Skip to main content

Troubleshooting: Resque

This guide covers common troubleshooting tasks for working with Resque background jobs.

Accessing the Resque Console​

Production Failed Jobs Dashboard​

View failed jobs in production at:

https://api.pairteam.com/resque/failed

Note: You need the Arc password to access this page.

Utility Methods​

The following utility methods can be run in the Bastion Rails console to help diagnose and manage failed jobs.

Count Failed Jobs by Queue and Class​

Returns a sorted list of failed job counts grouped by queue and job class:

def failed_jobs_counts(queue_name = nil)
  failed_jobs = Resque::Failure.all(0, 20000)

  if queue_name.present?
    failed_jobs.select! { |j| j["queue"] == queue_name }
  end

  failed_jobs
    .group_by { |j| j["queue"] + "-" + j["payload"]["args"][0]["job_class"] }
    .transform_values(&:count)
    .sort_by { |_, count| -count }
end

Usage:

# Count all failed jobs
failed_jobs_counts

# Count failed jobs for a specific queue
failed_jobs_counts("tasks")

Clean Up Failed Jobs by Class​

Removes all failed jobs for a specific job class:

def failed_jobs_cleanup(job_class)
  Resque::Failure.each do |index, job|
    Resque::Failure.remove(index) if job_class == job["payload"]["args"][0]["job_class"]
  end
  nil
end

Usage:

failed_jobs_cleanup("MyJobClass")

Built-in Resque::Failure Methods​

Resque provides several built-in methods for managing failed jobs. See the Resque::Failure API documentation for the complete reference.

Requeue All Failed Jobs from a Queue​

Resque::Failure.requeue_queue("tasks")

Clear Retried Jobs​

Removes failed jobs that have already been retried:

Resque::Failure.clear_retried

Other Useful Methods​

# Get total count of failed jobs
Resque::Failure.count

# Get a specific failed job by index
Resque::Failure.all(index, 1)

# Remove a specific failed job by index
Resque::Failure.remove(index)

# Requeue a specific failed job by index
Resque::Failure.requeue(index)

# Clear all failed jobs
Resque::Failure.clear