instl_bootd.1m (2012 03)

i
instl_bootd(1M) instl_bootd(1M)
NAME
instl_bootd - boot protocol server for Ignite-UX clients
SYNOPSIS
/opt/ignite/lbin/instl_bootd
[-s][-r reuse_time ][-t timeout ][-d debug_level ]
[
-c command ][-C command ][
-P port ][-b boot_file ][instl_boottab ][
-?]
DESCRIPTION
instl_bootd is a boot protocol daemon that responds to boot requests from clients wishing to install an
operating system using the Internet Boot Protocol (BOOTP) as defined in
RFC951 and RFC1048.
instl_bootd may respond to clients without the server’s prior knowledge of the client (unlike
bootpd
which requires a client to be registered with the server prior to a boot request).
When
instl_bootd receives a boot request, it allocates an Internet Protocol (IP) address from a list of
available addresses held in the instl_boottab
file. instl_bootd then responds to the client by
returning the IP address and the boot_file .
Options
instl_bootd recognizes the following options:
-s Allow instl_bootd to run as a stand-alone daemon. This option is used when
starting the daemon manually or via a system startup script, but not when starting
with inetd (see inetd (1M)). Without this option, instl_bootd expects an open
socket associated with the server port as its standard input. With this option,
instl_bootd opens the socket itself. Using inetd is the preferred startup
method.
-r reuse_time Specify the minimum amount of time (in minutes) that must elapse before an IP
address may be reused. The default value for reuse_time is five minutes.
reuse_time should correspond to the amount of time a client requires to perform the
initial phase of the boot process (that is, transferring the kernel into memory).
Once instl_bootd allocates an IP address to a client, it will not reallocate the
same address to another client for reuse_time minutes. This prevents IP address
collision during the client’s boot phase. instl_bootd does not respond to clients
(and clients cannot boot) if it cannot successfully allocate an IP address old enough
to satisfy the reuse_time requirement. In this case, a diagnostic message is logged
to syslogd (see syslogd (1M)). If this condition occurs frequently, adding more IP
addresses to instl_boottab may be necessary.
-t timeout Specify how long (in minutes) instl_bootd remains running after responding to
a boot request. A timeout value of zero means that instl_bootd will never exit.
The default timeout value is 15 minutes. Increasing timeout (or setting to it zero),
prevents instl_bootd from being started and stopped excessively during heavy
use. Decreasing timeout is useful if instl_bootd is rarely used, and its
resources need to be reclaimed sooner. The -t option has no meaning when used
with the -s option.
-d debug_level
Specify the detail of messages logged to syslogd (see syslogd (1M)). The default
value of zero means that only errors and warnings are logged. Values of 1, 2, or 3
cause increasingly more verbose message logging.
-c command Specify a command to be executed using system (see system (3S)) when
instl_bootd responds to a new client’s boot request. Two arguments are passed
to command: the LAN card station address and the IP address allocated to the
client.
-C command Similar to the -c option. Client systems often send multiple boot requests during
the boot process. The -c option causes instl_bootd to execute command only
the first time it responds to a boot request from a client. The -C option causes
instl_bootd to execute command every time it responds to a boot request.
-P port Override the default User Datagram Protocol (UDP) port numbers defined by ser-
vices (4) for instl_boots and instl_bootc. Default values are 1067 and
1068 respectively. This causes instl_bootd to listen and respond to boot
requests on port and port +1. port becomes the server port (instl_boots) and
HP-UX 11i Version 3: March 2012 − 1 − Hewlett-Packard Company 1