Linux and UNIX Man Pages

Linux & Unix Commands - Search Man Pages

icodd(1m) [hpux man page]

icapd(1M)																 icapd(1M)

NAME
icapd - Instant Capacity (iCAP) daemon. SYNOPSIS
icapd DESCRIPTION
The (formerly ) daemon is installed and started as part of the Instant Capacity software on all potential iCAP systems, and respawns itself if killed. If this daemon is not running, other Instant Capacity commands fail. The operations this daemon performs are vital in keeping the complexwide view of the Instant Capacity state current. The following entry is added to /etc/inittab in order to have start and respawn itself: icap:23456:respawn:/usr/lbin/icapd # Instant Capacity daemon This daemon is not started on hardware that is not supported under the Instant Capacity program. If is installed and running on a system with Instant Capacity components present (cores, cells, or memory), it sends daily asset report email to HP (if so configured), tracks tem- porary capacity, sends exception notifications, and maintains a healthy Instant Capacity state. For more information about the functions that performs for Instant Capacity systems, see the located at /usr/share/doc/icapUserGuide.pdf. The daemon reports errors via (see syslog(3C)). Exception notification email is sent to root and to the system contact email address (con- figured via the command (see icapmodify(1M)). The daemon performs periodic operations based on the time of day. The daemon is spawned by and gets its time zone specification from the /etc/default/tz file. By default, the time zone specified in /etc/default/tz is EST5EDT. You can specify which time zone the daemon uses to interpret its current time by modifying the /etc/default/tz file. For details about the time zone format, see environ(5). A restart of the daemon is required before the new time zone value takes effect (that is, kill the process). AUTHOR
was developed by HP. SEE ALSO
icapmodify(1M), icapstatus(1M), icapnotify(1M), icapmanage(1M), icap(5). icapd(1M)

Check Out this Related Man Page

intrbald(1M)															      intrbald(1M)

NAME
intrbald - daemon for balancing interrupts SYNOPSIS
interval | DESCRIPTION
Balancing interrupts is a process which distributes interrupts across the available CPUs in an optimal fashion. This action is performed by (see intctl(1M)). Performing frequent CPU migrations between system partitions may require you to perform periodic balancing of interrupts. This is espe- cially true with systems that use WLM (Workload Manager), iCAP (Instant Capacity), and vPars (Virtual Partitions) products. If the fre- quency of CPU migration is high, then balancing of interrupts should be performed more frequently. The daemon monitors the system for imbalance in interrupt distribution. If an imbalance is detected after the specified interval executes to balance the interrupt distribution. uses the parameters from the configuration file for balancing the interrupts. For more information about the command, see intctl(1M) and the configuration file. The daemon is started from the general purpose sequencer script (see rc(1M)), at run level 2 and is killed at run level 1. By default, the daemon is not enabled. The daemon configuration file must be edited and the parameter set to in order to enable running the daemon. After setting to for the first time, you can start the daemon manually by executing the script Otherwise, the daemon will automatically start the next time the system is restarted. Any actions performed by the daemon are logged in and details of interrupt migrations performed will be logged in the file Options recognizes the following options: Instructs the running daemon to reexamine the configuration file If the parameter has been updated, then the running daemon uses the new value. If the parameter is set to then the running daemon stops executing. The daemon checks if the system requires any balancing of interrupts after interval seconds; the valid range is 5 seconds to seconds (see limits(5)). This value can also be specified persistently in the configuration file by setting the parameter. The default value is 60 seconds. Setting this value too small can cause unnecessary interrupt migrations. Gracefully kills the running daemon. intrbald rc Configuration File is the configuration file. The following parameters can be persistently specified in this configuration file: Specifies if the daemon should be started automatically during the system bootup process by the script. The daemon will only be started automatically if this parameter is set to NOTE: This value is by default. Before using the daemon, needs to be changed to in the file. Then the daemon needs to be started, either manually by running or it will be started automatically at next boot via the script. The daemon checks if the system requires any interrupt balancing after this interval. The interval is specified in seconds; the valid range is 5 seconds to seconds (see lim- its(5)). The default is 60 seconds. Setting this value too small could cause unnecessary interrupt migrations. RETURN VALUE
Exit values are: Successful completion. An error condition occurred. WARNINGS
The command can be executed only by the superuser. This command should be used only by performance analysts for performance tuning pur- poses. If the interrupts are not redistributed properly, a possible decrease in overall system performance can occur. Some processors can become overloaded while other processors will not be optimally utilized. FILES
configuration file. See the section above. script. start symbolic link. kill symbolic link. daemon log file. SEE ALSO
intctl(1M), ioscan(1M), rc(1M), limits(5). intrbald(1M)
Man Page