SkipWhenPreviousJobIsRunningAttribute.cs

SkipWhenPreviousJobIsRunningAttribute.cs

// Zero-Clause BSD (more permissive than MIT, doesn't require copyright notice)
// 
// Permission to use, copy, modify, and/or distribute this software for any purpose
// with or without fee is hereby granted.
// 
// THE SOFTWARE IS PROVIDED "AS IS" AND THE AUTHOR DISCLAIMS ALL WARRANTIES
// WITH REGARD TO THIS SOFTWARE INCLUDING ALL IMPLIED WARRANTIES OF MERCHANTABILITY
// AND FITNESS. IN NO EVENT SHALL THE AUTHOR BE LIABLE FOR ANY SPECIAL, DIRECT,
// INDIRECT, OR CONSEQUENTIAL DAMAGES OR ANY DAMAGES WHATSOEVER RESULTING FROM LOSS
// OF USE, DATA OR PROFITS, WHETHER IN AN ACTION OF CONTRACT, NEGLIGENCE OR OTHER
// TORTIOUS ACTION, ARISING OUT OF OR IN CONNECTION WITH THE USE OR PERFORMANCE OF
// THIS SOFTWARE.

// Hangfire.Core 1.8+ is required, for previous versions please see revision from year 2022.

using System;
using System.Collections.Generic;
using Hangfire.Client;
using Hangfire.Common;
using Hangfire.States;
using Hangfire.Storage;

namespace ConsoleApp28
{
    public class SkipWhenPreviousJobIsRunningAttribute : JobFilterAttribute, IClientFilter, IApplyStateFilter
    {
        public void OnCreating(CreatingContext context)
        {
            // We can't handle old storages
            if (!(context.Connection is JobStorageConnection connection)) return;

            // We should run this filter only for background jobs based on 
            // recurring ones
            if (!context.Parameters.TryGetValue("RecurringJobId", out var parameter)) return;

            var recurringJobId = parameter as string;

            // RecurringJobId is malformed. This should not happen, but anyway.
            if (String.IsNullOrWhiteSpace(recurringJobId)) return;

            var running = connection.GetValueFromHash($"recurring-job:{recurringJobId}", "Running");
            if ("yes".Equals(running, StringComparison.OrdinalIgnoreCase))
            {
                context.Canceled = true;
            }
        }

        public void OnCreated(CreatedContext filterContext)
        {
        }

        public void OnStateApplied(ApplyStateContext context, IWriteOnlyTransaction transaction)
        {
            if (context.NewState is EnqueuedState)
            {
                ChangeRunningState(context, "yes");
            }
            else if ((context.NewState.IsFinal && !FailedState.StateName.Equals(context.OldStateName, StringComparison.OrdinalIgnoreCase)) || 
                     (context.NewState is FailedState))
            {
                ChangeRunningState(context, "no");
            }
        }

        public void OnStateUnapplied(ApplyStateContext context, IWriteOnlyTransaction transaction)
        {
        }

        private static void ChangeRunningState(ApplyStateContext context, string state)
        {
            // We can't handle old storages
            if (!(context.Connection is JobStorageConnection connection)) return;

            // Obtaining a recurring job identifier
            var recurringJobId = context.GetJobParameter<string>("RecurringJobId", allowStale: true);
            if (String.IsNullOrWhiteSpace(recurringJobId)) return;

            if (context.Storage.HasFeature(JobStorageFeatures.Transaction.AcquireDistributedLock))
            {
                // Acquire a lock in newer storages to avoid race conditions
                ((JobStorageTransaction)context.Transaction).AcquireDistributedLock(
                    $"lock:recurring-job:{recurringJobId}",
                    TimeSpan.FromSeconds(5));
            }

            // Checking whether recurring job exists
            var recurringJob = connection.GetValueFromHash($"recurring-job:{recurringJobId}", "Job");
            if (String.IsNullOrEmpty(recurringJob)) return;

            // Changing the running state
            context.Transaction.SetRangeInHash(
                $"recurring-job:{recurringJobId}",
                new[] { new KeyValuePair<string, string>("Running", state) });
        }
    }
}

Hi,
Thanks for sharing this helpful filter!
I've run into a critical issue when using it in a real-world setup:

✅ Context:

• ASP.NET Core 8 Web API
• Hangfire 1.8.20
• SQL Server storage
• Running recurring jobs
• Using SkipWhenPreviousJobIsRunningAttribute from this gist

🐞 Issue:
If the app is restarted while a job is running, the "Running" flag remains "yes" in SQL Server. This causes:

• All future executions of the job to be skipped permanently
• No reset of the "Running" state
• No automatic recovery

🎯 Expected Behavior:
Jobs should run at their scheduled times.
If one is already running (due to load or long processing), the next run can be skipped — but execution should resume normally on the next schedule, even after a crash or restart.

🧪 Test Case:
A recurring job scheduled every 1 minute
Contains Task.Delay(TimeSpan.FromMinutes(5), cancellationToken)
Decorated with:

[SkipWhenPreviousJobIsRunning]
[DisableConcurrentExecution(timeoutInSeconds: 0)]
[AutomaticRetry(Attempts = 0, LogEvents = true, OnAttemptsExceeded = AttemptsExceededAction.Fail)]

When I restart the app while the job is running, the "Running" flag stays "yes", and no future runs occur.
The root of the issue is this logic in the filter (slightly modified, but same effect):

var running = connection.GetValueFromHash($"recurring-job:{recurringJobId}", "Running");
if (!string.IsNullOrEmpty(running) && running.Equals("yes", StringComparison.OrdinalIgnoreCase))
{
    context.Canceled = true;
}

🙋 Question:
Is there a better way to implement this so it’s resilient to crashes and app restarts?
I'm new to Hangfire and would appreciate any guidance on how to safely handle this scenario.

Thanks.