7
在Mac OS X狮子使用Ruby v1.8.7和后台进程V1.1.8，我试图写一个消费过程，并把它办成一dameon：错误的文件描述符在Ruby中守护
# config[:name] => 'idx_my_delete_consumer'
# config[:daemon] => {:multiple => false,
# :backtrace => true,
# :dir_mode => :normal,
# :log_dir => '/Users/pprakash/consumer.log',
# :monitor => true,
# :dir => '/Users/pprakash/pids'}
Daemons.run_proc(config[:name], config[:daemon]) do
consumer = MyConsumer.new(config)
consumer.subscribe
end
然而，它不会启动，而是会抛出长时间的回溯，如下所示：
E, [2012-05-28T19:34:16.199770 #29357] ERROR -- : Bad file descriptor (Errno::EBADF) /opt/local/lib/ruby/gems/1.8/gems/daemons-1.1.8/lib/daemons/daemonize.rb:134:in `for_fd' /opt/local/lib/ruby/gems/1.8/gems/daemons-1.1.8/lib/daemons/daemonize.rb:134:in `close_io' /opt/local/lib/ruby/gems/1.8/gems/daemons-1.1.8/lib/daemons/daemonize.rb:134:in `initialize' /opt/local/lib/ruby/gems/1.8/gems/daemons-1.1.8/lib/daemons/daemonize.rb:134:in `new' /opt/local/lib/ruby/gems/1.8/gems/daemons-1.1.8/lib/daemons/daemonize.rb:134:in `close_io' /opt/local/lib/ruby/gems/1.8/gems/daemons-1.1.8/lib/daemons/daemonize.rb:75:in `call_as_daemon' /opt/local/lib/ruby/gems/1.8/gems/daemons-1.1.8/lib/daemons/application.rb:258:in `start_proc' /opt/local/lib/ruby/gems/1.8/gems/daemons-1.1.8/lib/daemons/application.rb:295:in `start' /opt/local/lib/ruby/gems/1.8/gems/daemons-1.1.8/lib/daemons/monitor.rb:51:in `watch' /opt/local/lib/ruby/gems/1.8/gems/daemons-1.1.8/lib/daemons/monitor.rb:51:in `fork' /opt/local/lib/ruby/gems/1.8/gems/daemons-1.1.8/lib/daemons/monitor.rb:51:in `watch' /opt/local/lib/ruby/gems/1.8/gems/daemons-1.1.8/lib/daemons/monitor.rb:45:in `each' /opt/local/lib/ruby/gems/1.8/gems/daemons-1.1.8/lib/daemons/monitor.rb:45:in `watch' /opt/local/lib/ruby/gems/1.8/gems/daemons-1.1.8/lib/daemons/monitor.rb:44:in `loop' /opt/local/lib/ruby/gems/1.8/gems/daemons-1.1.8/lib/daemons/monitor.rb:44:in `watch' /opt/local/lib/ruby/gems/1.8/gems/daemons-1.1.8/lib/daemons/monitor.rb:84:in `start_with_pidfile' /opt/local/lib/ruby/gems/1.8/gems/daemons-1.1.8/lib/daemons/monitor.rb:64:in `fork' /opt/local/lib/ruby/gems/1.8/gems/daemons-1.1.8/lib/daemons/monitor.rb:64:in `start_with_pidfile' /opt/local/lib/ruby/gems/1.8/gems/daemons-1.1.8/lib/daemons/monitor.rb:111:in `start' /opt/local/lib/ruby/gems/1.8/gems/daemons-1.1.8/lib/daemons/application_group.rb:149:in `create_monitor' /opt/local/lib/ruby/gems/1.8/gems/daemons-1.1.8/lib/daemons/application.rb:284:in `start' /opt/local/lib/ruby/gems/1.8/gems/daemons-1.1.8/lib/daemons/controller.rb:70:in `run' /opt/local/lib/ruby/gems/1.8/gems/daemons-1.1.8/lib/daemons.rb:197:in `run_proc' /opt/local/lib/ruby/gems/1.8/gems/daemons-1.1.8/lib/daemons/cmdline.rb:109:in `call' /opt/local/lib/ruby/gems/1.8/gems/daemons-1.1.8/lib/daemons/cmdline.rb:109:in `catch_exceptions' /opt/local/lib/ruby/gems/1.8/gems/daemons-1.1.8/lib/daemons.rb:196:in `run_proc' users/delete_consumer.rb:40
我不确定是什么原因导致此问题？目录名称，日志文件名都是有效的。我可以用这些配置创建MyConsumer的实例，并且能够从独立的程序/控制台正确执行#subscribe。
因此，在获得了Ruby守护进程的一些经验之后，我意识到任何这样的错误都意味着底层块（被妖魔化）有错误。修复底层块中的所有错误也会修复此错误。 –
我有同样的问题。感谢您分享您的发现！ – vpsz
为了帮助他人，请回答你自己的问题，以便它有一个正式的答案。 –