Module: RocketJob::Plugins::Retry

Extended by:
ActiveSupport::Concern
Defined in:
lib/rocket_job/plugins/retry.rb

Overview

Automatically retry the job on failure.

The following retry algorithm is used to automatically retry a job when it fails: Failed jobs are aborted so that they cannot be restarted since a new instance has there are workers available to run. For example if workers are busy working on higher priority jobs, then the job will only run once those jobs have completed, or their priority lowered. Additionally, while the job is queued no additional instances will be enqueued, even if the next cron interval has been reached.

Note:

  • After failure the job is scheduled to run again in the future.

  • The job will not be retried if:

    • The job has expired.

    • The job fails validations.

    • The number of retry counts has been exceeded.

  • To see the number of times a job has failed so far:

    job.failure_count

Example:

class MyJob < RocketJob::Job

include RocketJob::Plugins::Retry

# Set the default retry_count
self.max_retries = 3

def perform
  puts "DONE"
end

end

# Queue the job for processing using the default cron_schedule specified above MyJob.create!

# Replace the default retry_count MyCronJob.create!(max_retries: 10)

# Disable retries for this job instance MyCronJob.create!(max_retries: 0)

Instance Method Summary collapse

Instance Method Details

#rocket_job_retry_on_fail? ⇒ Boolean

Returns [true|false] whether this job will be automatically retried on failure

Returns:

  • (Boolean) 


61
62
63
 
# File 'lib/rocket_job/plugins/retry.rb', line 61

def rocket_job_retry_on_fail?
  failure_count > max_retries
end