xlhybridsOSS/healthcheck

health check

It polls backends and if they respond with HTTP 200 + an optional request body, they are marked good. Otherwise, they are marked bad. Similar to haproxy/varnish health checks.

new HealthCheck(options)

options can be an object.

Options:

• servers: An array containing servers to check.
• delay: Delay in msec between healthchecks. Defaults to 10000.
• timeout: How long in msec a healthcheck is allowed to take place. Defaults to 2000.
• failcount: Number of healthchecks good or bad in a row it takes to switch from down to up and back. Good to prevent flapping. Defaults to 2.
• send: What to send for the healthcheck. Defaults to '/'.
• expected: What to expect in the HTTP BODY, (meaning not the headers), in a correct response. If unset, just a HTTP 200 status code is required.
• https: If true, https indicates that uses https to health check. Defaults to false.
• logger: A function is invoked until every server finish in a healthcheck.

Example:

var dateformat = require('dateformat');
var HealthCheck = require("./").HealthCheck;
var Table = require('cli-table');

var instance = new HealthCheck({
    servers: [
        'localhost:3000',
        'localhost:3001'
    ],
    delay: 5000,
    timeout: 3000,
    failcount: 1,
    send: '/health.txt',
    expected: 'I_AM_ALIVE',
    https: true,
    logger: function(list) {
        var table = new Table({
            head: ['name', 'owner pid', 'action time', 'concurrent', 'since', "status", 'is down?']
        });
        var servers = Object.keys(list);

        servers.forEach(function(s) {
            var hc = list[s];
            var action_time = dateformat(hc.action_time, 'HH:MM:ss');
            var since = dateformat(hc.since, 'HH:MM:ss');
            table.push([s, hc.owner, action_time, hc.concurrent, since, hc.last_status, hc.down]);
        });
        console.log(table.toString());
    }
});

instance.status()

The status() method health status.

Example:

{
    'localhost:3000': {
        action_time: Wed Sep 10 2014 15: 28: 20 GMT + 0800(CST),
        concurrent: 1,
        down: false,
        failcount: 1,
        last_status: 'connect ECONNREFUSED',
        owner: 2300,
        since: Wed Sep 10 2014 15: 28: 20 GMT + 0800(CST)
    },
    'localhost:3001': {
        action_time: Wed Sep 10 2014 15: 28: 20 GMT + 0800(CST),
        concurrent: 1,
        down: false,
        failcount: 1,
        last_status: 'connect ECONNREFUSED',
        owner: 2300,
        since: Wed Sep 10 2014 15: 28: 20 GMT + 0800(CST)
    }
}

• owner: Worker pid processing this healthcheck.
• action_time: Date instance. Last time request was taken.
• concurrent: Number of concurrent bad or good responses.
• since: Date instance. How long this server's been concurrently bad or good.
• last_status: Status of last finished check.
• down: If true, the server is actually down.
• failcount: Number of concurrent bad responses.

instance.is_down(name)

Return true if the given server has failed its healthcheck.

Example:

instance.is_down("localhost:3000");