HP H06.25 RVU NonStop server software Software Installation and Upgrade Guide
Anuncio
Anuncio
H06.25 Software Installation and Upgrade
Guide
HP Part Number: 696595-002
Published: March 2014
Edition: H06.25 RVU
© Copyright 2012, 2014
Legal Notice
Confidential computer software. Valid license from HP 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 HP 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. HP 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.
Microsoft®, Windows®, Windows NT®, Windows® XP, and Windows Vista® are U.S. registered trademarks of Microsoft Corporation.
Intel, Pentium, and Celeron are trademarks or registered trademarks of Intel Corporation or its subsidiaries in the United States and other countries.
Java is a registered trademark of Oracle and/or its affiliates.
Motif, OSF/1, UNIX, X/Open, and the "X" device are registered trademarks, and IT DialTone and The Open Group are trademarks of The Open
Group in the U.S. and other countries.
Open Software Foundation, OSF, the OSF logo, OSF/1, OSF/Motif, and Motif are trademarks of the Open Software Foundation, Inc.
OSF MAKES NO WARRANTY OF ANY KIND WITH REGARD TO THE OSF MATERIAL PROVIDED HEREIN, INCLUDING, BUT NOT LIMITED TO,
THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE.
OSF shall not be liable for errors contained herein or for incidental consequential damages in connection with the furnishing, performance, or use of this material.
© 1990, 1991, 1992, 1993 Open Software Foundation, Inc. The OSF documentation and the OSF software to which it relates are derived in part from materials supplied by the following:
© 1987, 1988, 1989 Carnegie-Mellon University. © 1989, 1990, 1991 Digital Equipment Corporation. © 1985, 1988, 1989, 1990 Encore
Computer Corporation. © 1988 Free Software Foundation, Inc. © 1987, 1988, 1989, 1990, 1991 Hewlett-Packard Company. © 1985, 1987,
1988, 1989, 1990, 1991, 1992 International Business Machines Corporation. © 1988, 1989 Massachusetts Institute of Technology. © 1988,
1989, 1990 Mentat Inc. © 1988 Microsoft Corporation. © 1987, 1988, 1989, 1990, 1991, 1992 SecureWare, Inc. © 1990, 1991 Siemens
Nixdorf Informationssysteme AG. © 1986, 1989, 1996, 1997 Sun Microsystems, Inc. © 1989, 1990, 1991 Transarc Corporation.
OSF software and documentation are based in part on the Fourth Berkeley Software Distribution under license from The Regents of the University of California. OSF acknowledges the following individuals and institutions for their role in its development: Kenneth C.R.C. Arnold, Gregory S.
Couch, Conrad C. Huang, Ed James, Symmetric Computer Systems, Robert Elz. © 1980, 1981, 1982, 1983, 1985, 1986, 1987, 1988, 1989
Regents of the University of California.
Contents
About This Document...................................................................................11
1 Overview of Installing This RVU..................................................................14
Upgrading the DSM/SCM Client (Planner Interface) and the T6031H02 PVU............................20
2 Managing Microcode and Firmware...........................................................24
Updating Blade Complex Firmware for NS14000 Series Systems and NS16000 Series
3 Installation, Migration, and Fallback Considerations......................................36
Contents 3
4
OSM Uses Different User ID and Password to Monitor Maintenance Switch and UPS..................50
Contents
4 Upgrade Checklist....................................................................................62
5 Preparing to Upgrade the Server Software (Pre-installation Tasks)....................66
Contents 5
6 Installing a SUT for This RVU Using NonStop Software Essentials....................78
7 Installing a SUT for This RVU Using the Planner Interface................................91
8 Running ZPHIRNM ................................................................................106
9 Loading This RVU...................................................................................111
6 Contents
10 Resuming Normal Operations................................................................120
11 Backing Out the DSM/SCM Revision.......................................................127
12 Running ZPHIRNM to Reactivate the Previous Software Revision..................134
13 Falling Back on the CLIMs, Loading the Previous RVU, and Resuming Normal
Operations ..............................................................................................137
In the Halted State, Power Cycle Processors If Falling Back to the H06.03 or H06.04 RVU.........142
Contents 7
A Using DSM/SCM With NonStop Software Essentials to Install SPRs..............153
B Using DSM/SCM With the Planner Interface to Install SPRs..........................160
C Copying the DSM/SCM, NonStop Software Essentials, and SWID SPRs from the SUT Using NonStop Software Essentials..................................................167
Determine the Product Version of NonStop Software Essentials Server on the Host System and the
Determine the Product Version of DSM/SCM on the Host System and the SUT..........................168
Copy (RESTORE) the DSM/SCM Server, NonStop Software Essentials Server, and SWID Product
8 Contents
Receive the H02 DSM/SCM, NonStop Software Essentials Server, and SWID SPRs.......................169
Create a New Software Revision for H02 DSM/SCM, NonStop Software Essentials, and SWID
If Migrating to H06.23 or an Earlier RVU, Install the Latest NonStop Software Essentials Server SPR
D Installing SWID, DSM/SCM PVU, and Planner Interface (PI) Client Software ..176
E Updating SWAN Concentrator CLIP Firmware............................................190
Glossary..................................................................................................191
Index.......................................................................................................194
Contents 9
Tables
Updateable Firmware for the NonStop 6780 ServerNet Switch File Names and Locations..........34
Interoperability SPR for G-Series RVU Systems Connected Through Expand to H-Series or J-Series
Interoperability SPRs for H-Series RVU Systems Connected through ServerNet to J-Series RVU
Interoperability SPRs for G-Series RVU Systems Connected through ServerNet to H-Series RVU
H-Series and J-Series RVUs Needing SPRs for Nodes to Interoperate with SQL/MX 2.3.4 (H06.20
About This Document
This guide describes:
•
Upgrading the initial H06.25 RVU through the use of site update tape (SUT) refreshes, if necessary.
•
Falling back to a previous SYSnn.
This document might be revised between DVD publications. A more recent version of this document might exist online in the HP Integrity NonStop H-Series collection of the NonStop Technical Library
(NTL) at http://www.hp.com/go/nonstop-docs .
Supported Release Version Updates (RVUs)
This manual supports the H06.25.00 RVU and all subsequent H06.25.nn RVUs unless otherwise noted. Exceptions are described in the text.
Intended Audience
This guide is written for operations staff and support personnel and requires a working knowledge of Windows Server 2003 or Windows Server 2008, HP NonStop Open System Management
(OSM) Interface client and server components, system console software, NonStop Software
Essentials, DSM/SCM, Target Interface (ZPHITI), and ZPHIRNM applications.
New and Changed Information in This Edition
This edition includes these changes:
•
Multiple Providers per CLIM functionality and the requirments for CLIM fallback from H06.26
an earlier RVU are described in
“A. If Falling Back to H06.25 or an Earlier RVU” (page 113) .
•
When upgrading to H06.25 from an earlier RVU, the processors must be hard reset after halting and before loading the system as described in
“4. (Required for Upgrade to H06.25
or Later) Perform a Hard Reset” (page 114)
.
•
SQL/MX Release 3.2 replaces SQL/MX Release 2.3.4 as the default SQL/MX version on the
SUT. Users have a choice of staying with the latest SQL/MX Release 2.3.4 SPRs released as
Time Critical Fixes (TCFs) or to migrate to SQL/MX Release 3.2.
What This Guide Does Not Cover
•
System migration planning. System-wide issues involved in migrating to the latest H-series RVU are covered in:
◦
The H06.nn Release Version Update Compendium for this RVU
◦
The Interactive Upgrade Guide 2 (IUG2)
•
Installing an H-series RVU from a host system to a remote or separate network-connected target system using an Expand connection or tape. See the NonStop Software Essentials User's
Guide or DSM/SCM User's Guide for information on setting up target systems.
•
Installing and configuring HP NonStop Open System Services (OSS).
•
Upgrading HP NonStop system console workstation applications. To install or upgrade HP 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.
Supported Release Version Updates (RVUs) 11
•
Installing, configuring, and migrating to HP NonStop SQL/MX. The SQL/MX Installation and
Management Guide or SQL/MX Installation and Upgrade Guide provides step-by-step instructions for installing and configuring NonStop SQL/MX. For information about migrating a NonStop SQL/MX or NonStop SQL/MP database to an H-series RVU and falling the database back to G-series RVUs, see the SQL Supplement for H-Series RVUs.
•
Installing and configuring the I/O adapter module (IOAM) enclosure and its components. See your HP service provider.
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 May 2006, 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.
12
Related Information
NOTE: Many of the documents referenced in this guide are located in the NTL, which has moved to the HP Business Support Center (BSC) at http://www.hp.com/go/nonstop-docs .
•
The NonStop Firmware Matrices provides firmware levels for all RVUs.
•
The CLuster I/O Module (CLIM) Software Compatibility Reference provides the compatibility information of CLIM DVD SPRs and host RVUs.
•
The NonStop System Console Installer Guide contains information on installing OSM client-based components, and other applications on system consoles.
•
The H06.nn Release Version Update Compendium, for this RVU, provides summary information on the new features and migration issues.
•
The comForte MR-Win6530 online help provides information on configuring MR-Win6530 sessions.
•
The Support and Service library and guided procedures provide online help and web-based instructions for performing many tasks, including updating firmware. The Support and Service library is available to HP qualified support providers in the NonStop Technical Library (NTL).
•
The NonStop Software Essentials User's Guide provides detailed, task-oriented instructions for using NonStop Software Essentials. The DSM/SCM User's Guide provides detailed, task-oriented instructions for using DSM/SCM to perform the various tasks associated with installing and managing software revisions.
•
Planning guides for specific hardware platforms provide information for planning and configuring the installation of NonStop servers.
•
The NonStop Operations Guide provides information for operating and maintaining the
NonStop server.
•
Hardware installation manuals for specific hardware platforms provide detailed instructions and checklists for installing a NonStop NS-series server.
•
The OSM Configuration Guide provides an overview of all OSM applications and components, how to configure and get started with OSM, and how to use the primary OSM interface, OSM
Service Connection, to monitor and perform actions on system and cluster resources.
Publishing History
Part Number
696595–001
Product Version
N.A.
Publication Date
August 2012
HP Encourages Your Comments
HP 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.
Related Information 13
1 Overview of Installing This RVU
This chapter introduces the basic installation tools and requirements in H-series RVUs, including installing the site update tape (SUT) for a release version update revision (RVUR). For information on installing software product revisions (SPRs), see
“Using DSM/SCM With NonStop Software
Essentials to Install SPRs” (page 153)
.
SUT on DVD
Beginning with the H06.06 RVU, the H-series SUT is also available on DVD. For instructions on installing the SUT from the DVD, see the IPSetup User’s Guide.
WARNING!
See the Support Note S12017 (IPSetup on 64-bit Windows causes IE problems) for information about preventing or recovering from an Internet Explorer issue that can occur on 64-bit
Windows operating systems.
CLuster I/O Module (CLIM) Software for This RVU
If your system has CLIMs installed, the CLIM software is available on a separate DVD and in Scout for NonStop Servers. Before installing CLIM software on your NonStop system, you must install the software on the system console as described in the Cluster I/O Protocols (CIP) Configuration and
Management Manual.
For the CLIM minimum DVD version supported for this RVU, see the Cluster I/O Module (CLIM)
Software Compatibility Reference.
Disk Space Requirements
The maximum SUT with all standard products, plus all optional products available to customers, contains approximately 8 gigabytes (GB) of data, depending on optional products (however, actual disk space required can be much greater).
Operational Differences Between This RVU and G-series RVUs
For planning purposes, the following list represents known differences.
•
TAPEBOOT is not supported in H-series RVUs. You cannot make a system image tape (SIT).
•
In H-series RVUs, native compilers and linkers have new names. Therefore, automated scripts require changing.
•
The subvolume for public libraries is SYSnn in G-series RVUs. In H-series RVUs, it is ZDLLnnn.
•
In H-series RVUs, the DSM/SCM installation default is to manage OSS files. In G-series RVUs, the default is not to manage OSS files. For OSS users who request HP manufacturing to install
OSS, it is assumed that OSS is installed and configured.
•
In this RVU, tape storage on the NonStop system supplies both Fibre Channel (FC) and SCSI tape products. Tape devices can be directly connected to a NonStop server in two ways:
◦
Fibre channel tape devices can be connected directly to a Fibre Channel ServerNet adapter (FCSA) in an IOAM enclosure.
◦
SCSI tape devices can be connected through a T1200 FC to SCSI Converter device
(model M8201R), which enables connection to SCSI tape drives. For installation instructions, see the M8201R Fibre Channel to SCSI Router Installation and User’s Guide.
For a complete list of supported tape drives, including using SCSI to connect to G-series legacy tape drives, see the NonStop Storage Overview Manual.
14 Overview of Installing This RVU
•
Compaq TSM is not supported in H-series RVUs. You must use OSM. For many update tasks, you can still use the Subsystem Control Facility (SCF).
•
The OSM Low-Level Link features a new System Load Configuration dialog specifically for
H-series RVUs. This system load and save configuration feature is much more flexible than previous save configuration options used in G-series RVUs. The mirrored system disk can have an independent adapter and location from the primary system disk. $SYSTEM is the default in the OSM Low-Level Link Configuration shortcut menu.
•
On H-series systems, the OSM Low-Level Link physical view displays a tabular or inventory view of all the components with their attributes in the system.
•
Because of the different system architectures, updating processor boot code and other server firmware files on systems running H-series RVUs is completely different from updating server firmware on NonStop S-series servers.
•
The RCVDUMP program has two new parameters for H-series systems only, SLICE sliceId and PARALLEL. See the TACL Reference Manual for more information on syntax and usage.
Beginning with the H06.14 RVU, RCVDUMP creates a smaller extracted dump file in addition to the full dump. This smaller dump requires less time to transfer to analysts. You can suppress generation of the extracted dump file by specifying a new option, FULL. A new utility,
POSTDUMP, is provided to explicitly extract a smaller dump file from the full dump produced by RCVDUMP.
•
For H-series systems only, RELOAD has a new option OMITSLICE. The RELOAD command also includes the option NOPRIME, which is supported on both G-series and H-series systems.
For error messages for RCVDUMP and RELOAD see the TACL Reference Manual.
Investigating an H-Series RVU
Several resources are available to help you plan for installing an H-series RVU:
•
Product softdocs and the README file.
•
The H06.nn Release Version Update Compendium provides highlights of the new products and features for each H-series RVU. It also discusses the effect these changes have on installation and configuration, operations and management, data files, applications, and networks.
•
The Interactive Upgrade Guide 2 (IUG2) simplifies the planning of software and hardware for RVUs. Select the products and RVUs you use, and the IUG2 automatically displays customized migration instructions and lists new features.
•
The Managing Software Changes manual explains TRM2000 and provides information about system migration and installation processes, software product revision (SPR) analysis, and resources for investigating new RVUs and SPRs.
•
Scout for NonStop Servers is a Web-based SPR analysis and delivery tool. You can access
Scout through HP NonStop eServices Portal at: https://onepoint.nonstop.compaq.com/buildpage.asp.
Online help for using Scout is available on the Scout Web site.
•
Scout or the DSM/SCM Planner Interface can be used to view and print softdocs and release documents on the SUT after you receive new software into the archive. For more information, see the DSM/SCM User's Guide or Scout.
System Console Requirements
A NonStop system console (NSC) is an HP-supplied personal computer used to run OSM and other maintenance and diagnostic software for NonStop servers. For information on currently supported
Investigating an H-Series RVU 15
hardware models and other requirements, as well as the latest available console software, see the
NonStop System Console Installer Guide.
Updating NonStop System Console Software (Optional)
HP recommends that you always update console software to the versions available on the latest
NonStop System Console Installer DVD.
NOTE: HP recommends that you use the latest version of the NonStop System Console Installer
DVD.
If you have ordered new system consoles for this RVU, the latest client-side software is already installed; if not, obtain the latest NSC installer DVD. See the NonStop System Console Installer
Guide for detailed information on installing the system console software.
NOTE: HP recommends installing HP SIM and NonStop Software Essentials on a separate Central
Management Server (CMS), but does not support installing these products on the NonStop system console as well. Either way, the products can be installed or upgraded from the HP Insight Control for NonStop DVD that ships with the NSC DVD. For more information on whether installing HP
SIM and NonStop Software Essentials on the system console or a separate CMS makes sense for your LAN environment, see HP SIM for NonStop Manageability, located in the NonStop Technical
Library (NTL).
This guide shows you how to upgrade the RVU using either NonStop Software Essentials, an HP
SIM plug-in, or the DSM/SCM Planner Interface. If you are using a 64-bit console, you must use
NonStop Software Essentials, as the DSM/SCM client and its Planner Interface are not supported on 64-bit consoles. If not already installed, both HP SIM and the NonStop Software Essentials client component can be installed from the HP Insight Control for NonStop DVD that ships with the
NonStop System Console Installer DVD (Update 17 or later), as described in the NonStop System
Console Installer Guide. You can also download HP SIM and the NonStop Software Essentials client component from Scout, and download HP SIM from www.hp.com/go/hpsim . For more information on installing and getting started with NonStop Software Essentials, see the NonStop
Software Essentials Installation and Quick Start Guide.
Software Product
NonStop Software Essentials Server
NonStop Software Essentials Client (Windows)
NonStop Software Essentials Client (Linux)
T-Number
T0894
T0895
T0996
NOTE: NonStop Software Essentials requires that a server component be installed on the NonStop system. That server component is on the SUT for H06.24 and later, but can also be downloaded from Scout. If the NonStop Software Essentials server component is not installed on your NonStop systems, see the Nonstop Software Essentials Installation and Quick Start Guide.
Using OSM With This RVU
OSM Server-Based Component
For H06.25, the server-based OSM SPR is T0682H02^ACZ. This version may be superceded by a later version. For highlights of this server version, see the H06.25 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.
16 Overview of Installing This RVU
NOTE: HP recommends that you use the latest OSM SPRs, irrespective of what was shipped on your SUT.
OSM Client-Based Components
OSM client-based component upgrades are delivered on the NonStop System Console Installer
DVD. HP recommends upgrading to the very latest console software available.
The OSM client-based products are:
•
OSM Low-Level Link (T0633), which is used to provide down-system support, allowing you to communicate with your HP NonStop server even when the NonStop Kernel operating system is not running. In addition to performing the system loads, it is used to configure VIO and other modules, configure CLIMs, and update CLIM software. For information about other features and functionality, see the OSM Configuration Guide and the OSM Low-Level Link online help.
•
OSM Console Tools (T0634), which installs a variety of tools, including the Down System
CLIM Firmware Update Tools and the NonStop Maintenance LAN DHCP DNS Configuration
Wizard. For detailed information about these tools and other OSM Console Tools features, see the OSM Configuration Guide
Using DSM/SCM and NonStop Software Essentials With This RVU
NonStop Software Essentials and This RVU
The HP NonStop Software Essentials product is a software installation and management tool for all NonStop servers running H-series RVUs as of H06.09. NonStop Software Essentials uses
DSM/SCM server software and provides a replacement for DSM/SCM Planner Interface functions as well as certain Host Maintenance Interface functions that is more secure, easier to use, and overcomes other shortcomings of DSM/SCM. It coexists with DSM/SCM; you can use either
NonStop Software Essentials or the DSM/SCM Planner Interface for tasks in which there is overlapping functionality. For more information, see the NonStop Software Essentials Installation
and Quick Start Guide and the NonStop Software Essentials User's Guide.
These rules apply:
•
The NonStop Software Essentials server (T0894) and NonStop Software Essentials client
(T0895 or T0996) components installed on a system must both be the same NonStop System
Essentials release (1.0 or 2.0).
•
The SUTs for H06.24 RVU or and all later J-series RVUs contain the T0894H02 NonStop
Software Essentials server. The SUTs for J06.13 and earlier RVUs did not provide the NonStop
Software Essentials server, so you must install the T0894H02 SPR from Scout for these RVUs.
•
You cannot use the DSM/SCM Planner Interface on systems with 64-bit NonStop system consoles. If you have a 64-bit NonStop system console but do not have the NonStop Software
Essentials server installed on your system, you must manually install the server as described in the NonStop Software Essentials Installation and Quick Start Guide.
Using DSM/SCM and NonStop Software Essentials With This RVU 17
DSM/SCM and This RVU
CAUTION: Before you install the SUT, you must copy the SUT versions of the DSM/SCM server and client products to your host system. When you copy the server and client SPRs for DSM/SCM, you must also upgrade the SWID product (T9298H02) at the same time to avoid a fingerprint or mismatched file problem. If you are 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 167)
for the detailed instructions. If you are upgrading
DSM/SCM paired with the Planner Interface, see
“Installing SWID, DSM/SCM PVU, and Planner
Interface (PI) Client Software ” (page 176)
for the detailed instructions.
•
The DSM/SCM H02 PVU (T6031H02) is supported.
•
A new type of hardware platform and selectable option, TNS/E, has been added to the
Planner Interface (PI) screens. See the DSM/SCM User’s Guide for configuration information.
•
By default, the DSM/SCM H02 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.
•
CNFGINFO supports new system generation options for dynamic-link libraries (DLLs).
•
Creating a SIT is not supported in H-series RVUs.
•
Similar to G-series RVUs, you might have to update both the client and server-side SPRs. The
PI screen displays G02/H02 for the client side and H02 for the server side.
•
As of G06.27, you can manage an H-series target system from a G-series host system.
WARNING!
If an H06.12 or later host system is used to perform DSM/SCM builds for a target system that is running a release prior to H06.12, you must include the appropriate T0431 (OSBUILD)
SPR for that target system before performing the build. If this is not done, then the DSM/SCM
Sysgen (OSBUILD) will fail. The build will complete successfully if it is being done for an H06.12
or later system.
The appropriate T0431 SPRs are:
SPR
T0431H01^AAK
T0431H01^AAH
T0431H01^AAN
T0431H01^AAM
T0431H01^AAL
Target NonStop System RVU
H06.10 and H06.11
H06.09
H06.08
H06.07
H06.05 and H06.06
Any H06.12 or later RVU attempting to build an RVU prior to H06.12 must install these SPRs.
18 Overview of Installing This RVU
OSBUILD SPR
WARNING!
If an H06.12 or later host system is used to perform DSM/SCM builds for a target system that is running a release prior to H06.12, you must include the appropriate T0431 (OSBUILD)
SPR for that target system before performing the build. If this is not done, then the DSM/SCM
Sysgen (OSBUILD) will fail. The build will complete successfully if it is being done for an H06.12
or later system.
The appropriate T0431 SPRs are:
SPR
T0431H01^AAK
T0431H01^AAH
T0431H01^AAN
T0431H01^AAM
T0431H01^AAL
Target NonStop System RVU
H06.10 and H06.11
H06.09
H06.08
H06.07
H06.05 and H06.06
Any H06.12 or later RVU attempting to build an RVU prior to H06.12 must install these SPRs.
DSM/SCM and Public DLLs
DSM/SCM manages and controls the distribution, installation, and activation of public DLLs in
J-series RVUs the same way as H-series RVUs. Public DLLs are installed on the H-series and J-series target systems in their own subvolume.
To manage OSS files in J-series RVUs, use the T6031H02 or later SPR. The corresponding client version of NonStop Software Essentials or the Planner Interface must also be installed. For this
RVU, the SPR for the NonStop Software Essentials client is T0895V02^AAC for Windows and
T0996V92^AAC for Linux. For H06.28, the DSM/SCM SPR for the server side is T6031H02^ADD.
The SPR for the NonStop Software Essentials server is T0894H02^AAE.
DSM/SCM and OSS
When performing a Build/Apply, ensure that the setting of the Manage OSS Files option matches with 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 HP to install OSS, it is assumed that OSS is installed and configured. However, the OSS root fileset ($ZPNS) must be in the started state; otherwise, the Build/Apply fails.
Upgrading the NonStop Software Essentials Client
You can upgrade your NonStop Software Essentials client from the HP Insight Control for NonStop
DVD that ships with the NonStop System Console Installer DVD. For more information, see the
NonStop System Console Installer Guide. You can also download the NonStop Software Essentials client component from Scout, and download HP SIM from: www.hp.com/go/hpsim .
Software Product
NonStop Software Essentials Server
NonStop Software Essentials Client (Windows)
NonStop Software Essentials Client (Linux)
T-Number
T0894
T0895
T0996
Using DSM/SCM and NonStop Software Essentials With This RVU 19
For instructions on installing NonStop Software Essentials, see the NonStop Software Essentials
Installation and Quick Start Guide.
Upgrading the DSM/SCM Client (Planner Interface) and the T6031H02 PVU
To upgrade the DSM/SCM client, if you did not order new system consoles, the DSM/SCM client version must be downloaded from the ZDSMSCM subvolume. To download the client version
(Planner Interface) from ZDSMSCM to a separate PC console or your current console, follow the client or Planner Interface installation instructions in the DSM/SCM User’s Guide or
SWID, DSM/SCM PVU, and Planner Interface (PI) Client Software ” (page 176) .
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.
NOTE: The alternate name for the $SYSTEM drive has changed on H-series systems. For more information, see
“Build and Apply the New Software Revision” (page 98)
.
4.
Activate the new operating system image.
During the Build/Apply phase, DSM/SCM places product files in their target subvolumes
(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 it exists in the SYSnn being activated and if:
◦
The configuration being activated contains a ZCNFDEFS TSV
◦
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.
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 $ZCDB process
(T6586, Configuration Services).
20 Overview of Installing This RVU
Check or Alter the CONFTEXT File Before Installing This RVU SUT
The processor type (SYSTEM_PROCESSOR_TYPE) is specified in the CONFTEXT file. It is one of the values NSE-A or NSE-P.
IMPORTANT: For information about J-series systems, see the J06.nn Software Installation and
Upgrade Guide.
NOTE: Tape_Label_Processing is controlled by the Storage Subsystem ($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.
System
NS14x00 Series, NS16x00 Series
NS1000, NS1200
SYSTEM_PROCESSOR_TYPE
NSE-A
NSE-P
Check the CONFTEXT file to ensure that the following entries are present. They are needed for the implicit DLLs:
SYSTEM_IMPLICIT_INIT_FILES TANDEM^IMPLICIT^INIT^FILES;
SYSTEM_IMPLICIT_CORE_FILES TANDEM^IMPLICIT^CORE^FILES;
SYSTEM_IMPLICIT_FS_FILES TANDEM^IMPLICIT^FS^FILES;
SYSTEM_IMPLICIT_MCP_FILES TANDEM^IMPLICIT^MCP^FILES;
SYSTEM_IMPLICIT_ETC_FILES TANDEM^IMPLICIT^ETC^FILES;
This example shows a CONFTEXT file:
$TEMP CONFIGXX 2> type conftext
ALLPROCESSORS:
SYSTEM_PROCESSOR_TYPE NSE-A;
SYSTEM_VOLUME_SUBVOL $SYSTEM.SYS00;
SYSTEM_LIBRARY_CODE_FILES TANDEM^LIBRARY^CODE^FILES;
SYSTEM_PROCESS_CODE_FILES TANDEM^PROCESS^CODE^FILES;
SYSTEM_PROCESS_LIBRARY_FILES TANDEM^PROCESS^LIBRARY^FILES;
STANDARD_MICROCODE TANDEM^STANDARD^MICROCODE;
MICROCODE_FILES TANDEM^MICROCODE^FILES;
FILES_TO_COPY_TO_NEW_SYSTEM TANDEM^FILES^TO^COPY;
SYSTEM_IMPLICIT_INIT_FILES TANDEM^IMPLICIT^INIT^FILES;
SYSTEM_IMPLICIT_CORE_FILES TANDEM^IMPLICIT^CORE^FILES;
SYSTEM_IMPLICIT_FS_FILES TANDEM^IMPLICIT^FS^FILES;
SYSTEM_IMPLICIT_MCP_FILES TANDEM^IMPLICIT^MCP^FILES;
SYSTEM_IMPLICIT_ETC_FILES TANDEM^IMPLICIT^ETC^FILES;
FORMATTER_TEMPLATE_FILES TANDEM^FORMATTER^TEMPLATE^FILES;
INITIAL_COMINT_INFILE $DSMSCM.SYS.CIIN;
Installing CLIM Software
If your NonStop system has CLIMs installed, you must upgrade the CLIM software after installing the SUT. The procedure for upgrading CLIM software is described in the Cluster I/O Protocols
(CIP) Configuration and Management Manual.
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.
maps the major tasks involved with falling back.
Installing CLIM Software 21
Table 1 Tasks and Tools for Backing Out to the Previous SYSnn
To perform this task...
1. Submit a backout request.
Use this tool...
DSM/SCM Target Interface (ZPHITI)
Documented in this chapter...
OSM Service Connection online help 2. Return the firmware to previous versions. (Not recommended. If problems occur, firmware can be downgraded at this time.)
3. Load the HP NonStop operating system and, if CLIMs are installed, load the CLIM software from the previous revision into the first processor.
OSM Service Connection
DSM/SCM ZPHIRNM program
4. Load the HP NonStop operating system from the previous revision into the first processor.
OSM Low-Level Link
5. Reload the remaining processors.
TACL RELOAD command
Chapter 12: “Running ZPHIRNM to
Reactivate the Previous Software
Chapter 13: “Falling Back on the CLIMs,
Loading the Previous RVU, and Resuming
Chapter 13: “Falling Back on the CLIMs,
Loading the Previous RVU, and Resuming
IPs That Become SUT Products on H-Series
These products, which are IPs in G-series RVUs (and will remain IPs in G-series RVUs), become
SUT-based products in H-series RVUs:
Product Name iTP Secure WebServer, Strong,
Domestic, 128 Bit iTP Secure Webserver, Strong
International, 128 Bit
HP NonStop Pathway/iTS with HP
NonStop TS/MP
1
G-series Product ID
SJ98V5 becomes HSJ98
SJ95V6 becomes HSJ95
T-Numbers
T7951, T8997
T7951, T8997
SR76V1 T0524, T0526, T0527, T0528, T0529,
T0570, T0577
NonStop XML Parser
HP NonStop Simple Object Access
Protocol (SOAP)
SX20V5
SX21V3 becomes HSX21
T0535 (not moved to H-series), T0563
T0535 (not moved to H-series), T0563,
T0601, T0603, T1100 (not moving to
H-series)
T0592 (not moved to H-series), T0601 NonStop XSLT Translator SX22V2
NonStop SOAP Client SX24V1 T0624
NonStop Fast XML Parser SX25V1 T0625
1
NonStop Pathway/iTS with TS/MP (SR76V1) in G-series becomes NonStop Pathway with TS/MP in H-series. It includes iTS Gateway (T0524), SCOBOL Java Package (T0526), iTS Connection Router (T0527), iTS SCOBOL (T0528), iTS
Client Object Manager (T0529), XML4Pathway/iTS (T0570), Pathway iTS License (T0577), TS/MP License (T5812, already on the SUT), and Pathway/TS License (T5813, already on the SUT).
For IPs that are available for this RVU or for IPs that are not migrating to H-series, contact your HP representative.
Managing Microcode and Firmware
When you install a new RVU, you must ensure you are running the current firmware and microcode versions and upgrade those products, if necessary. The current firmware versions and upgrade procedures are given in
Chapter 2: “Managing Microcode and Firmware”
22 Overview of Installing This RVU
Installation, Migration, and Fallback
Many products have special considerations for installation, migration, and fallback, which you should be aware of before installing the new RVU. These considerations are described in
“Installation, Migration, and Fallback Considerations” .
Perform an RVU Upgrade on a Node
For instructions on how to perform an RVU upgrade on a node connected to any ServerNet Cluster topology (such as a BladeCluster Solution, ServerNet Cluster 6770, ServerNet Cluster 6780, or a combination of these topologies), see the procedure in the NonStop BladeCluster Solution Manual.
Installation, Migration, and Fallback 23
2 Managing Microcode and Firmware
NOTE: Trained HP service providers will update the firmware when they install any field-replaceable units (FRUs). Firmware, unless otherwise noted, is considered customer-updateable.
You must manage all microcode and firmware within your system properly, ensuring 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. Carefully review the NonStop Firmware
Matrices document, located in the Service and maintenance information section of the HP Integrity
NonStop Service Information collection of the NonStop Technical Library at http://www.hp.com/ go/nonstop-docs .
OSM displays firmware version information, including file names and a Compare State attribute, which compares the current version against the default or available version. Outside of OSM, you can use the FILEINFO command to see firmware files.
Enter this FILEINFO command, if you are connected to a NonStop S-series server, to check for location and file information for CPUCODE and SPCODE:
$SYSTEM STARTUP 7> FILEINFO $SYSTEM.SYS*.*CODE*
Enter this FILEINFO command to check for location and file information for the HSS firmware:
$SYSTEM STARTUP 7> FILEINFO $SYSTEM.SYS*.*FIRM*
Enter this FILEINFO command to check for location and file information for other firmware, such as ServerNet Clustering:
$SYSTEM STARTUP 7> FILEINFO $SYSTEM.SYS??.M????
These topics give the firmware file names and locations and describe how to upgrade SUT-based firmware. For firmware levels, see the NonStop Firmware Matrices document.
•
“Managing Maintenance Entity (ME) Firmware and ME FPGA Firmware”
•
“Managing NonStop Blade Complex Firmware”
•
“Managing ServerNet Adapter Firmware”
•
“Managing SWAN and SWAN 2 Concentrator Firmware”
•
“Managing NonStop Cluster Switch Firmware ”
•
For minimum firmware levels associated with RVUs, see the NonStop Firmware Matrices document, located in the Service and maintenance information section of the HP Integrity NonStop Service
Information collection of the NonStop Technical Library at www.hp.com/go/nonstop-docs .
Managing Maintenance Entity (ME) Firmware and ME FPGA Firmware
There is maintenance entity (ME) firmware on the ServerNet Switch Boards located in both processor switch and I/O adapter modules (IOAMs), and on VIO logic boards located in VIO enclosures.
Depending on your system type, the system has ME firmware in processor switches, I/O adapter modules (IOAMs), or VIO enclosures. The ME firmware and ME Field Programmable Gate Array
(FPGA) should be the same for all processor switches and IOAMs in your system. ME firmware and ME FPGA versions will be different for VIO modules (see
and
), but all VIO modules in your system should also have consistent versions. ME firmware and ME FPGA can be checked and, if necessary, updated from either the OSM Service connection or the Low-Level Link.
The following sections describe how to update ME firmware using the OSM Service Connection:
24 Managing Microcode and Firmware
To determine whether your system has processor switches, IOAMs, or VIO enclosures, see the planning guide for the system.
•
◦
Updating ME Firmware on a Processor Switch
◦
Updating ME Firmware on an I/O Adapter Module
◦
Updating ME Firmware on a VIO Enclosure
•
For information on using OSM Low-Level Link to perform ME firmware updates, see the online help available within the application.
If you have legacy NonStop S-series enclosures attached to your system, you will also have service processor (SP) firmware to check. For more information on updating SP firmware, see the latest
G-Series Software Installation and Upgrade Guide.
Updating ME Firmware
shows the ME firmware file name and location information for processor switches, IOAMs, and VIO enclosures. For firmware levels, see the NonStop Firmware Matrices document.
Table 2 Maintenance Entity (ME) Firmware File Names and Locations
Component
ServerNet Switch Board (IOAM and processor switch)
VIO enclosure logic board
File Name
M2805
M1805
Default Location
SYSnn subvolume
SYSnn subvolume
For more information on managing ME firmware, see:
• the planning guide or hardware installation manual for your server type.
• for Carrier Grade systems, the NonStop NS-Series Carrier Grade Server Manual.
• the online help in the OSM Low-Level Link or OSM Service Connection.
CAUTION: You should not use the Multi-Resource Actions dialog (in the OSM Service Connection) to update ME FW simultaneously on both fabrics, or a system outage might occur.
Updating ME Firmware on a Processor Switch
There is maintenance entity (ME) firmware on the ServerNet Switch Board object in each processor switch.
NOTE: In the OSM Low-Level Link, a ServerNet switch board is represented as an ME object.
To use the OSM Service Connection to update the ME firmware for processor switches:
1.
Locate the ServerNet Switch Board on one Internal ServerNet Fabric (in the tree pane, select
System > Fabric Group > Internal ServerNet Fabric > Processor Switch > ServerNet Switch
Board).
2.
Right-click the ServerNet Switch Board object and select Actions.
3.
In the Actions dialog box, select ME Firmware Update and click Perform Action.
4.
Monitor the update.
5.
Repeat for the processor switch on the other Internal ServerNet Fabric.
You can also check and update ME firmware using the OSM Low-Level Link. For more information, see the OSM Low-Level Link online help.
Managing Maintenance Entity (ME) Firmware and ME FPGA Firmware 25
Updating ME Firmware on an I/O Adapter Module (IOAM)
There is maintenance entity (ME) firmware on the ServerNet Switch Board object in each I/O
Adapter Module.
NOTE: In the OSM Low-Level Link, a ServerNet Switch Board is represented as an ME object.
To use the OSM Service Connection to update the ME firmware for IOAMs:
1.
Locate the ServerNet Switch Board on one Internal ServerNet Fabric.
•
For NS5000 and NS16000 series systems: In the tree pane, select System > Group > >
IOAM Enclosure > IOAM > ServerNet Switch Board.
•
For NS1000, NS3000, and NS14000 systems that contain IOAMs: Select System >
Fabric Group > IOAM Enclosure > IOAM > ServerNet Switch Board
2.
Right-click the ServerNet Switch Board object and select Actions.
3.
In the Actions dialog box, select ME Firmware Update and click Perform Action.
4.
Monitor the update.
5.
Repeat for the IOAM switch on the other Internal ServerNet Fabric.
Updating ME Firmware on a VIO Enclosure
There is Maintenance Entity (ME) firmware on the VIO Logic Board object in each VIO enclosure.
NOTE: In the OSM Low-Level Link, a VIO logic board is represented as an IME object.
To use the OSM Service Connection to update the ME firmware for VIO enclosures:
1.
Locate the VIO Logic Board: in the tree pane, select one of the following, depending on the type of server:
•
Select System > Fabric Group > Internal ServerNet Fabric > VIO Module > VIO Logic
Board.
or
•
Select System > Group nnn > VIO Module > VIO Logic Board
2.
Right-click the VIO Logic Board object and select Actions.
3.
In the Actions dialog box, select ME Firmware Update and click Perform Action.
4.
Monitor the update.
5.
Repeat for the VIO module in the other fabric.
Updating ME FPGA
shows the ME FPGA file name and location information for both processor switches and
IOAMs, and for VIO enclosures. For firmware levels, see the NonStop Firmware Matrices document located at: www.hp.com/go/nonstop-docs .
Table 3 Maintenance Entity (ME) FPGA File Names and Locations
Description
ME firmware (see Caution below)
Component
ServerNet Switch Board
(IOAM and processor switch)
ME firmware for IME in VIO enclosures (see Caution below)
VIO enclosure logic board
File Name
SAM0437
SAM0789
Default Location
SYSnn
SYSnn
CAUTION: You should not use the Multi-Resource Actions dialog (in the OSM Service Connection) to update ME FPGA simultaneously on both fabrics, or a system outage might occur.
26 Managing Microcode and Firmware
Use the OSM Service Connection to check and (if necessary) update the ME FPGA for all processor switches, IOAMs, and VIO modules just as you did when updating ME firmware, but choose the
ME FPGA Update action instead.
For information on using the OSM Low-Level Link to update ME FPGA, see the online help available within the application.
Managing NonStop Blade Complex Firmware
Different server types have different blade firmware requirements. This section lists the minimum required blade firmware versions for each type of server.
For more information on architecture types or on the individual server types, refer to the NonStop
NSxxxx Planning Guide for your server. For Carrier Grade systems, your service provider can refer to the NonStop NS-Series Carrier Grade Server Manual.
For NS14000 series and NonStop 16000 series systems, blade firmware is updated through an
OSM guided procedure, launched by performing the Firmware Update action on the Blade Complex object. For more information, see the online help available with the guided procedure.
NOTE: For NS14200 and mixed NS14000/NS14200 systems, and NS16200 and mixed
NS16000/NS16200 systems: You must perform the OSM Initialize Default Firmware File Names action.
To perform the Initialize Default Firmware File Names: When prompted, be sure you select the correct hardware platform from the Configuration list. For more information, see the hardware installation manual or upgrade procedures associated with your system in the Support and Services
Library in NTL.
While this action is not normally required for NS14000 and NS16000 systems, (OSM displays the proper firmware versions by default), you might be directed by the GNSC to use the action and choose Recommended Fix Firmware File Set from the list of options. This causes OSM to show the latest blade firmware available on the SUT as the Default File Version to be used for a firmware update. This may be necessary to incorporate a recent fix not in the original minimum base firmware for your server type.
For details specific to managing firmware in an NS1000, NS1200, NS3000, or NS5000 system, see the upgrade procedures associated with your system in the Support and Services Library in
NTL.
NOTE: In TACL commands and event messages, you might see:
•
The term slice, which is equivalent to the term NonStop Blade Element (NSBE).
•
The term processor complex, which is equivalent to the term NonStop Blade Complex (NSBC)
The OSM GUI now displays the newer Blade Complex and Blade Element terminology.
The following sections dealing with blade complex firmware apply to NS14000 series and
NS16000 series systems only. For other systems, your support provider can see the following sections in the Updating Blade Element Firmware in an NS2000, NS2000T, NS2000ST, NS1200
or NS1000 Server service procedure located in the Support and Service section of the NonStop
Technical Library:
•
Use the procedure described in the “Updating Blade Element Firmware in an NS2000,
NS2000T, NS2000ST, or NS1200 Server” section to update firmware on an NS3000AC server.
•
Use the procedure described in the “Updating Blade Element Firmware in an NS1000 Server” section to update firmware on an NS5000T or NS5000CG server. Note, however, that locations of buttons and LEDs shown in the illustrations differ for the NS5000T/NS5000CG.
Managing NonStop Blade Complex Firmware 27
Updating Blade Complex Firmware for NS14000 Series Systems and NS16000
Series Systems
You can update firmware one NonStop Blade Element at a time or all slices (Blade Elements) at the same time. However, reset and reintegration is performed on one NonStop Blade Element at a time. OSM notifies you to reset and then reintegrate the updated NonStop Blade Element with the running NonStop Blade Elements. OSM also alerts you if a reset action will result in the loss of a logical processor or in the loss of fault tolerance for a logical processor, and advises you on appropriate actions. After a reset action, reintegration begins automatically.
Each NonStop Blade Element contains two local flash memory devices. To ensure redundancy during a firmware update, a duplicate image (or flash image) of each firmware module (Primitive,
ISP1040, PAL/SAL, DISKBOOT, Halted State Services, and Baseboard Management Controller
Firmware) resides on the flash memory devices.
These flash or firmware images as defined and used in the Blade Complex Firmware Update guided procedure are:
•
The shadow image, which is the firmware image in the local flash memory that is not currently running. This firmware image is designed to be used only when the the current running image is unavailable or during a firmware update. Because this image is inactive, it can be updated.
•
The current image, which is the firmware image in the local flash memory that is currently running. Because this image is active, it cannot be updated.
If your firmware module files are downrev, use OSM and the guided procedure to update the shadow and current firmware images (as separate operations) for each NonStop Blade Element and for each NonStop Blade Complex. Before you begin the firmware update procedure, check for any yellow or red alarms associated with a processor (blade) complex. These alarms might indicate a severe problem. You can, however, begin the firmware update if the yellow alarm is associated with a downrev firmware file.
During the firmware update procedure, when you reset and reintegrate the NonStop Blade Element, the current image becomes the shadow image and the shadow image becomes the running image.
In OSM, under attributes, when the firmware update is completed for a NonStop Blade Element, both the current (running) and shadow images will display Same.
When you use the guided procedure:
•
The procedure's main window is not usable for running another guided procedure.
•
The Continue button is dimmed until the processing of a firmware task is completed.
•
Clicking Help in the guided procedure displays the help topic in the guided procedure. For definitions of actions and resources, see the OSM online help.
•
If you interrupt or stop the procedure before the firmware update is completed, the Blade
Complex firmware will be left in an incomplete and inconsistent state. You must restart the procedure from the beginning to perform the procedure again.
•
When the Confirm Firmware Modules Selected for Update dialog box appears, check that the firmware module is selected. (You can use a checkbox at the bottom of the dialog box to select all of the modules.)
•
In the Firmware Update Status screen, you can see the firmware modules updating. Some firmware modules take longer than others. The Continue button is dimmed until all of the modules have completed.
•
Every time you run the Blade Complex Firmware Update procedure, a new XML log file is created on the system console in the directory C:\Documents and
Settings\<username>\Application Data\HP\OSM\pdklogs
. The XML log file does not update automatically as the procedure runs. To update the information in the XML log file, click Refresh in your Web updateable browser.
28 Managing Microcode and Firmware
NonStop Blade Complex Firmware Update for Migrations From H06.03 or H06.04 to This RVU
CAUTION: Do not power cycle any NonStop Blade Element until the firmware update is complete.
If you interrupt or stop the procedure before the firmware update has completed, the NonStop
Blade Element firmware will be left in an incomplete and inconsistent state. Also, if there is an interruption during the firmware update, do not reset or power cycle the NonStop Blade Elements.
Instead, immediately start the procedure from the beginning.
Start and complete the Blade Complex Firmware Update guided procedure:
CAUTION: Ensure that the system is running this new RVU before starting this procedure.
1.
In the tree pane of the OSM Service Connection, right-click the Blade Complex resource.
2.
From the shortcut menu, select Actions.
3.
When the Actions window appears, select Firmware Update from the Available Actions list.
4.
Click Perform action.
5.
In the guided procedures interface, click Start in the Procedure Tasks List pane.
6.
Follow and complete the guided procedure dialog boxes using this summary: a.
Update NonStop Blade Complex (NSBC) firmware:
1) Using the OSM Service Connection and the Blade Complex Firmware Update guided procedure, complete the dialog boxes and update the shadow firmware on each
NonStop Blade Element.
NOTE: OSM will preselect all the firmware that it detects as down-rev.
NOTE: Update all NonStop Blade Elements in the complex at this time.
2) Using the Switch Blade Element Firmware Images dialog box, select Yes.
CAUTION: Do not reset the NonStop Blade Elements after this step.
3) Repeat steps 1 and 2 for the other NonStop Blade Complexes.
b.
Halt and then power cycle all processors:
1) Using OSM Low-Level Link, halt all the logical processors in the system.
2) Power cycle all processors as follows: a) Log on to the OSM Low-Level Link.
b) On the tool bar, click System Discovery.
c) In the Management Window, double-click the system button (labeled with the system name).
d) Right-click the GRP-400 button.
e) Select the Actions option.
f) From the GRP-400 actions list, select Power Cycle All Processors.
g) Click Perform Action.
h) Click OK when asked “Are you sure you want to do a Power Cycle All
Processors?” i) When the Action Status box confirms that the power cycle has successfully completed, close the dialog box.
NOTE: It should take approximately five minutes for all processors to complete
Power On initialization.
3) Load (coldload) the system with this RVU.
c.
Update NonStop Blade Complex (NSBC) firmware on the other flash, as follows:
Managing NonStop Blade Complex Firmware 29
1) Using the OSM Service Connection and the Blade Complex Firmware Update guided procedure, complete the dialog boxes and update the shadow firmware on each
NonStop Blade Element.
NOTE: OSM will preselect all the firmware that it detects as down-rev.
NOTE: Update all NonStop Blade Elements in the complex at this time.
2) Using the Switch Blade Element Firmware Images dialog box, select NO.
3) Repeat steps
and
for the other NonStop Blade Complexes.
NonStop Blade Complex Firmware Update for Migrations From H06.05 and Subsequent RVUs to
This RVU
CAUTION: Do not power cycle any NonStop Blade Element until the firmware update is complete.
If you interrupt or stop the procedure before the firmware update is completed, the NonStop Blade
Element firmware will be left in an incomplete and inconsistent state. Also, if there is an interruption during the firmware update, do not reset or power cycle the NonStop Blade Elements. Instead, immediately start the procedure from the beginning.
To start and complete the Blade Complex Firmware Update guided procedure:
1.
In the tree pane of the OSM Service Connection, right-click the Blade Complex resource.
2.
From the shortcut menu, select Actions.
3.
When the Actions window appears, select Firmware Update from the Available Actions list.
4.
Click Perform action.
5.
In the guided procedures interface, click Start in the Procedure Tasks List pane.
6.
Follow and complete the guided procedure dialog boxes using this summary: a.
Using the OSM Service Connection and the Blade Complex Firmware Update guided procedure, complete the dialog boxes and update the shadow firmware on each NonStop
Blade Element.
NOTE: HP recommends that you update all the NonStop Blade Elements at the same time.
b.
Using the Switch Blade Element Firmware Images dialog box, select Yes to switch the
NonStop Blade Element image from shadow to running.
c.
Reset the NonStop Blade Elements one at a time. Wait for each NonStop Blade Element to be reintegrated before proceeding. Use the OSM Service Connection Blade Element
Reset Action to reset the NonStop Blade Element.
NOTE: HP recommends that you reset and reintegrate each NonStop Blade Element one at a time. (The Reset action power cycles the NonStop Blade Element.) Resetting all
NonStop Blade Elements at the same time will take four processors offline.
Do not manually power cycle the processors. Doing so will leave the NonStop Blade
Elements and NonStop Blade Complex in an incomplete state.
The shadow firmware image that has been updated is now the running flash image. The previous running image is now the shadow image.
d.
Repeat steps a. through c. to update the new shadow image on all NonStop Blade
Elements or one NonStop Blade Element at a time until all the NonStop Blade Elements are done.
e.
Repeat steps a. through d. for the other NonStop Blade Complexes.
30 Managing Microcode and Firmware
Managing ServerNet Adapter Firmware
This manual covers only the ServerNet adapters supported on H06.14 and later RVUs:
•
Fibre Channel ServerNet adapter (FCSA)
•
Gigabit Ethernet 4-port ServerNet adapter (G4SA)
•
6763 Common Communication ServerNet Adapter (CCSA)
•
Token Ring ServerNet adapter (TRSA)
To update firmware for ServerNet adapters connected to legacy IOMF2 CRUs, see the latest
G-series Software Installation and Upgrade Guide.
NOTE: Before connecting your system to a NonStop S-series system through legacy I/O, you must first update the firmware for both the S-series Gigabit 4-port ServerNet adapter (G4SA) and the Fibre Channel ServerNet adapter (FCSA). Firmware for both the G4SA and FCSA adapters used for S-series systems is not initially compatible with other NonStop systems.
If you have a NonStop S-series system that is connected to the your system via Expand or ServerNet
Clusters, you do not need to initialize the G4SA or FCSA firmware.
To see if the ServerNet adapter firmware needs to be updated, navigate down the OSM tree pane for FCSA, G4SA, and CCSA:
CAUTION: Ensure that the system is running this new RVU before starting this procedure.
•
FSCA. The FCSA has two ports that allow connection to external fibre-channel-enabled devices.
You might need to update these firmware files using either SCF or OSM.
◦
The flash boot firmware (T0630G06) contains version information about the boot firmware located in the FCSA FLASH memory. The boot firmware runs on every FCSA after power on and after hardware or firmware errors, and is re-downloaded to initialize the adapter and to accept downloaded firmware from the processor. The AAN SPR is available for this T number.
◦
The flash firmware (T0612G06) contains version information about the operational firmware located in the FCSA FLASH memory. Flash firmware runs the FCSAs used for system load so that the processor boot millicode can read the system-load disk.
◦
RAM firmware: Operational firmware is downloaded to the FCSA when you issue the first START DISK command to the first disk drive in the fibre-channel loop.
◦
Use the OSM Service Connection to check the FCSA’s firmware version and update the firmware, if needed. If the Compare State shows Down-rev, update the firmware. The following example shows that the firmware is at an earlier version and needs to be updated:
Managing ServerNet Adapter Firmware 31
•
G4SA. The OSM Service Connection displays G4SA firmware attributes and action under the G4SA SAC object. To check or update G4SA firmware, locate and expand the G4SA, then select its subcomponent G4SA SAC object. In the following example, the firmware attributes show that the current firmware Version is the same as the available Default File
Version; no update is necessary. If the Compare State was Down-rev, you would use the
Firmware Update action to update the firmware:
•
CCSA. CCSA firmware is preinstalled on CCSAs in a NonStop S-series I/O enclosure. You might need to update the firmware using either OSM or SCF. The OSM Service Connection displays CCSA firmware attributes and action under the CCSA SAC object. To check or update
CCSA firmware using the OSM Service Connection, locate and expand the CCSA, then select its subcomponent CCSA SAC object. If the CCSA firmware attributes show that the current firmware Version is the same as the available Default File Version; no update is necessary. If the Compare State was Down-rev, you would use the Firmware Update action to update the firmware. For more information about the CCSA, refer to the LAN Configuration and
Management Manual and the NonStop Firmware Matrices
•
TRSA. To add or start the TRSA, use the SCF interface to the SLSA subsystem. To check and update the TRSA firmware, use the TSM Service Application as described in the Token-Ring
Adapter Installation and Support Guide. For more information, refer to the LAN Configuration
and Management Manual.
For information on updating FSCA, G4SA, and CCSA firmware, see the OSM online help on the
OSM Multi-Resource Action dialog box.
32 Managing Microcode and Firmware
shows the H-series ServerNet adapters that are supported in this RVU. Note that both the
FCSA and the G4SA are supported in G-series RVUs as well. For product levels, see the NonStop
Firmware Matrices document at: www.hp.com/go/nonstop-docs .
Table 4 H-Series ServerNet Adapters File Names and Locations
Description Component
Fibre Channel
ServerNet
Adapter (FCSA)
Flash firmware
FCSA
FCSA Flash Boot firmware
FCSA
G4SA firmware G4SA
File Name
I0630
C0612R00
C0612R00
Default Location
SYSnn subvolume
SYSnn subvolume
SYSnn subvolume
Tool
OSM Service
Connection or
SCF
OSM Service
Connection or
SCF
OSM Service
Connection or
SCF
Related Information
Fibre Channel
ServerNet Adapter
(FCSA) Installation and Support Guide
Fibre Channel
ServerNet Adapter
(FCSA) Installation and Support Guide
Gigabit Ethernet
4–Port Adapter
Installation and
Support Guide
Distributes SCSI
Boot Flash
Firmware download code for OSM
Update function.
The ISPFIRMW file is now contained in this product.
I/O Interface
Board
FCSA Flash firmware for
VIO
VIO enclosure
ISPFIRMW
I0630
SYSnn subvolume
SYSnn subvolume
OSM Service
Connection or
SCF
OSM Service
Connection or
SCF
FCSA Flash Boot firmware for
VIO
VIO enclosure
G4SA firmware for VIO
VIO enclosure
C0612R00
C0612R00
SYSnn subvolume
SYSnn subvolume
OSM Service
Connection or
SCF
OSM Service
Connection or
SCF
Fibre Channel
ServerNet Adapter
(FCSA) Installation and Support Guide
and hardware installation manuals for the NS1000,
NS1200, and
NS14000 servers
Fibre Channel
ServerNet Adapter
(FCSA) Installation and Support Guide
and hardware installation manuals for the NS1000,
NS1200, and
NS14000 series servers
Gigabit Ethernet
4–Port Adapter
Installation and
Support Guide and hardware installation manuals for NS1000,
NS1200, and
NS14000 series servers
Managing ServerNet Adapter Firmware 33
Managing SWAN and SWAN 2 Concentrator Firmware
Both SWAN and SWAN2 are supported in H-series RVUs and OSM is used to update the firmware.
SCF can also be used. For procedures to update the firmware, see either the SWAN Concentrator
Installation and Support Guide or the SWAN 2 Concentrator Installation and Support Guide.
NOTE: Firmware levels for each component are located in the NonStop Firmware Matrices document.
Managing NonStop Cluster Switch Firmware
To monitor and manage a ServerNet Cluster product configured with 6780 switches, use either the OSM Service Connection or SCF commands. See the ServerNet Cluster 6780 Operations
Guide for more information.
For information on required software, installation, migration, and fallback procedures, see the
ServerNet Cluster 6780 Planning and Installation Guide.
lists the file name and locations for the updateable firmware information for the cluster switch. Check with Scout for the latest SPRs. See the NonStop Firmware Matrices document at: www.hp.com/go/nonstop-docs for product levels.
Table 5 Updateable Firmware for the NonStop 6780 ServerNet Switch File Names and Locations
Firmware Name
Logic board configuration firmware
Router-2 ASIC* firmware
Component
6780 ServerNet
Switch
Cluster switch firmware
File Name
M2789
6780 ServerNet
Switch (logic board)
C2790
6780 ServerNet
Switch (FPGA** in logic board)
LBC2819
Default Location Tool
SYSnn subvolume SCF or OSM SC
SYSnn subvolume SCF or OSM SC
SYSnn subvolume SCF or OSM SC
Related
Information
ServerNet Cluster
6780 Operations
Guide
ServerNet Cluster
6780 Operations
Guide
ServerNet Cluster
6780 Operations
Guide
Managing CLIM Firmware
If your system has CLIMs attached, it is very important that your CLIM firmware is compatible with your system. Refer to the NonStop Firmware Matrices, located in the Service and maintenance
information section of the HP Integrity NonStop Service Information collection of the NonStop
Technical Library at http://www.hp.com/go/nonstop-docs , and ensure you have compatible firmware levels for all CLIM components.
For instructions on upgrading CLIM firmware when a system is halted and when it is running, see the Cluster I/O Protocols (CIP) Configuration and Management Manual.
In the past, the OSM Service Connection was used to perform the upgrade. The Down System
CLIM Firmware Update Tool now provides a faster, more efficient method for updating CLIM firmware in a single action while the system is halted during an RVU upgrade. To use this method, you must have:
•
OSM Service Connection version T0682H02^ACV or later.
•
OSM Console Tools product T0634 G06 ABB or later. When installed on the NonStop system console (as part of OSM Console Tools), the Down System CLIM Firmware Update Tool is launched from the Windows Start menu by selecting All Programs > HP OSM > Down System
CLIM Firmware Update Tool. For information on installing OSM Console Tools, see the NonStop
System Console Installer Guide.
34 Managing Microcode and Firmware
NOTE: See Support Note S12023 (OSM application issues with Administrator user on NSC) for instructions on preventing authentication failures from occurring on NonStop System Consoles when using these OSM tools: Down System CLIM Firmware Update, Update HSS and Copy HSS file actions in OSM Low-Level Link, and connecting to a peer NSC from the NonStop Maintenance
LAN DHCP DNS Configuration Wizard or the CLIM Boot Service Configuration Wizard.
A fix of this problem is automatically installed when you run the Master Installer on NonStop System
Console Installer DVD, Update 18 or later. Support Note S12023 describes another workaround option if you do not have the Update 18 or later DVD.
To perform the CLIM Firmware upgrade when the system is halted, before halting the system, use the OSM Service Connection and perform the Prepare for Down System CLIM Firmware Update action, then use the Down System CLIM Firmware Update Tool. For more information, see the
Cluster I/O Protocols (CIP) Configuration and Management Manual.
Managing CLIM Firmware 35
3 Installation, Migration, and Fallback Considerations
See the Readme file and the appropriate softdocs for the latest information on any migration or configuration steps. Considerations in this chapter are categorized according to the RVU at which they were introduced.
H06.03
Kernel-Managed Swap File (KMSF)
Servers running H-series RVUs require more swap space than do NonStop S-series servers. The default is 512 megabytes for each processor. Depending on your hardware and software configuration, more swap space might be needed. Using KMSF, monitor the actual virtual memory consumption. To check the size of the configured swap files for each processor, at a TACL prompt, start NSKCOM and enter the STATUS SWAPFILE command. For more information on commands and usage, see the Kernel-Managed Swap Facility (KMSF) Manual.
IOAM and Migrating Existing Data to an Enterprise Storage System (ESS)
G-series application programs that reside on an ESS and are run on a NonStop S-series system might require migration changes to run on a NonStop NS-series server and H-series RVU. For more information, refer to the H-Series Application Migration Guide.
The migration of existing data from internal SCSI-attached, 514-byte sector disks to industry standard, 512-byte sector disk subsystems includes an improved method for checksum protection.
The new checksum protection method employed for unstructured files requires the data stored on disk be contiguous and aligned in units of 28 kilobytes.
When a nonpartitioned unstructured file is created or migrated to an ESS, the extent sizes are rounded up to a mod-14 boundary, since an extent page is 2 KB. Rounding ensures continuity of the data set protected by the checksum. This design ensures the best possible performance for unstructured files, without adding significant overhead for checksums. Each 28 KB of data is protected with a 4-KB check block, which contains the checksum information.
Partitioned unstructured files cannot be migrated to an ESS without first performing a migration of all partitions of the unstructured file set to a mod-14 extent boundary. Partitioned unstructured files must use the same extent size and maximum extents for all partitions because positioning is dependent upon a consistent size for each of the partitions. Online migration attempts will abort when an unstructured partitioned file without mod-14 extent sizes is present on the source disk.
Before migrating to an ESS, rebuild all partitions, using a mod-14 extent size. Perform an offline copy to a new partition set.
The FCHECK utility (introduced in G06.24) includes a migration option that reports the files that cannot be migrated to ESS because of this format limitation, in addition to reporting the amount of disk space required for check block protection.
OSS Database Files
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
Interactive Upgrade Guide 2 for new OSS features and for a summary of migration or installation issues.
36 Installation, Migration, and Fallback Considerations
Installing OSS Products Not Managed by DSM/SCM
Use COPYOSS or PINSTALL or both to install products that are not managed by DSM/SCM.
•
If any OSS files are managed by DSM/SCM, use PINSTALL on the remaining pax archives.
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.
•
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. Then 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.
OSS General Alerts and Considerations
Also note these general OSS alerts and considerations:
•
Disk volumes that are accessed by the OSS environment, including the $SYSTEM volume where the OSS Monitor’s configuration files are kept, must not be configured for administration by the HP NonStop Storage Management Foundation (SMF).
•
For the OSS environment, certain configuration files require symbolic links from the OSS file system. For more information, see the OSS Management and Operations Guide.
•
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.
See the inetd(8) reference page in the Open System Services Shell and Utilities Reference
Manual for more information.
OSS Unautomated Configuration and Startup
Unautomated configuration and startup is described in Appendix B in the Open System Services
Management and Operations Guide.
OSS EasySetup (T0585)
The OSS EasySetup product (T0585) is available for automated setup, 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 or when starting, stopping, or removing an OSS subsystem. For setup and fallback purposes, OSS EasySetup provides:
•
A journal file of all installation and configuration actions
•
A file that records the directory structure of the initial OSS file installation
•
A configuration file to partially automate restarting the OSS subsystem. This file can be edited as needed when the configuration changes.
Starting with G06.23, OSS EasySetup no longer requires the process $NULL and the installation of SPI template files.
H06.03
37
PSTATE (T0705H01) and WHOHAS (T0706H01)
PSTATE and WHOHAS are two tools that are owned and supported by the Global NonStop
Solution Center (GNSC). As of H06.03 and G06.26, these two tools are delivered to customers on the SUT.
•
PSTATE provides a snapshot of the status of a running process. It provides everything that
TACL STATUS, DETAIL does plus information about all files open by the process and what
I/Os are outstanding.
•
WHOHAS provides a list of processes that have open files. WHOHAS gives the same information as FUP LISTOPENS but can report on all files, devices and processes in the system.
SANMAN and SNETMON
These two SCF SANMAN (T0502) displays have been enhanced:
•
SCF INFO CONN $ZZSMN
•
SCF STATUS CONN $ZZSMN
Previously, three SANMAN Event Management Service (EMS) events that called out a modular
ServerNet expansion board (MSEB) explicitly for S-series have been modified to report a more generic text message.
Before a your server node can be connected to an existing NonStop ServerNet Cluster with S-series nodes, an upgrade to SNETMON SPR T0294AAP is required on all S-series nodes if these nodes are not already running the minimum required T0294AAP SPR or a subsequent SPR. Failure to install the minimum T0294AAP SPR on the S-series nodes prevents those nodes from successfully discovering server nodes running H-series RVUs and vice-versa, thereby preventing the establishment of Expand-over-ServerNet lines between S-series and H-series server nodes.
T0294AAP can be installed online and does not require a system load.
Remote Database Facility (RDF)
In order to replicate the NonStop Remote Database Facility (RDF, T0346) software from an S-series system to an system running an H-series RVU, the S-series system must be on the T034A06^AAB or later version of RDF.
HP Tandem Failure Data System (TFDS)
HP Tandem Failure Data System (TFDS) is available on the TNS/E platform and provides hardware, software, and subsystem failure analysis and automated processor reload and dump capabilities.
•
TFDSCOM commands have been introduced, existing commands have been modified, and some commands have been eliminated. For example, the fast memory dump (FMD) that is
ON by default in G06.nn RVUs is not available on TNS/E platforms.
•
For new tokens and messages, EMS templates are adapted for servers supporting H-series
RVUs.
For more information on new commands and options, see the Interactive Upgrade Guide 2, and the TFDS Programmer’s Guide.
Minimum Memory Requirements
For information on logical processors available and their sizes, see the appropriate planning guide for your system.
Visual Inspect: Restart $IMON Required
If you use Visual Inspect, you must restart the Inspect Monitor process ($IMON) every time you restart the system.
38 Installation, Migration, and Fallback Considerations
C/C++ Default Floating-Point Type Change
Beginning in the H-series RVUs, the default floating-point format used by the C/C++ compiler is changed. In G-series, the default is Tandem floating-point format; on H-Series, the default IEEE format. The two formats are incompatible; an application must consist of either all Tandem format modules or all IEEE format modules. Certain HP-supplied libraries use the default IEEE format. If user programs also use the default format, there is no problem. However, any user program that explicitly specifies the option to use the Tandem format will encounter a linker error when attempting to load an HP module that uses IEEE format. The recommended workaround is to compile applications using IEEE format.
Providers of DLLs should also be aware of this default change, and take steps to provide libraries that are compatible with both floating-point formats. This can be accomplished in the following ways:
•
Write the library to be floating point neutral
•
Provide two versions of the library, one in Tandem format and one in IEEE format.
H06.05
TMF 3.3
The HP NonStop Transaction Management Facility (TMF) 3.3 was introduced in the H06.05 RVU.
If you do not use any new configurable features after migrating to TMF 3.3, falling back to the
RVU from which you migrated is not a problem. If you use the new TMF 3.3 feature that allows you to add auxiliary audit trails while TMF is started, however, you must first understand how that impacts any subsequent fallback. For H06.11 and later TMF 3.3 fallback and migration issues, see the TMF Planning and Configuration Guide.
The products T8697 (the older product version update of ARLIB) and T8923 (TIFSERVE) are not included in H-series RVUs. T8697 is being replaced by T2781 (ARLIB2). For more information, see the H06.05 Release Version Update Compendium and appropriate TMF softdocs.
TS/MP 2.1
HP NonStop TS/MP 2.1 was introduced in the H06.05 RVU. TS/MP 2.1 provides Pathsend users with higher link capacity to client processes. TS/MP 2.1 is intended only for Pathsend users using
TS/MP 2.0. It does not apply to SCOBOL users or TS/MP 3.0 users.
Migration
In general, all TS/MP 2.0 applications that use Pathsend requestors and TS/MP 2.0 scripts (also called OBEY files) can run in a TS/MP 2.1 environment. However, you might need to modify the
PATHMON process configuration to achieve optimum TS/MP 2.1 performance.
New ACS processes must be started via SCF at system startup and must be managed. These processes are $ZPnn, $ZLnn, and $ZCSnn.
For details on these issues, see the TS/MP Supplement for Limits Relief.
Fallback
Consult the Global NonStop Solution Center (GNSC) if you have installed TS/MP 2.1 and run applications that use Pathsend requestors, and then decide to fall back to TS/MP 2.0.
ServerNet Cluster
Starting in H06.05, the HP NonStop ServerNet Switch (model 6780) can support the long-distance option in a three-zone configuration of the layered topology. Before this RVU, the long-distance option was restricted to a two-zone configuration. To use the ServerNet cluster long-distance option
H06.05
39
in a three-zone configuration, you must install the H06.05 or later H-series RVU or the
T0502H01^AAP or later SPR.
HP recommends upgrading your servers to the H06.05 or later H-series RVU if the servers will be connected to a long-distance ServerNet Cluster. If the long-distance cluster also contains NonStop
S-series servers, HP recommends upgrading those servers to the G06.28 or later G-series RVU.
H-series RVUs starting with H06.05 contain T9050H02^AQD or later, and G-series RVUs starting with G06.28 contain T9050G09^AQB or later. These two T9050 SPRs correct a performance defect that affects Inter-Processor Communication (IPC) long-distance cluster transfers to and from
NonStop servers running H-series RVUs. For details, see the T9050H02^AQD and T9050G09^AQB softdocs.
Other than the T0502H02^AAP or later SPR, a three-zone long-distance configuration requires the same SPRs as a two-zone long-distance configuration.
SMF
Beginning in the H06.05 RVU, SMFINTRFCE (T0063) is merged into the SMF (T8472) product.
The merging of these T-numbers can cause a problem if you are installing the SMF bundle
(T8472H01^ACW or later) on a pre-H06.05 system without installing the entire SUT. (The problem does not occur if you are installing the entire G06.28 or later SUT.)
The problem occurs because the merged version of SMF and the existing T0063 component in your current (pre-H06.05) RVU both attempt to place the file ZSMFC in the ZSPIDEF subvolume.
This attempt results in a “duplicate file was found” error during the DSM/SCM installation of
T8472H01^ACW or later.
You can avoid this problem by manually removing T0063 before the upgrade, as follows: During the upgrade operation, DSM/SCM detects the duplicate SMF components and generates a message indicating that T0063 is already present in your current configuration. DSM/SCM then prompts you to either retain or remove it. Respond to this prompt by requesting removal of T0063.
Backup/Restore 2 Products Merged Into Single T-Number
Beginning in the H06.05 RVU, the Backup/Restore 2 products are merged into a single T-number,
T0744. When you use DSM/SCM to install the new SUT, the following screen will appear, prompting you to select the products to retain:
40 Installation, Migration, and Fallback Considerations
Since the new T-number T0744 is replacing the previous T27nn T-numbers, click Continue without selecting any of the T27nn T-numbers. DSM/SCM will remove those T-numbers from your configuration and install T0744.
TNS/R Compilers and Tools on H-Series Systems
Use One Platform for G-Series and H-Series Development
Beginning with H06.05, the G-series TNS/R compilers and development tools are supported in
H-series RVUs. These compilers and tools are:
NMC/NMCPLUS ptal
COBOL nld ld noft
TNS/R native C/C++ compiler
TNS/R pTAL compiler
TNS/R native COBOL compiler
TNS/R native linker (non-PIC)
TNS/R native linker (PIC)
TNS/R native object file tool
These tools enable you to use one environment, the H-series environment, to build programs for execution in either the TNS/R or the TNS/E environment.
NOTE: You must transfer the TNS/R object files to a TNS/R system for execution; you cannot execute them on the TNS/E system.
The H-series TNS/R compilers and tools have all the features and functionality of the G06.20 and later G-series products. TNS/R compilation on H-series systems is supported only for G06.20 and later G-series RVUs.
H06.05
41
For more information about support for G-series tools on H-series systems, see the H-Series
Application Migration Guide or the individual manuals for the tools.
TNS/R SRLs Consolidated Into a Single T-Number
The TNS/R cross compiler SRLs are consolidated into a single T-number, T8888 (named “TNS/R cross compilers SRLs”). This T-number contains all of the runtime libraries and object files that make up the cross compiler environment. The products that are now included in T8888 are:
T8432
T8432
T8473
T9050
T9228
T9627
T9661
T8888PAX
T7923
T7956
T8108
T8160
T8372
T8432
T8432
T8432
T1070
T1248
T1269
T2767
T2767
T2767
T5895
T6572
T-Number
T0179
T0179
T0179
T0248
T0372
T0428
ZSTFNSRL
ZOSSHSRL
ZCOBSRL
ZSECSRL
ZI18NSRL
ZCRTLSRL
CCPPMAIN
CRTLMAIN
CRTLNS
CRTLNS2
ZTLHSRL
ZOSSKSRL
ZZOSSWFSRL
ZOSSESRL
ZUTILSRL
File Name
ZCPLSRL
CPPINIT
CPPINIT2
ZXDRSRL
ZINETSRL
ZRLDSRL
ZOSSCSRL
ZSPTSRL
ZCRESRL
ZSTLSRL
CPPINIT3
CPPINIT4
ZRWSLSRL
ZICNVSRL
Product
C++ Runtime Library version 2
C++ Runtime Library version 2
C++ Runtime Library version 2
XDR Streams
Socket Library rld
OSS Encryption Routines
OSS POSIX Threads
High Performance Native CRE/RTL
C++ Runtime Library version 3
C++ Runtime Library version 3
C++ Runtime Library version 3
C++ Standard Library
ICONV Tables
Standard Functions Runtime Library
OSSFSLIB II
NMCOBOL Runtime Library
OSS Subject API
I18N Locales
Native C Runtime Library
Native C Runtime Library
Native C Runtime Library
Native C Runtime Library
Native C Runtime Library
TOOLS.H++ Library
NonStop Kernel
Filesys Libraries
OSSFSLIB
MC Runtime Library
OSS packaging for all the preceding
42 Installation, Migration, and Fallback Considerations
For reporting problems with any of these SRLs in H-series systems, you should specify the new
T-number T8888. The fix will be available via T8888. If you do report a problem against the old
T-number (for example, T8432 for ZCRTLSRL), the fix will still be available via T8888.
H06.06
OSS Large Files
Starting with the H06.06 RVU, OSS files larger than the previous 2 GB limit (large files) are supported. The OSS file system and utilities are able to handle file sizes up to 1 terabyte.
To implement this capability, the OSS file system and utilities are modified to support Format 2 files (previously, all files were created as Format 1). All large files are created as Format 2; small files (less than 2 GB) can be created as either Format 1 or Format 2. Format 1 files are automatically converted to Format 2 when opened using the OSS 64-bit utilities or Enscribe.
The ability to access large files can lead to problems if you install the H06.06 or later RVU, then need to fall back to an earlier RVU. Format 2 files are not supported in pre-H06.06 RVUs; if your applications have created new Format 2 files (large or small), those files cannot be accessed on a pre-H06.06 system. Following are descriptions of recommended fallback SPRs. For complete details on this issue, see Support Note S06056A, OSS Files Larger Than 2 GB for H-Series RVUs.
DP2
A DP2 fallback SPR, T9053AQM, is available. This fallback SPR is preinstalled in H06.05 systems, so for those systems, no action is required. You can install the fallback SPR on systems running
H06.03 or H06.04, as long as all the prerequisites are met:
•
If you are currently running the H06.03 RVU, consider upgrading to H06.05 before upgrading to this RVU to facilitate fallback if fallback becomes necessary.
•
If you are currently running the H06.04 RVU, HP strongly recommends that you install the fallback SPR before migrating to this RVU. If you do not install the fallback SPR before migrating to this RVU, you must install it during fallback to ensure a successful fallback.
The installation of T9053AQM requires a system load.
WARNING!
If you do not install the DP2 fallback SPR before migrating to this RVU and a fallback becomes necessary, you must perform a system load on the prior RVU, and then, before using any OSS files, you must install and coldload to the DP2 SPR.
Falling back to H06.03 is not supported.
Once the T9053AQM fallback SPR is installed:
•
Format 2 OSS files smaller than the 2 GB limit are converted to Format 1 when opened, so that the files remain usable. The file size cannot exceed the Format 1 limit.
•
Format 2 OSS files larger than the 2 GB limit are no longer accessible through the OSS file system and utilities, and DP2 returns an error for any operation other than purging the files.
However, the files can be accessed from the Guardian environment by using Enscribe.
Note that files smaller than 2 GB that were converted to Format 2 on H06.06 or later appear initially to most utilities (except DSAP/DCOM T9543H01^ABH) as zero-size Format 2 files after fallback. When the files are accessed for the first time after the fallback, they are converted to Format 1 and show their actual size.
If the DP2 fallback SPR is not installed, you cannot access any Format 2 files using either OSS or
Enscribe on a pre-H06.05 system, even if the files are smaller than 2 GB.
DP2 has a dependency on DSAP/DCOM. If you install the DP2 fallback SPR (T9053AQM), you must also install T9543H01^ABH as described under
H06.06
43
DSAP/DCOM
Migration
The DSAPDDL file is changed in the product revision of DSAP/DCOM delivered in the H06.06
and later RVUs (T9543H01^ABH), resulting in changes to the work file layout. You must use the
DSAPDDL file distributed with T9543H01^ABH when running ENFORM queries on the work file created using this SPR. The DSAPDDL file for T9543H01^ABH contains these changes to the directory-record structure:
•
The field named fifth is changed from type binary unsigned to type binary 32 unsigned.
•
These fields are added to the end of the structure:
◦
Field flab2_flags1 of type binary unsigned identifies OSS files larger than 2 GB.
◦
Field ext-0-fsu of type binary unsigned contains the extent size of large OSS files for extent number 0.
◦
Field ext-1-128-fsu of type binary unsigned contains the extent size of large OSS files for extent number 1-128.
Fallback
The H06.06 and later versions of DSAP/DCOM (T9543H01^ABH) are large OSS file safe, but pre-H06.06 versions are not. If you fall back from H06.06 or later to an earlier RVU, HP recommends that you install T9543H01^ABH during the fallback.
NOTE: An application that is “large file safe” causes no data loss or corruption when it encounters an OSS file larger than 2 GB. It may not necessarily have the capability to process OSS files larger than 2 GB, but it has the appropriate logic to handle errors or warnings detected during file manipulation operations and fail gracefully.
Backup and Restore 2 Fallback
OSS files larger than 2 GB that have been backed up cannot be restored to a pre-H06.06 system.
Install the Backup and Restore 2 fallback SPR T0744H01^AAA to restore files smaller than 2 GB in the following situations:
•
If you have a backup tape with a mix of OSS files smaller than 2 GB and larger than 2 GB.
You can restore the smaller files to an H06.05 file set by first installing T0744H01^AAA.
•
If you have created backup tapes on this RVU and you are falling back to H06.05.
Files larger than 2 GB are skipped. T0744H01^AAA does not support pre-H06.05 RVUs.
Guardian Compatibility Issue
OSS support for files greater than 2 GB has resulted on a compatibility issue for applications using
Enscribe APIs to access Guardian files created by OSS APIs.
Prior to the H06.06 RVU, Guardian files created by OSS APIs were always created as Format 1 with a 175 MB maximum size. As of H06.06, Guardian files created by the open() and creat() functions are Format 1 with an approximate 2 GB maximum size, and Guardian files created by the open64() and creat64() functions are Format 2 with an approximate 26 GB maximum size.
The standard OSS utilities, such as cp, pax, and pinstall, now use open64() and creat64(), and thus create a Guardian Format 2 file when given a /G output file name. For example: cp myfile /G/mydisk/mysubvol/myfile
Guardian applications that use file data and ignore file attributes are unaffected. An application that performs a 32-bit FILE_OPEN_ of a Format 1 file will successfully open a Format 2 file as long as the file’s EOF is less than the Format 1 limit of approximately 4 GB.
44 Installation, Migration, and Fallback Considerations
Guardian applications that use file attributes such as file format and maximum size are likely to be affected. These programs might fail when unexpected values are returned (such as Format 2 if only Format 1 is expected).
If a Format 2 file does cause a program to fail, the programmer can do one of the following:
•
Change the program to accept Format 2 files.
•
Protect the program by converting the file to Format 1 using the following commands:
FUP CREATE TEMP, FORMAT-1, TYPE U, CODE 180, &
EXT (28,1400), MAXTENTS 749
FUP COPY MYFILE, TEMP
FUP PURGE MYFILE
FUP RENAME TEMP, MYFILE
•
Protect the program by creating a Format 1 file, then appending the desired data to the file.
For example, replace the command: cp myfile /G/mydisk/mysubvol/myfile with the commands: gtacl -c ’FUP CREATE $mydisk.mysubvol.myfile, FORMAT 1, &
TYPE U, ODDUNSTR, CODE 180, EXT (28,1400), MAXTENTS 749’ cat myfile >> /G/mydisk/mysubvol/myfile
Upgrading Visual Inspect Client to Version 3.x
If you are currently running Visual Inspect client version 2.x, upgrading to version 3.x is optional.
However, to get the full benefit of the features and performance of the H-series Visual Inspect servers and client, you must upgrade to Visual Inspect client version 3.x. If you do plan to upgrade Visual
Inspect client, HP recommends that you upgrade to the latest version.
NetBatch
New NetBatch features require that Enform Plus (T0295) be installed on your system. If Enform
Plus is not installed or licensed, an error occurs.
Safeguard: Changed Attribute Defaults
The default values of certain attributes have been changed in the H03 version of Safeguard and
Standard Security in order to increase the “out of box” password security. Customers who do not wish to adopt the new defaults can follow the normal migration steps for Safeguard. Customers using Standard Security alone are impacted by this change.
Attributes specific to Safeguard configuration are:
Attribute
PASSWORD-ENCRYPT
PASSWORD-MINIMUM-LENGTH
Previous Default Value
OFF
0
New Default Value
ON
6
Attributes specific to the PASSWORD utility of Standard Security are:
Attribute
ENCRYPTPASSWORD
MINPASSWORDLEN
PROMPTPASSWORD
Previous Default Value
OFF
0
OFF
New Default Value
ON
6
BLIND
All attributes are applied as each user changes their password.
H06.06
45
Safeguard Password Encryption
The H03 and later versions of Safeguard and Standard Security improve the cryptology of the user passwords in the NonStop server environments.
If PASSWORD-ALGORITHM is DES or PASSWORD-ENCRYPT is OFF, the password (DES-encrypted or in clear text, respectively) is written to both the existing L/USERID and the new L/USERAX files.
This approach allows for direct fallback to prior versions of Safeguard and Standard Security.
If you enable the new HMAC256 encryption option, each subsequently changed password is encrypted using HMAC with the SHA256 algorithm and stored in L/USERAX. Because earlier versions of the security products do not understand HMAC, fallback requires extra steps. For additional information, see and
“Fallback With Standard Security (Safeguard Not Installed)”
. To assist fallback after PASSWORD-ALGORITHM is set to HMAC256, the DES or clear text version of each preexisting password is retained in L/USERID. When users change their password, the old password in L/USERID is marked as expired as of that date. For a new user added to the system after the algorithm is changed to HMAC256, the password in L/USERID file is deleted.
Migration in a Safeguard Environment
The following procedure is intended to handle any unexpected failures that might occur during
Safeguard migration. The procedure preserves the user/password database, which is necessary to restore the original system user/alias database.
1.
Use VPROC to determine the current versions of:
•
OSMP
•
OSMON
•
SAFEART
•
SAFECOM
2.
Back up current Safeguard files ($*.SAFE.*, $SYSTEM.SYSTEM.USERID, and
$SYSTEM.SYSTEM.USERAX).
3.
Use SAFECOM to build an OBEY file to save the current policy. To create an OBEY file, do the following in SAFECOM:
TACL> safecom/out $system.safe.safevalu
=display as commands on
=info safeguard, detail
The output from these commands is retained in a file named SAFEVALU located in
$SYSTEM.SAFE.
4.
To restore the original settings, run the OBEY file SAFEVALU, created in step 3, in SAFECOM.
If you do not follow the preceding migration steps and you do not want to accept the new default values, then after the new version is installed, use SAFECOM to modify the appropriate attributes.
For more details, see the Safeguard Administrator's Manual.
Fallback in a Safeguard Environment
Because of the new password encryption algorithm, fallback requires advance planning.
In all cases, fall back to the previous version of security software.
If PASSWORD-ENCRYPT is OFF or PASSWORD-ALGORITHM is set to DES, no extra fallback steps are required.
If PASSWORD-ENCRYPT is HMAC256, extra fallback steps are required. When users first change their password after HMAC256 is enabled, they must remember their immediate previous password.
This is especially important for the system administrator. After installing the previous version of
Safeguard and Standard Security:
46 Installation, Migration, and Fallback Considerations
1.
Before starting Safeguard, the system administrator must log on with the old super ID password.
The old password is the one used before the algorithm was changed to HMAC256.
2.
Start Safeguard.
3.
The system administrator must set a grace period for users to change their expired passwords.
4.
Users are prompted to change their password when logging into the system if:
•
Their user account existed before the installation of the H03 version of Safeguard, or
•
They are new users and their password was encrypted in DES or not encrypted at all before PASSWORD-ALGORITHM was changed to HMAC256.
When prompted, users should enter and re-enter a new password and log onto the system as usual.
5.
For new users that were added to the system after PASSWORD-ALGORITHM was changed to HMAC256, the system administrator must reset their passwords to enable them to log onto the system. Otherwise, the users cannot access the system after fallback.
Migration With Standard Security (Safeguard Not Installed)
Once the new version is installed, use the new PWCONFIG utility to modify the appropriate attributes if you do not want to accept the new default values.
Fallback With Standard Security (Safeguard Not Installed)
In all cases, install the previous version of Standard Security.
If ENCRYPTPASSWORD is OFF or ALGORITHM is set to DES, no extra fallback steps are required.
If ALGORITHM is set to HMAC256:
•
When users first change their password after HMAC256 is enabled, they must remember their immediate previous password.
•
After fallback, users must use their old password to log onto the system if:
◦
Their user account existed before the installation of the H03 version of Standard Security, or
◦
They are new users whose password was encrypted in DES or not encrypted at all before
ALGORITHM was changed to HMAC256.
•
If a user was added to the system after ALGORITHM was changed to HMAC256, they must use a blank password to log onto the system.
H06.07
Code Profiling Utilities
The Code Profiling utilities, introduced in the H06.07 RVU, consist of two products: the Code Cover
Utility (codecov, T0746) and the Profile Merge Utility (profmrg, T0747). To use these tools, applications must be compiled using the H06.07 or later versions of the TNS/E native compilers.
The SPR levels of these compilers are:
Compiler
TNS/E COBOL
TNS/E C/C++
TNS/E pTAL
Native c89 utility
Minimum SPR Level
T0356H01^AAI
T0549H01^AAD
T0561H01^AAC
T8164H01^ABE
H06.07
47
These compiler versions provide the required command line options for running the Code Profiling utilities.
Starting in the H06.09 RVU, the Profile Merge Utility (T0747) and Code Cover Utility (T0746) are included on the SUT and on the cross-compiler CD. Previously, the Profile Merge Utility was delivered only on the SUT.
Safeguard Support for OSS Access Control List
The H03 and later versions of Safeguard and Standard Security are enhanced to support the OSS
Access Control List (ACL) feature. A new security group, SECURITY-OSS-ADMINISTRATOR, and a new Safeguard configuration attribute, AUDIT-CLIENT-OSS, are provided to support this feature.
Fallback Considerations for OSS ACL
•
After fallback, the AUDIT-CLIENT-GUARDIAN attribute, which is a synonym for
AUDIT-CLIENT-SERVICE, is longer be applicable.
•
After fallback, if OSS-related operations are to be audited, enable the AUDIT-CLIENT-SERVICE attribute if it is not already enabled.
•
To restore backup tapes from an H06.08 or later RVU to the H06.05, H06.06, or H06.07
RVU, you must install the T0744H01^AAA Backup/Restore 2 SPR. This SPR is required because changes have been made to the information stored with each object. A Backup/Restore 2
SPR is not available for pre-H06.05 systems.
H06.08
OSS Version 1 Filesets Not Supported
As of the H06.08 RVU, Version 1 filesets are not supported. You must upgrade Version 1 filesets to Version 2 or Version 3 filesets.
The H06.08 RVU adds support for Access Control Lists (ACLs) for OSS regular files, directories, first-in/first-out (FIFO) special files, and bound AF_UNIX sockets in OSS Version 3 filesets. To use
OSS ACLs, you must first migrate to the H06.08 or later RVU, then migrate your existing filesets to Version 3 filesets. To migrate filesets, you must stop the OSS filesystem and the OSS environment and use the fsck DIAGNOSE FILESET, UPGRADE command as described in the Open System
Services Management and Operations Guide.
To restore backup tapes from H06.08 and later RVUs onto H06.05 through H06.07 RVUs, you must install the T0744 Backup/Restore 2 SPR. No T0744 Backup/Restore 2 SPR is available for pre-H06.05 RVUs.
Migration
If you migrate from the H06.04 RVU, HP strongly recommends that you install DP2 fallback SPR
T9053H02^AQM before installing this RVU, in case a fallback becomes necessary. The DP2 fallback SPR is included on H06.05 and later SUTs.
Fallback
To fall back, you must install the previous SPRs for the OSS products and for DP2. In addition, in either of these cases, you must run fsck on the fileset to downgrade the fileset to Version 2:
•
If the fsck utility has been run on a fileset to upgrade to Version 3 filesets
•
If the fileset was created on a system running an H06.08 or later H-series RVU
When the fileset is downgraded to Version 2, any optional ACLs associated with the files in the fileset are lost.
48 Installation, Migration, and Fallback Considerations
Safeguard Longer Password Length
Starting with the H04 version of Safeguard, the maximum allowed Safeguard password length is increased from 8 bytes to 64 bytes.
Migration
Users migrating to a version of Safeguard and Standard Security that supports longer passwords should be aware that two new attributes are introduced: PASSWORD-COMPATIBILITY-MODE and
PASSWORD-MAXIMUM-LENGTH. The default setting for PASSWORD-COMPATIBILITY-MODE is
ON, meaning that new passwords are limited to 8 bytes. To enable the longer password feature,
PASSWORD-COMPATIBILITY-MODE must be set to OFF, and the existing
PASSWORD-MAXIMUM-LENGTH attribute must be set to a value greater than 8.
For more information on these and other attribute changes, see the Safeguard Administrator’s
Manual.
There are no special migration considerations for the longer password feature. However, for a mitigation procedure for handling any unexpected failures that might occur during migration, see
“Migration in a Safeguard Environment” (page 46)
.
Fallback in a Safeguard Environment
From the H06.08 and later RVUs, you can fall back to the H06.06 and H06.07 RVUs.
Because longer Safeguard passwords are not recognized in previous RVUs, fallback requires advance planning. At the time of fallback, the super ID password must be less than or equal to 8 bytes. If the password exceeds this length, the system administrator must create a new super ID password of length less than or equal to 8 bytes and set the PASSWORD-COMPATIBILITY-MODE attribute to ON. After installing the previous version of Safeguard, the system administrator must do the following:
1.
Before starting Safeguard, log on using the 8-byte password.
2.
Use the PWCONFIG utility to change the PASSWORD-MINIMUM-LENGTH attribute to a value less than 8 (if it does not already have such a value).
3.
Start Safeguard.
Once these steps are completed, users whose password length is less than or equal to 8 bytes will be able to log on. Users whose password length exceeded 8 bytes before fallback will be unable to log on; the system administrator must change their passwords to have a length less than or equal to 8 bytes.
Fallback with Standard Security (Safeguard Not Installed)
At the time of fallback, the super ID password must be less than or equal to 8 bytes. The fallback steps are:
1.
If the super ID user password is longer than 8 bytes, set or change that password to be less than or equal to 8 bytes with PASSWORD-COMPATIBILITY-MODE set to ON.
2.
Install the previous version of Standard Security.
3.
Log on using the 8 character password.
4.
Use the PWCONFIG utility to change the PASSWORD-MINIMUM-LENGTH attribute to a value less than 8.
Once these steps are completed, users whose password length is less than or equal to 8 bytes will be able to log on. Users whose password length exceeded 8 bytes before fallback will be unable to log on; the system administrator must change their passwords to have a length less than or equal to 8 bytes.
H06.08
49
Compiler Default Change for Object Code Generation
Prior to the H06.08 RVU, the native C/C++, COBOL, and pTAL compilers generated preemptable object code by default. Beginning in H06.08, the compilers generate non-preemptable object code by default. Non-preemptable object code is more efficient than preemptable object code and results in faster compilation and execution. Preemptable object code is required only when building DLLs that provide preemptable symbols. The H06.08 versions of the compilers provide a command line option to specify to specify whether or not preemptable object code is generated.
Compiler Host
TNS/E Guardian
TNS/E OSS and Windows
Windows
Compiler Name Option Name
CCOMP, CPPCOMP, ECOBOL, EPTAL GLOBALIZED c89, ecobol eptal
-Wglobalized
-globalized
Because of the performance implications, developers should specify the option to generate preemptable object only when building DLLs that provide preemptable symbols.
COBOL H06.08 and Later Fallback
Starting with the H06.08 RVU, the ECOBOL compiler provides new syntax to support dynamic memory allocation. This new feature requires the H06.08 or later version of the COBOL Run-Time
Library; thus, applications that use the feature will not run on RVUs prior to H06.08 unless the
COBOL Run-Time Library (T0357AAC) is applied. This SPR is applicable only to H06.06 and later
RVUs. Applications using the new feature cannot be recompiled on RVUs prior to H06.08 unless the H06.08 version of the ECOBOL compiler (T0356AAK) is applied.
C H06.08 and Later Fallback
Beginning in the H06.08 RVU, the TNS/E C compiler (T8164ABH for c89, T0549AAE for CCOMP),
C Standard Headers (T8645H02), and the C run-time libraries (T8432H02 for the C Run-Time
Library, T1269H02 for CRE) provide a subset of new features from the new ISO/IEC 9899:1999 standard.
Applications that use the new run-time library features will not run on RVUs prior to H06.08 unless the new run-time library SPRs are applied. Applications that use the new compiler features cannot be recompiled on RVUs prior to H06.08 unless the new C compiler SPRs and standard header
SPR are applied.
For descriptions of the new c99 standard features currently supported by the C compiler and run-time library, see the C/C++ Programmer’s Guide.
H06.09
OSM Uses Different User ID and Password to Monitor Maintenance Switch and UPS
Beginning in the H06.09 RVU, OSM uses the Simple Network Management Protocol (SNMP) to monitor the Maintenance Switch and UPS (located under the Monitored Service LAN devices branch of the OSM Tree). The user ID and password are no longer used to start monitoring these devices. Instead, the SNMP read Access Community password is required for monitoring the
Maintenance Switch, and both the SNMP Read Access Community and SNMP Write Access
Community passwords are required to start monitoring the UPS. These passwords are established the administrator during initial configuration of the device. The default password for SNMP Read
Access Community is “public”; The default password for SNMP Write Access Community is
“private.”
The Identification Information attribute of the UPS is no longer displayed.
50 Installation, Migration, and Fallback Considerations
Migration Considerations
If a Maintenance Switch or UPS has been configured in a prior version of OSM, the OSM Access
Community Names will not be available. OSM will attempt to access it with the default values of
“public” for read access and “private” for Write access. If these are not the values with which the device has been configured, the device will be displayed as “Down.” To recover, do one of the following:
•
Stop monitoring, and restart monitoring with the correct values.
•
Change the values to the defaults by opening a Telnet session to the device and rediscovering the device in OSM.
For more information and instructions, see the OSM Configuration Guide.
Safeguard Password Quality Enhancements
The T6533H04^ABM and later versions of Standard Security and the T9750H04^AFC and later versions of Safeguard, initially released with the H06.09 RVU, provide these additional password enhancements:
•
Option to enforce certain rules governing user or alias passwords
•
Use of embedded spaces in user or alias passwords
The following new attributes are introduced:
Attribute
PASSWORD-UPPERCASE-REQUIRED
PASSWORD-LOWERCASE-REQUIRED
PASSWORD-NUMERIC-REQUIRED
PASSWORD-SPECIALCHAR-REQUIRED
PASSWORD-SPACES-ALLOWED
PASSWORD-MIN-QUALITY-REQUIRED
Description
When set to ON, passwords must contain at least one uppercase character.
When set to ON, passwords must contain at least one lowercase character.
When set to ON, passwords must contain at least one numeric character.
When set to ON, passwords must contain at least one special character (any nonalphanumeric character except for commas, double quotes, or semicolons).
When set to ON, passwords may optionally contain embedded spaces.
Specifies the minimum number of quality criteria that must be met when a password is set or changed. Valid values are 0 through 4.
The password quality attributes are disabled by default, and must be explicitly enabled through
SAFECOM (if Safeguard is running) or by using the PWCONFIG command. The password quality attributes take effect only if the PASSWORD-ENCRYPT attribute is set to ON and the
PASSWORD-ALGORITHM attribute is set to HMAC256. In addition, for
PASSWORD-SPACES-ALLOWED to take effect, PASSWORD-COMPATIBILITY-MODE must be set to OFF.
For more information on the password quality enhancements, see the Password Quality
Enhancements support note.
Migration
There are no migration considerations specific to the quality password feature. However, a mitigation procedure is available for handling any unexpected failures that might occur during migration.
This procedure is described under
“Migration in a Safeguard Environment” (page 46)
.
H06.09
51
Fallback
Embedded spaces in passwords are not supported in pre-H06.09 RVUs. Any user with a password that contains embedded spaces will be unable to login after fallback. Fallback considerations are:
•
Pre-H06.08 RVUs do not support the Safeguard longer password feature. Therefore, when falling back to a pre-H06.08 RVU, users must first follow the steps described under
“Fallback in a Safeguard Environment” (page 46)
.
•
When falling back to H06.08 or earlier:
◦
The super ID password must not contain embedded blanks.
◦
PASSWORD-COMPATIBILITY-MODE must be set to ON.
◦
Users whose password contains embedded spaces must request a new password.
C H06.09 and Later Fallback
Starting with the H06.09 RVU , the C++ Version 3 runtime library (ZCPP3DLL) is upgraded to the latest release of the Dinkumware C++ runtime library in order to support the ISO/IEC IS 14882 language specification and to upgrade the exception handling mechanism in the H-series C++
Version Neutral DLL (ZCPPDLL) for improved performance.
In the event of a fallback, applications compiled with the upgraded headers must be recompiled; otherwise, they will not execute on the earlier RVU.
H06.10
NonStop SQL/MX 2.3 Interoperability and Installation
NonStop SQL/MX Release 2.3 is introduced in the H06.10 RVU. NonStop SQL/MX Release 2.3, which has no G-series equivalent, provides this new functionality:
•
Provides an online variant of the POPULATE INDEX and MODIFY utilities
•
Generates modules files in Guardian directories
•
Projects only the requested number of rows from the table instead of scanning the complete table
•
Introduces a new Control Query Default statement (CQD)
NonStop SQL/MX Release 2.3 is interoperable with compatible releases of NonStop SQL/MX.
Interoperability between G-series and H-series RVUs is supported by NonStop SQL/MX on H06.03
and later RVUs For more information about the interoperability of NonStop SQL/MX releases, see the SQL/MX Database and Application Migration Guide.
If you have the H06.04 or later RVU on your system, you can install SQL/MX Release 2.3. For installation information, see the SQL/MX Installation and Management Guide.
Installing NonStop SQL/MX Release 2.3
Check that the HP NonStop Transaction Management Facility (TMF) is configured and running on the system node where you will install NonStop SQL/MX. See the TMF Planning and Configuration
Guide for TMF installation guidelines.
NOTE: All NonStop SQL/MX objects must reside on volumes audited by TMF. NonStop SQL/MX program files do not need to reside on audited volumes.
Before installing NonStop SQL/MX, review the instructions for preinstalling and installing NonStop
SQL/MX provided in the SQL/MX Installation and Management Guide.
52 Installation, Migration, and Fallback Considerations
Pathway/iTS 1.1
Pathway iTS/1.1, introduced in the H06.10 RVU, supports the following SCOBOL statements:
•
DIALOG-BEGIN
•
DIALOG-SEND
•
DIALOG-ABORT
•
DIALOG-END
These statements enable users to send more than 32000 bytes of data to the same Pathway server using a context-sensitive communication model.
These new features are provided by the PathTCP4 (T0700) component of Pathway/iTS.
Migration
Pathway/iTS 1.1 includes both PathTCP3 and PathTCP4. PathTCP3 is unchanged.
Migration to PathTCP4 is optional. If you select PathTCP4, examine the SCOBOL client programs for error handling and, optionally, implement the new SCOBOL verbs for dialog-style communication.
To carry the new features to the converted Java clients, first convert the modified SCOBOL application, and then deploy the converted application. To use the new features, you must configure
Path TCP4 as the TCP object in the Pathway environment configuration and determine on which
CPUs PathTCP4 will execute, since it is a Pathsend requestor.
Once modified for PathTCP4, an application cannot be used with PathTCP3.
OSI FTAM and MHS Longer Password Support
Starting with the H06.10 RVU, the OSI FTAM (T6982) and MHS (T7937) products support passwords up to 64 bytes long. The passwords can also include special characters. To use these features, the Standard Security T6533H04 SPR or later must be installed.
H06.11
Safeguard Logon Without a Password
Starting in the H06.11 RVU, Safeguard is enhanced to allow users to logon without a password.
(This capability can be used only if the SUPER.SUPER user has granted the program file that permission through Safeguard.)
Migration
There are no migration or fallback considerations specific to the new logon feature. However, you can use the mitigation procedure described under
“Migration in a Safeguard Environment” (page 46)
to handle any unexpected failures that might occur when migrating to the new version of Safeguard.
For more information, see the Safeguard Administrator's Manual.
NonStop Secure Shell Server
NonStop Secure Shell (SSH) Server was introduced in the H06.11 RVU. NonStop SSH Server provides secure communication between the MR-Win6530 terminal emulator (which replaces
OutsideView) and the NonStop server.
Migrating to SSH Server
NonStop SSH Server is an optional product, and must be set up for use with MR-Win6530. In terms of network management, NonStop SSH Server uses port 22, so that port must be enabled for NonStop SSH Server to work.
H06.11
53
H06.12
DSM/SCM and OSBUILD
If an H06.12 or later host system is used to perform DSM/SCM builds for a target system that is running a release prior to H06.12, you must install the appropriate T0431 (OSBUILD) SPR for that target system before performing the build. If this is not done, then the DSM/SCM Sysgen (OSBUILD) will fail. The build will complete successfully if it is being done for an H06.12 or later system. See
“DSM/SCM and This RVU” (page 18)
for a list of the required SPRs.
H06.13
DSM Tape Catalog Migration
DSM/TC MEDIADBM no longer abends when processing the DSM/TC MEDIACOM RECOVER
DISKFILE command. You no longer have to SQL/MP compile your MEDIADBM files if the system catalog is used. However, if you install the new DSM/TC products while the old version had its configuration tables in a user SQL catalog, DSM/TC will not find the configuration tables.
To prevent this problem from occurring, after the upgrade, SQL/MP compile MEDIADBM, specifying the previous user SQL catalog with the existing configuration tables. Then start to use the tape catalogs.
DSM/TC MEDIACOM now supports media types DLTIII, DLTIV, CART9840, DDS, DDS2, DDS3,
CART3490, SDLT, LTO1, LTO2, and LTO3 in all the tapevolume and pool related commands.
These new tapevolume types are displayed in Mediacom, but might not be visible in the VOLUME table, depending on whether or not the table is created with an older version of DSM/TC
MEDIADBM. For instance, a previously labeled DAT tape will appear in the output of INFO
TAPEVOLUME as “DAT”, while in the VOLUME table it appears as “CARTDAT”. If you want to use the old tapevolumes, then the pool and tapevolumes should be newly specified. After you upgrade, issue these commands:
ALTER POOL pool-name, …, TYPE ANY
ALTER TAPEVOLUME tape-name, POOL pool-name, …, TYPE type
ALTER POOL pool-name, …, TYPE type
H06.14
Minimum SPR Levels Required for Clustering Software
In order to add a BladeSystem node to an existing ServerNet cluster having both S-series and
Modular nodes, the following minimum SPR levels are required for ServerNet Cluster Monitor
(SNETMON, T0294) and ServerNet Cluster Fabric Manager (SANMAN, T0502):
Node
S-Series
Modular and BladeSystem
Minimum Requirement for T0294
T0294AAX
T0294AAW
Minimum Requirement for T0502
T0502AAX (only the PM module (that is, the ZSMNSCF file) has changed)
T0502AAW
TS/MP 2.3 Migration Considerations
TS/MP 2.3 is introduced in the H06.14 RVU.
Default Change
In TS/MP 2.3, the default SECURITY attribute setting is changed from N (any local or remote user) to O (owner only).
54 Installation, Migration, and Fallback Considerations
Migrating From TS/MP 2.0
After the system image is built using TS/MP 2.3, TS/MP 2.0 users need to remove #ROUT from
Persistent Manager ($ZZKRN) and configure #ZACS before starting TS/MP 2.3
Migrating From TS/MP 2.1
If you ordered and received TS/MP 2.1 on a prior H-series RVU, you will automatically receive
TS/MP 2.3 instead of TS/MP 2.1 when you receive an H06.14 or later SUT.
H06.15
Compiling C++ Programs to Run on RVUs Earlier Than H06.15
Objects compiled by the H06.15 and later C++ compilers might require new functions that are in the current C++ Common Runtime Library but not in RVUs earlier than H06.15. Specifically, programs that use either of the following features will fail to load because of unresolved externals
(assuming that they are not linked using the –rld_unresolved ignore option):
•
A try catch block
•
An exception specifier
To ensure that these programs run correctly on RVUs earlier than H06.15, you must specify the
-WRVU=H06.nn parameter on the c89 command (OSS or PC environment), or the RVU=H06.nn parameter on the CPPCOMP command (Guardian environment) where nn indicates the RVU on which the objects will run.
NOTE: nn must be between 08 and 14, inclusive, as the Runtime Library supports only RVUs
H06.08 and later.
You can fix the problem by installing T2831H01^AAD (Common Runtime Library) or later on all target systems that will run programs generated by the H06.15 or later version of the C++ compiler.
NOTE: You must reboot the system after installing T2831H01^AAD.
SQL/MX Versioning Feature Removed
The SQL/MX versioning enhancements that were introduced in the H06.13 RVU have been removed.
The following enhancements are no longer supported:
•
Ability of Expand-connected nodes with compatible SQL/MX versions to interoperate.
•
Predictable failure on access to or from nodes with incompatible SQL/MX versions.
•
Ability of each version of SQL/MX to determine compatibility.
H06.16
AF_UNIX Release 2 Migration and Fallback
Migrating to AF_UNIX Release 2
AF_UNIX Release 2 software completely replaces earlier versions of AF_UNIX Release 1 software.
There is no application migration required for AF_UNIX Release 2. The AF_UNIX Release 1 semantics remain available along with the AF_UNIX Release 2 semantics. AF_UNIX Release 1 semantics are the default semantics. You can choose which semantics to use either programmatically or by setting the Guardian DEFINE =_AFUNIX_PROCESS_NAME.
AF_UNIX Release 2 software replaces the $ZPLS process pair with a per-processor process, $ZLSnn, where nn is the processor number.
H06.15
55
When migrating to AF_UNIX Release 2:
•
If you start $ZPLS in a startup script using the SCF START SERVER command, and failure to start $ZPLS causes the script to be aborted and steps following the failing one to be omitted, modify the script to remove the command to start $ZPLS.
NOTE: In AF_UNIX Release 2, any changes to the #ZPLS attributes will have no effect.
•
If you use the STARTOSS utility to start OSS processes, modify the
target-volume.ZOSSINS.OSSINFIL
file by changing the line
SERVER #ZPLS to
* SERVER #ZPLS
NOTE:
target-volume
is normally $SYSTEM.
NOTE: #ZPLS cannot be deleted from ZOSSSERV but can be deleted from OSSINFIL.
Commenting out the #ZPLS statement from the OSSINFIL prevents the STARTOSS utility from issuing an error during execution.
The $ZLSnn processes are started automatically when the processors start; do not add them to any startup or shutdown scripts.
Falling Back to An Earlier Version of AF_UNIX
If fallback is required:
•
If you start $ZPLS in a startup script using the SCF START SERVER command, modify startup scripts to start $ZPLS using the SCF START SERVER command.
•
If you use the STARTOSS utility to start OSS processes, modify the
target-volume.ZOSSINS.OSSINFIL
file by changing the line
* SERVER #ZPLS to
SERVER #ZPLS
NOTE:
target-volume
is normally $SYSTEM.
This ensures that the OSS Socket Local Server is started correctly as system load.
The OSS Monitor included in AF_UNIX Release 2 software adds database records for the local server 2 (LS2) processes. If fallback is required, HP recommends that, after the system is no longer running an RVU containing AF_UNIX Release 2, you delete these records using the SCF DELETE
SERVER command:
ASSUME $ZPMON DELETE SERVER
#ZLS00 DELETE SERVER #ZLS01
...
DELETE SERVER #ZLS15
If you fail to delete the database records:
•
The SCF INFO SERVER command will show $ZLS00 through $ZLS15 as UNKNOWN servers with no parameters.
•
Attempts to START, STOP, or ALTER $ZLS00 through $ZLS15 will fail with the diagnostic
"Configuration contains invalid data".
For more information about AF_UNIX Release 2, see the Open System Services Programmer's
Guide.
56 Installation, Migration, and Fallback Considerations
CLuster I/O Module (CLIM)
The CLuster I/O Module (CLIM) is introduced in the H06.16 RVU. CLIM is a rack-mounted HP
ProLiant class server that provides connectivity between the NonStop host system and either the network or the storage devices. For CLIM upgrade procedures, see the Cluster I/O Protocols (CIP)
Configuration and Management Manual.
NOTE: Because H06.16 is the earliest H-series RVU to support CLIMs, you cannot fallback CLIMs on an H06.16 RVU system to an earlier H-series RVU.
CLIMs with software associated with H06.16 are not compatible with H06.17 or later RVUs.
However, if you are migrating from an H06.16 RVU to an H06.17 or later RVU, the software upgrade procedures in this manual are ordered so that the CLIM and RVU compatibility is assured because the RVU on the host is updated first, then, while the processors are halted, the CLIMs are updated to the CLIM software version associated with this RVU. When the system is loaded, the
CLIM and the NonStop host are at the same level and are therefore compatible.
If the CLIM is in the STARTED state, you can determine the CLIM software version by using OSM
Low-Level Link, OSM Service Connection or by entering:
SCF-> VERSION CLIM $ZZCIP.clim-name
If the CLIM is not in the STARTED state, you can use the CLIMCMD command with the Linux cat subcommand by entering:
TACL>climcmd clim-name cat /etc/vproc
Multiple versions of CLIM software are allowed on a system as long as all CLIMs have CLIM SPRs that are compatible with the host RVU. See the NonStop CLuster I/O Module (CLIM) Compatibility
Reference for all CLIM and host RVU compatibility information.
See the Cluster I/O Protocols (CIP) Configuration and Management Manual for troubleshooting tips about identifying an incompatibility in CLIM and host CIP software.
H06.17
CLuster I/O Module (CLIM) Software Compatibility
For H06.17 and subsequent J-series RVUs, CLIM software associated with the RVU is compatible with the current RVU and all preceding RVUs (H06.17 and above). See the NonStop CLuster I/O
Module (CLIM) Compatibility Reference for all CLIM and host RVU compatibility information.
NOTE: The NonStop host RVU is not compatible with every CLIM version. The CLIM must be at the NonStop host RVU version or above. Therefore, when you upgrade your NonStop host RVU, you must also update your CLIM to the software level associated with the RVU to which you are upgrading.
If you are migrating from a H06.17 or later RVU, you can perform an online upgrade to the CLIMs before upgrading the host system.
If the CLIM is in the STARTED state, you can determine the CLIM software version by using OSM
Low-Level Link, OSM Service Connection or by entering:
SCF-> VERSION CLIM $ZZCIP.clim-name
If the CLIM is not in the STARTED state, you can use the CLIMCMD command with the Linux cat subcommand by entering:
TACL>climcmd clim-name cat /etc/vproc
Multiple versions of CLIM software are allowed on a system as long as all CLIMs have CLIM SPRs that are compatible with the host RVU. See the NonStop CLuster I/O Module (CLIM) Compatibility
Reference for all CLIM and host RVU compatibility information.
See the Cluster I/O Protocols (CIP) Configuration and Management Manual for troubleshooting tips about identifying an incompatibility in CLIM and host CIP software.
H06.17
57
DSM/SCM Upgrade Issues/Interoperability SPR Requirements
If you are migrating from an H06.16 or earlier RVU to an H06.17 or later RVU, be aware that all the DSM/SCM target systems should be upgraded before the DSM/SCM host is upgraded. The snapshot file created by T6031 H02 ACL has changed and cannot be read by earlier versions of
DSM/SCM. See Support Note S09036 for more information.
H06.18
Falling Back to iLO Firmware T0848AAA
After updating the iLO firmware to T0848AAB or later (iLO version 1.77 or later), subsequent attempts to downgrade to T0848AAA (iLO version 1.50) using OSM Service Connection will generate an action failure message indicating that the downgrade operation has failed: iLo Firmware update failed. Please perform the following steps.
1. Retry the operation.
2. If the operation fails again, reset the iLo and then retry the operation again.
3. If the operation fails a third time, contact your service provider.
Additional information (may be useful to service provider): Firmware Update on Clim: C100251, failed with the following error: comForte SSH client version T9999H06_20Mar2009_comForte_SSH_0085
Downloading firmware (1.50) to iLo ... Failed Error: Firmware download failed.
Error (1): Operation failed.
STOPPED: $Z14N
CPU time: 0:00:00.007
2: Process terminated with fatal errors or diagnostics
Before retrying the downgrade operation, perform a CLIM Rediscovery from the OSM Service
Connection to determine if the iLO firmware has actually been downgraded. The iLO firmware downgrade operation should have succeeded in most cases even though the OSM Service
Connection Action result indicates a failure result. If, after the CLIM Rediscovery action is completed, the reported iLO level matches the expected level, then the failure result associated with the downgrade action can be safely ignored.
DSM/SCM Upgrade Issues/Interoperability SPR Requirements
If you are migrating from an H06.15 or earlier RVU to an H06.18 or later RVU, be aware that the USERID attributes of all persistent processes configured under $ZZKRN might change from
SUPER.SUPER to NULL.NULL. If this occurs on a system where $SYSTEM is using FCDM disk enclosures connected by FCSAs in IOAM enclosures, the system might halt when a system load to the new RVU is attempted. To prevent this problem, you must install DSM/SCM SPR T6031H02^ACJ or later on your system. See Hotstuff HS03058 for more information.
H06.19
NonStop SQL/MX
The H06.19 RVU introduced the large block support feature (32 KB blocks) for NonStop SQL/MX tables. This feature requires a new revision of the HP NonStop Remote Database Facility (RDF,
T0346). For systems running H6.19, the RDF version is compatible only with T0346H09^ACC and later SPRs of RDF. For more information, see the T0346 softdoc.
For the H06.19 RVU, the versions of RDF on primary and backup systems that are supported across different RVUs are:
Table 6 NonStop SQL/MX RDF Requirements
RVU on PRI
H06.19
H06.03 to H06.18
G06.03 and later
RVU on BKUP
H06.19
H06.19
H06.19
RDF SPR on PRI
H09^ACC
H09^ACC
A07^ACA
RDF SPR on BKUP
H09^ACC
H09^ACC
H09^ACC
58 Installation, Migration, and Fallback Considerations
H06.20
This section describes changes to default behavior introduced by new features at H06.20. For a complete list of new features, migration issues and fallback considerations, see the H06.20 Release
Version Update Compendium.
SQL/MX 2.3.4 Requires JDBC T2 Driver
JDBC T2 Driver must be installed on H06.20 and later H-series RVUs to support SQL/MX 2.3.4.
SQL/MX 2.3.4 requires interoperability SPRs. See
“Install Node Interoperability SPRs” (page 74)
for more information.
H06.21
NonStop SOAP 4
NonStop SOAP 4 migration issues exist for NonStop SOAP 3 to NonStop SOAP 4 upgrades. See the H06.21 Release Version Update Compendium for a list of migration issues and workarounds.
New Cluster I/O Module (CLIM) Storage Devices
HP introduces (controlled availability) support for next-generation CLIMs and storage devices. The new CLIMs and storage devices require H06.17 and later J-series RVUs. All software required for these components is included with the H06.21 RVU.
See the planning guide for your system for information about the CLIMs and storage devices.
The new devices require the H06.17 or later H-series RVU. If you must fall back to an earlier RVU, you cannot use the new CLIMs or SAS disk enclosures. If you fall back to an RVU between H06.17
and H06.20 inclusive, you must add SPRs to those RVUs to support the new CLIMs and disk enclosures because falling back to a previous RVU removes the required software. For a list of required SPRs for these components, see the CLuster I/O Module (CLIM) Software Compatibility
Reference.
The next-generation CLIMs and storage devices only work with each other. They cannot be mixed with the previous generation CLIMs and storage devices.
Other CLIM and CIP Changes
The Telco CLIM now supports CLIM failover.
The H06.21 RVU also includes new features to solve the issue of CLIMs trying to send data on down-link-pulse interfaces and to provide more control over routing.
CIP now supports audit logging for commands executed on a CLIM using CLIMCMD.
See the Cluster I/O Protocols (CIP) Configuration and Management Manual for more information about these features.
New Operating System Features
H06.21 adds several new operating system features, most of which have migration and fallback issues. See the H06.21 Release Version Update Compendium for more information about these features.
TMF 3.6
Fallback considerations exist for TMF 3.6 enhancements. See the H06.21 Release Version Update
Compendium for more information.
H06.20
59
iTP WebServer 7.2
Migration and fallback considerations exist for iTP WebServer 7.2. See the H06.21 Release
Version Update Compendium for more information.
Expand Enhancements.
Fallback considerations exist for the Expand enhancements. See the H06.21 Release Version
Update Compendium for more information.
H06.22
This section provides information only for new features and changes that affect the software installation and upgrade steps for H06.22. See the H06.22 Release Version Update Compendium for a complete list of new features and associated migration and fallback alerts.
Changes to SSOCLIM Privilege Requirement
The privilege for running SSOCLIM, which was SUPER.SUPER for H06.20, has been changed to any SUPER group user. SSOCLIM is run after a fallback. For details, see
.
H06.23
See the H06.23 Release Version Update Compendium for a complete list of new features and associated migration and fallback alerts associated with H06.23.
H06.24
This section provides information only for new features and changes that affect the software installation and upgrade steps for H06.24. See the H06.24 Release Version Update Compendium for a list of new features and associated migration and fallback alerts.
Safeguard Features
Migration and fallback considerations exist for the new Safeguard features. See the H06.24
Release Version Update Compendium for information.
NonStop Software Essentials
HP NonStop Software Essentials is a software installation and management tool for NonStop servers. Installed as a plug-in to HP SIM, NonStop Software Essentials uses DSM/SCM server software and provides a replacement for certain DSM/SCM Planner Interface functions that is more secure, easier to use, and overcomes other shortcomings of DSM/SCM. It coexists with
DSM/SCM; you can use either NonStop Software Essentials or the DSM/SCM Planner Interface for tasks in which there is overlapping functionality. For more information, see the NonStop Software
Essentials Installation and Quick Start Guide.
Also see
“NonStop Software Essentials and This RVU” (page 17) .
EXTDECSL
An SPR for T9003H01, ACC, exists in H06.24. If the application is using any of the new features introduced by this SPR, the application requires recompilation. If the application is not using any of the new features introduced by this SPR, recompilation using the SPR preceding the current SPR is enough. If the application is using any of the new features, remove the new features before recompiling the application with a prior SPR of EXTDECS.
60 Installation, Migration, and Fallback Considerations
TNS/E ECOBOL
An SPR for T0356H01, ABI, exists in H06.24. To fall back to a prior RVU, after fallback, recompile the application with the ECOBOL SPR that precedes this SPR.
IB CLIM Fallback Considerations
The IB CLIM is supported as of H06.23. If you fall back to H06.23 and have IB CLIMs, after you reimage the CLIMs, you must select Open for mode and the CLIM model should show 3 copper
2 IB
. You also need to install additional software. See
“Falling Back on the CLIMs, Loading the
Previous RVU, and Resuming Normal Operations ” (page 137)
.
H06.25
See the H06.25 Release Version Update Compendium for a list of new features and associated migration and fallback alerts.
Halt and Hard Reset Processors When Upgrading to H06.25
When upgrading to H06.25 from an earlier RVU, the processors must be hard reset after halting and before loading the system as described in
“4. (Required for Upgrade to H06.25 or Later)
Perform a Hard Reset” (page 114) .
Safeguard
Fallback and migration considerations exist for the new Safeguard features. See the H06.25
Release Version Update Compendium for more information.
SQL/MX Release 3.2
SQL/MX Release 3.2 replaces SQL/MX Release 2.3.4 as the default SQL/MX version on the SUT starting with H06.25 . Users have a choice of staying with the latest SQL/MX Release 2.3.4 SPRs released as Time Critical Fixes (TCFs) or to migrate to SQL/MX Release 3.2. Please refer to the instructions in hotstuff HS03173 for more details before installing H06.25.
NOTE: Starting with SQL/MX Release 3.2, mxtool includes the fixmxcsmd option, which supports the conversion of the SQL/MX Release 2.3.x CPU list format to the SQL/MX Release
3.2.x CPU list format in the ENVIRONMENTVALUES MXCS metadata table. To perform the CPU list format conversion, you must run the mxtool command with the fixmxcsmd option as described in the SQL/MX Connectivity Service Manual for SQL/MX Release 3.2.
H06.25
61
4 Upgrade Checklist
Use the checklist in
to track your progress as you perform the tasks required to upgrade to this RVU and the checklists
and
to record the
information needed when performing these tasks.
lists SPRs that should be installed before upgrading to this RVU.
CAUTION: The upgrade steps must be performed in the order specified in
.
NOTE: To perform these upgrade tasks, you need the NonStop System Console Installer Guide.
If your system has CLIMs, you also need 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 from H06.16 or a later RVU on a system with CLIMs installed, you can perform an online upgrade of the CLIMs before upgrading the host system. (See the Cluster
I/O Protocols (CIP) Configuration and Management Manual for online CLIM software upgrade procedures.) If you have not already upgraded your CLIMs, you will need to upgrade the CLIMs offline as described in this procedure.
You might have to update the service processor (SP) firmware on NonStop S-series systems for
IOMF2 PMF customer-replaceable units (CRUs).
Required Information Checklists
Gather the information in
before you start:
Table 7 Required Information Checklist, Before You Start
Type of Information
Volume where SYSnn subvolumes are located (usually $SYSTEM)
Value
Current SYSnn (before upgrading to this RVU)
FCSA firmware
G4SA firmware
•
ServerNet Switch firmware
•
ServerNet Switch FPGA firmware
CLIM software
CLIM firmware
•
Service Processor
•
Service Processor B
IOAM or VIO firmware
•
ME firmware
•
ME FPGA firmware
SWAN firmware
SWAN2 firmware
62 Upgrade Checklist
Table 7 Required Information Checklist, Before You Start (continued)
Value Type of Information
ServerNet Cluster firmware
•
Cluster Switch
•
Cluster Switch (logic board; router-2)
DSM/SCM product version (server and client)
Processor type
Record the information in
while you perform the upgrade.
Table 8 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
Upgrade Checklist
Go through this list in order and check off each task in the Done column after it is complete.
Table 9 Upgrade Checklist
Task Description
1. Perform all of the necessary pre-installation tasks before updating the system console and receiving the
SUT.
Reference
Chapter 5: “Preparing to Upgrade the Server Software
2. Before you receive and install the
SUT, first copy the DSM/SCM server and client product SPR and SWID SPR from the SUT to your host system.
“Copying the DSM/SCM, NonStop Software Essentials, and SWID SPRs from the SUT Using NonStop Software
or
PVU, and Planner Interface (PI) Client Software ”
3. Back up important files.
4. Record the current SYSnn.
“Back Up Important Files” (page 66)
“Record the Current SYSnn” (page 67)
Done?
5. Save current system configuration file.
6. Install required or recommended
SPRs and fallback SPRs, if necessary.
“Save the Current Configuration File (CONFxxyy)”
• “Using DSM/SCM With NonStop Software Essentials to Install SPRs” (page 153)
or
“Using DSM/SCM With the Planner Interface to Install SPRs”
• “DSM/SCM Upgrade Issues/Interoperability SPR
• “Install Node Interoperability SPRs” (page 74)
(list of required SPRs)
7. Upgrade system console software.
“Updating NonStop System Console Software
Upgrade Checklist 63
Table 9 Upgrade Checklist (continued)
Task Description
8. Verify system operations.
9. Install CLIM software on the
NonStop System Console.
Reference
“Verify System Processes” (page 68)
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)
Done?
10. For those with NonStop Software
Essentials, receive the RVU SUT into
DSM/SCM software archive
“Receive New Software Into the Archive” (page 79)
11. For those with DSM/SCM Planner
Interface, receive this RVU SUT into
DSM/SCM software archive.
“Receive New Software Into the Archive” (page 91)
12. For those with NonStop Software
Essentials, create this RVU software configuration revision.
“Create a New Software Revision” (page 81)
13. For those with DSM/SCM Planner
Interface, create this RVU software configuration revision.
“Create a New Software Revision” (page 95)
14. For those with NonStop Software
Essentials, build and apply this RVU software revision.
“Build and Apply the New Software Revision” (page 84)
15. For those with DSM/SCM Planner
Interface, build and apply this RVU software revision.
“Build and Apply the New Software Revision” (page 98)
16. Print DSM/SCM Operator
Instructions.
17. Prepare for Down System CLIM
Firmware Update action on the OSM
Service Connection.
18. Run ZPHIRNM.
19. Halt the processors.
“Print Operator Instructions From the Target Interface
• “Managing CLIM Firmware” (page 34)
•
OSM Service Connection online help
“2. Halt the Processors” (page 114)
20. If you are upgrading from H06.03
or H06.04, power cycle the processors.
“3. Power Cycle the Processors if Necessary” (page 114)
21. If you are migrating to H06.25 or a later RVY, perform a hard reset of the processors.
“4. (Required for Upgrade to H06.25 or Later) Perform a Hard Reset” (page 114)
22. Update CLIM software on all
“5. (Optional) Upgrade the CLIM Software” (page 115)
CLIMs by performing a CLIM software Cluster I/O Protocols (CIP) Configuration and upgrade using the OSM Low-Level
Link. (If you are migrating from
H06.17 or a later RVU and performed
Management Manual
an online upgrade of the CLIM to the
CLIM software released with this RVU before starting the host system upgrade, you can skip this step.)
“Managing CLIM Firmware” (page 34)
23. Update firmware for CLIM components as needed.
24. Load system with this new RVU.
25. Update ME and ME FPGA firmware, if necessary.
“6. Loading the System” (page 115)
“Managing Maintenance Entity (ME) Firmware and ME
and the CSSI web page
64 Upgrade Checklist
Table 9 Upgrade Checklist (continued)
Task Description
26. For systems with IOAMs, update
IOAM firmware, if necessary.
Reference
“Updating ME Firmware on an I/O Adapter Module
27. Update NonStop Blade Complex
(NSBC) firmware.
28. Update G4SA Download Code firmware, if necessary.
29. Install ISP1040 firmware, if necessary.
“Managing NonStop Blade Complex Firmware”
“Managing ServerNet Adapter Firmware” (page 31)
“Managing ServerNet Adapter Firmware” (page 31)
30. Update FCSA firmware, if necessary.
31. Recover the TCP/IP files, if necessary.
32. Check processor status.
“Managing ServerNet Adapter Firmware” (page 31)
“Update NonStop TCP/IP Configuration Files” (page 122)
33. Configure WANMGR, if necessary.
“Check Processor Status ” (page 68)
“Configure the WAN Manager” (page 125)
34. Update SWAN concentrator CLIP firmware , if necessary.
SWAN Concentrator Installation and Support Guide or
SWAN 2 Concentrator Installation and Support Guide
35. Start system software.
36. Configure OSS, if necessary.
37. Configure SMF, if necessary.
“Start User and System Applications” (page 120)
“Configure SMF (Optional)” (page 125)
38. Configure RDF, if necessary.
“Configure the Remote Database Facility (RDF) ”
“Re-configure iTP Webserver” (page 126)
39. Re-configure iTP WebServer, if installed on the previous RVU.
49. Re-deploy NonStop SOAP 4, if installed on the previous RVU.
41. Start applications.
“Re-deploy NonStop SOAP 4” (page 126)
Your application documentation
Done?
Fallback SPR Summary
Table 10 Recommended Fallback SPRs
Fallback SPR Current RVU
T9053AQM (DP2)
T9543ABH (DSAP/DCOM) pre-H06.05
pre-H06.06
T0744 (Backup/Restore 2) H06.05 only (not available for earlier RVUs)
Requires Coldload?
For More Information
Yes
No
No
See
See
See
Fallback SPR Summary 65
5 Preparing to Upgrade the Server Software (Pre-installation
Tasks)
This chapter describes tasks that are needed before updating the system console and receiving the SUT.
Back Up Important Files
Before you upgrade, create tape back-ups of critical host applications and files.
Back Up NonStop SQL/MP
HP 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 NonStop SQL/MX Installation and Management Guide.
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
Back Up TMF Files
Use the OBEYFORM option available in TMFCOM INFO commands to save current configuration information in command files.
Back Up CLIM Configuration Files
Back up CLIM configuration files for each CLIM using the TACL command CLIMBKUP.
NOTE: If you must fall back to this RVU, you need to restore these saved configuration files because the configuration file version must match the NonStop system RVU.
66 Preparing to Upgrade the Server Software (Pre-installation Tasks)
If you are upgrading from H06.16, create a subvolume to store the backups, then rename each backup file after you create it. The default file name for the backup is backupgz. Rename these as you create them. For example:
1.
CLIMBKUP N100251 $DATA1.BCKH0616
2.
FUP RENAME $DATA1.BCKH0616.BACKUPGZ, $DATA1.BCKH0616.N100251
3.
CLIMBKUP N100282 $DATA1.BCKH0616
4.
FUP RENAME $DATA1.BCKH0616.BACKUPGZ, $DATA1.BCKH0616.N100282
For upgrades from H06.17 and later RVUs, you can specify a unique filename in the same command. For example,
>CLIMBKUP N100251 $system.N100251.BCKH0617
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.
Record the Current SYSnn 67
Verify System Processes
Check Processor Status
1.
Log on to the OSM Low-Level Link.
The OSM Low-Level Link appears.
2.
Select Processor Status.
The Processor Status dialog box appears. The Processor Status for all processors should be
Executing NonStop OS
.
Check 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.
68 Preparing to Upgrade the Server Software (Pre-installation Tasks)
The System Status window indicates the overall health of a system by the color of the system icon.
From the Summary menu, select System Status. Yellow indicates a possible system problem.
For example:
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 service provider.
3.
For each NonStop Blade Complex (Processor Complex), double-click the NonStop Blade
Complex (Processor Complex).
4.
For each Logical Processor in the NonStop Blade Complex (Processor Complex): a.
Select the Logical Processor.
b.
In the Attributes tab, verify that the NSK State attribute has a value of Up and that the
Halt Flag attribute has a value of False.
NOTE: If the Halt Code or Halt Flag attributes do not match these values, see the
Processor Halt Codes Manual.
Verify System Processes 69
Verify the State of the Internal ServerNet Fabric
To verify that the paths and connections within the ServerNet system area network (ServerNet SAN) are properly functional:
1.
Log on to the OSM Service Connection.
2.
In the tree pane, double-click the Fabric Group.
3.
For each Internal ServerNet Fabric, double-click the fabric.
4.
Select each Processor Switch Power Supply.
5.
In the Attributes tab, if the Service State attribute is visible, verify that it has a value of OK.
(The Service State attribute is visible in OSM only if attention or service is required.)
6.
Check for alarms on the Internal ServerNet X and Y Fabric.
7.
If an alarm exists: a.
Select the fabric object displaying the alarm.
b.
Select the Alarms tab of the details pane.
c.
Select the individual alarm.
d.
Select Detail.
8.
Diagnose and resolve any problems that are causing the alarm. If necessary, contact your service provider before continuing with the migration.
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
Complex Firmware for NS14000 Series Systems and NS16000 Series Systems” (page 28)
and the OSM Service Connection User’s Guide.
Verify Critical System Processes
From a TACL prompt, log on to the system using the super ID (255,255). Enter the LISTDEV command:
QA2100.$SYSTEM.STARTUP 1> scf
SCF - T9082H01 - (23JUN11) (02MAY11) - 01/17/2013 10:44:02 System \QA2100
(C) 1986 Tandem (C) 2006 Hewlett Packard Development Company, L.P.
(Invoking \QA2100.$MAINT.ZQATOOLS.TIPCFSCF)
1-> listdev
70 Preparing to Upgrade the Server Software (Pre-installation Tasks)
LDev Name PPID BPID Type RSize Pri Program
0 $0 0,3 1,3 ( 1,0 ) 102 201 \QA2100.$SYSTEM.SYS00.OPCOLL
1 $NCP 0,14 1,13 (62,6 ) 3 199 \QA2100.$SYSTEM.SYS00.NCPOBJ
3 $YMIOP 0,5 1,5 ( 6,4 ) 80 205 \QA2100.$SYSTEM.SYS00.TMIOP
5 $Z0 0,7 1,7 ( 1,2 ) 102 200 \QA2100.$SYSTEM.SYS00.OCDIST
6 $SYSTEM 0,257 1,257 ( 3,51) 4096 220 \QA2100.$SYSTEM.SYS00.TSYSDP2
7 $ZOPR 0,8 1,8 ( 1,0 ) 102 201 \QA2100.$SYSTEM.SYS00.OAUX
79 $ZZKRN 0,302 1,417 (66,0 ) 4096 180 \QA2100.$SYSTEM.SYS00.OZKRN
81 $ZZWAN 0,299 1,373 (50,3 ) 132 180 \QA2100.$SYSTEM.SYS00.WANMGR
82 $ZZSTO 0,300 1,418 (65,0 ) 4096 180 \QA2100.$SYSTEM.SYS00.TZSTO
83 $ZZLAN 0,301 1,374 (43,0 ) 132 180 \QA2100.$SYSTEM.SYS00.LANMAN
84 $ZZCIP 0,362 1,356 (74,0 ) 132 200 \QA2100.$SYSTEM.SYS00.CIPMAN
106 $ZSNET 0,302 1,417 (66,0 ) 4096 180 \QA2100.$SYSTEM.SYS00.OZKRN
107 $ZSLM2 0,297 1,339 (67,0 ) 1024 221 \QA2100.$SYSTEM.SYS00.TZSLM2
108 $ZPTY 2,417 3,395 (46,0 ) 6144 150 \QA2100.$SYSTEM.ZSSH.STN
109 $ZPMON 0,365 (24,0 ) 4096 146 \QA2100.$SYSTEM.SYS00.OSSMON
113 $ZNET 0,18 1,17 (50,63) 3900 175 \QA2100.$SYSTEM.SYS00.SCP
126 $ZM03 3,289 (45,0 ) 132 201 \QA2100.$SYSTEM.SYS00.QIOMON
127 $ZM02 2,289 (45,0 ) 132 201 \QA2100.$SYSTEM.SYS00.QIOMON
128 $ZM01 1,290 (45,0 ) 132 201 \QA2100.$SYSTEM.SYS00.QIOMON
129 $ZM00 0,304 (45,0 ) 132 201 \QA2100.$SYSTEM.SYS00.QIOMON
130 $ZLOG 0,361 1,427 ( 1,0 ) 4024 150 \QA2100.$SYSTEM.SYS00.EMSACOLL
179 $ZCM03 3,365 (74,1 ) 57344 201 \QA2100.$SYSTEM.SYS00.CIPMON
180 $ZCM02 2,333 (74,1 ) 57344 201 \QA2100.$SYSTEM.SYS00.CIPMON
181 $ZCM01 1,386 (74,1 ) 57344 201 \QA2100.$SYSTEM.SYS00.CIPMON
182 $ZCM00 0,369 (74,1 ) 57344 201 \QA2100.$SYSTEM.SYS00.CIPMON
183 $YOSQA37 0,321 1,357 (63,0 ) 3 199 \QA2100.$SYSTEM.SYS00.LHOBJ
184 $YOSQA3 0,314 1,363 (63,0 ) 3 199 \QA2100.$SYSTEM.SYS00.LHOBJ
186 $SUITES 1,334 0,344 ( 3,51) 4096 220 \QA2100.$SYSTEM.SYS00.TSYSDP2
187 $SUITE2 2,311 3,337 ( 3,51) 4096 220 \QA2100.$SYSTEM.SYS00.TSYSDP2
188 $SQL11 3,334 2,301 ( 3,51) 4096 220 \QA2100.$SYSTEM.SYS00.TSYSDP2
189 $SQL10 2,302 3,332 ( 3,51) 4096 220 \QA2100.$SYSTEM.SYS00.TSYSDP2
190 $SQL09 1,331 0,345 ( 3,51) 4096 220 \QA2100.$SYSTEM.SYS00.TSYSDP2
191 $SQL08 0,346 1,329 ( 3,51) 4096 220 \QA2100.$SYSTEM.SYS00.TSYSDP2
192 $SQL07 3,330 2,303 ( 3,51) 4096 220 \QA2100.$SYSTEM.SYS00.TSYSDP2
193 $SQL06 2,304 3,326 ( 3,51) 4096 220 \QA2100.$SYSTEM.SYS00.TSYSDP2
194 $SQL05 1,326 0,347 ( 3,51) 4096 220 \QA2100.$SYSTEM.SYS00.TSYSDP2
195 $SQL04 0,348 1,324 ( 3,51) 4096 220 \QA2100.$SYSTEM.SYS00.TSYSDP2
196 $SQL03 3,324 2,305 ( 3,51) 4096 220 \QA2100.$SYSTEM.SYS00.TSYSDP2
197 $SQL02 2,306 3,322 ( 3,51) 4096 220 \QA2100.$SYSTEM.SYS00.TSYSDP2
198 $SQL01 1,322 0,349 ( 3,51) 4096 220 \QA2100.$SYSTEM.SYS00.TSYSDP2
199 $SQL00 0,350 1,319 ( 3,51) 4096 220 \QA2100.$SYSTEM.SYS00.TSYSDP2
200 $SCALE1 0,313 1,364 (63,0 ) 3 199 \QA2100.$SYSTEM.SYS00.LHOBJ
201 $SAS313 3,320 2,307 ( 3,51) 4096 220 \QA2100.$SYSTEM.SYS00.TSYSDP2
202 $SAS301 2,308 3,317 ( 3,51) 4096 220 \QA2100.$SYSTEM.SYS00.TSYSDP2
203 $SAS125 2,309 3,315 ( 3,51) 4096 220 \QA2100.$SYSTEM.SYS00.TSYSDP2
205 $QA2200 0,319 1,359 (63,0 ) 3 199 \QA2100.$SYSTEM.SYS00.LHOBJ
206 $POSQA1 0,320 1,358 (63,0 ) 3 199 \QA2100.$SYSTEM.SYS00.LHOBJ
207 $PITSMP1 0,312 1,365 (63,0 ) 3 199 \QA2100.$SYSTEM.SYS00.LHOBJ
208 $OSS3 3,313 2,310 ( 3,51) 4096 220 \QA2100.$SYSTEM.SYS00.TSYSDP2
209 $OSS2 2,291 3,311 ( 3,51) 4096 220 \QA2100.$SYSTEM.SYS00.TSYSDP2
210 $OSS1 1,317 0,351 ( 3,51) 4096 220 \QA2100.$SYSTEM.SYS00.TSYSDP2
211 $OSS 0,322 1,315 ( 3,51) 4096 220 \QA2100.$SYSTEM.SYS00.TSYSDP2
231 $MAINT 0,323 1,312 ( 3,51) 4096 220 \QA2100.$SYSTEM.SYS00.TSYSDP2
232 $LRGQA 0,317 1,361 (63,0 ) 3 199 \QA2100.$SYSTEM.SYS00.LHOBJ
233 $KMSF1 0,324 1,307 ( 3,51) 4096 220 \QA2100.$SYSTEM.SYS00.TSYSDP2
234 $J06ALT 2,292 3,309 ( 3,51) 4096 220 \QA2100.$SYSTEM.SYS00.TSYSDP2
235 $J0617 2,293 3,307 ( 3,51) 4096 220 \QA2100.$SYSTEM.SYS00.TSYSDP2
236 $J0616 2,294 3,304 ( 3,51) 4096 220 \QA2100.$SYSTEM.SYS00.TSYSDP2
237 $J0614 0,325 1,304 ( 3,51) 4096 220 \QA2100.$SYSTEM.SYS00.TSYSDP2
238 $GUEST 3,301 2,295 ( 3,51) 4096 220 \QA2100.$SYSTEM.SYS00.TSYSDP2
239 $FULL 3,299 2,296 ( 3,51) 4096 220 \QA2100.$SYSTEM.SYS00.TSYSDP2
240 $EMPTY2 3,297 2,297 ( 3,51) 4096 220 \QA2100.$SYSTEM.SYS00.TSYSDP2
241 $EMPTY1 2,298 3,295 ( 3,51) 4096 220 \QA2100.$SYSTEM.SYS00.TSYSDP2
242 $DUMP3 1,302 0,326 ( 3,51) 4096 220 \QA2100.$SYSTEM.SYS00.TSYSDP2
243 $DUMP2 1,300 0,327 ( 3,51) 4096 220 \QA2100.$SYSTEM.SYS00.TSYSDP2
244 $DUMP1 1,298 0,328 ( 3,51) 4096 220 \QA2100.$SYSTEM.SYS00.TSYSDP2
246 $BLSYSQA 0,318 1,360 (63,0 ) 3 199 \QA2100.$SYSTEM.SYS00.LHOBJ
247 $BLSCALE 0,310 1,367 (63,0 ) 3 199 \QA2100.$SYSTEM.SYS00.LHOBJ
248 $BLISTER 0,315 1,362 (63,0 ) 3 199 \QA2100.$SYSTEM.SYS00.LHOBJ
249 $BLESSIT 0,311 1,366 (63,0 ) 3 199 \QA2100.$SYSTEM.SYS00.LHOBJ
250 $AUDTR4 3,293 2,299 ( 3,51) 4096 220 \QA2100.$SYSTEM.SYS00.TSYSDP2
Verify System Processes 71
251 $AUDTR3 2,300 3,291 ( 3,51) 4096 220 \QA2100.$SYSTEM.SYS00.TSYSDP2
252 $AUDTR2 1,296 0,329 ( 3,51) 4096 220 \QA2100.$SYSTEM.SYS00.TSYSDP2
253 $AUDTR1 0,330 1,294 ( 3,51) 4096 220 \QA2100.$SYSTEM.SYS00.TSYSDP2
254 $AUDIT 0,331 1,292 ( 3,51) 4096 220 \QA2100.$SYSTEM.SYS00.TSYSDP2
268 $ZTCP0 0,424 1,419 (48,0 ) 57344 201 \QA2100.$SYSTEM.SYS00.CIPSAM
293 $ZTC0 0,450 1,391 (48,0 ) 57344 201 \QA2100.$SYSTEM.SYS00.CIPSAM
295 $ZTN0 0,451 1,413 (46,0 ) 6144 149 \QA2100.$SYSTEM.SYS00.TELSERV
296 $ZOLHD 0,531 1,448 ( 1,30) 132 150 \QA2100.$SYSTEM.SYS00.EMSDIST
299 $ZTNP0 0,455 1,421 (46,0 ) 6144 170 \QA2100.$SYSTEM.SYS00.TELSERV
300 $ZTCP1 1,422 0,456 (48,0 ) 57344 201 \QA2100.$SYSTEM.SYS00.CIPSAM
302 $ZTNP1 1,426 0,458 (46,0 ) 6144 170 \QA2100.$SYSTEM.SYS00.TELSERV
331 $ZEXP 0,61 1,38 (63,30) 132 149 \QA2100.$SYSTEM.SYS00.OZEXP
350 $ZRD9 1,447 2,427 ( 1,30) 132 150 \QA2100.$SYSTEM.SYS00.EMSDIST
918 $ZHOMR 0,89 ( 6,30) 80 151 \QA2100.$SYSTEM.ZSYSEX.HOMER
2->
Verify Disk Drives
1.
Enter the SCF STATUS DISK command:
> SCF STATUS DISK $*
2.
Verify that the STATUS DISK display includes all the disk drives and that the primary, backup, mirror, and mirror-backup paths are all described as STARTED:
$SYSTEM TOOLS 48> SCF STATUS DISK $*
SCF - T9082H01 - (01MAY05) (15MAR05) - 07/06/2005 10:55:16 System \ATLAS
(C) 1986 Tandem (C) 2003 Hewlett Packard Development Company, L.P.
STORAGE - Status DISK \ATLAS.$SYSTEM
LDev Primary Backup Mirror MirrorBackup Primary Backup
PID PID
6 *STARTED STARTED STOPPED STOPPED 0,257 1,257
STORAGE - Status DISK \ATLAS.$AD01
LDev Primary Backup Mirror MirrorBackup Primary Backup
PID PID
152 *STARTED STARTED STOPPED STOPPED 0,333 1,288
STORAGE - Status DISK \ATLAS.$ADAUD
LDev Primary Backup Mirror MirrorBackup Primary Backup
PID PID
151 *STARTED STARTED *STARTED STARTED 0,332 1,289
STORAGE - Status DISK \ATLAS.$ADBIT
LDev Primary Backup Mirror MirrorBackup Primary Backup
PID PID
150 *STARTED STARTED *STARTED STARTED 0,331 1,290
STORAGE - Status DISK \ATLAS.$ADSAV
LDev Primary Backup Mirror MirrorBackup Primary Backup
PID PID
149 *STARTED STARTED STOPPED STOPPED 0,330 1,291
Total Errors = 0 Total Warnings = 0
$SYSTEM TOOLS 49>
Verify Tape Drives (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. Note that OSM does not test tape libraries.
72 Preparing to Upgrade the Server Software (Pre-installation Tasks)
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 State Sub State Primary Backup Device Status
PID PID
71 STARTED UP 0,282 1,273 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 Present State Trace
N100251 Yes STARTED OFF
N100374 Yes STARTED 1, 2
N101254 Yes STARTED 2
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 Status TPStatus Trace
ZCSAM STARTED Started OFF
PROV1 STARTED Invalid (0,1,2,3,4,5,6,7,8,9,10,11,12,13,14,15)
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:
1> spoolcom $spls; collect, status
COLLECT STATE FLAGS CPU PRI UNIT DATA FILE %FULL
$S ACTIVE 3 , 1 154 4 $OPS0.SPL.DATA 44
Check for Sufficient Swap File Space
You must ensure that you have configured kernel-managed swap files of sufficient size for all your system and application software processes. The optimal amount of swap space varies, depending on the configuration of your system. You must be a super-group user (255, n) in order to change the KMSF configuration by adding, altering, stopping, or deleting swap space. Only one super-group user at a time can use NSKCOM to change the configuration.
Verify System Processes 73
KMSF swap files need to be larger than in G-series RVUs. The best practice is to monitor swap file usage. See the Kernel-Managed Swap Facility (KMSF) Manual for information on configuring swap file space.
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 (01OCT04) - Apr 15 2005
Copyright 1995 Compaq Computer Corporation
$SYSTEM.SYSTEM.ZSYSCFG
KMS.SWAPFILE = 0 $SYSTEM.ZSYSSWAP.SWAP00
KMS.SWAPFILE = 1 $SYSTEM.ZSYSSWAP.SWAP01
KMS.SWAPFILE = 2 $SYSTEM.ZSYSSWAP.SWAP02
KMS.SWAPFILE = 3 $SYSTEM.ZSYSSWAP.SWAP03
2.
Issue the 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 *
NSKCOM - T5838H01 BASE (01OCT04) - Jun 17 2005
Copyright 1995 Compaq Computer Corporation
$SYSTEM.SYSTEM.ZSYSCFG
KMS.SWAPFILE = 0 $SYSTEM.ZSYSSWAP.SWAP00
KMS.SWAPFILE = 1 $SYSTEM.ZSYSSWAP.SWAP01
KMS.SWAPFILE = 2 $SYSTEM.ZSYSSWAP.SWAP02
KMS.SWAPFILE = 3 $SYSTEM.ZSYSSWAP.SWAP03
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..
Install Node Interoperability SPRs
Node interoperability is the ability of the nodes (systems) in an Expand network to communicate and interact with each other. An Expand network can be interconnected using various technologies
74 Preparing to Upgrade the Server Software (Pre-installation Tasks)
such as Expand over IP and ServerNet Cluster. See the NonStop Networking Overview for more information about and diagrams of NonStop platform interoperability.
In addition to interoperability issues for nodes connected through Expand, there are software migration and software management issues to consider, such as accelerating TNS object files on a system running an RVU series for posterior use in a system running another RVU series, DSM/SCM software management in an environment composed of systems running different RVU series, and moving objects built on a system running an RVU series to a system running another RVU series by means of Expand, FTP, or media such as magnetic tapes.
NOTE: This manual does not list the SPRs needed for BladeCluster Solution interoperability. For specific SPRs required for BladeCluster Solution, see the BladeCluster Solution Manual.
lists a specific SPR that must be installed on systems running G-Series RVUs if you want the functionality it makes available through Expand-connected H-series or J-series RVU systems.
Table 11 Interoperability SPR for G-Series RVU Systems Connected Through Expand to H-Series or
J-Series RVU Systems
Minimum SPR
T9205D46^ACF
Product Name
TACL
Description
Required for remote NEWPROCESS of
OCA objects*
*T9205D46^ACF is required only if you want the functionality it provides.
Introduced in this RVU
G06.26
lists the specific SPRs that must be installed on systems running H-Series RVUs for compatibility and interoperability through a ServerNet Cluster with systems running J-Series RVUs.
Table 12 Interoperability SPRs for H-Series RVU Systems Connected through ServerNet to J-Series
RVU Systems
Minimum SPR
T0294H01^AAW
Product Name
SNETMON
Description
Required for ServerNet clustering*
Introduced in this RVU
H06.14
T0502H01^AAW SANMAN Required for ServerNet clustering* H06.14
*For more details, see the ServerNet Cluster Supplement for HP Integrity NonStop BladeSystems. To find out if your
NonStop server supports ServerNet clustering, see the planning guide for your NonStop server.
lists the specific SPRs that must be installed on systems running G-Series RVUs for compatibility and interoperability through a ServerNet Cluster with systems running H-Series RVUs.
Table 13 Interoperability SPRs for G-Series RVU Systems Connected through ServerNet to H-Series
RVU Systems
Minimum SPR
T0294G08^AAP
T0502G08^AAN
T0682G06^AAA
Product Name
SNETMON
SANMAN
OSM Service
Connection Suite
Description
Required for ServerNet clustering*
Required for ServerNet clustering*
Required for ServerNet clustering*
Introduced in this RVU
G06.26
G06.26
G06.25
*For more details, see the ServerNet Cluster Supplement for NonStop NS-Series Servers. To find out if your NonStop server supports ServerNet clustering, see the planning guide for your NonStop server.
The SPRs in
facilitate software migration and software management operations between nodes running G-series RVUs and nodes running H-series and J-series RVUs. Specifically:
•
The T9621D30^ABB, T9674D30^AAM, T9276D30^AAW, and T0448G06^AAA SPRs provide the option of using a G-series RVU system to construct objects that will ultimately run on an H-series or J-series RVU system. Possible activities that can be performed in the G-series
RVU system include accelerating objects that will eventually run on an H-Series or J-series RVU
Install Node Interoperability SPRs 75
system. Once the objects have been properly constructed on the G-series RVU system, they can be moved to the H-Series or J-series RVU system by whatever means preferred, for example, by Expand, FTP, or magnetic tape. For more details, refer to the H-Series Application Migration
Guide.
•
T6031G01^ABP provides the option to use DSM/SCM to manage software in an environment composed of systems running different RVU series. The systems need not be connected through
Expand. For more details, see the NonStop Software Essentials User's Guide.
•
T9617G03^AAY allows you to check the VPROC of H-series and J-Series object files that have been copied to a G-series RVU system or in situations where a G-series RVU node is used to manage software for a H-series or J-series RVU system. This functionality can also be used to query the VPROC of a remote object file through Expand.
Table 14 SPRs for Software Migration and Software Management Operations Between G-Series,
H-Series and J-Series RVU-Based Systems
Minimum SPR
T9617G03^AAY
T9621D30^ABB
T9674D30^AAM
T9276D30^AAW
T0448G06^AAA
T6031G01^ABP
Product Name
VPROC
Binder
PDTSHARE
Axcel
OCA
DSM/SCM
Description
Required to recognize 800 files
Required to recognize OCA region
Required to recognize OCA region
Required to recognize OCA region
Introduced in this RVU
G06.27
G06.27
G06.27
G06.27
Required to prepare TNS code files to run accelerated on a NonStop NS-series platform
G06.27
Required to enable receiving H-series SUTs and installing and managing software on a NonStop NS-series target system from a
NonStop S-series host (available to be applied to the G06.26 RVU and compatible to G06.21). The functions in
T6031ABP and T9298AAM are included in the G02 DSM/SCM and SWID product version updates released in G06.27 and later G-series RVUs.
Released post G06.26
and is backwards compatible to G06.21.
Also, the G02 SPR
(installed with G06.27
and later) meets the requirements.
With SQL/MX 2.3.4 installed on any node in a mixed-node environment, accessing any MX or
MP objects on a J or H-Series “downrev node” will fail with errors 1002 and 1100/1102/1103.
If T1050H23^AMC (SQL/MX MXCMP) and T0611H23^AMC (SQL/MX MODULES) are installed over SQL/MX 2.3.4, then accessing MX and MP objects on G-Series “downrev node” will fail with error 8016. A “downrev node” is any node running SQL/MX versions 2.3.3 or lower. To prevent this, install these SPRs on all H-series RVUs less than or equal to H06.20:
•
T1050H23^AMN (SQL/MX MXCMP)
•
T0611H23^AML (SQL/MX MODULES)
NOTE: See Hotstuff HS03051 for more information.
Table 15 H-Series and J-Series RVUs Needing SPRs for Nodes to Interoperate with SQL/MX 2.3.4
(H06.20 and Later RVUs)
H-Series
H06.10.00 H06.10.01 H06.11.00 H06.11.01
H06.11.02 H06.11.03 H06.12 H06.13 H06.14
H06.15.00 H06.15.01 H06.15.02 H06.16.00
H06.16.01 H06.17.00 H06.17.01 H06.17.02
H06.18.00 H06.18.01 H06.19 H06.20
J-Series
J06.03 J06.04.00 J06.04.01 J06.04.02 J06.05.00
J06.05.01 J06.06.00 J06.06.01 J06.06.02 J06.07.00
J06.07.01 J06.08 J06.09
76 Preparing to Upgrade the Server Software (Pre-installation Tasks)
Back Up System Information
To ensure you always have a backup of your system information, run your system with a mirror of
$SYSTEM. When you are loading a new RVU, you can assure a backup of your system by shutting down half of $SYSTEM and other important disks, loading the RVU on the other $SYSTEM, verifying the RVU and then reviving the other $SYSTEM. Alternatively, or in addition to running your system with a mirror of $SYSTEM and loading the RVU as described, create a disk image of $SYSTEM and store it in a safe place so you can physically install and boot the backup disk if needed.
If Necessary, Manually Install the NonStop Software Essentials Server
Product
HP recommends use of NonStop Software Essentials instead of the Planner Interface for upgrading your RVU. In addition to NonStop Software Essentials being the go-forward strategy, the Planner
Interface is not supported on 64-bit NonStop system consoles.
Systems running H06.24 or a later H-series RVU will already have the NonStop Software Essentials server component (T8094H02) installed.
If the NonStop Software Essentials server is not yet installed on your system or you want to upgrade from T8094H01 to T8094H02, you can install the latest server via SPR from Scout or use your installed NonStop Software Essentials software or the DSM/SCM Planner Interface to install the
T0894H02 server.
The NonStop Software Essentials Installation and Quick Start Guide identifies the circumstances under which you might need to manually install the T0894H02 NonStop Software Essentials server and related components and how to perform the manual installation.
NOTE: You cannot use or run DSM/SCM with the Planner Interface to install the NonStop Software
Essentials server on systems with 64-bit NonStop system consoles. If you have a 64-bit NonStop system consoles only but do not have the NonStop Software Essentials server installed, you must manually install the server and the other NonStop Server Essentials software.
See the NonStop Software Essentials Installation and Quick Start Guide for the NonStop Software
Essentials server installation procedures, including the manual installation procedure.
Back Up System Information 77
6 Installing a SUT for This RVU Using NonStop Software
Essentials
NOTE: If you do not have NonStop Software Essentials installed, see the NonStop Software
Essentials Installation and Quick Start Guide.
Installing a new RVU or a SUT involves these DSM/SCM tasks:
1.
“Receive New Software Into the Archive” (page 91)
2.
“Create a New Software Revision” (page 81)
3.
“Build and Apply the New Software Revision” (page 84)
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 153)
to Install SPRs.
Before you install a SUT or RVU, these requirements must be met:
•
When managing OSS files with DSM/SCM, check that the OSS Root Name Server $ZPNS is started and running on the DSM/SCM Target. (See
“Create a New Software Revision”
.) The volume where the SYSnn is being placed must be activated as $SYSTEM to install the OSS files. See Appendix B in the Open System Services Management and Operations
Guide, or if using EasySetup, see Chapter 1 in 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 68)
.
•
If necessary, update the system console.
•
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.
Back Up Important Files
•
Back up important files (such as, TMF, NonStop SQL/MP, NonStop SQL/MX, and so forth).
See
“Back Up Important Files” (page 66)
.
•
Drain the Spooler and back up spooler control files.
78 Installing a SUT for This RVU Using NonStop Software Essentials
Receive New Software Into the Archive
1.
Start TMF and DSM/SCM if they are not already running. For example:
> 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 APR 2005, 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.
Start HP SIM and then open a NonStop Software Essentials window by selecting the Manage
NonStop Software link on the NonStop server Tools & Links tab.
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 NonStop Software Essentials home page.
The Receive Software Request page appears. For example:
5.
In the Receive options group box, do one of the following, depending on your system environment:
•
Select Receive only changed files (default).
•
Select Receive all files.
Receive New Software Into the Archive 79
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.
11.
Click Submit. The request status page appears and is updated automatically.
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:
80 Installing a SUT for This RVU Using NonStop Software Essentials
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 the NonStop Software Essentials page open for the next task.
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.
Check that the OSS Root Name Server $ZPNS is running on the Target. The volume where the SYSnn is being placed must be activated as $SYSTEM to install OSS files. To determine the server process, issue an SCF STATUS SERVER command or an SCF LISTDEV command.
An example of the STATUS SERVER command is:
$ADBIT TESTBIT 11> SCF STATUS SERVER $ZPMON.*
SCF - T9082H01 - (01MAY05) (15MAR05) - 07/06/2005 08:58:59 System \ATLAS
(C) 1986 Tandem (C) 2003 Hewlett Packard Development Company, L.P.
OSS Status SERVER \ATLAS.$ZPMON.*
ServerName State LastError ErrorDetail
#ZMSGQ STARTED 0 0
#ZPLS STARTED 0 0
#ZPNS
STARTED 0 0
#ZTA00 STARTED 0 0
#ZTA01 STARTED 0 0
#ZTA02 STARTED 0 0
#ZTA03 STARTED 0 0
#ZTA04 STOPPED 0 0
#ZTA05 STOPPED 0 0
#ZTA06 STOPPED 0 0
#ZTA07 STOPPED 0 0
#ZTA08 STOPPED 0 0
#ZTA09 STOPPED 0 0
#ZTA10 STOPPED 0 0
#ZTA11 STOPPED 0 0
#ZTA12 STOPPED 0 0
#ZTA13 STOPPED 0 0
#ZTA14 STOPPED 0 0
#ZTA15 STOPPED 0 0
Total Errors = 0 Total Warnings = 0
$ADBIT TESTBIT 12>
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.
Create a New Software Revision 81
b.
Click the Output tab to view the default output values.
c.
Enter the appropriate values in the data fields.
NOTE: On H-series systems, the alternate name for the $SYSTEM drive now 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.
4.
Click OK to return to the Configuration Revisions page.
5.
Click the Create or Edit button 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 H06.25 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:
82 Installing a SUT for This RVU Using NonStop Software Essentials
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 from the system by DSM/SCM during the Build/Apply of the new configuration.)
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:
•
Select the products and click Continue.
•
To include all products, click the checkbox in the left column header to select all the products.
The New software revision list now displays the new revision level. For example:
Create a New Software Revision 83
7.
Keep the NonStop Software Essentials page 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.
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:
84 Installing a SUT for This RVU Using NonStop Software Essentials
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 85
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 FLAGS CPU PRI UNIT DATA FILE %FULL
$S ACTIVE 3 , 1 154 4 $OPS0.SPL.DATA 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:
86 Installing a SUT for This RVU Using NonStop Software Essentials
•
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.
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
==============================================================================
\WARRIOR Prior snapshot 2002-11-09:12:18:44
Print Operator Instructions From the Target Interface (ZPHITI) 87
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 2003-09-20:12:18:44
Selection number:
==============================================================================
Choose a target selection number from the list below. Status information is
shown for requests still in progress.
==============================================================================
1 NEW CONFIG
2 DEFAULT
3
4
5
6
7
8
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 Prior snapshot 2003-11-20:12:18:44
DEFAULT Last S/W applied on SYS 00 on 2002-11-22:12:17
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 2000-09-13:15:44
3 Review operator instructions for the last apply completed.
==============================================================================
88 Installing a SUT for This RVU Using NonStop Software Essentials
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 HSS - Halted State Services includes firmware changes.
T0630 FCSA F/W includes firmware changes.
T0287 PRM - Primitive State includes firmware changes.
T0287 IPF - Itanium(tm) Processor Firmware 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 reinitiate 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.
Recovery for Problems Installing This RVU SUT or SPRs 89
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.
90 Installing a SUT for This RVU Using NonStop Software Essentials
7 Installing a SUT for This RVU Using the Planner Interface
Installing a new RVU or a SUT involves these DSM/SCM tasks:
1.
“Receive New Software Into the Archive” (page 79)
2.
“Create a New Software Revision” (page 95)
3.
“Build and Apply the New Software Revision” (page 98)
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 the Planner Interface to Install SPRs” (page 160)
to install SPRs.
Before you install a SUT or RVU, these requirements must be met:
•
When managing OSS files with DSM/SCM, check that the OSS Root Name Server $ZPNS is started and running on the DSM/SCM Target. (See
“Create a New Software Revision”
.) The volume where the SYSnn is being placed must be activated as $SYSTEM to install the OSS files. See Appendix B in the Open System Services Management and Operations
Guide, or if using EasySetup, see Chapter 1 in 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 68)
.
•
If necessary, update the system console.
•
Obtain a user name and password authorized to use the DSM/SCM Planner Interface. See the DSM/SCM 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 DSM/SCM.
You can add a Safeguard alias anytime Safeguard is running.
Before using an existing alias for DSM/SCM, 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 DSM/SCM 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.
Back Up Important Files
•
Back up important files (such as, TMF, NonStop SQL/MP, NonStop SQL/MX, and so forth).
See
“Back Up Important Files” (page 66)
.
•
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:
> TMFCOM START TMF
> VOLUME $dsmscm_vol.ZDSMSCM
> RUN STARTSCM
Alerts for Installing a SUT 91
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 APR 2005, 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, STARTEDStart the DSM/SCM Planner Interface
2.
Select Start>Programs>dsm_scm>DSMSCM.
3.
When the Logon dialog box appears, enter the user ID or alias and password and click OK.
The Planner Interface screen displays two dialog boxes:
•
The Software in Archive dialog box lists the software that is currently in the DSM/SCM
Archive.
•
The Configuration Revisions dialog box lists the current software revisions.
4.
In the Software in Archive dialog box, click Receive software.
The Receive Software Request dialog box appears. For example:
5.
In the Receive Options group box, do one of the following, depending on your system environment:
•
Select Receive only changed files (default)
•
Select Receive all files (recommended for the new RVU)
92 Installing a SUT for This RVU Using the Planner Interface
6.
Under Software name, select Provided by planner and enter a name for the RVU.
7.
In the Options group box, click Input source.
The Request Source Options dialog box appears.
8.
Select Tape drive.
9.
From the list, select the tape drive that contains the SUT, and click OK.
The Request Source Options dialog box closes.
10.
In the Options group box of the Receive Software Request dialog box, click Scheduling.
The Request Scheduling Options dialog box appears. Make the appropriate selections and click OK.
The Request Scheduling Options dialog box closes.
11.
In the Receive Software Request dialog box, click OK.
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:
Receive New Software Into the Archive 93
•
EMS event messages to see if a cartridge on a nonautoloading drive of the SUT needs to be mounted. EMS messages are displayed in these windows:
◦
The EMS Event Stream window
◦
The EMS Events window displayed by DSM/SCM when you click Show EMS events from the Status window
◦
You can also check the Requests window to ensure that the request is finished. (Select
Requests>Requests list.) For example:
94 Installing a SUT for This RVU Using the Planner Interface
13.
When the request is finished, close the Status and Requests windows.
14.
Update the Software in Archive dialog box to show the newly received SUT. Select
Window>Refresh.
The new SUT appears in the window.
15.
Keep the Planner Interface open for the next task.
Create a New Software Revision
When you create a new software revision, DSM/SCM 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 are in the Planner Interface Configuration Revisions window.
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 DSM/SCM User’s Guide.
1.
Check that the OSS Root Name Server $ZPNS is running on the Target. The volume where the SYSnn is being placed must be activated as $SYSTEM to install OSS files. To determine the server process, issue an SCF STATUS SERVER command or an SCF LISTDEV command.
An example of the STATUS SERVER command is:
$ADBIT TESTBIT 11> SCF STATUS SERVER $ZPMON.*
SCF - T9082H01 - (01MAY05) (15MAR05) - 07/06/2005 08:58:59 System \ATLAS
(C) 1986 Tandem (C) 2003 Hewlett Packard Development Company, L.P.
OSS Status SERVER \ATLAS.$ZPMON.*
ServerName State LastError ErrorDetail
#ZMSGQ STARTED 0 0
#ZPLS STARTED 0 0
#ZPNS
STARTED 0 0
#ZTA00 STARTED 0 0
#ZTA01 STARTED 0 0
#ZTA02 STARTED 0 0
#ZTA03 STARTED 0 0
#ZTA04 STOPPED 0 0
#ZTA05 STOPPED 0 0
#ZTA06 STOPPED 0 0
#ZTA07 STOPPED 0 0
#ZTA08 STOPPED 0 0
#ZTA09 STOPPED 0 0
#ZTA10 STOPPED 0 0
#ZTA11 STOPPED 0 0
#ZTA12 STOPPED 0 0
#ZTA13 STOPPED 0 0
#ZTA14 STOPPED 0 0
#ZTA15 STOPPED 0 0
Total Errors = 0 Total Warnings = 0
$ADBIT TESTBIT 12>
1.
In the DSM/SCM Planner Interface, in the Configuration Revisions dialog box, select the target configuration.
2.
Select Maintain>Target Maintenance>Modify.
a.
Check that the Manage OSS Files option is selected.
Create a New Software Revision 95
b.
Click Output to view the default output values.
c.
Enter the appropriate values in the data fields.
NOTE: On H-series systems, the alternate name for the $SYSTEM drive now defaults to
$FC1 for fibre channel drives or $SCSI for SCSI drives. If this default has been changed, then the target volume specified in the Planner Interface for new subvolumes must reflect this change.
d.
Click OK. DSM/SCM returns you to the Target Information dialog box.
3.
Click OK to return to the Configuration Revisions dialog box.
4.
Click OK to return to the Planner Interface.
5.
Select New software revision.
DSM/SCM opens a New software revisions dialog box that lists the products in the last software revision that was built for the target system. It might take several minutes for the dialog box to appear.
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 H06.25 SUT) and click
OK.
d.
In the Software in Archive window, select the H06.25 SUT that was just received into the archive and click Copy to.
96 Installing a SUT for This RVU Using the Planner Interface
6.
Click Continue when the Confirmation message appears:
Note that if there are any discrepancies between the set of products included in the previous configuration and the set in your new configuration, a warning message appears listing any products missing in the current configuration that were present in the previous configuration.
(These products are removed from the system by DSM/SCM during the Build/Apply of the new configuration.)
You might receive a warning that certain OSS products are not DSM/SCM enabled. If this is the case, install them using PINSTALL/COPYOSS. See the appropriate OSS manuals for using
PINSTALL and COPYOSS.
The Non SUT Products screen might appear:
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:
•
Select the products and click Continue.
•
To include all products, click Select All and click Continue.
The New software revision list now displays the new revision level (H06.nn). For example:
Create a New Software Revision 97
7.
Keep the Planner Interface 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.
To avoid disrupting other system activity, schedule the Build to run at night and the Apply to run immediately after the Build. See the DSM/SCM 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 dialog box, 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 Build/Apply Request dialog box options and commands are dimmed and unavailable.
When the Build information is assembled, the Build/Apply Request dialog box is refreshed.
It displays default options and assigns a name to the Build/Apply request. For example:
98 Installing a SUT for This RVU Using the Planner Interface
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.
Click Edit instructions to display the Planner Instructions dialog box.
6.
Edit the instructions to provide the correct (new) SYSnn number, and then click OK.
The Planner Instructions box closes.
7.
Select the Scheduling, Output, SYSGEN, TSV Location, and File Placement options and configure as appropriate for your system. (See the DSM/SCM User’s Guide or the online help for more information about Build/Apply Request options.) When completed, click OK.
Build and Apply the New Software Revision 99
8.
(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 FLAGS CPU PRI UNIT DATA FILE %FULL
$S ACTIVE 3 , 1 154 4 $OPS0.SPL.DATA 4
9.
In the Build/Apply Request dialog box, click OK.
100 Installing a SUT for This RVU Using the Planner Interface
DSM/SCM runs the Build and Apply steps and displays a Status window, which is automatically updated at regular intervals. For example:
10.
Check EMS messages for more detailed information about your Build/Apply: a.
In the Events window, click Refresh to see newer messages.
b.
When finished viewing, click Close.
NOTE: If you encounter any problems, see the DSM/SCM User’s Guide or online help.
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:
Build and Apply the New Software Revision 101
•
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> Requests list).
11.
Close the EMS Events, Status, and Request windows.
12.
Exit the DSM/SCM Planner Interface.
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.
T6031H01 11NOV2004
(c) Hewlett Packard Development Company L.P., 1994-2004
102 Installing a SUT for This RVU Using the Planner Interface
All Rights Reserved
==============================================================================
F1-Logon F8-Clear entry
F15-Help SF16-Exit
Ready
2.
Enter your user ID or alias (if necessary) 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 2002-11-09:12:18:44
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 2003-09-20:12:18:44
Selection number:
==============================================================================
Choose a target selection number from the list below. Status information is
shown for requests still in progress.
==============================================================================
1 NEW CONFIG
2 DEFAULT
3
4
5
6
7
8
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.
Print Operator Instructions From the Target Interface (ZPHITI) 103
DSM Software Configuration Manager Target Menu - Select Action
==============================================================================
\WARRIOR Prior snapshot 2003-11-20:12:18:44
DEFAULT Last S/W applied on SYS 00 on 2002-11-22:12:17
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 2000-09-13:15:44
3 Review operator instructions for the last apply completed.
==============================================================================
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 SYSnn
(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 HSS - Halted State Services includes firmware changes.
T0630 FCSA F/W includes firmware changes.
T0287 PRM - Primitive State includes firmware changes.
T0287 IPF - Itanium(tm) Processor Firmware 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.
104 Installing a SUT for This RVU Using the Planner Interface
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 reinitiate 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 close the Status dialog box, reopen it by selecting Requests> Requests list and then double-clicking the request.
1.
Select the request to be canceled.
2.
Select Requests>Cancel request.
The cancel request action takes a few minutes to finish.
NOTE: If you cannot cancel the Build request through the DSM/SCM Planner Interface, use the DSM/SCM Target Interface (ZPHITI) to cancel the request. See the DSM/SCM User’s
Guide and online help.
Recovery for Problems Installing This RVU SUT or SPRs 105
8 Running ZPHIRNM
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 HP 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, except for the TMF subsystem, before running
ZPHIRNM.
•
You must have rename (purge) access to all files being renamed. Run ZPHIRNM as the super
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 to this RVU, install DSMSCM SPR T6031H02^ACJ or later. See
“DSM/SCM Upgrade Issues/Interoperability SPR Requirements” (page 58)
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.
106 Running ZPHIRNM
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 BladeCluster Solution Manual.
3.
Stop all other applications, except for TMF and the OSS file system.
The planning guides for NonStop systems contain detailed information on stopping applications, devices, and processes in an orderly fashion.
NOTE: Your site might have macros or OSS shell scripts set up to start and stop applications, devices, and processes.
Stop Applications 107
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 $collector-process, LISTOPENS command to identify processes that currently have jobs open.
2.
Use the SPOOLCOM;JOB (STATE PRINT) command to determine which jobs are currently printing.
Take the appropriate action to ensure that these jobs either terminate when finished or are stopped.
After receiving the SPOOLER DRAIN command, the supervisor stops accepting jobs for collection or printing (new opens are rejected with a file-system error 66—device downed). Each collector stops when it has no more open jobs. Each print process finishes printing active jobs and then stops.
After all collectors and print processes have stopped, the supervisor process stops itself. The Spooler enters the dormant state, ready to be warm started.
CAUTION: Never use the TACL STOP command to halt the Spooler. The control file can be corrupted during the next warm start or rebuild.
For more information on draining the Spooler, see Support Note S99086, NSK Spooler—Migration
and Draining Instructions and Chapter 1 and Appendix H in the Spooler Utilities Reference Manual.
Back Up the Spooler Control Files
While the spooler environment is closed, back up the spooler control files to render a known database from which you recover in case of a catastrophic spooler failure. Make sure that you drain the Spooler:
BACKUP/OUT SPLBKUP/$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 for H06.03:
Statistics:
TSVs:
TSVs in new configuration .................. 129
108 Running ZPHIRNM
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 ........................... 4038
Files processed in new configuration only ...................... 0
Files processed in old configuration only ...................... 9
Files processed that are shared by new and old configuration ... 4038
Files processed & unchanged between new and old configuration .. 4038
Files processed in obsolete TSVs ............................... 14
Files renamed from real name to fabricated name ................ 23
Files renamed from fabricated name to real name ................ 0
Files renamed from real name to fabricated (ZPG) name .......... 0
Files renamed from fabricated name to fabricated (ZPG) name .... 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 ....... 1/29/05 16:43:20
End ......... 1/29/05 16:43:37
Elapsed ..... 0:00:16
Errors & Warnings:
Total Rename Errors ..... 0
Total Rename Warnings ... 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 or DSM/SCM User's Guide.
Steps for Running ZPHIRNM
1.
At a TACL prompt, enter this command from the DSM/SCM subvolume:
> RUN ZPHIRNM $vol.SYSnn
ZPHIRNM 109
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 - (11NOV2004)
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 NonStop Software Essentials User's Guide or the DSM/SCM User's Guide for troubleshooting procedures and the DSM/SCM Messages Manual for a complete list of ZPHIRNM messages.
110 Running ZPHIRNM
9 Loading This RVU
This chapter describes how to perform a system load of this RVU into the memory of each processor in a NonStop system.
NOTE: For detailed information on performing a system load, see in the NonStop Operations
Guide.
By default, one $SYSTEM is already configured. You can select $SYSTEM, FCDM-Load,
CLIM-Attached-Load for systems that support this type of load), or an alternate system volume:
The $SYSTEM disks must contain the current RVU operating system files for the NonStop system.
and
identifies the possible load paths to use for an FCDM-Load and a
CLIM-Attached-Load, respectively:
Table 16 System Load Paths in Order of Use for FCDM-Load
9
10
11
12
5
6
7
8
13
14
15
16
2
3
4
Load Path
1
Description
Primary
Primary
Backup
Backup
Mirror
Mirror
Mirror backup
Mirror backup
Primary
Primary
Backup
Backup
Mirror
Mirror
Mirror backup
Mirror backup
Data Travels From
$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
0
0
0
0
1
1
1
1
0
0
0
To Processor
0
X
Y
X
Y
X
Y
X
Y
X
Y
X
Y
Y
X
Y
Over ServerNet Fabric
X
6
7
8
2
3
4
5
Load Path
1
Table 17 System Load Paths in Order of Use for CLIM-Attached-Load
Description
Primary
Primary
Backup
Backup
Mirror
Mirror
Mirror backup
Mirror backup
Data Travels From
$SYSTEM-P
$SYSTEM-P
$SYSTEM-B
$SYSTEM-B
$SYSTEM-M
$SYSTEM-M
$SYSTEM-MB
$SYSTEM-MB
0
0
0
0
0
0
0
To Processor
0
Y
Y
X
Y
Y
X
X
Over ServerNet Fabric
X
111
12
13
14
15
16
Load Path
9
10
11
Table 17 System Load Paths in Order of Use for CLIM-Attached-Load (continued)
Description
Primary
Primary
Backup
Backup
Mirror
Mirror
Mirror backup
Mirror backup
Data Travels From
$SYSTEM-P
$SYSTEM-P
$SYSTEM-B
$SYSTEM-B
$SYSTEM-M
$SYSTEM-M
$SYSTEM-MB
$SYSTEM-MB
1
1
1
1
1
1
1
To Processor
1
Y
Y
X
X
Y
X
X
Over ServerNet Fabric
Y
A CLIM-Attached-Load attempts to load the system from a system disk in the NonStop Storage
CLIMs connected to a ServerNet switch.
CLIM-Attached-Load for system with 2 CLIMs over $SYSTEM:
Path
Primary
Backup
Mirror
Group
100
100
100
Mirror Backup 100
2
2
Module
2
2
5
5
Slot
5
5
3
3
3
ServerNet Switch
Connector
3
Fiber
1
3
3
1
CLIM-Attached-Load for System with 4 CLIMs over $SYSTEM:
Primary LUN
101
101
N/A
N/A
Mirror LUN
N/A
N/A
201
201
Path
Primary
Backup
Mirror
Group
100
100
100
Mirror Backup 100
2
2
Module
2
2
5
5
Slot
5
5
4
4
3
ServerNet Switch
Connector
3
Fiber
1
1
3
3
Primary LUN
101
101
N/A
N/A
Mirror LUN
N/A
N/A
101
101
An FCDM-Load attempts to load the system from a system disk in the disk drive enclosure connected to IOAM enclosure group 110:
IOAM
Path
Primary
Backup
Group
110
110
Mirror
Mirror Backup
110
110
FCSA
3
2
Module
2
3
1
1
1
1
Disk Drive
Enclosure
Slot
2
2
SAC
1
1
1
1
Shelf
1
1
1
1
Bay
1
1
112 Loading This RVU
You can configure additional alternate system disks. To create an alternate system disk, see the hardware installation manual for your system. Then use OSM to make the disk available in the
Configuration drop-down menu in the System Load dialog box. The alternative system disks are displayed in the Attributes tab under the System object in the OSM Service Connection and are also displayed in the OSM Low-Level Link System Load dialog box in the Configuration shortcut menu.
Alerts
Before performing a system load:
•
Before halting the processors, stop applications that are still running, including TMF.
NOTE: To stop TMF, at a TACL prompt, enter:
TMFCOM STOP TMF
The NonStop Operations Guide contains detailed information on the best methods for stopping applications, devices, and processes.
•
Ensure that sufficient swap file space is available for the installation. Either use NSKCOM to ensure that there is at least one KMSF swapfile on $SYSTEM for each of CPU 0 and CPU 1, or ensure that there is more than 40MB of free space on $SYSTEM.
There are several ways to create additional free space on $SYSTEM, including:
◦ delete all existing zlog*.* files
◦ delete all existing *.zzsa* files
•
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 the NonStop Operations Guide.
1. Before Halting the Processors
A. If Falling Back to H06.25 or an Earlier RVU
Multiple Providers per CLIM functionality, which is first implemented with the H06.26 RVU, includes allowing a single CLIM to be associated with multiple data providers if the MULTIPROV attribute of the CLIM object is set to ON (enabled).
NOTE: If you have the H06.19 through H06.25 RVU and are not planning to migrate to the
H06.26 RVU, you can implement full Multiple Providers per CLIM functionality by acquiring the
T0853H01^ACF SPR from Scout and installing it.
If you need to fall back from H06.26 to H06.25 or an earlier H-series RVU that does not already have the T0853H01^ACF SPR installed, you must disable the MULTIPROV attribute to instruct
CLIMs to follow pre-Multiple Providers per CLIM behavior and use the sole provider specified by
Alerts 113
the PROVIDER attribute. For details, see the Cluster I/O Protocols (CIP) Configuration and
Management Manual.
B. When Upgrading From an H06.15 or Earlier RVU: Modify OSSINFIL
When upgrading from an H06.15 or earlier RVU, before halting the processors, you must remove the SERVER #ZPLS entry from the OSS subsystem maintenance file OSSINFIL on all systems with
OSS installed. The simplest way to do this is to use the TEDIT utility to change the SERVER #ZPLS entry to a comment, as follows:
TEDIT $SYSTEM.ZOSSINS.OSSINFIL
Then change
SERVER #ZPLS to
* SERVER #ZPLS
Failure to do this can result in an error during startup of OSS. This error occurs on systems with pre-T0585H01.
C. When Upgrading From the H06.03 or H06.04 RVU
If you are upgrading from the H06.03 or H06.04 RVU, then before halting the processors, update the shadow firmware on each NonStop Blade Element, as described in step 6a under
Blade Complex Firmware Update for Migrations From H06.03 or H06.04 to This RVU” (page 29)
.
D. When Upgrading From the H06.16 RVU
CLIMs configured with the CLIM software released with H06.16 are not compatible with H06.17
and later RVUs. Performing the CLIM upgrade when the processors are halted, as shown in this procedure, ensures the CLIM and host system will be compatible when you restart the system. For migrations from H06.16, do not perform an online upgrade to the CLIMs before the RVU upgrade.
Preliminary online CLIM upgrades are only supported for H06.17 and later H-series RVUs.
2. Halt the Processors
1.
Log on to the OSM Low-Level Link.
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 Halt.
5.
Click Perform action.
6.
Click OK when asked “Are you sure you want to do Halts?”
7.
When the Action Status box confirms that all processors have halted, close the Processor Status dialog box.
3. Power Cycle the Processors if Necessary
If you are upgrading from the H06.03 or H06.04 RVU, while the processors are halted, power cycle the processors as described in step 6b under
“NonStop Blade Complex Firmware Update for Migrations From H06.03 or H06.04 to This RVU” (page 29)
.
4. (Required for Upgrade to H06.25 or Later) Perform a Hard Reset
If you did not already perform a hard reset in
“3. Power Cycle the Processors if Necessary”
, you should now perform a hard reset on all processor blades, but only if you are migrating from H06.24 or an earlier H-series RVU to H06.25 or a later H-series RVU.
NOTE: You do not need to perform this operation again during subsequent system loads, provided the previously executing RVU was at least H06.25.
114 Loading This RVU
1.
Log on to the OSM Low-Level Link.
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.
5.
Click Perform action.
6.
Click OK when asked “Do you really want to perform a Hard Reset on the selected processor(s)?”
7.
When the Action Status box confirms that all processors have been reset, close the Processor
Status dialog box.
5. (Optional) Upgrade the CLIM Software
If you have not upgraded the CLIMs online before performing this RVU upgrade, upgrade the CLIM software on each CLIM now. See the instructions for upgrading CLIM software in the Cluster I/O
Protocols (CIP) Configuration and Management Manual.
6. Loading the System
If the automatic system load is not successful using one load path, the system load task attempts to use another path and keeps trying until all possible paths have been used or the system load is successful. Eight paths are available for loading.
The command interpreter input (CIIN) file is automatically invoked after the first processor is loaded.
The CIIN file contains the TACL RELOAD * command, which loads the remaining processors.
For details on load paths, see the hardware installation manual for your NonStop system.
Steps for Loading the System
A normal system load consists of loading the operating system from disk into the memory of processor 0 or processor 1 (this assumes the default location is the $SYSTEM volume):
1.
In the OSM Low-Level Link, select Start System. The System Load dialog box appears.
2.
a.
From the Configuration drop-down menu under System Load Configuration, select a system load volume. You can select $SYSTEM, FCDM-Load, SCSI-Load,
CLIM-Attached-Load, or an alternate system volume.
b.
If you are using non-default locations for the $SYSTEM volume:
•
For FCDM-Load, set the controller location and disk partition location using the Disk
Configuration dialog boxes in the OSM Low-Level Link.
•
For CLIM-Attached-Load, change the CLIM name, CLIM location, and disk location using the OSM Low-Level Link System Load dialog box and CLIM Attached Disk
Configuration dialog boxes as needed.
3.
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), the default configuration file that represents the configuration currently running.
b.
Under SYSnn and CIIN Option, verify that the CIIN Disabled option is not selected.
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.
For a description of the CIIN file, see the appropriate planning guide for your system.
4.
Under Configuration, ensure that all the check boxes for all paths are selected.
5. (Optional) Upgrade the CLIM Software 115
5.
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.
The Details dialog box displays Logon successful and Read configurations passed
(FCDM example):
This example shows the same dialog box for a CLIM-attached load:
116 Loading This RVU
One window opens with one CNSL and one CLCI session in it.
NOTE: If you are using OSM Low-Level Link version T0633G07^AAT or later, all four windows are contained within a single MR-Win6530 session window (as opposed to the four separate OutsideView sessions launched by earlier versions of the Low-Level Link.) Also,
MR-Win6530 hides the inactive Startup TACL and Event Stream windows, displaying only the active ones.
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
OSM Low-Level Link, reboot the system console, and restart the system load using the OSM
Low-Level Link again.
6. Loading the System 117
6.
Monitor the system startup process. Messages indicating the progress and completion of the system load, as well as 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
7.
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.
Completing the System Load (If Necessary)
Loading the processors with the CIIN file is usually automatic. However, 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 RELOAD TACL commands. You must be super ID (255,255) to load the processors.
1.
From the system console, log on to the system as a super-group user (255,nnn) and enter:
> RELOAD *, PRIME
2.
To load an individual processor or processors, enter:
> 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
For complete syntax, considerations, and examples of the TACL RELOAD command, see the
TACL Reference Manual.
After Coldloading the New RVU
If you are upgrading from the H06.03 or H06.04 RVU, after loading the new RVU, finish loading the rest of the NonStop Blade Complex firmware, as described in step 6c under
Complex Firmware Update for Migrations From H06.03 or H06.04 to This RVU” (page 29) .
After the System Load
After performing the system load as described in
“6. Loading the System” (page 115) , perform steps
6 through 9 from the procedure for performing an RVU upgrade on a node in the NonStop
BladeCluster Solution Manual.
Verifying That the System Is Started
Verify that each processor is running the operating system:
1.
Select Summary>Processor Status.
The Processor Status dialog box should show each processor to be Executing NonStop OS.
118 Loading This RVU
2.
Close the Processor Status dialog box.
3.
Verify that the NonStop operating system is working properly: a.
Check the startup TACL window to verify that the startup scripts completed successfully.
b.
Check the startup event stream window and the startup TACL window for error messages.
4.
To determine if at least one of the maintenance Providers is working: a.
From a TACL prompt, enter the command:
CLIMCMD clim-name hostname b.
The command can timeout and fail if it is issued while the SSH and TCP/IP processes are still coming up after the system load. Check the returned clim-name and Termination
Info value to confirm that the command was successful. If the returned clim-name does not match the name that was entered in the command and/or the Termination Info value is not 0, wait for five minutes and retry the command. If the returned values are still not correct, wait another five minutes and retry the command a third time. If the command is still not successful, see the Cluster I/O Protocols (CIP) Configuration and Management
Manual for troubleshooting tips about identifying an incompatibility in CLIM and host CIP software
5.
Close the System Startup dialog box.
For more information on system loads, the NonStop Operations Guide.
7. (Optional) Updating the CLIM Firmware
If necessary, (if you did not update the CLIM firmware while the system was down), update the
CLIM firmware using the OSM System Console as described under
. See the NonStop Firmware Matrices for firmware levels.
7. (Optional) Updating the CLIM Firmware 119
10 Resuming Normal Operations
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.
Start User and 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 next.
Before Starting DSM/SCM
Initialize SQL for NonStop SQL/MP
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.
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
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.
HP 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 the DSM/SCM User's Guide.
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
SQL/MX Installation and Management Guide.
120 Resuming Normal Operations
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.
Run merge_whatis
After coldloading the system, 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 either one or both of 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 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 cat7 whatis.frag
cat2 cat5 cat8 whatis_old
Run merge_whatis 121
cat3 cat6 whatis
/usr/share/man:
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.
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 106)
.
•
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
122 Resuming Normal Operations
>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
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 1 ICMP internet control message protocol ggp 3 GGP gateway-gateway protocol tcp 6 TCP transmission control protocol pup 12 PUP PARC universal packet protocol udp 17 UDP user datagram protocol ipv6 41 IP internet protocol for v6 icmpv6 58 ICMP internet control message protocol for v6
Step 3: Ensure That PORTCONF and SERVICES Files Are Complete
The PORTCONF file specifies the ports that the LISTNER process listens to and the corresponding server program it invokes when the request comes in.
If you copied files as documented in Step 1, for the DSM/SCM Planner Interface to work, you must add a services line corresponding to DSM/SCM in the PORTCONF file (shown in bold lettering) and save the file. If you did not copy the files as documented in Step 1, no action is necessary.
An example of the PORTCONF file:
58> type portconf
#
# This file tells the listner program which ports to
# listen to, and what programs to run
# Telnet is directly, and does not use the listner's
# services.
# To run the listner use:
# $system.ztcpip.listner / name.../ [config-file-name]
# where config-file-name is this file.
# ftp $system.ztcpip.ftpserv
finger $system.ztcpip.fingserv
7 $system.ztcpip.echoserv
phiwsock $dsmscm.zdsmscm.pdpexe
Configure Specific Products (Post-System Load) 123
Applications use the SERVICES file to retrieve the service port numbers and service names. If you copied files as documented in Step 1, check that the last lines of the SERVICES file have an entry similar to: phiwsock 1089/tcp #DSM/SCM Planner Interface
For DSM/SCM, the port number is user-definable. The setup default is 1089. In the ZDSMSCM subvolume, check the INITSTAT file for the recently configured TCP/IP port number. INITSTAT is created by DSM/SCM when INITENV is run.
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 H-series RVU and the 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 managed by DSM/SCM, use PINSTALL on the remaining pax archives.
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.
•
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. Then 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.
NOTE: Features introduced in H06.23 affect the default security settings in OSS. These changes impact where you install OSS packages into your system. See Support Note S12018 for detailed information.
See also the appropriate product’s softdoc for any pre- or post-installation instructions.
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.
124 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
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 OSS EasySetup product (T0585) is available for automated setup, 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 or when starting, stopping, or removing an OSS subsystem.
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 3.x 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:
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 (RDF)
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 for this RVU.
See the Release Version Update Compendium for your H-series RVU for new feature information.
Configure the WAN Manager
If you have a SWAN 1 or 2 concentrator:
Configure Specific Products (Post-System Load) 125
1.
Configure the WAN manager as described in the WAN Subsystem Configuration and
Management Manual.
2.
Update the SWAN CLIP firmware. See either the SWAN Concentrator Installation and Support
Guide or the SWAN 2 Concentrator Installation and Support Guide.
Re-configure iTP Webserver
If iTP WebServer 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 re-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.
126 Resuming Normal Operations
11 Backing Out the DSM/SCM Revision
This chapter describes:
•
Products that might have fallback implications. Before falling back or backing out the revision, review the appropriate product softdocs.
•
Using DSM/SCM to back out the revision.
NOTE:
•
You cannot revert a NonStop NS16200 system to the H06.11.03 RVU or to an RVU prior to
H06.11.01.
•
You cannot revert a NonStop NS1200 system to an RVU prior to H06.12.
•
You cannot revert a NonStop NS14200 system to an RVU prior to H06.13.
Products That Might Have Fallback Implications
Review the H06.nn Release Version Update Compendium for this RVU and product softdocs to make sure that your installed products do not have migration implications.
Open System Services (OSS)
•
Back up the OSS configuration database files before falling back. For more information, see the Open System Services Management and Operations Guide.
•
OSS Large Files support, introduced in the H06.07 RVU, has created fallback issues. For more
information, see “OSS Large Files” (page 43)
.
Storage Management Foundation (SMF)
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.
Safeguard
The following Safeguard features have fallback considerations.
Password Encryption
A new password encryption algorithm was introduced in H06.06. This change has fallback implications when falling back to a pre-H06.06 RVU, because new passwords will not be valid on the RVU to which you are falling back. The new passwords must be expired and users must revert to their previous passwords. For a detailed fallback procedure, see
.
Products That Might Have Fallback Implications 127
Longer Passwords
Starting with the version of Safeguard released with the H06.08 RVU, the maximum allowed password length is increased to 64 characters. This change has fallback implications when falling back to pre-H06.08 RVU, as users passwords longer than 8 characters will be unable to logon.
See
“Safeguard Longer Password Length” (page 49)
.
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 FLAGS CPU PRI UNIT DATA FILE %FULL
$S ACTIVE 3 , 1 154 4 $OPS0.SPL.DATA 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)
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:
128 Backing Out the DSM/SCM Revision
DSM Software Configuration Manager Main Menu
==============================================================================
\ELI Prior snapshot 2002-04-14:18:03:03
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 2002-04-14:18:03:03
Selection number:
3
4
5
6
7
8
==============================================================================
Choose a target selection number from the list below. Status information is
shown for requests still in progress.
==============================================================================
1 NEW CONFIG
2 DEFAULT
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 Prior snapshot 2003-04-14:18:03:03
DEFAULT Last S/W applied on SYS 00 on 2000-04-14:18:01
Selection number:
==============================================================================
Choose an action by entering the selection number of an item from the list
below.
==============================================================================
Submit a Backout Request 129
1 Perform/Review apply request
2 Perform backout to S/W Applied on SYS 03 on 2005-04-14:15:41
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 Prior snapshot 2002-04-14:18:03:03
DEFAULT Last S/W applied on SYS 00 on 2000-04-14:18:01
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 2002-04-14:18:01
To: SYS 01 applied on 2002-04-17:15:41
==============================================================================
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 2002-04-14:18:03:03
DEFAULT Last S/W applied on SYS00 on 2000-04-14:18:01
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
130 Backing Out the DSM/SCM Revision
F15-Help F16-Return SF16-Exit
DSM Software Configuration Manager Backout - Processing Backout
==============================================================================
\ELI Prior snapshot 2002-04-14:18:03:03
DEFAULT Last S/W applied on SYS00 on 2002-04-14:18:01
Backout 2000-04-18:11:00:39
==============================================================================
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 Prior snapshot 2002-04-14:18:03:03
DEFAULT Last S/W applied on SYS on 2000-04-14:18:01
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.
===============================================================================
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.
T0287 HSS - Halted State Services includes firmware changes.
T0287 PRM - Primitive State includes firmware changes.
T0287 IPF - Itanium(tm) Processor Firmware 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 F15-Help F16-Return SF16-Exit
Task completed successfully.
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.
Submit a Backout Request 131
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 the Planner Interface
1.
Log on to DSM/SCM (Start>Programs>dsm_scm>DSMSCM).
2.
Select Requests> Requests list.
The Requests dialog box appears, showing the type and status of various DSM/SCM requests.
3.
From the Request list, double-click your backout request.
The Status dialog box displays detailed information about your backout request.
This example indicates that the backout request finished successfully:
132 Backing Out the DSM/SCM Revision
Monitor the Backout Process Using the Planner Interface 133
12 Running ZPHIRNM to Reactivate the Previous Software
Revision
Fallback Considerations for NonStop Software Essentials After ZPHIRNM
If after migrating to this RVU, you decide to fall back to an RVU earlier than H06.24, you need to be aware of the fallback considerations for NonStop Software Essentials. NonStop Software
Essentials 2.0 or later components are not compatible with NonStop Software Essentials 1.0
components.
When you fall back, the ZPHIRNM step may rename the files in $SYSTEM.ZNSE, and you might not be able to continue using NonStopSoftware Essentials after reloading the system. These are the possible fallback scenarios you might encounter:
•
If the T0894H02 or later server was already installed on the fallback RVU, there should be no problems.
•
If the T0894H01 server was installed on the fallback RVU, you can either revert to using the
T0895V01 client from another CMS, or preferably you can install the T0894H02 or later server and T0884V02 client as described in the Nonstop Software Essentials Installation and
Quick Start Guide.
•
If no version of NonStop Software Essentials was installed on the fallback RVU, you can either revert to using the DSM/SCM Planner Interface, or you can install the T0894H02 or later server manually as described in the Nonstop Software Essentials Installation and Quick Start
Guide.
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 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 the 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:
134 Running ZPHIRNM to Reactivate the Previous Software Revision
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.
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
Run ZPHIRNM 135
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 or the DSM/SCM User's Guide for troubleshooting procedures and the DSM/SCM Messages Manual for a complete list of ZPHIRNM messages.
136 Running ZPHIRNM to Reactivate the Previous Software Revision
13 Falling Back on the CLIMs, Loading the Previous RVU, and Resuming Normal Operations
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.
NOTE: If your NonStop system does not have CLIMs installed, disregard the sections in this chapter that describe CLIM fallback procedures.
Alerts
Before performing a system load:
•
Before halting the processors, stop running applications, including TMF.
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.
•
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.
•
If you must stop the system startup process, click Abort. (The abort process should take effect within 30 seconds.)
•
Under Configuration File, select Saved Version (CONFigxx.yy:). List the saved configuration files by entering:
SCF FUP INFO $SYSTEM.ZSYSCONF.CONF*
You can refer to this listing later in the installation procedure.
•
For detailed information on performing a system load, see the information about starting and stopping a system in the NonStop Operations Guide.
CAUTION: If your intent is to remain on your current RVU, the NonStop Blade Complex firmware must be reverted back as explained under
“Revert to Previous NonStop Blade Complex
Task Overview
If you have CLIMs in your system and need to fall back to the previous version of the CLIM software, the steps to fall back on the 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 Compatibility Reference to determine whether reverting to the previous CLIM software is necessary for your fall back scenario. If your system does not have CLIMs installed, you can disregard these steps. Restoring the CLIMs assumes that you have saved the CLIM configurations by issuing the CLIMBKUP command before upgrading your system, as described in
“Back Up CLIM Configuration Files” (page 66)
and in the Cluster I/O
Protocols (CIP) Configuration and Management Manual. Falling back to previous versions of CLIM
Alerts 137
software adds approximately one hour for each CLIM to the process of loading the system on the previous RVU.
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.
Prepare to fall back to the previous CLIM firmware.
b.
Collect information about the CLIM environment and prepare to fall back to the previous
CLIM software.
c.
Modify the OSSINFIL file.
d.
If falling back to H06.15 or an earlier H-series RVU, migrate all CLIM services and disconnect the CLIMs.
e.
If falling back to the H06.03 or H06.04 RVU, update the shadow firmware on each
NonStop Blade element.
2.
Halt the processors.
3.
While the processors are halted: a.
If you are falling back to the H06.03 or H06.04 RVU, power cycle the processors.
b.
Fall back to the previous CLIM firmware using the Down System method.
c.
Reimage the CLIMs if you are reverting to the previous CLIM software.
d.
Prepare the CLIMs. Preparing the CLIMs involves configuring the interfaces that are used for the dedicated service LAN as well as for the Maintenance Providers, configuring the
Storage CLIM for its role in the system, and configuring all the CLIMs with the system information you gathered before halting the system.
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 to the previous NonStop Blade Complex if necessary.
7.
Install specific OSS files if necessary.
8.
Verify that the system is started.
Before Halting the Processors
Before halting the processors, perform these tasks.
If Falling Back to H06.24 or an Earlier RVU
Effective with the H06.25 RVU, a single CLIM can be associated with multiple data providers if the MULTIPROV attribute of the CLIM object is set to ON (enabled). If you need to fall back to
H06.24 or an earlier RVU, you must disable the MULTIPROV attribute to instruct CLIMs to follow pre-Multiple Providers per CLIM behavior and use the sole provider specified by the PROVIDER attribute. For detailed information, see the Cluster I/O Protocols (CIP) Configuration and
Management Manual.
If Falling Back to an RVU That Supports CLIMs
If you are falling back to an H-series RVU that supports CLIMs (H06.16 or later), perform these
CLIM fallback tasks.
138 Falling Back on the CLIMs, Loading the Previous RVU, and Resuming Normal Operations
(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
Prepare for Down System CLIM Firmware Update action in the OSM Service Connection. For more information, see the Cluster I/O Protocols (CIP) Configuration and Management Manual.
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.super TACL user ID.
3.
Obtain the DNS name of each CLIM iLO. The default name is system-climname-ILO, for example, mysys-n100251–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 a NonStop system console or a CLIM DVD. If you do not have the correct software image on a system console, install it from the H-series CLIM DVD by following the procedure in the Cluster I/O 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 either the CLIM Boot Service Configuration Wizard or the NonStop
Maintenance 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.
Before Halting the Processors 139
b.
To determine if one ore more of the CLIMs you are downgrading is being used for DHCP,
DNS, and BOOTP services, launch either the CLIM Boot Service Configuration Wizard or 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>HP 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.
Using the wizard, determine whether the CLIMs are being used for DHCP, DNS, and
BOOTP services. If they are, 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.
NOTE: If any CLIMs to be downgraded are being used for those services, you will use this information later, during the fallback step
“Migrate CLIM Services and Disconnect
.
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 $zzcip.c100251,detail
CIP Detailed Info CLIM \NSAA11.$ZZCIP.C100251
Mode....................... IP
140 Falling Back on the CLIMs, Loading the Previous RVU, and Resuming Normal Operations
Configured Location........ Group 100 , Module 2 , Slot 5 , Port 1
ConnPts.................... 2
X1 Location................ Group 100 , Module 2 , Slot 5 , Port 1
Y1 Location................ Group 100 , Module 3 , Slot 5 , Port 1
SvNet ID 1................. 0x000E0DC6
X2 Location................ Group 100 , Module 2 , Slot 5 , Port 2
Y2 Location................ Group 100 , Module 3 , Slot 5 , Port 2
SvNet ID 2................. 0x000E0EC6
Maintenance Interface IP... 16.107.184.198
Network CLIM Information:
MultiProv.................. ON
Provider................... CSAM2
Index...................... 1
The NonStop host system name in this example is \NSAA11 and the maintenance interface
IP address is 16.107.184.198.
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:
20> climcmd c100251 climconfig interface -info eth0 comForte SSH client version T9999H06_17May2009_comForte_SSH_0086
Interface : eth0
Interface Type : Physical Interface
Jumbo Frame : NO
IP Address : 16.107.184.198
Netmask : 255.255.255.0
ROUTE Details :
- Route Type : Default Route
Destination Address : 0.0.0.0
Netmask : 0.0.0.0
Gateway Address : 16.107.184.1
Metric : 0
Minimum RTO : Unspecified
Auto Negotiation Details :
- Status : on
Line Speed : Not Applicable
Duplex : Not Applicable
Software MAC Address : Unspecified
Termination Info: 0
In this example, the netmask is 255.255.255.0 and there is no gateway configured.
c.
Use OSM Low-Level Link to connect to the CLIM and obtain PCI Slot configuration information.
d.
Make note of all values.
If Falling Back to the H06.15 RVU or an Earlier RVU
Modify OSSINFIL
If you are falling back to an H06.15 or earlier H-series RVU, before halting the processors, you must restore the SERVER #ZPLS entry in the OSS subsystem maintenance file OSSINFIL on all systems with OSS installed. If you commented out this entry when installing this RVU, use TEDIT to modify the entry as follows:
TEDIT $SYSTEM.ZOSSINS.OSSINFIL
Then change
* SERVER #ZPLS to
SERVER #ZPLS
This change is required for correct startup of OSS.
Before Halting the Processors 141
Migrate CLIM Services and Disconnect CLIMs
The H06.15 RVU and earlier H-series RVUs do not support CLIMs. If you are using CLIMs and intend to fall back to H06.15 or an earlier H-series RVU, follow these general guidelines:
1.
Use the CLIM Boot Service Configuration Wizard or the NonStop Maintenance LAN DHCP
DNS Configuration Wizard to determine if services such as DHCP, DNS, or BOOTP are configured on the CLIMs or on the NonStop Console.
2.
If the services are configured on any of those CLIMs, you will need to migrate those services to other CLIMs or to system consoles. See Changing the DHCP, DNS, or BOOTP Server from
CLIMs to System Consoles or Changing the DHCP, DNS, or BOOTP Server from System
Consoles to CLIMs, as appropriate.
3.
Disconnect the CLIMs that are installed on the affected system.
If Falling Back to the H06.03 or H06.04 RVU
If you are falling back to the H06.03 or H06.04 RVU, before halting the processors, update the shadow firmware on each NonStop Blade Element, as described under
H06.04 From This RVU” (page 151) .
Halt the Processors
To halt the processors:
1.
Log on to the OSM Low-Level Link.
2.
On the toolbar, click Processor Status to display the Processor Status dialog box.
3.
In the Processor Status box, select all processors.
4.
From the Processor Actions list, select Halt.
5.
Click Perform action.
6.
Click OK when asked “Are you sure you want to do Halts?”
7.
When the Action Status box confirms that all processors have halted, close the Processor Status dialog box.
While the Processors are Halted
While the processors are halted, perform these tasks.
In the Halted State, Power Cycle Processors If Falling Back to the H06.03 or H06.04
RVU
If you are falling back to the H06.03 or H06.04 RVU, then while the processors are halted, power cycle the processors as described in step 6b under
“Reverting to H06.03 or H06.04 From This
In the Halted State, Fall Back to Previous CLIM Firmware (Optional)
If you are falling back to previous CLIM firmware using the down system method, use the Down
System CLIM Firmware Update Tool. For more information, see the Cluster I/O Protocols (CIP)
Configuration and Management Manual and the online help available from within the tool.
In the Halted State, Reimage the CLIMs
NOTE: Only reimage the CLIMs if you need 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 CLIMs must be performed by a NonStop authorized service provider. Authorized service providers can refer to Reimaging or Downgrading a CLIM for information about reimaging CLIMs.
H-series RVUs do not support Gen8 CLIMs.
142 Falling Back on the CLIMs, Loading the Previous RVU, and Resuming Normal Operations
In the Halted State, Prepare the CLIMs
1.
For each CLIM, set the IP address of eth0 via the iLO serial console.
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.
Use OSM Low-Level Link to perform the Configure action on each CLIM. In the Configure CLIM dialog box, enter all the information that you gathered before starting this procedure. For example:
Be sure to select the CLIM type for the type of CLIM you are configuring from the Mode of operation pull-down menu. After entering the required information in the dialog box, press
Configure. Here is an example screen showing the CLIM preparation has successfully completed:
While the Processors are Halted 143
4.
Click Yes. The CLIM reboots.
5.
If the system disk is a 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
6.
For the maintenance provider CLIMs, configure the maintenance provider interface: a.
For the two IP CLIMs configured with a Maintenance Provider (usually the first two IP
CLIMs, for example, N100251/N100252), 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 climconfig interface -info commands ( climcmd climname climconfig interface -info eth0:0
) 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
NOTE: The ipaddress and netmask values in the above command are examples, and these are provided by the CLIMCMD climconfig interface -info commands; however, eth0:0 is a hard-coded value for the CLIMs.
7.
When you receive the Remote disconnected without sending a disconnect message, click Ok.
144 Falling Back on the CLIMs, Loading the Previous RVU, and Resuming Normal Operations
Load the System
A normal system load consists of loading the operating system from disk into the memory of processor 0 or processor 1:
1.
In the OSM Low-Level Link, 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. For a description of the CIIN file, see the appropriate planning guide for your system.
3.
Under Configuration File, select Saved Version (CONFigxx.yy:)
Refer to the listing of saved 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
Load the System 145
OSM Low-Level Link, reboot the system console, and restart the system load using the OSM
Low-Level Link again.
6.
Monitor the system startup process. Messages indicating the progress and completion of the system load, as well as 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.
Complete the System Load (If Necessary)
Loading the processors with the CIIN file is usually automatic. However, 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 TACL RELOAD command as shown in the following steps.
You must be super ID (255,255) to load the processors.
1.
From the system console, log on to the system as a super-group user (255,nnn) and enter:
TACL> RELOAD *, PRIME
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
After Loading the Previous RVU
After loading the previous RVU, perform these tasks.
If Falling Back to the H06.03 or H06.04 RVU
If you are falling back to the H06.03 or H06.04 RVU, then after loading the previous RVU, finish loading the rest of the NonStop Blade Complex firmware, as described in step 6c under
“Reverting to H06.03 or H06.04 From This RVU” (page 151) .
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:
146 Falling Back on the CLIMs, Loading the Previous RVU, and Resuming Normal Operations
CLIM NAMEAdding a server to a VM cluster server group without an available connection (“sharer”) prevents connection to storage 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.
Complete CLIM Setup
1.
Establish trust between the CLIM and the NonStop host system.
If you are falling back to H06.20 (or to an RVU with T0682H02^ACC) or H06.21 (or to an
RVU with T0682H02^ACI), you must run SSOCLIM as SUPER.SUPER. If you are running or falling back to H06.22 or a later RVU, you can run SSOCLIM as any SUPER group user. 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 SUPER.SUPER (for a fallback to H06.20 or H06.21 or as any SUPER group user.
b.
Enter SSOCLIM.
The script prompts you for the CLIM name or IP address 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 66)
.
NOTE: The CLIM configuration files used during this procedure must be of the same RVU as the RVU to which you are falling back.
For the H06.16 RVU, rename and restore the backup files. For example, at the TACL prompt, enter: a.
>FUP RENAME $DATA1.BCKH0616.N100251, $DATA1.BCKH0616.BACKUPGZ
b.
>CLIMRSTR N100251 $DATA1.BCKH0616
c.
Repeat steps a. and b. for each CLIM.
For H06.17 and later RVUs, restore the backup files. For example, at the TACL prompt, enter: a.
>CLIMRSTR N100251 $SYSTEM.N10251.BCKH0617
b.
Repeat for each CLIM.
3.
Reboot the CLIMs: a.
For large systems, boot one of every pair of CLIMs on the system at the same time.
After Loading the Previous RVU 147
NOTE: For IP 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 n100251 psclim
If you do not see any running processes, use the CLIM command to clear the log and start. For example (IP CLIM):
>climcmd n100251 clim clearlog; clim start
For Storage CLIMs, after clearlog, you must reboot again. For example (Storage CLIM):
>climcmd s100253 clim clearlog
>climcmd s100253 clim reboot-nodebug 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 Low-Level Link from each system that relies on getting its HSS files as part of the
BOOTP service (all NonStop systems running J-series software except for NS2000) and use the HSS Files action to create and copy the system-specific HSS files to the CLIM. This action is available before and after system discovery.
a.
Before system discovery, click Copy HSS Files on the toolbar or select Copy HSS Files from the Tools menu.
b.
After system discovery, you must select Copy HSS Files from the Tools menu.
148 Falling Back on the CLIMs, Loading the Previous RVU, and Resuming Normal Operations
c.
Selecting Copy HSS Files brings up the Update HSS dialog box. It displays the current
HSS version for each processor on the system d.
Select the desired HSS version from the Available HSS versions drop-down menu.
e.
Enter the user name and password for every CLIM and the peer NSC to which the selected
HSS file needs to be copied and click Copy.
5.
If this is an IB CLIM, install third-party Informatica software for messaging and order sequencing.
After Loading the Previous RVU 149
NOTE: See Support Note SN 12023 (OSM application issues with Administrator user on NSC) for instructions on preventing authentication failures from occurring on NonStop System Consoles when using these OSM tools: Down System CLIM Firmware Update, Update HSS and Copy HSS file actions in OSM Low-Level Link, and connecting to a peer NSC from the NonStop Maintenance
LAN DHCP DNS Configuration Wizard or the CLIM Boot Service Configuration Wizard.
A fix of this problem is automatically installed when you run the Master Installer on NonStop System
Console Installer DVD, Update 18 or later. Support Note S12023 describes another workaround option if you do not have the Update 18 or later DVD.
Revert to Previous NonStop Blade Complex Firmware
Reverting to H06.05 or Later From This RVU
CAUTION: Do not power cycle any NonStop Blade Element until the firmware update is complete.
If you interrupt or stop the procedure before the firmware update is completed, the NonStop Blade
Element firmware will be left in an incomplete and inconsistent state. Also, if there is an interruption during the firmware update, do not reset or power cycle the NonStop Blade Elements. Instead, immediately start the procedure from the beginning.
Before starting this procedure:
•
Ensure that you are running the RVU to which you are reverting. If necessary, load (coldload) the system with that RVU.
•
If you are reverting to H06.12 or later, perform the OSM Initialize Default Firmware File
Names action.
To start and complete the Blade Complex Firmware Update guided procedure:
1.
In the tree pane of the OSM Service Connection, right-click the Blade Complex resource.
2.
From the shortcut menu, select Actions.
3.
When the Actions window appears, select Firmware Update from the Available Actions list.
4.
Click Perform action.
5.
In the guided procedures interface, click Start in the Procedure Tasks List pane.
6.
Follow and complete the guided procedure dialog boxes using this summary: a.
Using the OSM Service Connection and the Blade Complex Firmware Update guided procedure, complete the dialog boxes and update the shadow firmware on each NonStop
Blade Element.
NOTE: HP recommends that you update all the NonStop Blade Elements at the same time.
b.
Using the Switch Blade Element Firmware Images dialog box, select Yes to switch the
NonStop Blade Element image from shadow to running.
c.
Reset the NonStop Blade Elements one at a time. Wait for each NonStop Blade Element to be reintegrated before proceeding. Use the OSM Service Connection Blade Element
Reset Action to reset the NonStop Blade Element.
NOTE: HP recommends that you reset and reintegrate each NonStop Blade Element one at a time. (The Reset action power cycles the NonStop Blade Element.) Resetting all
NonStop Blade Elements at the same time will take four processors offline.
Do not manually power cycle the processors. Doing so will leave the NonStop Blade
Elements and NonStop Blade Complex in an incomplete state.
The shadow firmware image that has been updated is now the running flash image. The previous running image is now the shadow image.
150 Falling Back on the CLIMs, Loading the Previous RVU, and Resuming Normal Operations
d.
Repeat steps a. through c. to update the new shadow image on all NonStop Blade
Elements or one NonStop Blade Element at a time until all the NonStop Blade Elements are done.
e.
Repeat steps a. through d. for the other NonStop Blade Complexes.
Reverting to H06.03 or H06.04 From This RVU
CAUTION: If your intent is to remain on your current RVU, the NonStop Blade Complex firmware must be reverted back as described in this section.
CAUTION: Do not power cycle any NonStop Blade Element until the firmware update is complete.
If you interrupt or stop the procedure before the firmware update has completed, the NonStop
Blade Element firmware will be left in an incomplete and inconsistent state. Also, if there is an interruption during the firmware update, do not reset or power cycle the NonStop Blade Elements.
Instead, immediately start the procedure from the beginning.
To start and complete the Blade Complex Firmware Update guided procedure:
CAUTION: Ensure that the system is running the previous RVU before starting this procedure.
1.
In the tree pane of the OSM Service Connection, right-click the Blade Complex resource.
2.
From the shortcut menu, select Actions.
3.
When the Actions window appears, select Firmware Update from the Available Actions list.
4.
Click Perform action.
5.
In the guided procedures interface, click Start in the Procedure Tasks List pane.
6.
Follow and complete the guided procedure dialog boxes using this summary: a.
Update NonStop Blade Complex (NSBC) firmware:
1) Using the OSM Service Connection and the Blade Complex Firmware Update guided procedure, complete the dialog boxes and update the shadow firmware on each
NonStop Blade Element. The subvolume for each firmware element to be updated must be changed to point to the SYSnn of the RVU to which you are falling back.
NOTE: Update all NonStop Blade Elements in the complex at this time.
2) Using the Switch Blade Element Firmware Images dialog box, select Yes.
CAUTION: Do not reset the NonStop Blade Elements after this step.
3) Repeat steps
and
for the other NonStop Blade Complexes.
b.
Halt and then power cycle all processors:
1) Using OSM Low-Level Link, halt all the logical processors in the system.
2) Power cycle all processors as follows: a) Log on to the OSM Low-Level Link.
b) On the tool bar, click System Discovery.
c) In the Management Window, double-click the system button (labeled with the system name).
d) Right-click the GRP-400 button.
e) Select the Actions option.
f) From the GRP-400 actions list, select Power Cycle All Processors.
g) Click Perform Action.
Revert to Previous NonStop Blade Complex Firmware 151
h) Click OK when asked “Are you sure you want to do a Power Cycle All
Processors?” i) When the Action Status box confirms that the power cycle has successfully completed, close the dialog box.
NOTE: It should take approximately five minutes for all processors to complete
Power On initialization.
3) Load (coldload) the system with the RVU to which you are reverting.
c.
Update NonStop Blade Complex (NSBC) firmware on the other flash, as follows:
1) Using the OSM Service Connection and the Blade Complex Firmware Update guided procedure, complete the dialog boxes and update the shadow firmware on each
NonStop Blade Element.
NOTE: In the Select Firmware Modules for shadow image dialog box, select all parts that are noted as Up-Rev, except for BMC firmware (T0661); you must not revert to previous BMC firmware (see Support Note S08016).
2) Using the Switch Blade Element Firmware Images dialog box, select NO.
3) Repeat steps
and
for the other NonStop Blade Complexes.
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”
.
NOTE: Features introduced in H06.23 affect the default security settings in OSS. See Support
Note S12018 for detailed information.
Verify That the System Is Started
Verify that each processor is running the operating system:
1.
From the Summary menu in the OSM Low-Level Link, select Processor Status.
2.
The Processor Status dialog box should show each processor to be Executing NonStop OS.
152 Falling Back on the CLIMs, Loading the Previous RVU, and Resuming Normal Operations
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 H-series RVU.
See the H06.nn Release Version Update Compendium for this RVU and the Interactive Upgrade
Guide 2 (IUG2) for information on SPRs that you might need to apply to your current RVU before upgrading to this RVU and connecting to a G-series system.
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. You can use NonStop
Software Essentials to generate a number of 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 subvolumes.
See the NonStop Software Essentials User’s Guide for more information.
Tools for Managing SPRs 153
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 basically 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 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.
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 OCT 2004) SYSTEM \ELI Date 04 May 2005, 22:00:18
COPYRIGHT 2003 Hewlett-Packard Development Company, L.P.
System name \ELI
EXPAND node number 033
Current SYSnn SYS01
154 Using DSM/SCM With NonStop Software Essentials to Install SPRs
System number 000000
Software release ID H06.03.00
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
CODE EOF LAST MODIFIED OWNER RWEP PExt SExt
OSIMAGE 0 57354240 15DEC1999 9:57 255,255 NUNU 200 200
$SYSTEM.SYS01
CODE EOF LAST MODIFIED OWNER RWEP PExt SExt
OSIMAGE 0 57393152 05APR2000 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.
Start HP SIM and then open a NonStop Software Essentials window by selecting the Manage
NonStop Software link on the NonStop server Tools & Links tab.
3.
Click Receive software under the Software Installation group on the NonStop Software Essentials home page.
4.
Click the Input Source tab.
5.
If the SPR has been received on disk or has been restored from tape to disk, click Disk and load the DSVs into the table by entering a subvolume pattern and clicking the Find/Add button.
6.
If the SPR was received on tape, click Tape and select a tape drive from the list.
7.
In the Receive Software Request dialog box, click the Scheduling tab.
8.
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:
Install SPRs Using DSM/SCM With NonStop Software Essentials 155
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 Essentials User's Guide or the 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.
156 Using DSM/SCM With NonStop Software Essentials to Install SPRs
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 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.
Since the G06.10 RVU, 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 ***
HP 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
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:
If Running INSTALL^TEMPLATES Is Required 157
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: $SYSTEM.TEMPLATE.NEWNRES
*** INSTALL^TEMPLATES: Terminated due to error ***
The message indicates that the template and nonresident files specified in the INSTALL^TEMPLATES command are identical to those on the current subvolume. The INSTALL^TEMPLATES macro tries to create the newres and newnres files, but stops when it finds that these template files already exist.
To correct the problem:
1.
Rename the current files: tacl> rename newres, oldres tacl> rename newnres, oldnres
2.
Run the INSTALL^TEMPLATES macro again and use the file names newres and newnres.
This message indicates that the templates are successfully installed:
>$SYSTEM.SYS00.FUP /NAME $Z03M,INV :INBUF.1,OUTV :OUTBUF.1/
File Utility Program - T6553D45 - (14APR2000) System \TINY
Copyright Tandem Computers Incorporated 1981, 1983, 1985-2000
SECURE ($SYSTEM.TEMPLATE.NEWRES,$SYSTEM.TEMPLATE.NEWNRES),"NUNU"
*** INSTALL^TEMPLATES: Completed sucessfully ***
If a System Load Is Required
If OSM is installed and configured, use these instructions to load your RVU.
158 Using DSM/SCM With NonStop Software Essentials to Install SPRs
Before Halting the Processors
If you are falling back to the H06.03 or H06.04 RVU, then before halting the processors, update
the shadow firmware on each NonStop Blade Element, as described under “Reverting to H06.03
or H06.04 From This RVU” (page 151)
.
Halt the Processors
1.
If necessary, log on to OSM Low-Level Link.
2.
On the toolbar, select Processor Status to open the Processor Status dialog box.
3.
Select the processors to be halted.
4.
From the Processor Actions list, select Halt.
5.
Click Perform action.
6.
Click OK when asked if you are sure you want to halt.
7.
When the Action Status box confirms that all the processors have been halted, close the
Processor Status dialog box.
8.
Do not log off the OSM Low-Level Link.
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
Chapter 9 (page 111) , 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 10: “Resuming Normal Operations” .
Verify that your system is up and running. See
.
If a System Load Is Required 159
B Using DSM/SCM With the Planner Interface to Install
SPRs
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 H-series RVU.
See the H06.nn Release Version Update Compendium for this RVU and the Interactive Upgrade
Guide 2 (IUG2) for information on SPRs that you might need to apply to your current RVU before upgrading to this RVU and connecting to a G-series system.
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. You can use the
DSM/SCM Planner Interface to generate a number of 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 SYSnns.
See the DSM/SCM User’s Guide for more information.
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
160 Using DSM/SCM With the Planner Interface to Install SPRs
•
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 the Planner Interface
The steps for using DSM/SCM to install SPRs are basically 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 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.
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.
1.
At a TACL prompt, enter:
$SYSTEM SUPER 6> sysinfo
SYSINFO - T9268H01 - (01 OCT 2004) SYSTEM \ELI Date 04 May 2005, 22:00:18
COPYRIGHT 2003 Hewlett-Packard Development Company, L.P.
System name \ELI
EXPAND node number 033
Current SYSnn SYS01
System number 000000
Software release ID H06.03.00
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
CODE EOF LAST MODIFIED OWNER RWEP PExt SExt
OSIMAGE 0 57354240 15DEC1999 9:57 255,255 NUNU 200 200
Install SPRs Using DSM/SCM With the Planner Interface 161
$SYSTEM.SYS01
CODE EOF LAST MODIFIED OWNER RWEP PExt SExt
OSIMAGE 0 57393152 05APR2000 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.
Log on to the DSM/SCM Planner Interface.
3.
In the Software in Archive dialog box, click Receive software.
The Receive Software Request dialog box appears, and DSM/SCM automatically assigns the software input a name based on information received from the SPR input. You can enter a new name if you want.
4.
In the Receive Software Request dialog box, click Input source to display the Request Source
Options dialog box.
5.
If the SPR has been received on disk or has been restored from tape to disk, click
Volume.subvolume.
6.
Enter the name of the distribution subvolume containing the SPR, and then click Add.
7.
If the SPR was received on tape, click Tape drive and select a tape drive from the list.
8.
Click OK.
9.
In the Receive Software Request dialog box, click Scheduling.
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 OK.
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:
1.
In the Configuration Revisions dialog box, select the configuration you are updating, and click
New Software Revision.
The New software revision dialog box appears.
2.
In the Software in Archive dialog box, select SPRs to be added to the current configuration and click Copy to.
162 Using DSM/SCM With the Planner Interface to Install SPRs
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.
In the New software revision dialog box, click Build/Apply to list options for your Build/Apply.
The Build/Apply Request dialog box appears with all selections dimmed and unavailable.
In a few minutes, the dialog box is refreshed with defaults selected.
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 dialog box 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 options and configure as appropriate for your system. See the DSM/SCM User’s Guide or online help (Building and
Applying a New Configuration topic) for more information about these Build/Apply Request options.
6.
In the Build/Apply Request dialog box, click OK.
7.
When the Status window indicates that the request is done, select Window >Refresh.
This step updates the Software in Archive window to show the newly received software.
8.
Exit the DSM/SCM Planner Interface.
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.
Print Operator Instructions From the Target Interface (ZPHITI) 163
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.
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 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.
Since the G06.10 RVU, 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 ***
HP 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
164 Using DSM/SCM With the Planner Interface 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: $SYSTEM.TEMPLATE.NEWNRES
*** INSTALL^TEMPLATES: Terminated due to error ***
The message indicates that the template and nonresident files specified in the INSTALL^TEMPLATES command are identical to those on the current subvolume. The INSTALL^TEMPLATES macro tries to create the newres and newnres files, but stops when it finds that these template files already exist.
To correct the problem:
1.
Rename the current files: tacl> rename newres, oldres tacl> rename newnres, oldnres
2.
Run the INSTALL^TEMPLATES macro again and use the file names newres and newnres.
This message indicates that the templates are successfully installed:
>$SYSTEM.SYS00.FUP /NAME $Z03M,INV :INBUF.1,OUTV :OUTBUF.1/
File Utility Program - T6553D45 - (14APR2000) System \TINY
Copyright Tandem Computers Incorporated 1981, 1983, 1985-2000
If Running INSTALL^TEMPLATES Is Required 165
SECURE ($SYSTEM.TEMPLATE.NEWRES,$SYSTEM.TEMPLATE.NEWNRES),"NUNU"
*** INSTALL^TEMPLATES: Completed sucessfully ***
If a System Load Is Required
If OSM is installed and configured, use these instructions to load your RVU.
Before Halting the Processors
If you are falling back to the H06.03 or H06.04 RVU, then before halting the processors, update
the shadow firmware on each NonStop Blade Element, as described under “Reverting to H06.03
or H06.04 From This RVU” (page 151)
.
Halt the Processors
1.
If necessary, log on to the OSM Low-Level Link.
On the toolbar, select Processor Status to open the Processor Status dialog box.
2.
Select the processors to be halted.
3.
From the Processor Actions list, select Halt.
4.
Click Perform action.
5.
Click OK when asked if you are sure you want to halt.
6.
When the Action Status box confirms that all the processors have been halted, close the
Processor Status dialog box.
7.
Do not log off the OSM Low-Level Link.
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
Chapter 9 (page 111) , 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 10: “Resuming Normal Operations” .
Verify that your system is up and running. See
.
166 Using DSM/SCM With the Planner Interface to Install SPRs
C Copying the DSM/SCM, NonStop Software Essentials, and SWID SPRs from the SUT Using NonStop Software
Essentials
CAUTION: The NonStop Software Essentials server product (T0894H02) has been on the SUT since H06.2x. If you are preparing to install a SUT for H06.2x or an earlier H-series RVU, you will not find the NonStop Software Essentials SPR on the SUT. Instead, you must 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.
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 169)
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 at the same time 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 169)
, install the latest version of the NonStop Software Essentials client product as described in
“Install the Latest Version of the NonStop Software Essentials Client” (page 175)
.
•
If your system uses the DSM/SCM Planner Interface instead of NonStop Software Essentials, see
“Installing SWID, DSM/SCM PVU, and Planner Interface (PI) Client Software ” (page 176)
.
Alerts
•
If you are migrating to H06.23 or an earlier H-series RVU, the NonStop Software Essentials server SPR (T0894H02) will not be present on the SUT. Instead, you must use Scout to install the T0894H02^AAB SPR to your host system as described in
Earlier RVU, Install the Latest NonStop Software Essentials Server SPR Using Scout” (page 173)
.
•
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
,
“Determine the Product Version of DSM/SCM on the Host System and the SUT” (page 168)
, and
“Determine the Product Version of SWID on the Host System and SUT” (page 168) .
•
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 reinitiate the Build/Apply. See the DSM/SCM Messages Manual to learn the nature and severity of the problem.
Alerts 167
Determine the Product Version of NonStop Software Essentials Server on the Host
System and the SUT
CAUTION: The NonStop Software Essentials server product (T0894H02) has been on the SUT since H06.24. If you are preparing to install a SUT for H06.23 or an earlier H-series RVU, you will not find the NonStop Software Essentials SPR on the SUT. Instead, you must 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 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 T0894
> VPROC $SYSTEM.ZNSE.HTTPSRVR
For comparison, the NonStop Software Essentials server version on the SUT for this RVU is
T0894H02^AAB.
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
.
For information about installing the NonStop Software Essentials client product, see
Latest Version of the NonStop Software Essentials Client” (page 175) .
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
If the DSM/SCM 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 169) .
The correct DSM/SCM SPR for this RVU is T6031H02^ACY.
Determine the Product Version of SWID on the Host System and SUT
Using the VPROC command, identify the SWID SPR on the system.
For comparison, the SWID version on the SUT is T9298H02^AAV.
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 169) .
168 Copying the DSM/SCM, NonStop Software Essentials, and SWID SPRs from the SUT Using NonStop Software Essentials
Copy (RESTORE) the DSM/SCM Server, NonStop Software Essentials
Server, and SWID Product Versions from the SUT to Disk
CAUTION: The NonStop Software Essentials server product SPR (T0894H02) has been on the
SUT since H06.24. If you are preparing to install the SUT for H06.23 or an earlier H-series RVU, you will not find the NonStop Software Essentials SPR on the SUT during the
procedure. Instead, you must 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 H02 product version of DSM/SCM, NonStop Software Essentials server, and the
SWID product from this RVU to disk, at a TACL prompt, enter the RESTORE command:
>RESTORE/OUT $S.#nnn.DSM/$tape, ($*.E6031H02.*, *.E9298H02.*, E0894H02.*), VOL $DSMSCM, LISTALL where nnn is the RVU number (H06) and where $tape is the name of the tape drive containing this RVU.
The Windows (T0895) and Linux (T0996) versions of the NonStop Software Essentials clients are independent products and must be installed as SPRs from Scout as described in
Version of the NonStop Software Essentials Client” (page 175) .
Receive the H02 DSM/SCM, NonStop Software Essentials Server, and
SWID SPRs
1.
Start TMF and DSM/SCM if they are not already running:
2.
> TMFCOM START TMF > VOLUME $dsmscm_vol.ZDSMSCM > RUN STARTSCM
3.
Log on to NonStop Software Essentials.
4.
Click the Receive Software link on the Home Page.
5.
Under Software name, select Provided by planner and enter a name for the H02 DSM/SCM,
NonStop Software Essentials server, and SWID SPRs, up to 17 characters.
6.
Click the Input Source tab. NonStop Software Essentials displays the input source options.
7.
Select Subvolume.
8.
In the Find/Add text box, enter each distribution subvolume for the DSM/SCM files that you specified earlier in the RESTORE command:
$DSMSCM.E6031H02
nl
nl
$DSMSCM.E0894H02
nl
nl
$DSMSCM.E9298H02
Your Receive Software Input Source tab should look similar to this:
Copy (RESTORE) the DSM/SCM Server, NonStop Software Essentials Server, and SWID Product Versions from the SUT to
Disk
169
9.
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.
Create a New Software Revision for H02 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 set of products you received into the archive 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.
170 Copying the DSM/SCM, NonStop Software Essentials, and SWID SPRs from the SUT Using NonStop Software Essentials
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:
Added 01/07/14: DSM/SCM, NonStop Essentials server, and SWID c.
Click OK to save your changes to the revision notes.
Build and Apply the DSM/SCM H02 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.
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.
Build and Apply the DSM/SCM H02 Software Product Revision 171
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.super (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 157)
.
172 Copying the DSM/SCM, NonStop Software Essentials, and SWID SPRs from the SUT Using NonStop Software Essentials
If Migrating to H06.23 or an Earlier RVU, Install the Latest NonStop
Software Essentials Server SPR Using Scout
The NonStop Software Essentials server SPR is not present on the RVU SUTs H06.23 or earlier. If you are migrating to H06.23 or an earlier H-series RVU, you must install the T0894H02^AAB SPR from Scout to your system disk.
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 STARTSCM
These messages appear:
8> startscm
DSM/SCM Process Startup Function - STARTSCM - T6031H02 - (31MAR2003)
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 APR 2003, 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
If Migrating to H06.23 or an Earlier RVU, Install the Latest NonStop Software Essentials Server SPR Using Scout 173
$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:
At a TACL prompt, enter:
> VOLUME $dsmscm_vol.ZDSMSCM
> VOLUME TAEXE
> VPROC CBEXE
The VPROC command should now display the DSM/SCM SPR version T6031H02^ADB.
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 loaded from Scout, enter at the TACL prompt:
> VPROC T0894
> VPOROC $SYSTEM.ZNSE.HTTPSRVR
The VPROC command should now display the NonStop Software Essentials server version
T0894H02^AAB.
Verify that the SUT Version of SWID Is Installed
To verify that the SUT version of SWID is now installed on your system, use the VPROC command at the TACL prompt:
> \DRP08.$SYSTEM.SYSTEM 101> vproc swid
The VRPROC command should now display the SWID version T9298H02^AAV.
Recovery for Problems While Installing the H02 Version of DSM/SCM
If you encounter problems while installing the H02 version of DSM/SCM, see the DSM/SCM
User’s Guide.
If NonStop Errors Are Displayed
See the DSM/SCM Messages manual for information regarding the error message and the
DSM/SCM 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.
174 Copying the DSM/SCM, NonStop Software Essentials, and SWID SPRs from the SUT Using NonStop Software Essentials
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 the online help for more information.
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) or
Linux CMS (T0996) 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 HP Insight Control for NonStop DVD for the
RVU, which ships with the HP NonStop System Console Installer DVD, Update 17 or later.
For detailed pre-installation requirements and installation information for the Windows and Linux versions 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^AAB.
The NonStop Software Essentials client SPR for Linux is T0996V02^AAC.
If you have to fall back, you can use the newly installed client software.
Install the Latest Version of the NonStop Software Essentials Client 175
D Installing SWID, DSM/SCM PVU, and Planner Interface
(PI) Client Software
NOTE:
•
You must have installed both the DSM/SCM SPR and the corresponding client version before you receive an RVU
SUT. All newly shipped NonStop systems will have the latest DSM/SCM SPR installed. The latest DSM/SCM client version is available only on the SUT unless you have ordered new system consoles.
•
Always install the latest SWID PVU at the same time you install a new DSM/SCM PVU to avoid a fingerprint or mismatched file problem. (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.) Your newly shipped system will have the latest SWID SPR.
•
DSM/SCM is not installed and not supported on 64-bit consoles. If you have a Windows 2008 (64-bit) console, use the instructions for installing SWID and DSM/SCM using NonStop Software Essentials; see
DSM/SCM, NonStop Software Essentials, and SWID SPRs from the SUT Using NonStop Software Essentials”
Alerts
•
“Determine the Product Version of DSM/SCM on the Host System and the SUT” (page 168)
and
“Determine the Product Version of DSM/SCM on the Host System” (page 176)
.
•
Save the current configuration database. (See
“Save the Current Configuration File
).
•
Keep the startup TACL windows, the EMS Event Stream window, and the DSM/SCM 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 reinitiate the Build/Apply. See the DSM/SCM Messages Manual to learn the nature and severity of the problem.
Determine the Product Version of DSM/SCM on the Host System
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
If your VPROC commands do not display the version for this RVU, install the correct H02 DSM/SCM product version by following the procedures in
“Copy (RESTORE) the H02 DSM/SCM Product
Version to Disk (If Necessary)” (page 177) .
For comparison, the DSM/SCM version on the SUT is T6031H02^ADB.
Determine the Latest SWID Product
Using the VPROC command, identify the SWID SPR.
The SWID version for this RVU is T9298H02^AAV.
Determine the Product Version of DSM/SCM on the Client
To determine which product version of DSM/SCM client software is installed, start the DSM/SCM
Planner Interface. Before prompting you to log on, the About DSM/SCM window briefly displays version information. For example:
176 Installing SWID, DSM/SCM PVU, and Planner Interface (PI) Client Software
You can also bring up this screen with Help/About; this method leaves the window open until you close it.
To install the DSM/SCM client, see
“Install the DSM/SCM Client From ZDSMSCM ” (page 183)
.
Copy (RESTORE) the H02 DSM/SCM Product Version to Disk (If Necessary)
To copy the H02 product version of DSM/SCM and the SWID product from this RVU to disk, at a TACL prompt, enter the RESTORE command:
>RESTORE/OUT $S.#nnn.DSM/$tape, ($*.E6031H02.*, *.E9298H02.*), VOL $DSMSCM, LISTALL where nnn is the RVU number (H06) and where $tape is the name of the tape drive containing this RVU.
Receive the H02 SWID and DSM/SCM SPRs
1.
Start TMF and DSM/SCM if they are not already running:
2.
> TMFCOM START TMF > VOLUME $dsmscm_vol.ZDSMSCM > RUN STARTSCM
3.
Select Start>Programs>dsm_scm>DSMSCM to log on to the DSM/SCM Planner Interface.
4.
In the Software in Archive dialog box, select Receive software.
DSM/SCM displays the Receive Software Request dialog box.
5.
Under Software name, select Provided by planner and enter a name for the H02 DSM/SCM and SWID PVUs up to 17 characters.
6.
Under Options, click Input source.
DSM/SCM displays the Request Source Options dialog box.
7.
Select Volume.subvolume.
8.
In the Add text box, enter each distribution subvolume for the DSM/SCM files that you specified earlier in the RESTORE command:
$DSMSCM.E6031H02
$DSMSCM.E9298H02
Your Request Source Options dialog box should look similar to this:
Copy (RESTORE) the H02 DSM/SCM Product Version to Disk (If Necessary) 177
9.
Click OK.
You are returned to the Receive Software Request dialog box.
10.
Click OK to run the request immediately. (The request takes less than 5 minutes to finish.)
The Status dialog box appears. Wait until the display for the Request activity summary appears as Complete, and then close the Status window. For example:
178 Installing SWID, DSM/SCM PVU, and Planner Interface (PI) Client Software
11.
Select Window>Refresh to update the Software in Archive dialog box to include the received software.
Create a New Software Revision for H02 DSM/SCM and SWID SPRs
1.
In the Configuration Revisions dialog box, select the target system being upgraded and the current revision of the system being upgraded. Select New software revision.
DSM/SCM opens the New dialog box listing the products in the last software revision that was built for the target. Allow several minutes for the dialog box to appear.
2.
In the Software in Archive dialog box, select the DSM/SCM software you just received and click Copy to.
The DSM/SCM files are copied from the Archive to the new software revision.
3.
From the New software revision dialog box, verify that the new version of DSM/SCM 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:
Added 06/07/08: DSM/SWID c.
Click OK to add a written history of revisions to existing notes.
Build and Apply the DSM/SCM H02 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
Create a New Software Revision for H02 DSM/SCM and SWID SPRs 179
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.
1.
In the New software revision dialog box, 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 Build/Apply Request dialog box appears with all selections dimmed and unavailable.
In a few minutes, the dialog box is refreshed with defaults selected. The name assigned to the
Build/Apply request appears after Request name.
2.
Click Edit instructions to display the Planner Instructions box.
3.
Edit the instructions to provide the current SYSnn number, and then click OK.
4.
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 OK.
DSM/SCM runs the Build/Apply requests and displays a Status dialog box that is automatically updated at regular intervals.
5.
Check EMS messages for more detailed information about your Build/Apply:
•
From the Status window, click Show EMS events.
•
In the EMS Events window, click Refresh to see newer messages.
•
When finished viewing, click Close.
6.
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 requests are complete.
7.
Select Window>Refresh.
The Configurations Revision dialog box is refreshed and shows the new software configuration revision containing the H02 DSM/SCM SPR.
180 Installing SWID, DSM/SCM PVU, and Planner Interface (PI) Client Software
Rename the Applied DSM/SCM H02 Files By Running ZPHIRNM
1.
Stop DSM/SCM:
> VOLUME $dsmscm_vol.ZDSMSCM
> RUN STOPSCM where $dsmscm_vol is the DSM/SCM volume.
2.
Run ZPHIRNM:
> RUN ZPHIRNM $volume.SYSnn where SYSnn is the current SYSnn (because no SYSGEN or system load is required).
3.
When asked if you want to use a log file with the session, enter Y.
4.
When prompted, enter a file name for the log file.
ZPHIRNM displays this message:
Log file, filename, was successfully created.
ZPHIRNM processes the file renaming.
5.
Verify that ZPHIRNM ran successfully by reviewing the specified log file. For an example of output, see
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.super (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 157)
.
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 next.
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 ZDSMSCM subvolume to start DSM/SCM:
Rename the Applied DSM/SCM H02 Files By Running ZPHIRNM 181
> RUN STARTSCM
These messages appear:
8> startscm
DSM/SCM Process Startup Function - STARTSCM - T6031H02 - (31MAR2003)
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 APR 2006, 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 Correct SPR of DSM/SCM Is Installed
To verify that the appropriate product version of DSM/SCM is currently 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 $dsmscm_vol.ZDSMSCM
> vproc taexe
> vproc cbexe
The DSM/SCM version for this RVU is T6031H02^ACY.
Verify that SWID Is Installed
This example shows the H02 product version of SWID. Use the VPROC command. For example:
\DRP08.$SYSTEM.SYSTEM 101> vproc swid
The SWID version for this RVU is T9298H02^AAV.
182 Installing SWID, DSM/SCM PVU, and Planner Interface (PI) Client Software
Recovery for Problems While Installing the H02 Version of DSM/SCM
If you encounter problems while installing the H02 version of DSM/SCM, see the DSM/SCM
User’s Guide.
If NonStop Errors Are Displayed
See the DSM/SCM EMS Messages manual for information regarding the error message and the
DSM/SCM User’s Guide for information on deleting a Build/Apply request.
Cancel Pending Build/Apply Requests
NOTE: If you inadvertently close the Status dialog box, reopen it by selecting Requests> Requests list and then double-clicking the request.
1.
Select the request to be canceled.
2.
Select Requests>Cancel request.
The cancel request action takes a few minutes to finish.
NOTE: If you cannot cancel the Build request through the DSM/SCM Planner Interface, use the DSM/SCM Target Interface (ZPHITI) to cancel the request. See the DSM/SCM User’s
Guide and online help for more information.
3.
Using the Planner Interface, rename the old software inputs: a.
In the Software in Archive dialog box, select the software input.
b.
Select File>Show software input description to display the SUT Description dialog box.
c.
In the Software name box, edit the software input name so that another operator will know it is a DSM/SCM input that should not be used to build a new DSM/SCM revision.
d.
Click OK.
Install the DSM/SCM Client From ZDSMSCM
NOTE: Before you begin, back up your current DSM/SCM folder to save the configuration information and the Planner Interface version.
Check that unnecessary PC programs are not running before you begin the DSM/SCM client installation and that no DSM/SCM Build or Applies are in progress.
If you have to fall back, you can use the newly installed client software.
NOTE: HP recommends that both the DSM/SCM product version (T6031G02/H02) and
DSM/SCM client be at the same version level at all times. To make sure you have the right client, check the version using VPROC. Install it if necessary.
Determine Your DSM/SCM Client Software Version
To determine which version of DSM/SCM client software is installed, start the DSM/SCM Planner
Interface. Before prompting you to log on, the About DSM/SCM window briefly displays version information. See
“Determine the Product Version of DSM/SCM on the Host System” (page 176)
.
Determine the Current IP Address and Port Number
When you install DSM/SCM client software, you are prompted to enter the IP address of the host system with which DSM/SCM communicates and the IP port number for the DSM/SCM Winsock service.
1.
If the DSM/SCM client software on your system console is configured to communicate with a host system, determine the configured IP address and port number:
2.
Start Windows Explorer. Open My Computer if not already open.
3.
Double-click the hard drive icon to display its contents.
4.
Open the DsmScm folder to display its contents.
Recovery for Problems While Installing the H02 Version of DSM/SCM 183
5.
In the Contents pane, double-click dsmscm.ini.
Notepad opens the DSM/SCM initialization file. For example:
6.
Make a note of the IP address and port number.
7.
Close the dsmscm.ini file.
Copy the ZDSMSCM Client File From the Host
1.
From the Windows Start menu, select Run and enter the command CMD .
An example response is:
Microsoft Windows XP [Version 5.1.2600]
(C) copyright 1985–2001 Microsoft Corp.
C:\Documents and Settings\Administrator
>
2.
Enter: cd c:\
3.
If a DSMSCM directory does not exist, enter the following at the C:\ prompt to create it: mkdir DSMSCM
4.
At the C:\ prompt, enter: cd DSMSCM
5.
Enter: ftp <ip-address>, where ip-address is the IP address of your NonStop server.
For example, ftp 192.231.36.10.
An example response is:
Connected to 192.231.36.10.
220 osmlanx FTP SERVER T9552H02 (Version H02 TANDEM 03MAY2006) ready
User (192.231.36.10: (none)): super.super
331 Password required for SUPER.SUPER.
6.
Enter the super.super password.
An example response is:
230 User SUPER.SUPER logged in. OSS API enabled
7.
At the ftp> prompt, enter: quote guardian.
An example response is:
257 Guardian API enabled
8.
Enter: bin.
An example response is:
200 Type set to I
9.
Enter: cd $DSMSCM.ZDSMSCM.
An example response is:
184 Installing SWID, DSM/SCM PVU, and Planner Interface (PI) Client Software
250 CWD command successful
10.
Enter: get zdsmexe zdsm.exe.
An example response is:
200 PORT command successful.
150 Opening data connection for zdsmexe (192.231.36.1,3212d) (2334 bytes).
226 Binary Transfer complete.
ftp: 2334123 bytes received in 0.20 seconds 11441.78Kbytes/sec.
ftp>
11.
Move the zdsm.exe file into the desired folder on the NonStop console, such as c:\DSMSCM or C:\DSMH06nn.
12.
Double click on zdsm.exe to expand the files into the desired folder.
Install the DSM/SCM Client Software
1.
Using Windows Explorer, double-click Setup.exe.
A DSM/SCM Setup screen appears, and an Enter Object Manager Server Class dialog box appears with the default OM-001.
2.
Click Next to accept the default.
The Enter Software Product Group Type (shared or private) dialog box appears.
Install the DSM/SCM Client From ZDSMSCM 185
3.
Click Next to accept the default (private) or type in shared and click Next.
NOTE: Private software product groups can be used only by the planner who creates the grouping. Shared software product groups can be used by other planners. For more information on these options, see the DSM/SCM User’s Guide. (NonStop Software Essentials only supports shared product groups and HP recommends using shared product groups for PI as well.)
The Enter CNFGINFO Process Name dialog box appears with the default name $ZPHC.
4.
Click Next to accept the default.
The Setup Type dialog box appears:
186 Installing SWID, DSM/SCM PVU, and Planner Interface (PI) Client Software
5.
Check that WINSOCK is selected. Click Next.
The Choose Destination Location dialog box appears with the default path displayed, based on your Windows operating system:
6.
Click Next to accept the default.
You are now prompted to enter an IP address:
Install the DSM/SCM Client From ZDSMSCM 187
7.
Enter the IP address for the host system with which DSM/SCM communicates (see
“Determine the Current IP Address and Port Number” (page 183) ), and click Next.
NOTE: If DSM/SCM is not currently configured to communicate with a host system, accept the default IP address. You can always change this value later by editing the dsmscm.ini file.
You are now prompted to enter an IP port number:
8.
Enter the IP port number for the Winsock service (see
“Determine the Current IP Address and
Port Number” (page 183) ), and click Next.
NOTE: If DSM/SCM is not currently configured to communicate with a host system, accept the default IP port number. You can always change this value later by editing the dsmscm.ini
file.
You are now prompted to enter the host system name.
188 Installing SWID, DSM/SCM PVU, and Planner Interface (PI) Client Software
9.
Enter the host system name (or \noname if DSM/SCM is not currently configured to communicate with a host system), and click Next to continue.
You are asked if you would like to run DSM/SCM.
10.
Click No.
The setup program adds the DSM/SCM client software to the Windows Start>Programs menu.
You can start the DSM/SCM Planner Interface or CNFGINFO tool from the
Start>Programs>DSM_SCM submenu. For example:
Install the DSM/SCM Client From ZDSMSCM 189
E Updating SWAN Concentrator CLIP Firmware
If necessary, you can update the SWAN or SWAN 2 concentrator CLIP firmware after you have completed a system load of the new RVU.
See either the SWAN Concentrator Installation and Support Guide or the SWAN 2 Concentrator
Installation and Support Guide.
190 Updating SWAN Concentrator CLIP Firmware
Glossary
A7CINFO file
ALLPROCESSORS paragraph
Archive
A configuration file distributed with OSS products that contains information on the contents of all pax
(archive) files of an OSS DSV.
A required paragraph in the CONFTEXT configuration file that contains attributes defining the
HP NonStop™ operating system image for all system processors. The ALLPROCESSORS paragraph follows the optional DEFINES paragraph.
For DSM/SCM, a set of unstructured files used to collect the software received onto the host system. Files received as input are placed in the DSM/SCM archive, and attributes of the files are stored in the host database. You can use the Archive and Database Maintenance Interface to specify the archive location in the Configuration Manager profile.
A block-mode interface run by a database or system administrator at both the host system and target systems to perform Distributed Systems Management/Software Configuration Manager
(DSM/SCM) maintenance functions.
Archive and
Database
Maintenance
Interface concentrator manager process
(ConMgr)
CONFAUX file
A process provided as part of the wide area network (WAN) subsystem. The ConMgr process runs in each processor that supports WAN products and provides management functions to the
WAN subsystem and WAN products, such as downloading data link control (DLC) tasks to the communications line interface processors (CLIPs) on the ServerNet wide area network (SWAN) concentrator and selecting the preferred path for the DLC tasks.
The auxiliary configuration file created by the Distributed Systems Management/Software
Configuration Manager (DSM/SCM) tools. The CONFAUX file contains a list of the code files and system files that are needed to build the new HP NonStop operating system. HP recommends that you avoid making any changes to your CONFAUX file.
CONFBASE file
CONFIG file
In G-series or H-series release version updates (RVUs), the basic system configuration database file, which is stored on the $SYSTEM.SYSnn subvolume. See also
.
In G-series or H-series release version updates (RVUs), the current system configuration database file, which is stored on the $SYSTEM. ZSYSCONF subvolume. See also
.
configuration file In G-series and H-series release version updates (RVUs), one of the following files: CONFBASE,
CONFIG, one or more saved configuration files named CONFxxyy, and CONFSAVE.
configuration planner configuration revision
The person who manages system configuration changes and software configuration changes.
This person modifies the system configuration database for system configuration changes and creates a new operating system image for software configuration changes. See also
A planner-defined set of software products and related configuration information that the Distributed
Systems Management/Software Configuration Manager (DSM/SCM) can activate on a target system. Multiple configuration revisions might exist on a target system. A configuration revision is made up of the product versions named in its software revision list, its HP NonStop operating system image, and the relevant profile items, such as the location of the target subvolumes on the target system. It is created by a Build request and is included in the activation package sent to the target system.
configuration tag Each configuration tag identifies and configures the topology of a cluster switch and its unique position within the topology. The configuration tag defines which ServerNet node numbers the cluster switch supports.
configuration utility process
CONFTEXT file
The $ZCNF process that is the access process for the CONFIG file and starts and maintains the
$ZPM persistence manager process.
The configuration file used as input during system generation that contains a series of entries defining your HP NonStop™ operating system attributes. A G-series or H-series CONFTEXT file consists of one or two paragraphs: DEFINES (optional) and ALLPROCESSORS.
current image
CUSTFILE
The firmware image in the local flash memory that is currently running.
An EDIT file included on every site update tape (SUT) as $SYSTEM.Annnnnn.CUSTFILE, where
nnnnnn
is the system serial number of the target system. The CUSTFILE contains information on the software products on the SUT, their related files, and the destination and use of each file. HP customizes information in the CUSTFILE for each customer’s system.
191
Distributed Systems Management/Software Configuration Manager (DSM/SCM)
A graphical user interface (GUI)-based program that installs new software and creates a new HP
NonStop™ operating system. DSM/SCM creates a new software revision and activates the new software on the target system.
distribution subvolume (DSV)
A subvolume containing program files for a particular software product along with the release version update (RVU) document (softdoc) file for that product. The format for a DSV name is
Ynnnnrrr or Rnnnnrrr, where nnnn is the software product number and rrr is the base version identifier (such as D20) or software product revision (SPR) identifier (such as AAB).
duplex
Gigabit Ethernet
4-port ServerNet adapter (G4SA)
GNSC
Having two active processor elements, each contained in a separate NonStop Blade Element, per logical processor. Also called dual-modular redundancy (DMR).
A multiport ServerNet adapter that provides 1000 megabits/second (Mbps) data transfer rates between HP NonStop™ systems and Ethernet LANs.
Global NonStop Solution Center guided procedure A software tool that assists you in performing complex configuration or replacement tasks on an
HP NonStop™ server. OSM guided procedures are launched through actions in the OSM Service
Connection.
halted-state services (HSS)
HP NonStop Kernel operating system
A millicode module that provides a maintenance interface when a processor is in the stopped state. HSS is sometimes called boot millicode.
The former name of the operating system for HP NonStop systems.
Fault-tolerant, modular HP computers that support the HP NonStop™ operating system and are based on the Intel® Itanium® processor.
Identifies software used to manage or service HP NonStop servers.
HP NonStop
NS-series servers
HP NonStop Open
System
Management
(OSM) Interface
HP NonStop Open
System Services
(OSS)
HP NonStop operating system
HP NonStop servers
The product name for the OSS environment.
The operating system for HP NonStop systems.
The entire line of HP NonStop servers, including NonStop NS-series and NonStop S-series servers.
HP NonStop
Storage
Management
Foundation (SMF)
HP NonStop
Technical Library
(NTL)
HP NonStop
Transaction
Management
Facility (TMF)
HP Tandem
Advanced
Command
Language (TACL) millicode
A subsystem used by the storage subsystem that facilitates automation of storage management tasks by providing location-independent naming, storage pools, and virtual disks on HP NonStop™ systems.
The application for accessing, searching, and viewing technical publications and support information for the HP NonStop server. NTL replaces Total Information Manager (TIM).
HP software that provides transaction protection and database consistency in demanding online transaction processing (OLTP) and decision-support environments. It gives full protection to transactions that access distributed SQL and Enscribe databases, as well as recovery capabilities for transactions, online disk volumes, and entire databases.
The user interface to the HP NonStop™ operating system. The TACL product is both a command interpreter and a command language. Users can write TACL programs that perform complex tasks or provide a consistent user interface across independently programmed applications.
The system’s lowest-level machine-dependent code, often coded in assembler language. TNS/E millicode and TNS/R millicode are functionally similar to the microcode on TNS systems. The system has several types of millicode, including machine interrupt handlers, operating system
192 Glossary
Open System
Services (OSS) primitives, routines implicitly called from native-compiled code, emulators for TNS floating-point arithmetic, and emulators for privileged-only or long-running TNS machine operations.
An open system environment available for interactive or programmatic use with the HP NonStop operating system. Processes that run in the OSS environment usually use the OSS application program interface. Interactive users of the OSS environment usually use the OSS shell for their command interpreter. Synonymous with Open System Services (OSS) environment.
The HP NonStop Open System Services (OSS) application program interface (API), tools, and utilities.
Open System
Services (OSS) environment
Open System
Services (OSS)
Monitor planner
A Guardian utility that accepts commands affecting OSS objects through an interactive Guardian interface named the Subsystem Control Facility (SCF).
1. The Distributed Systems Management/Software Configuration Manager (DSM/SCM) user who is responsible for planning and managing new software revisions. The planner uses the
DSM/SCM Planner Interface to carry out these functions.
2. A person who participates in planning a new HP NonStop™ system.
Planner Interface A graphical user interface (GUI) to the Distributed Systems Management/Software Configuration
Manager (DSM/SCM) that runs on the host system. It provides an interface to all the host
DSM/SCM planner functions.
A firmware library based on the Intel® Itanium® architecture.
privileged architecture library
(PAL) release version update (RVU)
RELOAD shadow triplex
A collection of compatible revisions of HP NonStop operating system software products, identified by an RVU ID, and shipped and supported as a unit. An RVU consists of the object modules, supporting files, and documentation for the product revisions. An RVU also includes a set of documentation for the overall RVU.
An HP Tandem Advanced Command Language (TACL) command to load the HP NonStop operating system image from disk over the ServerNet fabrics into the memory of the processor.
The processor used to check the results of the master processor on a processor board.
Having three active processor elements per logical processor. Also called triple-modular redundancy (TMR).
193
Index
Symbols
$SYSTEM.ZTCPIP,
$SYSTEM.ZTCPIP subvolume,
$ZCDB,
$ZPLS,
6763 Common Communication ServerNet Adapter
(CCSA),
A
AUDITTRAILBUFFER parameter, DP2 and RDF performance,
B
Back up files backing up important files before upgrading,
ZTCPIP subvolume,
Backup and Restore,
Blade Complex Firmware guided procedure,
Build process, DSM/SCM,
Build/Apply request, defined,
C
CBEXE product determining version on host system,
verifying version,
CIIN file,
,
disabled,
CLCI sessions,
CLIM compatibility with host RVUs,
fallback procedure,
firmware,
installing CLIM software,
CNSL session,
Code Profiling Utilities,
Commands
RELOAD,
STATUS TAPE,
CONFIG file saving,
use in loading the previous RVU,
use in loading the system,
Current image,
D
DP2 fallback,
DSAP/DCOM,
DSM/SCM build process,
prepackaged reports,
Target Interface,
E
EasySetup,
ESS
FCHECK utility,
migrating data,
mod-14 extent boundary.,
partitioned unstructured files,
F
Fabricated file names,
Fallback summary of major tasks,
ZPHIRNM,
FCHECK utility,
Fibre Channel ServerNet Adapter (FCSA) ,
Firmware updates,
Floating-point default type,
G
Gigabit 4-port ServerNet Adapter (G4SA),
I
inetd process,
INITSTAT,
INSTALL^TEMPLATES instructions,
,
using with DSM/SCM,
Installing the DSM/SCM client from ZDSMSCM,
Interoperability audit format issues,
SPRs for NonStop BladeSystems and NonStop Systems,
IPs, SUT products in H-series RVUs,
K
Kernel-Managed Swap Facility,
Kernel-managed swap files,
L
Loading the J-series RVU,
Loading the system,
,
,
M
merge_whatis,
Migration issues,
MR-Win6530 terminal emulator startup windows on,
N
NetBatch,
New software revision, DSM/SCM,
New software revision, NonStop Software Essentials,
NonStop Cluster switch firmware,
NonStop Remote Database Facility,
NonStop Software Essentials, changing profile information,
NonStop SQL/MX,
NonStop System Console Installer DVD,
194 Index
NonStop TCP/IP configuration files,
NonStop TCP/IPv6 files,
NSKCOM,
O
OSS
EasySetup,
,
large files,
Root Name Server $ZPNS,
Version 1 filesets,
P
Pcleanup utility,
Planner Interface, changing profile information,
PORTCONF file,
Post-installation configuring OSS,
configuring RDF,
configuring SMF,
updating TCP/IP configuration files,
Preemptable object code,
PREVIEW mode,
Processor type in CONFTEXT file,
Profiles, target, changing information,
PROTOCOL file configuration file location,
PSTATE,
R
RELOAD command,
S
Safeguard attribute defaults,
longer passwords,
new defaults for attributes,
password encryption,
password quality enhancements,
support for OSS Access Control List,
SANMAN,
,
SCF testing the tape subsystem,
verifying critical processes,
verifying disks,
Scout for NonStop Servers,
ServerNet adapter firmware,
ServerNet Cluster,
SERVICES file,
Shadow image,
SMF,
SMF, interoperability and configuration,
SNETMON,
Software revision,
Spooler backing up, command for,
checking status of collector,
COLLECT, STATUS command,
,
DRAIN command,
draining,
SPRNOTES,
SPRs interoperability,
Status dialog box, reopening,
SUT, maximum size of,
SWAN and SWAN 2 concentrator firmware,
Swap file space, checking ,
SYSnn subvolumes listing,
System configuration file,
T
T0895,
T6031H02,
TACL RELOAD command,
TAEXE,
Tandem Failure Data System (TFDS),
TAPEBOOT,
Target subvolumes (TSVs),
Target systems, changing profile information,
TEMPLI program,
TFDS see Tandem Failure Data System (TFDS) ,
TMF,
TNS/R tools on H-series systems,
Token Ring ServerNet Adapter (TRSA),
TRM2000,
TS/MP,
U
Updating firmware user-updateable,
V
Visual Inspect,
VPROC command,
W
WHOHAS,
Z
ZcccGP files,
ZCNFDEFS,
ZDLLnnn,
ZINSAIDS subvolume,
ZMODGP,
ZMODGP, in Preview mode,
ZPHIRNM
PREVIEW mode,
reactivating previous software revision,
running,
ZPHITI canceling a Build/Apply request,
starting,
195
Descargar
Enlace público actualizado
Se ha actualizado el enlace público a tu chat. Puedes ver y gestionar tus enlaces de chat después del registro aquí.
Enlace público del documento
Anuncio