Module Hydromon_netplex

module Hydromon_netplex: sig .. end

Hydromon: Monitor ICE services by pinging them constantly

type config = {

	`state_file : string;`
	`shm_name : string;`
	`shm_size : int;`
	`ping_period : float;`
	`ping_timeout : float;`

}

Explanations see below, at hydromon_factory

val hydromon_processor : config -> Netplex_types.processor

Start the Hydromon processor for this configuration

val hydromon_factory : name:string -> unit -> Netplex_types.processor_factory

Reads the config from the config file, and starts the processor

The config must look like

         processor {
             type = "hydromon";
             state_file = "<filename>";
             shm_name = "<shmname>";
             shm_size = <size>;
             ping_period = <seconds>;
             ping_timeout = <seconds>;
         }
         workload_manager {
             type = "constant";
             jobs = 1
         }

The state_file is used to save the state, so when the service is restarted it can continue where it was stopped. It is legal to delete this file before startup, but of course nothing is watched then, and the clients may get confused.

The shm_name is a special name like "/hydromon.shm". It is used for efficient communication with the clients. The name is the name of a POSIX shared memory segment.

The shm_size is the size of the shared memory segment in bytes. It is also the maximum number of services that can be watched at the same time. A small number like 256.

The monitored services are pinged every ping_period seconds, and if there is no reply after ping_timeout seconds the ping is counted as failure.

This web site is published by Informatikbüro Gerd Stolpmann

Plasma	GitLab	Archive
Projects	Blog	Knowledge