Sidekiq (3.2.5) not respecting retry specification

[bobber205]

Got the following code

require 'sidekiq'
require 'sidekiq/api'

class MaintainerWorker
  include Sidekiq::Worker
  sidekiq_options :retry => 2
  def perform(action_data)
    action_data = JSON.parse action_data
    puts 1/0
  end
end

Couple issues

Without

require 'sidekiq/api'

it wouldn't retry jobs at all, I'm assuming without that include that sidekiq_options silently fails?
That would really tricky to figure out.
~~

While it's now retrying my job when it fails to divide by zero, it retries up 5+ times, at least as long as I've waited, not the time times I've specified. What's up?

[mperham]

1 should not be true. Please give us an error backtrace.

[seuros]

do you have any sidekiq plugin ?

[bobber205]

There was no error. Just me watching the sidekiq web interface seeing it process the job, get it sent failed and never retried.

[mperham]

Nothing in the logs?

[bobber205]

For #2
Seeing

 {"retry"=>true, "queue"=>"default", 

whenever my job retries in the console where I'm running the sidekiq process.

[bobber205]

weird. suddenly not able to repo #1.

[bobber205]

@seuros No plugins. Just installed sidekiq this morning.

[bobber205]

Going to just close this.
My bad. 👎
I was queueing using a PHP api and I didn't realize that the options for a job when queuing manually with have to be specified at time of enqueuing. Working as intended.

[mperham]

It's possible you have an old Sidekiq process running which loaded an old version of the code where that option wasn't set.

[mperham]

👍

[bobber205]

We're all really liking sidekiq so far. :D