我有几台专用服务器，在这些服务器上我的后台工作人员有时会在重新启动之前运行数天。发生这种情况时（它们似乎运行了 24 小时以上），然后我开始收到如下错误：
getaddrinfo: Temporary failure in name resolution
/home/deployer/.rbenv/versions/1.9.3-p194/lib/ruby/1.9.1/net/smtp.rb:540:in `open'
/home/deployer/.rbenv/versions/1.9.3-p194/lib/ruby/1.9.1/net/smtp.rb:540:in `tcp_socket'
/home/deployer/.rbenv/versions/1.9.3-p194/lib/ruby/1.9.1/net/smtp.rb:549:in `block in do_start'
/home/deployer/.rbenv/versions/1.9.3-p194/lib/ruby/1.9.1/timeout.rb:68:in `timeout'
/home/deployer/.rbenv/versions/1.9.3-p194/lib/ruby/1.9.1/timeout.rb:99:in `timeout'
/home/deployer/.rbenv/versions/1.9.3-p194/lib/ruby/1.9.1/net/smtp.rb:549:in `do_start'
/home/deployer/.rbenv/versions/1.9.3-p194/lib/ruby/1.9.1/net/smtp.rb:519:in `start'
来自该特定进程的所有远程调用都会发生该错误，直到它重新启动（不是硬件/服务器重新启动，只是 ruby 进程）。我已经仔细检查/确认重新启动该进程是一种修复，因为我试图通过 SSH 连接到服务器，并在失败的进程旁边运行一个新进程，这个进程可以完美运行，而有故障的进程仍然失败.
所以我的问题是，为什么会发生这种情况，除了每天重新启动流程之外，我该如何解决这个问题？
我在 ruby 1.9.3 p194 上，后台作业是用sidekiq gem 处理的（但是我不怀疑那个负责混乱）。
服务器位于http://www.hetzner.de（但我不认为他们应该受到责备，因为它正在发生在进程中）
更新
除了上述 smtp 调用错误之外，在执行简单的 http 请求时，我也会收到类似的错误。这也可以在进程重新启动后解决：
# (Errno::EHOSTUNREACH) "No route to host - connect(2)"
/home/deployer/apps/au/shared/bundle/ruby/1.9.1/gems/net-http-persistent-2.8/lib/net/http/persistent/ssl_reuse.rb:29:in `initialize'
/home/deployer/apps/au/shared/bundle/ruby/1.9.1/gems/net-http-persistent-2.8/lib/net/http/persistent/ssl_reuse.rb:29:in `open'
/home/deployer/apps/au/shared/bundle/ruby/1.9.1/gems/net-http-persistent-2.8/lib/net/http/persistent/ssl_reuse.rb:29:in `block in connect'
/home/deployer/.rbenv/versions/1.9.3-p194/lib/ruby/1.9.1/timeout.rb:54:in `timeout'
/home/deployer/.rbenv/versions/1.9.3-p194/lib/ruby/1.9.1/timeout.rb:99:in `timeout'
/home/deployer/apps/au/shared/bundle/ruby/1.9.1/gems/net-http-persistent-2.8/lib/net/http/persistent/ssl_reuse.rb:29:in `connect'
/home/deployer/.rbenv/versions/1.9.3-p194/lib/ruby/1.9.1/net/http.rb:755:in `do_start'
/home/deployer/.rbenv/versions/1.9.3-p194/lib/ruby/1.9.1/net/http.rb:750:in `start'
/home/deployer/apps/au/shared/bundle/ruby/1.9.1/gems/net-http-persistent-2.8/lib/net/http/persistent.rb:628:in `start'
/home/deployer/apps/au/shared/bundle/ruby/1.9.1/gems/net-http-persistent-2.8/lib/net/http/persistent.rb:888:in `reset'
/home/deployer/apps/au/shared/bundle/ruby/1.9.1/gems/net-http-persistent-2.8/lib/net/http/persistent.rb:567:in `connection_for'
/home/deployer/apps/au/shared/bundle/ruby/1.9.1/gems/net-http-persistent-2.8/lib/net/http/persistent.rb:926:in `request'
/home/deployer/apps/au/shared/bundle/ruby/1.9.1/gems/mechanize-2.5.1/lib/mechanize/http/agent.rb:258:in `fetch'
/home/deployer/apps/au/shared/bundle/ruby/1.9.1/gems/mechanize-2.5.1/lib/mechanize.rb:407:in `get'