test

Trusted Solaris Installation and Configuration

Using pfinstall to Test Profiles

When install_type initial_install is defined in a profile, you can use the pfinstall command to test the profile without actually installing the Trusted Solaris software on a workstation. pfinstall shows the results of how a workstation would be installed according to the specified profile, before you actually perform a custom JumpStart installation.

Ways to Use pfinstall

pfinstall(1M) enables you to test a profile against:

How to Use pfinstall to Test a Profile

The procedure to use pfinstall to test a profile involves testing the command in the JumpStart directory.

  1. On an installed and configured Trusted Solaris workstation, log in as a user who can assume the role root.

  2. As root at label admin_low, launch a terminal and see that the pfinstall(1M) command is available in the role's profile shell.


    # clist | grep pfinstall
    Note -

    The name profile shell refers to a shell that recognizes Trusted Solaris execution profiles. It does not refer to the machine profiles being tested here.

  3. To test the profile with a specific system memory size, set SYS_MEMSIZE to the specific memory size in Mbytes: 


    # SYS_MEMSIZE=memory_size
# export SYS_MEMSIZE

  4. Change the directory to the JumpStart directory where the profile resides:


    $ cd jumpstart_dir_path

    For example, the following command would change the directory to the jumpstart directory on the root file system.


    $ cd /jumpstart

  5. Run the pfinstall -d or pfinstall -D command to test the profile:


    $ /usr/sbin/install.d/pfinstall -D | -d disk_config [-c path] profile
    Caution - Caution -

    Without the -d or -D option, pfinstall will install the Trusted Solaris software on the workstation by using the specified profile, and the data on the workstation will be overwritten.

    In this command,

    -D

    Tells pfinstall to use the current workstation's disk configuration to test the profile against. You must be in the role root to execute pfinstall with the -D option.

    -d disk_config

    Tells pfinstall to use a disk configuration file, disk_config, to test the profile against.

    -c path

    Is the path to the Trusted Solaris CD. This is required if the Trusted Solaris CD is not mounted on /cdrom. (For example, use this option if you copied the Trusted Solaris CD image to disk or mounted the Trusted Solaris CD on a directory other than /cdrom).

    profile

    The name of the profile to test.

    Note -

    You should run pfinstall on a workstation running the same version of Trusted Solaris software that will be installed by the profile.

    Run pfinstall from the directory where the profile and disk_config files reside (which should be the JumpStart directory). If the profile or disk_config file is not in the directory where pfinstall is run, you must specify the path.

  6. Check to see if the results of pfinstall are as you expected. If not, change the profile and go to Step Step 5.

    You have completed testing the profile. To perform a custom JumpStart installation on a workstation, see Chapter 3, Installing a Workstation.

pfinstall Examples

Below are some examples of using pfinstall(1M) to test the basic_prof profile against the 104_test disk configuration file: 

/usr/sbin/install.d/pfinstall -D -basic_prof

/usr/sbin/install.d/pfinstall -d 104_test -basic_prof

/usr/sbin/install.d/pfinstall -D -c /export/install/ts7_sparc -basic_prof

SPARC: How to Create a Disk Configuration File for a SPARC System

A disk configuration file is a file that represents a structure of a disk (for example, bytes/sector, flags, slices). Disk configuration files enable you to use pfinstall from a single workstation to test profiles on different sized disks.

Overview - The procedure to create a disk configuration file for a SPARC workstation involves:

  1. Locate a workstation with a disk that you want to test a profile against.

  2. Log on as a user who can assume the role root.

  3. As root at label admin_low, launch a terminal and determine the device name for the workstation's disk.

  4. Redirect the output of prtvtoc to create the disk configuration file: 


    $ prtvtoc /dev/rdsk/device_name > disk_config

    In this command,

    /dev/rdsk/device_name

    Is the device name of the workstation's disk. device_name must be in the form cwtxdys2 or cwdys2.

    Note -

    Slice 2 must be specified in device_name.

    disk_config

    Is the disk configuration file name.

  5. Copy the disk configuration file to the JumpStart directory: 


    $ cp disk_config jumpstart_dir_path

    You have completed creating a disk configuration file.

    The following example creates a disk configuration file, 104_test, on a workstation with a 104-Mbyte disk, whose device name is c0t3d0s2. 


    $ prtvtoc /dev/rdsk/c0t3d0s2 > 104_test

    In this example, the 104_test file contains the following information:


    # cat 104_test

* /dev/rdsk/c0t3d0s2 partition map

*

* Dimensions:

*     512 bytes/sector

*      35 sectors/track

*       6 tracks/cylinder

*     210 sectors/cylinder

*    1019 cylinders

*     974 accessible cylinders

*

* Flags:

*   1: unmountable

*  10: read-only

*

*                          First     Sector    Last

* Partition  Tag  Flags    Sector     Count    Sector  Mount Directory

       0      2    00          0     16170     16169

       1      3    00      16170     28140     44309

       2      5    00          0    204540    204539

       6      4    01      44310    160230    204539

SPARC: How to Create a Multiple Disk Configuration File for a SPARC System

If you need to test a profile on multiple disks, you can concatenate disk configuration files together to create multiple disk configuration scenarios.

Overview - The procedure to create a multiple disk configuration file for a SPARC workstation involves:

  1. Concatenate the 104_test file with itself and save the output to another file:


    $ cat 104_test 104_test > dual_104_test

  2. Edit the disk configuration file so that each disk device name has a different target number.

    For example, the dual_104_test file is shown as follows: 


    # cat dual_104_test

* /dev/rdsk/c0t3d0s2 partition map

*

* Dimensions:

*     512 bytes/sector

*      35 sectors/track

*       6 tracks/cylinder

*     210 sectors/cylinder

*    1019 cylinders

*     974 accessible cylinders

*

* Flags:

*   1: unmountable

*  10: read-only

*

*                          First     Sector    Last

* Partition  Tag  Flags    Sector     Count    Sector  Mount Directory

0      2    00          0     16170     16169

1      3    00      16170     28140     44309

2      5    00          0    204540    204539

6      4    01      44310    160230    204539

* /dev/rdsk/c0t0d0s2 partition map

*

* Dimensions:

*     512 bytes/sector

*      35 sectors/track

*       6 tracks/cylinder

*     210 sectors/cylinder

*    1019 cylinders

*     974 accessible cylinders

*

* Flags:

*   1: unmountable

*  10: read-only

*

*                          First     Sector    Last

* Partition  Tag  Flags    Sector     Count    Sector  Mount Directory

0      2    00          0     16170     16169

1      3    00      16170     28140     44309

2      5    00          0    204540    204539

6      4    01      44310    160230    204539

    Because dual_104_test file was created by concatenating itself, the following editing was required:

    • The first disk device name was left as is.

    • The second disk device name was changed from /dev/rdsk/c0t3d0s2 to /dev/rdsk/c0t0d0s2 so it has a unique target number.

    You have completed creating a multiple disk configuration file.