NAME

icmpmonitor - multiple host monitoring tool

SYNOPSIS

icmpmonitor [-f cfg file] [-d] [-v] [-r]

DESCRIPTION

The icmpmonitor Using the InterNet Control Message Protocol (ICMP) "ECHO" facility, monitors several hosts, and notify admin if some of them are down.

OPTIONS

icmpmonitor accepts following options:

-d: if specified, the program will detach itself from control terminal and work in 'daemon' mode. All error messages will go to syslog.
-f <configuration file>: Name of configuration file.
-v: if specified, the program will produce more log messages describing its activity.
-r: Execute DOWN command repeatedly until host will be up again. If it is not specified, the command is excuted only once.

CONFIGURATION FILE FORMAT

You need to create the configuration file. You can find a sample one in the distribution named 'sample.cfg'. The file consists of one or more lines. Lines starting with "#" are comments and are ignored. Each line must contain following fields, separeated by spaces or tabs:

Hostname: Name of the host which will be monitored.
Ping Interval: Interval (in seconds) on which ICMP packets will be sent to the host to check it's status.
Max. Delay: Maximum delay (in seconds) after which host considered to be "down" if no response was received.
Up Command: Command which will be executed when host became accessible. This command is executed using "sh -c" and may contain shell syntax. If command contain spaces or tabs it should be quoted. If the command is executing some lenghy operation it is recommended to start it as detached process (for example using '&'), because icmpmonitor will not resume monitoring of this and other hosts until command is finished.
Down Command: Command which will be executed when host became not accessible. This command will be executed using "sh -c" and may contain shell syntax. If command contain spaces or tabs it should be quoted. If the command is executing some lenghy operation it is recommended to start it as detached process (for example using '&'), because icmpmonitor will not resume monitoring of this and other hosts until command is finished.
Startup Condition (optional): Condition of the host that should be assumed on startup. Possible values are "auto", "down", "none" and "up". The parameter is optional with "up" being the default value. The main purpose is controlling which up/down commands will be launched on startup. "auto" detects the state of the host silently on startup and doesn't launch any command first. "down" assumes the host is down and executes the up command on the first ping recieved. Similar to "up". "none" starts either up or down command on startup.

BUGS

It was reported that some systems (e.g. Solaris 2.5) do not have function vsnprintf. In such case icmpmonitor will use vsprintf instead, but this could cause buffer overflow problem.

If you find any other bugs, please follow link to bug tracking system from ICMPmonitor project home page to file them.

AUTHOR

Vadim Zaliva <lord@crocodile.org>

CONTRIBUTORS

I would like to thank following people for help in developement of this software: Alexandr S. Dubovicov <shurik@te.net.ua>, Andrey Arhipov <andy@paco.net>, Bernhard Lutzmann <belu@gmx.at>, Ingo Saitz <Salz@users.sourceforge.net>