Trailing-Edge
-
PDP-10 Archives
-
tops10v704_auxft
-
10,7/aux/sig.mem
There are 2 other files named sig.mem in the archive. Click here to see a list.
| FIELD TEST DRAFT
TOPS-10 Software Installation Guide
AA-P512C-TB
The TOPS-10 Software Installation Guide
provides the site administrator, system
manager, system programmer, or software
specialist with the information required to
install the TOPS-10 software on KL and KS
processors.
| OPERATING SYSTEM: TOPS-10 Version 7.04
|
| SOFTWARE: GALAXY Version 5.1
digital equipment corporation . marlboro, massachusetts
First Printing, February 1984
|
| Revised, April 1986
| Revised, March 1988
|
|
|
The information in this document is subject to change without notice
and should not be construed as a commitment by Digital Equipment
Corporation. Digital Equipment Corporation assumes no responsibility
for any errors that may appear in this document.
The software described in this document is furnished under a license
and may only be used or copied in accordance with the terms of such
license.
No responsibility is assumed for the use or reliability of software or
equipment that is not supplied by DIGITAL or its affiliated companies.
|
|
|
| Copyright (C) 1984, 1986, 1988 by Digital Equipment Corporation
| All Rights Reserved.
|
|
|
The postage-prepaid READER'S COMMENTS form on the last page of this
document requests the user's critical evaluation to assist us in
preparing future documentation.
The following are trademarks of Digital Equipment Corporation:
DEC DECnet IAS
DECUS DECsystem-10 MASSBUS
Digital logo DECsystem-20 PDT
PDP DECwriter RSTS
UNIBUS DIBOL RSX
VAX EduSystem VMS
VT
CONTENTS
PREFACE
PART 1: GETTING STARTED
CHAPTER 1 INTRODUCTION
1.1 READING PATH FOR NEW INSTALLATIONS . . . . . . . . 1-1
1.2 READING PATH FOR EXISTING INSTALLATIONS . . . . . 1-3
1.3 SUPPLEMENTARY INFORMATION . . . . . . . . . . . . 1-4
CHAPTER 2 PREPARATION
2.1 TOPS-10 SYSTEM INSTALLATION MATERIALS . . . . . . 2-2
2.2 KL FRONT END INSTALLATION MATERIALS . . . . . . . 2-3
2.3 REQUIRED HARDWARE CONFIGURATIONS . . . . . . . . . 2-3
2.3.1 KL Hardware . . . . . . . . . . . . . . . . . . 2-3
2.3.2 KS Hardware . . . . . . . . . . . . . . . . . . 2-4
2.4 POWERING UP THE SYSTEM . . . . . . . . . . . . . . 2-4
2.4.1 Powering up the KL10 . . . . . . . . . . . . . . 2-5
2.4.2 Powering up the KS10 . . . . . . . . . . . . . . 2-5
2.5 PREPARING INFORMATION . . . . . . . . . . . . . . 2-5
2.5.1 Worksheet 1 . . . . . . . . . . . . . . . . . . 2-5
2.5.2 Worksheet 2 . . . . . . . . . . . . . . . . . . 2-5
2.5.3 Worksheet 3 . . . . . . . . . . . . . . . . . . 2-6
PART 2: PREPARING THE SYSTEM
CHAPTER 3 BOOTING THE SYSTEM
3.1 BOOTING THE KL SYSTEM . . . . . . . . . . . . . . 3-1
3.1.1 Intializing the KL Front End . . . . . . . . . . 3-2
3.1.2 Booting the KL Monitor . . . . . . . . . . . . . 3-3
3.2 BOOTING THE KS SYSTEM . . . . . . . . . . . . . . 3-4
CHAPTER 4 STARTING THE BOOTSTRAP MONITOR
4.1 THE ONCE DIALOGUE . . . . . . . . . . . . . . . . 4-1
4.2 BRINGING UP A NEW SYSTEM . . . . . . . . . . . . . 4-3
iii
CHAPTER 5 BUILDING THE MONITOR FILE SYSTEM
5.1 ALLOCATING DISK SPACE FOR THE KL FRONT END . . . . 5-2
5.2 RESTORING BACKUP FROM TAPE . . . . . . . . . . . . 5-3
5.3 RESTORING FILES FROM TAPE TO DISK . . . . . . . . 5-3
5.3.1 Restoring the Bootstrap Monitor . . . . . . . . 5-3
5.3.2 Restoring the Monitor Sources and Associated
Software . . . . . . . . . . . . . . . . . . . . 5-4
5.4 CREATING THE TOPS-10 SYSTEM DIRECTORIES . . . . . 5-5
5.5 COPYING THE FILES INTO DIRECTORIES . . . . . . . . 5-6
CHAPTER 6 BUILDING THE SUPPLEMENTARY PROCESSOR'S FILE SYSTEM
6.1 BUILDING THE KL FRONT END FILE SYSTEM . . . . . . 6-1
6.1.1 Loading the Front End . . . . . . . . . . . . . 6-2
6.1.2 Booting the KL Front End from Disk . . . . . . . 6-5
6.2 CREATING THE KS MICROPROCESSOR FILE SYSTEM . . . . 6-6
6.2.1 Creating a Front end File System Directory Area 6-6
6.2.2 Writing the Bootstrap Program to Disk . . . . . 6-6
PART 3: UPGRADING AN EXISTING SYSTEM
CHAPTER 7 RESTORING THE MONITOR DISTRIBUTION TAPES
7.1 RESTORING KS MICROPROCESSOR FILES . . . . . . . . 7-2
7.2 RESTORING THE MONITOR SOURCES AND ASSOCIATED
SOFTWARE . . . . . . . . . . . . . . . . . . . . . 7-2
7.3 RESTORING TAPES USING COMMAND FILES . . . . . . . 7-3
7.4 USING SYSTEM FILES . . . . . . . . . . . . . . . . 7-4
CHAPTER 8 RESTORING THE SUPPLEMENTARY PROCESSOR'S FILE SYSTEM
8.1 THE DISTRIBUTED KL FRONT END SOFTWARE . . . . . . 8-1
8.2 USING RSXINS . . . . . . . . . . . . . . . . . . . 8-2
8.2.1 The RSXINS Dialogue . . . . . . . . . . . . . . 8-2
8.2.2 Installing the Front End File System . . . . . . 8-6
8.2.3 Automatic Front End Reload . . . . . . . . . . . 8-8
8.3 WRITING THE KS BOOTSTRAP PROGRAM TO DISK . . . . . 8-9
PART 4: BUILDING THE MONITOR AND TAILORING THE
SYSTEM
CHAPTER 9 BUILDING YOUR MONITOR
9.1 MONGEN . . . . . . . . . . . . . . . . . . . . . . 9-2
9.1.1 Running MONGEN . . . . . . . . . . . . . . . . . 9-3
iv
9.1.2 The MONGEN Dialogue Questions . . . . . . . . . 9-5
9.1.2.1 Running SYSGEN . . . . . . . . . . . . . . . 9-6
9.1.2.2 Running FGEN . . . . . . . . . . . . . . . . 9-14
9.2 ASSEMBLING THE MONITOR . . . . . . . . . . . . . 9-16
9.3 LINKING AND SAVING THE MONITOR . . . . . . . . . 9-17
9.3.1 Copying the New Monitor to SYS: . . . . . . . 9-17
9.4 BRINGING UP THE NEW MONITOR . . . . . . . . . . 9-18
9.5 CREATING THE SYSTEM BOOTABLE TAPE . . . . . . . 9-19
9.5.1 Creating the KL System Bootable Tape . . . . . 9-19
9.5.2 Creating the KS System Bootable Tape . . . . . 9-20
9.6 AUTOMATING THE BUILD PROCEDURE . . . . . . . . . 9-20
9.6.1 Customizing the Build Files . . . . . . . . . 9-22
9.6.2 The Build Procedure . . . . . . . . . . . . . 9-25
9.7 CUSTOMIZING BOOT . . . . . . . . . . . . . . . . 9-26
9.7.1 Customizing the KL BOOT . . . . . . . . . . . 9-26
9.7.2 Customizing the KS BOOT . . . . . . . . . . . 9-28
CHAPTER 10 STARTING UP AND TAILORING THE GALAXY SYSTEM
10.1 STARTING THE GALAXY SPOOLERS . . . . . . . . . . 10-2
10.2 TAILORING THE GALAXY SYSTEM . . . . . . . . . . 10-3
10.2.1 Running GALGEN . . . . . . . . . . . . . . . . 10-3
10.2.2 Assembling the Tailored GALAXY System . . . . 10-11
10.3 SHUTTING DOWN THE GALAXY SYSTEM . . . . . . . . 10-12
10.4 STARTING THE GALAXY SYSTEM . . . . . . . . . . . 10-14
CHAPTER 11 MAINTAINING THE SYSTEM AUTHORIZATION FILE
11.1 CONVERTING OLD ACCOUNTING FILES . . . . . . . . 11-1
11.2 ENTERING AND EXITING REACT . . . . . . . . . . . 11-1
11.3 GETTING HELP WHILE USING REACT . . . . . . . . . 11-2
11.4 DISPLAYING USER ACCOUNT INFORMATION . . . . . . 11-3
11.5 CREATING NEW ACCOUNTS . . . . . . . . . . . . . 11-4
11.5.1 Creating Multiple Accounts with the Same
Project Number . . . . . . . . . . . . . . . . 11-4
11.5.2 Creating a Single New User Account . . . . . . 11-5
11.6 CHANGING ACCOUNTS . . . . . . . . . . . . . . . 11-7
11.7 DELETING ACCOUNTS . . . . . . . . . . . . . . . 11-8
11.8 VERIFYING CHANGES TO ACCOUNTING FILES . . . . . 11-8
11.9 USING ACCOUNT VALIDATION . . . . . . . . . . . . 11-10
11.9.1 CREATING THE PROJCT.ACT FILE . . . . . . . . . 11-10
11.9.2 CREATING THE PROJCT.EXE FILE . . . . . . . . . 11-11
11.10 MAINTAINING LOGIN FAILURE RECORDS . . . . . . . 11-12
11.11 TOP-LEVEL REACT COMMANDS . . . . . . . . . . . . 11-13
11.12 USER-MODE COMMANDS . . . . . . . . . . . . . . . 11-15
CHAPTER 12 CREATING SYSTEM INITIALIZATION FILES
12.1 INITIA AT SYSTEM STARTUP . . . . . . . . . . . . 12-1
v
12.1.1 The SYSJOB.INI File . . . . . . . . . . . . . 12-2
12.1.1.1 Format of SYSJOB.INI . . . . . . . . . . . . 12-3
12.1.2 The TTY.INI File . . . . . . . . . . . . . . . 12-5
12.1.2.1 Format of TTY.INI . . . . . . . . . . . . . 12-5
12.1.2.2 Specifying Terminal Lines . . . . . . . . . 12-6
12.1.2.3 Specifying Terminal Options . . . . . . . . 12-7
12.1.2.4 Associated (Error) Messages . . . . . . . . 12-10
12.1.3 The INITIA Monitor Command . . . . . . . . . . 12-10
PART 5: SUPPLEMENTARY INFORMATION
APPENDIX A ONCE STARTUP OPTIONS
A.1 ONCE AT SYSTEM STARTUP . . . . . . . . . . . . . . A-2
A.2 RUNNING ONCE IN USER MODE . . . . . . . . . . . . A-6
A.3 STARTUP OPTIONS AND SWITCHES . . . . . . . . . . . A-7
A.3.1 CHANGing System Parameters . . . . . . . . . . A-11
A.3.2 DEFINing Structures and System Lists . . . . . A-14
A.3.3 DESTROYing All Structures . . . . . . . . . . A-20
A.3.4 DISSOLving a Single Structure . . . . . . . . A-23
A.3.5 EXITing to Monitor Level . . . . . . . . . . . A-24
A.3.6 GO Option . . . . . . . . . . . . . . . . . . A-24
A.3.7 NOINITIA Option . . . . . . . . . . . . . . . A-25
A.3.8 REFRESHing Selected Structures . . . . . . . . A-25
A.3.9 SHOWing System Parameters . . . . . . . . . . A-26
APPENDIX B MODIFYING THE MONITOR
B.1 REBUILDING THE MONITOR LIBRARY FILE . . . . . . . B-1
B.2 CHANGING COMMON SYMBOLS . . . . . . . . . . . . . B-2
B.2.1 Symbols Defined in COMMON . . . . . . . . . . . B-2
B.2.1.1 Decimal Symbols,Values . . . . . . . . . . . . B-2
B.2.1.2 Octal Symbols,Values . . . . . . . . . . . . . B-4
B.2.1.3 SIXBIT Symbols,Values . . . . . . . . . . . . B-5
B.2.2 Symbols Defined in COMDEV . . . . . . . . . . . B-5
B.2.2.1 Decimal Symbols,Values . . . . . . . . . . . . B-5
B.2.2.2 Octal Symbols,Values . . . . . . . . . . . . . B-6
B.2.3 Symbols Defined in COMMOD . . . . . . . . . . . B-7
B.2.3.1 Decimal Symbols,Values . . . . . . . . . . . . B-7
B.2.3.2 Octal Symbols,Values . . . . . . . . . . . . . B-7
B.2.3.3 SIXBIT Symbols,Values . . . . . . . . . . . . B-8
APPENDIX C SETTING SCHEDULER PARAMETERS
C.1 REQUIRED PRIVILEGES . . . . . . . . . . . . . . . C-1
C.2 CAUTIONS . . . . . . . . . . . . . . . . . . . . . C-1
C.3 SCDSET COMMANDS . . . . . . . . . . . . . . . . . C-2
C.3.1 EDIT . . . . . . . . . . . . . . . . . . . . . . C-2
vi
C.3.2 EXIT . . . . . . . . . . . . . . . . . . . . . . C-2
C.3.3 HELP . . . . . . . . . . . . . . . . . . . . . . C-3
C.3.4 READ/SET . . . . . . . . . . . . . . . . . . . . C-3
C.3.5 UPDATE JOBS . . . . . . . . . . . . . . . . . . C-5
C.4 THE SCDMAP.SYS FILE . . . . . . . . . . . . . . . C-6
APPENDIX D CREATING BOOTABLE FRONT END MEDIA
D.1 BUILDING THE PRIMARY BOOTABLE VOLUMES . . . . . . D-1
D.2 BUILDING THE THIRD VOLUME . . . . . . . . . . . . D-4
APPENDIX E SOFTWARE INSTALLATION GUIDE GLOSSARY
TABLES
9-1 Monitor Build Files . . . . . . . . . . . . . . 9-21
11-1 FAILUR.LOG File Format . . . . . . . . . . . . . 11-12
11-2 Top-level REACT Commands . . . . . . . . . . . 11-13
11-3 REACT Field Change Commands . . . . . . . . . . 11-15
12-1 Special Processing Commands . . . . . . . . . . 12-8
12-2 Options to Run Programs . . . . . . . . . . . . 12-9
A-1 Disk Parameters Defaulted By DESTROY Option . . A-21
A-2 Device Channels and Controllers . . . . . . . . A-22
C-1 SCDMAP.SYS Format . . . . . . . . . . . . . . . . C-6
PREFACE
This manual describes how to install, upgrade, customize, and start
the TOPS-10 operating system. It includes procedures for loading the
default monitor, copying the programs from the distribution tapes to
disk, building the monitor, starting the operating system, and
creating the bootstrap tape. The procedures for installing,
customizing, and starting the GALAXY batch and spooling system are
also included, as well as descriptions of the various system files
needed for timesharing on the TOPS-10 system. The procedures are
described and explained in a step-by-step method.
The TOPS-10 Software Installation Guide may also be used in the
following situations to:
o Start the system after a shutdown or failure.
o Change the monitor configuration.
o Install a new version of the monitor or GALAXY subsystem.
o Change disk parameters.
This manual is intended for experienced TOPS-10 software maintainers
who are familiar with TOPS-10 conventions for commands and symbols.
You should be able to operate the hardware components of the system
being installed. For example, instructions for mounting software
media (tapes and disks) are not documented in this manual because
those procedures are documented in the TOPS-10 Operator's Guide. The
following manuals are also valuable sources of information for
additional reference:
TOPS-10/TOPS-20 Batch Reference Manual
This manual provides information about writing and submitting
batch control files to the GALAXY batch processor.
TOPS-10 Operating System Commands Manual
vii
This manual provides information about monitor commands and
conventions.
TOPS-10 Operator's Command Language Reference Manual
This manual provides information about the OPR program and
operator-privileged commands.
TOPS-10 MACRO Reference Manual
This manual provides information required for making changes to
program source code.
TOPS-10 LINK Reference Manual
This manual provides information about TOPS-10's linking loader,
its switches, commands, and functions. LINK is used to build the
monitor after the software configuration has been specified.
TOPS-10 Monitor Calls Manual
This manual provides in-depth information about monitor calls and
monitor symbols that you may wish to modify in the monitor
sources.
TOPS-10/TOPS-20 RSX-20F System Reference Manual
This manual provides information about the KL system's front end
processor. Of particular importance are references to the KLINIT
dialogue and the PARSER commands.
TOPS-10/20 USAGE File Specification
This manual provides information about the formats of the files
created by the USAGE accounting system.
TOPS-10 ANF-10 Software Installation Guide
This manual provides information for building ANF-10 remote nodes.
|
| The procedures for installing a TOPS-10 system vary greatly between
| the first-time, new installation of a system, and the installation of
| a new version of the monitor on a system that has been running
| TOPS-10. This manual is organized into separate parts to address the
| specific needs of each case.
|
| Part 1, "Getting Started", provides two chapters that explain how to
| use the manual. Chapter 1 contains introductory material which will
| help you find the procedures that apply to your installation. Chapter
| 2 contains worksheets that are useful when building the TOPS-10
| monitor. DIGITAL Field Service personnel can help you fill out the
| hardware worksheet (Worksheet 1). As system manager you must provide
viii
| the information for the software worksheet (Worksheet 2) and for the
| disk parameters worksheet (Worksheet 3). Questions about software
| considerations can be answered by a DIGITAL Software Specialist.
|
| Part 2, "Preparing the System", consists of four chapters that
| describe the procedures TOPS-10 installers must complete before
| building the monitor on a new installation.
|
| Part 3, "Upgrading the System", contains two chapters that describe
| the procedures TOPS-10 installers must complete before upgrading the
| monitor on an existing system.
|
| Part 4, "Building the Monitor and Tailoring the System", provides four
| chapters that describe how to build your monitor and tailor it to the
| particular needs of your site. This part applies to both upgrades and
| new installations.
|
| Part 5, "Supplementary Information", contains 4 appendixes. The
| appendixes contain alternatives to the installation procedure and
| information for modifying the monitor software.
|
| A glossary is also included after the appendixes. The glossary
| defines the specialized terms used in this manual.
Prepare for the task of installing the monitor carefully. Read the
parts of this manual that you are not familiar with, and be sure to
set aside plenty of time to complete the installation. If you have
any problems with the information in this manual, please fill out the
Reader's Comment Card provided at the back of the book, and mail it to
Digital Equipment Corporation. This is an important form of feedback
for improving the quality of the documentation.
CONVENTIONS
The following conventions are used in this manual, according to the
definitions given here. Terms and file names are defined in the
Glossary in Appendix E.
Convention Meaning
? is used as a wildcard character in file
specifications. Any character in the filespec may
be replaced by a ?, thereby causing the reference
to include any files with the same characters;
except that any character will be accepted in
place of the ?, or no character may replace the ?.
For example, the filespec:
DSKB:MAN???.RPT[27,5434]
is satisfied by all of the following:
ix
DSKB:MANUAL.RPT[27,5434]
DSKB:MAN.RPT[27,5434]
DSKB:MANTM.RPT[27,5434]
* is also used as a wildcard character in file
specifications, but it replaces an entire field of
the specification; that is, the file name, file
extension, project number, programmer number, or
SFD name. Wildcard characters are described more
thoroughly in Chapter 1 of the TOPS-10 Operating
System Commands Manual.
<CTRL/x> represents a control-character. To enter a
control-character, hold down the CTRL key and
press the character represented by x.
<ESC> represents places where you press the ESCape key.
<RET> represents places where you press the RETURN key.
underlined text represents user input.
hh:mm:ss represents time on a 24-hour clock, where hh =
hours, mm = minutes, ss = seconds.
mmm-dd-yy represent a date, where mmm = month, dd =
or dd-mmm-yy day, yy = year.
V#(##) represent software version numbers.
VA##-##
VE##-##
x
xi
PART 1: GETTING STARTED
This part contains two introductory chapters which
explain how to use this manual. Chapter 1
contains introductory material which will help you
find the procedures that apply to the installation
you are responsible for. Chapter 2 contains
worksheets that are useful when building the
TOPS-10 monitor.
CHAPTER 1
INTRODUCTION
Before you install the system, read the cover letter distributed with
| the software package and the BEWARE file (MONITR.BWR) that is included
| on the Monitor/ANF-10 tape. They contain information about last
minute changes to installation procedures that could not be included
in this manual.
This manual uses the following conventions to denote the different
procedures for KL and KS systems:
o Sections appropriate to only one type of processor include
the processor type in the title.
o Within sections, specific references are made in the text to
each type of processor.
1.1 READING PATH FOR NEW INSTALLATIONS
The new system is assumed to have no data on the formatted disks.
Therefore, you must initialize the system and the system files. To
install the new system, first read and follow all of the instructions
in Part 1, Chapters 1 and 2. Then, read and follow all of the
instructions in each chapter of Part 2 that are appropriate for the
type of processor (KL or KS) that you are installing. Next, go on to
Part 4 and follow the procedures necessary to build your monitor and
tailor it to the needs of your particular site. Below is a summary of
the chapters in Parts 1, 2, and 4 that are appropriate for a new
installation.
PART 1: GETTING STARTED
Chapter 1: INTRODUCTION
This chapter will help you find the procedures that apply to the
installation you are currently responsible for.
1-1
INTRODUCTION
Chapter 2: PREPARATION
Chapter 2 describes how to check the hardware to be sure that it
was installed properly and is ready for software installation.
It contains worksheets to help you prepare the information that
you will need during installation.
PART 2: PREPARING THE SYSTEM
Chapter 3: BOOTING THE SYSTEM
This chapter describes how to load the bootstrap monitor from
tape and boot the system. For KL systems, this chapter describes
booting the front end processor as well. With this initial
system, you can build the TOPS-10 monitor.
Chapter 4: STARTING THE MONITOR
This chapter describes the ONCE dialogue, and describes how to
start the bootstrap monitor on a new installation.
Chapter 5: BUILDING THE MONITOR FILE SYSTEM
This chapter describes how to copy the monitor sources, CUSPs,
GALAXY system, and unbundled software from the distribution tapes
into disk areas that you prepare for this purpose. It also
describes how to create the system's administrative control files
and system program initialization files.
Chapter 6: BUILDING THE SUPPLEMENTARY PROCESSOR'S FILE SYSTEM
This chapter describes how to copy the front end processor (for
KL systems) or the microprocessor (for KS systems) from the
distribution media to disk, and how to boot the front end from
disk with the new software.
PART 4: BUILDING YOUR MONITOR AND TAILORING THE SYSTEM
Chapter 9: BUILDING YOUR MONITOR
This chapter describes how to tailor the TOPS-10 monitor using
MONGEN. MONGEN is used to create new monitor configuration
files. You can also assemble and load the new software, and
reload the system to start the new version of the monitor.
Chapter 10: STARTING AND TAILORING THE GALAXY SYSTEM
This chapter describes how to start up the GALAXY spoolers and
servers, how to tailor the GALAXY batch and spooling system to
the particular needs of the installation, and how to assemble and
1-2
INTRODUCTION
load the new GALAXY system.
Chapter 11: MAINTAINING THE SYSTEM AUTHORIZATION FILES
This chapter describes how to use the REACT program to create the
system accounting files. The system accounting files specify
PPNs and disk usage authorization for users.
Chapter 12: CREATING SYSTEM INITIALIZATION FILES
This chapter describes the INITIA program, and how to use INITIA
to start up TOPS-10 system components automatically when the
system is reloaded.
1.2 READING PATH FOR EXISTING INSTALLATIONS
If you are installing a new version of the monitor on a system that
has been running TOPS-10, you will probably use disks with system
information on them. You need not create system accounting or
initialization files and you will not need to boot the bootstrap
monitor from tape. However, it is necessary to restore files from
tape exactly as documented in these procedures.
Before building a new monitor for the existing installation, you
should save the existing monitor sources, CUSPs, GALAXY components,
and (for KL systems) front end file system on a backup disk or tape.
This will allow you to fall back to your old monitor if the new
monitor is difficult to install or run.
To install the new monitor, first read and follow all of the
instructions in Part 1, Chapters 1 and 2. Then, read and follow all
of the instructions in each chapter of Part 3 that are appropriate for
the type of processor (KL or KS) that you have. Next, go on to Part 4
and follow the procedures necessary to build your monitor and tailor
it to the needs of your particular site. Below is a summary of the
chapters in Parts 1, 3, and 4 that are appropriate for a system
upgrade.
PART 1: GETTING STARTED
Chapter 1: INTRODUCTION
This chapter will help you find the procedures that apply to the
installation you are currently responsible for.
Chapter 2: PREPARATION
Preparation for building the monitor is very important. Chapter
2 contains worksheets to help you prepare the information that
1-3
INTRODUCTION
you will need during installation. Experienced installers will
find the checklists and worksheets in this chapter useful.
PART 3: UPGRADING THE SYSTEM
Chapter 7: BUILDING THE MONITOR FILE SYSTEM
This chapter describes how to restore the monitor, CUSPs, and
unbundled software from the distribution tapes, and how to copy
them into the appropriate disk areas.
Chapter 8: RESTORING THE SUPPLEMENTARY PROCESSOR'S FILE SYSTEM
This chapter describes how to build the front end file system and
reload the front end (KL systems only) and how to create the
microprocessor file system (KS systems only).
PART 4: BUILDING THE MONITOR AND TAILORING THE SYSTEM
Chapter 9: BUILDING YOUR MONITOR
This chapter describes how to tailor the TOPS-10 monitor using
MONGEN. MONGEN is used to create new monitor configuration
files. You can also assemble and load the new software, and
reload the system to start the new version of the monitor.
Chapter 10: STARTING AND TAILORING THE GALAXY SYSTEM
This chapter describes how to start up the GALAXY spoolers and
servers, how to tailor the GALAXY batch and spooling system to
the particular needs of the installation, and how to assemble and
load the new GALAXY system.
Chapter 11: MAINTAINING THE SYSTEM AUTHORIZATION FILES
This chapter describes how to use the REACT program to create the
system accounting files. The system accounting files specify
PPNs and disk usage authorization for users.
Chapter 12: CREATING SYSTEM INITIALIZATION FILES
This chapter describes the INITIA program, and how to use INITIA
to start up TOPS-10 system components automatically when the
system is reloaded.
1.3 SUPPLEMENTARY INFORMATION
Supplementary information is provided in Part 5. This consists of
1-4
INTRODUCTION
four appendixes. The appendixes contain alternatives to the
installation procedure and information for modifying the monitor
software. The appendixes are summarized below:
Appendix A: ONCE STARTUP OPTIONS
This appendix describes the options to the ONCE program that
define disk characteristics and start up the monitor.
Appendix B: MODIFYING THE MONITOR
This appendix describes how to modify the monitor, build the
monitor library file by hand, and generate a monitor that is not
completely supported by Digital Equipment Corporation. It also
contains brief descriptions of the relevant symbols, source
modules, and symbol files.
Appendix C: SETTING SCHEDULER PARAMETERS
This appendix explains how to use the SCDSET program to set
system usage quotas, adjust scheduler parameters, update the
class of all logged in jobs, and manipulate a SCDMAP.SYS file.
Appendix D: MAKING THE DISTRIBUTION MEDIA BOOTABLE
This appendix describes the procedure for making the front end
media bootable for RSX-20F.
1-5
2-1
CHAPTER 2
PREPARATION
This chapter explains the steps and precautions to take before you
install the monitor on a new system or upgrade your existing monitor.
If you are installing a new system, a DIGITAL Field Service
Representative will install the hardware components of your system and
will run diagnostics to ensure that the hardware functions properly.
The materials you need to install a new system or upgrade an existing
system are:
o A supported hardware configuration. See Section 2.3.
To ensure that you have the hardware information that you
need to install the system, prepare Worksheet 1 in this
chapter. If you are installing a new system, the hardware
characteristics should be listed for you by the Field Service
Representative.
o A disk unit with plenty of free space.
At least 100,000 disk blocks are required for a system
upgrade. A newly formatted disk is required for a new
installation.
o Software distribution media: magnetic tapes, DECtapes, and
floppy diskettes.
Each type of processor requires software that is distributed
on a different set of media. The distribution package is
described in the following sections.
o A list of software components and options.
Use Worksheet 2 in this chapter to record this information.
o A list of disk parameters and search lists.
2-1
PREPARATION
Use Worksheet 3 in this chapter to record this information.
o System accounting files to record user accounts and access
privileges.
You can create these with the REACT program. REACT is
described in Chapter 11.
o A SYSJOB.INI file or OPR.ATO file to use for automatic system
startup.
| Chapter 12 explains how to create a SYSJOB.INI file. An
| OPR.ATO file can be used to automatically load and reload a
| communications front end by inclusion of the appropriate
| commands. For information on the OPR.ATO file, see the
| TOPS-10 Operator's Guide.
2.1 TOPS-10 SYSTEM INSTALLATION MATERIALS
The following magnetic tapes are provided in the software distribution
package:
|
| o The TOPS-10 Monitor/ANF-10 Tape
|
This magnetic tape contains the bundled monitor and ANF-10
software for TOPS-10. The tape is written at 1600
bits-per-inch density (BPI) by the BACKUP program. It
contains the following:
- Monitor source files
- TOPS-10 ANF-10 network files
|
| - MONITR.BWR file, which describes changes to the monitor.
o The TOPS-10 CUSP Tapes
"CUSP" stands for "Commonly Used System Program". A CUSP is
a system program that provides important functions for the
user. These tapes contain the latest versions of all of the
TOPS-10 monitor CUSPs and the files for the GALAXY batch and
spooling system. The tapes are written at 1600 BPI by the
BACKUP program. The CUSP tape contains a file named
CUSP.BWR, which describes changes to the programs on the CUSP
tape.
o The Customer-Supported Tape
This tape contains files that are supported by the customer
base, not by DIGITAL. This tape is also written at 1600 BPI.
2-2
PREPARATION
It contains the following files:
- customer-supported CUSP files
- customer-supported monitor modules
o The Tools Tape
This tape contains files that can be used for various
purposes, as supplied. The tools are provided "as is", and
may not be supported.
|
| o The Bootable Tape
This tape contains the TOPS-10 bootstrap monitor and related
| programs. The bootable tape is provided to allow you to
| start a standard operating system used to build a customized
monitor. The tape is written at 1600 BPI by the BACKUP
program.
o Unbundled Software Tapes
For each unbundled software product, magnetic tape(s) contain
the software necessary for running the software. Note that
unbundled options may require installation instructions that
are not included in this manual because the product contains
its own installation manual.
2.2 KL FRONT END INSTALLATION MATERIALS
The KL processor is also supplied with software for the RSX-20F front
end. The front end software is distributed on different types of
| media, depending upon the system. A KL10-D (DECsystem-1090) front end
| requires three DECtapes, labeled DT-A, DT-B, and DT-C. A KL10-E
| (DECsystem-1091) front end requires three floppy diskettes, labeled
| FLP-A, FLP-B, and FLP-C.
2.3 REQUIRED HARDWARE CONFIGURATIONS
The bootstrap monitor for TOPS-10 requires a minimal set of peripheral
hardware. The required configurations are different for the KL and KS
systems, and are specified below.
2.3.1 KL Hardware
For a KL processor, the minimally required configuration for the
2-3
PREPARATION
boostrap monitor is:
o One RH20 for disk, dual-ported between the front end and KL
processor
o One RH20 for tape
o CTY (console terminal)
2.3.2 KS Hardware
| For a KS 2020 processor, the minimally required configuration for the
| bootstrap monitor is:
o One RH11 (UBA#1) for disk
o One RH11 (UBA#3) for tape
o CTY (console terminal)
If you are installing a new system and have all the materials
available to begin installation, you are now ready to power up the
system; continue reading this chapter. If you are installing a new
version of the monitor on an existing system, go to Section 2.5.
2.4 POWERING UP THE SYSTEM
| Before powering up the system, perform the following steps:
|
| o For KL systems only, set the port controller select switch on
| the disk drive that is dual-ported between the KL10 and the
| front end to the A/B position.
|
| o Power up the disk drives.
|
| o Power up the magnetic tape drives.
|
| o Power up the line printers.
|
| o Turn on the CTY(s). Make sure each is loaded with paper and
| is on-line. Check the terminal setting for line speed. An
| LA36 terminal must be set to 300 baud. For an LA120, ask the
| Field Service Representative what the baud rate should be.
|
If a line printer, disk drive, or magtape drive is not powered up,
refer to the TOPS-10 Operator's Guide for powering-up procedures.
2-4
PREPARATION
2.4.1 Powering up the KL10
To power up the KL10, press the black switch on the upper control
panel to the position labeled POWER ON. When the system is powered
up, the light labeled POWER is lit. If you have never powered up the
system, refer to the TOPS-10 Operator's Guide for detailed
instructions.
2.4.2 Powering up the KS10
To power up the KS10, press the button labeled POWER. When the system
is powered up, the light labeled POWER is lit. This usually takes a
few seconds. 15 seconds after the power light comes on, the system
will try to load the microcode from the disk on drive 0. The message
BT AUTO will be printed on the CTY to indicate that this operation is
in progress. Because the microcode is not on the disk, repeated
attempts to load the microcode will be unsuccessful and the CTY will
continue to print BT AUTO every 15 seconds. Enter CTRL/C on the CTY
to stop the process of searching for the microcode on disk. When you
enter CTRL/C, the prompt KS10> will be printed on the CTY. Chapter 3
describes how to load microcode from tape.
Now that you have powered up the system, you can boot it as described
in Chapter 3. But first, prepare the information you will need by
following the instructions in the next section.
2.5 PREPARING INFORMATION
It is important to prepare the information that you will need during
installation. Because this information is lengthy, and may come from
several sources, worksheets are provided to record the information.
Use the completed worksheets as reference during installation and to
keep a record of the characteristics of the system.
2.5.1 Worksheet 1
If you are installing a new system, a Field Service Representative can
provide you with the information for Worksheet 1, Hardware
Configuration.
2.5.2 Worksheet 2
Use Worksheet 2 to record the software configuration. The software
configuration is largely the responsibility of the system manager.
2-5
PREPARATION
Information to make the necessary decisions is provided in this manual
and in other documents in the TOPS-10 Software Notebook Set.
Questions should be directed to a DIGITAL Software Specialist.
2.5.3 Worksheet 3
Use Worksheet 3 to record disk characteristics. The disk system must
be defined for the monitor. This is accomplished when you select the
DEFINE startup option for the ONCE dialogue (see Appendix A).
2-6
PREPARATION
Worksheet 1
|
| Hardware Worksheet
|
|
| System name (24 characters maximum): ______________________________
|
| CPU type (KL or KS): ______________
|
| Number of CPUs (KL only): _________
|
|
| CPU Serial Numbers
|
|
| CPU0: ____________________
|
| CPU1: ____________________
|
| CPU2: ____________________
|
|
| Total amount of system memory (KS 512K): ___________
|
| (KL 768K): ___________
|
| Clock ticks (60,50): ___________
|
|
2-7
PREPARATION
| Worksheet 1
|
|
| System Devices
|
|
| Answer YES or NO to the following questions.
|
|
| Does the system have any of the following:
|
|
| One or more CI-20 (SCA) subsystem interfaces? _____
|
| One or more NI-20 (Ethernet) subsystem interfaces? _____
|
| RP01/2/3 disks? _____
|
| Swapping drums? _____
|
| RS04/05 fixed-head disks? _____
|
| Any DX20/RP20 disk subsystems? _____
|
| CI-based disks (RA60/RA81)? _____
2-8
PREPARATION
| Worksheet 1
|
|
| System Devices cont'd
|
|
| Are any of the following kinds of magtape drives present:
|
|
| TM02/TM03 controller-based drives (TU45/TU77s)? _____
|
| DX10/TX01/TX02-based drives (TU70/71/72s)? _____
|
| DX20/TX02-based drives (TU70/71/72s)? _____
|
| TM78 controller-based (TU78/79s)? _____
|
| TM10B controller-based (TU40/41s)? _____
|
| TM10C controller-based (TU43)? _____
|
|
| Does the system have any of the following devices:
|
|
| Line printers connected to the I/O Bus? _____
|
| Line printers connected to the Console Front End? _____
|
| XY10 I/O bus plotters? _____
|
| CR10 I/O bus card readers? _____
|
| CD20 CFE-based card readers? _____
|
| CP10/CP10D I/O bus card punches? _____
|
| Paper-tape punches? _____
|
| Paper-tape readers? _____
2-9
PREPARATION
| Worksheet 1
|
|
| Front Ends
|
|
| Will any front ends be running: ANF-10? _____
|
| DN60 IBMCOMM? _____
|
| DECnet-10? _____
|
| How many terminal lines are there on CPU0? _____
|
| CPU1? _____
|
| CPU2? _____
|
| CPU3? _____
|
| CPU4? _____
|
| TTY number for OPR (usually CTY)? _____
|
| Which lines on your system are dataset lines? ________________
|
| ________________
|
| ________________
|
| ________________
2-10
PREPARATION
| Worksheet 2
|
|
| Software Worksheet
|
|
| Maximum number of jobs: _____________________
|
| Maximum amount of memory per job: _____________________
|
| Allow locking jobs? (Y or N): _____________________
|
| Minimum amount of memory for unlocked jobs: _____________________
|
| Number of real-time devices: _____________________
|
| Number of HPQs: ____________
|
| Number of PTYs: ____________
|
|
| Monitor Services
|
|
| Answer YES or NO to the following questions about user runtime
| accounting:
|
| Exclude monitor overhead? _____
|
| Use EBOX/MBOX clocks? (KL only) _____
|
| Include PI time? (KL only) _____
|
| Account verification? _____
|
| Include MSGSER? _____ PSISER? _____
|
| IPCF? _____ ENQ/DEQ? _____
2-11
PREPARATION
| Worksheet 3
|
|
| Disk Parameters Worksheet
|
|
| For each file structure, note the following:
|
|
| Structure Name Units in Structure
|
| ________________ __________________
|
| __________________
|
| __________________
|
|
| Consecutive blocks tried on output: _______________
|
| Overdraw blocks allowed per user: _______________
|
| Private structure (YES or NO): _____
|
| Owner PPN: _____________
|
| Amount of space for crash file: ___________K
|
| Blocks per cluster: ____________ Bits per cluster count: ____________
|
|
| Units
|
|
| For each unit, note the following:
|
|
| SAT Blocks
| Unit Name Unit-id SAT Blocks in Core Swap Space
|
|
| _________ _______ __________ __________ _________K
|
| _________ _______ __________ __________ _________K
|
| _________ _______ __________ __________ _________K
PART 2: PREPARING THE SYSTEM
This part is made up of four chapters that
describe the preparatory procedures that must be
accomplished before building the monitor and
tailoring the system. If you are upgrading an
existing system, you should go to Part 3 now.
CHAPTER 3
BOOTING THE SYSTEM
This chapter describes how to load the bootstrap monitor from tape and
how to boot the system. The procedure for booting a KL system is
described in Section 3.1. The procedure for booting a KS system is
described in Section 3.2.
3.1 BOOTING THE KL SYSTEM
Before booting the KL system, you must mount the required media on the
drives, load the front end, initialize the front end, and load the
bootstrap monitor. If you are unfamiliar with the procedure for
mounting software media, refer to the TOPS-10 Operator's Guide for
instructions on mounting DECtapes, floppy diskettes, disk packs, and
magtapes.
The monitor, supporting CUSPS, and unbundled software are distributed
on magnetic tape. The front end software is distributed on DECtapes
or floppy diskettes. Perform the following steps to mount the
distribution media and load the front end:
Step Explanation
1. Mount distribution media
For DECtapes: Mount the DECtape labeled DT-A on
Drive 0. Mount the DECtape labeled
DT-B on Drive 1.
For floppy diskettes: Mount the floppy labeled FLP-A in the
left-hand drive unit, DX0:. Mount
the floppy labeled FLP-B in the
right-hand drive unit, DX1:.
|
| Magnetic tape: Check to see if there is a write ring
| in the reel, and remove it before
| mounting the bootable tape on Drive
3-1
BOOTING THE SYSTEM
| 0.
|
Disk pack: You must also mount a disk pack that
the system can access. Mount a
properly formatted disk pack on RPmx,
where 'm' is the disk unit's
controller letter and 'x' is the unit
number. The drive must be
dual-ported between the front end and
the KL processor. The drive RPA0 is
used throughout this manual.
2. Set the data switches on To do this, raise the switches marked
the front end control panel 0, 1, and 2 on the switch register to
to 000007. the up position. The data switches
are off when down, on when up.
3. Check control switches The control switches to the right of
the data switches should all be
raised. This data switch setting
causes the front end to run the CPU
initialization dialogue (KLINIT) for
loading and configuring CPU memory.
4. Hold the ENABLE switch on, When you press these switches, the
press the SW/REG switch, front end software is loaded from
then release both. DECtape or floppy and the
initialization dialogue starts.
3.1.1 Intializing the KL Front End
At the CTY, type the responses underlined in the dialogue shown below
and press RETURN after typing each response. You cannot type ahead;
you must wait for the system to respond after you press RETURN. The
dialogue shown here is reproduced from a DECsystem-1091, KL10-E,
system startup. Note that the messages printed by the system will
differ slightly for DECsystem-1090, KL10-D, systems, but you should
type the same answers to the questions, regardless of the type of
system you are installing.
The KL initialization dialogue (KLINIT) starts automatically,
prompting for commands with KLI>. Enter the responses that are
underlined in the following example dialogue. Refer to the
TOPS-10/TOPS-20 RSX-20F Reference Manual for a more thorough
description of the KLINIT dialogue.
RSX-20F VE##-## hh:mm mmm-dd-yy
[SY0: REDIRECTED TO DX0:]
3-2
BOOTING THE SYSTEM
[DX0: MOUNTED]
[DX1: MOUNTED]
KLI -- VERSION VA##-## RUNNING
KLI -- ENTER DIALOGUE [NO,YES,EXIT,BOOT]?
KLI>YES<RET>
KLI -- KL10 S/N: 1042., MODEL B, 60 HERTZ
KLI -- KL10 HARDWARE ENVIRONMENT:
MCA25 CACHE PAGER
MOS MASTER OSCILLATOR
EXTENDED ADDRESSING
INTERNAL CHANNELS
CACHE
|
| The following question applies only to installations with an MCA25
| cache pager. If you have an MCA25 cache pager, answer this question
| with BOTH and press RETURN.
|
| KLI -- SELECT PAGE TABLE [FILE,BOTH,0,1]
| KLI>BOTH<RET>
KLI -- PAGE TABLE SELECTED: BOTH
KLI -- RELOAD MICROCODE [YES,VERIFY,FIX,NO]?
KLI>YES<RET>
KLI -- MICROCODE VERSION ### LOADED
KLI -- RECONFIGURE CACHE [FILE,ALL,YES,NO]?
KLI>ALL<RET>
KLI -- ALL CACHES ENABLED
KLI -- CONFIGURE KL MEMORY [FILE,ALL,REVERSE,FORCE,YES,NO]?
KLI>ALL<RET>
LOGICAL MEMORY CONFIGURATION
ADDRESS SIZE INT TYPE CONTROLLER
000000000 768K 4 MF20 11
KLI -- LOAD KL BOOTSTRAP [FILE,YES,NO,FILENAME]?
KLI>YES<RET>
KLI -- WRITE CONFIGURATION FILE[YES,NO]?
KLI>NO<RET>
KLI --BOOTSTRAP LOADED AND STARTED
| BOOT V#(##)
|
| BOOT>
3.1.2 Booting the KL Monitor
| At the BOOT> prompt, type the file specification of the monitor you
| are loading, followed by a switch to specify the tape drive controller
name, and press RETURN. The file specification is optional,
defaulting to the following:
DSKB:SYSTEM.EXE[1,4]/switch
3-3
BOOTING THE SYSTEM
Where '/switch' identifies the tape controller for the tape drive.
One of the following switches may be used:
/TM02 or /TM03 for TU45 and TU77 tape drives. For drives other
than Unit 0, specify /TM02:u or /TM03:u, where 'u'
is the unit number.
/TM10 for TU10, TU40, and TU41 tape drives.
/TX01 or /TX02 for TU70, TU71, TU72 tape drives on a DX10
channel.
/DX20 for TU70, TU71, or TU72 on a DX20 channel.
| /TM78:nx for TU78/79 tape drives, where 'n' is a single
| digit for the TM78's MASSBUS unit number and 'x'
| is a single digit for the slave unit number.
|
[Loading from DSKB:SYSTEM.EXE [1,4]]
When the system begins running, the system name is printed on the CTY:
KL10 mmm-dd-yy
The ONCE dialogue starts automatically when the monitor starts,
allowing you to give the monitor the information it requires to run.
Continue the installation procedure with Chapter 4, "Starting the
Bootstrap Monitor". If ONCE fails to start, repeat the procedure from
Step 2 in Section 3.1.
3.2 BOOTING THE KS SYSTEM
Before booting the KS system, you must mount the required media on the
drives. The monitor, supporting CUSPS, and unbundled software are
distributed on magnetic tape. If you are unfamiliar with the
procedure for mounting software media, refer to the TOPS-10 Operator's
Guide for instructions on mounting disk packs and magtapes.
| First, you will mount the bootable tape. Check to see if there is a
| write ring in the reel. If present, remove it and mount the bootable
| tape on tape drive 0. You may use a drive other than MTA0. Next,
| mount a formatted disk pack on disk drive 0.
Start the bootstrapping procedure by typing MT to the KS10> prompt on
| the CTY. The system responds by printing the BOOT> prompt on the CTY.
3-4
BOOTING THE SYSTEM
| KS10>MT<RET>
| KS10>USR MOD
|
| BOOT V#(##)
|
| BOOT>
|
| At the BOOT> prompt, type the file specification of the bootstrap
| monitor and a switch that specifies the tape drive controller, then
press RETURN. The file specification is optional, defaulting to the
following:
DSKB:SYSTEM.EXE[1,4]/switch
where '/switch' identifies the controller for the tape drive. The
following switches may be used:
/TM02 or /TM03 for TU45 and TU77 tape drives.
| BOOT starts the loading procedure.
|
| [Loading from DSKB:SYSTEM.EXE [1,4]]
|
| When the system begins running, the system name is printed on the CTY:
|
| KS10 mmm-dd-yy
|
| The ONCE dialogue starts automatically when the monitor starts,
| allowing you to give the monitor the information it requires to run.
| Continue the installation procedure with Chapter 4, "Starting the
| Bootstrap Monitor". If ONCE fails to start, repeat the procedure from
| Section 3.2.
3-5
4-1
CHAPTER 4
STARTING THE BOOTSTRAP MONITOR
This chapter describes how to start up the bootstrap monitor using
standard and default settings. The ONCE program always runs at
monitor startup and offers several options to the way it can run. For
complete information about ONCE, refer to Appendix A. This chapter
contains only the information and procedures needed to start the
bootstrap monitor on a new installation using default disk parameters.
|
| ONCE allows the user to:
|
| o Keep a record of the reason for starting or reloading the
| system.
|
| o Provide the date and time for the monitor.
|
| o Define, change, and show structure and unit parameters.
|
| o Define, change, and show the system lists -- the system
| search list, the active swapping list, and the system dump
| list.
|
| o Start the monitor.
4.1 THE ONCE DIALOGUE
ONCE begins by printing, on the CTY, the system name and the date that
the monitor was created. The system name can be changed when you run
| the MONGEN dialogue described in Chapter 9. After printing the system
| name and date, ONCE scans memory to be sure that it is accessible.
| Each monitor is built for a specific amount of memory. When you build
| your monitor using MONGEN, you will specify the amount of memory
| needed for your installation. Since you are installing a new system,
| you may receive the following message, indicating that the default
| monitor was built to expect more memory than that which is currently
| on-line:
4-1
| %Memory from xxxxxx to yyyyyy is off-line
| Set memory (DOWN,OFFLIN,ONLINE) [OFFLIN]:
|
| For a new installation, you should type DOWN and press RETURN.
|
| If you are starting a multi-processor KL system (SMP) but have not yet
| started the other CPUs, ONCE prints the following message:
|
| %CPU #nnnn is not running. Set CPU status (DOWN,UP) [DOWN]:
|
If you have an SMP system, you can start the CPU indicated by typing
UP and pressing RETURN. ONCE then asks for the date by printing:
Date:
Enter the date in one of the following forms:
MMM-DD-YY
or
DD-MMM-YY
Where MMM is replaced by the name of the month (JANUARY through
DECEMBER) abbreviated to the first three characters of the month name
(JAN through DEC), DD is the numerical day of the month (01 through
31) and YY is the year in a two digit year designation. Both the year
and the hyphens are optional. After you enter the date and press
RETURN, ONCE asks for the time:
Time:
Enter the time in one of the following forms:
HH:MM:SS
or
HHMMSS
Where HH is replaced by the hour, based on a 24-hour clock, MM is
replaced by the minutes past the hour, and SS is replaced by the
seconds past the hour, with optional colons separating the hours,
minutes, and seconds. Enter the time and press RETURN.
| It is very important that you respond with the correct date and time.
| If you do not, users' files may be written with the wrong creation
| dates, resulting in confusion and errors. If you enter the wrong date
| or time, you may correct the error later by selecting the CHANGE
| startup option followed by the DATE keyword. The CHANGE startup
| option is described in Appendix A.
|
| After receiving the time and date, ONCE asks the question:
|
| Why reload:
|
| This question is asked whenever the system is reloaded. For the
| purpose of installing a new system, you should answer this question
| with NEW, and press RETURN.
4-2
| If you press RETURN without typing an answer, ONCE prints a list of
| the possible responses and their meanings. If you type HELP and press
| RETURN, you will be provided with a list of the possible responses and
| a brief explanation of each. These are also listed and described in
| Appendix A, Section A.1.
|
| After this initial dialogue, ONCE prompts you for a startup option by
| printing:
|
| Startup option:
|
| If you type HELP and press return, ONCE lists the startup options and
| switches and a brief explanation of each. These are also listed and
| described in Appendix A, Section A.1. This chapter describes how to
| start the monitor quickly using default values. If you prefer to
| define file structures to suit the particular needs of your
| installation, refer to Appendix A for more information.
|
|
|
| 4.2 BRINGING UP A NEW SYSTEM
|
| The DESTROY option of the ONCE dialogue is a powerful procedure that
| ensures that the disk parameters are properly defined for the monitor.
| When the DESTROY option is used, the disks are refreshed, deleting all
| files on the disk. The HOME and SAT blocks are initialized, deleting
| all information about the disk's place in system lists and file
| structures.
|
| If used with disks that already contain data or structure definitions,
| the data and structure information will be DESTROYed and rewritten
| with default parameters. For this reason, the ONCE program ensures
| that you want to perform the DESTROY option by printing a warning
| message before accepting the DESTROY startup option. This warning is
| intended to prevent accidental destruction of disk data.
|
| However, when you are booting the monitor on a new installation with
| unused disk units and with no disk parameters set or defined, or if
| your disk units contain unknown information, you may find this option
| useful. Refer to Appendix A, Table A-1, for a list of the DESTROY
| option parameters.
|
| The DEFINE option provides reasonable default values for disk
| parameters, allows you to create file structures that are useable by
| the monitor, and to construct system lists from the units that are on
| line.
|
| To start the new installation quickly, follow the procedures in this
| section. To DEFINE structure and unit parameters to suit the needs of
| your particular installation now, refer to Appendix A for complete
| information about the DEFINE startup option.
|
| So far, you have provided a reason for the reload (NEW), have set the
| date (in the format MMM-DD-YY or DD-MMM-YY) and have set the time (in
| the format HH:MM:SS or HHMMSS). Now, you are prompted for a startup
4-3
| option:
|
| Startup option:
|
| Enter DESTROY and press RETURN. ONCE will respond with a warning
| message and will ask whether or not you want to proceed. This warning
| message helps to prevent accidental use of the DESTROY option. If
| your disks have usable data, or if your disk parameters are already
| defined, do not proceed. If you type NO or press RETURN in response
| to this question, ONCE will prompt you for another startup option.
|
| %WARNING: All structures will be refreshed. All disk files
| will be destroyed. All information on currently spinning packs
| will be lost. Do not proceed unless you are positive that you
| want to do this.
|
| Proceed (NO,YES) [NO]:
|
| Type YES and press RETURN. When you do this the parameters will be
| set and the disks will be refreshed. When ONCE is finished with that
| operation, it prints the following message:
|
| [HOME blocks initialized on all units]
|
| You are then prompted for a startup option. Enter DEFINE and press
| return:
|
| Startup option:DEFINE<RET>
| Define (ASL,SDL,SSL,STRUCT):STRUCT<RET>
|
| In the following dialogue, all numbers are decimal unless
| otherwise indicated. Values within square brackets are defaults
| which will be substituted if RETURN is pressed.
|
| Any question preceded with an asterisk indicates changing that
| parameter will require the structure to be refreshed.
|
|
| Note
| HOME blocks will be initialized immediately after all units in
| the structure have been entered.
| It is advisable that you define at least one structure to be DSKB.
| All examples in this manual use DSKB as the area to which files are
| copied.
|
| Structure to define:DSKB<RET>
|
| Disk drive for logical unit 0:RPxn<RET>
| Disk drive for logical unit 1:RPxn<RET>
| Do you want to see the bad regions (NO,YES) [NO]:NO<RET>
|
| [HOME blocks initialized on RPxn]
| Initialize BAT blocks (NO,YES) [NO]:YES<RET>
|
| [BAT blocks initialized on RPxn]
4-4
| In the following section, default values are enclosed in square
| brackets. Default values are calculated by ONCE according to the
| structure type. Accept the default value by pressing RETURN.
|
| Parameters for structure DSKB
| *Blocks per cluster (3 - 262143) [10]:
| Private structure (NO,YES) [NO]:
| Owner PPN [NONE]:
| Disk-set (1-36,ALL,NONE) [ALL]:
| *Number of K for CRASH.EXE (0 - 4096) [4096]:
| Blocks allowed for overdraw per user (0 - 307800) [500]:
| Consecutive blocks tried for on output 90 - 262143) [30]:
| *Bits per cluster count (0 - 18) [12]:
| Make this the only structure in the SSL (NO,YES) [YES]:
| Make this the only structure in the SDL (NO,YES) [YES]:
|
| In the following section, first specify a unit identification. This
| identification resides in the HOME blocks on the disk unit, for unique
| identification of the disk pack. Therefore, it should be a unique
| number that will never conflict with another code. Every pack should
| have a unit-id of six alphanumeric characters that may be a portion of
| the pack's serial number.
|
| Parameters for unit RPxn
| Unit ID [XXXXn]:
|
| For all other questions, accept the default value by pressing RETURN.
|
| *SAT blocks per unit (8 - 81) [8]:
| SAT blocks in core (1 - 8) [8]:
| *K for swapping (0 - 4807) [4807]:
| *First logical block for swapping (33090 - 269332) [134672]:
| Swapping class (0 - 1) [0]:
| Make this the only unit in the ASL (NO,YES) [YES]:
|
| Structure to define:
|
| Enter another structure to define or press RETURN and you will be
| prompted for another startup option. To start the monitor now, enter
| NOINITIA and press RETURN:
|
| Startup option:NOINITIA<RET>
|
| [Rebuilding the system search list from the HOM blocks]
|
| [Rebuilding the active swapping list from the HOM blocks]
|
| [Rebuilding the system dump list from the HOM blocks]
|
| [HOM blocks updated]
|
| To automatically log in under [1,2] type "LOGIN"
|
| TOPS-10 7.04 System Friday mmm-dd-yy hh:mm:ss
4-5
| .
4-6
CHAPTER 5
BUILDING THE MONITOR FILE SYSTEM
Before you can build your monitor you must build the monitor file
| system. To do this, you will restore files from the bootable tape,
| which you mounted before running ONCE, and from the Monitor/ANF-10
| tape, which contains the sources for the monitor and ANF-10. You will
| restore files from the CUSP, or Commonly Used System Program tape.
| The files for the GALAXY batch and spooling system are included on
| this tape.
|
This chapter also explains how to restore files from the unbundled
software tape, the customer-supported tape, and the TOOLS tape. All
procedures apply to both KL and KS systems unless otherwise noted in
the section title.
| Before you can restore files from the bootable tape, you must assign a
name to the tape. You mounted this tape when you booted the system,
before running ONCE. The tape drive is identified to the monitor for
your job using the following monitor command. On the CTY, type the
following:
.ASSIGN MTA0: TAPE:<RET>
The system responds with:
|
| MTA0 assigned
|
| This command assigns the name TAPE to the tape unit MTA0. If the
| bootable tape is on a different tape drive, use the correct physical
device name.
Next, if you are installing a KL system, Section 5.1 will show you how
to restore the file that is used to allocate space on disk for the
front end file system.
Then, for both KL and KS systems, Section 5.2 will show you how to
restore the BACKUP program, which will be used to copy the bootstrap
monitor from tape to disk. If you are installing a KS system, skip
Section 5.1 and go directly to Section 5.2.
5-1
5.1 ALLOCATING DISK SPACE FOR THE KL FRONT END
This section describes how to restore a file that is used to allocate
space for the front end file system. On the CTY, type the following:
.GET TAPE:FEFILE<RET>
The system will respond with:
Job setup
.
NOTE
This command may result in an error message of the
form:
?Transmission error
If a transmission error occurs, reposition the tape to
make another attempt to obtain the FEFILE program.
Push the rewind button on the tape drive and wait for
the tape to rewind. Then, type the GET command twice
as shown below:
.GET TAPE:FEFILE<RET>
Job setup
.GET TAPE:FEFILE<RET>
Job setup
.
Now that the FEFILE job is set up, you will allocate space on a disk
unit for the front end file system. At the . prompt, type START and
press RETURN:
.START<RET>
The system will respond with:
DISK UNIT NAME:
Enter a disk unit name and press RETURN. The following example
assumes that the front end file system will be stored on the disk unit
| RPA0. You may specify any physical disk unit for RPA0, but the disk
| must be dual-ported between the front end and the KL, and must be an
| RP06 drive type.
|
DISK UNIT NAME:RPA0:<RET>
The system will respond with:
SIZE OF FILE IN BLOCKS (<CRLF> GIVES DEFAULT OF 2000):
Press RETURN to accept the default file size. There may be a delay
before the front end completes this operation. When the operation is
5-2
complete, the system will respond with:
[2008 DATA BLOCKS ALLOCATED TO FE.SYS]
[FE.SYS AREA STARTS AT LOGICAL BLOCK 2301.]
[FRONT END FILE CREATED, HOM BLOCKS WRITTEN]
Next, you will restore the BACKUP program.
5.2 RESTORING BACKUP FROM TAPE
This section shows you how to restore the BACKUP program. You will
| use BACKUP to copy the bootstrap monitor from tape to disk and to
| restore the files from the Monitor/ANF-10 tape, CUSP tape,
| customer-supported tape, any unbundled software tapes, and the TOOLS
| tape.
The following example shows how to restore, save, and run BACKUP.
Enter the commands that are underscored in the example:
.GET TAPE:BACKUP<RET>
Job setup
.SAVE BACKUP<RET>
BACKUP saved
.RUN BACKUP<RET>
/
The / sign is the BACKUP prompt, indicating that BACKUP is running and
is ready to accept BACKUP commands. Next you will restore the
bootstrap monitor. When that is completed, you will restore the
monitor sources and associated software. The BACKUP program is
discussed in the TOPS-10 Operator's Guide. This manual shows only the
commands you need to type to restore the monitor and its supporting
software.
5.3 RESTORING FILES FROM TAPE TO DISK
The examples in this manual use DSKB as the structure to which the
files will be copied, and the area [10,7] for the directory area. The
BACKUP program recognizes the ersatz device name DEC as SSL:[10,7],
therefore the procedures in this manual use the device name DEC to
represent DSKB:[10,7]. The area [1,4] is reserved for ersatz device
SYS, from which system programs are run. Do not substitute another
directory area for [1,4]. The procedures in this manual use the
device name SYS to represent DSKB:[1,4].
5.3.1 Restoring the Bootstrap Monitor
| Now, you will restore the bootstrap monitor from the bootable tape.
| For KL systems, type the following commands after the / prompt:
5-3
/TAPE TAPE:<RET>
/SUPERSEDE ALWAYS<RET>
/RESTORE SYS:=*.*<RET>
"DONE
/
For KS systems, type the following commands:
/TAPE TAPE:<RET>
/SUPERSEDE ALWAYS<RET>
/INTERCHANGE<RET>
/RESTORE SYS:=*.*<RET>
"DONE
/RESTORE DEC:=*.*<RET>
"DONE
| Unload the bootable tape. To do this, for both KL and KS systems,
| type the following command:
|
| /UNLOAD TAPE:<RET>
| [MTA0: Read(C/H/S) = 7202000/0/0]
|
This command rewinds the tape and unloads it. Remove the tape from
the drive.
The next section describes how to use BACKUP to restore the files from
| the Monitor/ANF-10 tape, from any unbundled software tapes, from the
CUSP tape, the customer-supported tape, and the TOOLS tape.
|
|
|
| 5.3.2 Restoring the Monitor Sources and Associated Software
|
To copy the files from tape to disk, you must mount, restore, unload
and dismount each tape. You may restore the tapes in any order,
unless otherwise noted in the cover letter included with the specific
tape.
Repeat the procedure below for each distribution tape, starting with
| the Monitor/ANF-10 tape.
1. Mount the distribution tape.
2. Type the following commands to restore the files:
/TAPE TAPE:<RET>
/REWIND<RET>
/SUPERSEDE ALWAYS<RET>
/PROTECTION 155<RET>
/SSNAME ALL<RET>
/RESTORE DSKB:<RET>
!10,7 DSKB
"DONE
/UNLOAD TAPE:<RET>
[MTA0: Read(C/H/S) = 7752000/0/0]
5-4
3. Dismount the tape.
After the files are on disk, exit from BACKUP by typing EXIT and
pressing RETURN. Now you can create the TOPS-10 disk file system. To
do this, you will create the system directories then copy the
appropriate files into each directory.
5.4 CREATING THE TOPS-10 SYSTEM DIRECTORIES
The TOPS-10 monitor requires certain disk directories from which it
can access support programs and CUSPs. You will create these
directories on disk, then copy the files from the DSKB:[10,7] area to
those directories. The monitor recognizes the ersatz device name DEC
as DSKB:[10,7]. Therefore, the device name DEC is used in the
following procedures.
The program that creates directories is CREDIR. The CREDIR program is
described in the TOPS-10 User Utilities Manual. Only the commands you
need to complete this procedure are demonstrated below.
In the following procedure, the ersatz device names recognized by the
monitor are used to create the appropriate directories for storing
monitor files. Each ersatz device name results in a directory area
with the standard PPN that the monitor reserves for that purpose. A
directory for the specified PPN will be created on each file structure
in the system search list, unless you specify otherwise.
The ersatz device names used in this manual are defined as follows:
PPN Name Use
|
| [10,7] DEC: DEC-supplied software
|
[1,4] SYS: System library
[2,5] HLP: HELP files
[5,14] DOC: DOC files
[5,11] REL: REL files
[5,17] UNV: MACRO universal files
[1,7] ACT: Accounting daemon files (ACTDAE)
Use the following commands to create the ersatz device directories.
The directory areas will be created on each file structure in the
system search list. CREDIR will report on each directory it creates
on each file structure. In this example, DSKB is the only file
structure in the system search list. Type the commands that are
underscored in the following examples:
.RUN DEC:CREDIR<RET>
5-5
CREDIR will prompt you for the directory name you wish to create:
Create directory:HLP:<RET> Created DSKB0:[2,5].UFD/PROTECTION:775
Repeat for each directory name:
Create directory:DOC:<RET>
Created DSKB0:[5,14].UFD/PROTECTION:775
Create directory:REL:<RET>
Created DSKB0:[5,11].UFD/PROTECTION:775
Create directory:UNV:<RET>
Created DSKB0:[5,17].UFD/PROTECTION:775
Create directory:ACT:<RET>
Created DSKB0:[1,7].UFD/PROTECTION:775
To exit from CREDIR, type CTRL/C:
Create directory:<CTRL/C>
.
5.5 COPYING THE FILES INTO DIRECTORIES
This section shows you how to copy the files from DEC into the
| appropriate directories. You will copy the files to the appropriate
| areas from DEC, using the NFT (Network File Transfer) program. NFT is
| described in the TOPS-10 User Utilities Manual.
|
| The NFT prompt is an *. To run NFT, type the following command:
|
| .RUN DSKB:NFT[10,7,NFT]<RET>
| *
|
| To copy files you only need to use the COPY command as shown below.
| Use the COPY command to copy the files into their appropriate areas
| from DEC:
|
| *COPY SYS:/PROTECTION:055=DEC:*.EXE<RET>
|
| *COPY SYS:/PROTECTION:055=DEC:*.RAM,DEC:*.VFU<RET>
|
| *COPY SYS:/PROTECTION:055=DEC:DDT.REL,DEC:JOBDAT.REL<RET>
|
| *COPY SYS:/PROTECTION:055=DEC:OVRLAY.REL,DEC:TECO.ERR<RET>
|
| *COPY HLP:/PROTECTION:055=DEC:*.HLP<RET>
|
| *COPY DOC:/PROTECTION:055=DEC:*.DOC<RET>
5-6
| *COPY REL:/PROTECTION:055=DEC:*.REL<RET>
|
| *COPY UNV:/PROTECTION:055=DEC:*.UNV<RET>
|
| *COPY SYS:/PROTECTION:055=DEC:SYSTEM.CMD<RET>
|
| *COPY SYS:/PROTECTION:055=DEC:*.SYS<RET>
|
| *COPY SYS:/PROTECTION:055=DEC:*.INI<RET>
|
| For KL systems, also copy the following files as shown:
|
| *COPY SYS:=DEC:DXMCA.ADX,DEC:DXMPA.A8,DEC:DXMCD.ADX<RET>
|
| *COPY SYS:=DEC:KNICOD.BIN,DEC:FEUIC.TXT<RET>
|
| For KS systems, copy the following files as shown:
|
| *COPY SYS:=DEC:SMFILE.EXE<RET>
|
| To exit from NFT, type EXIT and press RETURN:
|
| *EXIT<RET>
| .
|
If you have not yet read the beware files, read them now. To find out
what .BWR files are available, type the following:
.DIR *.BWR[10,7,*,*,*,*,*]
You can display the files on your terminal using the TYPE command. Be
sure to read all the files listed by the DIR command.
Next, if you are installing a KL system, you will create the file
system for the front end. If you are installing a KS system, you will
create the file system for the microprocessor. Chapter 6 describes
these procedures.
5-7
6-1
CHAPTER 6
BUILDING THE SUPPLEMENTARY PROCESSOR'S FILE SYSTEM
The KL and KS processors are each supported by another processor. The
KL processor uses a PDP-11 front end for I/O, which requires space on
disk for its software. The KS system uses an 8080 microprocessor that
also requires disk space for storage. This chapter explains the
procedure for creating these disk file systems. If you are installing
a KL system, continue reading with Section 6.1. If you are installing
a KS system, go directly to Section 6.2.
6.1 BUILDING THE KL FRONT END FILE SYSTEM
To build the front end file system, you must load the front end from
the front end media, which you mounted according to the instructions
in Chapter 3. The front end software is distributed on either
DECtapes or floppy diskettes. The commands that you will type to the
RSX-20F command language processor, the PARSER, differ primarily in
the physical device names used. The examples used to illustrate your
communication with the PARSER, if not specifically separated by
appropriate headings, apply to both DECtapes and floppies. In all
examples, the commands that you will type are underscored.
The following procedures assume that the front end file system will be
stored on DSKB, mounted on RPA0. In fact, you can use any structure
by replacing DB0 in the commands with DBn, where 'n' is the unit
number in RPxn. Make sure that the disk unit is write-enabled and is
dual-ported between the front end and the KL.
KL systems may have more than one CPU by implementing Symmetric
Multi-Processing (SMP). In a multiple-CPU system, each CPU has a
front end, and each front end requires its own file system. If you
are installing an SMP system, you must follow the steps in Section
6.1.1 and 6.1.2 for each CPU, at each CTY.
For DECtapes:
First, check to make sure that the DECtapes are mounted correctly.
6-1
The DECtape labelled DT-A should be mounted on DECtape drive 0, DT0:,
and the unit selector switch should be set to 0. The DECtape labelled
DT-B should be mounted on DECtape drive 1, DT1:, and the unit selector
switch should be set to 1. On both DT0: and DT1:, set the
LOCAL/REMOTE switch to REMOTE and the READ/WRITE switch to WRITE
ENABLE.
For floppies:
First, check to make sure the floppy diskettes are mounted correctly.
The floppy diskette labelled FLP-A should be mounted in the left-hand
floppy drive DX0:, and the floppy diskette labelled FLP-B should be
mounted in the right-hand floppy drive DX1:.
6.1.1 Loading the Front End
Invoke the PARSER by typing CTRL/\ (control-backslash) on the CTY.
This control character is not echoed, but the PARSER responds with the
prompt PAR>, indicating that it is ready to accept commands.
NOTE
If you make an error at any time during your dialogue
with the PARSER, type CTRL/Z followed by CTRL/\. The
PARSER prompt will be printed, and you can retype the
command.
If you are loading RSX-20F from DECtape, allow at
least 45 minutes to copy the files.
Type the SET CONSOLE MAINTENANCE and RESET commands to ensure that the
KL is not running.
<CTRL/\>
PAR>SET CONSOLE MAINTENANCE<RET>
CONSOLE MODE: MAINTENANCE
PAR>RESET<RET>
PAR#
Set the data switches on the front end control panel to 000003, with
the first two switches (0 and 1) raised and the rest off.
Hold the ENABLE switch on and, at the same time, press the SW/REG
switch. The front end will print its software name and date and the
location of the front end software.
For DECtapes: For floppies:
RSX-20F VA##-## hh:mm dd-mmm-yy RSX-20F VA##-## hh:mm dd-mmm-yy
[SY0: REDIRECTED TO DT0:] [SY0: REDIRECTED TO DX0:]
6-2
[DT0: MOUNTED] [DX0: MOUNTED]
[DT1: MOUNTED] [DX1: MOUNTED]
Initialize the front end by running the INIalization program. The INI
program does not prompt when initialization is complete. When
initialization is complete, the lights on the drive panel stop
flashing. First, invoke the parser by typing CTRL/\, and type the
following commands:
<CTRL/\>
PAR#MCR INI<RET>
INI>DB0:<RET>
INI><CTRL/Z>
Mount the disk unit on which the front end file system will reside.
Run the MOUnt program by typing the following commands:
<CTRL/\>
PAR#MCR MOU<RET>
MOU>DB0:<RET>
MOU -- MOUNT COMPLETED
MOU><CTRL/Z>
Create the front end file system directory. Run the UFD program by
typing the following commands:
<CTRL/\>
PAR#MCR UFD<RET>
UFD>DB0:[5,5]<RET>
UFD><CTRL/Z>
The UFD program does not prompt the CTY when it is finished. When the
lights on the front panel stop flashing, the operation is complete.
If you are loading the front end from DECtapes, you will copy the
*.TSK files into directories before copying the rest of the files. If
you are loading the front end from floppies, you will copy all of the
files now. Run the PIP program to copy the files from DECtape or
floppy to disk. Exit from PIP by typing CTRL/Z. Type the following
commands:
For DECtapes: For floppies:
<CTRL/\> <CTRL/\>
PAR#MCR PIP<RET> PAR#MCR PIP<RET>
PIP>DB0:=DT0:F11ACP.TSK<RET> PIP>DB0:=DX0:,DX1<RET>
PIP>DB0:=DT0:PARSER.TSK<RET>
PIP>DB0:=DT1:PIP.TSK<RET>
PIP><CTRL/Z> PIP><CTRL/Z>
6-3
Redirect the system. Run the REDirect program to redirect the system
search area (SY0) from DECtape or floppy to disk. Type the following
commands:
<CTRL/\>
PAR#MCR RED<RET>
RED>DB0:=SY0:<RET>
RED><CTRL/Z>
If you are loading the front end from DECtape, you will copy the rest
of the files from DECtape to disk. This may take 25 minutes or more
to complete. When this has been completed, list the directory of all
the files in SY0 and exit from PIP. If you are loading the front end
from floppies, you have already copied the files and only need to list
the directory. Type the following commands:
For DECtapes: For floppies:
<CTRL/\> <CTRL/\>
PAR#MCR PIP<RET> PAR#MCR PIP<RET>
PIP>DB0:=DT0:,DT1:<RET>
PIP>DB0:/LI<RET> PIP>DB0:/LI<RET>
PIP><CTRL/Z> PIP><CTRL/Z>
Write front end monitor to SY0. Run the SAVe program to write the
front end monitor to SY0. It is already defined as DB0:[5,5]. The
DECtapes or floppies and disk are logically dismounted automatically.
PIP will print warning messages about the files that already exist,
but you may ignore these messages. Type the following commands:
For DECtapes: For floppies:
<CTRL/\> <CTRL/\>
PAR#SET CONSOLE OPERATOR<RET> PAR#SET CONSOLE OPERATOR<RET>
PAR#MCR SAV<RET> PAR#MCR SAV<RET>
SAV>SY0:/WB<RET> SAV>SY0:/WB<RET>
[DB0: DISMOUNTED] [DB0: DISMOUNTED]
[DT0: DISMOUNTED] [DX0: DISMOUNTED]
[DT1: DISMOUNTED] [DX1: DISMOUNTED]
For DECtapes:
Remove the DECtapes from drives 0 and 1, and mount the DECtape
labelled DT-C on drive DT0:.
For floppies:
Remove the floppies from DX0: and DX1: and mount the floppy labelled
FLP-C in drive DX0:.
Type the following commands:
6-4
For DECtapes: For floppies:
<CTRL/\> <CTRL/\>
PAR#MCR MOU<RET> PAR#MCR MOU<RET>
MOU>DT0:<RET> MOU>DX0:<RET>
MOU -- MOUNT COMPLETE MOU -- MOUNT COMPLETE
MOU><CTRL/Z> MOU><CTRL/Z>
Copy the files to disk. Run PIP to copy the files from the third
DECtape or floppy to disk and list the directory of files on DB0.
Then, exit from PIP, dismount the DECtape or floppy and remove the
media from the drive. Type the following commands:
For DECtapes: For floppies:
<CTRL/\> <CTRL/\>
PAR#MCR PIP<RET> PAR#MCR PIP<RET>
PIP>DB0:=DT0:<RET> PIP>DB0:=DX0:<RET>
PIP>TT:=DB0:/LI<RET> PIP>TT:=DB0:/LI<RET>
PIP><CTRL/Z> PIP><CTRL/Z>
<CTRL/\> <CTRL/\>
PAR#MCR DMO<RET> PAR#MCR DMO<RET>
DMO>DT0:<RET> DMO>DX0:<RET>
DMO -- DISMOUNT COMPLETED DMO -- DISMOUNT COMPLETED
DMO><CTRL/Z> DMO><CTRL/Z>
You can now boot the RSX-20F front end from the disk. This procedure
is described in Section 6.2
6.1.2 Booting the KL Front End from Disk
To initiate the load, hold the ENABLE switch on and press the DISK
switch at the same time. These switches are located on the drive
panel. Following is the output when the front end is loaded:
RSX-20F VA##-## hh:mm dd-mmm-yy
[SY0: REDIRECTED TO DB0:]
[DB0: MOUNTED]
KLI -- VERSION VA##-## RUNNING
KLI -- MICROCODE ### LOADED
KLI -- ALL CACHE ENABLED
LOGICAL MEMORY CONFIGURATION
ADDRESS SIZE INT TYPE CONTROLLER
00000000 768K 4 MF20 11
KLI -- CONFIGURATION FILE WRITTEN
KLI -- BOOTSTRAP LOADED AND STARTED
6-5
BOOT V#(##)
[Loading from DSKB:SYSTEM.EXE [1,4]]
BOOT>
You load the monitor by typing commands to the BOOT program. To load
the bootstrap monitor just press RETURN to the BOOT prompt. The
default file name is DSKB:SYSTEM.EXE[1,4].
The ONCE dialogue begins automatically. The startup options are
described in Appendix A. For a quick startup, use the GO startup
option.
If you are installing a multiple-processor KL, repeat the procedures
in Section 6.1 at the console for each CPU.
Section 6.2 describes the procedure for installing the microprocessor
file system on KS systems. If you are installing a KL system, choose
a ONCE startup option then go directly to Chapter 7.
6.2 CREATING THE KS MICROPROCESSOR FILE SYSTEM
This section describes the procedure for creating the file system for
the KS microprocessor, and for booting the monitor.
6.2.1 Creating a Front end File System Directory Area
If you are installing a new KS system, you must create a directory
area on disk for the microprocessor file system. This area must be
[6,2020], and it is recommended that you use the structure that
resides on RPA0. Do not use a multiple-unit file structure for the
microprocessor file system.
To create the directory area, type the following commands:
.R CREDIR<RET>
Create directory:DSKB:[6,2020]/EXIT<RET>
Created DSKB0:[6,2020].UFD/PROTECTION:775
.
6.2.2 Writing the Bootstrap Program to Disk
Next, you will write the bootstrap program to disk. To do this, run
the SMFILE program as shown below. You must supply the serial number
for your system's CPU. Note that the following example uses a CPU
serial number of 4097.
.RUN DEC:SMFILE<RET>
6-6
DECSYSTEM 2020 DIAGNOSTICS FE-FILE PROGRAM
VERSION #.#, TOPS-10, KS10, CPU =4097
[FOR HELP TYPE "HELP"]
|
| SMFILE>WRITE SETUP DSKB:<RET>
| SMFILE>WRITE RESET<RET>
| SMFILE>READ DEC:T10KL.RAM<RET>
| SMFILE>SERIAL 4097<RET>
| SMFILE>WRITE CRAM<RET>
| SMFILE>WRITE BOOT DEC:KSBOOT.EXE<RET>
| SMFILE>WRITE DONE<RET>
| SMFILE>EXIT<RET>
|
EXIT
.
The KS is now ready to run. Reload the monitor using the following
commands:
<CTRL/\>
KS10>HA<RET>
|
| When the CPU is halted, press the RESET buttons on the CPU panel.
|
| KS10 CSL.V#.#
| BT AUTO
| BOOT V#(#)
|
| BOOT><RET>
|
| [Loading from DSKB:SYSTEM.EXE [1,4]]
|
| You can load the monitor by typing commands to the BOOT program. To
| load the default monitor just press RETURN to the BOOT prompt. The
| default file name is DSKB:SYSTEM.EXE[1,4]. To load a different
| monitor, type the file name of the monitor to the BOOT prompt, taking
| care to include the structure name and the complete directory
| specification.
|
| The ONCE dialogue begins automatically. The startup options are
| described in Appendix A. For a quick startup, use the GO startup
| option. After you choose a startup option, go on to Chapter 9 for
| information about building your monitor.
6-7
PART 3: UPGRADING AN EXISTING SYSTEM
This part contains two chapters which describe how
to prepare your system for upgrade to a new
version of the monitor. If you are installing a
new system, go directly to Part 4, "Building the
Monitor".
CHAPTER 7
RESTORING THE MONITOR DISTRIBUTION TAPES
| Before you can customize the monitor for your installation you must
| copy the TOPS-10 monitor and CUSPs from the distribution tapes to
| disk. The monitor sources and associated software must be restored
| manually. The files on the DECnet, CUSP and TOOLS tape may be
| restored automatically using the BACKUP indirect command files
| supplied with each tape.
|
| Existing KS installations require some of the files from the bootable
| tape. These files are not compatible with previous versions and are
| required for building the microprocessor file system. If you are
| upgrading a KS system, you will copy these files from the bootable
| tape before copying the TOPS-10 monitor and CUSPs. If you are
| upgrading a KS system, mount the bootable tape now. If you are
| upgrading a KL system, mount the Monitor/ANF-10 tape.
|
| For existing systems already running GALAXY, ensure that the
| tape-drive you are using is not available to MDA by using the OPR
| command SET TAPE-DRIVE. MTA0 is used in the following example; if the
| distribution tape is on a different drive, use the correct physical
| device name.
|
| .R OPR<RET>
| OPR>SET TAPE-DRIVE MTA0 UNAVAILABLE
| Enter text and terminate with ^Z<CTRL/Z>
| ^Z
| OPR>
| 10:29:36 Device MTA0 -- Unavailable for use --
| OPR>EXIT<RET>
| .
|
For both KL and KS systems, assign a name to the tape by typing the
following command:
.ASSIGN MTA0: TAPE:<RET>
|
| The system responds with:
7-1
RESTORING THE MONITOR DISTRIBUTION TAPES
| MTA0 ASSIGNED
|
This command assigns the name TAPE to the tape unit MTA0.
The procedures for copying files from tape to disk use the BACKUP
program. At the . prompt, type the following:
.R BACKUP<RET>
/
The / sign is the BACKUP prompt, indicating that BACKUP is running and
is ready to accept BACKUP commands. The BACKUP program is discussed
in the TOPS-10 Operator's Guide. This manual shows only the commands
you need to type to restore the monitor and its supporting software.
If you are upgrading a KS system, continue with Section 7.1. If you
are upgrading a KL system, go directly to Section 7.2.
7.1 RESTORING KS MICROPROCESSOR FILES
The following commands allow you to restore the files needed to
upgrade a KS microprocessor file system. Type the commands that are
underscored in the following procedure:
| /TAPE TAPE:<RET>
| /SUPERSEDE ALWAYS<RET>
| /INTERCHANGE<RET>
| /RESTORE DEC:=SMFILE.EXE<RET>
| "DONE
| /RESTORE DEC:=T10KL.RAM<RET>
| "DONE
| /RESTORE DEC:=KSBOOT.EXE<RET>
| "DONE
| /UNLOAD TAPE:<RET>
| [MTA0: Read(C/H/S) = 7752000/0/0]
|
| Remove the bootable tape from the drive and mount the Monitor/ANF-10
| tape. Then, proceed with the instructions in Section 7.2.
7.2 RESTORING THE MONITOR SOURCES AND ASSOCIATED SOFTWARE
| The following procedure restores the monitor sources and associated
| software into DSKB:[10,7]. If you wish to restore the tape to an area
| other than DSKB:[10,7], you must specify this area in the /RESTORE
| command line as shown on the right in the example below. To restore
| the monitor sources and associated software, type the commands that
| are underscored:
7-2
RESTORING THE MONITOR DISTRIBUTION TAPES
| /TAPE TAPE:<RET>
| /REWIND<RET>
| /SUPERSEDE ALWAYS<RET>
| /PROTECTION 155<RET>
| /SSNAME ALL<RET>
| /RESTORE DSKB:<RET> or /RESTORE FOO:=DSKB:
!10,7 DSKB
"DONE
/UNLOAD TAPE:<RET>
| [MTA0: Read(C/H/S) = 7752000/0/0]
|
Remove the tape from the drive.
7.3 RESTORING TAPES USING COMMAND FILES
The DECnet tape, CUSP tape, and TOOLS tape each include a
corresponding BACKUP indirect command file. The command file is used
to restore the files directly into its appropriate UFD or SFD.
Each indirect command file is stored on the tape for which it can be
used:
o The first CUSP tape contains CUSPS.RST.
o The DECnet tape contains DECNET.RST.
o The TOOLS tape contains TOOLS.RST.
| When you use the supplied indirect command files, BACKUP does not
| store the files into the build area (ersatz device DEC) used in the
| manual procedure described Section in 5.4. It is very important that
| you check the commands in the indirect command file before you execute
| it, and adjust ersatz device references to apply to the directories
| used in your system.
|
| For each tape, you will:
|
| o Mount the tape and restore the indirect command file.
|
| o Check the ersatz device references contained in the indirect
| command file, and adjust them if necessary.
|
| o Execute the indirect command file.
|
The following example shows the commands to type to restore the
indirect command files from each tape.
/TAPE TAPE:<RET>
/REWIND<RET>
/INTERCHANGE<RET>
7-3
RESTORING THE MONITOR DISTRIBUTION TAPES
/SUPERSEDE ALWAYS<RET>
/PROTECTION 155<RET>
/SSNAME ALL<RET>
/RESTORE DSK:=DSKB:filename.RST<RET>
"DONE
/
Examine the ersatz device names in the indirect command file and check
them against existing directories to ensure that you will not
supersede necessary files.
Then, execute the indirect command file by typing the following
commands:
.R BACKUP<RET>
/@filename.RST<RET>
"DONE
/
7.4 USING SYSTEM FILES
The SYS: area includes several sample system files. The monitor uses
these to create user accounts and to initialize system programs.
After you complete the procedures in Chapter 8 and 9 you can tailor
these files to your installation.
o ACTDAE.SYS
This is the system's accounting file. It records the PPNs,
passwords, access information, and disk quotas for users and
operators. Without the ACTDAE.SYS file, no one (users or
operators) can log in.
o SYSJOB.INI
This file automatically starts system jobs.
o SYSTEM.CMD
This file can be read by OPR to automatically start batch and
spooling processes for GALAXY (refer to Chapter 10).
o TTY.INI
INITIA reads this file and initializes the CTY and TTY
STOMPER functions.
7-4
CHAPTER 8
RESTORING THE SUPPLEMENTARY PROCESSOR'S FILE SYSTEM
The KL and KS processors are each supported by another processor. The
KL processor uses a PDP-11 front end for I/O, which requires space on
disk for its software. The KS system uses an 8080 microprocessor that
also requires disk space for storage. Sections 8.1 and 8.2 explain
the procedure for restoring the front end file system for KL systems.
For KS systems, Section 8.2 explains the procedure for restoring the
bootstrap program to disk.
If you are upgrading a KL system, continue reading with Section 8.1.
If you are upgrading a KS system, go directly to Section 8.3.
|
|
|
| 8.1 THE DISTRIBUTED KL FRONT END SOFTWARE
|
| For existing installations, where modifications have been made to the
| front end software, you should copy the current front end file system
| from FE.SYS[1,4] to a backup disk area to resolve those modifications
| with the new version of the front end.
|
| The front end software for KL systems is distributed on the
| Monitor/ANF-10 tape. Additional installation tools are distributed on
| the CUSP tape. You must restore these tapes to disk, following the
| procedures in Chapter 7, before beginning the installation.
|
| If you have used DSKB:[10,7] as the distribution area on disk, the
| following Sub-File Directories (SFDs) will exist:
|
| DSKB:[10,7,RSX20F] Contains the distributions of RSX20F for both
| DECtape and floppy based systems. It also
| contains RSXINS.MIC, the master installation
| command file. RSXINS.MIC is described below.
|
| DSKB:[10,7,FELOAD] Contains FELOAD.BIN, a PDP-11 program which
| will be placed in the boot blocks of the disk
| containing the front end file system during
| the installation process.
8-1
| DSKB:[10,7,F11] Contains F11, a tool which is used to
| manipulate the front end file system. The
| installation process will use F11 to create
| and populate a new RSX20F file system.
|
| A Monitor Interpreted Commands (MIC) file, named RSXINS.MIC,
| distributed on DSKB:[10,7,RSX20F], performs all of the steps necessary
| to create a new FE.SYS file. It will:
|
| o Ask a set of questions about the installation, such as the
| target disk for the installation, where the distribution
| files are stored, and whether or not the console front end
| should be reloaded.
|
| o Use F11 to create and populate a new FE.SYS file on the
| target disk.
|
| o Use F11 to write the contents of FELOAD.BIN into the boot
| blocks of the target disk.
|
| o Run DTELDR to reload the front end with the new software.
| This is done if you answer YES to Question 7 of the
| installation dialogue. The installation dialogue is
| described in Section 8.2.1.
|
|
| The next section describes the RSXINS installation process in detail.
|
|
|
| 8.2 USING RSXINS
|
| Before running RSXINS, you must make sure of the following:
|
| o Your system must be running MIC version 11 or later.
|
| o The CUSP tape must be restored so that its subdirectory
| organization is preserved. This procedure is described in
| Chapter 7.
|
| o You must be logged in as [1,2]. The F11 program, which is
| invoked by RSXINS, uses Super USETI/USETO calls to do its
| work. These can only be executed by a job logged in as
| [1,2].
|
|
|
|
| 8.2.1 The RSXINS Dialogue
|
| A sample run of the RSXINS procedure is reproduced below. The
| dialogue questions are numbered for clarity in this text, but are not
| numbered in the dialogue itself. In the following example, RSXINS is
| used to install a new front end file on structure DSKE on a 1090
| (DECtape-based) KL system.
8-2
| RSXINS is invoked by the command:
|
| .DO DSKB:RSXINS[10,7,RSX20F] <RET>
|
| It will display the following:
|
| RSXINS -- RSX-20F Front-End File System Installation Procedure
|
| Installation started at 16:29:17 on Tuesday, 08-Mar-88
|
| This procedure will create, populate, and make bootable an
| RSX-20F front end file system on any TOPS-10 structure.
|
| In the following dialogue, default answers are enclosed within
| square brackets. If you would like further information about any
| question, answer it with a question mark (?). If you wish to
| return to the previous question asked, answer with a less-than
| sign (<).
|
| RSXINS checks to see if you are logged into [1,2]. If you are not, it
| will abort execution. Otherwise, it will begin a dialogue with you to
| obtain all of the installation information needed. In the dialogue,
| each question provides a default answer enclosed in square brackets
| ([]). To take the default answer, press the RETURN key.
|
| If you need help at any time with the answer to a particular question,
| you can respond with a question mark (?). RSXINS will then provide an
| explanation of the question and the appropriate type of reply it
| expects. Then, it will repeat the question.
|
| If you wish to go back to the previous question that was asked, you
| can type a less-than sign (<). RSXINS will repeat the previous
| question, using your previous reply as the default.
|
| RSXINS will first ask for the name of the target disk structure or
| unit:
|
| 1 * Which disk structure do you wish to put the FE.SYS area on
| [DSKB]? DSKE <RET>
|
| Respond to this question with either the four-character name of a disk
| structure or with the name of a specific disk unit (such as RPA0).
| Use the physical unit name if you are installing FE.SYS on a disk
| which is part of a multiple-unit structure.
|
| Next you will specify the location of the RSX20F distribution files.
| The area in which the executing RSXINS file is located is the default.
|
| 2 * Where is the distribution area of the front-end files located
| [default is DSKB:[10,7,RSX20F]]? <RET>
|
| Specify the location of FELOAD, the new PDP-11 bootstrap program. The
| default answer will be the FELOAD subdirectory on the same disk and
| PPN where the executing RSXINS file is located.
8-3
| 3 * Where is the distribution of FELOAD located [default
| DSKB:[10,7,FELOAD]]? <RET>
|
| Specify the location of F11, a tool used by the procedure to
| manipulate FE.SYS files. The default answer will be the F11
| subdirectory on the same disk and PPN where the executing RSXINS file
| is located.
|
| 4 * Where is the F11 program distribution area located [default
| DSKB:[10,7,F11]]? <RET>
|
| Your answer to the next question determines which variant of RSX20F
| will be installed on your system. The valid responses for this
| question are either F or D. If F is specified, the floppy-based
| version of RSX20F (VE##-##) will be installed; if a D is specified,
| the DECtape-based version (VA##-##) will be installed. The default is
| D, indicating a DECtape-based system.
|
| 5 * Is this for a Floppy (F) or DECtape (D) based system [D]? <RET>
|
| RSXINS will then ask you how large you wish to make the FE.SYS file.
| The default size is 2000 blocks. The FE.SYS file is allocated
| contiguously on the disk. If the new FE.SYS file is larger than the
| previous one, or if there is no previous file, the new file will be
| created beginning at a place on the disk that is:
|
| o On a block that is a multiple of 400 (decimal); and
|
| o In an area where the disk space is available contiguously.
|
| If the requested number of blocks for FE.SYS is not available with a
| beginning position on a multiple of 400 blocks, then the procedure
| will fail. In that case, it will be necessary to back up the disk,
| refresh it, run RSXINS to create the FE.SYS file, and then restore the
| rest of the disk's files.
|
| In addition, if the FE.SYS file is created on a disk which has user
| files on it, there can be a delay of up to 20 seconds between the
| messages "[SY0: redirected to DB0:]" and "[DB0: mounted]" when
| RSX20F is reloaded. This delay is caused by the search algorithm that
| RSX20F performs when it is looking for the FE.SYS area. It examines
| each block that is a multiple of 400 looking for a Files-11 HOM block.
| If the FE.SYS file is nearer to the logical 'end' of the disk rather
| than the 'beginning', the search can take up to 20 seconds. This
| behavior is normal.
|
| 6 * How many blocks should the new FE.SYS file be [2000]? <RET>
|
| RSXINS will then ask if the console front end is currently using the
| target structure. If you accept the default answer YES, then RSXINS
| will run DTELDR to reload the console FE using the target disk. If
| you answer NO, it will only perform the software installation.
|
| 7 * Is the target structure currently in use by RSX-20F [Y]? <RET>
8-4
| If you answered NO to this question, please skip to Question 9.
|
| If you answered YES, RSXINS will then display a cautionary message:
|
| ** Note ** It is important that the console FE running on
| the target structure (DSKE:) not access the FE.SYS area
| during the course of the installation, otherwise the
| front-end system could crash. This in itself is not a
| problem, because RSXINS will reload the FE after the new
| file system is built. However, if the KL10 CPU were to
| crash during the installation procedure before the update
| process was completed, then the front end would not be
| rebootable from the RP06 disk. It would be necessary to
| boot the front end from console media (floppies or
| DECtapes), after which you could then reload the Monitor.
|
| While the likelihood of this happening is quite minimal, it
| is possible. There are steps that you can take to minimize
| this risk further, such as:
|
|
| * Make sure that any unit-record devices connected to the
| front end (such as line printers or card readers) are
| idle during the update. If they are not idle and an
| error occurs, RSX20F could potentially log the error to
| disk and then invoke LOGXFR to transfer the log to the
| KL.
|
| * Avoid running this procedure if the time is within 15
| minutes of midnight. This prevents the possibility that
| RSX20F will attempt to run the MIDNIT task. If the time
| is close, refrain from answering the next prompt until
| it is 12:01 AM according to the system clock.
|
| * Avoid running this procedure if the Monitor is unstable
| or there is a high incidence of hardware failures. A
| system crash in the middle of the procedure would leave
| the FE.SYS file in a partially-updated state. The
| front-end would then have to be reloaded from console
| media.
|
|
| After this text is displayed it will ask you if you wish to see the
| message again. The default answer is NO; a YES will simply retype the
| text.
|
| If you are running a single-processor KL system, please skip down to
| Question 9. If you are running an SMP system, answer the following
| question with the CPU number whose console front end area you are
| installing. The default is for RSXINS to reload the front end on CPU
| 0.
|
| 8 * Which CPU's console front-end do you wish to reload [0]? <RET>
|
| All the parameters needed by RSXINS have now been entered. A summary
8-5
| of the selected parameters is displayed and you are asked if the
| summary is correct:
|
| Here is a summary of the options you have selected for
| this installation.
|
| Target disk structure for installation: DSKE
| Location of RSX20F distribution area: DSKB:[10,7,RSX20F]
| Location of FELOAD distribution area: DSKB:[10,7,FELOAD]
| Location of F11 distribution area: DSKB:[10,7,F11]
| Installation type (D=DECtape, F=Floppy): D
| Size of FE.SYS file to be created: 2000
| Structure currently in use by RSX20F: Y
| Console front end to be reloaded on CPU: 0
| FE.SYS currently exists on DSKE: Y
|
| Please indicate if these choices are in error. If you
| answer the following question with a NO, the entire RSXINS
| dialogue will be restarted.
|
| 9 * Are the above correct (Y or N) [No default] ? Y
|
| If you respond with N, then RSXINS will restart the dialogue. Your
| original responses will become the new defaults for each question. If
| you respond with Y, RSXINS will proceed to install the new front end
| software.
|
|
|
| 8.2.2 Installing the Front End File System
|
| RSXINS has asked all of its questions. It will now begin the
| installation process. It will display the following message:
|
| All questions have been asked. The procedure will now verify
| that required SFDs and files exist.
|
| RSXINS will check to see if the FELOAD, F11, and RSX20F subdirectories
| exist in the places you have specified. If it does not find one of
| them, it will abort the installation. If all files are located, you
| will see the following message:
|
| All files have been located. Commencing actual installation
| procedure.
|
| Now it will check to see if an FE.SYS file already exists on the
| target structure/unit. If an FE.SYS file does exist, RSXINS will run
| F11 and delete the file from the target disk:
|
| Deleting existing DSKE:FE.SYS[1,4] file
|
| .run F11RUN:f11
|
| F11> dest DSKE:
| [DSKE will be treated as RPC0]
8-6
| Do you really want to DESTROY RPC0:, YES or NO? yes
| [RPC0:FE.SYS[1,4] deleted, 2005. blocks freed]
| F11> exit
|
| RSXINS will now create and populate a new FE.SYS file. RSXINS will:
|
| o Initialize the FE.SYS area, allocating the number of blocks
| you requested in the dialogue.
|
| o Create a [5,5] UIC directory in the file.
|
| o Copy the file CORIMG.SYS, containing the bootable image of
| RSX20F, to the [0,0] directory, replacing the placeholder
| file created when the area was initialized.
|
| o Copy the rest of the RSX20F files to the [5,5] directory
| area.
|
| o Copy FELOAD.BIN to the boot blocks (blocks 1 and 4 through 7)
| of the disk unit.
|
| The output from RSXINS will look like this:
|
| Init front-end file, copy files to it, make structure bootable...
|
| MIC REVIVE
|
| .RUN F11RUN:F11
|
| F11> INI DSKE:/SIZE:2000
| [DSKE will be treated as RPC0]
| [Created RPC0:FE.SYS[1,4], 2005. blocks allocated]
| F11> UFD DSKE:[5,5]
| [DSKE will be treated as RPC0]
| +-Directories created:
| RPC0:[0,0]005005.DIR;1<164000=RWED,RWED,RWE,R>/ALLOCATE:1024.
| F11> DEL DSKE:[0,0]CORIMG.SYS
| [DSKE will be treated as RPC0]
| Files deleted:
| RPC0:[0,0]CORIMG.SYS;1
| 100. blocks freed
| F11> PUT DSKE:[0,0]/FILE:5=DSKB:[10,7,RSX20F,DECTAP]CORIMG.SYS
| [DSKE will be treated as RPC0]
| +-RPC0:[0,0]CORIMG.SYS;1=DSKB:CORIMG.SYS[10,7,RSX20F,DECTAP]
| F11> PUT DSKE:[5,5]=DSKB:[10,7,RSX20F,DECTAP].
| [DSKE will be treated as RPC0]
|
| F11 will output a line for each file it copies from the distribution
| area to the FE.SYS file. A total of 42 files will be copied during
| this step.
|
| RSXINS will next copy FELOAD.BIN from the distribution area to the
| boot blocks of the disk:
|
| F11> BOOT DSKE:=DSKB:[10,7,FELOAD]FELOAD.BIN
8-7
| [DSKE will be treated as RPC0]
| Loading 0 THRU 122
| Loading 124 THRU 661
| Loading 662 THRU 775
| Loading 4000 THRU 4312
| Loading 4314 THRU 4446
| Loading 4450 THRU 5102
| Loading 5104 THRU 5264
| Loading 5266 THRU 5630
| Loading 5632 THRU 6367
| Loading 6370 THRU 6416
| Loading 6420 THRU 7155
| Loading 7156 THRU 7322
| Loading 7324 THRU 7402
| Loading 7404 THRU 7407
|
| Xfer address = 4106
| Highest address loaded = 7407
| F11> EXIT
|
| If you answered NO to Question 7, RSXINS will finish at this point.
| If you answered YES, RSXINS will use DTELDR to reload the console
| front end.
|
|
|
| 8.2.3 Automatic Front End Reload
|
| RSXINS will run DTELDR to reload the console front end which is using
| the target disk. If this is a single-processor KL system, this will
| automatically be the front end 0 on CPU 0. If this is an SMP system,
| the console front end on the CPU specified in response to Question 8
| will be reloaded, completing the installation procedure for RSX20F:
|
| Reloading front-end 0...
|
|
| .R DTELDR
|
| */RELOAD:00/SWITCH:000205/NODUMP
| " front end 00 Loading from -11 disk.
| " front end 00 initializing
|
| " front end 00 started
| */EXIT
|
|
| RSXINS procedure finished at 16:34:06 on Tuesday, 08-Mar-88
|
|
| If you are installing a multiple-processor KL system, repeat the
| procedures in Section 8.1-8.2 for each CPU.
|
| Section 8.3 describes the procedure for writing the bootstrap program
| to disk for KS systems. If you are installing or upgrading a KL
8-8
| system, go on to Chapter 9.
|
|
|
| 8.3 WRITING THE KS BOOTSTRAP PROGRAM TO DISK
|
| It is not necessary to rebuild the KS microprocessor's file system for
| an upgrade to an existing system. However, before you can boot the
| monitor, you must write the bootstrap program to disk.
|
| To do this, run the SMFILE program as shown below. You restored this
| program from the bootable tape in Chapter 7. You must use the version
| of SMFILE shipped with TOPS-10 Version 7.04 and you must supply the
| serial number for your system's CPU. Note that the following example
| uses a CPU serial number of 4097.
|
| .RUN DEC:SMFILE<RET>
|
| DECSYSTEM 2020 DIAGNOSTICS FE-FILE PROGRAM
| VERSION #.#, TOPS-10, KS10, CPU =4097
| [FOR HELP TYPE "HELP"]
|
| SMFILE>WRITE SETUP DSKB:<RET>
| SMFILE>WRITE RESET<RET>
| SMFILE>READ DEC:T10KL.RAM<RET>
| SMFILE>SERIAL 4097<RET>
| SMFILE>WRITE CRAM<RET>
| SMFILE>WRITE BOOT KSBOOT.EXE<RET>
| SMFILE>WRITE DONE<RET>
| SMFILE>EXIT<RET>
|
EXIT
.
The KS is now ready to run. Reload the monitor using the following
commands:
<CTRL/\>
KS10>HA<RET>
When the CPU is halted, press the RESET buttons on the CPU panel.
KS10 CSL.V#.#
BT AUTO
BOOT V#(#)
BOOT><RET>
[Loading from DSKB:SYSTEM.EXE [1,4]]
You can load the monitor by typing commands to the BOOT program. To
load the default monitor just press RETURN to the BOOT prompt. The
default file name is DSKB:SYSTEM.EXE[1,4]. To load a different
monitor, type the file name of the monitor to the BOOT prompt, taking
8-9
care to include the structure name and the complete directory
specification.
The ONCE dialogue begins automatically. The startup options are
described in Appendix A. For a quick startup, use the GO startup
option.
8-10
PART 4: BUILDING THE MONITOR AND TAILORING THE SYSTEM
This part is made up of four chapters that
describe how to build your monitor and tailor the
system to the particular needs of your site. All
system installations require the procedures in
Part 4.
CHAPTER 9
BUILDING YOUR MONITOR
Even though you have a running monitor, the bootstrap monitor is only
a tool for configuring a monitor for the system. When you build the
monitor, you configure support for the peripheral devices, memory
units, terminal and network lines, and supply information that the
monitor needs to run CUSPs successfully. You will build the monitor
using the MONitor GENerator (MONGEN) program, assemble the monitor
with MACRO, and load and save the monitor with LINK. Then you will
bring up your new monitor and create a bootable monitor tape. Later
sections describe how to automate the build procedure and customize
BOOT.
This chapter describes the following steps to build the monitor:
1. Run MONGEN.
Run MONGEN and answer the questions in the dialogue section.
MONGEN creates symbol definition source files, containing
information obtained from your answers to the dialogue
questions. You use the source files to build your monitor.
MONGEN creates the following source files:
|
| o SYSCNF.MAC contains the symbol definitions for the
| hardware and software configuration.
|
| o F.MAC contains feature test switches and their settings.
|
| If you change the setting of any of the feature test switches
| in the FGEN dialogue, or if you add any feature test
| switches, you must then reassemble the bundled monitor
| modules to create a new monitor library file, TOPS10.REL.
| This procedure is described in Appendix B.
|
| 2. Assemble the source modules.
|
| The COMMON, COMDEV, and COMMOD modules of the monitor must be
| assembled with the symbol definitions recorded in the MONGEN
9-1
BUILDING YOUR MONITOR
| configuration file, SYSCNF.MAC.
3. Link and load the monitor.
The modules must be linked with the monitor library file,
TOPS10.REL, and with any unbundled software (such as FORTRAN
and DECNET) that your installation has chosen to run. These
are combined to form a runnable, executable monitor in a file
named SYSTEM.EXE.
9.1 MONGEN
MONGEN is a dialogue program that allows you to choose the software
modules and features that are most appropriate for your installation
and to define your hardware and software configuration. MONGEN builds
MACRO-10 source files. The MONGEN-created source files contain the
information needed to create an executable monitor that is specific to
your installation. Parameters that are assigned in these files
determine how the monitor data base is assembled.
Before running MONGEN, do the following:
1. Read this chapter to acquaint yourself with the current
version of the MONGEN dialogue.
2. Be thoroughly familiar with your system configuration.
3. Predetermine the answers you will provide to the questions in
the MONGEN dialogue. Worksheets 1 and 2, provided in Chapter
2, will help you determine your responses.
4. Obtain a copy of the product-specific installation guide for
each unbundled product you wish to use, and familiarize
yourself with any modifications to the following procedures
that might be necessary for installing unbundled software.
Special installation instructions for unbundled software
products are available in product-specific installation
guides.
The MONGEN dialogue is divided into the following sections, each
asking questions about a specific aspect of the system configuration:
| SYSGEN Defines the hardware and software configuration. This
| includes type and number of CPUs, processor
| characteristics, ports for data communications, tape
| and disk units, line printers, and other peripheral
| devices. SYSGEN defines the terminal configuration,
| including the OPR line number and the dataset line
| numbers. SYSGEN also offers some services of the
9-2
BUILDING YOUR MONITOR
| monitor that you may refuse or accept. SYSGEN is the
| longest portion of the MONGEN dialogue, and it creates
| the SYSCNF.MAC file.
FGEN Allows you to change the settings of feature test
options, thus enabling and disabling the software
features and symbol values that the monitor uses to
determine how to service users and how to perform
maintenance operations. FGEN creates the F.MAC file,
which is required for building the monitor. Note that
changing the default switch settings require that you
rebuild all the monitor library files, thus creating a
monitor that is not supported by DIGITAL. For more
information, refer to Appendix B.
|
| If you are building the monitor for the first time, answer all the
| questions in the SYSGEN dialogue first, then answer the questions in
| the FGEN dialogue. On subsequent occasions, select SYSGEN if you want
| to make changes to your harware or software configuration, and select
| FGEN if you want to change any feature test switch settings or add any
| new feature test switches.
9.1.1 Running MONGEN
You can run MONGEN from any disk area, and the source files that are
generated are stored in your job's default disk area (DSK) or any
other disk area you specify. Make sure that the disk area where the
monitor is to be built has at least 7000 free disk blocks. Clear the
directory area of any extraneous files to eliminate the possibility of
confusion when assembling the monitor.
|
| MONGEN has three modes of dialogue (SHORT, PROMPT, and LONG) to
| service users of different levels of experience. Throughout the
| dialogue, MONGEN questions take the form of the mode you choose.
| MONGEN questions are in the form:
|
| question (default,possible answers) [explanation]:
|
| An explanation of the three modes and the form the questions take in
| each mode, is provided below:
|
| Mode Effect
|
| SHORT MONGEN prints only the questions on your terminal.
| This mode assumes that you need no listing of
| default or possible answers, and no explanation of
| the question. MONGEN assumes that you are
| familiar with the choice of answers. SHORT mode
| is recommended only for experienced users.
9-3
BUILDING YOUR MONITOR
| PROMPT MONGEN prints each question, but adds (in
| parentheses) a choice of answers. The default
| value, if any, is printed first in the
| parenthesis, followed by the other answers you may
| type, separated by commas. Due to the nature of
| some of the questions, a default answer may not be
| offered. This is the default mode and it is used
| in this section to illustrate the dialogue.
|
| LONG MONGEN prints each question, the choice of answers
| (in parentheses), and adds an explanation of the
| question [in square brackets]. LONG mode is
| strongly recommended for all first-time users of
| this version of MONGEN.
|
| Each question, regardless of the mode chosen, ends with a colon. If
| you choose PROMPT or LONG mode, select your answer from the possible
| answers listed within parentheses and enter it after the colon. To
| accept the default answer, just press RETURN. If the question has no
| default, MONGEN prints an error message and repeats the question. The
| possible answers are listed in one of these forms:
|
| (d,m-n) A range of numbers is provided, where 'd' is the
| default value, 'm' is the smallest number and 'n'
| the largest number you may type. Choose a number
| in that range. Numbers are decimal unless
| specified otherwise in the question.
|
| (m-n) A range of numbers with no default value provided.
| You can specify multiple, consecutive terminal
| lines by answering with a range of numbers in the
| form (m-n).
|
| (YES,NO) Answer with YES or NO
|
| (a,b,c,...,z) Choose one of the values in the list.
|
| If you type an invalid or inappropriate response, MONGEN prints an
| error message and repeats the question. You may respond to the
| question again.
|
| To execute the version of MONGEN in DSK:[10,7], type the following
| command:
|
| .R SETSRC<RET>
|
| *CP [10,7,MON]<RET>
*<CTRL/C>
.RUN MONGEN<RET>
9-4
BUILDING YOUR MONITOR
MONGEN responds with:
|
| Write MONGEN.MIC (YES,NO):
|
| After you answer this question, MONGEN begins:
|
| MONGEN for 704 monitors
| MONGEN will generate any known TOPS-10 configuration. For
| information about which configurations are supported, please
| refer to the SPD.
| /HELP (PROMPT,SHORT,LONG):
|
| Since PROMPT is the default mode, MONGEN provides a list of possible
| answers in parentheses. If you now type /HELP, MONGEN provides the
| LONG version of this question:
|
| MONGEN for 704 monitors.
| MONGEN will generate any known TOPS-10 configuration. For
| information about which configurations are supported, please
| refer to the SPD.
| /HELP (PROMPT, SHORT, LONG)[
| All numbers are decimal unless stated otherwise
| Unique abbreviations are allowed everywhere to all questions
| /HELP reasks a question in the next longer mode
| /HELP:xxx sets the mode permanently. Modes are:
| SHORT is short
| PROMPT has choices in ()
| LONG has choices in () and an explanation in []]:
|
| As shown above, you may ask MONGEN for the next longer version of the
| current question by typing /HELP. If you wish to change modes for the
| remaining questions, you only need to type /HELP:mode, where 'mode'
| can be SHORT, PROMPT, or LONG. For example, if you choose SHORT or
| PROMPT mode and later want to change to LONG mode for the remaining
| questions, append the argument LONG to the switch /HELP, separating
| the two with a colon. For example, typing
|
| /HELP:LONG
|
| will set the mode to LONG for the remainder of the dialogue.
|
| To exit from MONGEN at any time, type <CTRL/Z>.
9.1.2 The MONGEN Dialogue Questions
This section lists the dialogue questions for each part of MONGEN. A
number precedes each MONGEN question. These numbers help you to
follow the dialogue more easily. They are not actually displayed when
you run MONGEN.
9-5
BUILDING YOUR MONITOR
NOTE
MONGEN asks some questions only for specific types of
systems; and asks others only if you answer a
preceding question in a certain way.
Before running MONGEN, read this chapter carefully. Decide the
answers for each question in the dialogue. Worksheets 1 and 2 are
provided in Chapter 2 to help you record your responses to the MONGEN
questions. If you like, you can create a batch control file to run
MONGEN, and answer the questions in the control file to save time.
The MONGEN control file is described in Section 9.5.
This section lists the MONGEN questions in PROMPT mode, which is the
default mode. Use LONG mode or type /HELP after a question to get
more information about each question.
|
| 1 Which GEN (SYS,F):
|
| Respond with SYS or F to indicate the section that you wish to run.
| If you respond with SYS, MONGEN prints the SYSGEN questions. When the
| SYSGEN dialogue is done, MONGEN asks Question 1 again, so that you can
| start the FGEN part of the MONGEN dialogue. The SYSGEN questions are
| listed in Section 9.1.2.1. The FGEN questions are listed in Section
| 9.1.2.2.
|
|
|
| 9.1.2.1 Running SYSGEN -
|
| SYSGEN asks questions about the hardware and software of your system.
| This includes the central processor(s), storage devices, and other
| peripheral devices.
|
| 2 Output (DSK:SYSCNF.MAC):
|
| 3 DECsystem10 (KL,KS):
|
| Question 4 is asked for KL systems only.
|
| 4 CPUs (1,1-6):
|
| 5 System name:
|
| Question 6 is asked for each CPU.
|
| 6 CPUn serial # (1-10000):
|
| 7 Exclude monitor overhead from user run time (YES,NO):
|
| Questions 8 and 9 are asked for KL systems only. Question 9 is asked
| only if you answer Question 8 with YES.
9-6
BUILDING YOUR MONITOR
| 8 EBOX/MBOX runtime accounting (YES,NO):
|
| 9 Exclude PI time from user runtime (YES,NO):
|
| 10 Account verification (YES,NO):
|
| Questions 11 and 12 are asked for KL systems only.
|
| 11 MOS memory (YES,NO):
|
| If you answer NO to Question 12, Question 20 will not be asked.
|
| Autoconfigured device options:
| 12 Include SCA (NO,YES,PROMPT):
|
| If you answer Questions 12 through 16, 20, 21, 26, and 31 through 39
| with PROMPT, you will be asked the following questions:
|
| Device code (CR,0-774):
| Drive, slave, or unit:
| Data:
| Do you want to specify more data (NO,YES):
|
| If you answer NO to Question 13, Question 108 will not be asked.
|
| 13 Include Ethernet service (NO,YES,PROMPT):
|
| If you answer Question 14 with NO, Questions 21 and 31 will not be
| asked.
|
| 14 Include SA10 device support (NO,YES,PROMPT):
|
| Questions 15 through 17 are asked for KL systems only.
|
| 15 Include RP01/2/3 disks (NO,YES,PROMPT):
|
| 16 Include Swapping drums (NO,YES,PROMPT):
|
| If you answer Questions 17, 19, 22, and 27 through 30 with PROMPT, and
| Question 18 with YES, you will be asked the following questions:
|
| Device code (CR,0-774):
| Massbus device (0,0-7):
| Drive, slave, or unit:
| Data:
| Do you want to specify more data (YES,NO):
|
| 17 Include RS04/5 Massbus disks (NO,YES,PROMPT):
|
| 18 Prompt for RP0x/RM0x Massbus disks (NO,YES):
|
| Questions 19 through 21 are asked for KL systems only.
9-7
BUILDING YOUR MONITOR
| 19 Include DX20/RP20 disks (NO,YES,PROMPT):
|
| 20 Include CI20/HSC disks (NO,YES,PROMPT):
|
| 21 Include SA10/IBM disks (NO,YES,PROMPT):
|
| 22 Include TM02/TM03 magtapes (NO,YES,PROMPT):
|
| Questions 23 through 25 are asked for KS systems only.
|
| 23 Include UNIBUS/LP20 printers (NO,YES,PROMPT):
|
| 24 Include UNIBUS/CD20 card readers (NO,YES,PROMPT):
|
| Local Terminal Configuration:
| 25 DZ-11 TTY lines (0-32):
|
| Questions 26 through 53 are asked for KL systems only.
|
| 26 Include DX10/TX01/TX02 tapes (NO,YES,PROMPT):
|
| 27 Include DX20/TX02 tapes (NO,YES,PROMPT):
|
| 28 Include TM78/TU78/TU79 magtapes (NO,YES,PROMPT):
|
| 29 Include TM10B magtapes (NO,YES,PROMPT):
|
| 30 Include TC10C magtapes (NO,YES,PROMPT):
|
| 31 Include SA10/IBM tapes (NO,YES,PROMPT):
|
| 32 Include BA10/LP100 printers (NO,YES,PROMPT):
|
| 33 Include CFE/LP20 printers (NO,YES,PROMPT):
|
| 34 Include XY10 plotters (NO,YES,PROMPT):
|
| 35 Include CR10 card readers (NO,YES,PROMPT):
|
| 36 Include CFE/CD20 card readers (NO,YES,PROMPT):
|
| 37 Include CP10/CP10D card punches (NO,YES,PROMPT):
|
| 38 Include PC09 paper tape punches (NO,YES,PROMPT):
|
| 39 Include PC09 paper tape readers (NO,YES,PROMPT):
|
| Types of DTE service to load:
| 40 Load ANF-10 DTE service (YES,NO):
|
| 41 Load IBMCOMM DTE service (YES,NO):
9-8
BUILDING YOUR MONITOR
| 42 Load DECnet DTE service (YES,NO):
|
| If you answer YES to Question 43, MONGEN asks Questions 44 through 53.
| If you answer NO, MONGEN asks Question 44 and goes directly to
| Question 54.
|
| Local terminal configuration:
| 43 Configure old terminal scanners (NO,YES):
|
| Question 44 is asked for each CPU.
|
| 44 How many RSX-20F TTYs on CPUn (0,0-128):
|
| 45 How many DC10s on CPUn (1,0-2):
|
| 46 How many DC68s on CPUn (1,0-2):
|
| 47 How many DC76s on CPUn (1,0-2):
|
| Questions 48 through 50 are asked for each DC10.
|
| For DC10 n:
| 48 # DC10B 8 line data groups (1-8):
|
| If Question 49 is answered with 0, Question 50 is not asked.
|
| 49 # DC10E Data set control groups (0-1):
|
| 50 Correspondance of DC10E lines to the DC10B lines (M-N,P):
|
| Question 51 is asked for each DC68.
|
| For DC68 n:
| 51 # Octal lines on DC68, including its console TTY (1-144):
|
| Questions 52 and 53 are asked for each DC76.
|
| For DC76 n:
| 52 Which DL10 port is the DC76 connected to (0,1-7):
|
| 53 Decimal lines on DC76 (1-129):
|
| 54 OPR octal line (CTY,0-3):
|
| 55 Data set lines
|
| 56 Number of Jobs (1-511):
|
| 57 Maximum pages of core for each job (0,0-16384):
|
| 58 Total amount of system core in K (512-4096):
9-9
BUILDING YOUR MONITOR
| Question 59 is asked for KS systems only.
|
| 59 Clock ticks per second (60,50):
|
| Question 60 is asked for KL systems only. If you answer Question 60
| with 0, MONGEN asks Question 61. Otherwise, MONGEN skips to Question
| 62.
|
| 60 Number of real-time devices (0,0-77):
|
| 61 Allow jobs to be locked in core (YES,NO):
|
| 62 Guaranteed number of pages for unlocked jobs (0,0-16384):
|
| 63 Number of pages reserved for locking jobs in Exec Virtual
| Memory (10,0-?):
|
| 64 Number of high priority queues (0,0-15):
|
| 65 Meter (NO,YES):
|
| 66 KASER (NO,YES):
|
| 67 MPXSER (YES,NO):
|
| 68 PSISER (YES,NO):
|
| 69 IPCF (YES,NO):
|
| 70 ENQ/DEQ (YES,NO):
|
| 71 Disk sets (ALL):
|
| If you are building a KL monitor and you answer Question 72 with YES,
| MONGEN asks Questions 73 through 86. Otherwise, MONGEN skips to
| Question 97.
|
| If you are building a KS monitor and you answer Question 72 with YES,
| MONGEN asks Questions 87 through 96. Otherwise, MONGEN skips to
| Question 97.
|
| 72 Configure non-autoconfigured hardware (NO,YES):
|
| For KL systems only, Questions 73 through 86:
|
| on CPUn:
| 73 DIS (NO,YES):
|
| 74 Type (VP10,340,VB10C):
|
| 75 # TD10S (0,0-2):
9-10
BUILDING YOUR MONITOR
| 76 # Units on DTn (1-8):
|
| 77 # of DA28s (0,0-4):
|
| 78 # of lines for TTY pool (8,0-512):
|
| 79 # of DAS78s (0,0-8):
|
| For DAS78 n:
| 80 Which DL10 port is the DAS78 connected to (0,0-7):
|
| 81 Decimal lines on the DAS78 (1-16):
|
| 82 # DN60s (0,0-7):
|
| For DN60 n:
| 83 Which port is the DN60 connected to (0,0-7):
|
| 84 Decimal lines on the DN60 (1-12):
|
| 85 How many DC75NPs or DN87s on CPUn (0,0-8):
|
| For front end number 1:
| 86 To which DL10 port is the DC75 or DN87 connected (0,0-7):
|
| For KS systems only, Questions 87 through 96:
|
| 87 # RX211s (0,0-2):
|
| 88 # Units on RXx (1-2):
|
| 89 # KMC/DUP Lines (0,0-2):
|
| 90 Type of line for KDPn (ANF10,DECNET,USER,IBM):
|
| 91 # DMR11 Lines (0,0-8):
|
| 92 DMRn line type (ANF10,DECNET,USER,IBM):
|
| 93 DMRn switched line (YES,NO):
|
| 94 DMRn half duplex (YES,NO):
|
| 95 Is DMRn secondary (YES,NO):
|
| 96 Load IBM communications software (YES,NO):
|
| 97 Number of PTYs (20,1-505):
|
| If you answer Question 98 with YES, MONGEN asks Questions 99 and 100.
| If you answer NO, MONGEN skips to Question 126.
9-11
BUILDING YOUR MONITOR
| 98 Network software (YES,NO):
|
| 99 Node name:
|
| 100 Number of remote TTYs (485,0-485):
|
| If you answer YES to Question 101, MONGEN asks Questions 102 through
| 114. If you answer NO, MONGEN skips to Question 115.
|
| 101 ANF-10 software (YES,NO):
|
| 102 Node name (node):
|
| 103 Node number of central site (1,1-77):
|
| 104 Remote terminals (YES,NO):
|
| 105 Virtual terminals (YES,NO):
|
| 106 Remote card readers (YES,NO):
|
| 107 Remote line printers (YES,NO):
|
| 108 Remote paper tape punches (NO,YES):
|
| 109 Remote paper tape readers (NO,YES):
|
| 110 Remote plotters (NO,YES):
|
| 111 Remote DN8x DDCMP devices (YES,NO):
|
| 112 Remote data entry terminals (YES,NO):
|
| 113 Remote task-to-task (YES,NO):
|
| 114 Number of connects (256,1-512):
|
| If you answer YES to Question 115, MONGEN asks Questions 116 through
| 121. If you answer NO, MONGEN skips to Question 122.
|
| 115 DECnet software (YES,NO):
|
| 116 Node name (node):
|
| 117 Area number of central site (1,1-63):
|
| 118 Node number of central site (1,1-1023):
|
| 119 Router type (ROUTING,NONROUTING):
|
| 120 Transmit password (DECNET20):
9-12
BUILDING YOUR MONITOR
| 121 Remote terminals (YES,NO):
|
| Question 122 is asked only if you answered Question 13 and 91 with
| YES. If you answer Question 122 with YES, MONGEN asks Questions 123
| through 125. If youu answer 122 with NO, MONGEN skips to Question
| 126.
|
| 122 LAT software (YES,NO):
|
| 123 Node name (node):
|
| 124 Number of circuits (20,1-100):
|
| 125 Service groups:
|
| 126 Decimal "symbol,value":
|
| 127 Octal "symbol,value":
|
| 128 SIXBIT "symbol,value":
|
| 129 Type "device-mnemonic,PI-channel" for special devices:
|
| 130 Type "device-mnemonic,PI-channel,no.-of-devices":
|
| 131 Type "device-mnemonic,PI-channel,highest-ac-to-save":
|
| 132 Type "ersatz-device,P,Pn,search-list-type":
|
| 133 Type "command-name,dispatch,flags,unique-bits,PRVTAB-name":
|
| If you enter a terminal class in Question 134, MONGEN asks Questions
| 135 through 144. Otherwise, MONGEN skips to Question 148.
|
| 134 Terminal class:
|
| 135 Width (80,17-255):
|
| 136 Length (0,0-255):
|
| 137 Fill (0,0-3):
|
| 138 Characteristics:
|
| 139 Address of erase table (V100EP):
|
| 140 Address of rubout table (VTXXBP):
|
| 141 ANSI conformance level (1,0-15):
|
| 142 DEC conformance level (0,0-15):
9-13
BUILDING YOUR MONITOR
| 143 Characteristic attributes:
|
| If you do not enter a member terminal type in Question 144, MONGEN
| repeats Question 134. If you enter a member terminal type, MONGEN
| first asks Questions 145 and 146, then repeats Question 134.
|
| Now enter member terminal types
| 144 Terminal type:
|
| 145 Additional attributes to set:
|
| 146 Attributes to clear:
|
| If a class to extend is entered in Question 147, Questions 144 through
| 146 are repeated. If no class to extend is entered, MONGEN asks
| Question 148.
|
| Additional terminals for classes:
| 147 Class to extend:
|
| 148 System default terminal type (TTY):
|
| This is the last question in the SYSGEN portion of the MONGEN
| dialogue. MONGEN responds with:
|
| File DSK:SYSCNF.MAC Closed [SYSGEN finished]
|
| The SYSGEN dialogue is finished when the source file is written to
| disk. Question 1 is repeated so that you may start the FGEN portion
| of the dialogue. If you wish to exit from MONGEN, type <CTRL/Z>.
|
|
|
| 9.1.2.2 Running FGEN -
|
| The FGEN dialogue simplifies the task of choosing appropriate software
| from a wide selection. For example, FGEN allows you to select
| non-standard TOPS-10 features for a given application. In the
| dialogue, the software options are called "feature-test switches."
|
|
| NOTE
|
| Only the standard settings of the feature-test
| switches have been tested and are fully supported.
| Although installations may alter the settings of
| feature-test options, nonstandard option settings have
| not been tested and are not supported.
|
If you change the settings of the feature-test switches, you must then
rebuild the monitor library file, TOPS10.REL. This procedure is
documented in Appendix B.
9-14
BUILDING YOUR MONITOR
After you select the FGEN portion of the dialogue, you may:
o List the feature-test switches with their standard settings.
o Request an explanation of the feature-test switch symbols.
o Change the feature-test switch settings.
o Add new feature-test switches.
The FGEN questions are listed below.
1 Which GEN (SYS,F):
2 Output (DSK:F.MAC):
|
| 3 Feature set (KL10,KS10):
|
| If you select the standard settings by answering Question 4 with YES,
| the FGEN portion of the dialogue is finished.
|
| If you answer Question 4 with NO, MONGEN prompts you with Question 5.
|
| 4 Standard settings (YES,NO):
|
| To list a standard switch and its setting, type the name of the switch
| and press RETURN. For example:
|
| Feature:KL10<RET>
| Hardware options
| FTKL10==:-1 ;KL10 support
|
| To list all standard switches and their settings, type ALL and press
| RETURN. All standard switches and settings will be listed, grouped
| according to the following categories: hardware options,
| communications and network features, I/O features, scheduler features,
| UUO features, miscellaneous features, and customer defined features.
|
| To change the value of a known switch, or to create a new switch,
| enter the name of the switch without the FT prefix, its value, and a
| comment if desired, separated by commas and press RETURN.
|
| 5 Feature:
|
| This is the last question in the FGEN portion of the MONGEN dialogue.
| MONGEN responds with:
|
| File DSK:F.MAC Closed
|
Question 1 is asked again. You may choose SYSGEN or FGEN, or type
<CTRL/Z> to exit from MONGEN.
9-15
BUILDING YOUR MONITOR
9.2 ASSEMBLING THE MONITOR
You must always assemble the following monitor source files:
COMMON.MAC
COMDEV.MAC
COMMOD.MAC
|
| With the configuration source file:
|
| SYSCNF.MAC
|
It is necessary to assemble the following universal files with them:
F.MAC
S.MAC
| NETPRM.MAC
| DTEPRM.MAC (KL only)
| DEVPRM.MAC
| D36PAR.MAC
|
It is important that F.MAC precede S.MAC in the compiler commands,
because S.MAC has conditional assemblies that depend on the contents
of F.MAC. Use the following monitor commands to compile the monitor:
.COMPILE/COMPILE F,S<RET>
MACRO: F
MACRO: S
EXIT
| .COMPILE/COMPILE DEVPRM,DTEPRM,NETPRM,D36PAR<RET>
| MACRO: DEVPRM
| MACRO: DTEPRM
| MACRO: NETPRM
| MACRO: D36PAR
|
| EXIT
|
| .COMPILE/COMPILE SYSCNF+<COMMON,COMDEV,COMMOD><RET>
| MACRO: COMMON
| MACRO: COMDEV
| MACRO: COMMOD
|
| EXIT
If you are using a standard combination of feature-test switch
settings and DIGITAL-supplied sources, you do not need to assemble the
remainder of the bundled monitor modules. These modules have already
been assembled and combined in a monitor library file named
| TOPS10.REL. TOPS10.REL must be located in the same disk area as that
| which you are using to build the monitor, before you can link and save
9-16
BUILDING YOUR MONITOR
| the monitor.
|
| For new KL installations, the TOPS10.REL library is in [10,7,MON,KL].
| For new KS installations, the TOPS10.REL library file is in
| [10,7,MON,KS].
If you are not using a standard combination of feature-test switch
settings or if you have made changes to the monitor sources, you must
assemble all the monitor sources, creating a new monitor library file.
These procedures are explained in Appendix B.
9.3 LINKING AND SAVING THE MONITOR
The next step in building the monitor is to link the monitor modules.
The LINK program used in the procedures that follow provides both a
runnable monitor (SYSTEM.EXE), and several options for information you
can receive about the linking process. This section lists only the
commands that the LINK program requires to build the monitor. For
full documentation on the LINK program, refer to the TOPS-10 Link
Reference Manual.
The following sequence of instructions loads and saves a monitor for a
standard KL configuration with no unbundled software:
| .R LINK<RET>
| *SYSTEM/SAVE/NOINITIAL/HASH:13K = /LOCALS-<CR>
| #COMMON.REL,COMDEV.REL,COMMOD.REL,TOPS10.REL-<CR>
| #/PATCH:200/SEARCH/COUNTERS/GO<CR>
|
EXIT
.
The file name SYSTEM is an arbitrary name chosen for the monitor; you
can use any file name. However, the bootstrap program defaults to
SYSTEM.EXE[1,4], so SYSTEM is a good choice for the file name.
To load unbundled software with the monitor, include the name of the
relocatable file(s) in the command line after COMMOD. For SMP support
use CPNSER.REL.
9.3.1 Copying the New Monitor to SYS:
Now copy your customized monitor to SYS so that it can be easily
loaded by the bootstrap program. To copy your monitor to SYS, type
the following monitor command:
.COPY SYS:=SYSTEM.EXE<RET>
9-17
BUILDING YOUR MONITOR
After SYSTEM.EXE is in SYS, you can start your new monitor.
9.4 BRINGING UP THE NEW MONITOR
Before you can load the new monitor, you must reload the system. To
do this, you will shut down the current monitor, then reload the
monitor. Use the GO startup option to start the monitor quickly
without changing disk parameters. The following procedures show a
quick startup for a KL and then for a KS. On both systems,
SYS:TTY.INI and SYS:SYSJOB.INI were created previously.
Procedure for KL Systems
<CTRL/\>
PAR>SHUTDOWN<RET>
DECSYSTEM-10 NOT RUNNING
[Dumping on DSKB:CRASH.EXE[1,4]]
[Loading from DSKB:SYSTEM.EXE[1,4]]
MONITOR mm-dd-yy
Why reload:NEW<RET>
Date:NOV-29-87<RET>
Time:18:36:12<RET>
Startup option: GO<RET>
MONITOR 18:36:41 CTY system 2476
Connected to Node CENTRA(0) Line #45
.LOGIN 1,2<RET>
.R OPR<RET>
OPR>EXIT<RET>
.
Procedure for KS Systems
<CTRL/\>
KS10>HA<RET>
(Press the RESET and BOOT switches on the console at the same time.)
KS10 CSL.V#.#
BTAUTO
BOOT V#(##)
BOOT><RET>
MONITOR mm-dd-yy
9-18
BUILDING YOUR MONITOR
Why reload:NEW<RET>
Date:NOV-29-87<RET>
Time:18:36:06<RET>
Startup option: GO<RET>
MONITOR 18:36:20 CTY system 4101
.LOGIN 1,2<RET>
.R OPR<RET>
OPR>EXIT<RET>
.
| Next, you will create a bootable magnetic tape for your system.
|
|
|
| 9.5 CREATING THE SYSTEM BOOTABLE TAPE
|
| This section describes the procedure for creating a bootable magnetic
| tape for your system. The bootable tape is used to get your monitor
| running in the event of a disk failure that prevents booting from
| disk. You should rebuild the bootable tape every time you load a new
| monitor.
|
| For KL systems, you can build the bootable tape using the instructions
| in Section 9.5.1. For KS systems, refer to Section 9.5.2. These
procedures save the monitor as DSKB:SYSTEM.EXE[1,4].
|
|
|
| 9.5.1 Creating the KL System Bootable Tape
|
| Use the following steps to create a bootable monitor tape for a KL
| system. Mount a magnetic tape that you will use for the bootable
| tape, ensuring that the write ring is on the tape. Tape drive MTA0 is
used here.
.ASSIGN MTA0: BACKUP:<RET>
.SET BLOCKSIZE BACKUP:512<RET>
.REWIND BACKUP:<RET>
.COPY BACKUP:=FEFILE.EXE<RET>
.COPY BACKUP:=SYS:BACKUP.EXE<RET>
Then, copy your new monitor and all the system files to tape by
typing:
.R BACKUP<RET>
9-19
BUILDING YOUR MONITOR
/SAVE DSKB:[1,4]=SYS:SYSTEM.EXE<RET>
| /SAVE DSKB:[1,4]=SYS:*.*,REL:*.*,UNV:*.*<RET>
|
|
|
| 9.5.2 Creating the KS System Bootable Tape
|
| Use the following steps to create a bootable monitor tape for a KS
| system. Mount a tape that you will use for the bootable tape,
| ensuring that a write ring is included. Tape drive MTA0 is used here.
.ASSIGN MTA0: BACKUP:<RET>
.SET BLOCKSIZE BACKUP:512<RET>
.REWIND BACKUP:<RET>
| .COPY BACKUP:=T10KL.RAM<RET>
|
| .COPY BACKUP:=BOOT.RDI<RET>
|
.COPY BACKUP:=SYS:BACKUP.EXE<RET>
Then, copy your new monitor and all the system files to tape by
typing:
.R BACKUP<RET>
/SAVE DSKB:[1,4]=SYS:SYSTEM.EXE<RET>
| /SAVE DSKB:[1,4]=SYS:*.*, REL:*.*, UNV:*.*<RET>
9.6 AUTOMATING THE BUILD PROCEDURE
This section describes how to automate the build procedure. After you
build the monitor through direct interaction, you can edit the
distributed build files to tailor them to the characteristics of your
installation. This section describes the files necessary to build a
monitor, tells you where they are located and how to customize them
and use them.
You can use the automated build procedure whenever the monitor must be
rebuilt (to install patches, upgrade the monitor version, or to make
changes to the configuration of the hardware and/or software). When
you need to make changes to your monitor that require rebuilding it,
follow the instructions in this chapter.
Before you submit the control files described in this section, start
the GALAXY batch and spooling system as described in Chapter 10.
9-20
BUILDING YOUR MONITOR
The following table lists each file and its function. An asterisk (*)
appears next to files that require customization. The next section
explains how to modify these files.
Some of the files pertain only to particular installations. Except
| where noted, all of the files below are on the Monitor/ANF-10 tape.
Table 9-1: Monitor Build Files
File Function
|
| BUILD.CTL compiles the files in the build areas. These
| areas include the COMMON modules (COMMON,
| COMDEV, and COMMON). In addition, BUILD.CTL
| builds TOPS10.GLB and MON.EXE.
|
* BUILDS.DCL contains command definitions.
* BUILDS.MIC submits all the build control files.
* BUILDS.PTH defines the build areas.
COMCTL.MIC provides a template for the control file that
COMPIL.MIC uses.
COMCTL.TEC reads COMCTL.MIC to generate the control file
that COMPIL.MIC uses.
COMPIL.CTL builds TOPS10.REL.
COMPIL.MIC creates and submits a control file that compiles
and lists a single monitor module.
* GEN.CTL runs the MONGEN dialogues.
MPE.FIL contains a list of MPE files to compile and the
names of the directory areas for the .REL files.
This file applies only to SMP sites. It is on
the MPE or CPNSER tape.
START.CTL performs the following operations in preparation
for full recompilation of the monitor:
o Deletes all files from BASE: except *.MAC
and *.MIC.
o Creates the LINK.CCL and GLOB.MIC files.
o Creates MPE.FIL, UNSUP.FIL, and DECNET.FIL
if they do not already exist in the build
9-21
BUILDING YOUR MONITOR
area.
o Uses the *.FIL files to create UNIV.CMD,
TOPS10.CMD, and TOPS10.CCL in each FGEN
area.
o Invokes TREE.TEC, and TOPS10.TEC.
|
| NOTE
|
| Submit START.CTL whenever you install a
| new version of the monitor, an autopatch
| update to the monitor, or when you
| change the value of any previously
| defined feature test symbols from a
| universal file.
|
* SWITCH.INI includes switches, runs the version of PATH
shipped with TOPS-10 Version 7.04, and sets
defaults for the SUBMIT commands.
* TOPS10.FIL contains a list of files to compile and the
names of the directory areas for the .REL files.
TOPS10.TEC creates the .CCL and .CMD files that COMPIL.CTL,
BUILD.CTL, and UNIV.CTL use.
TREE.TEC uses BUILDS.PTH and START.MIC to insure that
FGEN and BUILD areas exist and that they are
empty. TREE.TEC deletes START.MIC when it
completes this action.
* UNIV.CTL recompiles all the universal files in selected
areas, as defined by hard-wired paths.
UNSUP.FIL contains a list of customer-supported modules to
compile. This file applies only to sites with
unsupported configurations. It is on the
customer-supported tape.
9.6.1 Customizing the Build Files
Some of the files for building the monitor can be used without
alteration. However, the following files are only templates. You
must modify them to build a monitor that reflects your installation.
This section lists those files and explains how to customize them.
o BUILDS.DCL
9-22
BUILDING YOUR MONITOR
This file defines the PATH command to run SYS:PATH.EXE (as
used in the supplied .MIC and .CTL files). This command must
invoke the PATH program supplied on the 7.04
customer-supported CUSP tape. If this version of the PATH
program is installed in SYS: you do not need to include this
command definition.
o BUILDS.MIC
This file submits COMPIL.CTL in each FGEN area, and BUILD.CTL
in each build area. Edit it to use only your monitor areas.
o BUILDS.PTH
This file consists of three sections, separated by the
following string: !<*>! . Each section defines logical
names for one of the following areas: library areas, FGEN
areas, and monitor build areas.
1. LIBRARY AREAS
- BASE:
This name is required. Define it to point to the
area used to edit any locally modified sources for
building. The area should be the superior SFD (or
UFD) of each of the FGEN areas defined in the second
section of the file.
- CTLS:
This name is required. Define it to point to the
files used in the build procedures (*.MIC, *.CTL,
*.TEC, and *.FIL).
- LOGS:
This name is required. Define it to point to the
area that stores all the batch log files.
- MON:
You must define this name, and specify the /SEARCH
attribute. The /SEARCH attribute insures
accessibility of the files required during the build
procedure. It points to the distribution sources,
and files used in the build procedures.
- SYS:, UNV:, REL:
These may not be required, but you must use the
current version of MACRO and LINK to build the
9-23
BUILDING YOUR MONITOR
monitor.
2. FGEN AREAS
The FGEN areas are used to build a TOPS10.REL file for
linking monitors. There is one FGEN area for each FGEN
dialogue you complete in MONGEN. Most installations need
only one or two such areas: a KL area and a KS area.
3. MONITOR BUILD AREAS
The monitor build areas are used to build a MON.EXE for
each system at an installation. Define these areas to be
in SFDs subordinate to the appropriate FGEN areas.
However, if there is only one build area corresponding to
a given FGEN area, the two areas can point to the same
SFD. For example, in the supplied prototype, S could be
the same as KS.
| Finally, the BUILDS.PTH file contains DECLARE commands to
| define required monitor commands. This procedure also
| ensures that full directory scanning is enabled for the job.
|
| It may not be necessary to define commands, but directory
| scanning is required for the job. The commands must be
| defined if the System Search List definition is not [1,4].
| The automated build procedure requires that the PATH and MIC
| commands be defined, therefore those programs must exist in
| the build area if they are not in SYS.
o GEN.CTL
This file runs MONGEN for each FGEN and build area. Edit it
to generate only the configurations in your monitor areas.
o SWITCH.INI
This file should be in the UFD corresponding to the BASE:
area. Define it to run the version of PATH supplied on the
7.04 customer-supported CUSP tape and to process BUILDS.PTH.
It should also set up appropriate defaults for SUBMIT
commands.
o TOPS10.FIL
TOPS10.FIL contains a list of files to compile. For each
file, it lists four optional fields of information in the
following format:
<UNIV SPEC>|<COMPIL STRING>|<LIBRARY SPEC>|<AREA LIST>
9-24
BUILDING YOUR MONITOR
1. UNIV SPEC
This field determines whether the file is a universal
file, and consequently, whether its COMPIL STRING should
be duplicated in UNIV.CMD.
2. COMPIL STRING
This field contains the command to give to COMPIL to get
the file compiled. If the program is available as a .REL
file (such as EDDT), leave this field blank.
3. LIBRARY SPEC
This field contains the specification for the module to
be included in TOPS10.REL. If this file should not be
part of the monitor library file, leave this field blank.
4. AREA LIST
This field contains the list of FGEN and build areas for
which the file should be compiled.
o UNIV.CTL
This file forces compilation of all monitor sources. It
defines a path to each FGEN area to do this. Edit it to use
your FGEN areas.
9.6.2 The Build Procedure
The following example shows the complete build procedure. The column
to the left shows the commands that you type. The column to the right
of the commands contains explanatory remarks.
.SUBMIT START<RET>
.DO COMPILE MONGEN<RET> Include this COMPILE command only
if you have made changes to
MONGEN.MAC.
.SUBMIT UNIV<RET> Include the SUBMIT UNIV command
only if you have submitted START.
| .DO COMPILE modulename<RET> Include DO COMPILE if you have made
| changes to a module. The module
| will be recompiled and a listing
9-25
BUILDING YOUR MONITOR
| will be generated.
.DO BUILDS<RET> Compiles the monitor modules in the
FGEN areas, and builds MON.EXE and
TOPS10.GLB. If you have not
changed the universal files, this
is the only step you need to
perform.
9.7 CUSTOMIZING BOOT
| You can run the batch control file, BOOT.CTL, as distributed, to build
| BOOT after you have started GALAXY (refer to Chapter 10). The
| following command submits the control file for batch processing:
|
| .SUBMIT BOOT.CTL<RET>
|
| Or, before you bring up the monitor, you can perform the individual
| steps manually and modify the control file to eliminate the loading
| procedures for hardware that does not exist at your installation.
|
| For a KS system, you can modify the control file to include a step
| that creates binary files or a step to start EDDT.
|
| For a KL system, you can modify the control file to include or exclude
| microcode for various hardware components, as well. This procedure
| allows you to save space in memory, because the monitor contains all
| the microcode available.
|
| Section 9.7.1 describes the procedure for KL systems. Section 9.7.2
| describes the procedure for KS systems.
|
|
|
| 9.7.1 Customizing the KL BOOT
|
| To customize BOOT for your KL system, first enter the following
| command:
|
| .COMPILE/COMPILE BOOT.MAC<RET>
|
| The following output will appear on your terminal:
|
| MACRO: BTSPRM
| BOOT
| [Assembling Boot for a KL10]
| [Including DX10/TX01 driver]
| [Including TM02 driver]
| [Including DX20/TX02 driver]
| [Including TM78 driver]
9-26
BUILDING YOUR MONITOR
| EXIT
| .
|
| The files BTSPRM.UNV and BOOT.REL have been created. Next, run LINK
| to create BTONCE.EXE:
|
| .R LINK<RET>
| *BTONCE/SAVE=BOOT/LOCALS/SYMSEG:LOW,REL:EDDT/GO<RET>
| EXIT
| .
|
| Now, you can run BTONCE and, by providing Y or N answers to the
| questions, include or exclude microcode for various hardware
| components. Running BTONCE creates BOOT.EXE and BOOT.EXB. The BTONCE
| questions are as follows:
|
| .R BTONCE<RET>
| Create binary output file [Y/N]: Enter Y or N, then
| Preserve EDDT and symbols [Y/N]: press RETURN
| Load DX10/TX01 microcode [Y/N]:
| Load DX20/TX02 microcode [Y/N]:
| Load DX20/RP20 microcide [Y/N]:
| Load KLIPA/CI microcode [Y/N]:
| Load KLNI/NI microcode [Y/N]:
| .
|
| Lastly, run FILDDT to create a BOOT DDT that you can use to look at
| BOOT in the running monitor and examine problems that you might have
| with your BOOT procedure. The FILDDT prompt is File:. Enter the file
| name BTONCE followed by the switch /S, which directs FILDDT to load
| the symbol table from the file. After entering the file name and
| switch, press RETURN:
|
| .R FILDDT<RET>
| File:BTONCE/S<RET>
| [Extracting symbols from file DSKB:BTONCE.EXE[*.*]
| [1339 Symbols loaded from file]
| File:
|
| Enter the switch /M at the File: prompt and press RETURN. The switch
| /M directs FILDDT to look at the running monitor. In the following
| discussion, note that $ represents the escape character.
|
| Next, enter the command string 1$U, followed by RETURN. 1$U provides
| FILDDT with the physical page number of the CPU's EPT (Executive
| Process Table). $U is the commmand that sets the EPT page number in
| FILDDT; the EPT for CPU0 occupies physical page 1.
|
| Then, enter the command string -1,,-BTSVEC$8U, followed by RETURN.
| Because it is unlikely that the location of the assembled symbol
| values will match the arbitrary location selected by the monitor for
| BOOT, the assembled values must be relocated. $8U is the relocation
9-27
BUILDING YOUR MONITOR
| command, and -1,,-BTSVEC is the argument to that command.
|
| To exit from FILDDT, enter <CTRL/Z>. Then save the BTDDT file it
| created.
|
| File:/M<RET>
| [Looking at the running monitor]
| 1$U<RET>
| -1,,-BTSVEC$8U<RET>
| <CTRL/Z>
| .SAVE BTDDT<RET>
|
| These commands create a new version of BOOT and a BTDDT that you can
| now load. The file must be loaded into each front end on a
| multiple-processor KL system; repeat this procedure at the CTY of each
| processor in an SMP system. You must repeat the copying procedure
| every time you rebuild the front end file system.
The following procedure shows how to load BOOT into the front end for
a single-processor KL system:
.LOG 1,2<RET>
.RUN FE<RET>
.
On the CTY, type the following commands:
<CTRL/\>
PAR>MCR MOU<RET>
MOU>FE0:<RET>
MOU><CTRL/\>
PAR>MCR PIP<RET>
PIP>DB0:BOOT.EXB = FE0:[10,7]BOOT.EXB<RET>
PIP><CTRL/\>
PAR>MCR DMO<RET>
DMO>FE0:<RET>
DMO><CTRL/\>
<CTRL/\>
Now you can bring up your new monitor with the new BOOT.EXB file to
control microcode loading by the front end.
|
|
|
| 9.7.2 Customizing the KS BOOT
|
| To customize BOOT for your KS system, enter the following commands.
| These commands set the KS feature test switch to build BOOT for KS
| systems. Type CTRL/Z twice to exit from COMPILE:
|
| .COMPILE/COMPILE TTY:+DSU:BOOT.MAC<RET>
9-28
BUILDING YOUR MONITOR
| MACRO:FTKL10==0<RET>
| FTKS10==-1<RET>
| CTRL/Z
| CTRL/Z
| .
|
| Next, run LINK and create KSONCE.EXE:
|
| .R LINK<RET>
| *KSONCE/SAVE=BOOT/LOCALS/SYMSEG:LOW,REL:EDDT/GO<RET>
| EXIT
| .
|
| Now, you can modify the control file to include a step that creates
| binary output files and a step to start EDDT by running KSONCE.
| Include these functions by answering the following questions with YES:
|
| .R KSONCE<RET>
| Create binary output file [Y/N]: Enter Y or N, then
| Preserve EDDT and symbols [Y/N]: press RETURN
|
| To create KSBOOT.EXE from BOOT.EXE, type the following:
|
| .RENAME KSBOOT.EXE=BOOT.EXE<RET>
|
| Refer to Chapter 6 for the appropriate dialogue used to copy the file
| to the microprocessor file system.
9-29
10-1
CHAPTER 10
STARTING UP AND TAILORING THE GALAXY SYSTEM
After the TOPS-10 monitor is running, you can start the GALAXY batch
and spooling system. GALAXY is a set of programs that controls the
batch system, the operator interface, the output devices, and requests
for the disk and tape units of the operating system. The components
of the GALAXY system are:
Module Function
ACTDAE Accounting daemon
BATCON Batch controller
CATLOG Mountable device catalog daemon
CDRIVE Card reader spooler
GLXLIB Sharable GALAXY object-time system
LPTSPL Line printer spooler
| NEBULA DQS Spooler
|
OPR Operator interface
ORION Operator interface controller
PLEASE User-to-operator communications program
PULSAR Tape and disk label processor
| QUASAR GALAXY queue manager and scheduler
|
| QUEUE User spooling and mount request program
|
| SPRINT Card reader interpreter
|
10-1
STARTING UP AND TAILORING THE GALAXY SYSTEM
10.1 STARTING THE GALAXY SPOOLERS
Whenever the system is reloaded, the GALAXY spoolers must also be
started. You can start the spoolers by typing commands to the OPR
program, the operator interface for the GALAXY batch and spooling
system. After reloading the system, type the following command to run
OPR:
.R OPR<RET>
OPR>
OPR responds with its prompt, OPR>. To obtain a list of valid
commands to the OPR program, type ?. To obtain a list of valid
arguments to any command, type the command and a ?. To start the
spoolers, type START:
| OPR>START ? ONE OF THE FOLLOWING:
|
| BATCH-STREAM CARD-PUNCH FAL-STREAM NODE
|
| NQC-STREAM PAPER-TAPE-PUNCH PLOTTER PRINTER
|
| READER
|
| In this example, the ? was typed to show the possible arguments to
| START. To start the new GALAXY system, use each of the arguments with
| the START command:
|
| OPR>START BATCH-STREAM n<RET>
| OPR>START CARD-PUNCH n<RET>
| .
| .
| .
| OPR>START PRINTER n<RET>
| OPR>START READER n<RET>
If you receive error messages indicating that certain objects do not
exist, you may ignore them. The message shows that a certain
configuration does not exist at your installation. Continue with the
next command.
The spoolers can be started automatically by OPR if you include the
above commands in the file SYS:SYSTEM.CMD.
You can ensure that the appropriate GALAXY files have been started by
using the OPR>SHOW STATUS command.
The OPR program is described in the TOPS-10 Operator's Command
Language Reference Manual.
10-2
STARTING UP AND TAILORING THE GALAXY SYSTEM
10.2 TAILORING THE GALAXY SYSTEM
This section describes the procedure for tailoring the GALAXY system
to the requirements of your installation. It describes the procedures
necessary for running the GALGEN program, submitting the GALGEN batch
job, and starting the GALAXY system with the new parameters.
It is only necessary to run GALGEN if you have made changes to GALAXY
modules that are affected by GALGEN parameters, or if you wish to
change the default parameters described in the GALGEN program. Note
that DECnet Version 4 support is built into the distributed GALAXY
programs; therefore, it is not necessary to run GALGEN to include
DECnet support.
10.2.1 Running GALGEN
You can run GALGEN during the installation of GALAXY or after GALAXY
is started.
The GALGEN program consists of a set of questions that allows you to
change the standard configuration to match the hardware configuration
of your installation. If your answers to all of the GALGEN questions
match the default answers (shown in square brackets), it is not
necessary for you to run the GALGEN program. The GALGEN dialog is
shown below in the LONG format. Additional explanation is provided
for some questions.
You can save time and effort by building the GALAXY system in your own
disk area. Copy GALGEN.EXE into your area before running it, and
follow the procedure shown here:
.RUN GALGEN<RET>
GALGEN begins by printing the following:
GALGEN Version n(nnnn)
[Starting GALAXY Generation Procedure for TOPS-10 System]
[Writing GALAXY Configuration File DSKB:GALCNF.MAC[15,7]]
In the following dialog, all questions are of the form:
text of question (list or range of answers) [default answer]
The line starts with the actual text of the question. Following
the question is a description of the possible answers enclosed in
parentheses. This description might be a range of numbers, a
list of keywords, or a textual description. Following this
description is the default answer, enclosed in square brackets.
The default will be assumed if you only press the RETURN key.
10-3
STARTING UP AND TAILORING THE GALAXY SYSTEM
You have the choice of carrying on a LONG dialog in which an
explanation of each question is provided automatically, or a
SHORT dialog in which it is assumed that you are familiar with
the GALAXY generation procedure. In either case, you can always
obtain the help text by typing a question mark (?) in response to
any question.
Answer the following question with either LONG (for a long
dialog) or SHORT (for a short one). Simply pressing the RETURN
key will imply SHORT.
Dialog Length (SHORT,LONG) [SHORT]
Administrative Controls and Parameters
The questions in this section determine absolute GALAXY
parameters, which cannot be overridden by user or operator
commands.
| The system logs all operator interactions in a disk file in the
| spooling area. Your answer to the following question specifies
| the name to be used for this file.
|
| Operator log filename (3-6 Characters) [OPERAT]
|
| You have the option of maintaining a backup master queue file
| which may be used in the unlikely event that the primary master
| queue file is destroyed.
|
| Do you want redundant master queue file (YES,NO) [NO]
|
| The backup file for the master queue file is useful if GALAXY or
| the master queue file is accidentally deleted or damaged. It
| records all the current queue jobs so that they will not be lost.
|
| Any permanently mounted structure may be used for the master
| queues. The structure that is first in the system search list is
| preferred.
|
| File structure to use for master queue (3-6 Characters) [SSL]
|
| The structure on which the master queue file is stored should be
| easily accessible by GALAXY. SSL is the acronym for System Search
| List. Unless you are running a private version of GALAXY, you should
| press RETURN for this question.
|
| Users can specify a priority for their batch and spooling
| requests with the /PRIORITY switch. The allowable values for
| this parameter range between 1 (representing lowest priority) and
| 63 (representing the highest). The following two questions
| relate to this switch. The first question requests the maximum
10-4
STARTING UP AND TAILORING THE GALAXY SYSTEM
| priority which may be specified by a non-privileged user.
|
| Maximum priority non-privileged users may specify (1-63) [20]
|
| You may specify the value to be used if the user does not specify
| a /PRIORITY switch. This value must be between 1 and the maximum
| priority that you specified in the previous question.
|
| Default for /PRIORITY on batch and spooling requests (1-63) [10]
|
| A queue request that is queued without the /PRIORITY switch must take
| a default priority. The value you specify here will be used as the
| default for such requests. The value must be between 1 and the
| maximum that you specified in the previous question. If you use 1
| here, users will not be able to set some requests to a priority that
| is lower than the default. Therefore, it is recommended that the
| default that you specify here be greater than 1.
|
|
| [SYSTEM]INFO Parameters
|
| This section allows you to set any currently defined [SYSTEM]INFO
| control parameters.
|
| The [SYSTEM]INFO facility, which allows users to manipulate PIDs
| for IPCF communication by symbolic names, is now part of QUASAR.
| While [SYSTEM]INFO normally limits the maximum length of a
| symbolic PID name using the GETTAB item %IPMPL (maximum packet
| length), you have the option of specifying the maximum length of
| a PID name allowable at all.
|
| Maximum length of a PID name (29-79 Characters) [39]
|
|
| Applications
|
| The following section determines which optional applications you
| wish to run on your system. These applications will be accesible
| through OPR, the operator interface program. GALAXY supports by
| default the CATALOG, CONFIG, and QUOTA applications. If you wish
| to run additional applications (such as NCP) answer YES.
|
| If you answer this question with NO, the Applications portion of
| GALGEN is completed.
|
| Do you wish to run additional applications? (YES,NO) [YES]
|
| If you want to run addional applications, GALGEN asks:
|
| Include the LAT Control Protocol (LCP) commands? (YES,NO) [YES]
|
| Include the DECnet Network Control Protocol (NCP) commands?
10-5
STARTING UP AND TAILORING THE GALAXY SYSTEM
| (YES,NO) [YES]
|
| You can include installation-developed applications with your
| answers to the following questions.
| Do you wish to run any installation-developed applications?
| (YES,NO) [NO]
|
| If you answer this question with YES, GALGEN asks the following
| questions:
|
| Specify the name of the application. This name is used to access
| the application through OPR.
| Application name or carriage return to exit (1-20 characters)
|
| Each application must be associated with an application table
| name. This table file contains the command syntax parsing tree
| for the application. Specify only the file name.
| Table name for application (1-6 characters)
|
| The code to implement an application can exist either as a
| separate program (which communicates with users via ORION), or as
| an internal module within ORION itself.
| Internal ORION application? (YES,NO) [NO]
|
| If you answer this question with YES, GALGEN asks the following
| question:
|
| Please specify the name of the module containing the
| application-specific code. This module will be assembled and
| incorporated into ORION. You need only provide the file name.
| The extension of .MAC is assumed.
| Application module name (1-6 characters)
|
|
| Batch Job Defaults
|
| The questions in this section are used to define default values
| for the various options that a user can specify when submitting a
| batch job.
|
| The batch user can specify a maximum runtime for his batch job
| using the /TIME switch. If he does not specify this switch the
| following default will be used:
|
| Default batch job runtime limit (5-9999 seconds) [300]
|
| Every batch job has maximum limits for each type of spooled
| output. These limits can be set by the batch user with
| appropriate switches to the SUBMIT command. If the user doesn't
| specify these switches, the following defaults will be used:
|
| Line printer output limit (0-9999 pages) [200]
10-6
STARTING UP AND TAILORING THE GALAXY SYSTEM
| Card punch output limit (0-9999 cards) [1000]
| Paper tape punch output limit (0-5000 feet) [200]
| Plotter output limit (0-5000 minutes) [30]
|
| The user can specify whether or not the batch log file should be
| printed after the job terminates by using the /OUTPUT switch to
| the SUBMIT command. The action is specified by: LOG to always
| print the log file, NOLOG to never print the log file, and ERROR
| to print the log file only if the job ended with an unhandled
| error. If the user doesn't specify this switch, the following
| value will be used.
|
| Default for SUBMIT/OUTPUT (LOG,NOLOG,ERROR) [LOG]
|
| The system provides for the enforcement of a maximum physical
| memory limit on batch jobs. The user can specify a maximum
| memory parameter and if any program attempts to exceed this
| limit, the batch job goes virtual.
|
| If you answer the following question with NO, the Batch Job Defaults
| section of GALGEN is completed.
|
| Should memory limits be enforced (YES,NO) [YES]
|
| The user can specify a maximum memory limit by using the /CORE
| switch. If this switch is not specified by the user, the
| following default will be used.
|
| Memory limit (24-16384 pages) [512]
|
|
| Line Printer Defaults and Parameters
|
| The questions in this section are used to set default values for
| the line printer spooler.
|
| Job banner pages are printed at the beginning of each print
| request. The user's name and any /NOTE value is printed on them.
|
| Number of job banner pages (0-5) [2]
|
| Job trailer pages are printed at the end of each print request.
| The user's name and any spooler checkpoint information is printed
| on them.
|
| Number of job trailer pages (0-5) [2]
|
| File header pages are printed before each file or copy in each
| print request. The file name is printed in BOLD letters on each
| header page.
|
| Number of file header pages (0-5) [2]
10-7
STARTING UP AND TAILORING THE GALAXY SYSTEM
| The user can specify a special form for a print request with the
| /FORMS switch. If this switch is not specified the following
| default will be used.
|
| Name for standard output forms (4-6 Characters) [NORMAL]
|
| The number of characters that uniquely identifies a form is
| compared with the current form name to determine if a forms
| change is required for a specific user request.
|
| Number of characters which uniquely identifies form (2-6) [4]
|
| The forms type that a user specifies when submitting a request is
| compared with the forms type currently mounted in the printer and the
| forms type recorded in SYS:LPFORM.INI and SYS:SPFORM.INI. Your answer
| to this question specifies the number of characters in the name of the
| forms type that qualify as a match.
|
| There may be site-specific devices which require special device
| drivers loaded with LPTSPL as needed. For example, the LN01
| laser printer requires the module LPTL01 in order for LPTSPL to
| load fonts as requested.
|
| Do you want to load any special printer drivers? (YES,NO) [NO]
|
| If you answer this question with YES, GALGEN asks:
|
| Each device driver must have its own module to be loaded with the
| LPTSPL main module. This module's .REL file must be found on the
| build areas when LPTSPL is loaded.
| Name of device driver module (1-6 characters)
| Any more special drivers to load (YES,NO) [NO]
|
|
| Parameters for MDA
|
| The questions in this section are used to set the default
| parameters for the Mountable Device Allocator (MDA).
|
| The following question allows you to enable the deadlock
| avoidance code in MDA. This code prevents jobs from waiting for
| tape or disk resources that another job may already own, or that
| do not exist. Enabling the deadlock avoidance code causes MDA to
| use slightly more CPU time to insure that deadlocks will not
| occur.
|
| Enable deadlock avoidance? (YES,NO) [YES]
|
| You have the option of specifying which label type is the
| default.
10-8
STARTING UP AND TAILORING THE GALAXY SYSTEM
| NOTE
|
| Label type USER-EOT will allow a user program to
| see the end of tape condition on an UNLABELED
| tape only. All other label types will cause the
| user to block at end of tape while MDA
| automatically expands the volume set. In this
| case the user program will NOT see end of tape.
|
| Default tape label type (ANSI,BYPASS,IBM,USER-EOT,UNLABELED)
| [USER-EOT]
|
| You have the option of specifying the default density at which
| tapes are read and written.
|
| Default 9-track tape density (200,556,800,1600,6250) [1600]
| Default 7-track tape density (200,556,800) [800]
|
| You have the option of specifying the default track type on which
| to mount tapes.
|
| Default track type (7,9) [9]
|
| The system administrator can allow non-privileged users to bypass
| label processing.
|
| Allow bypassing for non-privileged users? (YES,NO) [NO]
|
| When mounting a labeled tape, a user can include the
| /LABEL-TYPE:BYPASS switch to specify that the tape label be ignored.
| If you want this function to be restricted to privileged users, answer
| NO to this question. If you want all users to be able to mount any
| tape regardless of the label type, answer YES to this question.
|
|
| Miscellaneous Defaults and Parameters
|
| The questions in this section refer to general GALAXY parameters.
|
| The following section establishes default bias and disk page to
| spooler unit conversion factors.
|
| Do you want the standard limit computation (YES,NO) [YES]
|
| This question refers to disk page conversion to spooler units. (Note
| that one disk page equals four disk blocks.) If you do not want this
| conversion, answer NO to the question. If you answer NO, GALGEN
| prints the following and prompts you for a value for each output
| device.
|
| The default output limit of a spooled file is calculated as:
10-9
STARTING UP AND TAILORING THE GALAXY SYSTEM
| limit = disk blocks * units per disk block * copies
|
| The output limit is expressed in pages for the printer, cards for
| the card punch, minutes for the plotter, and feet for the paper
| tape punch. The following questions establish the number of
| units of output per disk page.
|
| Number of printed pages per disk block [.7]
| Number of punched cards per disk block [8]
| Minutes of plotter output per disk block [0.07]
| Feet of paper tape per disk block [7]
|
| The default action to be taken when output exceeds the specified
| limit for an output spooler may be to ABORT the job and ignore
| all remaining output, ASK the operator what to do, or PROCEED to
| finish processing the job.
|
| Default output-limit-exceeded action (ABORT,ASK,PROCEED) [ASK]
|
| If file access between network nodes is to be done, each FAL -
| file access listener - stream needs a network type associated
| with it. The network types supported are ANF-10 and DECnet.
|
| Default network-type for FAL streams (ANF-10,DECnet) [ANF-10]
|
| Galaxy-10 spoolers (object processors) are now grouped into three
| categories and are started (logged-in and run) by QUASAR at
| different times depending on which category they fall into. The
| three categories and their corresponding characteristics are:
|
| %ONCE - "fired up" by QUASAR after QUASAR starts ORION,
| PULSAR, and CATLOG. The spooler will never logout on
| its own.
|
| %STCMD - "fired up" by QUASAR when an OPR START command is
| issued for a stream controlled by the spooler. The
| spooler will logout when all streams under its control
| are shut down.
|
| %DEMND - "fired up" by QUASAR when a job is in a queue that can
| be serviced by one of the spooler's job streams. These
| "demand" spoolers will logout after "n" minutes of
| idleness. You will be prompted below to define a value
| for "n".
|
| The PRCDAT macro in QSRMAC.MAC defines the category to which each
| supported object processor/spooler belongs.
|
| Number of minutes "demand" spoolers tolerate idleness (5-60) [10]
|
| Your site may wish to include additional symbols and values which
| are different from Digital's standard ones. By specifying these
10-10
STARTING UP AND TAILORING THE GALAXY SYSTEM
| values here, you may further tailor GALAXY to your particular
| installation.
|
| Do you wish to enter any site specific parameters? (YES,NO) [NO]
|
| If you answer this question with YES, GALGEN prints the following:
|
| The parameters are entered as lines of text in a form acceptable
| for assembly by MACRO-10. Valid examples are:
| XP SYMBOL,EXPRESSION ;comment
| ND SYMBOL,EXPRESSION ;comment
| SYMBOL==EXPRESSION ;comment
| Where SYMBOL is a legal symbol name for MACRO-10, and EXPRESSION
| is any valid MACRO-10 expression consisting of symbols and/or
| constants. The comment is any arbitrary string beginning with a
| semicolon.
| Enter site specific parameter text: (Empty line to end.)
|
| [End of GALGEN Dialog]
|
| When the dialog is finished, GALGEN writes the output file GALCNF.MAC
| into the disk area. Use this output file to configure your new GALAXY
| system.
10.2.2 Assembling the Tailored GALAXY System
Use the batch control file (GALAXY.CTL) provided to assemble the new
GALAXY system. To submit a batch job, the GALAXY system and spoolers
must be running. GALAXY is usually started at system startup from the
SYSJOB.INI or OPR.ATO file. You should create one of these files,
reload the system, and start the GALAXY spoolers (as described in
Section 10.1).
The GALAXY components listed below should be started automatically
whenever the system is reloaded. You can accomplish this by creating
a file named SYSJOB.INI, and including the commands to start the
components in that file. SYSJOB.INI is described in Chapter 12. The
GALAXY components that must be started automatically whenever the
system is reloaded are:
ACTDAE
QUASAR
The batch control file (GALAXY.CTL) that is included on the CUSP tape
uses the GALCNF.MAC file to generate a GALAXY system. The batch job
will generate the appropriate .EXE files for the GALAXY system. Copy
the batch control file into your area and submit it from there.
To start the batch job, type the monitor command and press RETURN:
10-11
STARTING UP AND TAILORING THE GALAXY SYSTEM
.SUBMIT GALAXY/TIME/RESTART<RET>
If the batch job ends in an error, try to correct the control file,
then delete the .LOG file and submit the control file again. If the
error occurs again, consult a software specialist or system
programmer.
NOTE
It is important that the batch job run successfully.
Check the log file carefully to be sure that no
warning (%) errors occurred. These errors do not end
the job but may cause problems later.
When the batch job finishes successfully, a series of messages will be
printed on the operator's terminal. One message will be printed for
each component of the GALAXY system as it is created.
When the components are successfully created, they are ready to be
copied into SYS. Note that the system should be stand-alone during
this process because no users can submit requests during this
procedure.
10.3 SHUTTING DOWN THE GALAXY SYSTEM
If a GALAXY system is already running, you must shut it down before
replacing it with new GALAXY modules. Before shutting down the GALAXY
system, you should obtain exclusive use of the operating system by
scheduling stand-alone time for this purpose or by ending timesharing.
Using the OPR program, stop the GALAXY system. To run OPR, type the
following command and press RETURN:
.R OPR<RET>
OPR>
OPR responds with its prompt, OPR>. To obtain a list of valid
commands to the OPR program, type ? at this point. To obtain a list
of valid arguments to any command, type the command and a ?.
| OPR>SHUTDOWN ? ONE OF THE FOLLOWING:
|
| BATCH-STREAM CARD-PUNCH FAL-STREAM NODE
|
| NQC-STREAM PAPER-TAPE-PUNCH PLOTTER PRINTER
|
| READER
|
| In this example, a ? was typed to show all the possible arguments to
| the SHUTDOWN command. To stop all the components of the GALAXY
10-12
STARTING UP AND TAILORING THE GALAXY SYSTEM
| system, type the SHUTDOWN command followed by the name of each
| component. For example:
|
| OPR>SHUTDOWN BATCH-STREAM n<RET>
| OPR>SHUTDOWN CARD-PUNCH n<RET>
| .
| .
| .
| OPR>SHUTDOWN PRINTER n<RET>
| OPR>SHUTDOWN READER n<RET>
During this procedure, OPR may print the following message on your
terminal:
--THERE ARE NO DEVICES STARTED--
In this case, there were no devices on line for that component. You
can ignore the message and go on to the next command.
After you shut down all the components, exit from OPR. Type EXIT and
press RETURN:
OPR>EXIT<RET>
.
After returning to monitor level, use the SYSTAT monitor command to
determine the jobs that are running for GALAXY. The GALAXY jobs may
be named any or all of the following:
|
| CDRIVE
| CATLOG
| NEBULA
| IBMSPL
|
Using the ATTACH command, attach to each job and log it out.
The GALAXY programs that are currently running must now be copied to a
backup area from SYS. In the following example, they are copied to
OLD ([1,3]).
|
| .COPY [1,3]=[1,4]CDRIVE.EXE,CATLOG.EXE,NEBULA.EXE,IBMSPL.EXE<RET>
|
| .
|
You must also delete the master queue files. The master queue files
can be deleted with the following command:
.DELETE SYS:QSRMS1.*<RET>
FILES DELETED:
SYS:QSRMS1.QSR
221 BLOCKS FREED
10-13
STARTING UP AND TAILORING THE GALAXY SYSTEM
Now you can copy the new files to SYS and bring up the new GALAXY
system.
10.4 STARTING THE GALAXY SYSTEM
After you stop the old GALAXY system, you can start the new GALAXY
system. The first step is to copy all the new GALAXY files into SYS.
The new files will automatically supersede the old files in SYS. Use
the COPY command to copy all the .EXE files from your area to SYS.
.COPY SYS:=*.EXE<RET>
You can start the GALAXY programs by reloading the system if the
system has a file SYS:SYSJOB.INI that logs in the system programs.
The SYSJOB.INI file is described in Chapter 12. To start the GALAXY
programs manually, reload the system by typing CTRL/\ on the CTY and
typing the SHUTDOWN command to the parser. Then reload the system, as
demonstrated here:
<CTRL/\>
PAR>SHUTDOWN<RET>
DECSYSTEM-10 NOT RUNNING
[Dumping on DSKB:CRASH.EXE[1,4]]
[Loading from DSK:SYSTEM.EXE[1,4]]
TOPS-10 704 Development mmm-dd-yy
Why reload:SCHED<RET>
Date:NOV-29-87<RET>
Time:17:55<RET>
Startup option:GO<RET>
TOPS-10 704 Development NOV-29-87
.
Now log in with the PPN [1,2] and start up the GALAXY spoolers by
running the OPR program, as described in Section 10.1.
10-14
CHAPTER 11
MAINTAINING THE SYSTEM AUTHORIZATION FILE
The REACT program enables you to maintain the system authorization
file, ACTDAE.SYS. This chapter describes how to use REACT commands to
create, change, and delete user accounts. It also describes how to
update the master accounting file, and enable account validation. To
change user accounts with REACT, you must have administrative
privileges (JP.ADM in JBTPRV).
The REACT Command Tables at the end of this chapter provide quick
reference to the commands discussed in Sections 11.4 through 11.8.
11.1 CONVERTING OLD ACCOUNTING FILES
The REACT shipped with TOPS-10 Version 7.04 requires that you use the
ACTDAE.SYS accounting file. If your system currently uses ACCT.SYS
and AUXACC.SYS files, you can use a program called NEWACT to create
the new accounting file, ACTDAE.SYS.
To convert ACCT.SYS and AUXACC.SYS files to the ACTDAE.SYS format, use
the NEWACT program as shown below:
1. Run NEWACT.
NEWACT looks up the ACCT.SYS, AUXACC.SYS, and USERS.TXT files
and uses the information in them to create the new ACTDAE.SYS
file. If you do not want the information in USERS.TXT
included in your new accounting file, delete the USERS.TXT
file.
2. Copy the new ACTDAE.SYS file to SYS.
11.2 ENTERING AND EXITING REACT
The following examples show how to enter and exit the REACT program.
11-1
To enter REACT type the following:
.R REACT<RET>
REACT>
To exit the REACT program and return to monitor command level, type
EXIT and press RETURN:
REACT>EXIT<RET>
.
11.3 GETTING HELP WHILE USING REACT
| There are three different ways to get help while using REACT: typing
| a question mark, typing HELP, and pressing the ESCape key.
|
| For information about REACT top-level commands, you can type a
| question mark to display a list of the command keywords:
|
| .R REACT<RET>
| REACT>? one of the following:
| CHANGE DELETE EXIT HELP INSERT LIST LOCK PURGE SELECT SHOW
| UNLOCK VERIFY
|
| If you type HELP and press return, REACT will print out a list of the
| commands and a brief description of each:
|
| REACT>HELP<RET>
| Help for REACT %33B(242)
| CHANGE Change profile entries
| DELETE Delete a profile
| EXIT Exit program
| HELP Gives information on commands
| INSERT Insert a profile into the accounting file
| LIST Write profile information to a file
| LOCK Lock accounting file against updates
| PURGE Purge expired profiles
| SELECT Select wildcarding criteria
| SHOW Display profile information on the terminal
| UNLOCK Unlock accounting file to allow updates
| VERIFY Verify the accounting file from a master file
|
| If you type HELP and a command name, REACT provides a description of
| that command:
|
| REACT>HELP CHANGE<RET>
| Help for REACT %33B(242) - CHANGE
| The CHANGE command enables you to enter user mode and modify user
| profiles. The syntax is: CHANGE user-id
|
| Once you have entered enough characters to uniquely identify a
| command, you may press the ESCape key. When you press the ESCape key,
| REACT completes the rest of the command name for you, and provides
11-2
| guidewords in parentheses to help you complete the command line. For
| example, if you type 'cha' and press the ESCape key:
|
| REACT>cha<ESC>
| REACT>chaNGE (user-id)
|
| Table 11-2 provides a complete list of top-level commands and their
| functions.
|
| When you are in user-mode, type a question mark to display commands:
|
| REACT>CHANGE [27,150]<RET>
| USER>? one of the following:
| ACCESS-TYPES CONTEXT-QUOTAS CORE-LIMITS
| DISTRIBUTION-LOCATION ENQ-DEQ-QUOTA EXPIRATION-DATE
| IPCF-QUOTAS LOGIN-TIMES MAILING-ADDRESS
| NAME PASSWORD PERSONAL-NAME
| PRIVILEGES PROFILE-DEFAULT PROGRAM-TO-RUN
| REQUIREMENTS SCHEDULAR-TYPE SPOOLED-DEVICES
| STRUCTURE-QUOTAS WATCH-BITS
| or one of the following:
| DEFAULT DONE HELP QUIT RESTORE SHOW
|
| In user-mode, you may ask for HELP on specific commands:
|
| USER>HELP ACCESS-TYPES<RET>
|
| ACCESS-TYPES specifies the types of access allowed to the
| specified user. ACCESS refers to any attempt by a user or in
| behalf of a user to gain access to the system using a correct
| combination of user-id and password.
|
| You may also use the ESCape key when you are in user-mode. For
| example, if you type 'acc' and press the ESCape key:
|
| USER>acc<ESC>
| USER>accESS-TYPES (allowed)
|
|
|
| 11.4 DISPLAYING USER ACCOUNT INFORMATION
|
| The SHOW command displays information about users listed in the
| accounting file. To display a user's profile, type the following:
|
| REACT>SHOW user-id/switch<RET>
|
| Where user-id can be the user name or PPN of an existing account
| profile. You may optionally include one of the following switches:
|
| /DETAIL lists all user profile information (default)
| /FAST lists only the user-name and PPN
|
| Characteristics of the profile that have not been changed from the
| default are designated by an asterisk at the beginning of the line.
11-3
| An example of a user profile is included in Section 11.5.2.
|
|
|
| 11.5 CREATING NEW ACCOUNTS
|
| The INSERT command allows you to create new accounts, singly or in
| groups, and to define the attributes of the accounts you create. The
| attributes you define for a user's account make up the user's profile.
| You must specify a user-id when you use the INSERT command.
|
|
|
| 11.5.1 Creating Multiple Accounts with the Same Project Number
|
| You can create a group of accounts with the same project number and
| different programmer numbers, using the INSERT command. To do this,
| you supply a wildcard character (*) for the programmer number.
|
| The example below shows how to create a group of 5 accounts with the
| project number 25, programmer numbers 100 through 500, using the
| default profile, with names and passwords supplied by REACT.
|
| .R REACT<RET>
| REACT>INSERT [25,*]<RET>
| Base programmer number: 100<RET>
|
| Supply the first programmer number in the series you want to create.
|
| Programmer number increment: 100<RET>
|
| Supply the number to increment each programmer number. By selecting
| 100, the programmer number for the first account, n, will be 100; for
| the second account, n+100; for the third account n+100+100; and so on.
|
| Count of PPNs to insert: 5<RET>
|
| Supply the number of PPNs you wish to create.
|
| Prompt for user's name? N<RET>
|
| Respond YES if you want REACT to prompt you for the user name for each
| account you create. If you respond NO, REACT generates a user name
| for each account it generates in this series, but you may change the
| name later.
|
| Prompt for password? N<RET>
|
| Enter YES if you want REACT to prompt you for the password for each
| account you create. If you enter NO, REACT generates a random
| password for each account it creates in this series, but you may
| change the password later. REACT creates the accounts:
|
11-4
| [READPL Default profile loaded: [%,%]]
| User [25,100] 25,100 inserted with password GLAMDIRK
| User [25,200] 25,200 inserted with password SNAKAIZ
| User [25,300] 25,300 inserted with password RENGLYPCH
| User [25,400] 25,400 inserted with password IZMOPWIQ
| User [25,500] 25,500 inserted with password FILZNIKR
| A total of 5 users inserted
| REACT>
|
| To SHOW the accounts you just created, type the SHOW command followed
| by the [project number, wildcard character]/SWITCH and press RETURN:
|
| REACT>SHOW [25,*]/FAST<RET>
| 25,100 25,100
| 25,200 25,200
| 25,300 25,300
| 25,400 25,400
| 25,500 25,500
| A total of 5 users listed
| REACT>
|
|
|
| 11.5.2 Creating a Single New User Account
|
| To create a single new user account, type INSERT, followed by the
| user-id. If you aren't sure which programmer numbers are available
| within a specific project number group, you can use the SHOW command
| combined with the switch /FAST to suppress profile information:
|
| REACT>SHOW [27,*]/FAST<RET>
| 27,100 OLDUSER
| 27,200 VERYOLDUSER
| 27,300 ANCIENTUSER
| A total of 3 users listed
| REACT>
|
| When you create a new account, the user-id may be a user name or PPN.
| After entering the user name, you may supply an existing user-id for
| REACT to use as a model for the new profile. If you do not supply an
| existing user-id, REACT uses a default profile to establish initial
| values for the account attributes. The following examples use the
| name NEWUSER for the new-user-id, and, in the second example, OLDUSER
| for the existing-user-id:
|
| REACT>INSERT NEWUSER<RET>
| New user PPN: [27,150]<RET>
| or
| REACT>INSERT NEWUSER=OLDUSER<RET>
| New user PPN: [27,150]<RET>
|
| If the new-user-id has already been assigned, REACT prints an error
| message:
|
| %REANAE Name NEWUSER is already taken by [P,PN]
11-5
| REACT>
|
| If the new-user-id is valid, but the PPN you have specified is already
| assigned, REACT prints an error message:
|
| %REAPAE PPN [P,PN] is already taken by OTHERUSER
| REACT>
|
| If both the new-user-id and the PPN are valid, and if you have not
| supplied an existing user-id to use as a model, REACT uses a default
| profile to establish initial values for the account attributes. After
| either of these have been done, REACT enters user mode and you can
| make changes to any of the attributes that do not fit the requirements
| of the new user.
|
| The following example shows how to create an account for user NEWUSER
| using the default profile:
|
| REACT>INSERT NEWUSER<RET>
| New user PPN: [27,150]<RET>
| [READPL Default profile loaded:[27,%]]
| USER>
|
| To display the default profile attributes of NEWUSER's account, type
| the SHOW command and press RETURN at the USER> prompt:
|
| USER>SHOW<RET>
| PPN: [27,150]
| User name: NEWUSER
| Profile default: [27,%] or [%,%]
| * Personal name: -none-
| * Distribution location: -none-
| * Mailing address: -none-
| * Expiration date: -none-
| * LOGIN times: Weekdays 0:23 Weekends 0:23
| * Access types: Card reader, Network file access, Local, ANF
| CTY, Dataset, Remote, Subjob of batch, Batch
| * Requirements for LOGIN:
| Account and remark strings are not required
| Name is not required
| Password under timesharing and batch
| Password change at next LOGIN
| Minimum password length: 6
| Password change interval: Every 90 days
| Password changes are allowed
| * Schedular type: 0
| * Program to run: -none-
| * Context-quotas: Context 4, Total pages 512
| * Core limits: Physical 512, Virtual 512
| * ENQ/DEQ quota: 100
| * IPCF quotas: Send 2, Receive 5, PIDs 2
| * Privileges: Disk priority: 3, HPQ: 15, System operator,
| ENQ-DEQ, IPCF, LOCK, METER, CPU, SPY-all-core, SPY-monitor,
| Unspooling
| * Spooled device bits: CDP, CDR, LPT, PLT, PTP
11-6
| * Watch-bits: MTA
| * Structure quotas:
| Structure Quota in Quota out Reserved Status
| --------- -------- --------- -------- ------
| DSKB 80000 20000 0
| DSKC 80000 20000 0
| Administrative data:
| Profile last changed by [27,300] at 29-Sep-87 14:33:20
| Last access -never-
| Last password change 29-Sep-87 14:33:20
|
| Complete the insertion with the command DONE:
|
| USER>DONE<RET>
| User [27,150] newuser inserted with password poobah
| A total of one user inserted
| REACT>
|
| To modify any of the attributes of NEWUSER's account, see the
| description of user-mode commands in the next section.
|
|
|
| 11.6 CHANGING ACCOUNTS
|
| The CHANGE command allows you to modify the attributes of an existing
| user profile. Before you change a user's profile, you may want to
| display the profile to check the current values and settings. Use the
| SHOW command and supply a user-id to display a profile.
|
| To modify user NEWUSER's profile, type CHANGE and the user-id or
| [P,PN] and press RETURN. After you have made changes, enter DONE and
| press RETURN. The following example shows how to change the
| expiration date of NEWUSER's account:
|
| REACT>CHANGE [27,150]<RET>
| USER>? one of the following:
| ACCESS-TYPES CONTEXT-QUOTAS CORE-LIMITS
| DISTRIBUTION-LOCATION ENQ-DEQ-QUOTA EXPIRATION DATE
| IPCF-QUOTAS LOGIN-TIMES MAILING-ADDRESS
| NAME PASSWORD PERSONAL-NAME
| PRIVILEGES PROFILE-DEFAULT PROGRAM-TO-RUN
| REQUIREMENTS SCHEDULAR-TYPE SPOOLED-DEVICES
| STRUCTURE-QUOTAS WATCH-BITS
| or one of the following:
| DEFAULT DONE HELP QUIT RESTORE SHOW
| USER>EXPIRATION-DATE 31-DEC-89<RET>
| USER>DONE<RET>
| User [27,150] newuser changed
| A total of one user changed
| REACT>
11-7
| 11.7 DELETING ACCOUNTS
|
| The DELETE command removes a user account from ACTDAE.SYS. You can
| use wildcards in a PPN to delete multiple accounts. The following
| example deletes user NEWUSER's account.
|
| REACT>SHOW NEWUSER/FAST<RET>
| 27,150 NEWUSER
| A total of one user listed
| REACT>DELETE [27,150]<RET>
|
| REACT responds with:
|
| User [27,150] newuser, expiration date: 31-Dec-89
| Are you sure?
|
| This question prevents you from accidentally deleting accounts. You
| can respond to this question with NO, QUIT, SHOW, or YES. Use NO or
| QUIT if you do not want to delete the account. Use SHOW to display
| the profile (the question "Are you sure?" will be repeated). Type YES
| only if you want to delete the account. If you type YES, REACT prints
| the following messages:
|
| Are you sure? YES<RET>
| User [27,150] newuser deleted
| A total of one user deleted.
| REACT>
|
| Use the SHOW command to check that you have deleted user NEWUSER's
| account.
|
| REACT>SHOW NEWUSER<RET>
| ?REANSU No such user newuser
|
|
|
| 11.8 VERIFYING CHANGES TO ACCOUNTING FILES
|
| To use the VERIFY command, you must first have a copy of ACTDAE.SYS in
| your disk area to use as a master accounting file. The default file
| name for this copy is MASTER.SYS. The VERIFY command is used to check
| for differences between the master file (MASTER.SYS) and the system
| file (ACTDAE.SYS), and to update MASTER.SYS to reflect valid changes
| that have been made to ACTDAE.SYS. You can list the differences
| between the two files and then decide whether or not you want REACT to
| update MASTER.SYS.
|
| Before issuing the VERIFY command to REACT, you must lock ACTDAE.SYS.
| Issuing the LOCK command opens ACTDAE.SYS in read-only mode and
| prevents other users from making changes to it while you are updating
| the master file. To lock ACTDAE.SYS, type LOCK and press RETURN:
|
| REACT>LOCK<RET>
| [REALOK Accounting file locked; changes are prohibited]
| REACT>
11-8
| To use the VERIFY command, type the following command to REACT:
|
| REACT>VERIFY master-file user-id/switch<RET>
|
| You may include a user-id (user name or PPN) after the master-file
| name, to check only that user profile against the same user profile in
| ACTDAE.SYS. If you include a PPN, you can also use the wildcard
| character (*) to VERIFY a group of accounts.
|
| You may also include one or more of the following switches:
|
| /REPORT Lists the account attributes of user profiles that
| are different from those in the system file. This
| action is taken by default.
|
| /NOREPORT Does not list the account attributes of the user
| profiles that are different from the system file.
|
| /UPDATE Allows you to update the master file with changes
| from the system file. This is the default, but
| after listing the differences between the master
| file and the system file, REACT will ask whether
| or not you want to UPDATE the master file.
|
| /NOUPDATE Does not allow you to update the master file with
| changes from the system file.
|
| To list differences throughout the entire master file, for all users:
|
| REACT>VERIFY MASTER.SYS<RET>
|
| Master file: DSKC:MASTER.SYS[1,2]
| Accounting file: DSKA:ACTDAE.SYS[1,4]
| Temporary file: DSKC:039REA.TMP[1,2]
| Verifying: [*,*] /REPORT /UPDATE
|
| REACT lists the profile information for any user profiles that are
| different from the system accounting file. This listing is identical
| to the display from the SHOW command, illustrated in Section 11.5.2.
| Then, to make sure you want to UPDATE the master file, REACT prints
| the following question:
|
| Preserve changes?
|
| You can respond with NO, QUIT, SHOW, or YES. If you want to delete
| the changes from your master file, type NO or QUIT. To display the
| changes, type SHOW. If you want to keep the changes, type YES.
|
| Then, unlock ACTDAE.SYS:
|
| REACT>UNLOCK<RET>
| [REAUNL Accounting file unlocked; changes are permitted]
11-9
| 11.9 USING ACCOUNT VALIDATION
|
| The USAGE file accounting system provides ASCII files containing data
| useful in billing users for system usage. The accounting daemon (the
| ACTDAE program) can provide validation of account strings when users
| log in, and write usage data in the USAGE files.
|
| For details on the format of the USAGE file entries, refer to the
| TOPS-10/20 USAGE File Specification. The following sections describe
| the procedure for enabling account validation.
|
| To enable account validation, do the following:
|
| 1. Answer Y to Question 10 in the SYSGEN portion of MONGEN
|
| 2. Require account validation for LOGIN for each PPN in
| ACTDAE.SYS. See Section 11.6.
3. Create the account validation file PROJCT.ACT.
4. Run the program PROJCT.EXE to convert PROJCT.ACT into
PROJCT.SYS.
5. Copy PROJCT.SYS into ersatz device SYS: (PPN is [1,4]).
11.9.1 CREATING THE PROJCT.ACT FILE
PROJCT.ACT is an ASCII file you can create to associate account
strings with PPNs. The format of each line in PROJCT.ACT is:
[P,PN]=ACCOUNT1,ACCOUNT2,ACCOUNT3,...
You can use the * and ? wildcard characters in the PPNs. You can
substitute the * for either the project number or programmer number,
or both. You can substitute the ? for a single digit in the PPN.
| For example, the PPN definition [27?,3007] would define account
| strings for [27n,3007], where 'n' equals any number from 0 to 7. The
| PPN definition [27,*] would define account strings for all programmer
| numbers with the project number 27, or [27,n] through [27,nnnnnn].
|
A single PPN can be defined to accept any number of account strings,
each of which is an ASCII sequence of up to 39 alphanumerical
characters. The account strings must match user input (at LOGIN),
exactly, including the casing of letters. If the account string
definition in PROJCT.ACT is all uppercase, for example, the user must
type the account string in all uppercase form.
You can allow the user to log in without typing an account string by
defining a default account string for the PPN. Use the /DEFAULT
switch after one of the account strings that is valid for the user's
PPN. Then, when the user logs in, (s)he can simply press RETURN in
11-10
response to the ACCOUNT: prompt. The session will be charged to the
default account string. For example, the following account definition
allows the user to default accounting data to the first account string
in the definition (PUBLICATIONS):
[27,5434]=PUBLICATIONS/DEFAULT,MARKETING,MAINTENANCE
The PPN entries in the PROJCT.ACT file must occur in numerically
ascending order. In the case of wildcard characters, the * is
equivalent to 777777, and the ? is equivalent to a 7. In case of a
conflict between a wildcarded number and an explicit number, the
wildcarded PPN must follow the explicit PPN. For example, the
following order is valid:
[10,10]=ABC
[10,2370]=DEF,GHI
[10,23??]=FOO
[10,*]=GHI
[*,*]=JKL
Note that, under this scheme, the user with PPN [10,10] may not log in
with account string JKL, because an explicit definition overrides any
subsequent definitions. However, the user with PPN [10,2370] can log
in with account string GHI, because any number of account strings can
be valid for a single PPN.
Account strings can also contain wildcard characters. The * in the
account string definition denotes account strings with any number of
characters (or no characters) to replace it. However, when the ? is
used to replace a character, it must be matched by a character.
Therefore, the account string ABC??FG* can be matched only by account
strings containing ABC for the first three characters, followed by any
two characters, containing FG for the sixth and seventh characters,
followed by no more or any number of characters.
11.9.2 CREATING THE PROJCT.EXE FILE
After you create PROJCT.ACT, and ensure that it exists in your
directory path, run the PROJCT.EXE program as shown below:
.R PROJCT<RET>
END OF JOB
EXIT
.
If an error occurs during processing the PROJCT.ACT file, PROJCT
displays a message indicating the field or problem in the input file.
Edit the PROJCT.ACT file to correct the problem, and run PROJCT again.
When successful, PROJCT creates the PROJCT.SYS file. Copy this file
into SYS:
When account validation has been enabled, the user will see an
11-11
additional prompt in the LOGIN dialogue (ACCOUNT:), after the
PASSWORD: prompt. See the TOPS-10 Operating System Commands Manual
for more information.
11.10 MAINTAINING LOGIN FAILURE RECORDS
REACT uses the FAILUR.LOG file to maintain records on login failures.
All records in the FAILUR.LOG file are written in 7-bit ASCII, and
vary in length. All validation failure entries have the same format
for the first eleven fields. The following table explains the first
eleven fields of the FAILUR.LOG file:
Table 11-1: FAILUR.LOG File Format
# Field Pos Data Type Comments
001 Error code 001 Octal Error code defined in ACTSYM
002 Usage style
Date/Time 003 Decimal Same format as usage entries
003 Job number 017 Decimal Job number of QUEUE./IPCF sender
004 PPN (LH) 020 Octal PPN (left half) of above job
005 PPN (RH) 026 Octal PPN (right half) of above job
006 Username 032 ASCII 12 charcacter name of above job
007 Program
name 044 ASCII 6 character program name
008 TTY name 050 ASCII Example: TTY45
009 Node name 056 ASCII Node name of job
(ANF-10, DECnet, or LAT)
010 Line name 072 ASCII Line name on above node
011 APC code 075 Octal Asynchronous port characteristic
The following table shows the format for fields that end the Invalid
account string (1) and Invalid password (6) error entries.
# Field Pos Data Type Comments
012 PPN (LH) 077 Octal PPN (left half) whose account
string was invalid.
013 PPN (RH) 083 Octal PPN (right half) as above
The job not privileged entry (3) only contains the first 11 fields
11-12
because the job number is already specified.
11.11 TOP-LEVEL REACT COMMANDS
The following table describes the basic action Commands for creating,
changing, and deleting accounts. It also describes commands for
updating the master accounting file and displaying user information.
Table 11-2: Top-level REACT Commands
Command Function
CHANGE userid Enables you to enter user-mode and
modify user profiles.
DELETE userid/ASK (default)
/NOASK Removes the specified account from
ACTDAE.SYS. You can specify a
wildcard in the userid. /ASK is the
default. /ASK causes REACT to
prompt you to confirm each deletion
you make. If you specify /NOASK,
REACT does not prompt you to confirm
each deletion.
EXIT Exits the REACT program and returns
you to monitor command level.
HELP Displays information about REACT
commands. To display a brief list
of REACT top-level commands, type
HELP and press RETURN. To get more
information about a particular REACT
command, type HELP and the command
name.
INSERT userid [=userid] Enables you to enter user-mode and
create user accounts. You can
supply an optional existing userid
to use as a guide for the new
profile.
LIST userid filespec/switch Writes information from the
specified user's profile to a file.
You can specify a wildcard in the
userid. The default file
specification is DSK:ACCT.LST[-].
Switches are: /DETAIL, which lists
all the information in a user's
profile, and /FAST, which lists only
the PPN and its associated username.
/DETAIL is the default.
11-13
LOCK Opens ACTDAE.SYS in READ-ONLY mode,
to prevent other users from making
changes to the file while you update
the master accounting file.
|
| PURGE Deletes expired profiles.
SELECT Allows you to mark accounts that
have certain profile attributes in
common. You use the SELECT command
to specify the attributes that you
want as identifiers for the group.
After you mark the accounts as a
group, you can change various
attributes of the profiles as a
group, rather than doing so
individually.
For example, to extend the
expiration date of all accounts that
expire on December 31, you do the
following: SELECT the accounts that
have the expiration date you want to
change, and CHANGE the expiration
date to the desired date.
SHOW userid Displays information from the
specified user's profile. You can
specify a wildcard in the userid.
Switches are: /DETAIL, which
displays all the information in a
user's profile, and /FAST, which
displays only the PPN and its
associated username. /DETAIL is the
default.
UNLOCK Reopens ACTDAE.SYS so that other
users may access it.
VERIFY [filespec] user-id/switch
| Checks for differences between a
| master accounting file (MASTER.SYS)
| and the system file (ACTDAE.SYS).
| You can specify a particular user
| account or verify all PPNs. This
| command allows you to list the
| differences between your master file
| and the system accounting file, and,
| optionally, update the master
| accounting file. The user-id can be
| either a [P,PN] or a user name. If
| the username includes characters
| other than A-Z, a-z, hyphen (-) or
| digits, you must enclose it in
11-14
| double quotes.
11.12 USER-MODE COMMANDS
User-mode commands are subcommands to the CHANGE, INSERT, and SELECT
commands, and consist of the following types of commands:
|
| o COMMANDS to perform the changes specified with Field Change
| Commands. These commands are listed in Table 11-2.
|
| o COMMANDS to change customer-defined fields that specify the
| customer-defined user account field that you want to change.
| You can define customer-defined FIELDS in REACTU.MAC to
| tailor the REACT program to the special needs of your
| installation.
|
| o COMMANDS to change account fields that specify the user
| account field that you want to change. These commands are
| listed in Table 11-3.
|
The following prompt indicates that you have entered user mode:
USER>
The following table lists REACT Commands to change fields.
Table 11-3: REACT Field Change Commands
Command Function
ALL Allows you to specify all of the options for
a given field change command.
AND Allows you to specify selection criteria
with the SELECT command.
ACCESS-TYPES Specifies the types of access allowed to the
specified user. An access is any attempt by
a user, or in behalf of a user, to gain
access to the system using a correct
combination of user-id and password.
ACCESS-TYPES include following: ANF-CTY,
BATCH, DATA-SET, FILES-ONLY, LOCAL,
NETWORK-FILE-ACCESS, PHYSICAL-CARD-READER,
REMOTE, SUBJOB-OF-BATCH.
CONTEXT-QUOTAS Specifies the limits governing the use of
job contexts. The context quota is the
number of contexts a user may have at any
one time. Each logged in job has at least
one (current) context. The maximum is 511.
A quota of zero indicates no limit. The
saved page quota is the number of pages of
11-15
swapping space a user may occupy with idle
contexts. A quota of zero indicates no
limit. Refer to the description of job
contexts in the TOPS-10 Operating System
Commands Manual for more information.
CORE-LIMITS Specifies a decimal value for the physical
and virtual limit of the number of pages.
The maximum number of physical pages is
8,192. The maximum number of virtual pages
is 16,384.
DISTRIBUTION-LOCATION Specifies text to be displayed on the banner
page(s) of spooled output. The text
indicates where the operator should
distribute the user's output.
DONE Writes the changes to the ACTDAE.SYS file.
ENQ-DEQ-QUOTA Specifies a decimal value for the number of
outstanding ENQ locks. The maximum number
is 511.
EXPIRATION-DATE Specifies the date and time when LOGINs to
this account are no longer allowed.
IPCF-QUOTAS Specifies a decimal value for the SEND,
RECEIVE, and PID quotas. The maximum value
for each quota is 511.
HELP Displays help text for the specified
user-mode command.
LOGIN-TIMES Specifies the time of the day the user is
allowed to LOGIN. Weekdays are divided into
24 one-hour segments. Weekends are divided
into 12 two-hour segments. Therefore, if a
user is permitted to LOGIN on Saturday at
7:00 AM, the user can actually LOGIN between
6:00 AM and 7:59 AM.
NAME Specifies the username for this profile.
The username can consist of any printable
character except the following: %, *, or
/ . It can be up to 39 characters long, and
it cannot begin with a numeric character.
NO Allows you to specify options to field
change commands
NONE Allows you to disallow all of the options to
a given field change command.
NOT Allows you to specify selection criteria
with the SELECT command.
11-16
OR Allows you to specify selection criteria
with the SELECT command.
PASSWORD Specifies a password the user must type to
gain access to the system. The password can
consist of any printable characters, and can
be up to 39 characters long.
PERSONAL-NAME Specifies a name other than the user's
offical accounting name (see NAME). The
personal name, if available, is printed on
the banner page(s) of a user's spooled
output. This is an unprivilged entry in a
profile, and as such may be modified by the
user.
PPN Specifies the user's [project, programmer
number]. The [P,PN] is the primary means of
identifying a user.
PRIVILEGES Specifies the privileged functions allowed
to the user. The privileged functions
are: CPU, DISK-PRIORITY, ENQ-DEQ, HPQ,
IPCF, LOCK, METER, MON-SPY, OPERATOR, POKE,
RTTRP, SPY, TRPSET, UNSPOOL
PROGRAM-TO-RUN Specifies the file specification of the
program to run after LOGIN is complete. The
file specification may include a device
name, program name, extension, and
directory, including up to five levels of
sub-file directories (SFD).
PROHIBIT Enables the system administrator to prevent
users from changing their passwords. This
is useful in situations where several people
share a single PPN.
|
| QUIT Terminates the current user-mode command
| without executing that command.
REQUIREMENTS Specifies additional information the user
must supply when logging in. Options are:
o ACCOUNT-STRING information associated
with the user
|
| o INTERVAL maximum time between password
| changes
|
| o LENGTH minimum characters for password
o NAME under timesharing or batch
11-17
o PASSWORD under timesharing or batch
|
| o PROHIBIT password changes
o REMARK to be stored in the usage files
RESTORE Deletes any changes you have made to a
profile in the current user-mode session,
and returns all attributes to their original
values.
SCHEDULER-TYPE Associates the specified user profile with a
group that has been assigned a scheduler
class. (See Appendix C for information
about assigning scheduler classes.)
SHOW Displays the current fields for the
specified user profile.
SPOOLED-DEVICES Specifies the physical devices not normally
available to the user. The options are CDP,
CDR, LPT, PTP, PLT and ALL.
STRUCTURE-QUOTAS Specifies the structure on which the user
has disk space allocated, and the amount of
disk space allocated on the structure.
LOGIN puts these structures in the user's
search list, and sets either of the
following two I/O status bits: READ-ONLY or
NO-CREATE.
WATCH-BITS Specifies any of the following watch bits
that LOGIN sets: CONTEXTS, DAY, FILES,
MTA, READ, RUN, VERSION, WAIT, WRITE. For
more information, see the SET WATCH command
in the TOPS-10 Operating System Commands
Manual.
11-18
CHAPTER 12
CREATING SYSTEM INITIALIZATION FILES
There are several files that the monitor can use to help start up the
system, initiate terminal communication, and provide information to
the user. The files can be created by the system adminstrator. If
they exist, the INITIA program reads them and passes the information
stored in them to the monitor. The files INITIA can read are:
| Filename When
|
| SYS:SYSJOB.INI system startup
| SYS:INITIA.TXT the TEXT option is specified
| SYS:NOTICE.TXT the NOTICE option is specified
| ALL:STR.TXT[1,4] the STRUCTURES option is specified
| SYS:TTY.INI a terminal requests connection
| DSK:SWITCH.INI[p,pn] you log in
| HLP:INITIA.HLP HELP is requested
|
| Users can also run INITIA from their terminals to set up terminal
| characteristics and display information about the terminal. The
| command options that direct INITIA to read these files are listed in
| Table 12-1, Special Processing Commands.
12.1 INITIA AT SYSTEM STARTUP
INITIA runs automatically at system startup. At system startup, it
reads SYS:SYSJOB.INI, an optional system startup file created by the
system administrator. You can prevent INITIA from running
automatically at startup by using the NO INITIA startup option, as
described in Appendix A. SYSJOB.INI contains the commands necessary
to start up system programs that normally run in a detached state
(that is, programs that do not normally require terminal I/O). Some
of the system programs that may be started from SYSJOB.INI are ACTDAE,
FILDAE, and QUASAR.
SYSJOB.INI allows you to start system components quickly and
automatically. Most of the programs that can be started automatically
12-1
by OPSER, from the OPR.ATO file, can be started more easily and
efficiently by INITIA through SYSJOB.INI. SYSJOB.INI can contain
monitor commands and commands to programs, but its intended purpose is
as a system initialization file, and great care should be taken to
ensure that the file can be read successfully.
INITIA also runs automatically whenever a terminal connects to the
| system. INITIA reads the file SYS:TTY.INI to establish information
| about the terminal. TTY.INI can also contain commands to INITIA to
| interrogate the connecting terminal to determine its characteristics
| and attributes.
NOTE
INITIA cannot access files that are protected from it.
Therefore, the system files SYSJOB.INI and TTY.INI
must have a protection code of <155>. Protection
codes are described in detail in the TOPS-10 Operating
System Commands Manual.
12.1.1 The SYSJOB.INI File
If it exists, SYS:SYSJOB.INI is read by INITIA at system startup.
TOPS-10 provides several system programs that are suitable for startup
by INITIA from SYSJOB.INI. System programs commonly started from
| SYSJOB.INI are the system DAEMON programs DAEMON, FILDAE, ACTDAE; the
| GALAXY component QUASAR; user utilities such as MIC; and site-specific
system programs. Any system program that normally runs in a detached
state is a candidate for startup from SYSJOB.INI.
INITIA reads SYSJOB.INI and sends the commands contained in it to the
monitor for special processing through the FRCLIN (force-line)
mechanism. FRCLIN is an internal mechanism of the monitor that allows
a privileged program to pass commands to the monitor as though the
commands had been issued from a terminal. INITIA uses FRCLIN to start
system programs because terminal input is handled by the monitor with
a minimum of delay. After INITIA is finished, FRCLIN is free for use
by other system components that must send commands to programs.
System programs that may be started from SYSJOB.INI detect the FRCLIN
mechanism and detach themselves automatically. To detect FRCLIN, a
program must compare the number of the terminal line that sent the
startup command with Word 162 (%CNFLN) of GETTAB Table 11 (.GTCNF),
where the FRCLIN line number is stored.
The FRCLIN mechanism ignores output from the programs that are started
with FRCLIN, and for that reason it may be difficult to determine why
SYSJOB.INI fails to complete. To detect the reason for an error in
SYSJOB.INI, set the DF.WFL bit in the monitor debugging word (DEBUGF),
and restart INITIA. Output from the programs that are started will
then be printed on the CTY.
12-2
12.1.1.1 Format of SYSJOB.INI -
SYSJOB.INI is formatted as a list of the commands that can be sent
through FRCLIN to the monitor. Any monitor commands can be placed in
SYSJOB.INI, but the following are the most commonly used:
SET DEFAULT ACCOUNT string Sets a default account string by
which the jobs on FRCLIN can be
logged in.
LOGIN Logs in a job through FRCLIN to
accommodate the program that is
started on the next line of the
file.
This command does not actually run the LOGIN program or use the LOGIN
UUO. The monitor performs special logging-in actions to prepare a job
for the program. The jobs are logged in under the PPN [1,2] by
default.
It is possible to log in programs with a PPN other than [1,2]. This
may be accomplished by omitting the LOGIN command, and by specifically
omitting the R or RUN command. The program will then be run from a
job that is not logged in, using PPN [2,5]. The program is
responsible for logging itself in, if desired, by using the LOGIN UUO
and any PPN that is valid.
program Specifies the name of a system
program to run.
Each program name is placed on a line following a LOGIN command. The
program is assumed to be run from SYS. Therefore, the command
BATCON
is actually interpreted as the command
R BATCON
The actual monitor command R or RUN may be included in the command
line.
The following is an example of a SYS:SYSJOB.INI file:
SET DEFAULT ACCOUNT MUMBLE ;set the account string as "MUMBLE"
LOG ;log in a SYSJOB
ACTDAE ;run the accounting daemon
LOG
DAEMON ;run SYS:DAEMON
LOG
FILDAE ;run SYS:FILDAE
LOG
QUASAR ;run SYS:QUASAR
.
.
12-3
.
NIK/DETACH ;run a local program in detached
LOG ;mode
SET HPQ 1 ;set a high priority run queue for
R MIC ;start MIC
.
To insure that input from other FRCLIN users does not interrupt
logically related groups of commands in SYSJOB.INI, INITIA groups the
SYSJOB.INI commands into strings delimited by LOGIN commands. For
example, the following list shows commands contained in a SYSJOB.INI
file:
LOGIN
ACTDAE
LOGIN
SET HPQ 1
DAEMON
INITIA groups the above commands into two strings. INITIA copies the
first string into FRCLIN's input buffer. The first string contains
the following commands:
LOGIN<CR>ACTDAE<CR>
Then it copies the the second string into FRCLIN's input buffer. The
second string contains the following commands:
LOGIN<CR>SET HPQ 1<CR>DAEMON<CR>
NOTE
INITIA considers any program name given in SYSJOB.INI
that starts with "LOG" to be a LOGIN command. This
does not affect the content of the input, but it may
cause conflicts, which the above technique attempts to
prevent. To differentiate between LOGIN commands and
other program names that begin with the letters LOG,
use the following format:
R LOG...
If errors occur during processing, the rest of the input file may not
be executed. Therefore, the order of the commands in SYSJOB.INI is
vital to successful system startup. It is important to place monitor
components first in the file, followed by CUSPS and other programs.
You should place ACTDAE first in the file, to ensure that accounting
is enabled when the rest of the programs are started. You may include
commands to the programs you are starting. If you include commands,
you must explicitly detach the program after the commands if the
program does not detach itself as a result of the commands.
12-4
12.1.2 The TTY.INI File
Before a user can log in, the terminal must be logically connected to
the system. When a terminal requests connection, INITIA reads the
| file SYS:TTY.INI to set up terminal and job characteristics. TTY.INI
| can contain commands to INITIA to:
|
| o Interrogate the connecting terminal to determine its
| characteristics and attributes.
|
| o Set terminal characteristics by explicit command.
|
| o Set options that specify information to be associated with
| the terminal.
|
|
| NOTE
|
| If you specified ONCE startup option NO INITIA, users
| must issue either the INITIA CHECK command or SETTY
| command to activate the options in TTY.INI.
12.1.2.1 Format of TTY.INI -
Each command line of TTY.INI, except continued lines, must begin with
one of the terminal identifiers listed in the table below. By using
the terminal identifiers in combination with the options listed in
Section 12.1.4 you can specify any number of terminals in a command
line, and you may use any number of options for each terminal. If
conflicting options are specified for a single terminal, the last
option in the command line for that terminal takes effect.
To continue a command line, put a hyphen (-) as the last character of
the line that is to be continued. To insert a comment, precede the
comment by a semicolon (;) or an exclamation point (!). File line
numbers are ignored. Multiple tabs and spaces are treated as one
space.
| The following is an example of a SYS:TTY.INI file:
|
| ALL: CHECK NORUN KSYS XONXOFF TEXT NOGAG LC-
| REMOTE
| APC:CTERM NOCHECK
| STOMP ACCOUNT "MUMBLE"
| CONNECT COMET_TTY0
| KL1026_TTY405 NO REMOTE NO TABS
| KL1026_TTY406 NO REMOTE NO TABS
| KL1026_TTY407 NO REMOTE NO TABS
| CTY: TYPE:LA120 NO REMOTE NO TABS GALOPR ACCOUNT-
| "MUMBLE"
| KL1026_TTY5 RCV:9600 XMT:9600 NO REMOTE
| KL1026_TTY6 RCV:300 XMT:300 REMOTE
12-5
| KL1026_TTY405 TYPE:LA120 ;CPU2 CTY
| KL1026_TTY406 TYPE:LA120 ;CPU1 CTY
| KL1026_TTY407 TYPE:LA120 ;CPU0 CTY (when not THE CTY)
| .
| .
| .
12.1.2.2 Specifying Terminal Lines -
The following commands identify a terminal or set of terminals. This
identifier is placed first in the command line, followed by the
options that are listed in the next section.
Identifier Terminal Lines Indicated
ALL: All terminal lines.
| APC:x Asynchronous Port Characteristic. All
| terminals of APC type x where 'x' is
| determined by certain attributes and
| characteristics. This identifier can be
| one of the following:
|
| unknown
| hardwired
| dataset
| Gandalf
| TSN
| autodial
| Micom
| NRT (Network Remote Terminal)
| LAT (Local Area Terminal)
| CTERM (Command Terminal)
CTY: The CTY (controlling terminal) only.
DETECT n Dataset lines that are set to n baud.
LINSPD n Lines that are set to n baud.
node_identifier Terminal at the node called "node." The
identifier following node_ can be any of
the identifiers in this list.
OTHER Any terminals not specified with a TTY:
identifier.
OTHSPD n All lines not set to n baud.
TTYnnn: Terminal line nnn. This identifier can be
abbreviated to TTnnn:.
TTYnnn-zzz: All terminal lines from nnn to zzz. This
12-6
identifier can be abbreviated to
TTnnn-zzz:.
(TTYaaa:, TTYbbb:,...,TTYzzz:)
All terminal lines listed.
Terminal line numbers range from 0-777 octal. Hard-wired terminals
are assigned permanent terminal line numbers in the form TTYnnn:,
where 'nnn' is the physical line number. For example, if the terminal
number TTY10: is assigned to a terminal that is hardwired to the
system, that number will not change. If the host system is named
MING, this terminal may also be referred to as MING_10.
Terminals connected through nodes have permanent line numbers in the
form nnn_lll, where 'nnn' is the node name or number and 'lll' is the
number of the line that connects the terminal to the node. When one
of these remotely-connected terminals is recognized by the system, the
monitor assigns a "terminal name" to it, in the form TTYnnn:, where
'nnn' is a temporary number assigned during the connection. These
terminal names are reassigned on a dynamic basis, as terminals
logically connect and disconnect from the system.
For example, terminal line 15 on node number 44, a terminal
concentrator with node name XENON, would be called 44_15, or XENON_15,
and this designation will not change. However, when terminal 44_15 is
connected to the host system, it may be assigned terminal name TTY130:
by the monitor. The terminal will be known to the monitor as TTY130:
as long as it remains connected to the host system. If the terminal
becomes disconnected and then reconnected, it might be assigned a
different terminal name, such as TTY52:.
12.1.2.3 Specifying Terminal Options -
Terminal identifiers are followed by options that specify information
to be associated with the terminal(s) identified at the beginning of
the command line.
There are three types of options:
o Commands that specify special processing by INITIA.
o Terminal characteristics to be set for the terminal when it
is connected.
o Programs to run automatically when the terminal is connected.
The options are separated by commas or spaces. For those options that
require an argument, the argument can immediately follow the option
name, or it can be separated from the option name by a colon or a
space. Option names can be abbreviated, as long as the abbreviation
is unique.
In conflicting situations, the option specified last in the command
12-7
line takes precedence over those specified earlier. You can turn many
options off by typing NO before the option word, with or without an
intervening space. A minus sign has the same effect as NO. Thus,
NO GAG, NOGAG, and -GAG have the same effect. In the table below, the
complement of the command is shown, if a complement is valid. The
default setting of the options is designated by (*).
The following table lists TTY.INI options to indicate special
processing to be done for the terminal at initialization:
Table 12-1: Special Processing Commands
Option Complement Meaning
| ATTRIBUTE Display terminal attributes.
|
| CHECK NOCHECK(*) Interrogate terminal to determine
| CHECK:default its class, model, and type, and
| set the attributes and
| characteristics associated with
| the class, model, and type found.
| When CHECK:default is specified,
| INITIA interrogates only if
| current terminal type matches the
| system default type specified in
| MONGEN Question 132.
CONNECT -- Connect networked terminal to the
host.
| HELP -- Read the file HLP:INITIA.HLP and
| print information about the INITIA
| command.
KSYS NOKSYS(*) Display KSYS time if it has been
set.
LOCATE node Make a logical connection with the
specified node, similar to the
monitor LOCATE command.
NAME(*) NONAME Print the system name.
NORUN -- Suppress the running of programs
that are set to run automatically
for the terminal.
NOTICE NONOTICE(*) Print SYS:NOTICE.TXT, if such a
file exists.
SETTTY NOSETTTY Set terminal from TTY.INI (on CTY)
or SWITCH.INI. If the job is
logged in, NOSETTTY is the
default. For more information
12-8
about SETTTY and SETTTY options,
see the TOPS-10 Operating System
Commands Manual.
STOMP -- Initiate the "terminal stomper"
function that disconnects terminal
lines that have been detached and
idle for two or more minutes.
STRUCTURES NOSTRUCTURES(*) Print individual structure notes
from ALL:STR.TXT[1,4].
TEXT NOTEXT(*) Print SYS:INITIA.TXT, if such a
file exists.
TTY NOTTY(*) Display all terminal settings for
this terminal.
| TTY:TTYnnn Display all terminal settings for
| TTYnnn.
|
|
| The following programs start automatically when INITIA runs for the
| terminal. A job is logged in for the program and INITIA creates a PPN
| for the job. Unless specified otherwise, INITIA constructs PPNs as
| follows:
|
| o If the terminal is a local terminal, INITIA uses [1,2].
|
| o If the terminal is remote, INITIA constructs a PPN using the
| algorithm [100+node-number,2]. For example, a terminal at
| the node that has node number 12 receives the PPN [112,2].
|
| INITIA creates a job search list for the job, which it copies
| from the current setting of the system search list.
|
|
| Table 12-2: Options to Run Programs
|
|
| Option Program
|
| AUTOMATIC Start OPSER and begin processing an automatic
| startup file. The file name that is read depends
| on the terminal from which OPSER was started.
|
| When the terminal is connected directly to the
| local system:
|
| o If device OPR:, OPR.ATO is used.
|
| o Otherwise, TTYnnn.ATO is used, where 'nnn' is
| the terminal line number.
|
| When the terminal is connected to a remote node:
12-9
| o If device OPR:, OPRnnn.ATO is used, where
| 'nnn' is the node number.
|
| o Otherwise, nnnlll.ATO is used, where 'nnn' is
| the node number and 'lll' is the terminal line
| number.
|
|
| CHKPNT Start CHKPNT under [1,2]. This program cannot be
| run from a remote terminal.
|
| CRASH Start FILEX under [10,1].
|
| DAEMON Start DAEMON under [1,2]. This program cannot be
| run from a remote terminal.
|
| GALOPR Start the OPR program, which begins processing the
| file SYS:SYSTEM.CMD, if the terminal is local. If
| the terminal is remote, SYS:nnnlll.CMD is used,
| where 'nnn' is the node number and 'lll' is the
| line number of the terminal where OPR is started.
|
| OMOUNT Start OMOUNT under [1,2]. This program cannot be
| run from a remote terminal.
|
| OPSER Start OPSER.
|
| STOMP Initiate the "terminal stomper" function, which
| disconnects terminal lines that have been detached
| and idle for longer than two minutes.
|
| SYSDPY Start the SYSDPY display program for the
| appropriate terminal type.
12.1.2.4 Associated (Error) Messages -
If there are errors in the TTY.INI file, INITIA prints:
?INTCER INITIA COMMAND ERROR
Correct the TTY.INI file and restart INITIA for the job. Any other
?INT... error message indicates a serious system problem that must be
investigated by a software specialist.
12.1.3 The INITIA Monitor Command
The INITIA monitor command enables users to initialize their terminals
at any time. To run INITIA from a terminal, use the INITIA monitor
command. You may include the INITIA command and your choice of
options in your SWITCH.INI file.
12-10
You can use INITIA to do the following:
|
| o Display terminal attribute settings, mode settings, and check
| terminal type.
|
| o Display the system banner and time of system shutdown (KSYS),
| if any.
|
| o Display general system notices.
|
| o Display operator notes about mounted structures.
|
| o Change terminal characteristics.
|
See the TOPS-10 Operating System Commands Manual for more information
about the INITIA command and the SWITCH.INI file.
12-11
PART 5: SUPPLEMENTARY INFORMATION
This part contains four appendixes. Appendixes A
though D describe alternatives to the installation
procedure and information for modifying the
monitor software.
APPENDIX A
ONCE STARTUP OPTIONS
| ONCE runs automatically at system startup. At system startup, the
| initial dialog questions allow you to provide a reason for the system
| reload and to specify the date and time. While the system is up and
| running, ONCE may also be run in user mode by running the monitor.
| You may refresh disks and change disk parameters when ONCE runs at
| system startup, or you may run ONCE in user mode without reloading the
| system.
|
| ONCE performs the following functions:
|
| o Checks memory for units that are offline and confirms the
| state of the memory units by a response from you.
|
| o Checks the HOME blocks on all the units for inconsistencies
| and optionally rewrites the HOME blocks. If any units are
| write-protected, ONCE reports the unit name, asking whether
| it should be write-enabled or write-protected.
|
| o Reports on disk units that are offline.
|
| o Checks the BAT blocks, listing any inconsistencies for each
| unit.
|
| o Lists the number of bad regions on any unit you specify.
|
| o Lists each file structure and the units in the file structure
| in their logical order.
|
| o Lists all the disk units that are not in a file structure.
|
| o Lists the units in the Active Swapping List. For new
| installations, there is no swapping list, and ONCE prints a
| message to indicate that.
|
| o Lists the structures in the System Search List, or SSL.
A-1
ONCE STARTUP OPTIONS
| o Lists the structures in the System Dump List, or SDL.
|
| o Allows you to change parameters of any structure.
|
| o Allows you to change parameters of any unit.
|
| o Allows you to change any system disk parameters.
|
| o Allows you to dissolve file structures, if any file
| structures exist.
|
| o Allows you to define file structures.
|
| o Allows you to specify the units on which you want HOME blocks
| rewritten.
|
| o Lists the structures that need to be refreshed, and allows
| you to specify the structures that you want refreshed.
|
| o Allows you to log in as [1,2], after ONCE has refreshed
| structures.
|
| o Starts running the null job and begins timesharing.
|
|
| The Disk Parameters Worksheet, Worksheet 3, is provided in Chapter 2
| to record the disk parameters for your system. Because you can define
| so many disk parameters with ONCE, you may find it useful to record
| the disk parameters for your system on the worksheet and to use this
| as an aid when you refresh structures, create new file structures, or
| redefine disk parameters.
|
|
|
| A.1 ONCE AT SYSTEM STARTUP
|
| ONCE begins when you reload the system, as described in Section 9.4.
| First, the system name and the date that the monitor was created are
| displayed. The system name is set by Question 5 of the SYSGEN portion
| of MONGEN (see Chapter 9).
|
| [Loading from DSKM:RL336A.EXE[1,4]]
|
| RL336A DEC-10 Development 16-Feb-88
|
| ONCE scans memory to be sure that it is accessible. Each monitor is
| built for a specific amount of memory. MONGEN Question 58 allows you
| to specify the amount of memory when you are building the monitor.
| Therefore, new installations may receive the following message,
| indicating that the default monitor was built to expect more memory
| than that which is currently online:
A-2
ONCE STARTUP OPTIONS
| %Memory from xxxxxx to yyyyyy is offline
| Set memory (DOWN,OFFLIN,ONLINE) [OFFLIN]:
|
| The default response appears in square brackets. If the memory
| indicated is not needed at this time, but will be needed later, press
| RETURN to accept the default answer. Memory that is set offline can
| be set online during system operation, with the SET MEMORY command.
| If all memory must be online, enter ONLINE and press RETURN. If the
| memory is not going to be required at all, enter DOWN and press
| RETURN. Memory that is down cannot be brought up with the SET MEMORY
| command.
|
| If you are starting an SMP system and have not yet started the other
| CPUs, ONCE prints the following message:
|
| %CPU #nnnn is not running. Set CPU status (DOWN,UP) [DOWN]:
|
| If you have an SMP system, you can start the CPU indicated by entering
| UP and pressing RETURN. To accept the default value, just press
| RETURN. After receiving an answer, ONCE asks for the date by
| printing:
|
| Date:
|
| Enter the date in the form:
|
| DD-MMM-YY
| or
| MMM-DD-YY
|
| Where DD is the numerical day of the month (01 through 31), MMM is the
| name of the month (JANUARY through DECEMBER) abbreviated as the first
| three letters of the month (JAN through DEC), and YY is replaced by
| the two digit year designation. The year is optional.
|
| If you respond with an incorrectly formatted date, ONCE asks the
| question again in more detail:
|
| Please type today's date as MMM-DD(-YY)
|
| If you specify a date before the last system crash or before the
| creation date of the monitor, the ONCE dialog issues an error message,
| supplies the last crash or creation date, and asks you to verify your
| response:
|
| %Specified date is prior to last crash.
| Last crash date:MMM-DD-YY
| or
| %Specified date is prior to monitor creation date.
| Creation date:MMM-DD-YY
| and
| Specified date:MMM-DD-YY
A-3
ONCE STARTUP OPTIONS
| Is this correct? (Y or N)
|
| To which you can respond Y if the date you typed was correct, or N, if
| you typed an incorrect date. If you enter N, the Date: prompt is
| reprinted so that you can enter the correct date.
|
| ONCE asks for the time after it receives the date:
|
| Time:
|
| Enter the time based on a 24-hour clock. The seconds are optional, as
| are the colons separating the hours, minutes and seconds.
|
| HH:MM:SS
| or
| HHMMSS
|
| If the format of the time is incorrect, ONCE prints the following
| message:
|
| Please type time as HHMMSS or HH:MM:SS
|
| If you specify a time before the last system crash, the ONCE dialog
| issues an error message, supplies the last crash time, and asks you to
| verify your response:
|
| %Specified time is prior to last crash.
| Specified time:HH:MM:SS
| Last crash time:HH:MM:SS
|
| Is this correct? (Y or N)
|
| If you enter N, the Time: prompt is reprinted so that you can enter
| the correct time.
|
| As with the date, it is important that you enter the correct time.
| The system date and time can be changed with the CHANGE DATE startup
| option; the system time can be changed with the SET DAYTIME monitor
| command. However, changing either the system date or time during
| system operation may adversely affect operation. Therefore, it is
| recommended that you reload the system if you find it necessary to
| change them.
|
| After receiving the date and time, ONCE then asks the question:
|
| Why reload:
|
| This question is asked whenever the system is reloaded. Several
| responses are allowed, depending on the reason for the reload. When
| you are installing the monitor, use one of the following answers:
|
| NEW if you are installing a new system
A-4
ONCE STARTUP OPTIONS
| SCHED if you are loading a new or different version of the
| monitor
|
| When you are reloading the system after a problem or crash, use one of
| the following answers:
|
| HALT if the system ended with a stopcode or HALT
|
| HARDWARE if the system ended with a hardware malfunction
|
| HUNG if the system would not respond
|
| LOOP if the monitor was looping in executive mode
|
| NXM if a non-existent memory stop occurred
|
| OPR if an operator error occurred
|
| PARITY if a memory parity error occurred
|
| POWER if a power failure occurred
|
| STATIC if static electricity caused the crash
|
| When you are bringing the system up after a scheduled maintenance
| session, or for scheduled standalone time, use one of the following
| answers:
|
| PM for preventative maintenance
|
| CM for corrective maintenance
|
| SA for scheduled stand-alone time
|
| If none of the responses you see here apply, use the following:
|
| OTHER followed by the reason for the reload
|
| You can include comments to further describe your reasons by preceding
| the comment with a semicolon or a space. Your answer to this question
| is stored in the system error log file ERROR.SYS.
|
| The ONCE program now prompts for a startup option.
|
| Startup option:
|
| ONCE provides several startup options. If you press RETURN without
| entering a startup option, you will be provided with a simple list of
| available options:
|
| Startup option:<RET>
A-5
ONCE STARTUP OPTIONS
| Options are: CHANGE DEFINE DESTRO DISSOL GO NOINIT
| REFRES SHOW
|
| If you type HELP and press RETURN you will be provided with a list of
| the startup options and switches and a brief explanation of each:
|
| Startup option:HELP<RET>
|
| Options are:
| CHANGE - Change system parameters
| DEFINE - Define structures and system lists
| DESTRO - Dissolve all structures in the system
| DISSOL - Dissolve a single structure
| GO - Start timesharing
| NOINIT - Start timesharing without system initialization
| REFRES - Refresh selected structures
| SHOW - Display system parameters
|
| Switches are:
| NOASL - Don't use reserved active swapping list
| NOCOPY - Don't start CRSCPY
| NOPRIM - Start timesharing in secondary protocol
| NOSDL - Don't use preserved system dump list
| NOSSL - Don't use preserved system search list
| NOVALI - Don't require account validation
| STANDA - Start timesharing with SCHED10 set
|
| Startup option:
|
|
|
| A.2 RUNNING ONCE IN USER MODE
|
| Running ONCE in user mode, by running the monitor, allows the operator
| to change disk parameters without reloading the system. To make
| changes to the disk file system, you must be logged in as [1,2].
|
| .R monitor-name<RET>
|
| Use scratch file (NO,YES) [default]:
|
| The first question allows you to store the disk data base in a .BIN
| file. The use of the scratch file provides the novice user the
| opportunity to learn how to use ONCE without any chance of file system
| damage. If you are not logged in as [1,2], the default answer is YES
| and you must accept the default answer. If you are not logged in as
| [1,2] and you answer NO, you will be returned to monitor level:
|
| ? Must be [1,2]
| .
|
| If you are logged in as [1,2] the default is NO. Normally, you
A-6
ONCE STARTUP OPTIONS
| respond to this question by pressing RETURN. Then, you are prompted
| for units to read:
|
| Read units:
|
| Specify the name of each disk unit you wish to change and press
| RETURN. Your entry specifies the units from which ONCE will read the
| home blocks. You may specify a single unit, or ALL. When you have
| specified all the units to read, press RETURN.
|
| Write units:
|
| Specify the disk units on which to write new HOME blocks for changed
| information. The ONCE program ensures that you will not inadvertently
| disrupt system operation by not allowing you to change the disk
| parameters for disks that you did not specify as 'Write units'.
|
| Startup option:
|
| If you press RETURN without entering a startup option, you will be
| provided with a simple list of available options:
|
| Startup option:<RET>
|
| Options are: CHANGE DEFINE DISSOL EXIT REFRES SHOW
|
| If you type HELP and press RETURN you will be provided with a list of
| the startup options and switches and a brief explanation of each:
|
| Startup option:HELP<RET>
|
| Options are:
| CHANGE - Change system parameters
| DEFINE - Define structures and system lists
| DISSOL - Dissolve a single structure
| EXIT - Return to monitor level
| REFRES - Refresh selected structures
| SHOW - Display system parameters
|
| Switches are:
| NOERRO - Re-write HOM blocks despite errors
|
| Startup option:
|
|
|
| A.3 STARTUP OPTIONS AND SWITCHES
|
| ONCE allows you to change many disk and search list parameters, and
| performs disk refreshing and restructuring for maintenance purposes.
| Certain startup options, startup option keywords, and switches are
| valid only when ONCE runs at system startup (executive mode); others
A-7
ONCE STARTUP OPTIONS
| are valid only when running ONCE in user mode. In following sections,
| each startup option is described and an example of its use is
| provided.
|
| When you select a startup option, the keywords applicable for that
| option are listed after the option in parentheses. For example, if
| you select the DEFINE startup option, the keywords valid for this
| option are listed after the option, in parentheses:
|
| Startup option:DEFINE<RET>
| Define (ASL,SDL,SSL,STRUCT):
|
| If you are already familiar with the keywords available for a specific
| startup option, you may type the startup option and the keyword on the
| same line:
|
| Startup option:DEFINE ASL<RET>
|
| Enter names of units...
|
| If you press RETURN without selecting a keyword you will be asked to
| specify a startup option again:
|
| Startup option:DEFINE<RET>
| Define (ASL,SDL,SSL,STRUCT):<RET>
|
| Startup option:
|
| The startup options and keywords, and the modes in which they are
| available, are listed and explained below. E denotes executive mode,
| and U denotes user mode.
|
| Option Mode Effect
|
| CHANGE E,U Use this option to change disk parameters.
| If you choose the CHANGE startup option when ONCE
| runs at system startup, the LOGIN command requires
| a PPN and password before you can log in a job.
|
| Mode Valid Keywords
|
| E,U ASL
| E DATE
| E NETWORk
| E,U SDL
| E,U SETS
| E,U SSL
| E,U STRUCTure
| E,U UNIT
|
| Section A.3.1 describes the CHANGE startup option.
A-8
ONCE STARTUP OPTIONS
| DEFINE E,U Use this option to define structures and
| system lists. If you choose the DEFINE startup
| option when ONCE runs at system startup, the LOGIN
| command requires a PPN and password before you can
| log in a job.
|
| Mode Valid keywords
|
| E,U ASL
| E,U SDL
| E,U SSL
| E,U STRUCTure.
|
| Section A.3.2 describes the DEFINE startup option.
|
| DESTROY E Use this option with extreme caution. The
| DESTROY option is used to refresh all structures
| and rewrite all HOME blocks. If you choose the
| DESTROY startup option, ONCE prints the following
| message on the CTY:
|
| To automatically log-in under [1,2] type "LOGIN"
|
| Then you can log in a job under the PPN [1,2] by
| simply typing LOGIN to the monitor prompt.
|
| Section A.3.3 describes the DESTROY startup
| option.
|
| DISSOLVE E,U Use this option to dissolve a single
| specified file structure.
|
| Section A.3.4 describes the DISSOLVE startup
| option.
|
| EXIT U Use this option to return to monitor level.
|
| Section A.3.5 describes the EXIT startup option.
|
| REFRESH E,U Use this option to refresh a single specified
| file structure. If you choose the REFRESH startup
| option, ONCE prints the following message on the
| CTY:
|
| To automatically log-in under [1,2] type "LOGIN"
|
| Then you can log in a job under the PPN [1,2] by
| simply typing LOGIN to the monitor prompt.
|
| Section A.3.6 describes the REFRESH startup
| option.
A-9
ONCE STARTUP OPTIONS
| SHOW E,U Use this option to display system parameters.
|
| Mode Valid keywords
|
| E,U ASL
| E DATE
| E,U NETWORk
| E,U SDL
| E,U SETS
| E,U SSL
| E,U STRUCTure
| E,U UNIT
|
| Section A.3.7 describes the SHOW option.
|
| GO E Use this option to start the monitor without
| any checking or dialog.
|
| Section A.3.8 describes the GO option.
|
| NOINITIA E Use this option to start the monitor without
| initializing terminal lines. Because terminal
| lines are not initialized, this option should be
| used only when running the system stand-alone.
|
| Section A.3.9 describes the NOINITIA option.
|
| The startup option can be followed by a number of switches to control
| the type of output to the CTY. The switches available in each mode
| are listed and explained below:
|
| Switch Mode Effect
|
| /NOASL E Use this switch to force the monitor to rebuild
| the Active Swapping List from disk after a reload.
| If you do not specify this switch, the monitor
| first tries to use the preserved copy in the
| Active Swapping List. If the monitor cannot use
| the preserved data, then it rebuilds the data from
| disk.
|
| /NOCOPY E Use this switch to prevent CRSCPY from starting
| automatically when the monitor is started up.
| When a continuable stopcode occurs, CRSCPY copies
| the current crash file (SYS:CRASH.EXE) into a
| backup file in the XPN area.
|
| /NOERRO U Use this switch to force the monitor to rewrite
| the HOME blocks.
|
| /NOPRIMARY E Use this switch to start timesharing in secondary
| protocol. This switch is valid for KL10s only.
A-10
ONCE STARTUP OPTIONS
| /NOSDL E Use this switch to force the monitor to rebuild
| the System Dump List from disk after a reload. If
| you do not specify this switch, the monitor first
| tries to use the preserved copy in the System Dump
| List. If the monitor cannot use the preserved
| data, then it rebuilds the data from disk.
|
| /NOSSL E Use this switch to force the monitor to rebuild
| the System Search List from disk after a reload.
| If you do not specify this switch, the monitor
| first tries to use the preserved copy in the
| System Search List. If the monitor cannot use the
| preserved data, then it rebuilds the data from
| disk.
|
| /NOVALIDATE E Use this switch to turn off account validation.
|
| /STANDALONE E Use this switch to bring up the monitor as a
| stand-alone system.
|
|
|
| A.3.1 CHANGing System Parameters
|
| This section describes the CHANGE startup option. You must DEFINE the
| system lists and structures before you can CHANGE them. Section A.3.2
| describes the DEFINE startup option and provides a complete
| description of the parameters that must be defined before they are
| changed.
|
| In the following examples, the CHANGE startup option is selected
| first, followed by the item to change. The text that you type is
| underlined.
|
| To CHANGE the Active Swapping List (E,U):
|
| Enter the names of the units in the Active Swapping List. Enter an
| extra RETURN when the list is complete.
|
| Startup option:CHANGE ASL<RET>
|
| Enter names of units to appear in the active swapping list, one
| per line. Terminate list with a blank line.
| RXxn<RET>
| <RET>
|
| Startup option:
|
| To CHANGE the date (E):
|
| Startup option:CHANGE DATE<RET>
| Date:dd-mmm-yy<RET>
A-11
ONCE STARTUP OPTIONS
| Time:hh:mm:ss<RET>
|
| Startup option:
|
| To CHANGE network information (E):
|
| The network information is initially provided by your answers to
| several questions in the MONGEN dialog. The appropriate MONGEN
| question numbers are shown next to the corresponding questions in the
| example below:
|
| Startup option:CHANGE NETWORK<RET> (MONGEN Question )
|
| System wide node name [name]:name<RET> ( 90)
| ANF-10 node name [name]:name<RET> ( 92)
| ANF-10 node number (1 - 77) [nn]:nn<RET> ( 94)
| Decnet node name [node name]:name<RET> (107)
| DECnet area number (1 - 36) [nn]:nn<RET> (108)
| DECnet node number (1 - 1023) [nnnn]:nnnn<RET> (109)
| LAT service name [name]:name<RET> (114)
|
| Startup option:
|
| To CHANGE the System Dump List (E,U):
|
| Enter the names of the structures in the System Dump List. Enter an
| extra RETURN when the list is complete.
|
| Startup option:CHANGE SDL<RET>
|
| Enter names of structures to appear in the system dump list, one
| per line. Terminate list with a blank line.
| DSKx<RET>
| <RET>
|
| Startup option:
|
| To CHANGE disk-set numbers (E,U):
|
| Disk-set numbers are assigned within the DEFINE STRUCTure startup
| option, and may also be changed within the CHANGE STRUCTure startup
| option.
|
| Startup option:CHANGE SETS<RET>
|
| Enter the desired disk-set numbers (1 - 36) or "ALL" for all
| disk-sets, one per line. Terminate with a blank line.
| ALL<RET>
| <RET>
|
| Startup Option:
A-12
ONCE STARTUP OPTIONS
| To CHANGE the System Search List (E,U):
|
| Startup option:CHANGE SSL<RET>
|
| Enter names of structures to appear in the system search list,
| one per line. Terminate list with a blank line.
| DSKx<RET>
| <RET>
|
| Startup option:
|
| To CHANGE STRUCTure parameters (E,U):
|
| The CHANGE STRUCTure startup option is used to change the parameters
| of an existing file structure. You can specify one structure to
| change only its parameters, or ALL to change all the structures.
|
| In the questions that follow, any question preceded with an asterisk
| indicates changing that parameter will require the structure to be
| refreshed. A range of acceptable values, if applicable, is enclosed
| in parentheses. A default value, if any, is enclosed in square
| brackets. The ranges of acceptable values are calculated according to
| the structure type previously defined. The default values provided
| are those values which were previously defined. All ranges and
| default values used in the following example are used for illustration
| only. To accept a default value, just press RETURN.
|
| Startup option:CHANGE STRUCT<RET>
| Structure to change:DSKx<RET>
|
| Change unit parameters for this structure (NO,YES) [NO]:
| Do you want to see the BAT blocks (NO,YES) [NO]:
| Parameters for structure DSKx
| Structure name [DSKx]:
| *Blocks per cluster (3 - 131071) [5]:
| Private structure (NO,YES) [NO]:
| Owner PPN (NONE,[p,pn]) [NONE]:
| Disk-set (1 - 36,ALL,NONE) [ALL]:
| *Number of K for CRASH.EXE (0 - 4096) [2048]:
| Blocks allowed for overdraw per user (0 - 307800) [500]:
| Consecutive blocks tried for on output (0 - 262143) [30]:
| *Bits per cluster count (0 - 18) [12]:
|
| Structure to change:
|
| When you have finished changing one or all structures, you may enter
| another stucture to CHANGE. If you wish to be prompted for another
| startup option, press RETURN.
|
| To CHANGE Unit paramters (E,U):
|
| The CHANGE UNIT startup option is used to change the unit parameters
A-13
ONCE STARTUP OPTIONS
| of an existing file structure.
|
| Enter a unit to change and press RETURN. A range of acceptable
| values, if applicable, is enclosed in parentheses. A default value,
| if any, is enclosed in square brackets. To accept a default value,
| just press RETURN.
|
| Startup option:CHANGE UNIT<RET>
|
| Any question preceeded with an asterisk indicates changing that
| parameter will require the structure to be refreshed.
|
| Unit to change:RXxn<RET>
| Parameters for unit RXxn
| Unit ID [XXXXn]:
| *SAT blocks per unit (5 - 81) [5]:
| SAT blocks in core (1 - 5) [5]:
| *K for swapping (0 - 21625) [20000]:
| *First logical block for swapping (1400 - 705504) [352752]:
| Swapping class (0 - 1) [0]:
| Make this the only unit in the ASL (NO,YES) [NO]:
|
| Unit to change:
|
| When you have finished changing a unit, you may enter another unit to
| CHANGE. If you wish to be prompted for another startup option, press
| RETURN.
|
|
|
| A.3.2 DEFINing Structures and System Lists
|
| This section describes the DEFINE startup option. In the following
| examples, the DEFINE startup option is selected first, followed by the
| item to define. The text that you type is underlined.
|
| To DEFINE the Active Swapping List (E,U):
|
| The Active Swapping List is the list of units that the system will use
| to store data that is swapped to disk. The units in the Active
| Swapping List should have swapping space allocated on them. Make this
| allocation with the DEFINE STRUCTure startup option. To change the
| amount of swapping space allocated to the units in the list, select
| the CHANGE UNIT startup option. For the most efficient swapping, the
| units should be listed in the order of fastest to slowest
| accessibility.
|
| The swapping list is divided into classes, where Class 0 is used
| first, then Class 1. All swapping units should be placed in Class 0,
| except units that have slow I/O. Be sure to include all or several
| controllers in the swap list. Use the DEFINE STRUCTure startup option
| to assign a unit to a swapping class. To change the swapping class of
A-14
ONCE STARTUP OPTIONS
| a unit, use the CHANGE UNIT startup option.
|
| Enter the names of the units in the Active Swapping List. Enter an
| extra RETURN when the list is complete.
|
| Startup option:DEFINE ASL<RET>
|
| Enter names of units to appear in the active swapping list, one
| per line. Terminate list with a blank line.
| RXxn<RET>
| <RET>
|
| Startup option:
|
| To DEFINE the System Dump List (E,U):
|
| Enter the names of the structures in the System Dump List. Enter an
| extra RETURN when the list is complete.
|
| The System Dump List is the list of structures where the system will
| attempt to write the crash file whenever the system is halted. Space
| should be allocated on every structure in the System Dump List for the
| crash file. Make this allocation with the DEFINE STRUCTure startup
| option. To change the amount of space allocated for the crash file,
| select the CHANGE STRUCTure startup option.
|
| Startup option:DEFINE SDL<RET>
|
| Enter names of structures to appear in the system dump list, one
| per line. Terminate list with a blank line.
| DSKx<RET>
| <RET>
|
| Startup option:
|
| To DEFINE the System Search List (E,U):
|
| Enter the name of each structure in the SSL, in the order in which the
| search list should be accessed. Type an extra RETURN when the list is
| complete.
|
| The System Search List is a list of file structures, listed in the
| order that they will be searched whenever the ersatz device SYS: is
| specified or implied. The system CUSPs must be accessible through the
| System Search List.
|
| Startup option:DEFINE SSL<RET>
|
| Enter names of structures to appear in the system search list,
| one per line. Terminate list with a blank line.
| DSKP<RET>
| <RET>
A-15
ONCE STARTUP OPTIONS
| Startup option:
|
| To DEFINE a structure (E,U):
|
| Enter the name of a structure to define and press RETURN. Then, enter
| the logical names of units within that structure.
|
| Startup option:DEFINE STRUCT<RET>
|
| In the following dialogue, all numbers are decimal unless
| otherwise indicated. Values within square brackets are defaults
| which will be substituted if RETURN is pressed.
|
| Any question preceeded with an asterisk indicates changing that
| parameter will require the structure to be refreshed.
|
|
| *** Note ***
| HOME blocks will be initialized immediately after all units in
| the structure have been entered.
| Structure to define:DSKx<RET>
| Disk drive for logical unit 0:RXxn<RET>
| Disk drive for logical unit 1:<RET>
|
| ONCE checks the BAT blocks for each unit. BAT blocks point to
| hardware errors on the disk pack. Two BAT blocks are written on the
| pack when it is formatted, and the system updates them when it finds
| blocks on the disk that cannot be read from or written to. The BAT
| blocks are first checked for consistency errors. They must be
| formatted properly, so that the system can read them. If
| inconsistencies are found in both blocks on the unit, they will be
| reported. If the first BAT block is all right, the second BAT block
| is not checked, because the second can be updated with a copy of the
| first.
|
| You have the option of listing the bad regions pointed to by the BAT
| blocks. This information would be useful if only the first BAT block
| had a consistency error. If both BAT blocks are inconsistent, then
| this information would be useless. Therefore, if the first BAT block
| is inconsistent, but the second is consistent, and if the list of bad
| regions on the disk is lengthy, you should consider having the disk
| reformatted.
|
| Do you want to see the bad regions (NO,YES) [NO]:
|
| ONCE initializes the HOME blocks. Each disk pack contains two HOME
| blocks that contain pertinent information about the disk unit, such as
| the file structure it is in, swapping space that has been allocated on
| the unit, and the other unit-specific information that ONCE allows you
| to change.
|
| [HOME blocks initialized on RXxn]
A-16
ONCE STARTUP OPTIONS
| If you ask to see the bad regions, the bad regions will be listed by
| unit according to the format shown below:
|
| % Unit RXxn bad blocks:
| First block No. CPU Cont Unit Error
| ------------ --- ----- ---- ---- -----
|
|
| You have the option of initializing the BAT blocks. Initialization
| rewrites the BAT blocks completely. All previous data about errors on
| the disk will be deleted and the BAT blocks will be filled in by the
| system as it attempts to read/write the disk. The BAT blocks should
| only be initialized if both have been reported to have
| inconsistencies, or if a disk channel has been erroneously reporting
| hard errors on the disk.
|
| When the BAT blocks are rewritten, the old BAT blocks will be lost,
| and the information about bad areas on the disk will have to be
| rewritten into the new BAT blocks.
|
| Initialize BAT blocks (NO,YES) [NO]:
|
| If you answer YES, you will be informed that the action has been
| completed:
|
| [BAT blocks initialized on RXxn]
|
| Enter a value to specify the cluster size for the structure, or accept
| the default value by pressing RETURN.
|
| The cluster size is critical to efficient use of the disk. A
| one-block data file requires 3 blocks on disk, including RIB, data
| block, and spare RIB. If a cluster size of 10 were assigned, 7 blocks
| would be wasted for this one-block file. The cluster size should not
| be set to less than 3, but the actual size depends upon the type of
| applications using the disk. The monitor requires a retrieval pointer
| for each cluster in the file it is accessing. Therefore, the cluster
| size should not be too small for the size of the average file. A
| large file with a small cluster size wastes monitor overhead.
|
| The default value listed by ONCE is determined by the type of disk
| unit. The best value depends on the type of application using the
| disk, but the default is often appropriate.
|
| Parameters for structure DSKx
| *Blocks per cluster (3 - 262143) [10]:
|
| Your next entry sets the privacy status of a structure. A private
| structure requires that users who wish to access it must use the MOUNT
| command to access that structure. A system structure is not private,
| and may be accessed by any user without explicitly mounting the
| structure for the job.
A-17
ONCE STARTUP OPTIONS
| Private structure (NO,YES) [NO]:
|
| A file structure need not have an owner PPN. However, if the owner
| PPN is set, the user with that PPN has full file access of the entire
| structure. Enter a PPN if you want to set or change the owner PPN of
| the structure. If you press return, the default value will be
| accepted and the current PPN will be cleared without replacing it.
|
| The PPN may be wildcarded. Note that a PPN of [*,*] allows all users
| full access to all files on the structure.
|
| Owner PPN [NONE]:
| Disk-set (1-36,ALL,NONE) [ALL]:
|
| If this structure will be included in the System Dump List, you must
| allocate space for the crash file. Enter a value to specify the
| amount of space to allow on the structure for a crash file. The crash
| file, CRASH.EXE, is the copy of memory that is written to disk
| automatically when the system crashes. This value should be at least
| the size of the maximum amount of memory used by your system. For
| RP20 and CI disks, however, this value should be 0.
|
| *Number of K for CRASH.EXE (0 - 4096) [4096]:
|
| Enter the number of blocks each user is allowed on the structure for
| overdraw purposes. The maximum allowable value is a variable that is
| dependant upon the type of disk unit.
|
| When the system attempts to write a file for a user whose disk-space
| quota on the structure is not large enough to accommodate the whole
| file, a number of blocks over the quota can be allowed: this number
| is the "overdraw". When the user reaches the disk quota, the monitor
| does not allow new files to be created. It attempts to write the
| files that are open at the time quota was exceeded.
|
| Blocks allowed for overdraw per user (0 - 307800) [500]:
|
| Determine the number of consecutive blocks that the system will check
| for when searching for free space on disk to write files.
|
| When the system attempts to write to disk, it tries to allocate a
| consecutive number of blocks that are free. You are allowed to
| specify this number. If the consecutive blocks are found, the file is
| written and the SAT block pointers are changed to reflect the change
| in free space. If the system cannot find a consecutive number of
| blocks of the size specified here, the file will be stored in the
| first free space, fragmented if necessary. Therefore, the value you
| give can affect the efficiency of file storage on the disk. It is
| recommended that you use a multiple of the cluster size. A number too
| small can result in fragmented disk files and excessive time spent in
| accessing files.
A-18
ONCE STARTUP OPTIONS
| Consecutive blocks tried for on output 90 - 262143) [30]:
|
| Enter the number of bits per cluster count, or press RETURN to accept
| the default.
|
| *Bits per cluster count (0 - 18) [12]:
|
| The system search list is a list of file structures, in the order that
| they will be searched whenever the ersatz device SYS: is specified or
| implied. The system CUSPs must be accessible through the system
| search list. Enter YES if you want this structure to be included in
| the SSL.
|
| Make this the only structure in the SSL (NO,YES) [YES]:
|
| The system dump list is the list of structures where the system will
| attempt to write the crash file whenever the system is halted. Space
| should be allocated on every structure in the system dump list for the
| crash file. Allocation for the crash file was made above.
|
| Make this the only structure in the SDL (NO,YES) [YES]:
|
| Specify a unit identification. This identification resides in the
| HOME blocks on the disk unit, for unique identification of the disk
| pack. Therefore, it should be a unique number that will never
| conflict with another code. Every pack should have a unit-id of six
| alphanumeric characters that may be a portion of the pack's serial
| number.
|
| Parameters for unit RXxn
| Unit ID [XXXXn]:
|
| Specify the number of SAT blocks on the unit. The default is
| calculated according to the type of disk unit. As a general rule, you
| should select the minimum number of SAT blocks, to reduce monitor
| overhead.
|
| *SAT blocks per unit (8 - 81) [8]:
|
| Enter the number of SAT blocks to be stored in memory for the unit.
| While the unit is mounted, the SAT blocks on the unit are stored in
| memory for fast access by the system. All of the SAT blocks should be
| stored in memory.
|
| SAT blocks in core (1 - 8) [8]:
|
| Specify the amount of swapping space allowed on the unit. The total
| allocated on all units should be at least 2 times memory size.
|
| If enough space is not allowed, swapping space may run out during
| timesharing. The only way to expand the swapping space is to refresh
| the disk, which destroys all user files. If the unit is or will be in
A-19
ONCE STARTUP OPTIONS
| the system's Active Swapping List, you must allocate some space for
| swapping on the unit.
|
| *K for swapping (0 - 4807) [4807]:
|
| ONCE calculates the address of the first logical block where the
| system will attempt to write swapped data. This address appears as
| the default value. The default is usually appropriate, because the
| swapping space is put in the middle of the disk, to minimize the
| average seek time when accessing the swapping space. However, you may
| change this value.
|
| *First logical block for swapping (33090 - 269332) [134672]:
|
| The swapping list is divided into classes, where Class 0 is used
| first, then Class 1.
|
| Swapping class (0 - 1) [0]:
|
| The active swapping list is the list of units that the system will use
| to store data that is swapped to disk. If you wish to include this
| unit in the active swapping list, enter YES below:
|
| Make this the only unit in the ASL (NO,YES) [YES]:
|
| If you have changed any parameters marked with an asterisk, ONCE
| reminds you that the structure must be refreshed:
|
| % Structure DSKx must be refreshed before it can be used.
|
| Structure to define:
|
| When you have finished defining a structure, you may enter another
| structure to DEFINE. If you wish to be prompted for another startup
| option, press RETURN.
|
|
|
| A.3.3 DESTROYing All Structures
|
| This section describes the DESTROY startup option. The DESTROY
| startup option is used to initialize the disk system using default
| parameters. This option is available only in executive mode.
|
| The DESTROY option is very powerful. When the DESTROY option is
| selected, the disks are refreshed, deleting all files on the disk, and
| the HOME and SAT blocks are initialized, deleting all information
| about the disk's place in system lists and file structures. For this
| reason, the ONCE program ensures that you want to select the DESTROY
| option by printing a warning message before accepting the DESTROY
| startup option. This warning is intended to prevent accidental
| destruction of disk data.
A-20
ONCE STARTUP OPTIONS
| The DESTROY option sets the disk parameters to defaults predetermined
| by the monitor. Those defaults are based on the types of disk units
| reported by the disk controllers, and values of the parameters are
| calculated by ONCE. The DESTROY startup option routine is discussed
| in Chapter 4.
|
| The parameters set by ONCE and the values defaulted by the DESTROY
| option are listed in Table A-1, below. The unit types in Table A-1
| are designated by controller names that are described in Table A-2.
|
| Table A-1 shows the types of disk controllers that are recognized by
| ONCE and the order in which they are characterized. The construction
| of file structures and system lists is dependent upon the type of disk
| controllers at each installation. For example, if the controller in
| the first column does not exist, ONCE configures starting at the
| second column, and so forth.
|
|
| Table A-1: Disk Parameters Defaulted By DESTROY Option
|
|
| STRUCTURE PARAMETERS
|
|
| Parameter Type Default Parameter Values
|
| Unit type | FH | FS | RP | DP | RN | RA
| ---------------------------------------------------------------
| Max. no. | | | | | |
| of units | 0 | 32 | 16 | 32 | 16 | 8
| ---------------------------------------------------------------
| Starting | | | | | |
| structure | | | | | |
| name | ---- | DSKA | DSKB | DSKE | DSKF | DSKI
| ---------------------------------------------------------------
| No. of structures | | | | | |
| to build | 0 | 1 | 3 | 1 | 3 | 3
| ---------------------------------------------------------------
| No. of | | | | | |
| consecutive | | | | | |
| blocks | | | | | |
| tried for | | | | | |
| on output | 10 | 10 | 30 | 30 | 30 | 30
| ---------------------------------------------------------------
| No. of | | | | | |
| blocks | | | | | |
| guaranteed | | | | | |
| to each | | | | | |
| user | 0 | 0 | 0 | 0 | 0 | 0
| ---------------------------------------------------------------
| No. of | | | | | |
| blocks | | | | | |
A-21
ONCE STARTUP OPTIONS
| allowed for | | | | | |
| over-draw | | | | | |
| for each | | | | | |
| user | 200 | 100 | 500 | 500 | 500 | 500
| ---------------------------------------------------------------
| Private | | | | | |
| structure? | NO | NO | NO | NO | NO | NO
| ---------------------------------------------------------------
| #K for | | | | | |
| CRASH.EXE | 0 | 0 | -1 | -1 | 0 | 0
| ---------------------------------------------------------------
| No. of | | | | | |
| blocks per | | | | | |
| cluster | 1 | 1 | 10 | 5 | 50 | 10
| ---------------------------------------------------------------
| No. of | | | | | |
| bits per | | | | | |
| cluster | | | | | |
| count | 12 | 12 | 12 | 12 | 12 | 12
|
|
| UNIT PARAMETERS
|
| Unit type | FH | FS | RP | DP | RN | RA
| ---------------------------------------------------------------
| #K for | | | | | |
| swapping | 0 | 200 | 1000 | 0 | 0 | 0
| Unit-id Defaults to ONCnnn, where nnn is random
|
|
|
| Table A-2: Device Channels and Controllers
|
|
| Channels Controller Kont-name Unit Types
|
| RH20 --> DX20 | RP20 (master) | RNx | RP20 (slaves)
| |---------------------------------------------
| | | |
| | TM02 | MTX | TU45, TU70,
| | | | TU71, TU72, TU77
| -------------------------------------------------------------
| | | |
| RH10, RH20 | Massbus | FSx | RS04
| | |-----------------------------
| | | |
| | | RPx | RP04, RP05
| | | | RP06, RP07
| -------------------------------------------------------------
| | | |
| RH11 (KS only) | Massbus | RPx | RM03, RP06
A-22
ONCE STARTUP OPTIONS
| |---------------------------------------------
| | | |
| | TM0n | MTx | TU45, TU78
| -------------------------------------------------------------
| | | |
| DF10, DF10C | RP10 | DPx | RP02, RP03
| |---------------------------------------------
| | | |
| | RC10 | FHx | drums
| -------------------------------------------------------------
| | | |
| DX10 | TX01 | MTx | TU70, TU71, TU72
| -------------------------------------------------------------
| | | |
| RH20 | TM0n | MTXn | TU45, TU77
| |---------------------------------------------
| | | |
| | TM78 | MTx | TU78
| -------------------------------------------------------------
| | | |
| RH20 --> CI20 | HSC50 | RAX | RA60, RA80, RA81
| -------------------------------------------------------------
|
|
|
|
| A.3.4 DISSOLving a Single Structure
|
| This section describes the DISSOLVE startup option. The DISSOLVE
| option is available in executive and user modes.
|
| In the following example, the DISSOLVE startup option is selected
| first, followed by the structure to dissolve. The text that you type
| is underlined.
|
| A file structure is dissolved by deleting all reference to that file
| structure on the unit. The units in that structure will not be listed
| as part of any file structure until you define a file structure as
| containing those units.
|
| Startup option:DISSOLVE<RET>
| Structure to dissolve:DSKx<RET>
| All data will be destroyed. Are you sure (NO,YES) [NO]:YES<RET>
| Structure to dissolve:
|
| Enter another structure to dissolve or press RETURN to select another
| startup option.
A-23
ONCE STARTUP OPTIONS
| A.3.5 EXITing to Monitor Level
|
| This section illustrates the EXIT startup option. EXIT returns you to
| monitor level.
|
| You may EXIT only when running ONCE in user mode.
|
| Startup option:EXIT<RET>
| .
|
|
|
| A.3.6 GO Option
|
| This section describes the GO startup option. This option is
| available only in executive mode.
|
| Startup option:GO<RET>
|
| The GO option allows you to start up the monitor without changing any
| disk parameters. INITIA is allowed to run, but offline disk units are
| not checked for or reported. Because the GO startup option starts
| INITIA, one of the following files must exist:
|
| SYS:TTY.INI to set up terminal line characteristics.
|
| SYS:OPR.ATO to be read by OPSER automatically.
|
| INITIA first attempts to read TTY.INI. If the file does not exist in
| SYS, it runs OPSER, which attempts to read OPR.ATO. If OPR.ATO does
| not exist, OPSER prints an error message. You can exit from OPSER
| typing CTRL/C and continue.
|
| If TTY.INI exists, then INITIA follows the instructions in TTY.INI.
| The following example assumes that TTY.INI exists, and contains the
| line:
|
| CTY GALOPR
|
| INITIA reads the line, and starts OPR on the CTY, where the commands
| are entered. INITIA reads SYS:SYSJOB.INI, starting the system
| programs, and then waits for input. Type the EXIT command to exit
| from OPR:
|
| MONITOR FEB-16-88
| Why reload:OTHER ;TEST<RET>
| Date: FEB-20-88<RET>
| Time: 11:35<RET>
|
| Startup option: GO<RET>
|
| MONITOR 11:35:32 CTY system 1026
A-24
ONCE STARTUP OPTIONS
| Connected to Node CENTR(0) Line # 45
| .LOGIN 1,2
| .R OPR
| OPR>EXIT<RET>
|
| .
|
|
|
| A.3.7 NOINITIA Option
|
| This section describes the NOINITIA startup option. This option is
| available only in executive mode.
|
| You can start up the monitor without running INITIA to initialize the
| terminal lines by using the NOINITIA option. Note that this also
| prevents INITIA from reading the SYSJOB.INI file as well. The
| SYSJOB.INI file is described in Chapter 10.
|
| The NOINITIA startup option is identical to the GO option, except that
| it prevents the INITIA program from running. Therefore, it is a quick
| way to start the monitor without changing any disk parameters.
| However, the NOINITIA option should not be used to start timesharing
| because INITIA is required to initialize terminal lines. INITIA can
| be run later from the CTY using the INITIA SYS monitor command. See
| Chapter 10 for more information about INITIA.
|
| TOPS-10 7.04 system FEB-16-88
| Why reload:OTHER;TEST<RET>
| Date:FEB-16-88<RET>
| Time:11:44<RET>
|
| Startup option: NOINITIA<RET>
|
| After entering NOINITIA and pressing RETURN, ONCE checks for offline
| disk units and lists them. However, you are not allowed to change
| them.
|
| %UNIT-NAME is offline
|
| TOPS-10 7.04 system FEB-16-88
|
| .
|
|
|
| A.3.8 REFRESHing Selected Structures
|
| This section describes the REFRESH startup option. In the following
| example, the REFRESH startup option is selected first, followed by the
| structure to refresh. The text that you type is underlined.
A-25
ONCE STARTUP OPTIONS
| Structures may be refreshed in executive and user modes.
|
| WARNING
|
| When you refresh a file structure you destroy all data
| on the structure. Therefore, ONCE attempts to make
| absolutely certain that you intend to refresh the
| structure before doing so.
|
| The REFRESH option forces a refresh procedure to initialize
| characteristics that require refreshing. When you select the REFRESH
| startup option, the HOME blocks of all units are checked to determine
| which file structures need to be refreshed. Then, you are asked to
| enter the name of a file structure to be refreshed. After refreshing,
| you will again be asked for the name of a file structure to be
| refreshed. If you press RETURN without providing the name of a file
| structure, you will be prompted to enter another startup option.
|
| Startup option:REFRESH<RET>
| Structure to refresh:DSKx<RET>
| Structure does not need refreshing.
| Are you sure (NO,YES) [NO]:NO<RET>
| Structure to refresh:<RET>
|
| Startup option:
|
| Enter another structure to refresh or press RETURN.
|
|
|
| A.3.9 SHOWing System Parameters
|
| This section describes the SHOW startup option. In the following
| examples, the SHOW startup option is selected first, followed by the
| item to show. The text that you type is underlined.
|
| To SHOW the Active Swapping List (E,U):
|
| Startup option:SHOW ASL<RET>
|
| Swapping units:
| Unit UnitID Class Space (K)
| ---- ------ ----- ---------
| RPA0 DSKD0 0 4807
| RPA6 KLAD0 0 4807
| RPB4 DSKP0 0 20000
| (Memory size = 2048K)
|
| Startup option:
|
| The Active Swapping List is the list of units that the system will use
| to store data that is swapped to disk. The swapping list is divided
A-26
ONCE STARTUP OPTIONS
| into classes, where Class 0 is used first, then Class 1, Class 2, and
| so on.
|
| To SHOW the date (E):
|
| Startup option:SHOW DATE<RET>
| Thursday 4-Feb-88 :27:58
|
| Startup option:
|
| To SHOW networks (E,U):
|
| Startup option:SHOW NETWOR<RET>
|
| Networks:
| System-wide node name: DEC10
| ANF-10 node DEC10(10)
| DECnet node DEC10(7.169)
| LAT service: DEC10
|
| Startup option:
|
| To SHOW the System Dump List (E,U):
|
| Startup option:SHOW SDL<RET>
| System dump list: DSKD, DSKP
|
| Startup option:
|
| To SHOW the System Search List (E,U):
|
| Startup option:SHOW SSL<RET>
|
| System search list:
| Structure Position
| --------- --------
| DSKD 0
|
| Startup option:
|
| To SHOW disk sets (E,U):
|
| Startup option:SHOW SETS<RET>
| Disk sets: 2, 6, 11, 26
|
| Startup option:
|
| To SHOW a structure (E,U):
|
| Enter a structure to SHOW. If you enter ALL, ONCE lists the
| parameters for all the structures that are online and repeats the
| question so that you can SHOW parameters for another structure. When
A-27
ONCE STARTUP OPTIONS
| you enter a structure to show, you are provided with a list of
| parameters for that structure or structures. This listing can be
| useful if you wish to change the parameters of a file structure. If
| you do not wish to SHOW a structure, press RETURN and you will be
| prompted for another startup option.
|
| Startup option:SHOW STRUCT<RET>
| Structure:DSKD<RET>
|
| Parameters for structure DSKD
| *Blocks per cluster: 10
| Private structure: NO
| Owner PPN: [10,56]
| Disk-set: ALL
| *Number of K for CRASH.EXE: 4096
| Blocks allowed for overdraw per user: 500
| Consecutive blocks tried for on output: 0
| *Bits per cluster count: 15
| Position in the SSL: 0
| Unit in the SDL: YES
| Structure:<RET>
|
| Startup option:
|
| To SHOW a unit (E,U):
|
| Enter a unit to SHOW. When you enter a unit to show, you are provided
| with a list of the parameters of any disk unit or all disk units.
| This listing is useful when changing the parameters of a disk unit.
| If you type ALL, parameters of all of the disk units will be listed.
| If you specify a unit, ONCE lists the unit parameters and then repeats
| the question so you can SHOW another unit. If you do not wish to SHOW
| a unit, press RETURN and you will be prompted for another startup
| option.
|
| Startup option:SHOW UNIT<RET>
| Unit: RPA0<RET>
|
| Parameters for unit RPA0
| Unit ID: DSKD0
| *K for swapping: 4807
| *First logical block for swapping: 134672
| Swapping class: 0
| *SAT blocks for unit: 8
| *SAT blocks in core: 8
| Unit in the ASL: YES
| Unit:<RET>
|
| Startup option:
A-28
APPENDIX B
MODIFYING THE MONITOR
B.1 REBUILDING THE MONITOR LIBRARY FILE
The monitor library file, TOPS10.REL is distributed with the TOPS-10
monitor in a standard form. It is linked with the monitor common
modules (COMMON, COMDEV, and COMMOD), as explained in Section 9.3 of
this manual. For new installations, the procedure in Chapter 9 is
sufficient for building the monitor.
The monitor library file contains assembled code for all the bundled
monitor modules. If an installation has made modifications to the
monitor, or if you changed any settings of feature test options in the
FGEN portion of MONGEN, you must reassemble the monitor library file.
NOTE
The symbol values and configuration options possible
with this procedure are so numerous that it is
impossible to predict results in every combination.
Therefore, you should be aware that the resulting
operating system may not be supported by Digital
Equipment Corporation.
Several modules must appear in a specific order. They are:
|
| F.MAC
| S.MAC
| NETPRM.MAC
| DTEPRM.MAC
| SYSCNF.MAC+COMMON.MAC
| SYSCNF.MAC+COMDEV.MAC
| SYSCNF.MAC+COMMOD.MAC
| EJBDAT.REL
| FILFND.MAC
| FILIO.MAC
| FILUUO.MAC
| RHXKON.MAC
| RPXKON.MAC
B-1
These files must precede all other modules, and must be listed in this
order. The following files must be placed, in this order, at the end
of the list:
|
| PATCH.MAC
| EDDT.REL
| USRONC.MAC
|
For more information about building the monitor library file, read the
batch control file COMPIL.CTL.
B.2 CHANGING COMMON SYMBOLS
The COMMON, COMDEV, and COMMOD modules contain symbols that may be
redefined with MONGEN. This section lists those symbols. When
entering the symbols to MONGEN, type them in the form:
symbol,value
The symbols and their default values are listed here as they might be
typed in the MONGEN dialog.
| The following symbols may be changed in Questions 109 through 111 of
| the SYSGEN portion of MONGEN. Decimal values are changed with
| Question 109, Octal values are changed with Question 110, and SIXBIT
| values are changed with Question 111.
|
|
|
| B.2.1 Symbols Defined in COMMON
|
| The following symbols are defined in COMMON. The symbols are listed
| in the form symbol,default-value.
|
|
|
| B.2.1.1 Decimal Symbols,Values -
|
| The following symbols are defined with values in decimal radix. These
| default values can be changed with Question 109 of the SYSGEN portion
| of MONGEN.
ERPTMX,80 Specifies the length of block for errors.
EVBN,128 Blocks available for mapping buffers and
IOWDs.
HIGHXC,80 Highest extended channel.
IFC0,5 Swapper fairness count maximum.
LIM1HS,30 Limit of stopcodes in any 1 hour period.
B-2
LOGSIZ,34 17K (current size of LOGIN CUSP).
M.BMAX,13 Maximum number of simultaneous batch job
streams.
M.CBAT,10 Number of bad addresses in per CPU bad.
M.CLSN,16 Number of CPU classes for scheduler.
M.CMCT,32 Number of calls to CHKTAL routine.
M.DNSZ,30000 DECnet free core size.
M.DTLS,90 CPU time limit (seconds) to set.
M.EBPS,12500000 Default number of EBOX ticks per second.
M.ENQD,100 Default ENQ quota.
M.ENQM,500 Maximum number of active queues.
M.ETSZ,3000 Ethernet free core size.
M.KNSZ,3000 KLNI free core size.
M.LASZ,15000 LAT free core size.
M.LCST,0 Length of the Line Data Block reserved for
customer definition.
M.MBPS,9750000 Default MBOX references per second.
M.OMSM,60 Number of minutes between calls to BIGBEN.
M.PCST,0 Length of the Process Data Block reserved for
customer definition.
| M.PFHT Default Page Fault Handler timer trap
| interval.
M.SCSZ,4000 SCA free core size.
MAXNBP,64 Maximum number of SNOOP UUO breakpoints.
MEMITL,4 Number of ways memory can be interleaved.
| MINCOR,Jobn*300 Assume 3 DDBS per job.
MINMAX,1024*12 Smallest legal CORMAX.
MLACTS,39 Maximum length of the account string.
NSPMEM,1000 1000 nano-seconds per memory cycle.
PAVJSP,20 Predicted average job size.
B-3
RLDTIM,112 Time to auto-reload.
SFC0,20 CPU0 scheduler fairness count.
SFC1,20 CPU1 scheduler fairness count.
TTDMOS,16 Maximum output string for RSX-20F.
UFC0,10 UUO fairness counter maximum.
UFC1,10 UUO fairness counter for slave.
B.2.1.2 Octal Symbols,Values -
| The following symbols are defined by values in octal radix. These
| default values can be changed with Question 110 of the SYSGEN portion
| of MONGEN.
BPTBIT,0 Initial user "SET DDT BREAKPOINT" setting.
CTYDCR,0 Delay for RETURN on CTY during ONCE.
CTYDLF,0 Delay for LINE FEED on CTY during ONCE.
| DEFBPT,0 Initial CPU breakpoint mask (one bit per
| CPU).
|
DEFDEB,0 Debugging word.
DTTRY,4 Number of times to retry on DECtape errors.
INDPPN,0 0 means project programmer numbers are
dependant across projects. That is, owner of
PPN is determined by programmer number only.
LIM1MS,4 Limit of stopcodes in any 1 minute period.
LIM8MS,7 Limit of stopcodes in any 8 minute period.
M.BMIN,0 Minimum number of guaranteed batch jobs.
M.CLSS,1 Scheduler is class system scheduler.
M.DCLS,0 Scheduler class for new jobs.
M.EMRT,0 Default is regular runtime.
M.ITA,0 Default is to use interval timer.
M.JMAX,M.JOB-M.BMIN Maximum number of simultaneous interactive
jobs.
M.PIDN,JOBN*2 Size of PIDTAB.
B-4
M.RCMP,0 Prohibits execution of restricted command
features from remote terminals.
M.STOF,0 Frequency to call start TTY output (must be
2^n-1).
M.WCH,JW.WMT Watch bits used to set initial.
M.XFFA,0 Do file access protection checking on [1,2]
and JACCT.
SCHEDN,0 Schedule value.
STDENS,5 Standard density (6250 BPI).
SYSSEG,-1 Minus one for fast GETSEG from SYS.
TTXTRA,0 Number of extra TTY DDBs for slave lines.
B.2.1.3 SIXBIT Symbols,Values -
| The following symbols are defined by values in SIXBIT radix. These
| default values can be changed with Question 111 of the SYSGEN portion
| of MONGEN.
|
| M.CRSH,CRSCPY Default crash copy program name.
|
| M.MAIL,MS Name of mail program
B.2.2 Symbols Defined in COMDEV
| The following symbols are defined in COMDEV. The symbols are listed
| in the form symbol,default-value.
B.2.2.1 Decimal Symbols,Values -
The following symbols are defined with values in decimal radix. These
default values can be changed with Question 109 of the SYSGEN portion
of MONGEN.
DSETIM,320 Maximum time to do a data security erase
(seconds).
IDLSEC,90 Any terminal idle this number of seconds gets
disconnected.
LPTSIZ,27+2 LPT buffer size in words (132 column
printer).
B-5
M.AIDL,n The number of seconds before idle ANF-10
lines are disconnected, where n defaults to
the current value for IDLESEC (see above).
M.DIDL,n The number of seconds before idle dataset
lines are disconnected, where n defaults to
the current value for IDLESEC (see above).
M.LIDL,n The number of seconds before idle LAT lines
are disconnected, where n defaults to the
current value for IDLESEC (see above).
M.NIDL,n The number of seconds before idle NRT/CTERM
lines are disconnected, where n defaults to
the current value for IDLESEC (see above).
|
| MSGMAX,512 Maximum message size in the network.
MTSIZ,128 Magtape buffer size in words.
NOISE,12 Noise record length in bytes.
|
| NTTPLN,32 Size of a terminal Process Control Block.
TAPTIM,60 Maximum time to skip 10 records (seconds).
| TTCHKS,8 Size of words in a TTY chunk, in powers of 2,
| with a minimum of 8. Increasing the size of
| a TTY chunk can increase TTY throughput.
|
TTYRMT,1 Remote terminal. If 1, terminal is remote.
If 2, terminal is local.
B.2.2.2 Octal Symbols,Values -
| The following symbols are defined by values in octal radix. These
| default values can be changed with Question 110 of the SYSGEN portion
| of MONGEN.
MAXODR,6 Maximum number of data requests for input
devices.
|
| MSGXMX,512 Maximum message size the -10 will transmit.
MTDAEM,10 Number of times per reel to call DAEMON for
magtape error reports.
MTDLTP,LT.BLP Default label type.
MTELOG,-1 Zero to log all recovered magtape errors.
B-6
B.2.3 Symbols Defined in COMMOD
| The following symbols are defined in COMMOD. The symbols are listed
| in the form symbol,default-value.
B.2.3.1 Decimal Symbols,Values -
| The following symbols are defined with values in decimal radix. These
| default values can be changed with Question 109 of the SYSGEN portion
| of MONGEN.
BLKMAX,100000 Maximum number of blocks to transfer (KL10).
BLKMAX,252 Maximum number of blocks to transfer (KS10).
CHVIFP,10 Standard initial fairness count for
positioning.
CHVIFT,10 Standard initial fairness count for
transferring.
DSKTRY,10 Number of error trys on disk data errors.
FIL4WD,9 Number of 8 word core blocks per job (for
NMB,PPB,UFB,ACC).
| M.BNDT,16 Number of seconds before binding CI disks at
| ONCE time.
|
MAXUSI,8 Maximum negative USETI (maximum number of
extended RIBs).
MFDSIZ,30 Number of blocks for initial MFD set by
refresher.
PTRLEN,7 Number of in-core retrieval pointers per
file.
RCLTRY,10 Number of times to try 'DSKTRY' data errors.
SERTRY,3 Number of times to try on search errors.
SWCLSN,7 Highest class for swapping.
SWPMAX,8 Maximum number of units for swapping.
UNVRSF,500 Reciprocal safety factor for each unit.
B.2.3.2 Octal Symbols,Values -
| The following symbols are defined by values in octal radix. These
B-7
| default values can be changed with Question 110 of the SYSGEN portion
| of MONGEN.
CCWMAX,1 Maximum number of data channel command words.
LBFSIZ,1000 Default large-buffer size (OP.LBF on an
OPEN). Specify the octal value in words in
multiples of 200.
LIMLVL,5 Maximum number of nested SFDs for this
monitor.
|
| LNMMXL,144 Maximum length of any logical name specified.
|
| LNMMAX,77 Maximum number of logical names.
M.CBMX,JOBN-1 Default blocks in disk cache.
M.DFLB,3 Default number of large disk buffers.
M.DFNB,6 Default number of disk buffers.
MFDPRV,555000 MFD privileges (all can READ, LOOKUP).
PRVFIL,057 Standard file privilege.
PRVSPL,077 Protection for spooled output.
PRVSYS,155 Protection for most system programs.
PRVUFD,775 Standard UFD privilege.
PRYSYS,157 Protection for SYS:*.SYS.
RIBERN,0 RIB error threshold for RIB errors on
structures in system search list.
SYNRPR,357000 Privileged files in SYS.
SYRDPR,355000 Non-privileged files in SYS.
SYSPRV,775000 SYS UFD privileges (project 1 can read/create
on SYS).
|
|
|
| B.2.3.3 SIXBIT Symbols,Values -
|
| The following symbols are defined by values in SIXBIT radix. These
| default values can be changed with Question 111 of the SYSGEN portion
| of MONGEN.
|
| M.QSTR, Default structure for device queues. The
| default value for M.QSTR is blank. The value
| is the SIXBIT name of the file structure for
B-8
| queuing programs set up by ONCE-only code to
| be the first stream up.
B-9
APPENDIX C
SETTING SCHEDULER PARAMETERS
The SCDSET program enables a privileged user to accomplish the
functions of the SCHED. monitor call, and to manipulate a SCDMAP.SYS
file. For example, you can use SCDSET to do the following:
o Set system usage quotas
o Adjust scheduler parameters
o Create and modify a SCDMAP.SYS file
o Move a SCDMAP.SYS file into the SYS area
o Update the class of all logged-in jobs automatically
C.1 REQUIRED PRIVILEGES
The privileges required to use SCDSET depend on the functions you want
to accomplish. The following list describes the privileges required
for different functions.
o To manipulate a SCDMAP.SYS file, you need only to have the
privileges necessary to move the file into the SYS area.
o To set system usage quotas, and adjust scheduler parameters,
you must be logged in as [1,2], or the SCDSET program must
have JACCT privileges.
o To use the READ/SET functions, you must be logged in as
[1,2].
C.2 CAUTIONS
If you misunderstand how the scheduler works, or make typing mistakes
C-1
while using SCDSET, you may unintentionally halt all work being done
on the system. Therefore, practice working with SCDSET and the
scheduler during non-production timesharing until you are familiar
with both of them.
To prevent system quotas and sytem load from affecting your work, run
SCDSET in a High Priority Queue (HPQ) except when you create or modify
a SCDMAP.SYS file. If you use an HPQ when you create or modify a
SCDMAP.SYS file, you adversely affect system performance.
C.3 SCDSET COMMANDS
This section describes the top-level SCDSET commands. It also
describes EDIT subcommands, and the parameters you can read and set
with the READ/SET commands.
C.3.1 EDIT
Calls the MAKMAP routine. The MAKMAP routine enables you to create or
modify a SCDMAP.SYS file. EDIT uses the following subcommands:
SUBCOMMAND FUNCTION
COPY/SET Copies a file from a area or device to the SYS area.
EXIT Returns to top-level.
HELP Displays EDIT help text.
ONE CLASS Changes one scheduler type to a scheduler class in the
SCDMAP.SYS file.
RANGE Changes a range of scheduler types to scheduler classes
in the SCDMAP.SYS file.
READ Reads in a SCDMAP.SYS file for modification.
TYPE Displays a specified range of scheduler types.
WRITE Writes a SCDMAP.SYS file to a device or area.
C.3.2 EXIT
Returns your job to monitor level.
C-2
C.3.3 HELP
Displays help text.
C.3.4 READ/SET
Execute functions of the SCHED. monitor call. You can read or set
the following parameters:
PARAMETER MEANING
AVG SWAP TIME
The estimated average swap time for a typical job on
the system. The AVG SWAP TIME controls the rate at
which the swapper changes from one class to the next
when it chooses jobs to swap in.
BB CLASS
The class to run as background batch. Background batch
class jobs run whenever the system is otherwise idle.
PARAMETER MEANING
BB SWAP TIME
The estimated time required to swap a background batch
job. The scheduler uses the BB SWAP TIME to decide
when the system should permit a background batch job to
be swapped in.
CORE SCHEDULING INTERVAL
The rate at which the system stops accounting for
in-core protect time and stops using in-core protect
time to requeue jobs. The CORE SCHEDULING INTERVAL is
important for systems with a large amount of memory, or
very few users.
CPU SCAN ORDER
The scan table used by a CPU or CPUs.
DEFAULT CLASS
The class new jobs assume when they first log in. The
DEFAULT CLASS applies to a new job until LOGIN reads
the job's class from SCDMAP.SYS and changes the class
appropriately.
FREE CORE GOALS/LIMITS
The percentage of free and potentially free memory that
must be available to the swapper. If this percentage
is not available, the swapper swaps out jobs in long
term wait.
HIGH SEGMENT RETENTION TIME
The dormant high segment retention time is high segment
C-3
in-core protect time. It prevents the deletion from
memory of high segments which have become dormant until
a specified time has elapsed.
IN-CORE FAIRNESS FACTOR
The factor that controls the rate at which out of core
jobs and in-core jobs are scheduled for swapping.
JOB CLASS
The scheduler class of all logged in jobs.
MS INTERVAL
The Micro Scheduling interval (in jiffies). The MS
interval specifies how often the scheduler selects a
class to be scheduled.
PRIMARY PERCENTAGES
The system usage percentages for a particular job
class. The sum of all primary percentages may be
fixed. In this case, the sheduler only chooses a job
to run when its class has the highest priority. When
the sum is fixed, a class cannot exceed its class
quota. If the only runnable jobs are in this class,
the scheduler runs the null job.
PROT
A constant used to calculate the minimum core usage
quota. The measurement is in microseconds. PARAMETER
MEANING
PROT1
The in-core timeslice used after a job has expired its
initial in-core protect time. PROT1 controls the rate
at which I/O bound jobs circulate around the run
queues.
PROTM
The maximum in-core protect time assigned to a job. It
specifies an upper limit for very large jobs.
PROTOT
A constant used to calculate the minimum core usage
quota. The measurement is in microseconds.
RESPONSE FAIRNESS
The rate at which the scheduler chooses jobs to run.
The rate depends on the order of the jobs in the "just
swapped in list", and the allocations specified by the
job classes.
RUNTIME BY CLASS
The amount of runtime each class has used since the
last time that a class quota was changed. The
measurement is in jiffies. This parameter also
displays the percentage of runtime each class has used.
C-4
SCHEDULER FAIRNESS
The rate at which PQ2 jobs are scheduled ahead of PQ1
jobs. SCHEDULER FAIRNESS allows compute bound jobs to
run even when there are many interactive jobs running.
SECONDARY ALLOCATION
The maximum amount that a class will be given after
using up the primary percentage.
SWAPPER FAIRNESS
The rate at which PQ2 jobs are swapped in ahead of PQ1
jobs.
TIME BASE
The base run quanta (in milliseconds) for either run
queue. Note that internally, the scheduler stores this
value in local jiffies. Therefore, it may not be
possible to adjust the time slice by a few
milliseconds.
TIME MAXIMUM
The maximum quantum runtime that a job uses.
TIME MULTIPLIER
The parameter that asssigns a larger quantum runtime
for large jobs. The quantum runtime is the result of
the following:
(TIME BASE)+(TIME MULTIPLIER)*(job size)
The scheduler compares the quantum runtime to a run
quantum maximum, and assigns the smaller of the two
values to large jobs.
C.3.5 UPDATE JOBS
The UPDATE JOBS command updates all the job parameters defined above
for logged in jobs, and writes the changes to files if requested.
UPDATE JOBS accomplishes this through the FIXJOB routine, which does
the following:
o Asks if you want to move a new SCDMAP.SYS file to the SYS
area. If you answer yes, you move the SCDMAP.SYS file. If
you answer NO, FIXJOB calls the MOVMAP routine to move the
SCDMAP.SYS file.
o Gets each logged in job's PPN.
o Determines whether to use the batch half or timesharing half
of the SCDMAP.SYS file currently on SYS.
o Looks in ACTDAE.SYS for the user's scheduler type.
C-5
o Executes the SCHED. monitor call to put that job in the
correct scheduler class.
o Prints an error message for each logged in job it does not
find in ACTDAE.SYS, or for which the SCHED. monitor call
failed.
The FIXJOB routine does not execute the SCHED. monitor call for [2,5]
jobs, or for its own job.
C.4 THE SCDMAP.SYS FILE
You must have the SCDMAP.SYS file on SYS. The file contains 1024
(decimal) nine bit entries (256 decimal words). The first 512 entries
constitute the map for timesharing users, the second 512 entries
constitute the map for batch users. Each entry contains a number
between zero and thirty-one inclusive (0-31) that specifies a
scheduler class. Each of the 512 scheduler types for both timesharing
and batch has an entry. Thus, at login time, LOGIN looks in
ACTDAE.SYS for the PPN's scheduler type and then looks in SCDMAP.SYS
for the current scheduler class for that scheduler type and job type
(timeshare or batch). If you do not want to differentiate between
timeshare and batch, the second half of the map should be identical to
the first half.
The format of the SCDMAP.SYS file is as follows:
Table C-1: SCDMAP.SYS Format
BITS
0-8 9-17 18-26 27-35
Word 0 n0 n1 n2 n3
Word 1 n4 n5 n6 n7
Word 2 n8 n9 n10 n11
Word 127 n508 n509 n510 n511
Word 128 m0 m1 m2 m3
Word 129 m4 m5 m6 m7
Word 130 m8 m9 m10 m11
C-6
Word 255 m508 m509 m510 m511
nXXX is the location of the Scheduler Class for timesharing users
of Scheduler Type XXX (as stored in ACTDAE.SYS), and mXXX is the
location of the Scheduler Class for batch users of Type XXX.
C-7
APPENDIX D
CREATING BOOTABLE FRONT END MEDIA
| This appendix documents the procedure for creating bootable front end
| media for RSX-20F. These media provide a fallback capability if the
| front end file system on your system's RP06 disk is damaged or lost.
|
| The installation process comprises two parts. Part 1 of the process,
| described in Section D.1, creates the "A" and "B" DECtapes or
| floppies, and requires that your system be standalone. Part 2,
| described in Section D.2, creates the "C" DECtape or floppy, and can
| be performed at any time, either before or after part 1 is performed,
| without shutting down timesharing on the KL.
|
|
|
| D.1 BUILDING THE PRIMARY BOOTABLE VOLUMES
|
| This procedure assumes:
|
| o that your system has a bootable front end file system mounted
| on the front end's hard disk drive.
|
| o that the system has been shut down, as this process must be
| done in a standalone environment.
|
| o that you have installed your front end files using the RSXINS
| procedure described in chapter 8.
|
| o that you have a set of three DECtapes or floppies in good
| condition to hold the front end files.
|
|
| The example below illustrates the procedure for both DECtapes and
| floppies. The left-hand column displays the DECtape procedure; the
| right-hand displays the floppy procedure.
|
| 1. Mount the disk pack containing a bootable RSX-20F system in the
| front end RP06.
D-1
| 2. Mount a pair of DECtapes or floppies on drives 0 and 1. These
| will become the A and B volumes of the set you are creating.
|
| 3. If you are writing DECtapes, ensure that both DECtape drives are
| write enabled.
|
| 4. Set the PDP-11 console switches to 000203.
|
| 5. Press the SW/REG and ENABLE switches at the same time.
|
| 6. Initialize the media on drive 0:
|
| For DECtapes: For Floppies:
|
| <CTRL/\> <CTRL/\>
| PAR>MCR INI<RET> PAR>MCR INI<RET>
| INI>DT0:/INDX=MID<RET> INI>DX0:<RET>
|
| The INI program will now initialize the volume. This process will
| take approximately 5 minutes for a DECtape or 30 seconds for a floppy.
| The INI program will not reprompt when it is complete.
|
| 7. Repeat step 6 for drive 1, substituting a 1 for the 0 in the
| command line for INI.
|
| 8. Now tell RSX-20F to mount the media:
|
| For DECtapes: For Floppies:
|
| <CTRL/\> <CTRL/\>
| PAR>MCR MOU<RET> PAR>MCR MOU<RET>
| MOU>DT0:<RET> MOU>DX0:<RET>
| MOU -- MOUNT COMPLETE MOU -- MOUNT COMPLETE
| MOU>DT1:<RET> MOU>DX1:<RET>
| MOU -- MOUNT COMPLETE MOU -- MOUNT COMPLETE
| MOU><CTRL/Z> MOU><CTRL/Z>
|
| 9. Create the [5,5] UFD on the volume on unit 0, using the UFD
| utility:
|
|
| For DECtapes: For Floppies:
|
| <CTRL/\> <CTRL/\>
| PAR>MCR UFD<RET> PAR>MCR UFD<RET>
| UFD>DT0:[5,5]<RET> UFD>DX0:[5,5]<RET>
|
| UFD, like INI, does not reprompt when it is finished. The UFD
| creation process takes about 3 minutes on a DECtape and about 15
| seconds on a floppy.
|
| 10. Repeat steps 8 and 9 for the volume on drive 1.
|
| The remaining commands are common to both DECtape and floppy systems.
D-2
| 11. Invoke the PIP utility to copy the files to both volumes:
|
| <CTRL/\>
| PAR>MCR PIP<RET>
| PIP>@BLDFE1<RET>
|
| BLDFE1 is a command file that tells PIP which files to put on each of
| the two volumes. It instructs PIP to copy the files to each volume
| and then to do a directory of the newly-created files. The process
| will take about 40 minutes for DECtapes and about 10 minutes for
| floppies.
|
| 12. Exit from PIP and boot the unrun image of RSX-20F from the file
| RSX20F.SYS. This image will use the newly created front end
| volumes as the system device SY:.
|
| PIP><CTRL/Z>
| <CTRL/\>
| PAR>MCR BOO<RET>
| BOO>DB0:RSX20F.SYS<RET>
|
| 13. Wait a few seconds and then type control-backslash. This will
| cause the newly-booted RSX-20F to search for the PARSER on
| DECtape or floppy 0.
|
| Please note that all commands will take somewhat longer to perform, as
| the files must now be read in from DECtape or floppy.
|
| <CTRL/\>
| INS -- SY MOUNTED, TSK=PARSER
|
| PAR -- [PARSER] ESE - EBOX STOPPED - EXAMINE
| PAR>
|
| 14. If you are building a floppy-based system, please skip this
| step. The DECtape version of RSX-20F does not mount DECtape 1,
| so you must do it manually:
|
| PAR>MCR MOU<RET>
| MOU>DT1:<RET>
| MOU -- MOUNT COMPLETE
| MOU><CTRL/Z>
|
| 15. Next, set some PARSER parameters and save the full RSX-20F image
| to the front end media. The last command, which directs SAV to
| save the system core image, will take about 10 minutes to
| execute. Once the save is completed, the newly saved software
| will be reloaded automatically.
|
| <CTRL/\>
| PAR>SET CONSOLE MAINTENANCE<RET>
| CONSOLE MODE: MAINTENANCE
| PAR>SET FAULT-CONTINUATION<RET>
| PAR>SET DATE<RET>
| DATE: 16-MAR-88<RET>
D-3
| TIME: 16:11<RET>
| CURRENT SYSTEM DATE:
| WEDNESDAY, 16-MARCH-88 16:11
| VALIDITY FLAG IS:ON
| PAR>MCR SAV<RET>
| SAV>SY0:/WB<RET>
| [Dn0: dismounted]
| [Dn1: dismounted]
|
|
| RSX-20F Vx##-## 16-MAR-88
|
| [SY0: redirected to Dx0:]
| [Dn0: mounted]
| [Dn1: mounted]
|
| PAR -- [PARSER] ESE - EBOX STOPPED - EXAMINE
| PAR>
|
| 16. Lower and raise the HALT switch on the PDP-11 console, set the
| PDP-11 console switches to 000203, and press the SW/REG and
| ENABLE switches at the same time.
|
| You may now remove the DECtapes or floppies from the drives. Please
| be sure to place labels on them identifying the volumes -- the one on
| drive 0 should be labelled "A"; the one on drive 1, "B".
|
| You now have a minimal bootable RSX-20F system on the first two
| DECtapes or floppies. If you wish, you may build the third volume
| now, or you may reboot the KL system and build the third volume after
| timesharing has been restarted.
|
|
|
| D.2 BUILDING THE THIRD VOLUME
|
| To have a complete copy of RSX-20F on console media, you must create a
| third volume of front end media. This volume will hold additional
| files that will not fit on the first two of the set.
|
| The steps involved can be done immediately after Part 1 is completed,
| or they may be done later. It is not necessary to shut down the KL
| before starting this procedure.
|
| 1. Mount the floppy or DECtape that is to receive the files on
| drive 0.
|
| 2. Initialize the media on drive 0:
|
| For DECtapes: For Floppies:
|
| <CTRL/\> <CTRL/\>
| PAR>MCR INI<RET> PAR>MCR INI<RET>
| INI>DT0:/INDX=MID<RET> INI>DX0:<RET>
D-4
| The INI program will now initialize the volume. This process will
| take approximately 5 minutes for a DECtape or 30 seconds for a floppy.
| The INI program will not reprompt when it is complete.
|
| 3. Now tell RSX-20F to mount the device, making it available for
| use:
|
| For DECtapes: For Floppies:
|
| <CTRL/\> <CTRL/\>
| PAR>MCR MOU<RET> PAR>MCR MOU<RET>
| MOU>DT0:<RET> MOU>DX0:<RET>
| MOU -- MOUNT COMPLETE MOU -- MOUNT COMPLETE
| MOU><CTRL/Z> MOU><CTRL/Z>
|
| 4. Create the [5,5] UFD on the volume, using the UFD utility:
|
| For DECtapes: For Floppies:
|
| <CTRL/\> <CTRL/\>
| PAR>MCR UFD<RET> PAR>MCR UFD<RET>
| UFD>DT0:[5,5]<RET> UFD>DX0:[5,5]<RET>
|
| UFD, like INI, does not reprompt when it is finished. The UFD
| creation process takes about 2 minutes on a DECtape and about 10-15
| seconds on a floppy.
|
| 5. Now invoke PIP, and tell it to copy the files. The commands in
| this example are common to both DECtape and floppy systems.
|
| <CTRL/\>
| PAR>MCR PIP<RET>
| PIP>@BLDFE2<RET>
|
| BLDFE2, like the BLDFE1 file used in Part 1, tells PIP which files
| belong on this volume. It will also do a directory of the volume when
| it is finished copying those files. The process will take about 20
| minutes for DECtapes and 5 minutes for floppies.
|
| 6. Once the file copy is complete, exit from PIP and dismount the
| volume:
|
| For DECtapes: For Floppies:
|
| PIP><CTRL/Z> PIP><CTRL/Z>
| <CTRL/\> <CTRL/\>
| PAR>MCR DMO<RET> PAR>MCR DMO<RET>
| DMO>DT0:<RET> DMO>DX0:<RET>
| DMO -- DISMOUNT COMPLETE DMO -- DISMOUNT COMPLETE
| DMO><CTRL/Z> DMO><CTRL/Z>
|
| 7. Remove the DECtape or floppy from the drive. Label it as volume
| "C" of the set.
|
| The installation process is complete.
D-5
APPENDIX E
SOFTWARE INSTALLATION GUIDE GLOSSARY
This appendix lists and explains the terms and conventions used in
this manual.
AC
An accumulator.
Accounting Files
The files that are used to control access and use of computer
resources. ACTDAE.SYS and QUOTA.SYS are the accounting files.
See Chapter 9 for more information about these files.
AUTCON
The monitor module to automatically configure tape and disk
drives. AUTCON is part of the monitor, that detects changes in
status of I/O channels. It is also invoked at system startup to
build the appropriate data bases for mass-bus and CI bus devices.
BAF block
The BAT block that is stored in memory.
BAT block
The data table, stored on disk, containing information about bad
areas on disk.
CRASH.EXE
E-1
The file automatically written by BOOT when the system fails.
CRASH.EXE can be used to reload a suspended monitor.
CTY:
The console terminal for the front end and CPU. The CTY is
usually a hard-copy terminal, so that important information about
the system that is printed can be easily perused when the system
is not running.
CUSP
A Commonly Used System Program that works closely with the
monitor to perform system functions. For example, LOGIN, PIP,
and BACKUP are CUSPs.
DAY-MON-YEAR
The format of the system date you supply to the ONCE-only dialog.
Ersatz Device Names
The monitor associates certain device names with PPNs. Such a
device name is called an "ersatz device." The list of ersatz
device names can be expanded with Question 49 of the HDWGEN
portion of MONGEN.
File Structure
A file structure is one or more disk packs, mounted and spinning,
that have been defined to the monitor as a single entity to be
referenced and accessed by the file structure name. File
structure information is entered into the disk HOME blocks by the
ONCE program.
HH:MM:SS
The time you input for the system time in the ONCE-only dialog.
HIGH-ADDR
The high-bound physical address in memory that is referenced in
the ONCE dialog.
HOME block
The data block that is stored on disk which contains information
E-2
about the unit's file structure, access information, and amount
of swapping space.
Initial Job Search List
The job's initial search list (established by LOGIN from the
information in the accounting files.)
Job Search List
Those file structures that constitute DSK for the job.
LOW-ADDR
The lower-bound address of a physical address reference specified
in the ONCE dialog.
MFD
The Master File Directory of all the User File Directories. An
MFD is kept for each file structure.
ONCE
The monitor module that contains the ONCE-only mandatory system
startup dialog and I/O device initialization. In debugging mode,
ONCE preserves the monitor's symbol table. If the data is not
available, ONCE asks why the system is being reloaded, and allows
any of several startup options to change disk parameters and/or
start up the system.
ONCMOD
The monitor module containing the optional ONCE-only system
startup dialog for disk initialization. Structure and unit
parameters are changed using ONCMOD. TWICE runs ONCMOD during
timesharing.
OPR device
A designation made for the benefit of the monitor, which
indicates the terminal from which the operator's controlling job
is running. Monitor messages for the operator are directed to
the OPR device.
OPR program
E-3
The operator-interface program to the various components of the
GALAXY batch and spooling system.
PPN
A project-programmer number. [PPN] in square brackets represents
the User File Directory of the user with that PPN.
REFSTR
The monitor module that is called by ONCE and TWICE to refresh
structures.
SAB block
The SAT block that is stored in memory.
SAT block
The Storage Allocation Table, stored on each disk unit,
containing information about the status of the clusters on the
disk and their use bits, which tell whether clusters on the disk
are "in use," storing data, or "free" for storage of data.
STR-NAME
The name of a defined file structure, which you provide to the
ONCE dialog.
SYSINI
The monitor module responsible for system initialization. SYSINI
runs only in the policy, or boot, CPU. It invokes the ONCE
dialog, initializes device data blocks, maps memory, and sets up
monitor service routines, such as the RSX-20F interface. In
time, it causes CPU startup.
System Dump List
The list of file structures that will be searched for space to
store CRASH.EXE when the system fails.
System Search List
Those files structures that are searched when some ersatz device
names, such as SYS: or DEC:, are used.
E-4
UFD
The User File Directory that lists the files of each user in a
file structure.
UNIT-NAME
The physical unit name of a disk specified in the ONCE dialog.
E-5
INDEX
Account string definition, 11-10 switches, 9-14
Account validation FEFILE, 5-2
enabling, 9-7 FGEN, 9-14
setting up, 11-10 File structures
Assembling defining, A-16
monitor sources, B-1 modifying, A-11
the monitor, 9-16 refreshing, A-25
File system
Beware files, 5-7 front end, 6-1
Boot file specification KS microprocessor, 6-6
for KL, 3-3 TOPS-10, 5-5
for KS, 3-5 FRCLIN mechanism, 12-2
BOOT program, 6-6 Front end
Bootable tape, 9-19 file system, 6-1
Booting installation material, 2-3
from DECtape, 3-1
from floppies, 3-1 GALAXY spoolers, 10-2
KL front end, 3-2 GALAXY system
assembling, 10-11
Changing components, 10-1
disk parameters, A-13 starting, 10-14
Configuring stopping, 10-12
the monitor, 9-1 tailoring, 10-3
CPU serial number, 9-6 GALGEN, 10-3
CREDIR program, 5-5 GO startup option, A-24
Decimal symbols, B-2, B-5, B-7 INITIA monitor command, 12-10
Default INITIA program
disk parameters, A-20 disabling, A-25
Defining system startup, 12-1
disk units, A-13 Installation
file structures, A-16 materials, 2-1
structure owner, A-18 tapes, 2-2
swapping lists, A-11
system dump list, A-12 KS microprocessor file system,
system search list, A-13 6-6
DESTROY Startup Option, A-20
DESTROY startup option, 4-3 Loading
Device names, 5-5 the monitor, 9-17
Disk parameters unbundled software, 9-17
changing, A-13 Login failures, 11-12
defaults, A-20
Disk units, A-13 Minimum hardware configuration,
2-3
Enabling account verification, Modifying file structures, A-11
9-7 MONGEN
Ersatz device names, 5-5 program, 9-1
questions, 9-4
FAILUR.LOG file, 11-12 Monitor configuration, 9-1
Feature-test Monitor-build files, 9-1
configuration file, 9-1 Multi-CPU system front ends, 6-1
Index-1
Octal symbols, B-4, B-6, B-7 Specifying monitor configuration,
ONCE dialog, A-1 9-1
ONCE program, 4-1 Standard directories, 5-5
startup options, A-7 Starting the monitor, 9-18
OPR.ATO file, 12-10 Startup option
OPSER auto file, 12-10 DESTROY, 4-3
GO, A-24
Powering up the KL10, 2-5 REFRESH, A-25
Powering up the KS10, 2-5 Structure owner, A-18
Printing Beware files, 5-7 Swapping list, A-11
PROJCT.ACT file, 11-10 SYSCNF configuration file, 9-1
PROJCT.EXE file, 11-11 SYSGEN, 9-6
SYSJOB.INI
commands, 12-3
REACT program creating, 12-3
getting help, 11-2 example, 12-3
privileges required to run, file, 12-1
11-1 System dump list, A-12
profile attributes, 11-4 System name, 9-6
user profile, 11-4 System search list, A-13
REFRESH startup option, A-25
Refreshing file structures, A-25 Terminal lines
Reloading KS systems, 6-7, 8-9 number, 12-7
Restoring option, 12-8
FEFILE, 5-2 Terminal names, 12-7
TOPS-10 file system, 5-5
TTY.INI
Saving the monitor, 9-19 creating, 12-5
SCDMAP.SYS file, C-6 file, 12-2, 12-5
SCDSET program, C-1
Setting date and time, 4-2 USAGE accounting, 11-10
SIXBIT symbols, B-5, B-8 User runtime, 9-6
Index-2