3

我有一些使用 Boost 的 SSL ASIO 库的多线程代码。该代码是多线程的，但每个 SSL 连接都有一个互斥锁，并且对async_*函数的调用是在持有互斥锁的情况下完成的。

我有时会看到我的代码停止运行。有一个卡住的线程。线程卡在bio_write. 堆栈跟踪如下所示：

#0  0x00000000010a974f in bio_write ()   

#1  0x00000000010a3529 in BIO_write ()

#2  0x000000000105fe72 in ssl3_write_pending ()

#3  0x0000000001060b02 in ssl3_write_bytes ()

#4  0x0000000000cce43a in boost::asio::ssl::detail::engine::do_write
(this=0x1e618a0, data=0x234c7b0, length=189) at
/usr/include/boost/asio/ssl/detail/impl/engine.ipp:294

#5  0x0000000000cce109 in boost::asio::ssl::detail::engine::perform(this=0x1e618a0, op=(int
(boost::asio::ssl::detail::engine::*)(boost::asio::ssl::detail::engine *
const, void *, std::size_t)) 0xcce3fc<boost::asio::ssl::detail::engine::do_write(void*, unsigned long)>,
data=0x234c7b0, length=189, ec=..., bytes_transferred=0x7fff8ad74d48) at
/usr/include/boost/asio/ssl/detail/impl/engine.ipp:219

#6  0x0000000000ccdd23 in boost::asio::ssl::detail::engine::write
(this=0x1e618a0, data=..., ec=..., bytes_transferred=@0x7fff8ad74d48: 0) at
/usr/include/boost/asio/ssl/detail/impl/engine.ipp:137

#7  0x0000000000dac504 in boost::asio::ssl::detail::write_op<boost::asio::mutable_buffers_1>::operator()
(this=0x7fff8ad74d20, eng=..., ec=..., bytes_transferred=@0x7fff8ad74d48: 0)
at /usr/include/boost/asio/ssl/detail/write_op.hpp:51

#8  0x0000000000da9d8a in boost::asio::ssl::detail::io_op<boost::asio::basic_stream_socket<boost::asio::ip::tcp,
boost::asio::stream_socket_service<boost::asio::ip::tcp> >,
boost::asio::ssl::detail::write_op<boost::asio::mutable_buffers_1>,
boost::asio::detail::write_op<boost::asio::ssl::stream<boost::asio::basic_stream_socket<boost::asio::ip::tcp,
boost::asio::stream_socket_service<boost::asio::ip::tcp> > >,
boost::asio::mutable_buffers_1, boost::asio::detail::transfer_all_t,
boost::_bi::bind_t<void, boost::_mfi::mf2<void, Peer,
boost::system::error_code const&, unsigned long>,
boost::_bi::list3<boost::_bi::value<boost::shared_ptr<Peer> >, boost::arg<1>
(*)(), boost::arg<2> (*)()> > > >::operator() (this=0x7fff8ad74d10, ec=..., 
bytes_transferred=0, start=1) at /usr/include/boost/asio/ssl/detail/io.hpp:136

#9  0x0000000000da7f42 in boost::asio::ssl::detail::async_io<boost::asio::basic_stream_socket<boost::asio::ip::tcp,
boost::asio::stream_socket_service<boost::asio::ip::tcp> >,
boost::asio::ssl::detail::write_op<boost::asio::mutable_buffers_1>,
boost::asio::detail::write_op<boost::asio::ssl::stream<boost::asio::basic_stream_socket<boost::asio::ip::tcp,
boost::asio::stream_socket_service<boost::asio::ip::tcp> > >,
boost::asio::mutable_buffers_1, boost::asio::detail::transfer_all_t,
boost::_bi::bind_t<void, boost::_mfi::mf2<void, Peer,
boost::system::error_code const&, unsigned long>,
boost::_bi::list3<boost::_bi::value<boost::shared_ptr<Peer> >, boost::arg<1> (*)(),
boost::arg<2> (*)()> > > > (next_layer=..., core=..., op=...,
handler=...) at /usr/include/boost/asio/ssl/detail/io.hpp:322

#10 0x0000000000da634d in
boost::asio::ssl::stream<boost::asio::basic_stream_socket<boost::asio::ip::tcp,
boost::asio::stream_socket_service<boost::asio::ip::tcp> >
>::async_write_some<boost::asio::mutable_buffers_1,
boost::asio::detail::write_op<boost::asio::ssl::stream<boost::asio::basic_stream_socket<boost::asio::ip::tcp,
boost::asio::stream_socket_service<boost::asio::ip::tcp> > >,
boost::asio::mutable_buffers_1, boost::asio::detail::transfer_all_t,
boost::_bi::bind_t<void, boost::_mfi::mf2<void, Peer,
boost::system::error_code const&, unsigned long>,
boost::_bi::list3<boost::_bi::value<boost::shared_ptr<Peer> >, boost::arg<1>
(*)(), boost::arg<2> (*)()> > > > (this=0x1e61880, buffers=..., handler=...)
at /usr/include/boost/asio/ssl/stream.hpp:502

#11 0x0000000000da3032 in
boost::asio::detail::write_op<boost::asio::ssl::stream<boost::asio::basic_stream_socket<boost::asio::ip::tcp,
boost::asio::stream_socket_service<boost::asio::ip::tcp> > >,
boost::asio::mutable_buffers_1, boost::asio::detail::transfer_all_t,
boost::_bi::bind_t<void, boost::_mfi::mf2<void, Peer,
boost::system::error_code const&, unsigned long>,
boost::_bi::list3<boost::_bi::value<boost::shared_ptr<Peer> >, boost::arg<1>
(*)(), boost::arg<2> (*)()> > >::operator() (this=0x7fff8ad74f00, ec=...,   
bytes_transferred=0, start=1) at /usr/include/boost/asio/impl/write.hpp:250 

#12 0x0000000000d9ffbb in
boost::asio::async_write<boost::asio::ssl::stream<boost::asio::basic_stream_socket<boost::asio::ip::tcp,
boost::asio::stream_socket_service<boost::asio::ip::tcp> > >,
boost::asio::mutable_buffers_1, boost::_bi::bind_t<void,
boost::_mfi::mf2<void, Peer, boost::system::error_code const&, unsigned
long>, boost::_bi::list3<boost::_bi::value<boost::shared_ptr<Peer> >,  
boost::arg<1> (*)(), boost::arg<2> (*)()> > > (s=..., buffers=...,     
handler=...) at /usr/include/boost/asio/impl/write.hpp:585

有任何想法吗？

4

2 回答 2

5

这是多线程问题的结果。

对于大多数 Boost.Asio 对象，在一个对象上挂起多个异步操作是安全的；它只是指定对对象的并发调用是不安全的。尽管如此，它通常不会在某些类型上出现问题，例如ip::tcp::socket. 但是，ssl::stream强调：

应用程序还必须确保所有异步操作都在相同的隐式或显式链中执行。

在您的情况下，解决方案是wrap()每个组合操作的完成处理程序由同一条链完成。这将导致在链中调用所有中间操作。有关线程安全性和股线细微差别的更多详细信息，请考虑阅读此答案。

于 2013-03-12T16:48:41.073 回答
0

如果您可以提供代码，它可能会有所帮助。仅从堆栈跟踪中很难分析它。如果由于某种原因您无法提供代码，那么这里有一些想法：

bio_write 是调用以将数据写入套接字的开放 SSL 方法。如果您尝试直接调用它，可能会出现一些问题。 几个月前，有人在尝试使用它时寻求帮助，但得到的答案为零。

我也在使用带有 SSL 的 ASIO。我的代码使用一个线程将消息写入套接字，并使用另一个线程处理从套接字读取消息。我的客户端正在与多台服务器通信，但总共有 2 个线程来处理所有服务器上的所有套接字 I/O。这是我写出套接字的代码，到目前为止在 Windows 下没有问题：

void SSLSocket::SendWorkerThread(SSLSocket* psSLS)
{
   // This thread method that gets called to process the messages to be sent to the server.
   //
   // Since this has to be a static method, call a method on the class to handle server requests.
   psSLS->ProcessSendRequests();
}

void SSLSocket::ProcessSendRequests()
{
   // This method handles sending msgs to the server.
   //
   std::stringstream ss;
   DWORD WaitResult;
   Log.LogString("SSLSocket::ProcessSendRequests: Worker thread " + Logger::NumberToString(boost::this_thread::get_id()) + " started.\n", LogInfo);
   // Loop until the user quits, or an error of some sort is thrown.
   try
   {
      do
      {
         // If there are one or more msgs that need to be sent to a server, then send them out.
         if (SendMsgQ.Count() > 0)
         {
            Message* pMsg = SendMsgQ.Front();
            SSLSocket* pSSL = pMsg->pSSL;
            SendMsgQ.Pop();
            const Byte* pBuf = pMsg->pBuf;
            const int BytesInMsg = pMsg->BytesInMsg;
            boost::system::error_code Error;
            {
               Locking CodeLock(SocketLock); // Single thread the code.
               // unsigned int BytesTransferred = boost::asio::write(*pSSL->pSocket, boost::asio::buffer(pBuf, BytesInMsg), Error);
               boost::asio::async_write(*pSSL->pSocket, boost::asio::buffer(pBuf, BytesInMsg), boost::bind(&SSLSocket::HandleWrite, this,
                  boost::asio::placeholders::error, boost::asio::placeholders::bytes_transferred));
            }
            ss << "SSLSocket::ProcessSendRequests: # bytes sent = " << BytesInMsg << "\n";
            Log.LogString(ss.str(), LogDebug2);
            Log.LogBuf(pBuf, BytesInMsg, DisplayInHex, LogDebug3);
         }
         else
         {
            // Nothing to send, so go into a wait state.
            WaitResult = WaitForSingleObject(hEvent, INFINITE);
            if (WaitResult != 0L)
            {
               Log.LogString("SSLSocket::ProcessSendRequests: WaitForSingleObject event error.  Code = " + Logger::NumberToString(GetLastError()) + ". \n", LogError);
            }
         }
      } while (ReqAlive);
      Log.LogString("SSLSocket::ProcessSendRequests: Worker thread " + Logger::NumberToString(boost::this_thread::get_id()) + " done.\n", LogInfo);
   }
   catch (std::exception& e)
   {
      stringstream ss;
      ss << "SSLSocket::ProcessSendRequests: threw an error - " << e.what() << ".\n";
      Log.LogString(ss.str(), LogError);
      Stop();
   }
}

请注意，握手必须在尝试写入套接字之前完成。学到了一个艰难的方式。

如果这没有帮助，请尝试打印出它试图写出的字节数。如果它试图写出大量字节，那么也许值得研究。ASIO 可以接受大量数字，但 SSL 可能不行。您也可以尝试将程序置于调试状态，并在调用 write 语句时设置断点。尝试查看缓冲区结构以确保它以及其他对象（如 io_service 对象和套接字对象）完好无损，以查看是否有任何东西被踩到或可能被无意破坏。

我正在使用 SSL 库的版本 c，因为 OpenSSL 网站声称这是最稳定的版本。因此，如果您使用的是比该版本更新的版本并且所有其他方法都失败了，您可以尝试使用版本 c 来代替。

于 2013-03-12T04:17:09.740 回答