This question already has an answer here:
Closed 10 years ago.
Possible Duplicate:
Send buffer empty of Socket in Linux?
I want to create a socket server sending some data to a connecting client and disconnect him again.
I'm using non-blocking sockets so I don't know how to figure out if all data have at least been sent (send?) correctly (in short: no more data in my send buffer).
I don't want to keep the connection established if it's not neccessary anymore because I can't ensure that the client disconnects on his own.
Currently I'm just shutting down the client using shutdown() and later close(). But testing showed me a client does not always recieve all data before the connection gets closed.
There must be a way to ensure all data got send before closing the connection on non-blocking sockets, too, isn't there? Hope my question is clear enough and you can help me (:
The only way you can know your data has been sent prior to ending the connection is for the peer to acknowledge it in the application protocol. You can ensure that both ends get to EOS at the same time by shutting down for output at both ends and then reading to EOS at both ends, then closing the socket at both ends.
you can send the file size prior to data of the file. While closing the socket just check the file size and take appropriate action to close or resend the file....
Related
This question already has answers here:
close vs shutdown socket?
(9 answers)
Closed 6 years ago.
On this MSDN page:
Sending and Receiving Data on the Client
It recommends closing the sending side of the socket by using:
shutdown(SOCK_ID, SD_SEND);
Why should I?
Maybe I dont have to, and its just a recommendation? Maybe its for saving memory? Maybe for speed?
Does anyone have an idea?
The answer is in the shutdown() documentation:
If the how parameter is SD_SEND, subsequent calls to the send function are disallowed. For TCP sockets, a FIN will be sent after all data is sent and acknowledged by the receiver.
...
To assure that all data is sent and received on a connected socket before it is closed, an application should use shutdown to close connection before calling closesocket. One method to wait for notification that the remote end has sent all its data and initiated a graceful disconnect uses the WSAEventSelect function as follows :
Call WSAEventSelect to register for FD_CLOSE notification.
Call shutdown with how=SD_SEND.
When FD_CLOSE received, call the recv or WSARecv until the function completes with success and indicates that zero bytes were received. If SOCKET_ERROR is returned, then the graceful disconnect is not possible.
Call closesocket.
Another method to wait for notification that the remote end has sent all its data and initiated a graceful disconnect uses overlapped receive calls follows :
Call shutdown with how=SD_SEND.
Call recv or WSARecv until the function completes with success and indicates zero bytes were received. If SOCKET_ERROR is returned, then the graceful disconnect is not possible.
Call closesocket.
...
For more information, see the section on Graceful Shutdown, Linger Options, and Socket Closure.
In other words, at least for TCP, calling shutdown(SD_SEND) notifies the peer that you are done sending any more data, and that you will likely be closing your end of the connection soon. Preferably, the peer will also do the same courtesy for you. This way, both peers can know the connection was closed intentionally on both ends. This is known as a graceful disconnect, and not an abortive or abnormal disconnect.
By default, if you do not call shutdown(SD_SEND), closesocket() will attempt to perform a graceful shutdown for you UNLESS the socket's linger option is disabled. It is best not to rely on this behavior, you should always call shutdown() yourself before calling closesocket(), unless you have good reason not to.
It is unnecessary and redundant except in the following cases:
You want to achieve a synchronized close as described in the documentation quoted by Remy Lebeau.
The socket has been duplicated somehow, e.g. it is shared with child or parent processes or via the API, and you want to ensure the FIN is sent now.
Your application protocol requires that the peer receive a shutdown but needs to continue to send. This can arise for example when writing a proxy server.
You may have unread data in your socket receive buffer and you want to close and ignore it and send a FIN before provoking a connection reset, which will happen when you close if there is unread pending data.
These are the only cases I've ever come across in about 30 years: there may be others but I'm not aware of them.
There are no specific resources associated with sending or receiving operation on the socket, the socket is either used or closed. There reason for shutdown is not related to resource-management. Shutting down the socket is implementation of so-called graceful shutdown protocol, which allow both sides of the communication to realize the connection is going down and allows to minimize loss of data.
This question already has answers here:
Are parallel calls to send/recv on the same socket valid?
(3 answers)
Closed 7 years ago.
Sockets can generally two way communicate, therefore the same socket can be used to send and recv.
If I wanted to send some data (on another thread) while the socket is getting read, what would the kernel do? This is applied for both parts.
Consider this example: the server is sending you a file and say it will take a lot (low uplink or a very big file). The user gets bored and decides to SIGINT you. You catch it and tell the server to stop sending the file (with some kind of message).
Will you be able to send to tell the server to stop sending even though you're reading from it? And of course, that's applied to the server-side as well.
Hopefully I've been enough clear.
If I wanted to send some data (on another thread) while the socket is getting read, what would the kernel do?
Nothing special... sockets aren't like garden hoses... there's just some meta-data added to a packet that's sent between the machines, so the reading and writing happen independently (apart perhaps from if one side calls recv() on a socket that has unsent data in the local buffers due to the Nagle algorithm, which bunches up data into sensible size packets, it might time-out immediately and send whatever it can, but any tuning of that would be an implementation latency-tuning detail and doesn't change the big picture or way the client and server call the TCP API).
Consider this example: the server is sending you a file and say it will take a lot (low uplink or a very big file). The user gets bored and decides to SIGINT you. You catch it and tell the server to stop sending the file (with some kind of message). Will you be able to send to tell the server to stop sending even though you're reading from it? And of course, that's applied to the server-side as well.
The kernel accepts a limited amount of data to be sent, and a limited amount of data received, after which it forces the sending side to wait until some has been consumed before sending more. So, if you've sent data to a server, then get a local SIGINT and send a "oh, cancel that" in the same way, the server must read all the already-sent data before it can see the "oh, cancel that". If instead of sending it "in the same way" you turn on the Out Of Band (OOB) flag while sending the cancel message, then the server can (if it's written to do so) detect that there's OOB data and read it before it's completed reading/processing the other data. It will still need to read and discard whatever in-band data you've already sent, but the flow control / buffering mentioned above means that should be a manageable amount - far less than your file size might be. Throughout all this, whatever you want to recv or the server sends is independent and unaffected by the large client->server send, any OOB data etc..
There's a discussion and example code from GNU at http://www.gnu.org/software/libc/manual/html_node/Out_002dof_002dBand-Data.html
Thread 1 can safely write to the socket (with send) whilst thread 2 reads from the socket (with recv). What you need to be careful of is that at the point where you close() the socket the threads are synchronised, else the file descriptor may be used elsewhere, so the other thread (if not synchronized) could read from a file descriptor now used for something else. One way to achieve this would be for your reading thread to shutdown the file descriptor, which should cause the other end to drop the connection and thus error an in-progress send.
I'm using boost::asio and sending a list to a client and closing the socket when finished. Somehow the client sometimes gets an End Of File error before he has received everything.
I'm guessing this has to do with the server closing the socket right after sending the last list entry. Is there an easy way to solve this async_send to call the handler only after the data has been successfully sent?
Or is my End Of File error coming from something else?
Boost.Asio is an operating system independent abstraction layer over TCP and UDP sockets. They provide no guarantee that the other application has received and processed the data. You will need to include this logic in your application, you may want to study the OSI model.
If you're closing the socket immediately after async_send() returns, this is incorrect. You should close it only after the completion handler is invoked.
Suppose I have a server application - the connection is over TCP, using UNIX sockets.
The connection is asynchronous - in other words, clients' and servers' sockets are non-blocking.
Suppose the following situation: in some conditions, the server may decide to send some data to a connected client and immediately close the connection: using shutdown with SHUT_RDWR.
So, my question is - is it guaranteed, that when the client call recv, it will receive the (sent by the server) data?
Or, to receive the data, recv must be called before the server's shutdown? If so, what should I do (or, to be more precise, how should I do this), to make sure, that the data is received by the client?
You can control this behavior with "setsockopt(SO_LINGER)":
man setsockopt
SO_LINGER
Waits to complete the close function if data is present. When this option is enabled and there is unsent data present when the close
function is called, the calling application is blocked during the
close function until the data is transmitted or the connection has
timed out. The close function returns without blocking the caller.
This option has meaning only for stream sockets.
See also:
man read
Beej's Guide to Network Programming
There's no guarantee you will receive any data, let alone this data, but the data pending when the socket is closed is subject to the same guarantees as all the other data: if it arrives it will arrive in order and undamaged and subject to TCP's best efforts.
NB 'Asynchronous' and 'non-blocking' are two different things, not two terms for the same thing.
Once you have successfully written the data to the socket, it is in the kernel's buffer, where it will stay until it has been sent and acknowledged. Shutdown doesn't cause the buffered data to get lost. Closing the socket doesn't cause the buffered data to get lost. Not even the death of the sending process would cause the buffered data to get lost.
You can observe the size of the buffer with netstat. The SendQ column is how much data the kernel still wants to transmit.
After the client has acknowledged everything, the port disappears from the server. This may happen before the client has read the data, in which case it will be in RecvQ on the client. Basically you have nothing to worry about. After a successful write to a TCP socket, every component is trying as hard as it can to make sure that your data gets to the destination unharmed regardless of what happens to the sending socket and/or process.
Well, maybe one thing to worry about: If the client tries to send anything after the server has done its shutdown, it could get a SIGPIPE and die before it has read all the available data from the socket.
What is the easiest way to check if a socket was closed on the remote side of the connection? socket::is_open() returns true even if it is closed on the remote side (I'm using boost::asio::ip::tcp::socket).
I could try to read from the stream and see if it succeeds, but I'd have to change the logic of my program to make it work this way (I do not want data to be extracted from the stream at the point of the check).
Just check for boost::asio::error::eof error in your async_receive handler. It means the connection has been closed. That's the only proper way to do this.
Is there a boost peek function available? Most socket implementations have a way to read data without removing it from the queue, so you can read it again later. This would seem to satisfy your requirements.
After quickly glancing through the asio docs, I wasn't able to find exactly what I was expecting, but that doesn't mean its not there.
I'd suggest this for starters.
If the connection has been cleanly closed by the peer you should get an EOF while reading. Otherwise I generally ping in order to figure out if the connection is really alive.
I think that in general once you open a socket, you should start reading it inmediately and never stop doing so. This way you can make your server or client to support both synchronous and asynchronous protocols. The moment the client closes the connection, the moment the read will tell you this.
Using error_code is able to check the condition whether the client is connected or not. If the connection is success, the error_code error.value() will return 0, else return other value. You can also check the message() from the error_code.
boost::asio::socket_base::keep_alive keepAlive(true);
peerSocket->set_option(keepAlive);
Enable keep alive for the peer socket. Use the native socket to adjust the keepalive interval so that as soon as the connection is closed the async_receive handler will get EOF while reading.
Configuring TCP keep_alive with boost::asio