Question: What exactly are stalled jobs and what happens to running jobs when consumer/ worker crashes?

[mai1x9]

Hello there, can anyone please explain what exactly the stalled job in Bull mean and when does a job becomes stalled.

As per documentation, it says any job which hangs cpu will be regarded as stalled job.

My question is,

• On what factors bull decide its a stalled job. Is it based on event loop latency or delay?? Document says event loop is busy for several seconds but how many seconds.

• Secondly, when a queue is paused for 60 seconds, the job is becoming stalled. Why does job go into stalled when the entire queue is paused from processing jobs.

• thirdly, when a running job suddenly crashes (worker crashes let's say due to mem limit etc..), will the job becomes failed job or stalled Job?

• What configurations to be set in order to tell bull to retry the stalled jobs. I see there's limit for number of stalled jobs, does it means all jobs within the max allowed limit are retried automatically and the stalled jobs exceeding max allowed limit must be manually added into queue for reprocessing or should be permanently removed or canceled?

• Is it wise to cancel the stalled job instead of reprocessing it.