hwmgr(8) System Manager's Manual hwmgr(8)
NAME
hwmgr - Command line interface to hardware device data
SYNOPSIS
/sbin/hwmgr command [subsystem] [parameters]
Brief Description
The hwmgr utility works with the KSM hardware set and the kernel hardware management module. It enables you to manage hardware components
and the hardware sub-systems that maintain information about hardware components. A hardware component can be a storage peripheral, such as
a disk or tape, or a system component such as a CPU or a bus. Use the hwmgr utility to manage hardware components on either a single system
or on a cluster. The default file permission is -rwxr-xr-x and you must be root to use the hwmgr utility.
The hwmgr utility provides two types of commands, internal and generic. Internal commands do not contain a subsystem identifier on the com-
mand line. Generic commands are characterized by a subsystem identifier after the command name.
Internal Command Synopsis
Use the following syntax for internal /sbin/hwmgr command options. Options can be abbreviated to the minimum unique string, such as -m
instead of -member. /sbin/hwmgr -flash light [-dsf dev-special-filename] [-bus scsi-bus | -target scsi-target | -lun scsi-lun] [-seconds
N-seconds] [-nopause]
/sbin/hwmgr -get attribute [saved | default | current] [-a attribute...] [-a attribute=value] [-a attribute!=value] [-id hardware-compo-
nent-id] [-category hardware-category] [-member cluster-member-name] [-cluster]
/sbin/hwmgr -get category
/sbin/hwmgr -help [command | subsystem]
/sbin/hwmgr -set attribute [saved | current] -a attribute=value [-a attribute=value]... [-id hardware-component-id] [-member cluster-mem-
ber-name] [-cluster]
/sbin/hwmgr -view cluster
/sbin/hwmgr -view devices [-dsf device-special-filename] [-category hardware-category] [-member cluster-member-name] [-cluster]
/sbin/hwmgr -view hierarchy [-id hardware-component-id [-instance component-instance]] [-member cluster-member-name]
/sbin/hwmgr -view env
/sbin/hwmgr -view timestamps [-reverse] [-a timestamp-attribute...] [-a timestamp-attribute=time] [-id hardware-component-id] [-category
hardware-category] [-member cluster-member-name] [-cluster]
/sbin/hwmgr -view transaction [-member cluster-member-name] [-cluster]
Generic Commands
Generic commands are characterized by a subsystem identifier after the command name. The current subsystems are: component, scsi and name.
The component subsystem maintains information on all hardware devices specified in the binary /etc/dec_hwc_ldb and /etc/dec_hwc_cdb data-
bases. These databases contain information on all hardware components currently registered with hardware management. This includes most
devices on a system. The name subsystem maintains information on all hardware components in the binary /etc/dec_hw_db database, often
referred to as the hardware topology. The database contains hardware persistence information, maintained by the kernel driver framework and
includes data for buses, controllers and devices. The scsi subsystem maintains information on all SCSI devices in the binary
/etc/dec_scsi_db database. The SCSI database contains entries for all devices managed by the SCSI/CAM architecture.
Note that some generic hwmgr commands are duplicated in more than one subsystem. You should use the subsystem most closely associated with
the type of operation you want to perform. The current list of generic commands is: -add -delete -edit -locate -redirect -refresh -reload
-scan -show -unconfigure -unload
Generic Command Syntax
Usage for generic commands is unique for each commmand and subsystem combination. Use the help command to view usage for all commands sup-
ported under a subsystem: /sbin/hwmgr -help [component | name | scsi]
FLAGS
The following flags are used with many of the commands. Other parameters are unique to the individual commands and are defined with the
commands. Perform the operation on every member of the cluster. This parameter is valid only when the system is a cluster member. Per-
form the operation on the specified cluster member. This parameter is valid only when the system is a cluster member. Display extended
details where they are provided. Provide detailed information about the return status of this command. Note that the -verbose switch must
be spelled out completely, and cannot be abbreviated. To make the verbose operation default, define the environment variable HWMGR_VERBOSE
= TRUE.
COMMAND OPTIONS
The following sections describe the flags and parameters for each command set in turn. Note that you can abbreviate command words to the
minimum number of unique letters, such as -m for -member.
Internal Commands
Displays information on the state of each cluster member. This command is only useful in a cluster environment. Displays the internal
Hardware Manager (hwmgr) environment variable settings, which you can set in your environment configuration file such as Use this variable
to set the location of the main hwmgr data file. Use this variable to set debugging on or off. Use this variable to set numbers to hexa-
decimal. Causes the display output to be truncated at 80 columns rather than wrapped. Use this variable to set hwmgr to display verbose
output by default. This option can also be specified using the -verbose flag on the command line. Displays event information such as the
registration time of any devices, and the host on which it occurred. Displays information on the transactions that the hardware management
kernel code uses to perform some requests, including the status of: Any ongoing hardware management transactions Whether the hardware man-
agement cluster database is locked The last hardware management transaction that was completed. Use the devices option to display all
devices on the system. All devices on the local host will be returned by default, unless one of the following parameters are supplied:
Specifies a hardware category, such as disk or tape on which the operation should be performed. Specifies the host name of a cluster mem-
ber on which the operation should be performed. Specifies that the operation should be performed cluster-wide. If this option is not
specified, only data for the local host will be returned.
Note that for the purpose of this command, a device is considered to be any hardware component that has the dev_base_ name
attribute. Displays the current hardware component hierarchy. If no cluster member name is specified this operation defaults to the
local system. Displays the hierarchy from the specified component downwards. Specify a hardware identifier (HWID). Specifies an
instance of a hardware component. Some devices may be in the component hierarchy in more than one place. For example, SCSI disks on
a shared bus would be seen in the hierarchy under two buses. Each entry for the component in the hierarchy is considered a differ-
ent instance. Specifies the host name of a cluster member on which the operation should be performed.
The display hierarchy command always uses the hardware set. It will display the currently registered hardware components which have
been placed in the system hierarchy. Use this option to flash the display light (LED) on a SCSI disk device for a default time
period of 30 seconds. Note that this operation may not work on all devices, and you may have to open a cover to access the light on
some systems, particularly where the disk is installed in an internal bay. Specifies a device by its special filename such as
/dev/disk/dsk4. Specifies the device(s) on the given SCSI bus number. Specifies the device(s) at the given SCSI target number.
Specifies the device at the given logical unit number. Specifies a duration of N seconds for the light to flash. The default is 30.
Specifies no pauses between flashes. Returns attribute values for a device. You can specify the device attributes to return,
according to their type and one or more optional matching parameters. The value of an attribute can be: This is the value of the
attribute that has been configured and stored in the database using the -set saved command option. When you set the saved value of
an attribute, you are setting the value to be saved across boots and used on subsequent boots. This is the default value of an
unconfigured attribute. When you add a new device and boot the system, the device attributes of the new device will be at their
default values. This is the value of the attribute that the driver is currently using in its operation of the device. If you set
an attribute using -set current, the saved value is unchanged. When you shut down and reboot the system, the value of the attribute
reverts to the saved value in the database. If you want the value you are using to persist across boots, you must use the -set saved
command option. When using the -get command, the current values are returned by default.
Note that not all attributes can always have a current, saved, or default value. Attribute values may be assigned to a device by the
system at startup, so that the saved value shows 0, but the current value may be different. Most attributes have read-only values
and are not settable. Typically, you are only able to set a few attributes for a given device.
Settable attributes are marked (settable) when you use the (-get attributes) option. The following truncated display shows an exam-
ple: # /sbin/hwmgr -get attributes
path_fail_limit = 5
device_starvation_time = 25 (settable)
cluster_disables = 0
For each attribute status (saved, default, current) you can specify the following optional parameters: Use the -a option to return
values of an individual attribute, such as path_fail_limit, which is a SCSI disk attribute defining the limit for path failures.
Use the -a option with the =value parameter to return attributes that match the specified value. For example, to search for devices
that support power management, where the saved value of power management is enabled (1), use the following command: # hwmgr -get
attribute -a power_mgmt_capable=1 Use the -a option with the !=value parameter to return attributes that do not match the specified
value. # hwmgr -get attribute -a power_mgmt_capable!=1 Returns the attribute values for the specified hardware device identifier
(HWID). For example, the following command returns all the current attribute values for hardware component 18: # hwmgr -get
attribute current -id 18 Specifies a hardware category, such as bus or ide_adapter on which the operation should be performed. You
can display the available categories using the -get category command option. Specifies the host name of a cluster member on which
the operation should be performed. Specifies that the operation should be performed cluster-wide. If this option is not specified,
only data for the local host will be returned. Returns a list of all Hardware Component Categories available on the system, such as
platform, scsi_bus, or disk. Displays the complete online help screen for the hwmgr utility. You can optionally add the following
parameters to restrict the output to more specific information: Displays help only for the specified command. Displays help only
for the specified subsystem, which can be component, name or scsi. Sets either the saved (permanent) or current (in use) value for
the specified attribute(s). Refer to the -get flag for more information. Note that default attribute values cannot be set.
The following parameters can be specified: At least one attribute and value must be specified, you can specify other attributes to
set as required, subject to limitations on the command input buffer. Specifies a hardware component identifier (HWID) on which the
operation is to be performed. Specifies the host name of a cluster member on which the operation is to be performed. Specifies
that the operation is to be performed cluster-wide. If the system is part of a cluster and this option is not specified, the opera-
tion will only be performed on the local host.
Note that the values of the attributes of a particular hardware component may not be the same for each member of the cluster. One
hardware component may have different attributes, or different values for the same attributes, for each separate cluster member.
Warning
You should specify a hardware identifier when setting an attribute. Failure to do so will set the attribute value on all hardware
components.
The following sections lists the command options for generic commands, which can target a subsystem. There are three supported subsystems:
component, name and scsi. Syntax for each subsystem is described in turn.
The component Subsystem
The component subsystem references all hardware devices specified in the binary /etc/dec_hwc_ldb and /etc/dec_hwc_cdb databases. These
databases contain information on all hardware components currently registered with hardware management. This includes most devices on a
system.
The component subsystem supports the following command options: Deletes a hardware component from all members of the cluster, and removes
all occurrences of the hardware component from all system databases. A hardware component identifier must be supplied. Finds a hardware
component. The locate operation is designed to help you physically find a piece of hardware. How this is actually implemented depends on
the hardware being located.
If the locate operation is directed to a SCSI disk component, the disk will attempt to flash the activity light. This is the same
as using the command: # hwmgr -flash light
The following optional flags can be specified for locate: Specifies the device special filename, such as /dev/disk/dsk1. Specifies
the bus, target, lun (n/n/n/) address. Specifies a duration of N seconds for the light to flash. The default is 30 seconds. Speci-
fies no pause between flashes. Note that this operation works only on SCSI disks and may not work on some devices. This command
will delete (from every database) all hardware components which are not currently registered. Use this command only if the system or
cluster is set up in an optimal configuration and you want to remove obsolete hardware entries from all databases.
Any devices that are not plugged in and have not been seen during this boot session will be removed. Scans the system for new hard-
ware components and if a new component is found, configure it. If you do not specify any arguments, the command will scan all hard-
ware on the system. Note that the -scan component option is asynchronous. When you issue the command it completes immediately,
although the scan can still be working in the kernel. To test for completion of a scan, use the Event Manager (EVM) evmwatch and
evmshow commands to monitor for an event with the following event name: sys.unix.hw.scan_completed. For example: # evmwatch -f
'[name sys.unix.hw.scan_completed]' | evmshow A hardware scan has just completed.
Alternatively, use the graphical EVM Viewer interface to monitor completion of the scan. (See the EVM(5) reference page for an
introduction to EVM.)
The following flags can be specified with -scan: Specifies the hardware component identifier (HWID) for the device on which the
-scan operation should be performed. Specifies the hardware category, such as disk or tape, on which the -scan operation should be
performed. See also the -get category option. Performs a recursive scan downward in the system topology. If this option is not
specified, the scan operation will not scan recursively to hardware components below the starting component, unless recursion occurs
as part of the normal scanning process for the component. A bus, for example, would always be scanned recursively (since that is
how the scan code for a bus is written). A scan of a platform (CPU) would not normally be recursive. Specifies an instance of a
hardware component on which the scan should be performed Some devices may be in the component hierarchy in more than one place. For
example, SCSI disks on a shared bus would be seen in the hierarchy under two buses. Each entry for the component in the hierarchy
is considered a different instance. Specifies the host name of a cluster member on which the operation should be performed. (See
the RESTRICTIONS section.) Display hardware components from the hardware component databases. This includes all hardware compo-
nents, including hardware components that were previously registered but may not be currently registered.
The -show component command returns a FLAGS field as part of the output. The FLAGS field is a series of characters that give infor-
mation on the state of a hardware component. The following characters are used in the FLAGS output for show: The component is cur-
rently registered with hardware management. The component has device nodes associated with it. The component has a cluster wide
unique name or has cluster-wide available device special files. The component has saved attributes associated with it in the hard-
ware set. A WARNING field that this component has an inconsistency in the hardware component database.
The following flags can be specified with -show to control the output: Shows all hardware components that are currently registered
with hardware management. Shows all hardware components that are NOT currently registered with hardware management. These compo-
nents were previously registered at one time. Shows all hardware components that are cluster shared. A cluster shared hardware
component is a hardware component that has a name which is guaranteed to be unique anywhere in the cluster, or has cluster-wide
available device special files. Shows all hardware components that are NOT cluster shared. See the -cshared flag for a definition
of cluster shared. Shows all hardware components that have device nodes associated with them. Shows all hardware components that
do NOT have device nodes associated with them. Shows all hardware components that have saved attributes. See the -get attribute
command for a description of saved attributes. Shows all hardware components that do NOT have saved attributes. See the -get
attribute command for a description of saved attributes. Shows all hardware components that have software inconsistencies in the
hardware component database. An inconsistency is a possible internal error with the component database. Use the -full switch with
the show command to view detailed information on what may be wrong with the hardware component.
Note that this command does not fix database inconsistencies; it only detects inconsistencies. One possible fix may be to reboot
the cluster. Shows all hardware components that do NOT have software inconsistencies. Shows detailed information on the hardware
component. Displays output from the show command in an alternate format. Currently the only other supported output for the show
component command is tcl. Specifies a cluster member on which the -show operation should be performed. Perform the operation on
every member of the cluster. This parameter is valid only when the system is a cluster member. Specifies the hardware component
identifier (HWID) for the device on which the -show operation should be performed.
If no member ID is specified and the -cluster option was not used, the -show operation defaults to the local cluster member. Uncon-
figures a hardware component. Unconfiguring a hardware component will unregister the component from the running system but will not
remove the persistence of the component from the databases. The following options are supported: Unconfigure database entries with
the specified hardware component identifier. Specifies an instance of a hardware component on which the operation should be per-
formed. Some devices may be in the component hierarchy in more than one place. For example, SCSI disks on a shared bus would be
seen in the hierarchy under two buses. Each entry for the component in the hierarchy is considered a different instance. Performs
a recursive operation downward in the system topology. If this flag is not specified, the operation will not recurse to hardware
components below the starting component. Perform the operation on the specified cluster member. This parameter is valid only when
the system is a cluster member.
The name Subsystem
The name subsystem references all hardware components in the /etc/dec_hw_db binary database which often referred to as the hardware topol-
ogy. The database contains hardware persistence information, maintained by the kernel driver framework, including data for controllers and
buses. To add a bus, controller or device entry to the name subsystem, use the following syntax: hwmgr -add name
Specifies the base persistence name that should be used for this persistence entry. For example scsi will be the persistence name
for a SCSI adapter. Specifies the persistence number to be associated with this hardware component. For example, the first SCSI
adapter will be 0, making the persistence entry for the adapter scsi0. Specifies the type for a hardware persistence entry. Speci-
fies the persistence name of the parent bus or controller, at which location the new entry will persist. For example, a SCSI
adapter might persist under parent bus tza. Specifies the instance number of the parent bus or controller of the hardware compo-
nent. For example, the instance number for a SCSI adapter persisting on a parent tza bus might be instance 0, for tza0. Specifies
the slot number occupied by this hardware component. For example, A SCSI adapter might persist at slot 2 of the tza0 bus (tza0 slot
2).
To add other (generic) types of entries to the name subsystem, use the following command syntax: hwmgr -add name
As for bus, controller, or device above. As for bus, controller, or device above. Specifies the type for a hardware persistence
entry. Specifies a key used to identify the persistence entry. Specifies a cluster member on which the -add operation should be
performed. (See the RESTRICTIONS section.) Deletes a hardware persistence entry. When you delete a hardware component it cannot be
used. For example, if the system has a network adapter named tu0 and you add a second network adapter in a lower slot, the new
adapter will be automatically named tu1. However, if you delete tu0 then both adapters will be automatically renamed in the order
they are discovered (when the system probes buses and controllers). The new network adapter in the lower slot becomes tu0 and the
original adapter is renamed to tu1. (See also the -remove operation.)
The following parameters should be specified: Specifies the name of the hardware persistence entry that you want to delete. Speci-
fies a cluster member on which the -delete operation should be performed. (See the RESTRICTIONS section.) This option enables you
to edit hardware persistence of a bus, controller, or device using the following command syntax: hwmgr -edit name
Specifies the name of the hardware persistence entry that you want to edit: Specifies a changed parent name for this persistence
entry. Specifies a changed parent instance number name for this persistence entry. Specifies a changed slot number name for this
persistence entry. Specifies a cluster member on which the -edit operation should be performed. (See the RESTRICTIONS section.)
To edit a generic persistence entry, use the following command syntax: hwmgr -edit name
As for bus, controller, or device above. Specifies a new value for the key for this persistence entry. Specifies a cluster member
on which the -edit operation should be performed. (See the RESTRICTIONS section.) This operation does an unload followed by a load,
and usually operates on a driver. (See the -load and -unloadoptions.) Specifies which driver should be reloaded. Specifies that
the kernel should not call the driver configuration routine when the driver is reloaded. Specifies a cluster member on which the
-reload operation should be performed. (See the RESTRICTIONS section.) This operation removes a hardware persistence entry from the
database but does not affect any hardware component currently using the name. It only affects the persistence of the name across
reboots. (See also the -delete operation.) Specifies the name of the hardware persistence entry that is to be removed. Specifies a
cluster member on which the -remove operation should be performed. (See the RESTRICTIONS section.) This operation scans a hardware
persistence entry for new devices. Specifies the name of the hardware persistence entry that is to be scanned. Specifies a cluster
member on which the -scan operation should be performed. (See the RESTRICTIONS section.) Specifies that the -scan operation is to
be performed on every cluster member. This operation displays persistence entries. Specifies a cluster member on which the -show
operation should be performed. (See the RESTRICTIONS section.) This operation unconfigures an instance of a hardware persistence
entry. Specifies the name of the hardware persistence entry that is to be unconfigured. Specifies a cluster member on which the
-unconfigure operation should be performed. (See the RESTRICTIONS section.) This operation unloads a module (usually a driver) from
memory after unconfiguring the module. Specifies the suffix for the driver that is to be unloaded from memory. For example, tu,
the driver suffix for the tulip network interface card
The scsi Subsystem
The scsi subsystem references all SCSI devices in the binary /etc/dec_scsi_db database. The SCSI database contains entries for all devices
managed by the SCSI/CAM architecture. Shows entries from the SCSI/CAM Logical ID persistence database. If no cluster member identifier is
specified and the cluster-wide option was not used, this operation defaults to the local system. The following flag options can be speci-
fied for the -show command: Show any SCSI database entries with the specified hardware component identifier. Show any SCSI database
entries with the specified SCSI device identifier. The SCSI device ID is a SCSI specific identifier used by the SCSI/CAM subsystem, it is
not related to the hardware ID used by hardware management. Show any SCSI entries with a path (valid or stale) at the specified bus. A
stale path occurs when a device that was previously seen is no longer accessible at that path. Show any SCSI entries at the specified tar-
get. Show any SCSI entries at specified lun (logical unit number). Specifies a cluster member on which the -show operation should be per-
formed. Specifies a SCSI device type, such as disk or tape. Note that SCSI device types are specified in the SCSI standards that are sup-
ported in this release of the operating system software. Supported standards and restrictions are defined in the Software Product Descrip-
tion. Specifies that only SCSI components with valid paths should be displayed. If a device can be accessed, the path is valid. Specifies
that only SCSI components with stale paths should be displayed. A path becomes stale when a device that was previously seen is no longer
accessible at that path. Specifies that if detailed information on a device is available, that data should be displayed. Use this command
to manually assign a unique name to a legacy SCSI device that does not provide such a name. If a SCSI device does not provide a unique
name, it will be seen as a different device for each path from which it can be accessed.
Use the -edit scsi command for every place from which the device is seen, assigning the same name each time. CAM will then provide
this name to the hardware manager, enabling the device to be seen as the same device no matter what access path is used. The device
will also receive a unique device special file. Assign the device special file names, cluster dev_t values, local dev_t values,
Logical ID, and hardware ID of one hardware component to another. Source and target scsi logical device identifiers (did) must be
supplied. This command will get the hardware component for the entry being deleted and pass the request to the component subsystem
handler to finish the delete. (The -member option is not supported on the SCSI -delete operation.)
If the entry is not registered in the kernel with HWC (only under unusual circumstances) the -delete option will remove the entry
from the CAM database without calling the component subsystem. The SCSI device identifier. The scan command probes the scsi sub-
system for new hardware. If a new SCSI device has been added to the system you can use the -scan scsi option to find the new hard-
ware.
The following flag options can be specified for the -scan scsi command: Specifies the device(s) on the given SCSI bus number Speci-
fies the device(s) at the given SCSI target number Specifies the device at the given logical unit number Perform the operation on
the specified cluster member. (See the RESTRICTIONS section.)
DESCRIPTION
Use this utility for managing hardware, gathering information on system status, and diagnosing device problems. The hwmgr utility has the
following features: The hwmgr utility is specific to the hardware management subsystem in the kernel, and uses only the KSM hardware set
and functions provided by the enhanced management subsystem in the kernel. Provides a wide range of hardware management functions under a
single command. Enables you to manage hardware components that are currently unregistered in the KSM hardware set. These may be hardware
components seen on a previous system boot, but not currently seen in the active configuration. Enables you to manage hardware components
in multiple system databases. Enables you to propagate a management request to multiple members of a cluster.
See the System Administration guide for more information on device names and device special files and a definitive list of the supported
device names. Note that some hwmgr commands can be run directly from the SysMan Menu, and many device properties and attributes can be
monitored from the SysMan Station.
RESTRICTIONS
The following notes and restrictions apply:
When working on a cluster, if you do not specify a member name the operation defaults to the local member. Some command options may require
that you specify a member name, but in general it is an option.
Currently the -locate component -id command is only implemented for SCSI disks, using the disk activity indicator light (LED).
The following operations on the name subsystem are not supported by all drivers: The -reload name operation. The -unconfigure name opera-
tion. The -unload name operation.
ERRORS
The command returns an int with an errorno from <errno.h>.
EXAMPLES
These examples have been reformatted for ease of reference. The actual formatted output from commands may be slightly different. If your
system is a member of a cluster, use hwmgr to send hardware management requests to any node in the cluster. To verify that Hardware Man-
ager recognizes all members in the cluster use the following command: # hwmgr -view cluster
Member ID State Member HostName
--------- ----- ---------------
1 UP rene (localhost)
31 UP witt
24 UP freu
10 UP rogr The following example displays devices on the system: # hwmgr -view devices
HWID: Device Name Model Location -----------------------------------------------------------
15: /dev/disk/floppy0c 3.5in fdi0-unit-0
18: /dev/disk/dsk0c RZ1 CB-CA bus-0-targ-0-lun-0
19: /dev/disk/cdrom0c RRD46 bus-0-targ-4-lun-0
The information shown in the output is as follows: The hardware id number assigned to the device. The device special file name for
the device, which is located in the dev directory, The vendor's model number, as shown on the device or its casing. The physical
location of the device, bus, target, and logical unit number (lun).
Note that on this system, no tape device is present and the command: # hwmgr -view devices -type tape
returns no data, while the following command displays the same output as shown in the preceding example; # hwmgr -view devices -type
disk This example shows all devices on a cluster: # hwmgr -view devices HWID: Device Name Mfg Model Host Location
--------------------------------------------------------
3: /dev/kevm jully 28: /dev/disk/floppy0c 3.5in floppy jully fdi0-unit-0 40: /dev/disk/dsk0c
DEC RZ28M (C)DEC jully bus-0-targ-0-lun-0 41: /dev/disk/dsk1c DEC RZ28L-AS (C)DEC jully bus-0-targ-1-lun-0 42: /dev/disk/dsk2c
DEC RZ28 (C)DEC jully bus-0-targ-2-lun-0 43: /dev/disk/cdrom0c DEC RRD46 (C)DEC jully bus-0-targ-6-lun-0 44: /dev/disk/dsk3c
DEC RZ28M (C)DEC phoey bus-1-targ-1-lun-0 45: /dev/disk/dsk4c DEC RZ28L-AS (C)DEC jully bus-1-targ-2-lun-0 45: /dev/disk/dsk4c
DEC RZ28L-AS (C)DEC phoey bus-1-targ-2-lun-0 46: /dev/disk/dsk5c DEC RZ29B (C)DEC jully bus-1-targ-3-lun-0 46: /dev/disk/dsk5c
DEC RZ29B (C)DEC phoey bus-1-targ-3-lun-0 47: /dev/disk/dsk6c DEC RZ28D (C)DEC jully bus-1-targ-4-lun-0 47: /dev/disk/dsk6c
DEC RZ28D (C)DEC phoey bus-1-targ-4-lun-0 48: /dev/disk/dsk7c DEC RZ28L-AS (C)DEC jully bus-1-targ-5-lun-0 48: /dev/disk/dsk7c
DEC RZ28L-AS (C)DEC phoey bus-1-targ-5-lun-0 49: /dev/disk/dsk8c DEC RZ1CF-CF (C)DEC jully bus-1-targ-8-lun-0 49: /dev/disk/dsk8c
DEC RZ1CF-CF (C)DEC phoey bus-1-targ-8-lun-0 50: /dev/disk/dsk9c DEC RZ1CB-CS (C)DEC jully bus-1-targ-9-lun-0 . . .
The following command enables you to display information on a particular device by specifying its device special file: # hwmgr -view
devices -dsf /dev/disk/cdrom0c
HWID: DSF Name Mfg Model Location
--------------------------------------------------------
31: /dev/disk/cdrom0c DEC RRD47(C)DEC bus-0-targ-4-lun-0 The following example command displays the device hierarchy on the sys-
tem: # hwmgr -view hierarchy
HWID: Hardware component hierarchy ----------------------------------
1: platform AlphaServer 800 5/500
2: cpu CPU0
4: bus pci0
5: scsi_adapter isp0
6: scsi_bus scsi0
18: disk bus-0-targ-0-lun-0 dsk0
19: disk bus-0-targ-4-lun-0 cdrom0
20: graphics_controller trio0
8: bus eisa0
9: serial_port tty00
10: serial_port tty01
11: parallel_port lp0
12: keyboard PCXAL
13: pointer PCXAS
14: fdi_controller fdi0
15: disk fdi0-unit-0 floppy0
16: network tu0
17: network tu1
You can obtain more specific output by identifying the cluster member for which you want to obtain data, or by specifying a particu-
lar device identifier. The following example displays the time that each hardware component was initially registered on a small
standalone system (truncated): # hwmgr -view timestamps
HWID: Component Attribute Name Hostname Time occurred ------------------------------------------------------------
1: AlphaServer 80 registration_time cy1.k3. <time>
2: CPU0 registration_time cy1.k3. <time>
3: kevm registration_time cy1.k3. <time>
4: pci0 registration_time cy1.k3. <time>
5: pci0slot5 registration_time cy1.k3. <time>
7: pci0slot6 registration_time cy1.k3. <time>
9: pci0slot7 registration_time cy1.k3. <time>
Note that in the above output, <time> is displayed as weekday month date HH:MM:SS year, for example: Sat May 15 14:42:18 1999. The
following command displays the devnodes from the component subsystem: # hwmgr -show comp devnodes
HWID: HOST FLAGS SERVICE COMPONENT NAME -----------------------------------------------
3: pmoba r-d-- none kevm
20: pmoba r-d-- none tty00
22: pmoba r-d-- none tty01
24: pmoba r-d-- none lp0
27: pmoba r-d-- iomap FDI-fdi0-unit-0
35: pmoba rcd-- iomap SCSI-WWID:0410004c:"DEC RZ28 0034766791"
36: pmoba rcd-- iomap SCSI-WWID:04100024:"DEC RZ25F 14295981"
37: pmoba r-d-- iomap SCSI-WWID:0710002c:"DEC RRD43 6l00000"
38: pmoba r-d-- iomap SCSI-WWID:0710002c:"DEC TLZ06 04l00000"
40: pmoba --d-i iomap SCSI-WWID:0710002c:"DEC RX26 0l00000"
42: pmoba rcd-- iomap SCSI-WWID:0410004c:"DEC RZ26L 574435"
43: pmoba rcds- iomap SCSI-WWID:0410003a:"DEC RZ26L 2191192"
This example shows any hardware components that have device nodes associated with them. If a hardware component has device nodes,
it means that it has a device special file that can be used to access the device. To see more information on the device special
file, such as the dev_t information you can use the --full option with the --show comp command, as follows: # hwmgr -show comp -id 3
-full
HWID: HOSTNAME FLAGS SERVICE COMPONENT NAME -----------------------------------------------
3: pumba r-d-- none kevm
DSF GROUP
INSTANCE GRPFLAGS GROUPID SUBSYSTEM BASENAME L1 L2
---------------------------------------------------------
0 40 1 kevm kevm pseudo kevm
DEVICE NODE
ID LB LC CB CC B C Class Suffix L3B L3C
devT devT devT devT Flags Flags
---------------------------------------------------------------
0 0 4f00000 0 0 0x0 0x865 0x0 (null) (null) (null)
2 0 4f00002 0 0 0x0 0x865 0x0 .pterm (null) (null)
1 0 4f00001 0 0 0x0 0x875 0x0 .debug (null) (null)
This example shows the device special file basename of pseudodevice kevm. It also shows the device special file group instance and
dev_t information. This example shows SCSI devices on the system: # hwmgr -show scsi
SCSI DEVICE DEVICE DRIVER NUM DEVICE FIRST HWID: DID HOSTNAME TYPE SUBTYPE OWNER PATH FILE VALID PATH
---------------------------------------------------------------
22: 0 ftiwod disk none 2 1 dsk0 [0/3/0]
23: 1 ftiwod cdrom none 0 1 cdrom0 [0/4/0]
24: 2 ftiwod disk none 0 1 dsk1 [2/2/0]
25: 3 ftiwod disk none 0 1 dsk2 [2/2/1]
29: 4 ftiwod disk none 0 1 scp2 [2/2/2]
To view all the paths to a particular SCSI device you can use the -show scsi -full option, as follows: # hwmgr -show scsi -full
SCSI DEVICE DEVICE DRIVER NUM DEVICE FIRST HWID: DID HOSTNAME TYPE SUBTYPE OWNER PATH FILE VALID PATH
----------------------------------------------------- --------- 43: 7 pmoba disk none 0 2 dsk3 [0/2/0]
WWID:0410003a:"DEC RZ26L (C) DECPCB= ; HDA=000052191192"
BUS TARGET LUN PATH STATE
------------------------------
1 2 0 stale
0 2 0 valid
The following command displays the hardware categories available on the system: # hwmgr -get category
Hardware Categories ------------------- category = undefined category = platform category = cpu category = pseudo category = bus
category = scsi_adapter category = scsi_bus category = graphics_controller category = serial_port category = parallel_port category
= keyboard category = pointer category = fdi_controller category = disk category = network The following command shows partial out-
put of all attributes for all devices on the system: # hwmgr -get attribute 1:
name = AlphaServer 800 5/500
category = platform 2:
name = CPU0
category = cpu
sub_category = EV5.6 (21164A)
power_mgmt_capable = 0
state = 1
speed = 500 <long display output truncated>
You can return specific information from specific devices, as shown in the following command example: # hwmgr -get attribute -id 4
4:
name = pci0
category = bus
# hwmgr -get attribute -id 3
More specific information can be specified by naming required attributes, as follows:
# hwmgr -get attribute -a name -id 4 4:
name = pci0
For some attributes, you can specify the saved, default, or current values. The following example shows how you set attributes that
are flagged as (settable) in the output from /sbin/hwmgr -get attributes: # hwmgr -get attr saved -id 18 -a device_starvation_time
18:
device_starvation_time = 25 # hwmgr -set attr saved -id 18 -a device_starvation_time=30 18:
saved device_starvation_time = 30
The first command displays the saved attribute named device_starvation_time for device id 18, a hard disk. The second command
increases the saved value from 25 to 30. The following example shows how you can use the redirect option to replace a failed device
(did=3) . This command assumes that: You used the -show scsi (or other) option to obtain the device identifier (did) for the failed
device. The replacement device has been physically installed according to the directions in the Owner's Manual. You used the -scan
scsi option to probe for the device and register it. You used the -show scsi option to obtain the device identifier (did) for the
new device and found it to be 53. # hwmgr -redirect scsi -src 3 -dest 53 Note that you can use the Event Manager (EVM) to monitor
your system for hardware events. For example, the following command will watch for hardware events at the terminal: # evmwatch -f
'[name sys.unix.hw]' | evmshow
Alternatively, use the graphical EVM Viewer interface to monitor such events. (See the EVM5 reference page for an introduction to
EVM.)
FILES
The executable image The data file for the HWC component subsystem The data file for the NAME component subsystem The data file for the CAM
SCSI component subsystem The main data file Location of the device special files and sub-directories The default device class sub-directo-
ries, containing device special files for the named devices The device information databases, which specify characteristics of device mod-
els
SEE ALSO
Commands: dsfmgr(8), sysman(8), sysman_station(8),
System Administration
hwmgr(8)