RediBox Schedule

Allows functions to run at set times, taking into consideration multi-server environments for hassle free scheduling.

Works well with Jobs.

Installation

First ensure you have RediBox installed.

Install Schedule via npm:

npm install redibox-hook-schedule --save

Usage

Configure schedules

Within your redibox config, we'll setup a new schedule object containing a schedules array. Each array item consists of a runs function, data and an interval.

• runs [Function/String] A function or string (a globally available function as a dot notated string i.e. some.fooBar function which would resolve to global.some.fooBar automatically).

• data [Array/Object/Primitive] Any data to pass along to each schedule function.

• interval [String] A string of the interval time, compatible with Later.js.

• multi [Boolean]

  • default: false If true, the schedule will run on every server. If false it'll be locked to a single server only.

{
  schedule: {
    schedules: [
      runs: function(schedule) {
        // do something every 5 minutes
        console.log('The value of foo is: ' + schedule.data.foo);
      },
      data: {
        foo: 'bar',
      },
      interval: 'every 5 minutes',
    ],
  },
}

Accessing schedule data

If passing in a function directly (like above), the schedule is available to the runs function as the first argument, where the data can be accessed via schedule.data.

Schedule Timing

Under the hood, the schedule uses Later. If a schedule is set to run every 1 minute, it will run on the minute, every minute. This ensures uniform schedules across multiple environments. It does not run 1 minute after the server booted.

Multi-server environments

Typically a large application will deploy many servers running the same code base. As expected the schedule will run on each individual server. If you have 20 servers deployed, and a schedule runs every minute to query an external API then update your database, you don't want 20 servers doing this at once.

Luckily, by default only a single server can only run a schedule at any one time. This is handled by utilising Redis locks. Once a schedule is picked up by a server, it is locked on Redis and cannot be run again until it is unlocked (which is performed automatically).

There might be however use cases where running a scheduled task across all servers is required. In this case, simply set the multi option to true on the schedule object.

License

MIT