5    Performing Default, Custom, and Cloned Installations

This chapter describes the information you supply for a default or custom installation and how to respond to prompts if you are performing a cloned installation.

The goal of this chapter is to provide the information necessary for you to perform the user actions shown in Table 5-1.

Table 5-1: Summary of User Actions

Remember that neither the cloned, default, nor custom installation preserves user or data files, so make sure your current operating system has been backed up. If errors occur that prevent a successful installation of Version 4.0 , you may need to restore the previous version of the operating system.

Before beginning the installation, ensure that you have performed all prerequisite tasks as described in Section 1.4.

5.1    What You See After Booting the System

What you see after you boot your system from the distribution media depends on the following:

• If you are performing a default or custom installation, refer to Section 5.1.1 if your system console has graphics capabilities or refer to Section 5.1.2 if your system console does not have graphics capabilities.

• If your system is registered for a configuration description file (CDF) for a cloned installation, refer to Section 5.1.3

5.1.1    If Your System Has Graphics Capabilities

If your system console has graphics capability and 32 MB or more of memory to support a graphic installation, the X Server is started and an Installation Setup window displays.

5.1.1.1    Unique Features of the Graphical User Interface

This section describes the unique features of the graphical user interface:

• Click on the fields in the Installation Setup window to enter your responses to the installation procedure.

• With the exception of the Root Password and Date and Time fields, you can enter the information on the Installation Setup window in any sequence. When you specify a password in the Root Password field, you are required to validate the password by entering it again. When you enter either a date or a time, you are required to enter the other as well.

• The graphical user interface contains extensive online help. To display the online help, click on the Help push button.

• For ease of use, you can click on the Setup Done button as soon as the Installation Setup window displays because default responses have been provided in all fields where an entry is required. Keep in mind that if you do this, only the mandatory software subsets are installed and the installation will occur on the default disk that is chosen by the installation procedure.

• If you do not provide certain essential site-specific information (such as root password, your system's host name, the date and time, and location and time zone), you will be prompted to enter that information during the configuration phase of the installation process.

• If you are performing a custom installation, you can select to install all software subsets at once by clicking on the Add All pushbutton in the Software Selection window.

• You can access the Disk Configuration application directly from the Installation Setup window to view or modify disk partitions.

5.1.2    If Your System Does Not Have Graphics Capabilities

If your system console does not have graphics capability or has less than 32 MB of memory, the following text-based installation procedure is displayed:

 

       Welcome to the Digital UNIX Installation Procedure

 

This procedure installs Digital UNIX onto your system.  You will
be asked a series of system configuration questions.  Until you
answer all questions, your system is not changed in any way.

 

During the question and answer session, you can go back to any
previous question and change your answer by entering:  history

 

There are two types of installations:

 

o  The Default Installation installs a mandatory set of
   software subsets on a predetermined file system layout.

 

o  The Custom Installation installs a mandatory set of
   software subsets plus optional software subsets that you
   select.  You can customize the file system layout.

 

The UNIX Shell option puts your system in single-user mode with
superuser privileges.  This option is provided for experienced
UNIX system administrators who want to perform file system or
disk maintenance tasks before the installation.

 

The Installation Guide contains more information about installing
Digital UNIX.

 

1) Default Installation
2) Custom Installation
3) UNIX Shell

 

Enter your choice: 

You may want to refer to Section I.1 and Section I.2 in Appendix I to review sample text-based installations before you actually start your own text-based installation. Reviewing these sample installations gives you an idea of what to expect. You can also use these sample installations to follow the progress of your own system installation.

5.1.2.1    Unique Features of the Text-Based Interface

This section describes the unique features of the text-based interface:

• The text-based interface presents a series of questions that you must answer before any disk configuration or software installation begins. Until you answer all questions and start the installation, your system is not changed in any way.

• At any time during the question and answer session, you can enter the word history to go back and change your answers.

• Online help is available for every question by entering the word help or a ? (question mark) at the corresponding prompt.

Section 5.2 summarizes the information you must enter.

5.1.3    If Your System Is Registered for a Cloned Installation

If you booted over the network to start the installation and the RIS server administrator registered your system to a RIS environment and to a configuration description file (CDF), a cloned installation procedure starts. The screen display is similar to the following:

*** Performing RIS Installation from server
 
A Configuration Description File,  filename.cdf was specified  for use during the installation of this system.

The Configuration Description File can be used by the installation to
provide the information related to file system layout and software
selection.

 

If you choose to use the Configuration Description File, the installation
will proceed and you will not have to answer any questions.  If you decide
not to use the Configuration Description File, the installation will
continue interactively.

 

Would you like to use the Configuration Description File? (y/n) [y]:

• If you decide not to use the CDF for a cloned installation and enter n, an interactive installation procedure begins. What you see next depends on whether or not the system console has graphics capabilities. Refer to Section 5.1.1 or Section 5.1.2 as appropriate.

• If you enter y, the CDF is retrieved from the RIS server and is validated for use on your system. Validation includes ensuring that the disk name and disk type specified in the CDF exists on your system. Refer to Section 5.1.3.1 and Section 5.1.3.2 for more information about CDF validation.

5.1.3.1    Successful CDF Validation

If CDF validation is successful, the installation procedure continues as shown in Section 5.16. File systems and swap space are created as specified in the CDF and the software subsets defined in the CDF begin loading after file system creation. Site-specific information such as host name, geographic location and time zone, and date and time is obtained from the RIS server. Later in the installation process, you will be required to boot off the newly installed disks, as shown in Section 5.18 and to enter a root password as shown in Section 5.5 (if a password is not already contained in the CDF).

5.1.3.2    CDF Validation Failures

This section describes the types of CDF validation failures that may occur. If you encounter CDF validation failures, you should verify with your RIS system administrator that your system is registered to the correct CDF.

• If CDF validation fails because the disk name specified in the CDF does not match a disk name attached to your system, the following message displays:

  *** Validating CDF prior to starting installation...
  *** Validation Error:
  
   
  
  The disk name "rz8" was not found on this system when
  attempting to validate the following filesystem(s): root usr
  

• If CDF validation fails because the CDF file includes a disk type for a specific device but the device on your system has a different disk type, the following message displays:

  *** Validating CDF prior to starting installation...
  *** Validation Error:
  
   
  
  The rz25 disk type specified in the Configuration Description
  File does not match the name/type found on this system for the
  following device(s): rz3/rz26l
  

The previous validation errors are saved in the /var/tmp/install.log file for your reference until you reboot the system.

You may restart the cloned installation procedure by doing one of the following:

• Enter restart at the root (#) system prompt

• Reboot the system from the RIS server

5.2    Summary of User Input

Regardless of the user interface (graphical or text-based), you should be prepared to provide the information shown in Table 5-2 for default and custom installations.

Table 5-2: User Input Required for Installation Procedure

Table Notes:

1. This site-specific information is obtained from the RIS server during Remote Installation Services (RIS) installations and cannot be edited. Enter this information only if you are using the CD-ROM as the distribution source of the software. If you do not make an entry in these fields, you are prompted for this information during the installation configuration phase.

2. If you do not specify a root password, you are prompted for one during the installation configuration phase.

3. The boot commands you have to enter are supplied by the installation procedure. When appropriate, enter the boot command sequence exactly as shown on your screen.

Section 5.3 through Section 5.20 provide additional reference information for each piece of information required by the installation procedure.

5.3    Choosing the Installation Type

Table 5-3 compares the features of the default and custom installation types to help you decide which installation type best suits your needs.

Table 5-3: Comparison of the Default and Custom Installations

Table Note:

1. You can choose the file system type for a custom installation, but for systems with less than 32 MB of memory, you do not have the option to use AdvFS as the file system type.

You should perform a custom installation if any one of the statements shown in Table 5-4 is true.

Table 5-4: Criteria for Deciding to Perform a Custom Installation

Table Notes:

1. You can choose the file system type for a custom installation, but for systems with less than 32 MB of memory, you do not have the option to use AdvFS as the file system type.

2. If you are installing a system that you plan to use as a dataless management services (DMS) server, you must perform a custom installation because you need more than just the mandatory software subsets on a DMS server. Instead of creating a link from /var to a mount point, you may want to create a separate file system for /var because the file systems needed for the dataless areas are located in the /var/adm/dms directory. Refer to Sharing Software on a Local Area Network for information about setting up disk partitions for DMS servers, software requirements for DMS servers, and installing software on DMS clients.

3. If you plan to install software on a remote installation services (RIS) server, you must perform a custom installation because you need more than just the mandatory software subsets on a RIS server. Instead of creating a link from /var to a mount point, you may want to create a separate file system for /var because the file systems needed for RIS areas are located in the /var/adm/ris directory. Refer to Sharing Software on a Local Area Network for information about setting up RIS servers.

You should perform a default installation if all of the statements shown in Table 5-5 are true.

Table 5-5: Criteria for Deciding to Perform a Default Installation

Table Note:

1. The size of the default disk partitions were designed so that a default installation fits on the disk. The default disk layout is: root is on partition a, /usr is on partition g, the var area is a directory under /usr, and the swap area is on partition b. If you perform a default installation, you cannot preserve existing custom partition tables or disk labels. Custom partition tables are partition tables that have been changed to sizes other than the default values. If a customized partition table exists, it will be overwritten.

If you are planning to install additional layered products later or may use this system as a DMS or RIS server, the default installation will not suit your needs.

The default installation is recommended for systems with limited disk space and less than 32 MB of memory.

5.3.1    Navigating Through the Installation Setup Window

Table 5-6 describes how to navigate through the Installation Setup window for custom installations:

Table 5-6: Quick Navigation Through the Custom Installation Setup Window

Table 5-7 describes how to navigate through the Installation Setup window for default installations:

Table 5-7: Quick Navigation Through the Default Installation Setup Window

5.4    Specifying a Host Name

If the host name was not obtained from the server (during a RIS installation), you must give your system a name. The host name is used to identify your system on the network. The following lists the guidelines for host names:

• Host names may contain from 2 to 63 alphanumeric upper or lower case characters (a-z, A-Z, 0-9).

• The first character in a host name must be a letter.

• Hyphens (-) are permitted in host names. Periods (.) are permitted only if you are entering the fully qualified domain name. Fully qualified host names can contain a maximum of 254 characters.

Following are examples of correct and incorrect host names:

Correct:	mysystem	mysystem.com	abc-university.edu
Incorrect:	my_system	1996.com	binary

Note

Do not use the words generic or binary for your host name because they have been reserved for use by the operating system.

You may want to consult your site system administrator before choosing your host name because site-specific restrictions (such as maximum length) may have been defined. You also do not want to choose a host name that is already being used by another system. If your system is already running a previous version of Digital UNIX and is connected to a network, you should keep the same host name because changing it would impact how your system is recognized on the network.

If you do not supply a host name, you are prompted for one during the installation configuration phase.

5.5    Specifying a Root Password

Every operating system has a superuser who has permissions that supersede those of ordinary users. This superuser is often referred to as the root user. The superuser is usually the system administrator. This user has access to all files and all devices and can make any changes to the operating system. The root user is said to have superuser privileges. For that reason, the root user (or root account), needs a special password.

Passwords should contain a combination of upper and lower case letters and a minimum of six to a maximum of 16 characters. The Digital UNIX operating system verifies only the first eight characters. Digital suggests using numbers and special characters such as the dollar sign ($), the percent sign (%), the number sign (#), the period (.), the hyphen (-), the underscore (_), or the at sign (@) in your password.

When choosing a password, you should not use:

• Words found in any dictionary (in any language)

• Personal information about you or your family such as names, addresses, birthdays, social security numbers, telephone numbers, names of pets

• Any combination of words in the dictionary and personal information

Do not choose a password that can be easily guessed by someone who knows you. Because the root user has absolute power over the operating system, the root password should be carefully protected.

The following are examples of correct and incorrect root passwords:

Correct:	U8one2too	wht%IZ-dne	DL_wrks@9	9Pnt.99%
Incorrect:	lowercase	nonumbers	Spot	7-6-58

Remember this password because you will need it to log in as the user root the first time you log in to the system after the installation. You can change your root password at any time (using the passwd command). When you enter the password, it is not displayed on your screen for security reasons. You must enter your new password again for verification.

If you do not supply a root password, you will be prompted for one during the installation configuration phase.

5.6    Entering the Date and Time

If the date and time was not obtained from the server (during a RIS installation), you must enter the current date and time. If you enter either the date or the time, you must enter the other.

If you do not supply a date and time, you are prompted to enter the date and time during the installation configuration phase.

The date is entered in the format mm-dd-yy where mm represents the month, dd represents the day of the month, and yy represents the year. For example, the digits entered in the order 03 15 96 represent March 15, 1996. The text-based interface requires that a hyphen (-) separate each set of digits, for example, 03-15-96.

Enter the time in digits using the 24-hour clock in the format hh:mm; the hh represents the current hour, for example, 14 represents 2 o'clock in the afternoon; the mm represents the minutes, for example, 06 represents the sixth minute of the hour.

If you do not specify the time, messages displayed during the installation procedure are shown in Greenwhich Mean Time.

If you do not specify a date and time for CD-ROM installations, the date and time recorded by the installation procedure might be later or earlier than the date and time for your time zone because the procedure has no way to determine date and time until the configuration phase when you will be required to enter a date and time. Although the absolute time displayed in the installation is incorrect, relative time elapsed is accurate. Therefore, you can still use time-stamping to determine how long the installation is taking.

5.7    Entering the Location and Time Zone

If the location was not obtained from the server (during a RIS installation), a menu lists the locations that are available. The location is used to set the time zone. If the location has more than one time zone, for example the United States, you must specify a time zone for the location. You should select the location that best describes your geographic location. If you do not select a location during a text-based installation, the default is Greenwich Mean Time (GMT). Table 5-8 describes the acronyms that appear in the location menu:

Table 5-8: Definitions of Location Acronyms

Table Note:

1. The W-SU time zone follows MET time zone rules. W-SU is provided only for backward compatibility and the MET time zone should be used instead.

If you do not supply a geographic location and time zone, you are prompted for this information during the installation configuration phase.

5.8    Choosing the Disk for the Root File System

The installation procedure requires that you specify a disk to contain the root file system. If you are performing a default installation, this is the only decision you have to make.

There are two requirements for the disk that contains the root file system:

1. The disk you choose for the root file system must be one of the supported disks shown in Table 3-2.

2. The root file system is always located on partition a of the disk you choose. Partition a must be at least 98,304 blocks (48 MB) and must start at block 0 (zero), the beginning of the disk. Digital recommends selecting a disk where the size of partition a is 64 MB (131,000 blocks).

The following information is displayed for each disk connected and available to your system:

• Device name, for example, rz0

• Disk type, for example, RZ26

• Disk number, for example, 0

• Controller type to which the disk is connected, for example, SCSI

• Controller number to which the disk is connected, for example, 0

The unit number for the disk that contains the root file system must be in the range 0 to 255 for ra type devices, in the range 0 to 47 for re type devices (SCSI disks employing RAID technology), and in the range 0 to 511 for rz type devices. This information is pertinent if your system has, for example, more than 511 rz disks because the disks with unit numbers greater than 511 will not be displayed for selection during the installation. The installation procedure automatically displays the device name (with unit numbers) for each disk connected to your system.

5.8.1    Disk Size Restrictions

If you are using the text-based installation interface, disks that do not have partitions large enough to hold the root and /usr file systems and the swap1 area on the same disk are marked with an asterisk to the left of the Disk Type column. As an example, RZ55 and RZ24L disks have a partitions large enough to contain the root file system, but neither disk has a g (or other) partition large enough for the /usr file system. Neither disk type is able to hold all file systems and another disk is required.

This is not a problem during custom installations because you have the opportunity to select different disks and partitions. If you are performing a custom installation, you can, for example, use RZ55 and RZ24L disks as required and allocate the /usr (and other) file system to other disks and disk partitions on your system. The default installation uses only default partitions and puts all file systems on a single disk. Therefore, disk types such as the RZ55 and RZ24L cannot be used for a default installation.

If you are performing a default installation with the text-based interface, and you try to install Digital UNIX Version 4.0 on a disk that is too small, you will see a message similar to the following:

The disk you selected is too small for a Default installation.
You may switch to a Custom installation to distribute the
software on multiple disks, or you may select a larger disk
to continue the Default installation.

 

1) Switch to Custom installation
2) Select a different disk

 

Enter your choice:

If you select option 1, you can use the custom installation procedure to allocate partitions on other disks for the /usr and other file systems or spread the software across multiple disks. Review Chapter 1 and read Chapter 3 before you continue with the custom installation.

Option 2 lets you choose a different disk. You must choose a disk large enough to contain root, and /usr file systems and swap1 area to continue the default installation or the same message is repeated. If your system has another disk and it is not marked with an asterisk, you can choose this option and continue with the installation. Refer to Appendix F and Appendix G to ensure that the disk you choose has enough space.

5.8.2    Disk Label Handling

The disk selected to contain the root file system always has a new disk label written to it, regardless of whether one already exists or not. This ensures that a valid bootstrap exists on the disk. Without one, the disk is not bootable.

Before writing a new disk label, the disk is checked for a preexisting disk label. If none is found, which is typical for a disk that has never been used for Digital UNIX before, a disk label is written using the default partition information.

The following describes how disk labels are handled by the installation procedure:

• During a default installation, if the disk chosen for root already had a default disk label, the disk label is preserved, new boot blocks are written, and the disk label is restored.

  If the disk chosen for root had a customized disk label, a default disk label is applied, and the partitions used for root, /usr, and var and any partitions that overlap may be overwritten.

• During a custom text-based installation, if no disk label exists, a disk label containing the default partition information is used. If a disk label does exist, its partition information is compared to the default partition information. If the disk labels differ, you can choose either the customized or the default.

• During a custom graphical installation, if no disk label exists, a disk label containing the default partition information is used. A warning displays if an existing disk label does not match the default partition table and the default file system layout was used. You then have the option to invoke the Disk Configuration application to change the partition table.

If an existing disk label is selected, it is saved and rewritten to the disk with a valid bootstrap. The new disk label is identical to the previous disk label. If the default disk label is selected, a disk label containing the default partition information is used.

5.8.3    Preserving Data on an Existing Disk

During a custom installation, you can choose the disk partitions on which the root, /usr, and /var, file systems and swap areas will be installed. If the chosen partitions contain data or user files, the data is lost (overwritten). Selecting a partition for one of these file systems that overlaps the partitions containing the data to be preserved has the same effect. Also, modifying the partition information in the disk label in such a way that the partitions containing data have their size, offset, or both modified causes the data to be lost. However, if the partitions containing the data are undisturbed, their contents will be untouched.

Assuming that the partitions to be preserved were undisturbed during the installation, the partitions can be used on the newly-installed system. If the partition contained a file system, it should be capable of being mounted and accessed. If the partition you are preserving were using the Advanced File System (AdvFS), make sure you select AdvFS as the file system type.

5.8.4    ULTRIX Partition Tables

This section applies only if you chose the custom installation.

If the disk you choose to contain the root file system was previously used on an ULTRIX system, it will be formatted with ULTRIX partition tables. You will see a message similar to the following:

ULTRIX compatible partition data found.
This data may be different than the standard
partition layout information in /etc/disktab.

 

ULTRIX partition table layout is:
partition     bottom        top       size    overlap
    a              0      40959      40960    c,h
    b          40960     163839     122880    c
    c              0     832526     832527    a,b,d,e,f,g,h
    d         163840     386735     222896    c,g
    e         386736     609631     222896    c,g
    f         609632     832526     222895    c,g
    g         163840     832526     668687    c,d,e,f
    h              0          0          0    a,c

 

Use the ULTRIX-style partition data? [y]:

If you enter y to use the ULTRIX layout, a Digital UNIX disk label is created for the disk. This label will correspond to the existing ULTRIX partition table. If you enter n, the default partitions for Digital UNIX are created.

5.8.5    Using the Default or Existing Disk Partition Table

During a custom installation, the graphical user interface detects a customized partition table when root, /usr, and var are on the same disk. When you click on the Select Software... pushbutton, a dialog box notifies you that the existing, customized disk partition table does not match the default partition table.

• If you want to view or change the existing partition table, click on Cancel to dismiss the dialog box and then click on Partition Disks....

• If you want to use the customized disk partition table and proceed directly to software selection, click on OK to dismiss the dialog box and display the software selection window.

If you are using the text-based user interface to perform a custom installation and your system has nonstandard disk partitions on the disk that contains the root file system, you are prompt to keep the nonstandard partitions or replace them with the disk's default partitions.

If you have carefully planned your partition layout and you want to keep the partitions you have on the disk, choose the existing partition table. If neither the default nor existing partition tables are suitable, exit the installation procedure and use either the Disk Configuration application or the disklabel command to modify the partitions on the disk.

The following information is shown for each disk partition:

• Partition - provides the name of the partition (the letters a through h).

• Start - the block number at which the partition begins.

• Size - the total number of 512-byte blocks in the partition. One block equals -1/2 kilobyte (kB) or 512 bytes. One megabyte (MB) equals 1024 kB (1,048,576 bytes), or 2048 blocks. If you need to determine the size in MB, divide the size in blocks by 2048.

• End - the block number at which the partition ends.

• Overlaps - displays the other partitions with which the partition overlaps.

If the disk you chose has a customized partition table, the display is similar to the following:

The rz3 disk has a non-default partition table.

 

           Partition   Start   Size      End     Overlaps
 Default        a          0  131072    131071   c
                b     131072  262144    393215   c
                c          0 2050860   2050859   a b d e f g h
                d     393216  552548    945763   c g
                e     945764  552548   1498311   c g h
                f    1498312  552548   2050859   c h
                g     393216  819200   1212415   c d e
                h    1212416  838444   2050859   c e f
 Existing       a          0  263340    263339   c
                b     263340 1787520   2050859   c d e f g h
                c          0 2050860   2050859   a b d e f g h
                d     393216  552548    945763   b c g h
                e     945764  552548   1498311   b c g h
                f    1498312  552548   2050859   b c h
                g     393216  819200   1212415   b c d e h
                h     263340 1787520   2050859   b c d e f g

 

Choose which partition table to use.
1) Default table
2) Existing table

 

Enter your choice:

5.9    Description of File System Types: UFS and AdvFS

This section describes the two file system types, UFS and AdvFS, that are available for custom installations. This information may help you decide whether or not to perform a custom installation because the default installation does not give you the option to choose file system type.

The custom installation lets you choose between the UNIX file system (UFS) or the Advanced File System (AdvFS) as the file system type for the root, /usr, and /var file systems.

Unless you choose otherwise, UFS is the default file system on Digital UNIX systems. On systems with less than 32 MB of memory, you do not have the option to use AdvFS as the file system type; the UNIX file system (UFS) is chosen automatically.

UFS has a more rigid hierarchy than AdvFS. In a UFS file system, each disk (or disk partition) contains one separate file system. The UFS file system is characterized by a hierarchical structure, the ability to create and delete files, dynamic growth of files, the protection of file data, and the treatment of peripheral devices.

UFS is compatible with the Berkeley 4.3 Tahoe release. UFS allows a pathname component to be 255 bytes, with the fully qualified pathname length restriction of 1023 bytes. The Digital UNIX implementation of UFS supports a maximum file size equivalent to the largest supported file system (128 GB).

Refer to the System Administration guide for more information about UFS.

The POLYCENTER Advanced File System (AdvFS) is a journaled local file system that provides higher availability and greater flexibility than traditional UNIX file systems. Using transaction journaling, AdvFS recovers file domains in seconds rather than minutes after an unexpected restart such as a power failure. AdvFS journaling also provides increased file system integrity. AdvFS provides greater flexibility by allowing filesets (file systems) to share a single storage pool and enabling hard and soft fileset quotas in addition to user and group quotas. AdvFS supports a maximum file size of 128 GB.

Refer to the System Administration guide or the POLYCENTER Advanced File System and Utilities for Digital UNIX, Guide to File System Administration for more information about AdvFS. Contact your Digital representative about obtaining POLYCENTER documentation.

5.10    Choosing the Location and File System Type for /usr

This section applies only if you are performing a custom installation.

If you chose not to use the default file system layout, you must choose a disk and partition on which the /usr file system will reside. You have the option to choose between UFS or AdvFS as the file system type for the /usr file system.

You can go back and change the disk and partition for /usr if the partition is too small to hold the optional software you select later on in the installation procedure.

If you need more information about the contents of the /usr file system, refer to Section 3.7.

5.11    Choosing the Location of the var Area

This section applies only if you are performing a custom installation.

You can place the var area either as a directory under the /usr file system or create a separate var file system.

If the system you are installing is a Dataless Management Services (DMS) server, you should allocate a separate file system for /var because all dataless environments reside in /var/adm/dms on the server. Putting var under /usr could mean that your system will run out of disk space when you create dataless environments. Refer to Sharing Software on a Local Area Network for more information about allocating the var area and calculating disk space for DMS environments and RIS servers.

If you plan to set up your system as a RIS server, you should allocate a separate file system for /var because all RIS environment information is stored in the /var/adm/ris directory. In addition, if the RIS user chooses to extract the data for the RIS area from the distribution media rather than symbolically linking to the area, this data will also be stored in /var/adm/ris. Refer to Sharing Software on a Local Area Network for more information about allocating the var area and calculating disk space requirements for RIS servers.

If you decide to create a separate file system for /var, you can choose between UFS or AdvFS as the file system type for var.

If you need more information about the contents of the /var file system, refer to Section 3.8.

5.12    Choosing the Location of Swap Space

You must select the disk and partition on which you want to allocate the primary swap space. Digital recommends a minimum of 128 MB of swap space. If your swap partition selections do not amount to 128 MB, a warning message is displayed as a reminder that you should allocate more swap space.

Although you cannot choose the swap strategy modes during the installation procedure, there are two strategies for swap allocation: immediate and over-commitment. The swap strategy mode for Digital UNIX systems with greater than 32 MB of memory is immediate mode which means that swap space is allocated when modifiable virtual address space is created. This mode requires more swap space than over-commitment mode because it guarantees that there will be enough swap space if every modifiable virtual page is modified.

On systems with less than 32 MB of memory, the swap strategy mode is deferred or over-commitment (also known as lazy swap). This means that swap space is not allocated until the system needs to write a modified virtual page to swap space. To optimize performance on systems with limited capacity, swap space is not allocated until a process needs it, not when a process starts up.

Refer to System Administration for more information about swap allocation strategies and how to switch from one swap allocation mode to the other after the installation.

If you need more information about planning swap space, refer to Section 3.9.

5.12.1    Allocating a Second Swap Area

This section applies only if you are performing a custom installation.

You have the option to allocate a second swap area during a custom installation.

To optimize the performance of your swap space, spread out your swap space across multiple devices and use the fastest disks for swap devices. To ensure the best performance, place each swap area on a separate disk instead of placing multiple swap areas on the same disk.

5.13    Installing Mandatory Software Subsets

The following software subsets are the minimum required for the Digital UNIX Version 4.0 operating system. A default installation automatically installs only these software subsets; a custom installation installs these software subsets plus the optional software subsets you select. Some of the software subsets designated as mandatory depend on your system's hardware. For example, there are four supported keyboard types; only the software subset supporting the keyboard type connected to your system is mandatory.

   Base System
   Base System - Hardware Support
   Base System Management Applications and Utilities
   Basic Networking Configuration Applications
   Basic Networking Services
   Compiler Back End
   Hardware Kernel Header and Common Files
   Hardware Kernel Modules
   Kernel Header and Common Files
   Keyboard Support
   Local Printer Support
   NFS(tm) Configuration Application
   NFS(tm) Utilities
   Standard Kernel Modules
   Tcl Commands

Depending on your system's graphics options, either DECwindows 75dpi Fonts or DECwindows 100dpi Fonts is mandatory. The mandatory X Server software subset depends on whether your system has a TurboChannel bus, QVision graphics adapter, or PCI bus. If your system has graphics capability, the following windowing and graphical applications software subsets are installed as mandatory automatically:

   Adobe Fonts
   Basic X Environment
   CDE Desktop Environment
   CDE Mail Interface
   CDE Minimum Runtime Environment
   DECwindows Fonts
   Graphical Base System Management Utilities
   Graphical Print Configuration Application
   Graphical System Administration Utilities
   Netscape Navigator V1.12I
   Old X Environment
   Tk Toolkit Commands
   X Fonts
   X Servers Base
   X Servers 

If Asynchronous Mode Transfer (ATM) hardware is detected during the installation process, the following software subsets are also installed as mandatory:

   ATM Commands
   ATM Kernel Header and Common Files
   ATM Kernel Objects
   ATM Kernel Modules

The following POLYCENTER AdvFS software subsets are usually optional. They become mandatory during a custom installation if AdvFS is chosen as the file system type for root, /usr, or /var:

   POLYCTR advfs
   POLYCTR advfs Kernel Modules

Refer to Appendix C for descriptions of the mandatory software subsets.

The amount of free space remaining in the root, /usr, and /var file systems is displayed to indicate if the disk partitions you chose for those file systems are large enough to hold the mandatory software subsets. Space remaining is shown in gigabytes (GB), megabytes (MB), or kilobytes (kB):

• If you click on View Software... for default installations or Select Software... for custom installations, the graphical user interface shows the amount of free space remaining in the File System Status fields at the bottom of the software selection window.

• The text-based interface shows space remaining in a display similar to the following:

  Free space remaining (root/usr/var): 18.4MB/176MB/192MB
  

The installation procedure will prevent you from selecting a disk that is too small to hold the mandatory software subsets. However, if you feel that the disks will not have enough free space remaining after the installation of the mandatory software subsets, go back and select another larger disk. You can also go back and perform a custom installation where you can customize the file system layout.

5.14    Selecting Optional Software Subsets

This section applies only if you are performing a custom installation because you cannot select optional software during a default installation. Default installations only let you view the mandatory software that will be installed automatically.

When you select optional software subsets, the amount of free space remaining in the root, /usr, and /var file systems is displayed to indicate if the disk partitions you chose are large enough to hold the software you are selecting. Space remaining is shown in gigabytes (GB), megabytes (MB), or kilobytes (kB).

• The graphical user interface shows the amount of space remaining in the File System Status fields at the bottom of the Software Selection Window. The amount of space remaining is displayed and updated as you select each optional software subset.

• The text-based interface shows space remaining in a display similar to the following:

  Free space remaining (root/usr/var): 18.4MB/176MB/192MB
  

  The figures are updated after pressing the Return key as each optional software subset is selected.

If you select a software subset that has a dependency with another subset that is not yet selected, the other subset is selected automatically.

• The graphical interface displays a dialog box alerting you of dependent software; clicking on OK automatically selects the other software for installation.

• The text-based interface automatically installs software subset dependencies. When you select a software subset with a dependency, a message similar to the following displays:

  The chosen subset(s) require one or more additional subset(s)
  which will be loaded automatically:
    * Doc. Preparation Tools  (OSFDCMT400)
  

Some optional software subsets are hardware specific; that is they are optional because you do not have the hardware or graphics capabilities which they support. The Windowing Environment category, which provides support for numerous keyboard types, is an example of this situation. Therefore, during text-based installations, it is not necessary to select ALL mandatory and all optional subsets, or during a graphical installation to click on Add All because you will install hardware-specific software that your system does not need.

Section I.2.1 in Appendix I provides a list of the optional software subsets that are available to install regardless of the interface you are using. Refer to Appendix C for descriptions of the optional software subsets.

5.14.1    Selecting Optional Software Using the Text-Based Interface

When making software subset selections with the text-based interface, separate multiple selections with a space and enter consecutive ranges with a hyphen between the beginning and ending range of numbers. There may be more optional software subsets than can fit on one screen. You may enter your selections screen by screen or all at once at the end of the list.

If you want to change your selections, press the Return key until you reach the end of the software subset list. Select the option to CANCEL selections and redisplay menus to start the selection process again.

It may not be wise to select ALL software subsets because you will select hardware-specific software subsets that your system does not need (such as fonts, keyboard types, and Xservers).

When you are finished making optional software subset selections, press Return at the prompt displayed at the end of the software subset list:

 


.
.
.


 

The following choices override your previous selections:
77) ALL mandatory and all optional subsets
78) MANDATORY subsets only
79) CANCEL selections and redisplay menus

 

Add to your choices, or press RETURN to confirm previous choices.
Free space remaining (root/usr/var): 18.4MB/176MB/192MB

You have the opportunity to confirm your selections before software subset loading begins.

5.14.1.1    If File Systems Are Full After Selecting Optional Software

As you are selecting optional software subsets using the text-based interface, free disk space is calculated automatically. Review these numbers as you make your selections because if your file systems are near capacity, you have a few options:

• Use the history command to go back and start software selection again or choose the selection number that corresponds to CANCEL selections and redisplay menus to reselect optional software subsets. This time, select only those software subsets actually needed.

• Use the history command to go back and select a different disk with larger partitions to contain the root, /usr, and /var file systems and swap space.

• Use the history command to go back and place the file systems and swap space on separate disks.

• If you are an experienced UNIX user, enter the history command to return to the first screen and choose the UNIX Shell option. Then, use the disklabel command to resize your disk partitions.

5.14.2    Selecting Optional Software Using the Graphical User Interface

When making software subset selections with the graphical user interface, you can click on an individual software subset or a software subset category (such as Reference Pages). Then, click on the Add button to add the software subset or software subset category to the list of selected software to install. Double-clicking on an individual subset or subset category has the same effect.

It may not be wise to click on the Add All button because you will select hardware-specific software subsets that your system does not need. However, an alternative to selecting one software subset (or category) at a time is to Add All subsets in one step and then double-click on the subsets you do not want. Adding all subsets results in more software than your system needs, so remember to remove all of the hardware-specific subsets related to keyboard types, Xservers, and fonts. The user interface will not let you remove software subsets that are mandatory for your system's hardware configuration.

If you want to remove (or deselect) optional software selections, position the cursor in the Selected Software window and double click on the software subset or software subset category you want to remove. Another way is to click on the software subset or software subset category and then click on the Remove button to remove the software subset or software subset category from the list. The Remove button is enabled only if the selected software subset or software subset category is removable.

When you are finished selecting optional software, click on OK to return to the Installation Setup window. Click on Setup Done to start the installation procedure. Then, click on OK to verify that you want the installation to begin.

5.14.2.1    If File Systems Are Full After Selecting Optional Software

As you are selecting optional software subsets using the graphical interface, free disk space is calculated automatically and is shown at the bottom of the window. Review this information periodically to make sure your file systems are not full. You cannot proceed if a file system is full. If the file systems are at or near capacity, you have a few options:

• To free up disk space, remove (or deselect) optional software subsets in the Selected Software window.

• Click on Remove All to remove all selected optional software. Then, only mandatory software will remain in the Selected Software window.

• Return to the Installation Setup window and click on the Partition Disks... pushbutton to resize the disk partitions.

• Return to the Installation Setup window and select a different, larger partition on the same or different disk or put file systems and swap space on separate disks.

5.15    Verifying the Start of the Installation Procedure

Regardless of whether you chose a default or custom installation, you must indicate that you are ready to begin the installation. Up until this point, except for disk label changes (if any), your system is not changed in any way. This is the last chance you have to verify your disk, partition, and software selections.

• If you are using the text based interface, press the Return key at the following prompt to start the installation:

  You have now answered all questions needed to install
  Digital UNIX on this system.  Press CTRL/C to cancel the
  installation; or type "history" to modify your earlier
  answers; or press RETURN to proceed with installation:
  

• If you are using the graphical interface, click on OK in the confirmation dialog box to begin the installation.

5.16    File System Creation

When the installation procedure starts, the root, /usr, and /var file systems and swap areas are created on the disks and partitions you selected for a custom installation, or are created on the default disk layout for a default installation.

The screen display looks similar to the following.

*** Creating the root file system on device rz1a ***

*** Creating the usr file system on device rz1g ***

*** Creating the swap1 file system on device rz1b ***

5.17    Loading Software Subsets

Software subsets are loaded after file systems are created. Even though disk space was checked during software selection, disk capacity is checked again before actual software subset loading begins. The software load display shows you how many software subsets will be installed and includes an incremental counter to show progress as the load proceeds. The software load display looks similar to the following:

Checking file system space required to install specified subsets:

 

File system space checked OK.

 

*** Loading the operating system software subsets ***

 

The installation procedure will now load the software on your
disk partitions.  This process will take from 45 to 120 minutes
to complete depending on your distribution media and processor
type.

 

Loading 1 of 28 subset(s)....

 

Base System
   Copying from system9 (inet)
	Working....Fri Mar 15 13:21:30 EDT 1996
   Verifying
	Working....Fri Mar 15 13:23:31 EDT 1996

 

Loading 2 of 28 subset(s)....

 

Base System - Hardware Support
   Copying from system9 (inet)
	Working....Fri Mar 15 13:24:18 EDT 1996
   Verifying

 

Loading 3 of 28 subset(s)....

 

Compiler Back End
   Copying from system9 (inet)
	Working....Fri Mar 15 13:24:59 EDT 1996
   Verifying

 


.
.
.


 

Loading 26 of 28 subset(s)....

 

Graphical Base System Management Utilities
   Copying from system9 (inet)
   Verifying

 

Loading 27 of 28 subset(s)....

 

Graphical System Administration Utilities
   Copying from system9 (inet)
   Verifying

 

Loading 28 of 28 subset(s)....

 

Graphical Print Configuration Application
   Copying from system9 (inet)
   Verifying

 

28 of 28 subset(s) installed successfully.

5.18    Rebooting the System

When you began the installation process, you booted either from the CD-ROM or over the network. Now, after software subsets are loaded, the installation procedure requires that you boot the system off the newly installed disks. The screen displays the boot_osflags variable, the bootdef_dev variable and the boot command that you must enter to reboot your system. At the console prompt (>>>), enter the boot command sequence shown on your screen. The boot device you use depends upon your processor type and the installation media you are using. Do not enter the boot variables that are shown in the example.

Note

If your system has a graphics device on its ISA bus that requires a kernel device driver, you must modify the isacfg entry to match the kernel device driver before rebooting the system. When you issue this command, let the input line wrap; do not press the Return key in the middle of the line.

>>>  isacfg -mod -slot  slot_number -dev  device_number
    -handle  vendor_handle  -etyp 1 -enadev 1

In the previous example, replace vendor_handle with the handle supplied in the vendor's installation documentation.

If you performed a RIS installation from a RIS area that already has a kernel device graphics device driver installed, and you already set the handle to the one specified in the vendor's installation documentation, you do not have to reset the handle at this time.

Your screen will look similar to the following:

Issue the following console commands to set your default bootpath
variable and to boot your system disk to multiuser:

 

        >>> set boot_osflags A
        >>> set bootdef_dev DKA0
        >>> boot

 

syncing disks... done
CPU 0: Halting... (transferring to monitor)

 

?05 HLT INSTR
  PC= FFFFFC00.0044CA90 PSL= 00000000.00000005

Enter the boot commands at the console mode prompt (>>>) as instructed:

>>>  set boot_osflags A [Return]
BOOT_OSFLAGS = A
 
>>>  set bootdef_dev DKA0 [Return]
BOOTDEF_DEV = DKA0

 
>>>  boot [Return]

Write down the boot commands here in case you need them again: >>> >>> >>>

Software configuration begins after the system boots. Section 5.19.2 provides samples of system configuration screens. A kernel build procedure begins after software configuration.

5.19    Software Configuration

Software configuration occurs automatically and refers to the process of tailoring the software subsets, setting the host name, root password, date and time, and time zone, system tuning, and building a kernel for use by the operating system and by your hardware.

5.19.1    Configuring Kernel Device Driver Graphics Kits

During the software configuration phase, if the installation procedure detects a graphics card not supported in the Digital UNIX base operating system, you are prompted to insert the media containing the graphics driver as shown in the following prompt:

This system requires a driver for its graphics card.

 

Enter the device where the kit can be found (e.g. rz4c),
or <return> to cancel:

• If the graphics driver is located on a floppy disk, insert the floppy disk into the drive and enter the floppy device location, for example, fd0c, and press Return.

• If the graphics driver is located on a CD-ROM, remove the Digital UNIX CD-ROM from the CD-ROM drive, and insert the CD-ROM containing the graphics driver. Then, enter the location of the CD-ROM device, for example, rz4c, and press Return.

When you enter the device where the graphics driver is located and press Return, the setld command displays a menu for each of the software kits on the distribution media. When you select the appropriate graphics driver, the graphics driver is copied to the system. Software subset configuration begins next.

5.19.2    Configuring Base Operating System Software Subsets

The name of each software subset is displayed as it is being configured. Your output depends upon the software subsets you chose to install and your processor type. The display is similar to the following:

** SYSTEM CONFIGURATION ***

 

Configuring "Base System " (OSFBASE400)

 

Configuring "Base System - Hardware Support " (OSFHWBASE400)

 

Configuring "Compiler Back End " (OSFCMPLRS400)

 


.
.
.


 

Configuring "Graphical Base System Management Utilities"
(OSFXSYSMAN400)

 

Configuring "Graphical System Administration Utilities"
(OSFXADMIN400)

 

Configuring "Graphical Print Configuration Application"
(OSFXPRINT400)

If you did not provide certain essential site-specific information (such as a root password, your system's host name, the date and time, and location and time zone) earlier in the installation procedure, you will be prompted to enter that information now.

What happens after software configuration completes depends on whether you performed a default, custom, or cloned installation. Proceed to Section 5.20 for more information.

5.20    Building the Kernel

If you performed a default installation, the kernel is built automatically with the mandatory kernel parameters for your system configuration. None of the options shown in Section 5.20.1 will be included in the kernel. After the kernel build, continue with Section 5.21, which shows you how to log in to your system for the first time. If you want to build a kernel with selected options after the default installation, refer to the doconfig(8) reference page.

If you performed a custom installation or invoked the update installation with the -i option, go to Section 5.20.1 to select kernel options.

If you performed a cloned installation, how the kernel build occurs is defined in the configuration description file (CDF). If the CDF was originally created during a default installation, the kernel is built automatically. If the CDF was originally created during a custom installation, you have the opportunity to select kernel options.

5.20.1    Selecting Kernel Options

The kernel options you see on the Kernel Option Selection menu depend on the software subsets that were installed. The installation (or update installation) of certain base operating system software subsets contain a kernel component, and the installation procedure gives you the option to include or exclude the use of the software in the kernel.

If you do not select the kernel option for the product, you will not be able to use that product. For example, if you do not select the Logical Storage Manager (LSM) kernel option, even though you installed the LSM software subsets, you will not be able to use LSM. When you select a kernel option, additional code to support the option is loaded into the kernel. Therefore, selecting All of the above kernel options significantly increases the size of the kernel.

The following kernel subsystems are mandatory on all systems except systems with less than 32 MB of memory:

• The /proc File System - required for System V Environment and used by debuggers

• Quotas - UNIX File System (UFS) file quotas

• Serial Line Interface Protocol (SLIP)

• Network File System (NFS) Server

• STREAMS protocol

If your system has less than 32 MB of memory, the kernel subsystems available for your system have been optimized and the selection of certain optional kernel subsystems has been disabled. If you choose any of these kernel options for systems with less than 32 MB system, you may negatively impact system performance.

The Kernel Option Selection menu has a Help option that displays online help about each kernel option.

The Kernel Option Selection menu is similar to the following:

*** KERNEL CONFIGURATION AND BUILD PROCEDURE ***

 

*** KERNEL OPTION SELECTION ***

    Selection   Kernel Option
--------------------------------------------------------------

 

        1       LAN Emulation over ATM (LANE)
        2       Classical IP over ATM (ATMIP)
        3       ATM UNI 3.0/3.1 Signalling for SVCs
        4       Asynchronous Transfer Mode (ATM)
        5       Advanced File System (ADVFS)
        6       System V Devices
        7       Kernel Breakpoint Debugger (KDEBUG)
        8       NTP V3 Kernel Phase Lock Loop (NTP_TIME)
        9       Packetfilter driver (PACKETFILTER)
        10      Point-to-Point Protocol (PPP)
        11      STREAMS pckt module (PCKT)
        12      X/Open Transport Interface (XTISO, TIMOD, TIRDWR)
        13      File on File File System (FFM)
        14      ISO 9660 Compact Disc File System (CDFS)
        15      Audit Subsystem
        16      ACL Subsystem
        17      Logical Storage Manager (LSM)
        18      All of the above
        19      None of the above
        20      Help
        21      Display all options again

 

--------------------------------------------------------------

The following is a description of each kernel option:

LAN Emulation over ATM (LANE)

This option provides the ATM Forum standard LAN Emulation module. This should be configured when interoperating with other LANE end-systems or routers. This option is not required for Classical IP over ATM.

Classical IP over ATM (ATMIP)

This option provides IETF standard IP over ATM (RFC 1577). This module should be configured when interoperating with other end-systems and routers that also use Classical IP. This option is not required in configurations that communicate using only LANE.

ATM UNI 3.0/3.1 Signalling for SVCs

This option provides ATM Forum standard UNI 3.0 and UNI 3.1 signaling protocols. This option should be configured when connected to an ATM network that supports SVCs via UNI signaling. This module does not need to be configured when using PVCs only.

In general, UNI signaling and one or both of Classical IP over ATM or LANE will be configured. At least Classical IP over ATM or LANE is required to use TCP/IP over ATM.

Asynchronous Transfer Mode (ATM)

The ATM option installs the kernel software and applications necessary to use an ATM network. ATM technology is a connection-oriented wide area/local area technology based on the high-speed switching of 53-byte cells across a network. ATM can traverse microwave, copper, and fiber and is not limited to one kind of data. The ATM software subsets are installed automatically if the installation process detects ATM hardware. The Digital UNIX ATM subsystem supports the ATM Forum User-Networking (UNI) Version 3.0 specification, including ILMI for registration of a single address, UNI signaling for point-to-point connections, and QOS class 0 with best effort delivery.

System V Devices

The System V Devices option includes kernel options for devices required for the System V environment product. These devices are /dev/prf (System V kernel profiler) and FFM_FS (File On File File System).

Advanced File System (AdvFS)

The Advanced File System is a log-based, local file system that allows modification and expansion of file systems by mounting on different devices or adding devices to that file system.

If you chose AdvFS as the file system type for root, /usr, or /var, the subset is mandatory will not be displayed in the menu; AdvFS will be configured automatically.

NTP V3 Kernel Phase Lock Loop (NTP_TIME)

The NTP_TIME kernel option enables the kernel phase lock loop (PLL) time adjusting algorithm described by RFC 1589, for use with the NTP V3 daemon.

The NTP V3 daemon can be used without the NTP_TIME kernel option. If the NTP_TIME kernel option is configured, a new system call is available to xntpd that uses a PLL algorithm in the kernel for improved accuracy when adjusting the system clock frequency. A detailed description of the PLL algorithm can be found in RFC 1589. Refer to Network Administration for more information about NTP.

Kernel Breakpoint Debugger (KDEBUG)

The Kernel Breakpoint Debugger loads the kernel debugger kdebug and provides physical memory space for debugging custom kernels.

/proc Application Debugger

The /proc file system enables running processes to be accessed and manipulated as files by the system calls open, close, read, write, lseek, and ioctl. While the /proc file system is most useful for debuggers, it enables any process with the correct permissions to control another running process. Thus, a parent/child relationship does not have to exist between a debugger and the process being debugged.

Packetfilter driver

The packetfilter is a software interface that allows an application to send and receive packets directly to or from a local area network (Ethernet or FDDI). The packetfilter provides flexible demultiplexing (filtering) of incoming packets, so that many such applications may run simultaneously.

The Digital UNIX packetfilter supports two filtering models: the original CMU/Stanford model, as supported in ULTRIX, and the BSD Packet Filter (BPF), which provides more flexible and efficient filtering. (BPF was developed by the University of California, Lawrence Berkeley Laboratory.) Several public domain applications that use the packetfilter are integrated in Digital UNIX, including rarpd, tcpdump, tcpslice, nfswatch, and nfslogsum. Refer to the packetfilter(7) reference page for more information.

Point-to-Point Protocol (PPP)

The Point-to-Point Protocol (PPP) supports an asynchronous serial line or a pseudo-device terminal so that users can transfer files or NFS mount file systems across phone lines. Applications such as telnet, ftp, ping, Worldwide Web browsers, or any X program can be run over the IP network facility of the PPP data link layer. PPP is more configurable and robust than SLIP.

Refer to Network Administration for more information about PPP.

STREAMS pckt module (PCKT)

The STREAMS pckt module provides emulation for the SVR4 pty packet module.

Data Link Bridge (DPLI V2.0 Service Class 1)

Provides a DLPI V2.0 (Service Class 1) interface to BSD IFNET based network interfaces that allows STREAMS based protocol stacks to utilize BSD IFNET based network interfaces.

This kernel option is mandatory for LAT (Local Area Transport). If you installed LAT, Data Link Bridge does not appear as a kernel option.

Serial Line Interface Protocol (SLIP)

Interface protocol support for a serial line so that users can transfer files or NFS-mount file systems across phone lines.

This functionality is mandatory on systems with greater than 24 MB of memory.

Refer to Network Administration for more information about SLIP.

Quota

UFS disk quotas provide users with the ability to establish a limit on the number of blocks and inodes (or files) that a user or a group of users can allocate.

This functionality is mandatory on systems with greater than 24 MB of memory. Disk quotas for AdvFS are configured into the kernel by default, on systems in which AdvFS is installed.

STREAMS protocol (STREAMS, STRKINFO, LDTTY, RPTY)

The STREAMS framework provides an alternative to sockets. The STREAMS interface was developed by AT&T and consists of system calls, kernel routines, and kernel utilities that are used to implement everything from networking protocol suites to device drivers. Applications in user space access the kernel portions of the STREAMS framework using system calls such as open, close, putmsg, getmsg, and ioctl.

This functionality is mandatory on systems with greater than 24 MB of memory. The STREAMS framework is required by many of the personal computer protocols and DECnet.

X/Open Transport Interface (XTISO)

The X/Open Transport Interface (XTISO) defines an application interface that is independent of any transport provider. Programs written to XTI can be run over a variety of transport providers, such as Transmission Control Protocol (TCP) or User Datagram Protocol (UDP). The application specifies which transport provider to use.

Selecting this kernel option allows DECnet/OSI to load its kernel modules dynamically; you do not have to rebuild the kernel and reboot.

File on File File System (FFM)

File on File (FFM) is a file system that permits mounting a regular, character, or block special file on top of a regular file to support the STREAMS subsystem. This feature is used mainly by SVR4 compatible system calls. It can be viewed as a dynamic version of named pipes. The contents of the covered file are still available to any process that had the file open at the time of the FFM mount. FFM is required for the System V Environment.

ISO 9660 Compact Disc File System (CDFS)

CDFS provides the ability to mount CD-ROMs formated to the ISO 9660 standard or the High Sierra Group (HSG) format.

Audit Subsystem

The Audit Subsystem provides a security audit system configurable through the audit_setup command. The subset provides additional security on system features such as passwords, account ownership, and remote access.

ACL Subsystem

Loading the optional ACL kernel subsystem causes the system to enforce any access control lists (ACLs) that may be present on files. ACLs can be set and displayed with or without having the ACL subsystem loaded. An ACL is an extension to the traditional permission bits that normally control access to a file. ACLs allow additional users, groups, or both to be specified along with the permission bits that should apply for a matching request. Refer to the acl(4), getacl(1), and setacl(1) reference pages for more information.

Simple Network Management Protocol (SNMP)

The Simple Network Management Protocol (SNMP) is the Internet standard protocol for exchanging network management information. The SNMP agent provides a local or remote network manager with system information, network interface data, address resolution information (ARP), information about the routing layer (IP and ICMP), and information about the transport layer (TCP and UDP). The operating system includes an SNMP agent that allows a host to be managed by a network manager.

Local Area Transport Support

The LAT protocol provides an efficient means of logically connecting terminal servers to one or more nodes on the same local area network (LAN).

LAT software has the features required for a host to function as a service node so that requests for connections can be made by server users. LAT also permits host applications to initiate connections to the server's ports, designated as applications ports, to access remote devices such as printers. Digital UNIX supports 1500 logins using LAT.

If you chose the LAT subset when you made your subset selections, the option is not displayed in the kernel option menu and is configured automatically. If you add the LAT subset after the installation, you must reconfigure the kernel using the doconfig command.

Refer to Network Administration for more information about LAT.

Logical Storage Manager (LSM)

The Logical Storage Manager is an integrated, host-based disk storage management tool that protects against data loss and improves disk input/output (I/O) performance. Basic LSM functionality includes disk spanning and concatenation. System administrators use LSM to perform disk management functions without disrupting users or applications accessing data on those disks.

NFS Server

The Network File System Server option allows systems to export various file systems to other nodes.

This functionality is mandatory on systems with greater than 24 MB of memory.

Refer to Network Administration for more information about NFS.

After entering your choice of kernel options, the system displays a list of options you selected and asks you to verify your choice. For example:

You selected the following kernel options:

 

        Asynchronous Transfer Mode (ATM)
        System V Devices
        Logical Volume Manager (LVM)
        Kernel Breakpoint Debugger (KDEBUG)
        Packetfilter driver (PACKETFILTER)
        STREAMS pckt module (PCKT)
        Data Link Bridge (DLPI V2.0 Service Class 1)
        X/Open Transport Interface (XTISO, TIMOD, TIRDWR)
        File on File File System (FFM)
        ISO 9660 Compact Disc File System (CDFS)
        Audit Subsystem
        Local Area Transport Support
        Logical Storage Manager (LSM)

 

Is that correct? (y/n) [y]:

If the displayed kernel options are the ones you want, enter y or press Return to accept the default response. If the displayed kernel options are not the ones you want, enter n at the prompt. The Kernel Option Selection menu is redisplayed for you to select kernel options again.

5.20.2    Editing the Configuration File

This section applies to custom installations or to update installations that were invoked with the -i option.

After you select kernel options, you have the option to edit the configuration file. The configuration file is a text file that defines the components built into the kernel; it is located in the /usr/sys/conf/ SYSTEM_NAME file.

You may want to edit the configuration file to add devices, tune parameters, enable realtime preemption, or add third party layered product support. You may also want to recreate any customizations you may have made in a previous version of this file. Section 5.20.2.1 describes how to edit the configuration file to enable realtime preemption.

The Command and Shell User's Guide contains a tutorial that teaches you how to use the ed text editor (as well as the vi text editor). Refer to System Administration for information about the contents of the configuration file and the entries that you may want to edit.

You will see a prompt similar to the following:

Do you want to edit the configuration file? (y/n) [n]:

If you enter n or press Return (to enter the default response) to skip the edit of the configuration file, the kernel build process begins as shown in Section 5.20.3.

If you enter y to edit the configuration file, the following message displays:

Using 'ed' to edit the configuration file.  Press return
when ready, or type 'quit' to skip the editing session:

Enter the word quit if you want to skip the editing session and start the kernel build process or press the Return key if you want to edit the configuration file.

The following example shows an editing session using the ed text editor where the value of maxusers is changed to a value of 64:

Using ed to edit the configuration file.  Press return  when ready, or type quit to skip the editing  session:  [Return]
1907                           [1]
/maxuser                       [2]
maxusers          32           [3]
s/32/64                        [4]
maxusers          64           [5]
w                              [6]
1907                           [7]
q                              [8]

1. The number of lines in the configuration file. [Return to example]

2. Entering /maxuser searches for the line containing the word maxusers. [Return to example]

3. The line containing the word maxusers displays. [Return to example]

4. Entering /s/32/64 substitutes the value 32 for the value 64. [Return to example]

5. When you press the Return key, the line is redisplayed with the changed value. [Return to example]

6. Entering w writes (or saves) the change you made. [Return to example]

7. The total number of lines in the file is displayed, 1907. Because new lines were not added to the file, the number of lines displayed is the same as when the editing session began. [Return to example]

8. Entering q quits (or exits) the editing session. [Return to example]

The kernel build begins when you write and quit the editing session.

5.20.2.1    Enabling Realtime Preemption

The Digital UNIX kernel provides options to enhance the performance of realtime applications conforming to POSIX 1003.1b-1993 (formerly 1003.4 Draft 14). The realtime kernel makes it possible for the operating system to guarantee that an application has access to resources in a timely and predictable manner.

The realtime kernel that supports kernel preemption was previously installed as an option during base system installation. Now, the POSIX 1003.lb portions are included in the kernel automatically and a separate kernel is not needed. Preemption capabilities are disabled by default, but can be selected and enabled when the kernel is configured and built.

To enable realtime preemption in the kernel during the installation procedure, edit the configuration file and set the rt-preempt-opt parameter equal to 1 as shown in the following example:

rt-preempt-opt=1

5.20.3    Kernel Build Messages

When the subsets are configured and the configuration file is completed, the installation procedure invokes the doconfig utility to automatically make the device special files needed by the hardware and build the kernel for your system. Messages similar to the following are displayed:

       The system will now automatically build a kernel
       and then reboot. This will take approximately 15
       minutes, depending on the processor type.

 

       When  the login prompt appears after the system
       has rebooted, use 'root' as the  login name and
       the SUPERUSER  password that was entered during
       this procedure, to log into the system.

 


 

*** PERFORMING KERNEL BUILD ***
        Working....Fri Mar 15 15:45:24 EST 1996
        Working....Fri Mar 15 15:47:24 EST 1996
        Working....Fri Mar 15 15:49:25 EST 1996
        Working....Fri Mar 15 15:51:26 EST 1996

Your system may boot to single-user mode if the boot_osflags variable was not set as described in Section 5.18. The system reboots using the new kernel when doconfig completes. To bring the system to multiuser mode, press Ctrl/d at the root prompt (#). You are prompted to enter the run level. Four run levels are available:

• 0 specifies the halt state

• S or s specifies single-user mode

• 2 specifies multiuser mode without network services

• 3 specifies multiuser mode with network services

Another method to bring the system to multiuser mode is to use the following syntax for the init command and specify one of the run levels shown in the previous list:

init [ 0 | s | S | 2 | 3 ]

5.21    Logging in to the System for the First Time

Logging in to a system means typing in a user name and password to gain access into the operating system. If the user name and password match an account name on the system, the user is permitted access to that account. On newly-installed systems, the only user name recognized by the system is the user root. After installation is complete, the system administrator sets up an account for each user. Chapter 6 describes how to set up the system for general use.

What happens when you log in for the first time depends upon whether you have a graphics workstation or a text-based terminal without graphics capabilities:

• If you have a graphics workstation, a Common Desktop Environment (CDE) login window is displayed. Enter root as the user name and then enter the root password you specified during the installation procedure. The root password is sometimes called the superuser password. When you correctly enter this information, the following displays:

  • A Help Viewer window, Introducing the Desktop. This online help describes basic desktop skills and how to activate online help for the desktop.

  • A dialog box labeled Action Required. This dialog box provides information about reading the new /.dtprofile file that describes how to modify your .login or .profile files to interact correctly with the new Common Desktop Environment (CDE).

  • The File Manager application displays the contents of the root (/) directory.

  • A SysMan Configuration Checklist contains a list of the configuration applications that you should run to set up your system for general use. Section 6.1 provides more information about the checklist.

  • The CDE Front Panel displays at the bottom of the screen. This Front Panel has replaced the DECwindows environment Session Manager.

    Refer to the CDE Companion guide for an introduction to CDE and for information about migrating from DECwindows Motif to CDE.

• If you do not have a system console with graphics capabilities, at the login: prompt, enter root as the login name and at the password: prompt, enter the root password you specified during the installation procedure.

Go to Chapter 6 for information about setting up your system for general use.