Question

我有一个Rails 5应用程序，它利用延迟作业从外部API提取数据并将其保存在数据库存储中，

但是，当我从API发出要进行播种的请求时，它在同一个种子请求中排队两次，而不是处理一次。

初步观察，我确实看到两条记录的处理程序与另一条记录完全不同。下面发布的是我尝试过的代码片段。

需要了解为什么会发生这种情况。寻求任何帮助将不胜感激。预先感谢。

控制器代码

class MovieController < ApplicationController

  def new_movie_seed_request
    @languages = Language.all.by_lang_name
    @movie = Movie.new
  end

  def seed_request_for_movie
    @movie_title = params["mov_title"]
    @movie_language = params["mov_language"]
    if params.present? && params["mov_title"].present? && params["mov_language"].present?
      Delayed::Job.enqueue SeedMovieViaTmdbJob.new(params["mov_title"], params["mov_language"])
      flash.now[:success] = 'Request for seeding has been submitted successfully'
    else
      flash[:danger] = 'Please enter valid movie title and movie language in order to proceed with the seeding request'
      redirect_to action: "new_movie_seed_request"
    end
  end

end

将请求排队后的延迟的工作表记录。

INSERT INTO "delayed_jobs" ("handler", "run_at", "queue", "created_at", "updated_at") VALUES ($1, $2, $3, $4, $5) RETURNING "id"  [["handler", "--- !ruby/object:ActiveJob::QueueAdapters::DelayedJobAdapter::JobWrapper\njob_data:\n  job_class: SeedMovieViaTmdbJob\n  job_id: b21ae8e7-65ec-4960-b697-d6f056c275da\n  provider_job_id: \n  queue_name: default\n  priority: 0\n  arguments:\n  - Gully Boy\n  - hi\n  executions: 0\n  locale: en\n"], ["run_at", "2019-03-22 09:20:41.852410"], ["queue", "default"], ["created_at", "2019-03-22 09:20:41.852566"], ["updated_at", "2019-03-22 09:20:41.852566"]]

INSERT INTO "delayed_jobs" ("handler", "run_at", "queue", "created_at", "updated_at") VALUES ($1, $2, $3, $4, $5) RETURNING "id"  [["handler", "--- !ruby/object:SeedMovieViaTmdbJob\narguments:\n- Gully Boy\n- hi\njob_id: b21ae8e7-65ec-4960-b697-d6f056c275da\nqueue_name: default\npriority: \nexecutions: 0\n"], ["run_at", "2019-03-22 09:20:41.877060"], ["queue", "default"], ["created_at", "2019-03-22 09:20:41.877122"], ["updated_at", "2019-03-22 09:20:41.877122"]]

请注意已创建的两个记录的处理程序属性中的更改。

Answer 1

啊哈！

挂钩：https://github.com/collectiveidea/delayed_job#hooks

Delayed :: Job有一个入队钩子，当入队时，该钩子在Job类上调用enqueue方法。如果您的工作恰好是ActiveJob，这将触发ActiveJob :: Base类定义的入队方法，该方法实际上再次将工作排队！（这次是主动工作）

解决方案：如果要通过调用Delayed :: Job.enqueue创建要排队的自定义作业，请不要从ActiveJob继承！只需使用带有perform方法的普通旧Ruby对象即可。（或者，您可以使用不执行任何操作的方法来覆盖enqueue方法，但实际上根本不需要从ActiveJob继承）

延迟的工作会将每个请求两次插入后端？

1 个答案: