAGS
3 ,
CPU
1
PRI
154
UNIT
4
DATA FILE
$OPS0.SPL.DATA
%FULL
4
8.
In the Build/Apply Request page, click Submit Build/Apply.
DSM/SCM runs the Build and Apply steps and displays a Status window, which is automatically updated every five seconds. For example:
9.
Check EMS messages for more detailed information about your Build/Apply.
Click the system name in the System column to view EMS events on that system for this
Build/Apply request.
When the Build/Apply finishes successfully and a snapshot is received into the host database:
•
The Request activity summary in the Status window shows that the Build, Apply,
Create snapshot, and Receive snapshot activities finished. For example:
54 Installing a SUT for this RVU Using NonStop Software Essentials 5.0
•
The last message in the EMS Events window shows that the input snapshot finished successfully.
•
The Requests window shows that the request is finished. (Select Requests> Show
Requests).
10. Exit NonStop Software Essentials.
Print Operator Instructions From the Target Interface (ZPHITI)
Print the operator instructions that you edited.
1.
Enter the following from the DSM/SCM subvolume to display the TI Logon screen:
RUN ZPHITI
DSM Software Configuration Manager TI Logon
==============================================================================
User ID:_________________________
Password:__________
==============================================================================
==============================================================================
This software is licensed by Hewlett Packard Development Company L.P. and is subject to restrictions on use, disclosure and copying and is Commercial
Software. If a user is an entity of the United States Government, only
LIMITED RIGHTS or RESTRICTED RIGHTS are granted.
HP and the HP logo are trademarks of
Hewlett Packard Development Company L.P.
T6031H03 11NOV2015
(c) Hewlett Packard Development Company L.P., 1994-2004
All Rights Reserved
==============================================================================
F1-Logon
F15-Help
F8-Clear entry
SF16-Exit
Ready
2.
Enter your user ID or alias and password and press F1 to log on to the Target Interface.
The Main Menu appears:
DSM Software Configuration Manager Main Menu
==============================================================================
\WARRIOR Prior snapshot 2015-10-09:12:18:44
Print Operator Instructions From the Target Interface (ZPHITI) 55
Selection number: 0
==============================================================================
Choose an action by entering the selection number of an item from the list below.
==============================================================================
1 Create system report
2 Create system snapshot
3 Perform target activities
4 Perform audit
5 Perform database verify
9 Enable trace (for support use only)
==============================================================================
F1-Select F4-Show Messages
F15-Help F16-Return SF16-Exit
Ready
3.
From the ZPHITI Main Menu, enter 3 (Perform target activities) in the Selection number field, and press F1 to display the Target Menu - Select Target screen.
DSM Software Configuration Manager Target Menu - Select Target
==============================================================================
\WARRIOR Prior snapshot 2015-10-20:12:18:44
Selection number:
==============================================================================
Choose a target selection number from the list below. Status information is shown for requests still in progress.
==============================================================================
1 NEW CONFIG
6
7
8
2 DEFAULT
3
4
5
9
===========================More Targets PgDn/PgUp=============================
F1-Select
F15-Help F16-Return SF16-Exit
Ready
4.
In the Selection number field, enter the number corresponding to the logical target to which you applied the software (selection 2 in this example), and press F1 to display the Target
Menu - Select Action screen.
DSM Software Configuration Manager Target Menu - Select Action
==============================================================================
\WARRIOR
DEFAULT
Prior snapshot 2015-11-20:12:18:44
Last S/W applied on SYS 00 on 2015-11-22:12:17
Selection number:
==============================================================================
Choose an action by entering the selection number of an item from the list below.
==============================================================================
1 Perform/Review apply request
2 Perform backout to S/W Applied on SYS 01 on 2015-09-13:15:44
3 Review operator instructions for the last apply completed.
56 Installing a SUT for this RVU Using NonStop Software Essentials 5.0
==============================================================================
F1-Select
F15-Help F16-Return SF16-Exit
Ready
5.
Enter 3 (Review operator instructions for the last apply completed) in the Selection number field, and press F1 to display the operator instructions.
6.
Enter a printer location in the Print operator instructions field, and press F9 to print the operator instructions.
An example instruction screen might look like this:
Activation Instructions for \BITDRP BITDRP
These are general instructions that were determined from configuration attributes.
Refer to the release documentation for detailed information on migrating to this software configuration.
1. Stop all applications. Keep OSS running.
2. RUN ZDSMSCM.ZPHIRNM SYS01
(Note: replace nn with correct number)
3. Check for $0 EMS messages.
Correct any errors indicated in the $0 EMS messages.
Do not proceed until this step has been successfully completed.
4. Coldload is required.
5. Run merge_whatis if ZPHIRNM indicates it is needed.
6. Update FIRMWARE.
Run OSM (or TSM) to check firmware and perform firmware updates.
T2805 MAINTENANCE ENTITY includes firmware changes.
T2805 MAINTENANCE ENTITY includes firmware changes.
T0287 HALTED STATE SVCS includes firmware changes.
T0630 FCSA F/W includes firmware changes.
T0287 HALTED STATE SVCS includes firmware changes.
T0287 HALTED STATE SVCS includes firmware changes.
7. Restart all applications.
7.
Press SF16 to exit the Target Interface.
In the Activation instructions, you might see information regarding the merge_whatis OSS command:
Run merge_whatis if zphirnm indicates it is needed.
When you finally run ZPHIRNM, the following lines are written to the log:
Note: merge_whatis needs to be run to update the man pages.
Files in a whatis.frag directory have changed.
The merge_whatis is a shell script in /bin. It creates and updates the whatis database used by the OSS shell man, apropos, and whatis commands. You must run the merge_whatis command once against each directory that contains the man page files to be used.
You will run merge_whatis after using ZPHIRNM. If you have OSS configured to run the OSS merge_whatis utility automatically, you can ignore running merge_whatis.
Recovery for Problems Installing This RVU SUT or SPRs
If you receive an error indication in the Requests, Status, or EMS Events windows, you might have to re-initiate the Build/Apply. See the DSM/SCM Messages Manual to determine the nature and severity of the problem.
Cancel Pending Build/Apply Requests
NOTE:
If you inadvertently navigate away from the Status page, reopen it by clicking on the
Request Name link on the Request List page.
Click the Cancel button on the Request Status page.
Recovery for Problems Installing This RVU SUT or SPRs 57
The cancel request action takes a few minutes to finish.
NOTE:
If you cannot cancel the Build request through NonStop Software Essentials, use the
DSM/SCM Target Interface (ZPHITI) to cancel the request. See the NonStop Software Esssentials
User’s Guide and the online help for more information.
58 Installing a SUT for this RVU Using NonStop Software Essentials 5.0
7 Running ZPHIRNM
NOTE:
For vNS, see the Virtualized NonStop Deployment and Configuration Guide.
During the Build/Apply phase, DSM/SCM places product files in their subvolumes with temporary
(fabricated) file names so that new file names do not conflict with the file names of currently running programs.
This chapter describes how to run the ZPHIRNM program on the target system to:
•
Rename the files that currently have actual file names with fabricated names. These files become obsolete when the new files are renamed.
•
Rename new files, which have fabricated names, with their actual names.
Warning, error, and completion messages are written to a log file if you choose to create one.
Error and completion messages are also displayed in the terminal-emulation window.
After ZPHIRNM has completed renaming all OSS and Guardian files, it will run ZMODGP if it exists in the SYSnn being activated and if:
•
The configuration being activated contains a ZCNFDEFS TSV:
104 Processing TSV $SYSTEM.ZCNFDEFS Mapfile ZMP0007N (102)
•
One or more of the ZcccGP files in the ZCNFDEFS TSV are new or changed (as compared to the last configuration activated). If none of the ZcccGP files change, ZMODGP is not run.
Alerts
•
Back up copies of NonStop TCP/IP configuration files in the ZTCPIP subvolume before running ZPHIRNM.
•
Applications that use files that are being replaced should not be running. To avoid any possibility of conflict, stop all applications and subsystems except TMF and the OSS subsystem if OSS is managed by DSM/SCM.
•
You must have rename (purge) access to all files being renamed. Run ZPHIRNM with the super group ID (255,255).
•
If you are connected to an Expand network, you might see a message stating that the
NonStop SQL file, ZZSQLCI2, was not renamed during ZPHIRNM. However, on further examination of the log file, you might find that ZPHIRNM did finally rename the file, in which case you can ignore the error message and continue.
•
If you are using DSM/SCM to manage OSS files, ZPHIRNM notifies you if and when to run the merge_whatis command. Run merge_whatis after loading the new system.
•
If ZPHIRNM is run in PREVIEW mode, ZMODGP is run in TRIAL mode. However, any errors in the new and changed files will not be discovered in preview or trial mode because
ZPHIRNM in preview mode does not actually rename the files. Therefore, ZMODGP will not be run against the same files as when the real file activation is done.
•
Prior to performing an RVU migration, install DSM/SCM SPR T6031H03^ACJ or later. See
Hotstuff HS03058 for more information.
Back Up (Copy) Files in the $SYSTEM.ZTCPIP Subvolume (Best Practice)
Make backup copies of these files in the $SYSTEM.ZTCPIP subvolume before running ZPHIRNM, which overwrites them with sample versions.
Alerts 59
You might need previous versions for reference or fallback:
•
DNSBAK
•
DNSBOOT
•
DNSCACHE
•
DNSHOSTS
•
DNSLOCAL
•
DNSREV
•
DNREVBAK
•
HOSTS
•
NETWORKS
•
PORTCONF
•
PROTOCOL
•
RESCONF
•
SERVICES
•
SMTPCONF
The file names of NonStop TCP/IPv6 that reside on the SUT are:
•
ZTCPIP.SMPLHOST
•
ZTCPIP.SMPLPROT
•
ZTCPIP.SMPLNETW
•
ZTCPIP.SMPLPORT
•
ZTCPIP.SMPLRESC
•
ZTCPIP.SMPLSERV
Stop Applications
Before running ZPHIRNM:
1.
Stop DSM/SCM if it is still running. For example:
> VOLUME $dsmscm_vol.ZDSMSCM
> Run STOPSCM
STOP CNFGINFO server process $ZPHC
STOP DSM/SCM Pathway system $YPHI
PATHCOM $YPHI;SHUTDOWN !,WAIT
$Z02H: TCP TCP-H, STOPPED
$Z02H: TCP TCP-T, STOPPED
STOP DSM/SCM Alternate EMS Collector $ZPHI where $dsmscm_vol is the DSM/SCM volume.
2.
Perform steps 1 through 4 of the procedure for performing an RVU upgrade on a node in the NonStop X Cluster Solution Manual.
3.
Stop all applications and subsystems except TMF and the OSS subsystem if it is managed by DSM/SCM.
NOTE:
Your site might have macros or OSS shell scripts set up to start and stop applications, devices, and processes.
60 Running ZPHIRNM
Drain the Spooler (Optional)
Use the SPOOLCOM;SPOOLER,DRAIN command to perform an orderly shutdown of the Spooler.
Failure to do so can cause problems during a warm start of the Spooler.
Before executing the SPOOLCOM;SPOOLER,DRAIN command:
1.
Use the SPOOLCOM;COLLECT $collector-process, LISTOPENS command to identify processes that currently have jobs open.
2.
Use the SPOOLCOM;JOB (STATE PRINT) command to determine which jobs are currently printing.
Take the appropriate action to ensure that these jobs either terminate when finished or are stopped.
After receiving the SPOOLER DRAIN command, the supervisor stops accepting jobs for collection or printing (new opens are rejected with a file-system error 66—device downed). Each collector stops when it has no more open jobs. Each print process finishes printing active jobs and then stops.
After all collectors and print processes have stopped, the supervisor process stops itself. The
Spooler enters the dormant state, ready to be warm started.
CAUTION:
Never use the TACL STOP command to halt the Spooler. The control file can be corrupted during the next warm start or rebuild.
For more information on draining the Spooler, see Support Note S99086, NSK Spooler—Migration
and Draining Instructions and in the Spooler Utilities Reference Manual.
Back Up the Spooler Control Files
While the spooler environment is closed, back up the spooler control files to render a known database from which you recover in case of a catastrophic spooler failure. Make sure that you drain the Spooler:
BACKUP/OUT SPLBKUP/$TAPE,($SYSTEM.SPOOLER.*,$DATA01.SPOOLER.DATA),LISTALL
ZPHIRNM
During the Build/Apply phase, DSM/SCM places product files in their subvolumes with temporary
(fabricated) file names so that the new file names do not conflict with the file names of currently running programs. The ZPHIRNM program renames files that currently have actual file names with fabricated names and renames new files, which have fabricated names, with their actual names. ZPHIRNM is run from the DSM/SCM subvolume.
Using ZPHIRNM PREVIEW Mode
You can do a preview run of ZPHIRNM to check that the target subvolumes (TSVs) and associated map files are consistent. Running ZPHIRNM in PREVIEW mode simulates the ZPHIRNM process without actually updating files or activating the configuration. In PREVIEW mode, ZPHIRNM indicates whether TMF needs to be up when running ZPHIRNM. For example, if no SQL program files need running, TMF is not required to be running along with the ZPHIRNM program.
If ZPHIRNM is run in PREVIEW mode, ZMODGP will be run in TRIAL mode. The new ZcccGP files will be named ZFBccncc, but cannot be easily distinguished from non-ZcccGP files also named ZFBccncc.
Example of ZPHIRNM:
Statistics:
TSVs:
Drain the Spooler (Optional) 61
TSVs in new configuration ..................
129
TSVs in old configuration ..................
130
TSVs in new configuration only .............
0
TSVs in old configuration only .............
1
TSVs shared by new and old configurations ..
129
TSVs processed in new config ...............
129
TSVs processed in old config (obsolete) ....
1
Files Processed:
Files processed in new configuration ...........................
Files processed in new configuration only ......................
Files processed in old configuration only ......................
Files processed that are shared by new and old configuration ...
Files processed & unchanged between new and old configuration ..
Files processed in obsolete TSVs ...............................
Files renamed from real name to fabricated name ................
Files renamed from fabricated name to real name ................
Files renamed from real name to fabricated (ZPG) name ..........
Files renamed from fabricated name to fabricated (ZPG) name ....
4038
0
9
4038
4038
14
23
0
0
0
Total Number of files renamed ..................................
-------
23
Number of files renamed back during undo processing ............
0
Generic Process MODification(ZMODGP) is not required because no ZcccGP files in the $SYSTEM.ZCNFDEFS TSV have changed.
Process Time:
Start .......
1/29/08 16:43:20
End .........
1/29/08 16:43:37
Elapsed .....
0:00:16
Errors & Warnings:
Total Rename Errors .....
Total Rename Warnings ...
0
0
ZPHIRNM for $SYSTEM($DRP07).SYS04 finished successfully
PREVIEW mode also reveals most errors that ZPHIRNM would encounter if it were run. If errors are revealed, you can evaluate and correct them interactively.
The syntax for running ZPHIRNM in PREVIEW mode is (from the DSM/SCM subvolume):
> RUN ZPHIRNM [$vol.]SYSnn , PREVIEW [STRICT] [,option] ...
If you enter a RUN ZPHIRNM command with no arguments, ZPHIRNM help text displays information about PREVIEW mode. For more information, see the NonStop Software Essentials
User’s Guide.
Steps for Running ZPHIRNM
1.
At a TACL prompt, enter this command from the DSM/SCM subvolume:
> RUN ZPHIRNM $vol.SYSnn
62 Running ZPHIRNM
where:
•
$vol
specifies the default current system load volume, usually $SYSTEM.
•
SYSnn specifies the subvolume to which you just applied software. The SYSnn subvolume must match the activation SYSnn that you specified earlier during the
Build/Apply phase.
2.
When asked if you want to use a log file with the session, enter Y.
3.
When prompted, enter a file name for the log file:
6> run zphirnm $system.sys00
DSM/SCM Utility - ZPHIRNM - T6031H01 - (11NOV2015)
Copyright Hewlett Packard Development Company L.P. 1993-2004
Do you want to use a log file with this session? (Y/N): y
You may use any file for logging.
If you request to use a disk file that doesn't exist, it will be created as an edit file.
Please enter a filename: elist00
Log file is open, Zphirnm will now begin processing.
ZPHIRNM begins processing and provides status of its progress.
4.
If necessary, you can compare the creation dates of the whatis file to the whatis_old file to see if changes occurred:
a.
Run the more command against the whatis file: more /usr/share/man/whatis
b.
After reviewing the displayed file, exit by entering the letter q at the percentage (%) prompt.
c.
Repeat Steps a. and b. for the whatis_old file.
Verify that ZPHIRNM Ran Successfully
1.
Review the log file.
2.
If ZPHIRNM did not run successfully, correct whatever problems occurred and run it again.
Two types of error and warning messages are issued during the renaming process:
•
If a fatal error occurs, ZPHIRNM immediately issues an error message and stops.
•
If a nonfatal error occurs, ZPHIRNM issues an error message and attempts to restore the subvolume it is currently processing to its original state. A nonfatal error is one that affects processing only for a single subvolume.
◦
If the subvolume is successfully restored, ZPHIRNM continues with the next subvolume.
◦
If the subvolume cannot be restored, ZPHIRNM issues an additional error message and continues with the next subvolume.
◦
If the subvolume was successfully processed, you can ignore warning messages.
See the DSM/SCM User’s Guide for troubleshooting procedures and the DSM/SCM
Messages Manual for a complete list of ZPHIRNM messages.
3.
If the OSS subsystem is being managed by DSM/SCM, upon a successful ZPHIRNM and befor halting any processors, make sure to stop all the OSS fileset mount points. This can be done by using the OSS EasySetup (T0585) STOPOSS utility as described in the Open
ZPHIRNM 63
System Services Installation Guide or by manually stopping each mount point as described in the Open System Services Management and Operations Guide.
64 Running ZPHIRNM
8 Loading this RVU
12
13
14
15
16
8
9
10
11
4
5
6
7
1
2
3
NOTE:
For vNS, see the Virtualized NonStop Deployment and Configuration Guide.
This chapter describes how to perform a system load on this RVU from the operating system to disk into the memory of processor 0 or processor 1 in a NonStop X system.
NOTE:
For additional information about performing a system load of a NonStop system, see the NonStop Operations Guide.
By default, one $SYSTEM is already configured. You can select $SYSTEM, CLIM-Attached-Load, or an alternate system volume:
The $SYSTEM disks must contain the current RVU operating system files for the NonStop system.
identifies the possible load paths you would use to perform a
CLIM-Attached-Load.
Table 5 System Load Paths in Order of Use for CLIM-Attached-Load
Load Path Description
Primary
Primary
Backup
Backup
Mirror
Mirror
Mirror backup
Mirror backup
Primary
Primary
Backup
Backup
Mirror
Mirror
Mirror backup
Mirror backup
Data Travels From To Processor
$SYSTEM-P
$SYSTEM-P
$SYSTEM-B
$SYSTEM-B
$SYSTEM-M
$SYSTEM-M
$SYSTEM-MB
$SYSTEM-MB
$SYSTEM-P
$SYSTEM-P
$SYSTEM-B
$SYSTEM-B
$SYSTEM-M
$SYSTEM-M
$SYSTEM-MB
$SYSTEM-MB
1
1
1
1
1
0
1
1
1
0
0
0
0
0
0
0
X
X
Y
Y
Y
X
Y
X
X
X
X
Y
Y
Y
Y
Over InfiniBand
Fabric
X
A CLIM-Attached-Load attempts to load the system from a system disk in the NonStop Storage
CLIMs connected to an InfiniBand fabric.
CLIM-Attached-Load for system with two CLIMs over $SYSTEM:
Path
Primary
Backup
Mirror
Mirror Backup
CLIM
SCLIM000
SCLIM001
SCLIM001
SCLIM000
Primary LUN
101
101
N/A
N/A
Mirror LUN
N/A
N/A
201
201
65
CLIM-Attached-Load for a system with four CLIMs over $SYSTEM:
Path
Primary
Backup
Mirror
Mirror Backup
CLIM
SCLIM000
SCLIM001
SCLIM002
SCLIM003
Primary LUN
101
101
N/A
N/A
Mirror LUN
N/A
N/A
201
201
This is the System Startup Tool screen display for the system load configuration of two Storage
CLIMs.
You can configure additional alternate system disks. To create an alternate system disk, contact your HPE authorized service provider to perform the task. Then use the OSM System Startup
Tool to make the disk available. If the alternate system disk information was saved in OSM
Service Connection, it can be displayed in the Attributes tab under the System object in the OSM
Service Connection and from the OSM System Startup Tool.
66 Loading this RVU
Alerts
Before performing a system load:
•
Before halting any processor, make sure all applications are stopped, then stop TMF and, if necessary, the OSS subsystem.
NOTE:
To stop TMF, enter this at a TACL prompt:
TMFCOM STOP TMF
•
If the OSS subsystem is being managed by DSM/SCM, upon a successful ZPHIRNM and before halting any processors, make sure to stop all the OSS fileset mount points. This can be done by using the OSS EasySetup (T0585) STOPOSS utility as described in the Open
System Services Installation Guide or by manually stopping each mount point as described in the Open System Services Management and Operations Guide.
The NonStop Operations Guide contains information about the best methods for stopping applications, devices, and processes.
•
Ensure that sufficient swap file space is available for the installation. For more information, see
“Check for Sufficient Swap File Space” (page 44)
HPE recommends that you review the content on $SYSTEM and delete any files that are no longer needed to free up space. For assistance determining which files are no longer needed, contact your system operations team.
•
Do not close the System Load Startup dialog box while startup is in progress. If you attempt to close the dialog box, a message dialog box appears notifying you that the system startup process will be stopped if you continue.
•
If you stop a system startup before it has finished, the state of your system will be unpredictable. You might need to perform another system startup to enable the system to resume normal operation.
•
If you must stop the system startup process, click Abort. (The abort process should take effect within 30 seconds.)
•
In the System Load dialog box, either select all four disk paths, or check that the selected disk path is correct. If you do not select all four or select the incorrect disk path, the system does not load properly.
•
If you have problems performing a system load, see
“Recovering from a System Load Failure”
Perform Updates
1. Before Halting the Processors
Before halting the processors on your system, perform these actions as needed.
Determine If the System Firmware Version Will Need to Be Upgraded
Before you halt the processors on your system as described in
“2. Halt the Processors” (page 69) ,
find out if your system firmware needs to be updated as you prepare load to load the new RVU.
NOTE:
Before powering up the blade, first make sure that the HSS file installed on the CLIM or NonStop system console is the correct HSS version for the L17.02 RVU.
Unless you are migrating to a new system that has not yet been fully installed, you should take this opportunity to identify the latest supported system firmware versions for your system and determine which system firmware version is currently installed on your system.
Alerts 67
1.
See these chapters in the latest edition of the NonStop Firmware Matrices:
•
The chapter “NonStop L-Series Firmware Delivered on the NonStop System Console
Installer DVD” identifies and describes the firmware components that are delivered on the NonStop system console DVD.
•
The chapter “Valid System Firmware and HSS Combinations” identifies the server firmware bundle versions (which contains the system firmware and iLO firmware) and the HSS firmware versions released on the most recent System Firmware Dates by system and processor type.
NOTE:
The Server Firmware Bundle Version of the system firmware version must be compatible with (in the same table row with) the HSS firmware SPR Revision for your system.
•
The chapter “Identification of BL860c i4, BL860c i2, and rx2800 Server Firmware
Versions” identifies, by processor-specific firmware bundle version number, the system firmware versions and iLO firmware versions that are currently supported and those that are not.
2.
If the blade servers that will be used with the new RVU are already installed, identify the system firmware version currently installed on each of them. On NonStop X systems, the system firmware version is stored on the system board of each blade. Invoke iLO from the
OSM Service Connection. The iLO is displayed from a browser, from which you select
Firmware Versions to display the current system firmware version.
NOTE:
If necessary, contact your HPE authorized service provider for assistance with these procedures:
•
On NonStop X systems, invoke iLO from the OSM Service Connection. The iLO is displayed from a browser, from which you select “Firmware Versions” to display the current system firmware version.
3.
If your currently-installed system firmware version is no longer supported, perform a system firmware update as described in
“3. (Conditional): Update the System Firmware” (page 70) .
4.
If the currently-installed system firmware version is supported, you do not need to perform a system firmware update.
Determine If the HSS Firmware Version Needs to Be Upgraded
NOTE:
For HSS on vNS, see the Virtualized NonStop Deployment and Configuration Guide.
Before you halt the processors on your system as described in
“2. Halt the Processors” (page 69) ,
find out if your HSS firmware will need to be updated as you prepare to load the new RVU.
HSS firmware would need to be updated when:
•
You are migrating your current NonStop system to another NonStop system to run the new
RVU. However, you cannot definitively identify the new HSS firmware version until the new system has been fully installed and powered on.
•
You receive a support note or Hotstuff message instructing you to update your HSS firmware.
Unless you are migrating to a new system that has not yet been fully installed, you should take this opportunity to identify the latest supported HSS firmware versions for your system and determine which HSS firmware version is currently installed on your system.
1.
Identify the supported HSS firmware version(s) for your system by referring to the chapter
“Valid System Firmware and HSS Combinations” in the latest edition of the NonStop Firmware
Matrices. This chapter identifies both the server firmware bundle versions and the HSS firmware versions released on the most recent System Firmware Dates by system and processor type.
68 Loading this RVU
NOTE:
The Server Firmware Bundle Version of the system firmware version must be compatible with (in the same table row with) the HSS firmware SPR Revision for your system.
2.
Identify the HSS firmware version that is currently installed on your system’s blade servers.
Identify the current HSS firmware version running on each processor by using the OSM
System Startup Tool's Processor Status window. For detailed instructions, see the OSM online help for the System Startup Tool and “Verify Halted State Services (HSS)” in the
NonStop System Console Installer Guide. If necessary, contact your HPE authorized service provider for assistance with these procedures.
3.
If your currently-installed HSS firmware version is no longer supported, perform a system firmware update as described in
“4. (Conditional): Update the HSS Firmware” (page 70) .
4.
If your currently-installed HSS firmware is supported, you do not need to perform an HSS firmware update.
Run the Prepare for Down System CLIM Management Action in the OSM Service Connection
Before you halt the processors on your system as described in
“2. Halt the Processors” (page 69) ,
run the Prepare for Down System CLIM Management action in OSM Service Connection on the CLIMs object:
1.
Launch the OSM Service Connection client in Internet Explorer.
2.
Right-click on the CLIM object in the tree and select Actions.
3.
Select the Prepare for Down System CLIM Management action from the dropdown and press Perform Action.
4.
Enter the SYSnn of the new RVU (e.g. $SYSTEM.SYS01).
5.
Enter the IP address of the NonStop system console.
6.
Enter the username and password for the console.
7.
Press OK.
8.
When prompted to transfer the firmware files to the console, press OK.
2. Halt the Processors
You must use the OSM System Startup Tool to halt the processors whenever you migrate your host system from one RVU to another. Perform these steps:
1.
Logon and open the OSM System Startup Tool to display the Processor Status window.
Perform Updates 69
Under Processor Status, select all of the processors.
2.
From the Processor Actions list, select Halt.
3.
Click Perform action.
4.
Click OK when asked “Are you sure you want to do Halts?”
5.
When the Action Status box confirms that all processors have halted, close the Processor
Status dialog box.
3. (Conditional): Update the System Firmware
To update your system firmware:
1.
Identify the system firmware version that is currently installed on your system and determine whether or not it needs to be updated for the new RVU as described in
System Firmware Version Will Need to Be Upgraded” (page 67) .
2.
If the system firmware currently installed on your system does not need to be updated, skip this section and proceed to
“4. (Conditional): Update the HSS Firmware” (page 70)
.
4. (Conditional): Update the HSS Firmware
Make sure that the HSS firmware version on your NonStop X system needs to be updated as described in
“Determine If the HSS Firmware Version Needs to Be Upgraded” (page 68) .
70 Loading this RVU
To upgrade the HSS firmware, use the HSS and HCA Firmware Management Tool. Refer to the
OSM online help for the HSS and HCA Firmware Management Tool and the NonStop System
Console Installer Guide for detailed instructions:
1.
Uninstall the previous version of HSS firmware.
2.
Install the new HSS firmware on the latest NSC DVD update applicable to your system.
3.
After installing the new HSS firmware version on the NSC, launch the HSS and HCA Firmware
Management Tool to perform the Update HSS action to update the version of HSS running on the processor.
NOTE:
Upon successful completion of the Update HSS action, the HSS and HCA Firmware
Management Tool prompts you to choose whether or not you want to initiate a Hard Reset now.
4.
Click OK when asked “Do you really want to perform a Hard Reset on the selected processor(s)?” The HSS and HCA Firmware Management Tool performs a hard reset of the system processors.
5. Upgrade the CLIM Software and Firmware
If you have not upgraded the CLIMs online before performing this RVU upgrade, use the CLIM
Management Tool to upgrade the CLIM software and/or firmware on each CLIM now. See the
OSM online help and
“Managing CLIM Firmware and Software” (page 21)
for more information.
6. Prepare to Load the System
Before you begin to load the system, start the MR-Win6530 terminal emulator and open one window each for the CNSL session display and the CLCI session display.
7. Load the System Using the System Startup Tool
A normal system load uses the OSM System Startup Tool to load the operating system from disk into the memory of processor 0 or processor 1. Perform these steps:
1.
From the System Startup Tool, select the Start System from the Operations drop-down menu at the top of the window.
2.
In the System Startup pop-up dialog box, select CLIM attached, or a saved system disk configuration, if one exists.
a.
In the SYSnn field, enter nn, which is the number of the system subvolume that contains the version of the operating system you will load.
b.
Under SYS nn and CIIN Option, verify that the CIIN Disabled check box is unchecked.
NOTE:
HPE recommends that you use the CIIN file to assist with the startup procedure of a system. Instructions for activities such as loading CPUs and starting subsystem resources can be added to this file so that each system load is performed in a similar manner. If the system's CIIN file contains instructions under SYSnn and CIIN Option, verify that the CIIN Disabled check box is unchecked.
Perform Updates 71
3.
Under Configuration File, select Saved Version (CONFigxx.yy:).
IMPORTANT:
HPE recommends using Current (CONFIG) unless your system startup procedures require that the system load to a known saved configuration. For more information, see
“Save the Current Configuration File (CONFxxyy)” (page 38)
.
4.
Under Configuration, ensure that the correct paths are checked.
CAUTION:
The system will attempt a system load against each checked path. Once all checked paths have been exhausted and the operating system is not in an Executing state, the system load is considered to have failed. Otherwise, the system load processor will succeed by showing the state of "Executing NonStop OS." As long as the system is attempting to start up using a disk path, one is not allowed to close the System Load dialog box. Once the system load completes, whether it is successful or not, the System Load dialog box can be closed.
5.
Click Start System. The system load begins.
If you want to stop the load process, click Abort in the System Load dialog box. Wait 30 seconds for the abort to take effect.
72 Loading this RVU
6.
In the System Startup dialog, click Start System. The system load begins. The Startup
TACL and Event Stream windows should open.
7.
Close the System Startup dialog after the message Loaded System Successfully is displayed in the activity window at the bottom of the dialog box. Wait up to 30 seconds for the abort to take effect.
CAUTION:
You are not allowed to close the System Load dialog box while the system load is in progress. However, should you close the System Startup Tool, both windows will close and the load in progress will stop after the last action in progress has completed. To continue, you must start again from the System Startup Tool/Operations/Start System. The
System Startup dialog will attempt to continue where if previously left off.
Load the System Using a Saved System Disk Configuration
A normal system load consists of loading the operating system from disk into the memory of processor 0 or processor 1.
1.
Use the OSM System Startup Tool to perform the system load. See the OSM online help for more information.
Perform Updates 73
a.
From the Configuration drop-down menu under System Load Configuration, select a system load volume. You can select a saved system disk configuration like L1702 in the preceding screen capture.
b.
If you are using non-default locations for the $SYSTEM volume for a
CLIM-Attached-Load, change the CLIM name, CLIM location, and disk location using the OSM System Startup Tool as needed.
NOTE:
If you are using a system disk drive that is not yet saved as a system disk configuration (for example, L1702), change all of the Primary, Backup, Mirror and
Mirror-Backup paths to point to the physical location in the disk bay enclosure for that system drive by changing the CLIM Name, Enclosure ID, and the Bay ID as needed.
2.
In the SYSnn box, enter nn, which is the number of the system subvolume that contains the version of the operating system you will load, the version created by DSM/SCM.
a.
Under Configuration File, select Current (CONFIG) unless your startup procedures require a system load to a known saved configuration.
b.
The CIIN file must be filled and enabled if it contains the commands needed to start the system and you do not plan to manually start the system.
NOTE:
HPE recommends that you use the CIIN file to assist with the startup procedure of a system. Instructions for activities such as loading CPUs and starting subsystem resources can be added to this file so that each system load is performed in a similar manner. If the system's CIIN file contains instructions under SYSnn and CIIN Option, verify that the CIIN Disabled check box is unchecked. The startup TACL process performs the commands in the CIIN file and then logs off.
3.
Click Start System.
a.
A dialog box appears and asks if you want to continue.
b.
Click the Yes button to halt all processors and proceed with the system load.
c.
If you want to stop the load process, click Abort in the System Load dialog box. You might wait up to 30 seconds before the abort takes effect.
CAUTION:
Do not close the System Load dialog box while startup is in progress. If you attempt to close the dialog box, an OSM message box indicates that the system startup process will be stopped if you continue. If you stop a system startup before the operation finishes, the state of your system cannot be predicted. You might need to perform another system startup to enable the system to resume normal operation.
The following example shows the dialog box for a CLIM-attached load:
74 Loading this RVU
If one or more of the startup event stream windows or startup TACL windows does not launch on the system console after a few minutes (5 to 8 minutes), abort system startup, close the dialog box, reboot the system console, and restart the system load.
4.
While the system load is underway, monitor the progress from the MR-Win6530 CNSL amd
CLCI sessions. EMS events are displayed in the CNSL session and any saved start up procedures that are save in the CIIN file are displayed in the CLCI session.
Select Open the Startup TACL and For Event Streams from the Operations pull-down menu.
a.
Check the Startup TACL window to confirm that the startup scripts completed successfully.
b.
Check the Startup TACL window and the Startup Event Streams window for error messages.
c.
Close all System Startup Tool dialog windows after all steps have been completed to verify that the system loaded successfully.
Perform Updates 75
When the system load is complete, the startup TACL will appear in the CLCI session and in the Detailed box in the System Load dialog box displays:
SYSTEM STARTUP COMPLETE,
NSK RUNNING ON PROCESSOR n
5.
If the system load fails, look for halt codes or messages displayed in the Detailed box on
System Startup dialog box. For recovery procedures, see the Processor Halt Codes Manual or Operator Messages Manual.
Verify the System Load
Use the Processor Status dialog box to verify the state of processors:
1.
Use the System Startup Tool to check the processor status. When opened, the System
Startup Tool displays Processor Status information. It also displays HSS version information.
If a processor succeeded, the status will display “Executing NonStop OS.”
76 Loading this RVU
2.
Verify that the NonStop operating system is working properly by performing these tasks:
a.
Check if any part of the system startup procedures saved in the CIIN file encountered problems.
b.
If any new generic NSK processes have been created, save a new configuration
CONFxxyy using SCF SAVE CONFIGURATION.
c.
Perform any startup procedures not that are not contained in the CIIN file (for example, starting customer applications).
Perform Updates 77
9 Resuming Normal Operations
NOTE:
For vNS, see the Virtualized NonStop Deployment and Configuration Guide.
After the System Load
After performing the system load:
1.
Update firmware as described in
“Update Firmware (If Necessary)” (page 78) .
2.
Have your HPE authorized service provider perform an RVU upgrade and all necessary subsequent steps as described in the NonStop X Cluster Solution Manual.
Update Firmware (If Necessary)
Update the following firmware as necessary:
•
For any additional Server Blade firmware updates (For any additional firmware updates, the update procedures may vary based on the system type. See the documentation specific to your Nonstop system for detailed information.)
•
Any additional Blade firmware, if applicable to your system
•
InfiniBand Blade firmware
•
InfiniBand Clustering firmware
• c7000 enclosure firmware, including the Onboard Administrator (OA) firmware. See the
NonStop Firmware Matrices for the minimum firmware version required version for this RVU.
•
Update CLIM and CLIM component firmware if needed. See
, the OSM online help for the CLIM Management Tool, and the CLuster
I/O Module (CLIM) Software Compatibility Reference for details.
See
Chapter 2: “Managing Firmware on NonStop X Systems”
for a description of the firmware installation procedure.
To find out what firmware is required for your NonStop X system, see the NonStop Firmware
Matrices.
Revive the Down Half of the Mirrored Disk
If you shut down half of $SYSTEM and other mirrored disks before applying the new RVU, revive the disks now. For instructions on how to bring up the downed half of a mirrored disk, see the
Guardian User's Guide.
Your Core License
For NonStop X systems, use the OSM Read Core License action and verify that the license information is correct for your system. See the NonStop Core Licensing Guide for more information.
Start System Applications
You can now start your application software. Follow your site’s procedures for executing the system configuration and startup (command) files to accomplish this.
DSM/SCM is often started automatically by a system command file. If not, you must start it manually as described in
“Before Starting DSM/SCM” (page 112)
and
78 Resuming Normal Operations
Before Starting DSM/SCM
Initialize SQL for NonStop SQL/MP
NOTE:
ZZSQLCI2 is a temporary file that contains the SQLCI2 program. SQLCI2 is the process through which the NonStop SQL/MP conversational interface (SQLCI) communicates with the
NonStop SQL/MP executor to request various functions. After SQL initialization is complete,
ZZSQLCI2 is renamed and becomes the permanent SQLCI2 program.
HPE recommends that you always have a current backup copy of the
$SYSTEM.SYSTEM.ZZSQLCI2 file in case you need to initialize SQL later for some reason. If you cannot find a copy of ZZSQLCI2 on the $SYSTEM disk, use DSM/SCM to export the file from the software archive. (ZZSQLCI2 is a component file of the TSQLCI2 product, T9198.)
For more information, see the SQL/MP Installation and Management Guide. For information on exporting the file from the archive, see NonStop Software Essentials online help.
1.
Check that TMF is started:
> TMFCOM STATUS TMF
2.
Save a copy of the existing ZZSQLCI2 file on the $SYSTEM disk. For example:
> FUP DUP $SYSTEM.SYSTEM.ZZSQLCI2,$SYSTEM.BKUPZSQL.*, SAVEALL
3.
NOTE:
Although NonStop SQL/MP requires that you initialize SQL before you start
DSM/SCM, NonStop SQL/MX does not require this step. For information about installing
NonStop SQL/MX, see the appropriate SQL/MX Release 3.x Installation and Upgrade Guide.
Enter:
> SQLCI INITIALIZE SQL; EXIT;
4.
Copy the ZZSQLCI2 file saved previously back to the $SYSTEM disk. For example:
> FUP DUP $SYSTEM.BKUPZSQL.ZZSQLCI2, $SYSTEM.SYSTEM.*, SAVEALL
Cold Start the Pathway Configuration
1.
Verify your current configuration against the cold start configuration in PATHHOST (or
PATHTARG) to check that the current configuration reflects all changes you have made.
2.
Run the PATHCOLD macro from the ZDSMSCM subvolume to cold start the Pathway configuration.
Start DSM/SCM
1.
From the ZDSMSCM subvolume, start DSM/SCM:
> RUN STARTSCM
Because the $ZHOME reliable home-term process is not an interactive process and is always paused, no action is necessary.
2.
Press Enter. STARTSCM automatically runs the IP6031 TACL macro. This step can take several minutes.
Start System Applications 79
Run merge_whatis
NOTE:
Make sure the OSS subsystem has been started. The automated method uses the
OSS EasySetup utility as described in
“Using OSS EasySetup (T0585)” (page 83) . The
unautomated configuration and startup method manually starts the OSS subsystem as described in
“Using OSS Unautomated Configuration and Startup” (page 83)
.
After coldloading the system, you might need to run the OSS merge_whatis command.
(ZPHIRNM notifies you if you need to run it.) (The merge_whatis command is a shell script in /bin that creates and updates the whatis database used by the OSS shell man, apropos, and whatis commands.) You might have a shell script that runs the following, depending on your OSS environment:
/bin/merge_whatis
NOTE:
You can run the Pcleanup utility on any file that is managed and audited by DSM/SCM before running the merge_whatis command. However, the OSS utility, Pcleanup cleans up
OSS files, but not DSM/SCM files, in the OSS file system. Only the DSM/SCM CLEANOSS macro cleans up DSM/SCM files in the OSS file system. The CLEANOSS macro is run only if you fall back and then migrate again to a new PVU. During a Build/Apply, DSM/SCM automatically prompts you to run the CLEANOSS macro.
1.
Log on to OSS as the super group ID (255,255):
Current volume is $SYSTEM.STARTUP
$SYSTEM STARTUP 1> osh
2.
Change directories by using the cd command. For example:
/G/SYSTEM/STARTUP: cd /usr/share/man
3.
Run the merge_whatis command:
/usr/share/man: merge_whatis
4.
Enter the list (ls) command:
/usr/share/man: ls cat1 cat4 cat2 cat5 cat3
/usr/share/man: cat6 cat7 cat8
whatis
whatis.frag
whatis_old
After you run merge_whatis, the directory contains a file named whatis, containing the database. If you have previously installed OSS, a file named whatis_old is present, containing the previous version of the database.
Configure Specific Products (Post-System Load)
This section contains information about the configuration of various products such as OSS and
SMF. See product-specific documentation for more information on installation and configuration.
80 Resuming Normal Operations
NOTE:
HPE recommends that you use the CIIN file to assist with the startup procedure of a system. Instructions for activities such as loading CPUs and starting subsystem resources can be added to the CIIN file so that each system load is performed in a similar manner. If the system has CIIN content, you can create and/or update any startup procedures within CIIN.
Update NonStop TCP/IP Configuration Files
NOTE:
You must have all the NonStop TCP/IP configuration files in your $SYSTEM.ZTCPIP
subvolume. Otherwise, applications dependent on proper network connections will fail.
Alerts:
•
See
“Back Up (Copy) Files in the $SYSTEM.ZTCPIP Subvolume (Best Practice)” (page 59) .
•
Keep your current PROTOCOL file in your $SYSTEM.ZTCPIP subvolume. Update it (if necessary) with the new lines from the sample PROTOCOL file (SMPLPROT).
Step 1: Copy the Sample TCP/IP Files Into $SYSTEM.ZTCPIP (Optional)
You need to copy the sample files if you do not have any of the following NonStop TCP/IP configuration files in your $SYSTEM.ZTCPIP subvolume:
•
HOSTS
•
PROTOCOL
•
NETWORKS
•
PORTCONF
•
RESCONF
•
SERVICES
Perform a FUP DUP operation to rename the sample configuration files from this RVU SUT to the real file name.
NOTE:
Copy these configuration files only if you do not already have versions of these files.
Depending on your applications, if you copy the configuration files, you might have to insert previous network port information into each individually copied file.
The sample configuration files are named:
•
SMPLHOST
•
SMPLPROT
•
SMPLNETW
•
SMPLPORT
•
SMPLRESC
•
SMPLSERV
To copy the sample files:
>FUP DUP $SYSTEM.ZTCPIP.SMPLHOST, $SYSTEM.ZTCPIP.HOSTS
>FUP DUP $SYSTEM.ZTCPIP.SMPLPROT, $SYSTEM.ZTCPIP.PROTOCOL
>FUP DUP $SYSTEM.ZTCPIP.SMPLNETW, $SYSTEM.ZTCPIP.NETWORKS
>FUP DUP $SYSTEM.ZTCPIP.SMPLPORT, $SYSTEM.ZTCPIP.PORTCONF
>FUP DUP $SYSTEM.ZTCPIP.SMPLRESC, $SYSTEM.ZTCPIP.RESCONF
>FUP DUP $SYSTEM.ZTCPIP.SMPLSERV, $SYSTEM.ZTCPIP.SERVICES
Configure Specific Products (Post-System Load) 81
Step 2: Check the PROTOCOL File (Optional)
CAUTION:
For both CIP and NonStop TCP/IPv6 (even if running in IPv4 mode), you must make sure that the PROTOCOL file contains ipv6 and icmpv6 entries. If these entries are absent, a network failure might occur because the new socket routines present in TELSERV, LISTNER, and FTP access the entries in the PROTOCOL file.
If the PROTOCOL configuration file exists in $SYSTEM.ZTCPIP, you must add two lines from the sample PROTOCOL file (SMPLPROT). Find the lines in SMPLPROT from the SUT that refer to IPV6 and ICMPV6 and copy those lines into your $SYSTEM.ZTCPIP.PROTOCOL file. For example:
12> TYPE SMPLPROT
#
# @(#)protocols 1.1 Tandem
#
# Internet (IP) protocols
# This file is never consulted when the yellow pages are running
# ip 0 IP # internet protocol, pseudo protocol number icmp ggp tcp pup udp ipv6 icmpv6
1
3
6
12
17
41
58
ICMP
GGP
TCP
PUP
UDP
IP
ICMP internet control message protocol gateway-gateway protocol transmission control protocol
PARC universal packet protocol user datagram protocol internet protocol for v6 internet control message protocol for v6
Configure OSS
See the Open System Services Management and Operations Guide and the Open System
Services Installation Guide for information about updating and maintaining OSS database files.
See the Release Version Update Compendium for your L-series RVU and the Interactive Upgrade
Guide 2 (IUG2) for new OSS features and for a summary of migration or installation issues.
Install Specific OSS Files Using COPYOSS or PINSTALL
For those products that are not managed by DSM/SCM, you must install them using COPYOSS or PINSTALL or both.
•
If any OSS files are not managed by DSM/SCM, follow the installation procedure for each product. See the OSS Management and Operations Guide for a description of PINSTALL use or the pinstall(1) reference page in the OSS Shell and Utilities Reference Manual.
•
The COPYOSS utility is the preferred method to use over using the Guardian utility PINSTALL or the equivalent OSS utility /bin/pax.
•
If no OSS files are managed by DSM/SCM, use COPYOSS for all of the OSS pax archives in the ZOSSUTL subvolume. See the OSS Management and Operations Guide for a description of COPYOSS use or the copyoss(8) reference page in the OSS Shell and
Utilities Reference Manual. You can use PINSTALL to install any pax archives in other subvolumes. See the OSS Management and Operations Guide for a description of PINSTALL use, or the pinstall(1) reference page in the OSS Shell and Utilities Reference Manual.
See also the appropriate product’s softdoc for any pre- or post-installation instructions.
82 Resuming Normal Operations
NOTE:
To preserve the original inetd.conf file—before using COPYOSS or PINSTALL—the following steps must be done to prevent COPYOSS or PINSTALL from overwriting the original inetd.conf
file:
•
Save the original inetd.conf file by using the command:
> mv inetd.conf smplinetd.conf
•
After running COPYOSS or PINSTALL, if the original inetd.conf file needs to be restored, use the command:
> mv smplinetd.conf inetd.conf
Run the inetd Process
Run the inetd process immediately after loading the OSS product files into the OSS environment.
The inetd process listens for connections on certain Internet sockets. When a connection is found on one of its sockets, inetd determines what service the socket corresponds to and invokes a program to service the request.
For those products that are not managed by DSM/SCM, you must install them using COPYOSS or PINSTALL. See the appropriate product’s softdoc and installation instructions.
Using OSS Unautomated Configuration and Startup
Unautomated configuration and startup is described in Appendix B in the Open System Services
Management and Operations Guide.
Using OSS EasySetup (T0585)
The automated setup method uses the OSS EasySetup product (T0585) as described in the
Open System Services Installation Guide. OSS EasySetup is a series of TACL scripts that assist a site when configuring and installing a minimal OSS subsystem for the first time. The STARTOSS and STOPOSS utilities can be used to manage existing OSS subsystem resources by updating the OSSINFIL file when the new OSS mount points or servers are created.
Configure SMF (Optional)
If you plan to configure SMF, an optional product, note:
•
Migration of data from direct volumes to SMF logical volumes is documented in the Storage
Management Foundation User's Guide.
•
Performance-critical applications that optimize the physical location of their data might need to specify the PHYSVOL option when creating data files, instead of relying on the SMF file-placement algorithm.
•
NonStop SQL/MP must be up and running before you start SMF.
CAUTION:
To maintain data integrity, do not place these items on a virtual disk process
(VDP):
◦
SWAP files
◦
TMF audit data
◦
DSM/SCM files
To ensure system configuration integrity, always:
Configure Specific Products (Post-System Load) 83
1.
Save the system configuration after SMF configuration changes are made.
2.
Reload the system from the current system configuration file or the latest version with SMF changes.
See the softdoc T8472ACK or later for SMF configuration and interoperability considerations.
Configure the Remote Database Facility
The AUDITTRAILBUFFER parameter is a modifier to DP2, which improves the performance of
RDF. Ensure that an appropriate value is specified for the AUDITTRAILBUFFER modifier for the master audit-trail disk process. The AUDITTRAILBUFFER parameter allows you to specify how many megabytes of buffer space to allocate for the audit-trail buffers.
See the SCF Reference Manual for the Storage Subsystem for information on configuring the buffer size and check the DP2 softdoc for more information.
See the Release Version Update Compendium for your L-series RVU for new feature information.
Set Up iTP WebServer
If iTP WebServer, an OSS product, was installed and was being used with the previous RVU, the same iTP WebServer installation would not work after the RVU upgrade. To get the iTP
WebServer installation working (both T8996 and T8997) after an RVU upgrade, you must re-configure and set up iTP WebServer at the same OSS location using the setup script.
Re-deploy NonStop SOAP 4
If NonStop SOAP 4 was deployed and was being used with the previous RVU, the same SOAP
4 installation would not work after the RVU upgrade. To use NonStop SOAP 4 after an RVU upgrade, you must deploy the NonStop SOAP 4 version available with the new RVU in the existing
NonStop SOAP 4 deployment directory, using the NonStop SOAP 4 deploy script.
See the L17.02 Release Version Update Compendium for information about NonStop SOAP 4.2.
84 Resuming Normal Operations
10 Backing Out the DSM/SCM Revision
NOTE:
For vNS, see the Virtualized NonStop Deployment and Configuration Guide.
Products That Might Have Fallback Implications
Review the Release Version Update Compendium for this RVU and product softdocs to make sure that your installed products do not have migration and fallback implications.
Open System Services
Back up the OSS configuration database files before falling back. For more information, see the
Open System Services Management and Operations Guide.
Storage Management Foundation
The record format of the system configuration database sometimes changes between RVUs and is automatically converted to the new format the first time the system is loaded. To ensure system configuration integrity, always:
1.
Save the system configuration after SMF configuration changes are made.
2.
Reload the system from the current system configuration file or the latest version with SMF changes.
CAUTION:
Create a saved version of the configuration before loading a new RVU or revision. If this step is omitted, you cannot fall back to a previous RVU or revision.
For more information, see the Storage Management Foundation User’s Guide and the required
SMF softdocs.
Before Backing Out the Revision
1.
Confirm that the Spooler is active and not 100 percent full.
a.
At a TACL prompt, start SPOOLCOM and issue a COLLECT, STATUS command. For example:
12> spoolcom $spls; collect, status
COLLECT STATE
$S ACTIVE
FLAGS CPU
3 , 1
PRI UNIT
154 4
DATA FILE
$OPS0.SPL.DATA
%FULL
4
If the spooler is not active, start the spooler.
You will use the Spooler to print out operator instructions when backing out the revision.
2.
Start the TMF subsystem and DSM/SCM if they are not already running. For example:
4> TMFCOM START TMF
5> volume $dsmscm_vol.ZDSMSCM
6> RUN STARTSCM where $dsmscm_vol is the DSM/SCM volume.
Start the Target Interface (ZPHITI)
NOTE:
With NonStop Software Essentials 4.0 and later, you can also use NonStop Software
Essentials instead of ZPHITI to back out a software revision that has already been applied. For more information, see the NonStop Software Essentials User’s Guide or the online help. The procedure that follows describes only the ZPHITI method.
Products That Might Have Fallback Implications 85
1.
From the DSM/SCM subvolume, enter:
RUN ZPHITI
The TI Logon screen appears:
DSM Software Configuration Manager TI Logon
==============================================================================
User ID:______________________
Password:____________
==============================================================================
==============================================================================
This software is licensed by Hewlett Packard Development Company L.P. and is subject to restrictions on use, disclosure and copying and is Commercial
Software. If a user is an entity of the United States Government, only
LIMITED RIGHTS or RESTRICTED RIGHTS are granted.
HP and the HP logo are trademarks of
Hewlett Packard Development Company L.P.
T6031H01 11NOV2004
(c) Hewlett Packard Development Company L.P., 1994-2004
All Rights Reserved
==============================================================================
F1-Logon F8-Clear entry
F15-Help SF16-Exit
Ready
2.
Enter your user ID or alias and password and press F1 to log on to the Target Interface.
The Main Menu appears:
DSM Software Configuration Manager Main Menu
==============================================================================
\ELI Prior snapshot 2015-10-14:18:03:03
Selection number: 0
==============================================================================
Choose an action by entering the selection number of an item from the list below.
==============================================================================
1 Create system report
2 Create system snapshot
3 Perform target activities
4 Perform audit
5 Perform database verify
9 Enable trace (for support use only)
==============================================================================
F1-Select F4-Show Messages
F15-Help F16-Return SF16-Exit
Ready
Submit a Backout Request
1.
From the ZPHITI Main Menu, enter 3 (Perform target activities) in the Selection number field, and press F1.
The Target Menu - Select Target screen appears:
DSM Software Configuration Manager Target Menu - Select Target
==============================================================================
\ELI Prior snapshot 2015-10-14:18:03:03
Selection number:
86 Backing Out the DSM/SCM Revision
5
6
7
==============================================================================
Choose a target selection number from the list below. Status information is shown for requests still in progress.
==============================================================================
1 NEW CONFIG
2 DEFAULT
3
4
8
9
===========================More Targets PgDn/PgUp=============================
F1-Select
F15-Help F16-Return SF16-Exit
Ready
2.
In the Selection number field, enter the number corresponding to the logical target that you want to back out of (selection 2 in this example), and press F1.
The Target Menu - Select Action screen appears:
DSM Software Configuration Manager Target Menu - Select Action
==============================================================================
\ELI
DEFAULT
Prior snapshot 2015-10-14:18:03:03
Last S/W applied on SYS 00 on 2015-10-14:18:01
Selection number:
==============================================================================
Choose an action by entering the selection number of an item from the list below.
==============================================================================
1 Perform/Review apply request
2 Perform backout to S/W Applied on SYS 03 on 2015-10-14:15:41
3 Review operator instructions for the last apply completed.
==============================================================================
F1-Select
F15-Help F16-Return SF16-Exit
Ready
3.
In the Selection number field, enter 2 (Perform backout to S/W), and press F1.
The Backout - Verify Request screen appears, which shows:
•
The SYSnn, date, and timestamp of the current software revision and of the previous revision, which becomes the current revision when the backout request finishes
•
The location where DSM/SCM will place the snapshot
•
A warning that the request makes the last software applied on this target inaccessible
DSM Software Configuration Manager Backout - Verify Request
==============================================================================
\ELI
DEFAULT
Prior snapshot 2015-10-14:18:03:03
Last S/W applied on SYS 00 on 2015-10-14:18:01
Snapshot location: \ELI.$SPR09
==============================================================================
ATTENTION - This request is not reversible. Once the current configuration for this target is backed out, it cannot be made current again.
==============================================================================
Backout target revision will be performed:
From: SYS 02 applied on 2015-10-14:18:01
Submit a Backout Request 87
To: SYS 01 applied on 2015-10-17:15:41
==============================================================================
F1-Submit request
F15-Help F16-Return SF16-Exit
Ready
4.
Press F1 to submit the backout request.
DSM/SCM automatically reapplies the previous software revision.
While the backout request is processing, DSM/SCM displays the Backout - Processing
Backout screen, which gives a description of the request, its processing status, and the location of the previous SYSnn software affected by the request. The screen is updated with current status information approximately every 15 seconds, as in these examples:
DSM Software Configuration Manager Backout - Processing Backout
==============================================================================
\ELI Prior snapshot 2015-10-14:18:03:03
DEFAULT Last S/W applied on SYS00 on 2015-10-14:18:01
Backout 2000-04-18:10:59:34
==============================================================================
Request is being processed. Status information will appear below indicating progress.
==============================================================================
Backout Awaiting start.
2000-04-18:10:59:34
==============================================================================
F3-Restart F4-Show messages F6-Cancel request
F15-Help F16-Return SF16-Exit
DSM Software Configuration Manager Backout - Processing Backout
==============================================================================
\ELI
DEFAULT
Prior snapshot 2015-10-14:18:03:03
Last S/W applied on SYS00 on 2015-10-14:18:01
Backout 2015-10-18:11:00:39
==============================================================================
Request is being processed. Status information will appear below indicating progress.
==============================================================================
Backout Processing.
2000-04-18:11:00:39
==============================================================================
F3-Restart F4-Show messages F6-Cancel request
F15-Help F16-Return SF16-Exit
After the software backout is finished, the Backout - Complete Backout screen appears with operator instructions.
A complete set of operator instructions might look like this:
DSM Software Configuration Manager Backout - Complete Backout
==============================================================================
\ELI
DEFAULT
Prior snapshot 2015-10-14:18:03:03
Last S/W applied on SYS on 2015-10-14:18:01
Print operator instructions on:
==============================================================================
Placement of files for this Backout is complete. To activate the software that has been placed follow the operator instructions displayed below.
88 Backing Out the DSM/SCM Revision
===============================================================================
Activation Instructions for \BITDRP BITDRP
These are general instructions that were determined from configuration attributes.
Refer to the release documentation for detailed information on migrating to this software configuration.
1. Stop all applications. Keep OSS running.
2. Update FIRMWARE.
Run OSM to check firmware and perform firmware updates.
T0902 HALTED STATE SVCS includes firmware changes.
T0891 HALTED STATE SVCS includes firmware changes.
3. RUN ZDSMSCM.ZPHIRNM SYS00
4. Check for $0 EMS messages.
Correct any errors indicated in the $0 EMS messages.
Do not proceed until this step has been successfully completed.
5. Coldload is required.
6. Run merge_whatis if ZPHIRNM indicates it is needed.
7. Restart all applications.
==============================================================================
F4-Show Messages
F9-Print operator instructions
Task completed successfully.
F15-Help F16-Return SF16-Exit
5.
Enter a printer location and press F9 to print the operator instructions. You might need to modify these instructions before returning to the previous RVU. For instance, you might not need to change the firmware unless you have to replace hardware that can be supported only by the new RVU.
6.
Press SF16 to exit the Target Interface.
Monitor the Backout Process Using NonStop Software Essentials
1.
Log on to NonStop Software Essentials.
2.
Select Requests->Show requests.
The Requests status page appears, showing the type and status of various DSM/SCM requests.
3.
From the Request list page, click your backout request name.
The Request Status page displays detailed information about your backout request.
This example indicates that the backout request finished successfully:
Monitor the Backout Process Using NonStop Software Essentials 89
11 Running ZPHIRNM to Reactivate the Previous Software
Revision
NOTE:
For vNS, see the Virtualized NonStop Deployment and Configuration Guide.
This step runs the ZPHIRNM program again to change the temporary names of the previous product files back to their actual names.
Alerts
Before running ZPHIRNM:
•
Applications that use files that are being replaced should not be running. To avoid any possibility of conflict, stop all applications, except for the TMF subsystem, before running
ZPHIRNM.
•
You must have rename (purge) access to all files being renamed. You can ensure this access by running ZPHIRNM as the super group ID (255,255).
•
If you are connected to an Expand network, you might see a message stating that the
NonStop SQL file, ZZSQLCI2, was not renamed during ZPHIRNM. However, on further examination of the log file, you may find that ZPHIRNM did finally rename the file, in which case you can ignore the error message and continue.
•
Make sure that you drain the Spooler before running ZPHIRNM. If you drain the Spooler
after running ZPHIRNM, an error message appears indicating that the spooler program and library versions are different, and Spooler terminates with errors.
Drain the Spooler (Optional)
Before executing the SPOOLCOM;SPOOLER,DRAIN command:
•
Use the SPOOLCOM;COLLECT $collector-process, LISTOPENS command to identify processes that currently have jobs open.
•
Use the SPOOLCOM;JOB (STATE PRINT) command to determine which jobs are currently printing.
•
Take the appropriate action to ensure that these jobs either terminate when finished or are stopped.
•
Drain the Spooler. To drain the Spooler, log on as a super group user (255,n), access
SPOOLCOM, and enter the following command:
SPOOLCOM> SPOOLER, DRAIN
This command causes the Spooler to move from the active to the dormant state.
CAUTION:
Do not use the TACL STOP command to drain or stop the Spooler. Also, do not issue the TACL STOP command after you have issued the SPOOLCOM DRAIN command. The control file can be corrupted during the next warmstart or rebuild.
For more information on draining the Spooler and recovery, see Chapter 1 in the Spooler
Utilities Reference Manual.
90 Running ZPHIRNM to Reactivate the Previous Software Revision
Run ZPHIRNM
Using PREVIEW Mode
You can do a preview run of ZPHIRNM to check that the target subvolumes (TSVs) and associated map files are consistent. Running ZPHIRNM in PREVIEW mode simulates the ZPHIRNM process without actually updating files or activating the configuration. In PREVIEW mode, ZPHIRNM indicates whether or not TMF needs to be up when running ZPHIRNM.
Run ZPHIRNM
1.
At a TACL prompt, enter from the DSM/SCM subvolume:
> RUN ZPHIRNM $vol.SYSnn where:
•
$vol
specifies the default current system load volume, usually $SYSTEM.
•
SYSnn specifies the SYSnn subvolume named previously in the backout request.
For example:
> RUN ZPHIRNM $system.SYS00
2.
When asked if you want to use a log file with the session, type Y and press Enter.
3.
When prompted, type a file name for the log file, and press Enter.
4.
ZPHIRNM issues this message and begins processing:
Log file, <filename>, was successfully created.
5.
Verify that ZPHIRNM ran successfully.
6.
Review the log file.
7.
If ZPHIRNM did not run successfully, correct the problem and rerun.
Two types of error and warning messages are issued during the renaming process:
•
If a fatal error occurs, ZPHIRNM immediately issues an error message and stops.
•
A nonfatal error is one that affects processing only for a single subvolume. If a nonfatal error occurs, ZPHIRNM issues an error message and attempts to restore the subvolume it is currently processing to its original state.
◦
If the subvolume is successfully restored, ZPHIRNM continues with the next subvolume.
◦
If the subvolume cannot be restored, ZPHIRNM issues an additional error message and continues with the next subvolume.
◦
If the subvolume was successfully processed, you can ignore warning messages.
See the NonStop Software Essentials User’s Guide for troubleshooting procedures and the DSM/SCM Messages Manual for a complete list of ZPHIRNM messages.
Run ZPHIRNM 91
12 Falling Back on the CLIMs, Loading the Previous RVU, and Resuming Normal Operations
NOTE:
For vNS, see the Virtualized NonStop Deployment and Configuration Guide.
This chapter describes how to perform a system load of the previous RVU and how, as needed, to revert the CLIM software and/or firmware to the previous versions.
When You Must Reimage a CLIM or All CLIMs in a System
Before you reimage one or more CLIMs in a NonStop X system running L17.02 or a later L-series
RVU, you must consider these issues and perform these actions:
•
If you reimage a CLIM in a NonStop X system with the CLIM DVD for L17.02 (T0853L02^BAL) or later, you must also update T0682 to the latest L-series SPR version (T0682L02^BAE for
L17.02 or later).
•
After you reimage the CLIM as described in Reimaging or Downgrading a CLIM and after you have updated T0682 to an L17.02 or later SPR version, you must execute SSOCLIM against every CLIM in the system.
•
If you update T0682 to T0682L02^BAE or later, you must also execute SSOCLIM against every CLIM in the system.
You must reimage all of the CLIMs in the system with a pre-L17.02 CLIM DVD if all of these conditions are true:
•
You were previously running the L17.02 CLIM DVD (T0853L02^BAL).
•
You revert T0682 to a pre-L17.02 T0682 SPR.
•
You reimage any CLIM in the system with a pre-L17.02 CLIM DVD.
If these conditions are all true, you must reimage all of the CLIMs in the system with a pre-L17.02
CLIM DVD.
See the CLuster I/O Module (CLIM) Software Compatibility Reference for information about compatibility.
Alerts
Before performing a system load:
•
Before halting any processor, make sure all applications are stopped, stop TMF and, if necessary, stop the OSS subsystem. If the OSS subsystem is being managed by DSM/SCM, upon a successful ZPHIRNM and prior to halting any processors, make sure to stop all the
OSS fileset mount points. This can be done by using the OSS EasySetup (T0585) STOPOSS utility as described in the Open System Services Installation Guide or by manually stopping each mount point as described in the Open System Services Management and Operations
Guide.
NOTE:
To stop TMF, at a TACL prompt, enter: TMFCOM STOP TMF
•
Do not close the System Startup dialog box while startup is in progress. If you attempt to close the dialog box, a message dialog box appears to alert you that the system startup process will be stopped if you continue.
•
If you stop a system startup before it has finished, the state of your system will be unpredictable. You might need to perform another system startup to enable the system to resume normal operation.
92 Falling Back on the CLIMs, Loading the Previous RVU, and Resuming Normal Operations
•
If you must stop the system startup process, click Abort. (The abort process should take effect within 30 seconds.)
•
For detailed information on performing a system load, see the NonStop Operations Guide.
•
List your saved configuration files by entering:
FUP INFO $SYSTEM.ZSYSCONF.CONF*
You might want to refer to this list later in the procedure.
•
If you are falling back and using next generation CLIMs and SAS disk enclosures, the system to which you are falling back needs SPRs to support those components. See the CLuster
I/O Module (CLIM) Software Compatibility Reference for this information.
•
Check the list of fallback issues in the Release Version Update Compendium for the RVU to which you are falling back.
Task Overview
Should you need to fall back to the previous version of CLIM software for one or more CLIMs in your system, the steps for falling back the software for a given CLIM are included in these procedures.
If reverting to the previous CLIM software is necessary or desirable, the CLIM must be reimaged, as described in these procedures. Refer to the CLuster I/O Module (CLIM) Software Compatiblity
Reference to determine whether reverting to the previous CLIM software is necessary for your fallback scenario.
Restoring the CLIMs assumes that you have saved the CLIM configurations by issuing the
CLIMBKUP command before upgrading your system, as described in
and in the Cluster I/O Protocols (CIP) Configuration and Management Manual.
Falling back to previous versions of CLIM software adds approximately one hour for each CLIM to the process of loading the system on the previous RVU.
NOTE:
A CLIM configured with a T0853L02^BAA DVD version can only be restored with a backup of the same version. CLIMRSTR exists with errors if an incompatible backup file is specified.
NOTE:
When you fall back, you must restore CLIM configuration files saved from the same
RVU as the NonStop system RVU to which you are falling back.
The procedure for falling back on the CLIM and loading the NonStop system on the previous
RVU is:
1.
Before halting the processors:
a.
Collect information about the CLIM environment.
b.
Modify the OSSINFIL file. (OSSINFIL is an OSS file for T0585 OSS EasySetup.)
2.
Halt the processors.
3.
While the processors are halted:
NOTE:
Do not reimage both of the ME Unit CLIMs simultaneously. Reimage one ME Unit
CLIM first and configure it using the OSM System Configuration Tool. Then reimage the other ME Unit CLIM and configure it.
•
Revert the server blade firmware, hard reset the system, and reimage the CLIMs.
•
Prepare the CLIMs:
a.
Launch the System Configuration Tool.
b.
Select the Configure CLIM option.
Task Overview 93
c.
Log on to one of the ME Units for the system.
d.
Any CLIMs that have been reimaged will prompt for credentials.
e.
Verify that all the CLIMs are displayed in the table.
f.
If any CLIMs are not present, enter the IP address of the CLIM in the Start IP
Address field and press Discover CLIMs.
g.
Press Next to perform the configuration.
4.
Load the system.
5.
Once the system is loaded, complete the CLIM setup. This step involves running a script that establishes a secure relationship between the CLIMs and the system. In addition, you restore more of the CLIM configurations by running a script that restores various aspects of the CLIM configuration from the stored file.
6.
Revert firmware if necessary.
7.
Install specific OSS files if necessary.
8.
Verify that the system is started.
NOTE:
For detailed information about other tasks required to load the NonStop X system on the previous RVU, see the procedures described in
“Loading this RVU” (page 65) .
Before Halting the Processors
(Optional) Prepare to Fall Back to Previous CLIM Firmware
If you will be falling back to previous CLIM firmware using the down system method, perform the appropriate action in the OSM Service Connection, as described in
“Managing CLIM Firmware and Software” (page 21)
and the OSM online help for the CLIM Management Tool.
Prepare to Fall Back to Previous CLIM Software
NOTE:
If the system disk is a SAS disk, the SAS disk enclosures that contain the system disks
(both primary and mirror) must have LUNs assigned prior to a reimage and must not be power-cycled during the reimage process.
Following are preparatory steps for downgrading to a previous version of the CLIM software.
1.
Obtain the password for the root user ID on the CLIM.
2.
Obtain a password for the super group TACL user ID.
3.
Obtain the DNS name of each CLIM iLO. The default name is system-climname-ILO (for example, mysys-NCLIM000–ILO).
4.
If you determined that reverting to the previous CLIM software is necessary, ensure you have the CLIM software image for the RVU you are falling back to on either the NonStop system console or a CLIM DVD. If you do not have the correct software image on a system console, install it from the CLIM DVD for L-series by following the procedure in the CLuster
IO Protocols (CIP) Configuration and Management Manual.
5.
Determine if one or more of the CLIMs you are downgrading is being used for DHCP, DNS, and BOOTP services instead of the system consoles on your LAN. This step uses the OSM
Service Connection and the NonStop LAN DHCP DNS Configuration Wizard, depending on which wizard you have installed on your NonStop system console.
a.
Determine the IP address of the CLIM(s) you are going to downgrade. Use the OSM
Service Connection to identify the Maintenance Interface IP Address for the CLIM.
94 Falling Back on the CLIMs, Loading the Previous RVU, and Resuming Normal Operations
b.
To determine if one ore more of the CLIMs you are downgrading is being used for DHCP,
DNS, and BOOTP services, launch the NonStop Maintenance LAN DHCP DNS
Configuration Wizard. Whichever wizard you have installed on your NonStop system console is launched from the Windows Start menu by selecting All Programs>OSM>, followed by the name of the wizard.
NOTE:
For detailed information about the appearance of and instructions for using the wizard you have installed on your NonStop system console, see the online help for that wizard.
c.
If the wizard confirms that any of the CLIMs you are downgrading are being used for
DHCP, DNS, and BOOTP services, check the listed IP addresses to see if they correspond to the CLIMs you are downgrading. For information on how to use the wizard, see the online help for the wizard you are using.
Before Halting the Processors 95
6.
Obtain the system name, netmask, gateway, and IP address for the CLIMs. If this information is unavailable from your environmental documentation, continue with the next step to obtain it directly from your system. Otherwise, skip the next step.
7.
Obtain and record the current CLIM configuration information:
a.
Using the SCF INFO CLIM, DETAIL command, obtain the maintenance interface IP address and NonStop host system name of the CLIM(s) to be downgraded. For example:
->INFO CLIM $ZZCIP.NCLIM000, DETAIL
b.
Use the climcmd clim-name climconfig interface -info eth0 command from a TACL prompt for the CLIMs to obtain the netmask and gateway. For example:
>climcmd NCLIM000 climconfig interface -info eth0 comForte SSH client version T9999J06_07Sep2015_comForte_SSH_0080
Interface
Interface Type
: eth0
: Physical Interface
Jumbo
IP
Frame
Address
Netmask
: NO
: 192.168.37.83
: 255.255.0.0
In this example, the netmask is 255.255.0.0 and there is no gateway configured.
c.
Use the OSM System Startup Tool to connect to the CLIM and obtain PCI Slot configuration information.
d.
Make note of all values.
Halt the Processors
NOTE:
Whenever a processor is halted, the information provided under “Performing an Upgrade on a Node” in the NonStop X Cluster Solution Manual should also be considered. Contact your
HPE authorized service provider for assistance.
1.
Logon to the System Startup Tool by highlighting the affected system name, then supplying a username and password if prompted.
96 Falling Back on the CLIMs, Loading the Previous RVU, and Resuming Normal Operations
2.
Check the topmost Processor # selection box so that the System Startup Tool selects all processors.
Halt the Processors 97
3.
From Processor Actions, select Halt from the pull-down menu, then select Perform Action.
4.
When the Processor Action dialog box appears, select Yes..
5.
Within minutes, each processor’s status will report Halted by operator (Halt %100).
98 Falling Back on the CLIMs, Loading the Previous RVU, and Resuming Normal Operations
6.
When the Action Status box confirms that all processors have halted, close the Processor
Status dialog box.
While the Processors Are Halted
NOTE:
If you are falling back to a previous HSS version, use the HSS and HCA Firmware
Update Tool to load that version of HSS. Doing so will automatically hard reset the processors to pick up the new version.
NOTE:
Firmware is backward compatible. HPE recommenads that you retain current firmware, and revert to previous firmware only when necessary.
Perform a Hard Reset:
1.
Open the System Startup Tool.
2.
On the toolbar, click Processor Status to display the Processor Status dialog box.
3.
Under Processor Status, select all processors.
4.
From the Processor Actions list, select Hard Reset.
While the Processors Are Halted 99
5.
Click Perform action.
6.
Click OK when asked “Do you really want to perform a Hard Reset on the selected processor(s)?”
100 Falling Back on the CLIMs, Loading the Previous RVU, and Resuming Normal Operations
7.
Make sure the Action Status box confirms that all processors have been reset.
(Optional) Fall Back to Previous CLIM Firmware
If you will be falling back to previous CLIM firmware using the down system method, perform the appropriate action in the OSM Service Connection, as described in
“Managing CLIM Firmware and Software” (page 21)
and the OSM online help for the CLIM Management Tool.
Reimage the CLIMs
NOTE:
CLIMs are reimaged to revert their software to an older version. Check the CLuster
I/O Module (CLIM) Software Compatibility Reference to determine which versions of the CLIM are compatible with the target RVU.
Reimaging a CLIM must be performed by an HPE authorized service provider. Users can refer to Reimaging or Downgrading a CLIM for detailed instructions about reimaging CLIMs.
Prepare the CLIMs
Perform this procedure only if the CLIMs have been reimaged or the CLIM software reverted.
1.
For each CLIM, set the IP address of eth0 through the iLO serial console.
Prepare the CLIMs 101
a.
At the Linux prompt, log onto the CLIM. CLIMs are shipped with two default maintenance interface usernames and passwords:
• root/hpnonstop
• user/clim
b.
At the Linux prompt, add the interface by entering this command: climconfig interface –add eth0
c.
Add the IP address and bring up the eth0 interface by entering: climconfig ip –add eth0 –ipaddress ip-address -netmask netmask
2.
Reboot the CLIMs by entering: reboot
3.
Run the OSM System Configuration Tool and perform the Configure CLIM action. If an ME
Unit CLIM that was reimaged, perform the Change Maintenance Entity Unit action instead.
102 Falling Back on the CLIMs, Loading the Previous RVU, and Resuming Normal Operations
4.
If the system disk is an SAS disk and you have SAS disk enclosures that contain the system disks (both primary and mirror), issue the following command on one of the storage CLIMs that connect to the system disks: lunmgr ––approve yesall
5.
For the two IP CLIMs configured with a Maintenance Provider (usually the first two IP CLIMs, for example, NCLIM000/NCLIM001 in a NonStop X system), configure the eth0:0 interface.
Using the IP address and netmask of ZTCP0 and ZTCP1 you collected by using the SCF
INFO CLIM DETAIL and CLIMCMD interface -info commands prior to starting the downgrade, and using the iLO serial console, enter these commands at the Linux prompt: climconfig interface -add eth0:0 climconfig ip -add eth0:0 -ipaddress 192.168.36.10 -netmask 255.255.0.0
6.
If you reimaged both CLIMs and they were being used to host DHCP, DNS, and BOOTP services on your dedicated service LAN, as determined in
“Prepare to Fall Back to Previous
Prepare the CLIMs 103
CLIM Software” (page 94) , you must log on to the OSM HSS and HCA Firmware Management
Tool and perform the Copy HSS Files action—before loading the system—as described in
“Complete CLIM Setup” (page 106) .
Load the System
A normal system load consists of using the OSM System Startup Tool to load the operating system from disk into the memory of processor 0 or processor 1:
NOTE:
For even more detailed instructions on performing a system load, see
.
1.
Select Start System. The System Load dialog box appears.
2.
In the System Load dialog box, select $SYSTEM or a saved configuration.
a.
In the SYSnn field, enter nn, which is the number of the system subvolume that contains the version of the operating system you will load.
b.
Under SYSnn and CIIN Option, verify that the CIIN Disabled check box is unchecked.
The CIIN file must be enabled because it contains commands needed to start the server.
The startup TACL process performs the commands in the CIIN file and then logs off.
3.
Under Configuration File, select Saved Version (CONFigxx.yy:).
Refer to the list of configuration files generated before beginning this procedure.
4.
Under Configuration, ensure that the correct paths are checked.
5.
In the System Load dialog box, click Start System. The system load begins.
If you want to stop the load process, click Abort in the System Load dialog box. You might wait up to 30 seconds before the abort takes effect.
CAUTION:
Do not close the System Load dialog box while startup is in progress. If you attempt to close the dialog box, an OSM message box indicates that the system startup process will be stopped if you continue. If you stop a system startup before the operation finishes, the state of your system cannot be predicted. You might need to perform another system startup to enable the system to resume normal operation.
If one or more of the startup event stream windows or startup TACL windows does not launch on the system console after a few minutes (5 to 8 minutes), abort system startup, close the
System Startup Tool, reboot the system console, and restart the system load using the
System Startup Tool again.
6.
Monitor the system startup process. Messages indicating the progress and completion of the system load, and the reload of processors indicated in the CIIN file, appear in these display locations on the system console.
When the system load is complete (the operating system is loaded successfully into the memory of processor 0 or processor 1), the Detailed box on the System Load dialog box displays:
SYSTEM STARTUP COMPLETE,
NSK RUNNING ON PROCESSOR n
•
If the system load fails, check for the appearance of any halt codes or messages. For recovery procedures, see the Processor Halt Codes Manual or Operator Messages
Manual.
•
After the operating system is running on the server, messages are no longer sent to the Details box. See the event messages sent to the startup event stream window.
104 Falling Back on the CLIMs, Loading the Previous RVU, and Resuming Normal Operations
Complete the System Load (If Necessary)
Loading the processors with the CIIN file can be automatic if the CIIN file is not empty. If the operating system loaded successfully into processor 0 or processor 1, but the commands in the
CIIN file do not reload all remaining processors, use the following TACL RELOAD command.
You must be a super group ID (255,nnn) to load the processors.
NOTE:
The CIIN file can be updated to include steps to help bring up subsystems and, if your site allows it, parts of your applications.
NOTE:
HPE recommends that you use the CIIN file to assist with the startup procedure of a system. Instructions for activities such as loading CPUs and starting subsystem resources can be added to the CIIN file so that each system load is performed in a similar manner. If the system has CIIN content, you can create and/or update any startup procedures within CIIN.
1.
From the system console, log on to the system as a super-group user (255,nnn) and enter:
TACL> RELOAD *
2.
To load an individual processor, enter:
TACL> RELOAD nn where nn is the processor number of the processor you want to reload.
If the reload is initiated successfully, this message appears in a startup TACL window on the system console screen:
PROCESSOR RELOAD: nn
Using the SSOCLIM * Option to Run SSOCLIM on All Available CLIMs
For this RVU, SSOCLIM can be made to run on all available CLIMs by using the * option instead of the CLIM IP or CLIM name in the SSOCLIM command line:
SSOCLIM [{CLIM NAME | IP | *}{CLIM USER} {PASSWORD}][FORCED]
Where:
CLIM NAME is the name of the CLIM to be used.
IP is the IP address of the CLIM to be used.
* runs SSOCLIM on all available CLIMs.
CLIM USER is the user name to be used against a CLIM when CLIM NAME or IP is used or against all available CLIMs when * is used.
PASSWORD is the password to be used against a CLIM when CLIM NAME or IP is used or against all available CLIMs when * is used.
FORCED
, if specified, suppresses the warning instruction to run SSOCLIM on all the CLIMs.
Using the SSOCLIM * Option to Run SSOCLIM on All Available CLIMs 105
Complete CLIM Setup
1.
Establish trust between the CLIM and the NonStop host system.
If you are falling back to an earlier L-series RVU, you must run SSOCLIM as SUPER.SUPER.
In all cases, you must run SSOCLIM twice, once with the root username and hpnonstop password, and once with the user username and clim password.
a.
At a TACL prompt, log on as any SUPER group user.
b.
Enter SSOCLIM.
The script prompts you for the CLIM name, IP address or DNS name of the CLIM. Enter that parameter. Then the script prompts you for the username and password. Use the root username and hpnonstop password.
c.
Re-enter the SSOCLIM command using the user username and clim password.
2.
Use CLIMRSTR to restore all the CLIMs with the CLIM backup files from the previous CLIM version. Restore the files that you stored when you followed the CLIM software upgrade procedures in
“Back Up CLIM Configuration Files” (page 38)
.
NOTE:
The CLIM configuration files used during this procedure must be of the same RVU as the RVU to which you are falling back.
Rename and restore the backup files. For example, at the TACL prompt, enter:
a.
>FUP RENAME $DATA1.BCKJ0605.NCLIM000, $DATA1.BCKJ0605.BACKUPGZ
b.
>CLIMRSTR NCLIM000 $DATA1.BCKJ0605
c.
Repeat steps a. and b. for each CLIM.
Restore the backup files. For example, at the TACL prompt, enter:
a.
>CLIMRSTR NCLIM000 $SYSTEM.NCLIM000BCKJ0606
b.
Repeat for each CLIM.
3.
Reboot the CLIMs:
a.
For large systems, boot only one of the CLIMs from every failover pair of CLIMs in the system at the same time.
NOTE:
For IP CLIMs and Telco CLIMs, you must know which CLIMs are configured as failover pairs. For Storage CLIMS, you must know which pair of CLIMs are attached to a particular path or you will lose access to the disk.
b.
Wait for all the CLIMs to be restored and wait for disk paths or IP addresses to return before rebooting the next CLIM.
For Storage CLIMs, use SCF status disk $*. For IP and Telco CLIMs, use CLIMCMD psclim
.
>climcmd NCLIM000 psclim
If you do not see any running processes, use the CLIM command to clear the log and start. For example (IP CLIM):
>climcmd NCLIM000 clim clearlog; clim start
For Storage CLIMs, after clearlog, you must reboot again. For example (Storage
CLIM):
>climcmd SCLIM000 clim clearlog
>climcmd SCLIM000 clim reboot-nodebug
106 Falling Back on the CLIMs, Loading the Previous RVU, and Resuming Normal Operations
c.
Reboot the other set of CLIMs, that is, the other CLIM in each failover pair or the other
CLIM in each pair that connects to a particular disk.
d.
If the system disk is a SAS disk and you have SAS disk enclosures that contain the system disks (both primary and mirror), check the LUN manager status to be sure they are functional. For example:
1)
Check lunmgr status:
>climcmd clim-name lunmgr –p
2)
If the above command did not display any LUN information, issue these commands:
>CLIMCMD clim-name lunmgr -a yesall
SCF> allow all errors
SCF> reset disk $*
SCF> start disk $* -
Say Yes when it prompts
SCF> status disk $*-*, sel not started
4.
If you reimaged a CLIM being used for DHCP, DNS, and BOOTP services on your LAN, use the OSM HSS and HCA Firmware Management Tool to perform the Copy HSS Files action as described in
“Complete CLIM Setup” (page 106) .
Revert Firmware (If Necessary)
HPE recommends running the latest firmware; however, if necessary, revert the following firmware to the firmware on the SYSnnsubvolume from which you loaded the system:
•
IB switch firmware
•
IB Cluster firmware
•
Server Blade firmware (Requirements vary according to system type. See the documentation specific to your NonStop system for details.)
• c7000 enclosure firmware, if applicable to your system (Onboard Administrator firmware installed through FCO, if necessary)
• c7000 enclosure firmware, if applicable to your system
•
CLIM firmware, if your system has CLIMs attached, and you did not update CLIM firmware while the system was down
See the softdocs for details.
Install Specific OSS Files (If Necessary)
For those products that are not managed by DSM/SCM, you must install them using COPYOSS or PINSTALL or both. For details, see
“Install Specific OSS Files Using COPYOSS or PINSTALL”
Verify That the System Is Started
Launch the OSM System Startup Tool and use it to verify that each processor is executing the
NonStop operating system.
Revert Firmware (If Necessary) 107
13 Support and other resources
Accessing Hewlett Packard Enterprise Support
•
For live assistance, go to the Contact Hewlett Packard Enterprise Worldwide website:
www.hpe.com/assistance
•
To access documentation and support services, go to the Hewlett Packard Enterprise Support
Center website:
www.hpe.com/support/hpesc
Information to collect
•
Technical support registration number (if applicable)
•
Product name, model or version, and serial number
•
Operating system name and version
•
Firmware version
•
Error messages
•
Product-specific reports and logs
•
Add-on products or components
•
Third-party products or components
Accessing updates
•
Some software products provide a mechanism for accessing software updates through the product interface. Review your product documentation to identify the recommended software update method.
•
To download product updates, go to either of the following:
◦
Hewlett Packard Enterprise Support Center Get connected with updates page:
www.hpe.com/support/e-updates
◦
Software Depot website:
www.hpe.com/support/softwaredepot
•
To view and update your entitlements, and to link your contracts and warranties with your profile, go to the Hewlett Packard Enterprise Support Center More Information on Access
to Support Materials page:
www.hpe.com/support/AccessToSupportMaterials
IMPORTANT:
Access to some updates might require product entitlement when accessed through the Hewlett Packard Enterprise Support Center. You must have an HPE Passport set up with relevant entitlements.
Websites
Website
Hewlett Packard Enterprise Information Library
Hewlett Packard Enterprise Support Center
108 Support and other resources
Link www.hpe.com/info/enterprise/docs www.hpe.com/support/hpesc
Website
Contact Hewlett Packard Enterprise Worldwide
Subscription Service/Support Alerts
Link www.hpe.com/assistance www.hpe.com/support/e-updates
Software Depot
Customer Self Repair
Insight Remote Support for NonStop
Serviceguard Solutions for HP-UX
www.hpe.com/support/softwaredepot www.hpe.com/support/selfrepair www.hpe.com/info/insightremotesupport/docs www.hpe.com/info/hpux-serviceguard-docs
Single Point of Connectivity Knowledge (SPOCK) Storage compatibility matrix
www.hpe.com/storage/spock
Storage white papers and analyst reports
www.hpe.com/storage/whitepapers
Customer self repair
Hewlett Packard Enterprise customer self repair (CSR) programs allow you to repair your product.
If a CSR part needs to be replaced, it will be shipped directly to you so that you can install it at your convenience. Some parts do not qualify for CSR. Your Hewlett Packard Enterprise authorized service provider will determine whether a repair can be accomplished by CSR.
For more information about CSR, contact your local service provider or go to the CSR website:
www.hpe.com/support/selfrepair
Remote support
Remote support is available with supported devices as part of your warranty or contractual support agreement. It provides intelligent event diagnosis, and automatic, secure submission of hardware event notifications to Hewlett Packard Enterprise, which will initiate a fast and accurate resolution based on your product’s service level. Hewlett Packard Enterprise strongly recommends that you register your device for remote support.
For more information and device support details, go to the following website:
www.hpe.com/info/insightremotesupport/docs
Documentation feedback
Hewlett Packard Enterprise is committed to providing documentation that meets your needs. To help us improve the documentation, send any errors, suggestions, or comments to Documentation
Feedback ([email protected]). When submitting your feedback, include the document title, part number, edition, and publication date located on the front cover of the document. For online help content, include the product name, product version, help edition, and publication date located on the legal notices page.
Customer self repair 109
A Using DSM/SCM With NonStop Software Essentials to
Install SPRs
NOTE:
Before performing the steps in this appendix install NonStop Software Essentials by using the procedures in the NonStop Software Essentials Installation and Quick Start Guide.
SPRs provide new features between RVUs or correct software problems from a previous RVU.
This appendix describes tools available for investigating SPRs and instructions on how to install
SPRs.
CAUTION:
Do not apply any SPR to an RVU without first verifying (by using Scout for NonStop
Servers, for example) that the SPR can be installed on that specific L-series RVU.
See the Release Version Update Compendium for this RVU and the Interactive Upgrade Guide
2 for information on SPRs that you might need to apply to your current RVU before upgrading to this RVU.
Tools for Managing SPRs
Always keep an SPR log file that lists which SPRs have been reviewed, selected, and installed on your system:
•
Update the SPR log regularly to maintain accurate and current information about the SPRs installed on your system.
•
Document the reasons for a decision not to install SPRs that have been issued for products you run on your system.
Scout for NonStop Servers
Using Scout for NonStop Servers, you can display detailed information about RVUs, SPRs, and prerequisites for SPRs. You can also access:
•
Outage Prevention Notifications (OPNs)
•
Hotstuff messages
•
Support Notes
•
RVU (softdocs) and softdoc addendums
After researching available SPRs, you can download those you consider appropriate for your system directly to your workstation and then use FTP to transfer downloaded SPRs to your server, or you can request tape delivery.
DSM/SCM
DSM/SCM has automatic features that can help you keep records of SPRs. Use NonStop Software
Essentials to generate prepackaged reports, including reports that list:
•
The products on a system and the SPRs applied to those products
•
The SPRs that have been replaced by a specific SPR
•
The requisite SPRs for a selected SPR
You can also use the DSM/SCM Archive to export SPRs to different volumes.
See the NonStop Software Essentials User’s Guide for more information.
110 Using DSM/SCM With NonStop Software Essentials to Install SPRs
SPRNOTES
You can use SPRNOTES to research an SPR before installing it. Read the installation instructions and considerations documented in the SPRNOTE, which includes:
•
A summary of SPR contents
•
Lists of SPRs required for this SPR
•
Installation considerations, including tasks to perform before and after running DSM/SCM
•
A summary of problems corrected
•
A summary of known problems remaining
Keep a copy of each SPRNOTE and record the status of the related SPR. For example, record whether the SPR is not applicable, is already installed, or needs to be installed.
Install SPRs Using DSM/SCM With NonStop Software Essentials
The steps for using DSM/SCM to install SPRs are the same as those for installing a SUT.
Save the Current Configuration
First, save the current system configuration database CONFIG file. Use the SCF SAVE command:
1.
At a TACL prompt, start SCF:
2> SCF
2.
List the saved configuration files. At the SCF prompt, enter:
-> FUP INFO $SYSTEM.ZSYSCONF.CONF*
3.
Save the current configuration by using a unique file name. At the SCF prompt, enter:
-> SAVE CONFIGURATION xx.yy where xx.yy is a number from 00.00 through 99.99.
This example saves the current system configuration to the file $SYSTEM.ZSYSCONF.CONF1101:
-> SAVE CONFIGURATION 11.1
The configuration file $SYSTEM.ZSYSCONF.CONF1101 has been created.
If you specify a configuration file that already exists, SCF asks you if you want to replace it. If you specify that you do not want to replace the existing file, SCF does not overwrite the file. You must enter another SAVE command to save the current configuration file.
Determine the Current SYSnn
There are several ways to record the current SYSnn subvolume on your system.
You must use the current SYSnn subvolume if you need to fall back to your current RVU. You must use a different SYSnn subvolume to create a new software revision if NonStop Software
Essentials indicates that a SYSGEN is required.
1.
At a TACL prompt, enter:
$SYSTEM SUPER 6> sysinfo
SYSINFO - T9268H01 - (01 SEP 2015) SYSTEM \ELI Date 04 Sep 2015, 22:00:18
COPYRIGHT 2003 Hewlett-Packard Development Company, L.P.
System name
EXPAND node number
Current SYSnn
\ELI
033
SYS01
Install SPRs Using DSM/SCM With NonStop Software Essentials 111
System number 000000
Software release ID L15.08.00
2.
Issue a TACL FILEINFO command on $system.*.osimage to list the SYSnn subvolumes on your system and determine which one is open.
$SYSTEM.SYS00
OSIMAGE O
CODE EOF LAST MODIFIED OWNER RWEP
0 57354240 15OCT2015 9:57 255,255 NUNU
PExt SExt
200 200
$SYSTEM.SYS01
OSIMAGE O
CODE EOF LAST MODIFIED OWNER RWEP PExt SExt
0 57393152 05OCT2015 8:56 255,255 NUNU 200 200
3.
At a TACL prompt, enter a STATUS 0,0 command to see which SYSnn subvolume is currently being used:
3> status 0,0
System \ELI
Process Pri PFR %WT Userid Program file Hometerm
0,0 201 P R 000 255,255 $SYSTEM.SYS01.NMONTOR
$YMIOP.#CLCI
Swap File Name: $SYSTEM.#0
Receive the SPR Into the DSM/SCM Archive
Before Starting DSM/SCM
Start TMF if it is not already running:
> TMFCOM START TMF
Bring the SPR Into the DSM/SCM Archive
1.
Start DSM/SCM if it is not already running:
> volume $dsmscm_vol.ZDSMSCM
> RUN STARTSCM
2.
Open an Internet Explorer window and enter the URL for NonStop Software Essentials
(windows-system-address:9988/nse).
3.
Perform the on-screen instructions to logon.
4.
Click Receive software under the Software Installation group on the NonStop Software
Essentials home page.
5.
Click the Input Source tab.
6.
If the SPR has been received on disk or has been restored from tape to disk, click inside the Enter volume subvolume pattern input field.
7.
Enter the name of the distribution subvolume containing the SPR, and then click Find/Add.
8.
If the SPR was received on tape, click Tape and select a tape drive from the list.
9.
In the Receive Software Request dialog box, click the Scheduling tab.
10. Enter the time you want your Receive Software Request to run. The default option is for the request to run immediately. If this is what you want, click Submit.
Create a New Software Configuration That Includes the SPR
After the SPR has been received into the Archive, you can copy it. To create a new software configuration that includes the SPR:
112 Using DSM/SCM With NonStop Software Essentials to Install SPRs
1.
From the Configuration Revisions page, select the configuration you are updating and click
Create in the New Revision group. The new software revision page appears.
2.
Select SPRs to be added to the current configuration and click Copy to.
Build the New Revision
After you have created a new software revision, prepare to install it on your target systems by building it into an executable form:
1.
On the New software revision page, click Build/Apply to load the Build/Apply options page.
2.
Select Build (if not connected to an Expand network) or Build, apply & process snapshot
(if connected to an Expand network).
3.
Include in the activation package either only changed files or all files. If you are installing an
SPR, you can select changed files.
4.
In the Activation SYSnn: box, enter the current SYSnn (if no SYSGEN is required) or the new SYSnn (if SYSGEN is required).
NOTE:
If the Build/Apply Request page informs you that SYSGEN is required, check that the Spooler is active and not 100 percent full before executing the Build/Apply request. At a TACL prompt, start SPOOLCOM and issue a COLLECT, STATUS command.
5.
Select Scheduling, Output, SYSGEN, TSV Location, and File Placement tabs and configure as appropriate for your system. See the NonStop Software or online help (Building
and Applying a New Configuration topic) for more information about these Build/Apply
Request options.
6.
On the Build/Apply page, click Submit.
7.
Monitor the request from the request status page.
Print Operator Instructions From the Target Interface (ZPHITI)
1.
Make sure that Spooler is active.
2.
On the DSM/SCM subvolume, enter:
RUN ZPHITI
The TI Logon screen appears.
3.
Enter your user ID/alias and password and press F1 to log on to the Target Interface.
The Main Menu appears.
4.
Enter 3 (Perform target activities) in the Selection number field and press F1.
The Target Menu - Select Target screen appears.
5.
In the Selection number field, enter the number corresponding to the logical target on which the new software was applied and press F1.
The Target Menu - Select Action screen appears.
6.
In the Selection number field, enter 3 (Review operator instructions for the last apply completed) and press F1 to display the operator instructions.
7.
Enter a printer location in the Print operator instructions field and press F9 to print the operator instructions.
Run ZPHIRNM
During the Build/Apply phase, DSM/SCM places product files in their subvolumes with temporary
(fabricated) file names so that the new files do not conflict with the file names of currently running programs. After the Build/Apply finishes, run the ZPHIRNM program on the target system to:
•
Rename the files that currently have actual file names with fabricated names. These files become obsolete when the new files are renamed.
•
Rename the new files, which have fabricated names, with their actual names.
Print Operator Instructions From the Target Interface (ZPHITI) 113
1.
At a TACL prompt, enter this command from the DSM/SCM subvolume:
> RUN ZPHIRNM $vol.SYSnn where:
•
$vol
specifies the default current system load volume, usually $SYSTEM.
•
SYSnn specifies the current or new (depending on whether SYSGEN was required)
SYSnn subvolume.
For example:
> RUN ZPHIRNM $system.SYS00
2.
When asked if you want to use a log file with the session, enter Y and press Enter.
3.
When prompted, enter a file name for the log file.
ZPHIRNM issues this message and begins processing:
Log file is open, Zphirnm will now begin processing.
4.
Verify that ZPHIRNM ran successfully by reviewing the log file.
5.
If ZPHIRNM did not run successfully, correct the problem and rerun.
See the DSM/SCM User’s Guide for troubleshooting procedures and the DSM/SCM Messages
Manual for a complete list of ZPHIRNM messages.
If Running INSTALL^TEMPLATES Is Required
If the operator instructions state that you should run INSTALL^TEMPLATES to update EMS template files, you must be logged on as the super group ID (255,255) user to run the
INSTALL^TEMPLATES macro. This macro cannot be run with the NOWAIT option and takes approximately ten minutes to finish.
If there is a resident template file in any of the input subvolumes, the INSTALL^TEMPLATES macro displays this error:
Invalid nonresident template file -- not key sequenced: $SYSTEM.ZTEMPL.NEWRES
*** INSTALL^TEMPLATES: Terminated due to error ***
HPE recommends that you designate a unique subvolume to hold the template files created by
INSTALL^TEMPLATES. When you follow the suggested procedure, a nonresident template file is not present in ZTEMPL, and the error does not occur.
1.
Set the volume and subvolume to $SYSTEM.ZINSAIDS:
-> volume $system.zinsaids
2.
Load the TACLMACS file:
-> load /keep 1/ TACLMACS
114 Using DSM/SCM With NonStop Software Essentials to Install SPRs
NOTE:
If you run the INSTALL^TEMPLATES macro with no arguments, you are prompted to provide required information. For information on the INSTALL^TEMPLATES arguments, print the usage notes:
-> volume $system.zinsaids -> load /keep 1/ TACLMACS -> install^templates /out outdev/ help where outdev is the name of the desired output device.
3.
Choose a location on $SYSTEM to keep updated template files: tacl>volume $system.template
4.
Rename any existing templates: tacl> rename newres, oldres tacl> rename newnres, oldnres
5.
Run INSTALL^TEMPLATES to create updated template files: tacl>install^templates newres newnres $system.ztempl
NOTE:
Running this macro permanently changes the location of the EMS template files.
DSM/SCM creates new EMS templates when it creates a new SYSnn. However, when the new system is loaded, the old template files named in the last execution of
INSTALL^TEMPLATES are opened instead of the new files. To use the new RTMPLATE and TEMPLATE EMS files created by DSM/SCM, enter this SCF ALTER command:
->
ASSUME SUBSYS $ZZKRN-> ALTER, RESIDENT_TEMPLATES
$SYSTEM.SYSTEM.RTMPLATE, & NONRESIDENT_TEMPLATES
$SYSTEM.SYSTEM.TEMPLATE
The files newres and newnres are meant to be used as temporary names until the next
SYSGEN and system load.
The following error message might occur when running the INSTALL^TEMPLATES macro:
>$SYSTEM.SYSTEM.SCF /NAME $Z019,INV :INBUF.1,OUTV :OUTBUF.1/
Current Nonresident template file is: $SYSTEM.TEMPLATE.NEWNRES
TEMPLATE and NONRESIDENT files must be different: $SYSTEM.TEMPLATE.NEWNRES
*** INSTALL^TEMPLATES: Terminated due to error ***
The error message indicates that the template and nonresident files specified in the
INSTALL^TEMPLATES command are identical to those on the current subvolume. The
INSTALL^TEMPLATES macro tries to create the newres and newnres files, but stops when it finds that these template files already exist.
To correct the problem:
1.
Rename the current files: tacl> rename newres, oldres tacl> rename newnres, oldnres
2.
Run the INSTALL^TEMPLATES macro again and use the file names newres and newnres.
This message indicates that the templates are successfully installed:
>$SYSTEM.SYS00.FUP /NAME $Z03M,INV :INBUF.1,OUTV :OUTBUF.1/
File Utility Program - T6553D45 - (14APR2000) System \TINY
Copyright Tandem Computers Incorporated 1981, 1983, 1985-2000
If Running INSTALL^TEMPLATES Is Required 115
SECURE
($SYSTEM.TEMPLATE.NEWRES,$SYSTEM.TEMPLATE.NEWNRES),"NUNU"
*** INSTALL^TEMPLATES: Completed sucessfully ***
If a System Load Is Required
If OSM is installed and configured, perform these procedures to load your RVU.
NOTE:
If the OSS subsystem is being managed by DSM/SCM, upon a successful ZPHIRNM and before halting any processors, make sure to stop all the OSS fileset mount points. This can be done by using the OSS EasySetup (T0585) STOPOSS utility as described in the Open System
Services Installation Guide or by manually stopping each mount point as described in the Open
System Services Management and Operations Guide.
Halt the Processors
Halt the processors as described in
“Halt the Processors” (page 96)
.
NOTE:
Whenever a processor is halted, the information provided under “Performing an Upgrade on a Node” in the NonStop X Cluster Solution Manual should also be considered. Contact your
HPE authorized service provider for assistance.
Perform a System Load
A normal system load consists of loading the operating system from disk into the memory of processor 0 or processor 1.
See
, for information on performing a system load.
Start Your Applications and Processes
After you have reloaded the processors and the Processor Status screen shows all processors executing the NonStop operating system, you can start your applications and processes. See
Chapter 9: “Resuming Normal Operations” .
Verify that your system is up and running. See
.
116 Using DSM/SCM With NonStop Software Essentials to Install SPRs
B Copying the DSM/SCM, NonStop Software Essentials, and SWID SPRs from the SUT Using NonStop Software
Essentials
To better ensure that the SUT installation proceeds smoothly, copy the DSM/SCM, NonStop
Software Essentials server, and SWID SPRs located on the SUT to your system and perform the other tasks described in
“Copying the DSM/SCM, NonStop Software Essentials, and SWID
SPRs from the SUT Using NonStop Software Essentials” (page 117)
before you receive and install the RVU SUT. In the upgrade checklist table in
“Upgrade Checklist” (page 34) , the step
that describes and links to the procedures described in this appendix is located at the start of the table.
NOTE:
•
In order for the procedure described in
“Copy (RESTORE) the DSM/SCM Server, NonStop
Software Essentials Server, and SWID Product Versions from the SUT to Disk” (page 118)
to display the NonStop Software Essentials client screens needed to perform it, you must already have a functioning and interoperating set of NonStop Software Essentials server and client products installed on your system.
•
You must copy the SWID SPR when you copy the DSM/SCM SPR from the SUT to your system. (DSM/SCM uses SWID to determine if a file has changed by a method called fingerprinting. If an existing file has a different fingerprint from a newer file, the newer version replaces the existing file.)
•
After you copy the DSM/SCM, NonStop Software Essentials server, and SWID SPRs from the SUT as described in
“Copy (RESTORE) the DSM/SCM Server, NonStop Software
Essentials Server, and SWID Product Versions from the SUT to Disk” (page 118) , install the
latest version of the NonStop Software Essentials client product as described in
Latest Version of the NonStop Software Essentials Client” (page 123) .
Alerts
•
For instructions on identifying the product versions of the NonStop Software Essentials,
DSM/SCM, and SWID SPRs currently installed on the host system and on the SUT, see
“Install the Latest Version of the NonStop Software Essentials Client”
•
Save the current configuration database. (See
“Save the Current Configuration File
).
•
Keep the startup TACL windows, the EMS Event Stream window, and the NonStop Software
Essentials client interface windows open whenever you are performing a Build/Apply.
Otherwise, Build/Apply errors will result.
•
If you receive an error indication in the Request Status window, you might have to re-initiate the Build/Apply. See the DSM/SCM Messages Manual to learn the nature and severity of the problem.
Determine the Product Version of NonStop Software Essentials Server on the Host
System and the SUT
To determine which product version of NonStop Software Essentials server (T0894H02) is currently installed on your system, use the VPROC command.
At a TACL prompt, enter:
> VPROC $SYSTEM.ZNSE.HTTPSRVR
Alerts 117
If the NonStop Software Essentials server product version on the SUT does not match the version on your system, copy the SUT version to your system as described in
DSM/SCM Server, NonStop Software Essentials Server, and SWID Product Versions from the
For information about installing the NonStop Software Essentials client product, see
Latest Version of the NonStop Software Essentials Client” (page 123) .
Determine the Product Version of DSM/SCM on the Host System and the SUT
To determine which product version of DSM/SCM is currently installed on your system, use the
VPROC command to determine the product version of TAEXE (Apply product of DSM/SCM) and
CBEXE (Build product of DSM/SCM).
At a TACL prompt, enter:
> VOLUME $dsmscm_vol.ZDSMSCM
> VPROC TAEXE
> VPROC CBEXE
For comparison, the DSM/SCM version on the SUT is T6031H03^ADJ.
If the DSM/SCM server product version on the SUT does not match the version on your system, copy the SUT version to your system as described in
“Copy (RESTORE) the DSM/SCM Server,
NonStop Software Essentials Server, and SWID Product Versions from the SUT to Disk” (page 118) .
Determine the SWID Product Version on the Host System and the SUT
Using the VPROC command, identify the SWID SPR on the system:
> VPROC $SYSTEM.SYSTEM.SWID
For comparison, the SWID version on the SUT is T9298H03^AAZ.
If the SWID product version on the SUT does not match the version on your system, copy the
SUT version to your system as described in
“Copy (RESTORE) the DSM/SCM Server, NonStop
Software Essentials Server, and SWID Product Versions from the SUT to Disk” (page 118) .
Copy (RESTORE) the DSM/SCM Server, NonStop Software Essentials
Server, and SWID Product Versions from the SUT to Disk
CAUTION:
You can use Scout to install the latest orderable NonStop Software Essentials server SPR on your system when you copy the DSM/SCM and SWID SPRs from the SUT.
To copy the H03 product version of the DSM/SCM server, NonStop Software Essentials server, and the SWID product from this RVU to disk, enter this RESTORE command from the TACL prompt:
>RESTORE/OUT $S.#nnn.DSM/$tape, ($*.X6031???.*, *.X9298???.*, *.X0894???.*),
VOL $DSMSCM, LISTALL where nnn identifies the RVU number (L17) and where $tape is the name of the tape drive containing this RVU.
The Windows (T0895) version of the NonStop Software Essentials 5.0 client is an independent product and can be installed as an SPR from Scout as described in
“Install the Latest Version of the NonStop Software Essentials Client” (page 123) .
Receive the H03 DSM/SCM, NonStop Software Essentials Server, and
SWID SPRs
1.
Start TMF and DSM/SCM if they are not already running:
> TMFCOM START TMF
> VOLUME $dsmscm_vol.ZDSMSCM
118 Copying the DSM/SCM, NonStop Software Essentials, and SWID SPRs from the SUT Using NonStop Software Essentials
> RUN STARTSCM
2.
Log on to NonStop Software Essentials.
3.
Click the Receive Software link on the Home Page.
4.
Under Software name, select Provided by planner and enter a name for the H03 DSM/SCM,
NonStop Software Essentials server, and SWID SPRs, up to 17 characters.
5.
Click the Input Source tab. NonStop Software Essentials displays the input source options.
6.
Select Subvolume.
7.
In the Find/Add text box, enter each distribution subvolume for the DSM/SCM files that you specified earlier in the RESTORE command:
$DSMSCM.X6031H03
$DSMSCM.X0894H03
$DSMSCM.X9298H03
Your Receive Software Input Source tab should look similar to this:
8.
Click Submit to run the request immediately. (The request takes less than 5 minutes to finish.) The Request Status page is loaded into the work panel.
Receive the H03 DSM/SCM, NonStop Software Essentials Server, and SWID SPRs 119
Create a New Software Revision for DSM/SCM, NonStop Software
Essentials, and SWID SPRs
1.
From the New Revision page, select the target system being upgraded and the current revision of the system being upgraded. Select Create.
The New Revision page is loaded with the last software revision that was built for the target in the new revision table.
2.
Load the software you just received into the source table and click Copy to.
The set of products you received into the archive: T6031, T0894, and T9298 are copied from the Archive to the new software revision.
3.
From the New software revision table, verify that the new version of DSM/SCM, NonStop
Software Essentials server, and SWID are included in the new software revision.
4.
Edit the revision notes for the new revision:
a.
Click Edit revision notes.
The Revision Notes box appears, containing the existing revision notes.
b.
Enter a description of the changes to the existing text. For example:
Rev SYSnn Date Description
--- ----- -------- ------------------------------------------------------------
23 SYS00 12/24/15 Installing DSMSCM, NSE & SWID from L15.08 SUT
c.
Click OK to save your changes to the revision notes
Build and Apply the DSM/SCM Software Product Revision
The Build process creates an activation package of products contained in the new software revision. During the Apply process, the activation package is transferred to the target system.
When the Apply request is finished, a snapshot is automatically created and returned to the host.
Because you are building and applying only a few files, it should take less than an hour before the Build and Apply requests are finished and the snapshot is returned. When the Build and
Apply requests are finished, the new software is placed in the current SYSnn subvolume.
120 Copying the DSM/SCM, NonStop Software Essentials, and SWID SPRs from the SUT Using NonStop Software Essentials
1.
On the New software revision page, click Build/Apply to display the options for your
Build/Apply.
2.
Edit the instructions to the operator to provide the current SYSnn number and then click
OK.
3.
Complete the Build/Apply Request dialog box:
a.
Under Perform steps, select Build, apply & process snapshot.
b.
Enter the current SYSnn in the Activation SYSnn text box (because no SYSGEN is required).
c.
Click Submit Build/Apply.
NonStop Software Essentials runs the Build/Apply requests and displays a Status page that is automatically updated at regular intervals.
4.
Check EMS messages for more detailed information about your Build/Apply:
•
Click the system name in the System column to view EMS events on that system for this Build/Apply request.
•
When finished viewing, close the window.
5.
To confirm that the Build and Apply tasks finish successfully and that a snapshot has been received into the host database, wait until the status shows that all activities are complete.
Run the INSTALL^TEMPLATES Macro
EMS templates are used by EMS to format EMS messages. After running ZPHIRNM, and if notified to do so by DSM/SCM, run the INSTALL^TEMPLATES TACL macro to install DSM/SCM
EMS templates in their proper locations.
NOTE:
The files newres and newnres are used as temporary names until the next
SYSGEN and system load.
The INSTALL^TEMPLATES macro is located in the $SYSTEM.ZINSAIDS subvolume and invokes the TEMPLI program to build new resident and nonresident EMS template files.
You must be logged on as a super group ID (255, 255) user to run the INSTALL^TEMPLATES macro. This macro cannot be run with the NOWAIT option and takes approximately ten minutes to finish.
For detailed instructions on running the INSTALL^TEMPLATES macro, including troubleshooting information, see
“If Running INSTALL^TEMPLATES Is Required” (page 114)
.
Restart Applications
You can now start your application software. Follow your site’s procedures for executing the system configuration and startup (command) files to accomplish this.
DSM/SCM is often started automatically by a system command file. If not, you must start
DSM/SCM manually as described in
.
Before Starting DSM/SCM
1.
Check that TMF is started:
> TMFCOM STATUS TMF
2.
Verify your current configuration against the cold start configuration in PATHHOST (or
PATHTARG) to check that the current configuration reflects all changes you have made.
3.
If necessary, run the PATHCOLD macro from the ZDSMSCM subvolume to cold start the
Pathway configuration.
Start DSM/SCM
1.
Run the STARTSCM macro from the ZDSM/SCM subvolume to start DSM/SCM:
Run the INSTALL^TEMPLATES Macro 121
> RUN STARTSCM
These messages appear:
8> startscm
DSM/SCM Process Startup Function - STARTSCM - T6031H03 - (30SEP2015)
Copyright Hewlett Packard Development Company L.P. 1993-2003
\
NOTICE: * If the hometerm chosen during INITENV, $zhome,
* is not currently in a paused state, STARTSCM and DSM/SCM
* server processes may hang indefinitely.
Be sure this
* hometerm has been paused!
Because the $ZHOME reliable home-term process is not an interactive process and is always paused, no action is necessary.
PATHWAY CONTROL FILE DATED: 21 OCT 2015, 03:52:16
$Z03N: WARN - *1079* SERVER CB-001, NO STATIC SERVERS DEFINED
$Z03N: WARN - *1079* SERVER IS-001, NO STATIC SERVERS DEFINED
$Z03N: WARN - *1079* SERVER MISERVER, NO STATIC SERVERS DEFINED
$Z03N: SERVER OM-001, STARTED
$Z03N: SERVER SM-001, STARTED
$Z03N: SERVER SM-T-001, STARTED
$Z03N: SERVER T9154-HELP-SERV, STARTED
$Z03N: WARN - *1079* SERVER TA-001, NO STATIC SERVERS DEFINED
$Z03N: WARN - *1079* SERVER TISERVER, NO STATIC SERVERS DEFINED
$Z03N: WARN - *1079* SERVER US-001, NO STATIC SERVERS DEFINED
$Z03N: TCP TCP-H, STARTED
$Z03N: TCP TCP-T, STARTED
2.
If necessary, press Enter to have STARTSCM run the IPM6031 TACL macro automatically.
This step takes several minutes.
STARTSCM: A new version of DSM/SCM has been installed and IPM6031 needs to be run.
You can either exit and run IPM6031 manually or continue and IPM6031 will be run for you.
Enter <CR> to continue or BREAK to exit.
Verify that the SUT Version of DSM/SCM is Installed
To verify that the SUT version of DSM/SCM is now installed on your system, use the VPROC command to display the product version of TAEXE (Apply product of DSM/SCM) and CBEXE
(Build product of DSM/SCM). At a TACL prompt, enter:
> volume $dsmscm_vol.ZDSMSCM
> VPROC TAEXE
> VPROC CBEXE
The VPROC command should now display the DSM/SCM SPR version T6031H03^ADJ.
Verify that the SUT or Scout Version of the NonStop Software Essentials Server is
Installed
To verify that the latest version of NonStop Software Essentials server SPR is now installed on your system, either copied from the SUT or installed from Scout, enter at the TACL prompt:
> VPROC $SYSTEM.ZNSE.HTTPSRVR
The VPROC command should now display the NonStop Software Essentials server version
T0894H02AAK.
122 Copying the DSM/SCM, NonStop Software Essentials, and SWID SPRs from the SUT Using NonStop Software Essentials
Verify that the SUT Version of SWID Is Installed
To verify that the SUT version of SWID is now installed on your system, enter at the TACL prompt:
> vproc swid
The VRPROC command should now display the SWID version T9298H03^AAZ.
Recovery for Problems While Installing DSM/SCM
If you encounter problems while installing DSM/SCM, see the NonStop Software Essentials
User’s Guide.
If DSM/SCM Errors Are Displayed
See the DSM/SCM EMS Messages manual for information regarding the error message and the
NonStop Software Essentials User’s Guide for information on deleting a Build/Apply request.
Cancel Pending Build/Apply Requests
NOTE:
If you inadvertently close the Status page, reopen it by clicking on the Request Name link on the Request List page.
Click the Cancel button.
The cancel request action takes a few minutes to finish.
NOTE:
If you cannot cancel the Build request through NonStop Software Essentials, use the
DSM/SCM Target Interface (ZPHITI) to cancel the request. See the NonStop Software Essentials
User’s Guide and online help for more information.
NOTE:
If you inadvertently close the Status dialog box, reopen it by selecting Requests>
Requests list and then double-clicking the request.
Install the Latest Version of the NonStop Software Essentials Client
After you have successfully copied the latest versions of DSM/SCM, NonStop Software Essentials server, and SWID SPRs from the SUT to the system, install the latest version of the NonStop
Software Essentials client.
The NonStop Software Essentials client product that you install on a Windows console (T0895) is not provided on the SUT. Instead, use Scout to install the appropriate client’s SPR or use the client installer that is provided on the Insight Control for NonStop DVD for the RVU, which ships with the NonStop System Console Installer DVD, Update 17 or later.
For detailed pre-installation requirements and installation information for the Windows version of the NonStop Software Essentials client, see the NonStop Software Essentials Installation and
Quick Start Guide.
The NonStop Software Essentials client SPR for Windows is T0895V02^AAK.
Recovery for Problems While Installing DSM/SCM 123
C Recovering from a System Load Failure
How to Recover from a System Load Failure
If a system load is not successful:
1.
Check for messages in the System Load dialog box of the OSM System Startup Tool.
2.
Check the Processor Status dialog box of the OSM System Startup Tool for halt codes. Look up the halt codes in the Processor Halt Codes Manual for further information about the cause of failure and the appropriate recovery procedure.
3.
After you load a system from a saved version (CONFxxyy) of the system configuration database file, verify that no pending changes to system attributes appear. From a TACL prompt:
INFO SUBSYS $ZZKRN
124 Recovering from a System Load Failure
Index
Symbols
Fallback
H
A
accessing
AUDITTRAILBUFFER parameter, DP2 and RDF
B
Back up files
backing up important files before upgrading, 37
Build/Apply request, defined, 18
C
CBEXE product
determining version on host system, 118
CIIN file
CLIM
Commands
CONFIG file
use in loading the previous RVU, 104
contacting Hewlett Packard Enterprise, 108
I
INSTALL^TEMPLATES
iTP WebServer considerations, 84
K
L
Loading the system, 73, 104, 116
M
MR-Win6530 terminal emulator
N
New software revision, NonStop Software Essentials, 50
NonStop SOAP 4 considerations, 84
NonStop Software Essentials, changing profile
NonStop System Console Installer DVD, 14
NonStop TCP/IP configuration files, 81
O
OSS
D
Determine if HSS firmware needs to be upgraded, 68
Determine if system firmware needs to be upgraded, 67
documentation
DSM/SCM
E
F
P
Post-installation
updating TCP/IP configuration files, 81
Processor type in CONFTEXT file, 19
Profiles, target, changing information, 51
PROTOCOL file
configuration file location, 82
125
R
S
SCF
testing the tape subsystem, 43
verifying critical processes, 42 verifying disks, 42
SMF, interoperability and configuration, 84
SOAP 4 see NonStop SOAP 4
Spooler
checking status of collector, 44
COLLECT, STATUS command, 54, 85, 113
Status dialog box, reopening, 123
support
Hewlett Packard Enterprise, 108
Swap file space, checking , 44
SYSnn subvolumes
System configuration file, 111
T
Target systems, changing profile information, 51
U
Update the system firmware, 70
updates
V
W
Z
ZPHIRNM
126 Index
reactivating previous software revision, 90
ZPHITI
Scarica
Link pubblico aggiornato
Il link pubblico alla tua chat è stato aggiornato. Puoi visualizzare e gestire i tuoi link di chat dopo la registrazione qui.
Link pubblico del documento
Pubblicità