H06.25 Software Installation and Upgrade Guide HP Part Number: 696595-002 Published: March 2014 Edition: H06.
© 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.
Contents About This Document...................................................................................11 Supported Release Version Updates (RVUs)................................................................................11 Intended Audience..................................................................................................................11 New and Changed Information in This Edition............................................................................
IOAM and Migrating Existing Data to an Enterprise Storage System (ESS).................................36 OSS Database Files...........................................................................................................36 Installing OSS Products Not Managed by DSM/SCM.............................................................37 OSS General Alerts and Considerations...............................................................................37 OSS Unautomated Configuration and Startup........
Compiling C++ Programs to Run on RVUs Earlier Than H06.15................................................55 SQL/MX Versioning Feature Removed..................................................................................55 H06.16..................................................................................................................................55 AF_UNIX Release 2 Migration and Fallback..........................................................................55 CLuster I/O Module (CLIM).....
Verify Critical System Processes............................................................................................70 Verify Disk Drives...............................................................................................................72 Verify Tape Drives (If Necessary)..........................................................................................72 Verify That the CLIMs Are in the Started State........................................................................
Verifying That the System Is Started....................................................................................118 7. (Optional) Updating the CLIM Firmware...............................................................................119 10 Resuming Normal Operations................................................................120 Revive the Down Half of the Mirrored Disk...............................................................................120 Start User and System Applications.........
Complete CLIM Setup.......................................................................................................147 Revert to Previous NonStop Blade Complex Firmware................................................................150 Reverting to H06.05 or Later From This RVU........................................................................150 Reverting to H06.03 or H06.04 From This RVU....................................................................
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 SPRs...................................................................................................................................170 Build and Apply the DSM/SCM H02 Software Product Revision.................................................171 Run the INSTALL^TEMPLATES Macro.............................................
Tables 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 Tasks and Tools for Backing Out to the Previous SYSnn .........................................................22 Maintenance Entity (ME) Firmware File Names and Locations................................................25 Maintenance Entity (ME) FPGA File Names and Locations.....................................................26 H-Series ServerNet Adapters File Names and Locations........................................................
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.
• 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.
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.
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.
• 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.
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: DVD.
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.
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.
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 Target NonStop System RVU T0431H01^AAK H06.10 and H06.
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.
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.
Table 1 Tasks and Tools for Backing Out to the Previous SYSnn To perform this task... Use this tool... Documented in this chapter... 1. Submit a backout request. DSM/SCM Target Interface (ZPHITI) Chapter 11: “Backing Out the DSM/SCM Revision” 2. Return the firmware to previous versions. (Not recommended. If problems occur, firmware can be downgraded at this time.) OSM Service Connection OSM Service Connection online help 3.
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 Chapter 3: “Installation, Migration, and Fallback Considerations”.
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.
To determine whether your system has processor switches, IOAMs, or VIO enclosures, see the planning guide for the system. • • “Updating ME Firmware” ◦ Updating ME Firmware on a Processor Switch ◦ Updating ME Firmware on an I/O Adapter Module ◦ Updating ME Firmware on a VIO Enclosure “Updating ME FPGA” For information on using OSM Low-Level Link to perform ME firmware updates, see the online help available within the application.
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. 2. 3. 4. 5.
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.
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.
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.
1) 2) 3) 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. Using the Switch Blade Element Firmware Images dialog box, select NO. Repeat steps 1 and 2 for the other NonStop Blade Complexes.
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.
• 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.
Table 4 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.
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.
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.
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.
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.
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.
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.
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.
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.
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.
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.
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.
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.
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.
1. 2. 3. 4. 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. Start Safeguard. The system administrator must set a grace period for users to change their expired passwords.
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.
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.
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.
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.
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.
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.
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.
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.
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.
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.
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.
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.
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.
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.
4 Upgrade Checklist Use the checklist in “Upgrade Checklist” (page 63) to track your progress as you perform the tasks required to upgrade to this RVU and the checklists Table 7 and Table 8 (page 63) to record the information needed when performing these tasks.Table 10 (page 65) lists SPRs that should be installed before upgrading to this RVU. CAUTION: (page 63).
Table 7 Required Information Checklist, Before You Start (continued) Type of Information Value ServerNet Cluster firmware • Cluster Switch • Cluster Switch (logic board; router-2) DSM/SCM product version (server and client) Processor type Record the information in Table 8 while you perform the upgrade.
Table 9 Upgrade Checklist (continued) Task Description Reference 8. Verify system operations. “Verify System Processes” (page 68) 9. Install CLIM software on the NonStop System Console. 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) 10.
Table 9 Upgrade Checklist (continued) Task Description Reference Done? 26. For systems with IOAMs, update IOAM firmware, if necessary. “Updating ME Firmware on an I/O Adapter Module (IOAM)” (page 26) 27. Update NonStop Blade Complex (NSBC) firmware. “Managing NonStop Blade Complex Firmware” (page 27) 28. Update G4SA Download Code firmware, if necessary. “Managing ServerNet Adapter Firmware” (page 31) 29. Install ISP1040 firmware, if necessary. “Managing ServerNet Adapter Firmware” (page 31) 30.
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.
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.
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.
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. 4.
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.
LDev 0 1 3 5 6 7 79 81 82 83 84 106 107 108 109 113 126 127 128 129 130 179 180 181 182 183 184 186 187 188 189 190 191 192 193 194 195 196 197 198 199 200 201 202 203 205 206 207 208 209 210 211 231 232 233 234 235 236 237 238 239 240 241 242 243 244 246 247 248 249 250 Name PPID $0 0,3 $NCP 0,14 $YMIOP 0,5 $Z0 0,7 $SYSTEM 0,257 $ZOPR 0,8 $ZZKRN 0,302 $ZZWAN 0,299 $ZZSTO 0,300 $ZZLAN 0,301 $ZZCIP 0,362 $ZSNET 0,302 $ZSLM2 0,297 $ZPTY 2,417 $ZPMON 0,365 $ZNET 0,18 $ZM03 3,289 $ZM02 2,289 $ZM01 1,290 $ZM00
251 252 253 254 268 293 295 296 299 300 302 331 350 918 2-> $AUDTR3 $AUDTR2 $AUDTR1 $AUDIT $ZTCP0 $ZTC0 $ZTN0 $ZOLHD $ZTNP0 $ZTCP1 $ZTNP1 $ZEXP $ZRD9 $ZHOMR 2,300 1,296 0,330 0,331 0,424 0,450 0,451 0,531 0,455 1,422 1,426 0,61 1,447 0,89 3,291 0,329 1,294 1,292 1,419 1,391 1,413 1,448 1,421 0,456 0,458 1,38 2,427 ( 3,51) 4096 220 \QA2100.$SYSTEM.SYS00.TSYSDP2 ( 3,51) 4096 220 \QA2100.$SYSTEM.SYS00.TSYSDP2 ( 3,51) 4096 220 \QA2100.$SYSTEM.SYS00.TSYSDP2 ( 3,51) 4096 220 \QA2100.$SYSTEM.SYS00.
1. Enter the SCF STATUS TAPE command: > SCF STATUS TAPE $* 2. Verify that the STATUS TAPE display includes a least one tape drive configured and running, as shown: STORAGE - Status TAPE $T0150 LDev 71 3. 4. State Sub State STARTED UP Primary PID 0,282 Backup PID 1,273 Device Status READY Use the SCF STATUS TAPE, DETAIL command to display the current state of the tape drive. If the status of the tape drive is stopped or not ready, start the tape drive and put the tape drive online.
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.
such as Expand over IP and ServerNet Cluster. See the NonStop Networking Overview for more information about and diagrams of NonStop platform interoperability.
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.
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.
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.
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.
6. 7. 8. 9. 10. Under Software name, select Provided by planner and enter a name for the RVU. Click the Input Source tab. Select Tape drive. From the list, select the tape drive that contains the SUT. 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: 80 • The status of the request in the Status window.
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.
b. c. Click the Output tab to view the default output values. 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. 5. Click OK to return to the Configuration Revisions page.
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.
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.
2. 3. Under Perform steps, select Build, apply & process snapshot. 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. 5. 6. Make a note of the new SYSnn.
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 $S 8. STATE ACTIVE FLAGS 3 , CPU 1 PRI 154 UNIT 4 DATA FILE %FULL $OPS0.SPL.DATA 4 In the Build/Apply Request page, click Submit Build/Apply.
• 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.
Selection number: 0 ============================================================================== Choose an action by entering the selection number of an item from the list below.
F1-Select F15-Help F16-Return SF16-Exit Ready 5. 6. Enter 3 (Review operator instructions for the last apply completed) in the Selection number field, and press F1 to display the operator instructions. 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.
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.
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.
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.
6. 7. Under Software name, select Provided by planner and enter a name for the RVU. In the Options group box, click Input source. The Request Source Options dialog box appears. 8. 9. Select Tape drive. 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.
• 94 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.
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.
b. c. Click Output to view the default output values. 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. 3. 4. 5. Click OK. DSM/SCM returns you to the Target Information dialog box.
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.
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.
2. 3. Under Perform steps, select Build, apply & process snapshot. 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. 5. 6. Make a note of the new SYSnn.
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 $S 9. STATE ACTIVE FLAGS 3 , CPU 1 In the Build/Apply Request dialog box, click OK. 100 Installing a SUT for This RVU Using the Planner Interface PRI 154 UNIT 4 DATA FILE %FULL $OPS0.SPL.
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.
• 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.
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.
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.
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.
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.
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.
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.
TSVs TSVs TSVs TSVs in old in new in old shared configuration .................. configuration only ............. configuration only ............. by new and old configurations .. 130 0 1 129 TSVs processed in new config ............... TSVs processed in old config (obsolete) .... 129 1 Files Processed: Files Files Files Files Files Files Files Files Files Files processed in new configuration ........................... processed in new configuration only ......................
where: 2. 3. • $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. When asked if you want to use a log file with the session, enter Y. When prompted, enter a file name for the log file: 6> run zphirnm $system.
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: Guide. For detailed information on performing a system load, see in the NonStop Operations 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.
Table 17 System Load Paths in Order of Use for CLIM-Attached-Load (continued) Load Path Description Data Travels From To Processor Over ServerNet Fabric 9 Primary $SYSTEM-P 1 Y 10 Primary $SYSTEM-P 1 X 11 Backup $SYSTEM-B 1 X 12 Backup $SYSTEM-B 1 Y 13 Mirror $SYSTEM-M 1 Y 14 Mirror $SYSTEM-M 1 X 15 Mirror backup $SYSTEM-MB 1 X 16 Mirror backup $SYSTEM-MB 1 Y A CLIM-Attached-Load attempts to load the system from a system disk in the NonStop Storage CLIMs connecte
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.
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.
1. 2. 3. 4. 5. 6. 7. Log on to the OSM Low-Level Link. On the toolbar, click Processor Status to display the Processor Status dialog box. Under Processor Status, select all processors. From the Processor Actions list, select Hard Reset. Click Perform action. Click OK when asked “Do you really want to perform a Hard Reset on the selected processor(s)?” When the Action Status box confirms that all processors have been reset, close the Processor Status dialog box. 5.
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.
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.
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.
2. 3. Close the Processor Status dialog box. 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. 5.
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.
Cold Start the Pathway Configuration 1. 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. 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.
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.
>FUP >FUP >FUP >FUP DUP DUP DUP DUP $SYSTEM.ZTCPIP.SMPLNETW, $SYSTEM.ZTCPIP.SMPLPORT, $SYSTEM.ZTCPIP.SMPLRESC, $SYSTEM.ZTCPIP.SMPLSERV, $SYSTEM.ZTCPIP.NETWORKS $SYSTEM.ZTCPIP.PORTCONF $SYSTEM.ZTCPIP.RESCONF $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.
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.
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.
1. 2. Configure the WAN manager as described in the WAN Subsystem Configuration and Management Manual. 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.
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.
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.
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 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.
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.
Monitor the Backout Process Using NonStop Software Essentials 1. 2. Log on to NonStop Software Essentials. 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. 2.
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.
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.
6. 7. Review the log file. 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.
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.
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.
(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.
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.
Configured Location........ ConnPts.................... X1 Location................ Y1 Location................ SvNet ID 1................. X2 Location................ Y2 Location................ SvNet ID 2................. Maintenance Interface IP... Group 100 , Module 2 Group 100 , Module Group 100 , Module 0x000E0DC6 Group 100 , Module Group 100 , Module 0x000E0EC6 16.107.184.
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.
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: b. • root/hpnonstop • user/clim 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.
4. 5. Click Yes. The CLIM reboots. 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.
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.
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.
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.
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.
5. 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. e. Select the desired HSS version from the Available HSS versions drop-down menu. 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. If this is an IB CLIM, install third-party Informatica software for messaging and order sequencing.
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.
d. e. 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. 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.
h) i) Click OK when asked “Are you sure you want to do a Power Cycle All Processors?” 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) c. Load (coldload) the system with the RVU to which you are reverting.
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.
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.
System number Software release ID 2. 000000 H06.03.00 Issue a TACL FILEINFO command on $system.*.osimage to list the SYSnn subvolumes on your system and determine which one is open. $SYSTEM.SYS00 OSIMAGE CODE 0 EOF LAST MODIFIED OWNER 57354240 15DEC1999 9:57 255,255 CODE 0 EOF LAST MODIFIED 57393152 05APR2000 8:56 RWEP NUNU PExt 200 SExt 200 $SYSTEM.SYS01 OSIMAGE 3.
1. 2. 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. 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.
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. 3. When asked if you want to use a log file with the session, enter Y and press Enter. When prompted, enter a file name for the log file.
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.
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. 2. 3. 4. 5. 6. 7. 8. If necessary, log on to OSM Low-Level Link. On the toolbar, select Processor Status to open the Processor Status dialog box. Select the processors to be halted. From the Processor Actions list, select Halt.
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.
• 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.
$SYSTEM.SYS01 CODE 0 OSIMAGE 3. EOF LAST MODIFIED 57393152 05APR2000 8:56 OWNER RWEP 255,255 NUNU PExt SExt 200 200 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 0,0 201 P R 000 255,255 $SYSTEM.SYS01.NMONTOR Swap File Name: $SYSTEM.#0 Hometerm $YMIOP.
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. 3. 4.
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. • 1.
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 $system.template 4.
SECURE ($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.
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.
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.
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.
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 170 1. From the New Revision page, select the target system being upgraded and the current revision of the system being upgraded. Select Create.
3. 4. 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. 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.
NonStop Software Essentials runs the Build/Apply requests and displays a Status page that is automatically updated at regular intervals. 4. 5. 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.
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.
$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 to continue or BREAK to exit.
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.
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.
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.
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.
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.
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.
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. 4. When asked if you want to use a log file with the session, enter Y. When prompted, enter a file name for the log file. ZPHIRNM displays this message: Log file, filename, was successfully created.
> 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.
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.
5. In the Contents pane, double-click dsmscm.ini. Notepad opens the DSM/SCM initialization file. For example: 6. 7. Make a note of the IP address and port number. 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. 3.
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.
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.
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.
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.
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.
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.
Glossary A7CINFO file A configuration file distributed with OSS products that contains information on the contents of all pax (archive) files of an OSS DSV. ALLPROCESSORS paragraph 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.
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.
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. Open System Services (OSS) 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.
Index FCHECK utility, 36 migrating data, 36 mod-14 extent boundary., 36 partitioned unstructured files, 36 Symbols $SYSTEM.ZTCPIP, 122 $SYSTEM.
NonStop TCP/IP configuration files, 122 NonStop TCP/IPv6 files, 107 NSKCOM, 74 OSS EasySetup, 37, 125 large files, 43 Root Name Server $ZPNS, 81, 95 Version 1 filesets, 48 SPRNOTES, 154, 160 SPRs interoperability, 74 Status dialog box, reopening, 105, 183 SUT, maximum size of, 14 SWAN and SWAN 2 concentrator firmware, 34 Swap file space, checking , 73 SYSnn subvolumes listing, 67 System configuration file, 154, 161 P T Pcleanup utility, 121 Planner Interface, changing profile information, 96 PORTCONF f