Minor Cron Concurrency Clarification

[jonahgeorge]

Based on #690, I'm under the impression it's entirely safe to run multiple workers with cron enabled. You would just suffer a slightly higher resource utilization. Is this actually the case?

If so, would you accept a minor README update to clarify this? I originally assumed this but doubted myself when I saw the snippet below without any other explanation:

# Enable cron in this process; e.g. only run on the first Heroku worker process
config.good_job.enable_cron = ENV['DYNO'] == 'worker.1' # or `true` or via $GOOD_JOB_ENABLE_CRON

Thanks a ton for your work on this project!

[bensheldon]

@jonahgeorge yep! it is safe to run multiple workers with cron enabled....by default, as long job records are preserved because a unique index is used on the cron key+time in the database and if the record is deleted and there is enough clock-drift another process could re-enqueue with the same key+time. But that is probably a rare risk.

That snippet could use some explanation, or you could remove it entirely.

[jonahgeorge]

Awesome, thanks!