0

Also, what are the best practices around managing the Resque built in "management structures" (like the ones in the title)? Should I ever be clearing them with jedis.del(String key) or something similar?

jlewkovich
  • 2,725
  • 2
  • 35
  • 49

1 Answers1

1

  1. resque:failed is the failure queue in Resque itself whereas resque:stat is the statistics layer of Resque, so without digging into the code, I'm guessing that resque:stat:failed is the statistics count of how many failures Resque has encountered.

  2. That's up for personal opinion. For me, the best way to manage them is always through the UI first, then through the console / backend if that isn't possible. I try to avoid clearing / setting keys manually using redis/jedis as I don't want to risk messing something up.

CDub
  • 13,146
  • 4
  • 51
  • 68
  • So do failed jobs just continue to pile up in `resque:failed`? If I handle them in some way, shouldn't I clear them from `resque:failed` and then decrement from `resque:stat:failed`? I'm just confused on how this data is meant to be used, whether it's something the user should be managing or if it's just there as a data bucket. – jlewkovich Dec 27 '13 at 20:37
  • I think the `resque:stat:failed` should not be decremented - could be good to know how many failures you've had. Once you fix any issues that fall into the `resque:failed` queue, yes, best practice would be to clear that queue. – CDub Dec 27 '13 at 20:53