[trac554] First stage of adding protocol-dependent upstream fetch

Admin tasks:
* Split out io_error.h from asiolink.h
* Made test files follow naming convention of main files
* More discriminatory includes in some files
* Updates tests/Makefile.am

Coding tasks:
* Add additional methods to io_socket.* and put in dummies in
  {tcp,udp}_socket.h
* Incorporated basic IO Fetch code from Scott

src/lib/asiolink/Makefile.am

@@ -21,6 +21,7 @@ libasiolink_la_SOURCES += dns_answer.h
 libasiolink_la_SOURCES += simple_callback.h
 libasiolink_la_SOURCES += interval_timer.h interval_timer.cc
 libasiolink_la_SOURCES += recursive_query.h recursive_query.cc
+libasiolink_la_SOURCES += io_error.h
 libasiolink_la_SOURCES += io_socket.cc io_socket.h
 libasiolink_la_SOURCES += io_message.h
 libasiolink_la_SOURCES += io_address.cc io_address.h

src/lib/asiolink/asiolink.h

@@ -32,6 +32,7 @@
 #include <asiolink/io_endpoint.h>
 #include <asiolink/io_message.h>
 #include <asiolink/io_socket.h>
+#include <asiolink/io_error.h>
 
 /// \namespace asiolink
 /// \brief A wrapper interface for the ASIO library.
@@ -83,20 +84,6 @@
 /// the placeholder of callback handlers:
 /// http://think-async.com/Asio/asio-1.3.1/doc/asio/reference/asio_handler_allocate.html
 
-namespace asiolink {
-
-
-/// \brief An exception that is thrown if an error occurs within the IO
-/// module.  This is mainly intended to be a wrapper exception class for
-/// ASIO specific exceptions.
-class IOError : public isc::Exception {
-public:
-    IOError(const char* file, size_t line, const char* what) :
-        isc::Exception(file, line, what) {}
-};
-
-
-}      // asiolink
 #endif // __ASIOLINK_H
 
 // Local Variables: 

src/lib/asiolink/io_completion_cb.h

+// Copyright (C) 2011  Internet Systems Consortium, Inc. ("ISC")
+//
+// Permission to use, copy, modify, and/or distribute this software for any
+// purpose with or without fee is hereby granted, provided that the above
+// copyright notice and this permission notice appear in all copies.
+//
+// THE SOFTWARE IS PROVIDED "AS IS" AND ISC DISCLAIMS ALL WARRANTIES WITH
+// REGARD TO THIS SOFTWARE INCLUDING ALL IMPLIED WARRANTIES OF MERCHANTABILITY
+// AND FITNESS.  IN NO EVENT SHALL ISC BE LIABLE FOR ANY SPECIAL, DIRECT,
+// INDIRECT, OR CONSEQUENTIAL DAMAGES OR ANY DAMAGES WHATSOEVER RESULTING FROM
+// LOSS OF USE, DATA OR PROFITS, WHETHER IN AN ACTION OF CONTRACT, NEGLIGENCE
+// OR OTHER TORTIOUS ACTION, ARISING OUT OF OR IN CONNECTION WITH THE USE OR
+// PERFORMANCE OF THIS SOFTWARE.
+
+#ifndef __IO_COMPLETION_CB_H
+#define __IO_COMPLETION_CB_H
+
+#include <asio/error_code.hpp>
+#include <coroutine.h>
+
+
+/// \brief Asynchronous I/O Completion Callback
+///
+/// The stackless coroutine code requires that there be an "entry function"
+/// containing the coroutine macros.  When the coroutine yields, its state is
+/// stored and when the "entry function" is called again, it jumps to the
+/// line when processing left off last time.  In BIND-10, that "entry function"
+/// is the Boost asynchronous I/O completion callback - in essence operator().
+///
+/// This class solves the problem of circularity in class definitions.  In
+/// BIND10, classes such as IOFetch contain the coroutine code.  These include
+/// objects of classes such as IOSocket, whose signature has to include the
+/// callback object - IOFetch.  By abstracting the I/O completion callback into
+/// this class, that circularity is broken.
+///
+/// One more point: the asynchronous I/O functions take the callback object by
+/// reference.  But if a derived class object is passed as a reference to its
+/// base class, "class slicing" takes place - the derived part of the class is
+/// lost and only the base class functionality remains.  By storing a pointer
+/// to the true object and having the base class method call the derived class
+/// method through that, the behaviour of class inheritance is restored. In
+/// other words:
+/// \code
+/// class derived: public class base {
+/// :
+/// };
+/// derived drv;
+///
+/// // Call with pointer to base class
+/// void f(base* b, asio::error_code& ec, size_t length) {
+///    b->operator()(ec, length);
+/// }
+///
+/// // Call with reference to base class
+/// void g(base& b, asio::error_code& ec, size_t length) {
+///    b.operator()(ec, length);
+/// }
+///
+/// void function xyz(derived *d, asio::error_code& ec, size_t length) {
+///    f(d, ec, length);  // Calls derived::operator()
+///    g(*d, ec, length); // Also calls derived::operator()
+/// }
+/// \endcode
+
+class IOCompletionCallback : public coroutine {
+public:
+
+    /// \brief Constructor
+    IOCompletionCallback() : self_(this)
+    {}
+
+    /// \brief Virtual Destructor
+    virtual ~IOCompletionCallback()
+    {}
+
+    /// \brief Callback Method
+    virtual void operator()(asio::error_code ec = asio::error_code(),
+        size_t length = 0) {
+        (*self_)(ec, length);
+    }
+
+private:
+    IOCompletionCallback*   self_;      ///< Pointer to real object
+};
+
+
+#endif // __IO_COMPLETION_CB_H

src/lib/asiolink/io_error.h

+// Copyright (C) 2011  Internet Systems Consortium, Inc. ("ISC")
+//
+// Permission to use, copy, modify, and/or distribute this software for any
+// purpose with or without fee is hereby granted, provided that the above
+// copyright notice and this permission notice appear in all copies.
+//
+// THE SOFTWARE IS PROVIDED "AS IS" AND ISC DISCLAIMS ALL WARRANTIES WITH
+// REGARD TO THIS SOFTWARE INCLUDING ALL IMPLIED WARRANTIES OF MERCHANTABILITY
+// AND FITNESS.  IN NO EVENT SHALL ISC BE LIABLE FOR ANY SPECIAL, DIRECT,
+// INDIRECT, OR CONSEQUENTIAL DAMAGES OR ANY DAMAGES WHATSOEVER RESULTING FROM
+// LOSS OF USE, DATA OR PROFITS, WHETHER IN AN ACTION OF CONTRACT, NEGLIGENCE
+// OR OTHER TORTIOUS ACTION, ARISING OUT OF OR IN CONNECTION WITH THE USE OR
+// PERFORMANCE OF THIS SOFTWARE.
+
+
+#ifndef __IO_ERROR_H
+#define __IO_ERROR_H
+
+#include <exceptions/exceptions.h>
+
+namespace asiolink {
+
+/// \brief An exception that is thrown if an error occurs within the IO
+/// module.  This is mainly intended to be a wrapper exception class for
+/// ASIO specific exceptions.
+class IOError : public isc::Exception {
+public:
+    IOError(const char* file, size_t line, const char* what) :
+        isc::Exception(file, line, what) {}
+};
+
+
+}      // asiolink
+
+#endif // __IO_ERROR_H

src/lib/asiolink/iofetch.cc → src/lib/asiolink/io_fetch.cc

@@ -18,13 +18,12 @@
 #include <sys/socket.h>
 #include <netinet/in.h>
 
-#include <boost/bind.hpp>
-
 #include <asio.hpp>
 #include <asio/deadline_timer.hpp>
 #include <asio/ip/address.hpp>
 
 #include <boost/shared_ptr.hpp>
+#include <boost/bind.hpp>
 #include <boost/date_time/posix_time/posix_time_types.hpp>
 
 #include <dns/buffer.h>
@@ -35,7 +34,7 @@
 #include <dns/rcode.h>
 
 #include <asiolink.h>
-#include <internal/coroutine.h>
+#include <coroutine.h>
 #include <internal/udpdns.h>
 #include <internal/tcpdns.h>
 #include <internal/iofetch.h>
@@ -50,137 +49,59 @@ using namespace isc::dns;
 
 namespace asiolink {
 
-struct TcpFetch::UdpData {
-    // UDP Socket we send query to and expect reply from there
-    udp::socket socket;
-    // Where was the query sent
-    udp::endpoint remote;
-    // What we ask the server
-    Question question;
-    // We will store the answer here
-    OutputBufferPtr buffer;
-    OutputBufferPtr msgbuf;
-    // Temporary buffer for answer
-    boost::shared_array<char> data;
-    // This will be called when the data arrive or timeouts
-    Callback* callback;
-    //Callback* callback;
-    // Did we already stop operating (data arrived, we timed out, someone
-    // called stop). This can be so when we are cleaning up/there are
-    // still pointers to us.
-    bool stopped;
-    // Timer to measure timeouts.
-    deadline_timer timer;
-    // How many milliseconds are we willing to wait for answer?
-    int timeout;
-
-    UdpData(io_service& service,
-        const udp::socket::protocol_type& protocol,
-        const Question &q,
-        OutputBufferPtr b, Callback *c) :
-          socket(service, protocol),
-          question(q),
-          buffer(b),
-          msgbuf(new OutputBuffer(512)),
-          callback(c),
-          stopped(false),
-          timer(service)
-    { }
-
-
-};
-
-struct TcpFetch::TcpData {
-    // TCP Socket
-    tcp::socket socket;
-    // tcp endpoint
-    tcp::endpoint remote;
-    // What we ask the server
-    Question question;
-    // We will store the answer here
-    OutputBufferPtr buffer;
-    OutputBufferPtr msgbuf;
-    // Temporary buffer for answer
-    boost::shared_array<char> data;
-    // This will be called when the data arrive or timeouts
-    Callback* callback;
-    // Did we already stop operating (data arrived, we timed out, someone
-    // called stop). This can be so when we are cleaning up/there are
-    // still pointers to us.
-    bool stopped;
-    // Timer to measure timeouts.
-    deadline_timer timer;
-    // How many milliseconds are we willing to wait for answer?
-    int timeout;
+// Constructor for the IOFetchData member
 
-    TcpData(io_service& service,
-        const tcp::socket::protocol_type& protocol,
-        const Question &q,
-        OutputBufferPtr b, Callback *c) :
-          socket(service, protocol),
-          question(q),
-          buffer(b),
-          msgbuf(new OutputBuffer(512)),
-          callback(c),
-          stopped(false),
-          timer(service)
-    { }
-
-
-};
-
-UdpFetch::UdpFetch(io_service& io_service, const Question& q,
-                 const IOAddress& addr, uint16_t port,
-                 OutputBufferPtr buffer, Callback *callback, int timeout) :
-    data_(new UdpData(io_service, 
-          addr.getFamily() == AF_INET ? udp::v4() : udp::v6(), 
-          q, buffer, callback))
-{
-    data_->remote = UDPEndpoint(addr, port).getASIOEndpoint();
-    data_->timeout = timeout;
-}
-TcpFetch::TcpFetch(io_service& io_service, const Question& q,
-                 const IOAddress& addr, uint16_t port,
-                 OutputBufferPtr buffer, Callback *callback, int timeout) :
-    tcp_data_(new TcpData(io_service, 
-          addr.getFamily() == AF_INET ? udp::v4() : udp::v6(), 
-          q, buffer, callback))
+/// \brief Constructor
+///
+/// Just fills in the data members of the IOFetchData structure
+///
+/// \param io_service I/O Service object to handle the asynchronous
+///     operations.
+/// \param question DNS question to send to the upstream server.
+/// \param address IP address of upstream server
+/// \param port Port to use for the query
+/// \param buffer Output buffer into which the response (in wire format)
+///     is written (if a response is received).
+/// \param callback Callback object containing the callback to be called
+///     when we terminate.  The caller is responsible for managing this
+///     object and deleting it if necessary.
+/// \param timeout Timeout for the fetch (in ms).  The default value of
+///     -1 indicates no timeout.
+/// \param protocol Protocol to use for the fetch.  The default is UDP
+
+IOFetch::IOFetchData::IOFetchData(IOService& io_service,
+    const isc::dns::Question& query, const IOAddress& address, uint16_t port,
+    isc::dns::OutputBufferPtr buff, Callback* cb, int wait, int protocol)
+        :
+    socket((protocol == IPPROTO_UDP) ?
+        static_cast<IOSocket*>(new UDPSocket(io_service, address)) :
+        static_cast<IOSocket*>(new TCPSocket(io_service, address))
+    ),
+    remote((protocol == IPPROTO_UDP) ?
+        static_cast<IOEndpoint*>(new UDPEndpoint(address, port)) :
+        static_cast<IOEndpoint*>(new TCPEndpoint(address, port))
+    ),
+    question(query),
+    buffer(buff),
+    msgbuf(new OutputBuffer(512)),         // TODO: Why this number?
+    data(new char[IOFetch::MAX_LENGTH]),
+    callback(cb),
+    stopped(false),
+    timer(io_service.get_io_service()),
+    timeout(wait)
 {
-    tcp_data_->remote = TCPEndpoint(addr, port).getASIOEndpoint();
-    tcp_data_->timeout = timeout;
 }
 
-//
-/// The following functions implement the \c IOFetch class.
-///
-/// The constructor
-IOFetch::IOFetch(io_service& io_service, const Question& q, 
-                 const IOAddress& addr, uint16_t port,
-                 OutputBufferPtr buffer, Callback *callback, int timeout,
-                 int protocol)
+
+
+/// IOFetch Constructor - just initialize the private data
+IOFetch::IOFetch(IOService& io_service, const Question& question,
+    const IOAddress& address, uint16_t port, OutputBufferPtr buffer,
+    Callback *callback, int timeout, int protocol) :
+        data_(new IOFetch::IOFetchData(io_service, question, address, port,
+            buffer, callback, timeout, protocol)
+        )
 {
-    if (protocol == IPPROTO_TCP)
-    {
-        TcpFetch(io_service, q, addr, port, buffer, callback, timeout);
-/*
-        tcp_data_ = new TcpData(io_service, 
-          addr.getFamily() == AF_INET ? udp::v4() : udp::v6(), 
-          q, buffer, callback);
-        tcp_data_->remote = TCPEndpoint(addr, port).getASIOEndpoint();
-        tcp_data_->timeout = timeout;
-*/
-    }
-    else
-    {
-        UdpFetch(io_service, q, addr, port, buffer, callback, timeout);
-/*
-        data_(new UdpData(io_service, 
-          addr.getFamily() == AF_INET ? udp::v4() : udp::v6(), 
-          q, buffer, callback));
-        data_->remote = UDPEndpoint(addr, port).getASIOEndpoint();
-        data_->timeout = timeout;
-*/
-    }
 }
 
 /// The function operator is implemented with the "stackless coroutine"
@@ -198,7 +119,7 @@ IOFetch::operator()(error_code ec, size_t length) {
         {
             Message msg(Message::RENDER);
             
-            // XXX: replace with boost::random or some other suitable PRNG
+            // TODO: replace with boost::random or some other suitable PRNG
             msg.setQid(0);
             msg.setOpcode(Opcode::QUERY());
             msg.setRcode(Rcode::NOERROR());
@@ -207,7 +128,7 @@ IOFetch::operator()(error_code ec, size_t length) {
             MessageRenderer renderer(*data_->msgbuf);
             msg.toWire(renderer);
             dlog("Sending " + msg.toText() + " to " +
-                data_->remote.address().to_string());
+                data_->remote->getAddress().toText());
         }
 
 
@@ -220,10 +141,17 @@ IOFetch::operator()(error_code ec, size_t length) {
                 TIME_OUT));
         }
 
+        // Open a connection to the target system.  For speed, if the operation
+        // was a no-op (i.e. UDP operation) we bypass the yield.
+        bool do_yield = data_->socket->open(data->remote.get(), *this);
+        if (do_yield) {
+            CORO_YIELD;
+        }
+
         // Begin an asynchronous send, and then yield.  When the
         // send completes, we will resume immediately after this point.
-        CORO_YIELD data_->socket.async_send_to(buffer(data_->msgbuf->getData(),
-            data_->msgbuf->getLength()), data_->remote, *this);
+        CORO_YIELD data_->socket->async_send(data_->msgbuf->getData(),
+            data_->msgbuf->getLength(), data_->remote.get(), *this);
 
         /// Allocate space for the response.  (XXX: This should be
         /// optimized by maintaining a free list of pre-allocated blocks)
@@ -231,15 +159,16 @@ IOFetch::operator()(error_code ec, size_t length) {
 
         /// Begin an asynchronous receive, and yield.  When the receive
         /// completes, we will resume immediately after this point.
-        CORO_YIELD data_->socket.async_receive_from(buffer(data_->data.get(),
-            MAX_LENGTH), data_->remote, *this);
+        CORO_YIELD data_->socket->async_receive(data_->data.get(),
+            static_cast<size_t>(MAX_LENGTH), data_->remote.get(), *this);
+
         // The message is not rendered yet, so we can't print it easilly
-        dlog("Received response from " + data_->remote.address().to_string());
+        dlog("Received response from " + data_->remote->getAddress().toText());
 
-        /// Copy the answer into the response buffer.  (XXX: If the
+        /// Copy the answer into the response buffer.  (TODO: If the
         /// OutputBuffer object were made to meet the requirements of
         /// a MutableBufferSequence, then it could be written to directly
-        /// by async_recieve_from() and this additional copy step would
+        /// by async_receive_from() and this additional copy step would
         /// be unnecessary.)
         data_->buffer->writeData(data_->data.get(), length);
 
@@ -248,6 +177,13 @@ IOFetch::operator()(error_code ec, size_t length) {
     }
 }
 
+// Function that stops the coroutine sequence.  It is called either when the
+// query finishes or when the timer times out.  Either way, it sets the
+// "stopped_" flag and cancels anything that is in progress.
+//
+// As the function may be entered multiple times as things wind down, the
+// stopped_ flag checks if stop() has already been called.  If it has,
+// subsequent calls are no-ops.
 void
 IOFetch::stop(Result result) {
     if (!data_->stopped) {
@@ -255,19 +191,23 @@ IOFetch::stop(Result result) {
             case TIME_OUT:
                 dlog("Query timed out");
                 break;
+
             case STOPPED:
                 dlog("Query stopped");
                 break;
-            default:;
+
+            default:
+                ;
         }
         data_->stopped = true;
-        data_->socket.cancel();
-        data_->socket.close();
-        data_->timer.cancel();
+        data_->socket->cancel();    // Cancel outstanding I/O
+        data_->socket->close();     // ... and close the socket
+        data_->timer.cancel();      // Cancel timeout timer
+
         if (data_->callback) {
-            (*data_->callback)(result);
+            (*(data_->callback))(result);   // Call callback
         }
     }
 }
 
-}
+} // namespace asiolink

src/lib/asiolink/io_fetch.h

+// Copyright (C) 2010  Internet Systems Consortium, Inc. ("ISC")
+//
+// Permission to use, copy, modify, and/or distribute this software for any
+// purpose with or without fee is hereby granted, provided that the above
+// copyright notice and this permission notice appear in all copies.
+//
+// THE SOFTWARE IS PROVIDED "AS IS" AND ISC DISCLAIMS ALL WARRANTIES WITH
+// REGARD TO THIS SOFTWARE INCLUDING ALL IMPLIED WARRANTIES OF MERCHANTABILITY
+// AND FITNESS.  IN NO EVENT SHALL ISC BE LIABLE FOR ANY SPECIAL, DIRECT,
+// INDIRECT, OR CONSEQUENTIAL DAMAGES OR ANY DAMAGES WHATSOEVER RESULTING FROM
+// LOSS OF USE, DATA OR PROFITS, WHETHER IN AN ACTION OF CONTRACT, NEGLIGENCE
+// OR OTHER TORTIOUS ACTION, ARISING OUT OF OR IN CONNECTION WITH THE USE OR
+// PERFORMANCE OF THIS SOFTWARE.
+
+#ifndef __IOFETCH_H
+#define __IOFETCH_H 1
+
+#include <netinet/in.h>
+
+#include <config.h>
+
+#include <asio.hpp>
+#include <boost/shared_array.hpp>
+#include <boost/shared_ptr.hpp>
+#include <asio/deadline_timer.hpp>
+
+#include <dns/buffer.h>
+#include <dns/question.h>
+
+#include <asiolink/asiolink.h>
+#include <asiolink/ioaddress.h>
+#include <asiolink/iocompletioncb.h>
+#include <asiolink/iocompletioncb.h>
+
+
+#include <asiolink/iosocket.h>
+#include <asiolink/ioendpoint.h>
+#include <coroutine.h>
+
+
+
+namespace asiolink {
+
+/// \brief Upstream Fetch Processing
+///
+/// IOFetch is the class used to send upstream fetches and to handle responses.
+/// It is a base class containing most of the logic, although the ASIO will
+/// actually instantiate one of the derived classes TCPFetch or UDPFetch.
+/// (These differ in the type of socket and endpoint.)
+class IOFetch : public IOCompletionCallback  {
+public:
+
+    /// \brief Result of Upstream Fetch
+    ///
+    /// Note that this applies to the status of I/Os in the fetch - a fetch
+    /// that resulted in a packet being received from the server is a SUCCESS,
+    /// even if the contents of the packet indicate that some error occurred.
+    enum Result {
+        SUCCESS = 0,        ///< Success, fetch completed
+        TIME_OUT,           ///< Failure, fetch timed out
+        STOPPED             ///< Control code, fetch has been stopped
+    };
+    // The next enum is a "trick" to allow constants to be defined in a class
+    // declaration.
+
+    /// \brief Integer Constants
+    enum {
+        MAX_LENGTH = 4096   ///< Maximum size of receive buffer
+    };
+    /// \brief I/O Fetch Callback
+    ///
+    /// Callback object for when the fetch itself has completed.  Note that this
+    /// is different to the IOCompletionCallback; that is used to signal the
+    /// completion of an asynchronous I/O call.  The IOFetch::Callback is called
+    /// when an upstream fetch - which may have involved several asynchronous
+    /// I/O operations - has completed.
+    ///
+    /// This is an abstract class.
+    class Callback {
+    public:
+        /// \brief Default Constructor
+        Callback()
+        {}
+
+        /// \brief Virtual Destructor
+        virtual ~Callback()
+        {}
+
+        /// \brief Callback method called when the fetch completes
+        ///
+        /// \brief result Result of the fetch
+        virtual void operator()(Result result) = 0;
+    };
+
+    /// \brief IOFetch Data
+    ///
+    /// The data for IOFetch is held in a separate struct pointed to by a
+    /// shared_ptr object.  This is because the IOFetch object will be copied
+    /// often (it is used as a coroutine and passed as callback to many async_*()
+    /// functions) and we want keep the same data).  Organising the data this
+    /// way keeps copying to a minimum.
+    struct IOFetchData {
+
+        // The next two members are shared pointers to a base class because what
+        // is actually instantiated depends on whether the fetch is over UDP or
+        // TCP, which is not known until construction of the IOFetch.  Use of
+        // a shared pointer here is merely to ensure deletion when the data
+        // object is deleted.
+        boost::shared_ptr<IOSocket> socket;     ///< Socket to use for I/O
+        boost::shared_ptr<IOEndpoint> remote;   ///< Where the fetch was sent
+
+        isc::dns::Question          question;   ///< Question to be asked
+        isc::dns::OutputBufferPtr   buffer;     ///< Received data held here
+        isc::dns::OutputBufferPtr   msgbuf;     ///< ... and here
+        boost::shared_array<char>   data;       ///< Temporary array for the data
+        Callback*                   callback;   ///< Called on I/O Completion
+        bool                        stopped;    ///< Have we stopped running?
+        asio::deadline_timer        timer;      ///< Timer to measure timeouts
+        int                         timeout;    ///< Timeout in ms
+
+        /// \brief Constructor
+        ///
+        /// Just fills in the data members of the IOFetchData structure
+        ///
+        /// \param io_service I/O Service object to handle the asynchronous
+        ///     operations.
+        /// \param query DNS question to send to the upstream server.
+        /// \param address IP address of upstream server
+        /// \param port Port to use for the query
+        /// \param buff Output buffer into which the response (in wire format)
+        ///     is written (if a response is received).
+        /// \param cb Callback object containing the callback to be called
+        ///     when we terminate.  The caller is responsible for managing this
+        ///     object and deleting it if necessary.
+        /// \param wait Timeout for the fetch (in ms).  The default value of
+        ///     -1 indicates no timeout.
+        /// \param protocol Protocol to use for the fetch.  The default is UDP
+
+        IOFetchData(IOService& io_service, const isc::dns::Question& query,
+            const IOAddress& address, uint16_t port,
+            isc::dns::OutputBufferPtr buff, Callback* cb, int wait = -1,
+            int protocol = IPPROTO_UDP);
+    };
+
+    /// \brief Constructor.
+    ///
+    /// Creates the object that will handle the upstream fetch.
+    ///
+    /// \param io_service I/O Service object to handle the asynchronous
+    ///     operations.
+    /// \param question DNS question to send to the upstream server.