PiBox.Plugins.Jobs.Hangfire 1.0.54

PiBox.Plugins.Jobs.Hangfire

PiBox.Plugins.Jobs.Hangfire is a plugin that allows PiBox based services to use the hangfire package to manage and schedule (cron) jobs.

Installation

To install the nuget package follow these steps:

dotnet add package PiBox.Plugins.Jobs.Hangfire

or add as package reference to your .csproj

<PackageReference Include="PiBox.Plugins.Jobs.Hangfire" Version="" />

Appsettings.yml

Configure your appsettings.yml with these properties

For example

hangfire:
  Host: localhost
  Port: 5432
  Database: postgres
  User: postgres
  Password: postgres
  InMemory: true
  enableJobsByFeatureManagementConfig: false
  allowedDashboardHost: localhost # you need to set this configuration to be able to access the dashboard from the specified host

featureManagement: # we can conveniently can use the microsoft feature management system to enable jobs based on configuration
  hangfireTestJob: true # if you have enabled the 'enableJobsByFeatureManagementConfig: true' then you can configure here if your jobs should run on execution or not, useful for multiple environments etc.

HangfireConfiguration.cs

[Configuration("hangfire")]
public class HangfireConfiguration
{
    public string? Host { get; set; }
    public int Port { get; set; }
    public string? Database { get; set; }
    public string? User { get; set; }
    public string? Password { get; set; }
    public bool InMemory { get; set; }
    public string AllowedDashboardHost { get; set; }
    public bool EnableJobsByFeatureManagementConfig { get; set; }
    public int? PollingIntervalInMs { get; set; }
    public int? WorkerCount { get; set; }
    public string ConnectionString => $"Host={Host};Port={Port};Database={Database};Username={User};Password={Password};";
}

Configuration in your plugin host

public class SamplePluginHost : IPluginServiceConfiguration
{
    public void ConfigureServices(IServiceCollection serviceCollection)
    {
        // here you can register your recurring jobs or just annotate them with the [RecurringJob] attribute
        serviceCollection.ConfigureJobs((jobRegister, _) =>
            {
                jobRegister.RegisterParameterizedRecurringAsyncJob<HangfireParameterizedTestJob, int>(Cron.Daily(), 1); // every day at 0:00 UTC
                jobRegister.RegisterRecurringAsyncJob<HangfireTestJob>(Cron.Hourly(15)); // every hour to minute 15 UTC
            });
    }
}

Usage

Please take a look at the official Hangfire documentation https://docs.hangfire.io/en/latest/ and best practises https://docs.hangfire.io/en/latest/best-practices.html

Every job does out of the box 10 retries (with increasing wait times) if not configured otherwise.

Auto registration of jobs

You can also define recurring jobs as simple classes and annotate them with the [ReccuringJob] attribute which then auto-registers the job for you at startup

[RecurringJob("0 0 * * *")]
public class HangfireTestJob : AsyncJob
{
    public HangfireTestJob(ILogger<HangfireTestJob> logger) : base(logger)
    {
    }

    protected override Task<object> ExecuteJobAsync(CancellationToken cancellationToken)
    {
        Logger.LogInformation("TEst");
        return Task.FromResult<object>(cancellationToken);
    }
}

Define jobs

Without parameters

public class HangfireTestJob : AsyncJob
{
    public HangfireTestJob(ILogger<HangfireTestJob> logger) : base(logger)
    {
    }

    protected override Task<object> ExecuteJobAsync(CancellationToken cancellationToken)
    {
        Logger.LogInformation("TEst");
        return Task.FromResult<object>(cancellationToken);
    }
}

With parameters

Job input parameters and results get serialized to json and are displayed in the hangfire dashboard, be aware of the possibility of leaking sensitive information!

public class MyResult
{
    public int JobParameter {get;set;}
}
public class HangfireParameterizedTestJob : ParameterizedAsyncJob<int>
{
    public HangfireParameterizedTestJob(ILogger<HangfireParameterizedTestJob> logger) : base(logger)
    {
    }

    protected override Task<MyResult> ExecuteJobAsync(int value, CancellationToken cancellationToken)
    {
        var result = new MyResult(){ JobParameter = value};
        Logger.LogInformation("Test job for {Value}", value);
        return result; //
    }
}

Fire and forget jobs or enqueue dynamically

You can create new jobs from anywhere in your service, see also https://docs.hangfire.io/en/latest/background-methods/index.html

BackgroundJob.Enqueue<IEmailSender>(x => x.Send("hangfire@example.com"));
BackgroundJob.Enqueue(() => Console.WriteLine("Hello, world!"));

Attributes

UniquePerQueueAttribute

If you want the job only to be executed as one instance at any given point in time use the

[UniquePerQueueAttribute("high")]

This ensures that there is only one job of the same type/name and method parameters in processing or queued at any given point

JobCleanupExpirationTimeAttribute

With this you can specify how many days the results of a job should be kept until it gets deleted

[JobCleanupExpirationTimeAttribute(14)]

Filters

LogJobExecutionFilter

This filter logs the start and finish of an job execution.

EnabledByFeatureFilter

This filter works in conjunction with the microsoft feature management system. If you would like to be able to enable or disable the execution of your jobs based on configuration this is the right tool for it.

Default Feature management with file based configuration

hangfire:
  enableJobsByFeatureManagementConfig: true

featureManagement:
  hangfireTestJob: true
  neverRunThisJob: false

This allows you to enable jobs based on configuration files. If you have enabled the setting

enableJobsByFeatureManagementConfig: true

then you can configure here, if your jobs should run on execution or not, useful for multiple environments etc.

If your service supports hot reloading of configuration files, you can enable/disable jobs at run time.

Feature management with the pibox unleash plugin

This works in conjunction with the plugin PiBox.Plugins.Management.Unleash. This replaces the ability of setting the features via files. Instead one can use the unleash api/service and use feature flags for enabling the jobs. Just make sure that the name of the job matches the name of the feature flag you are creating in unleash.

The pibox unleash plugin then should do the rest of the heavy lifting.

Since the attribute resolves the feature on before executing the job, changes to the configuration can be done at runtime with a maximal delay based on how often the pibox unleash plugin refreshes its cache. You can find more information in the documentation of the pibox unleash plugin.