I am trying to implement a web server in C. I am sending a response of a request using the send command in a loop.
buf_len = 1;
while (buf_len > 0)
{
buf_len = read(fh, out_buf, BUF_SIZE);
if (buf_len > 0)
{
send(client_s, out_buf, buf_len, 0);
}
}
Should I terminate the last character of by char out_buf[BUF_SIZE] with '\0'.
Also when I am trying to send a png file with the same code, Sometimes I receive the entire image and sometimes only half of it. It loads fully when I refresh the page.
For send function, see man page
RETURN VALUE
On success, these calls return the number of characters sent. On error, -1 is returned, and errno is set appropriately.
So it doesn't guarantee that buf_len bytes are sent out, you can check the return value.
As Hardy said above, send does not guarantee that buf_len bytes are sent out. So one of many ways to do it is
buf_len = 1;
while (buf_len > 0) {
buf_len = read(fh, out_buf, BUF_SIZE);
if(buf_len < 0) {
perror("read");
exit(1);
}
int sent = 0;
int count = buf_len;
while(count) {
int n = send(client_s, out_buf + sent, count, 0);
count -= n;
sent += n;
}
}
How is out_buf defined?
Related
I am trying to write and read Integer value into/from C socket. Sometimes ntohs() return very big values like 55000 , 32000 etc...Though client is always sending value <1500. If I run the program it happens after 10-15 minutes...Sometimes after 20-30 minutes.
Can you please check below code and tell me
Why this line getting printed ?
printf("Garbage value - ntohs problem ..Exiting... ");
// write exactly n byte
inline int write_n(int fd, char *buf, int n) {
int nwrite, left = n;
int totalwrite = 0;
while (totalwrite != n) {
if ((nwrite = write(fd, buf, left)) <= 0) {
break;
} else {
totalwrite = totalwrite + nwrite;
left -= nwrite;
buf += nwrite;
}
}
if (totalwrite == 0)
return nwrite;
return totalwrite;
}
// send exactly n byte
inline int send_n(int fd, char *buf, int n) {
int nwrite, left = n;
int totalwrite = 0;
while (totalwrite != n) {
if ((nwrite = send(fd, buf, left, MSG_NOSIGNAL)) <= 0) {
break;
} else {
totalwrite = totalwrite + nwrite;
left -= nwrite;
buf += nwrite;
}
}
if (totalwrite == 0)
return nwrite;
return totalwrite;
}
uint16_t nread, len, plength, nsend;
int MTU = 1500;
char buffer[2000];
// Server receive ( Linux 64 bit)
while (1) {
// read packet length
nread = read_n(TCP_SOCKFD, (char *) &plength, sizeof(plength));
if (nread <=0) {
break;
}
len = ntohs(plength);
if (len <=0 || len > 1500 ) {
**printf("Garbage value - ntohs problem ..Exiting... "); // WHY ?**
break;
}
// read packat data
nread = read_n(SOCKFD, buffer, len);
if (nread != len) {
break;
}
}
//---------------------
// CLIENT send ( Android 5 )
while (1) {
nread = read(tunfd, buffer, MTU);
if (nread <= 0 || nread > 1500) { // always <=1500
break;
}
plength = htons(nread);
// send packet lenght
nsend = send_n(TCP_SOCKFD, (char *) &plength, sizeof(plength));
if (nsend != sizeof(plength)) {
break;
}
// send packet data
nsend = send_n(TCP_SOCKFD, buffer, nread);
if (nsend != nread) {
break;
}
}
Thank you
We cannot tell you with certainty what's happening because you cannot provide a verifiable example. Additionally, you've not presented the implementation of read_n(), but supposing that it follows the same model as write_n() and send_n(), we can nevertheless perform some analysis.
Each of the data transfer functions returns a short count in the event that data transfer is interrupted by an error. The client code watches for this, and breaks out of its loop if it detects it. Well and good. The server code does not do this when reading plength, however. Since plength, as a uint16_t, is two bytes in size, a partial read is possible and would go unnoticed by your server code.
In your example, plength is modified only via the one read_n() call presented. Network byte order is big-endian, so the most-significant byte is read first. It is possible that the combination of that byte with the stale one left over from the previous read would represent a number exceeding 1500. For example, if a 221(0x00dd)-byte packet is followed by a 1280(0x0500)-byte packet, and a partial read occurs on the second packet size, then the combined result will be 1501(0x05dd).
I don't presently see any reason to think that the client sends data different in nature than you think it does, and I don't presently see any other way that your server code could give the appearance of receiving different data than the client sends, especially since client and server each abort at the first recognized sign of trouble.
Do note, however, that this code could still be made more robust. In particular, consider that read(), write(), and send() can fail even when there is no problem with the underlying socket or data transfer request. In particular, they can fail with EINTR if the call is interrupted by a signal, and if the socket is in non-blocking mode then they can fail with EAGAIN. There may be others. It does not seem useful to operate your socket in non-blocking mode, but you might indeed want to watch for EINTR and resume reading after receiving it.
I would also suggest that, at least during development, you emit more data about the nature of the error. Call perror(), for example, and afterward print the bad data. You might even consider logging data sent and received.
I have a problem with receiving the correct values on the server-side of my client-server program.
header file included in both server and client:
#define CHUNK_SIZE 1024
#define ARR_LEN 3
client:
int uids[ARR_LEN] = {1994, 2423, 1222};
unsigned int uidlen = 0;
char uidbuffer[CHUNK_SIZE] = {0};
for(int i = 0; i < ARLL; i++)
{
uidlen = strlen(uids[i])+1;
snprintf(uidbuffer, uidlen, "%s", uids[i]);
if(send(socket, uidbuffer, strlen(uidbuffer), 0) < 0)
DIE("Write Error");
if(recv(socket, uidbuffer, sizeof(uidbuffer), 0) < 0)
DIE("Acknowledge Error");
memset(uidbuffer, 0, sizeof(uidbuffer));
}
server:
char uid_buff[CHUNK_SIZE];
for(int i = 0; i < ARR_LEN; i++)
{
memset(uid_buff, 0, sizeof(uid_buff));
// receiving the UID and storing it directly
if(recv(client_sock, uid_buff, sizeof(uid_buff), 0) < 0)
DIE("Receive Error");
printf("buffer content: %s\n", uid_buff);
uid_str = uid_buff;
uids[i] = (uid_t)strtol(uid_str, (char **)NULL, 10);
if(send(client_sock, uid_buff, sizeof(uid_buff), 0) < 0)
DIE("Acknowledge Error");
}
These are only parts of my program. I tried to only include the relevant parts. The output is this:
buffer content: 1994
buffer content: 24231222
buffer content:
While I want it to be:
buffer content: 1994
buffer content: 2423
buffer content: 1222
What could be the problem? I know that it's not so easy and that server–client communication is carried out in a stream of bytes rather than messages, but I want to mimmic that functionality by acknowledging every received "message". Could you please clue me what to do? I'm getting desperate.
You will need a protocol.
For example, you define that each message in your application has following format:
xx | message
this means first two bytes (mind the endianness) you receive indicate the length of the message that follows. Now you should first receive first two bytes - check the length - and then receive exactly that number of bytes. After that you know you have successfully received that message. And then you can proceed to other messages (which could/should have similar format: length + message itself).
Example:
Say you want to send three messages:
char s1[]="message1";
char s2[]="message2";
char s3[]="message3";
//You do this(client side):
int x1 = strlen(s1); // length of message1
int intsize = 4; // just size of integer -we'll need in next call
sendall(socket, &x1, &intsize); // send length of first message
sendall(socket, s1, &x1); // Now send the message
//On server:
int x = 0;
int y = 4; //size of integer most probably
receiveall(socket,&x,&y);//get length first; y=4 because that is size of integer
receiveall(socket, buffer, &x); // now we know how many bytes to expect - x - so request that number of bytes only
you can repeat this logic for other messages too.
Finally, you want to use such functions (here) instead of send and receive (because send and receive might not send/receive the number of bytes you tell it to):
int sendall(int s, char *buf, int *len)
{
int total = 0; // how many bytes we've sent
int bytesleft = *len; // how many we have left to send
int n;
while(total < *len) {
n = send(s, buf+total, bytesleft, 0);
if (n == -1) { break; }
total += n;
bytesleft -= n;
}
*len = total; // return number actually sent here
return n==-1?-1:0; // return -1 on failure, 0 on success
}
you will need a similar receiveall function.
You do not show the code that establishes the connection.
If you are using UDP sockets, each message is stand alone, you will receive a separate message for each message sent, but not necessarily in the same order.
If you are using TCP or Unix sockets, data may be received in chunks of different sizes than those used for sending. Only the sequence of bytes is preserved, not the chunk sizes. You need to specify a protocol: unless you specify a fixed number of bytes per message, you need to send some sort of separator to allow the server to tell where each message ends.
I'm writing a small and simple server (in C language for Linux stations).
A client requests a file to my server, my server asks this file to another server which sends it to my server.
My server should NOT receive ALL the file before sending it to the client BUT must send the bytes of the file so as they arrive.
This is an exercise in school so I can not dissociate myself from this requirement.
I have implemented the function explained below. The problem is that the client receives a non-deterministic number of bytes and NEVER the entire file.
int Recv_and_send_file (int socketa, int socketb, char *buffer, size_t file_size){
size_t n;
ssize_t nread;
ssize_t nwritten;
char c;
for (n=1; n<file_size; n++)
{
nread=recv(socketa, &c, 1, 0);
if (nread == 1)
{
nwritten = send(socketb,&c,1,0);
}
else if (nread == 0)
{
*buffer = 0;
return (-1); /* Errore */
}
else
return (-1); /* Errore */
}
}
*buffer = 0;
return (n);
}
Someone could kindly tell me where I'm wrong?
Is it an stupid idea to change the values SO_SNDBUF and SO_RCVBUF on both the server and the client?
Assuming the file_size is the total number of bytes you want to send, then your for loop will only send file_size - 1 bytes. In other words, you are off by one. Start from 0 instead to fix this:
for (n=0; n<file_size; n++)
{ //..
You capture the return value of send(), but you do not check to see if it was successful or not.
You are treating a 0 return value from recv() the same as an error. Since you do not show what you do after returning -1 from your function, I don't know if this may be contributing to your problem or not.
Certain errors on send() and recv() are "soft", in that you are allowed to retry the operation for those particular errors. One such error is EINTR, but check the documentation on your system to see if there are others.
In order to optimize performance and simplify your code, you can use splice()+pipes. Sendfile enables you to "forward" data between file descriptors, without the copy to user space.
Are you sure you have copied the correct code? That part as it is would not compile, there is a } in the last else which don't match with a corresponding {.
Also, how you get to know the file size? if it's send thru the socket as an integer, bear in mind the possible byte order of the source and destination machines.
Anyway, you are reading one byte at a time, you should improve it this way:
EDIT: use buffer and not the extra buff[2048];
int Recv_and_send_file (int socketa, int socketb, char *buffer, size_t file_size){
ssize_t nread;
ssize_t nwritten;
ssize_t bLeft=file_size;
while (bLeft > 0)
{
nread=recv(socketa, buffer, bleft, 0);
if (nread > 0)
{
nwritten = send(socketb, buffer, nread, 0);
bLeft -= nread;
buffer+=nread;
}
else if (nread == 0)
{
// I think this could raise a memory exception, read below
*buffer = 0;
return (-1); /* Errore */
}
else
{
return (-1); /* Errore */
}
}
// If buffer is allocated with file_size bytes this one will raise a memory exception
// *buffer = 0;
return (file_size-bLeft);
}
I am testing my TCP echo server, with Telnet, I can see that the client connects to the server and sends a charcter and in return the server returns a string to the client.
Now my problem is by using this recv() in a infinite loop I can only receive one character (even though the client tends to send a string).
This is how I am doing to receive the datagram from the client
TCP SERVER
while(1)
{
socket = accept(server_socket, (struct sockaddr *)&client_address, (socklen_t)&client_length);
recv(socket, recv_buffer, sizeof(recv_buffer), 0);
printf("Received string from client is %s", recv_buffer);
/*then I send my string to the client*/
send(socket, send_buffer, sizeof(send_buffer), 0);
}
Here is my problem that my recv() routine reads only one character even though the client wants to send a whole string. Is there a way how I can make this recv() routine wait before it receives all the characters from the client and then send a response to the client.
Any suggestions would be appreciated
Regards
Well, you are doing something wrong. Look at the definition of recv:
int recv(int s, void *buf, size_t len, int flags);
So it recieves len amount of bytes. You passed sizeof(recv_buffer) as the len parameter. Now I'm guessing recv_buffer is defined as a char*. Getting the sizeof of a pointer means that you get the amount of bytes necessary to store that pointer, instead of the memory it points to.
Do something like this instead:
const int buf_len = 100;
char recv_buffer[buf_len];
recv(socket, recv_buffer, buf_len, 0);
printf("Received string from client is %s", recv_buffer);
You need to build up the string you are receiving yourself in a loop, using the return value of recv() to find how many bytes you actually got. TCP/IP does not guarantee that all the data sent with one call to send() can be received with one call to recv(). And you must examine the return value of every sockets function you call to check for actual lengths sent/received, and for errors.
Your code is a disaster (sorry for being blunt, but it is best to be straight).
recv() returns the number of bytes actually read. Not only that but it will not clear the previous contents of the buffer and it will fill up right to the the end of the buffer if there is data available. All this means that you cannot treat the content of the buffer as a null terminated string.
You need to do something like:
ssize_t bytesRead = 1;
char recv_buffer[SOME_SIZE];
while (bytesRead > 0)
{
int bytesRead = recv(socket, recv_buffer, SOME_SIZE, 0);
if (bytesRead > 0)
{
// do something with the bytes. Note you cannot guarantee that the buffer contains a valid C string.
}
}
if (bytesRead == -1)
{
// report error from errno
}
else
{
// bytesRead == 0 have reached end of file (i.e. socket closed at other end)
}
There is no way to get recv to wait until the buffer is full before returning. It will wait until there are some bytes available and then return. The same applies to send by the way. You can't assume with one call to send that all of your bytes have actually been sent. You need to put send in a loop too:
ssize_t totalBytesWritten = 0;
ssize_t bytesWritten = 0;
while (bytesWritten >= 0 && totalBytesWritten < bytesToWrite)
{
bytesWritten = send(socket, sendBuffer + totalBytesWritten, bytesToWrite - totalBytesWritten, 0);
if (bytesWritten > 0)
{
totalBytesWritten += bytesWritten;
}
}
if (bytesWritten == -1)
{
// error
}
I'm trying to receive a single packet at a time from the server, since packets are going too fast, and each is of undefined size, calling recv() with number of bytes to read will read the first packet and maybe a part of the second packet. Since each packet is NULL terminated, I thought reading byte by byte until a NULL byte is received.
int recvLen = 0;
char TB;
char recvBuffer[1024];
while (recv(Socket, &TB, 1, 0) > 0 && TB != 0 && recvLen < 1024)
{
recvBuffer[recvLen] = TB;
recvLen++;
}
I don't think this method is efficient at all. If the server sent 1024 bytes, recv() will be called 1024 times.
Is there any other method to recv() until a NULL char is received, or some better method than this one I'm using?
EDIT:
i added the packet size infront of the data sent from the server, but now, if a false packet or even sometimes for no reason, packets gets messed up and no correct data is received. here is my code
#define UPLOAD_LEN 2755
int PacketSize, recvLen;
char Size[4];
char recvBuffer[UPLOAD_LEN+1];
while(1)
{
if(recv(Socket,Size,4,0)>0)
{
Size[4] = '\0';
PacketSize = atoi(Size);
if (PacketSize > UPLOAD_LEN || PacketSize <= 0) continue;
recvLen = recv(Socket, recvBuffer, PacketSize, 0);
} else recvLen = -1;
if (recvLen > 0)
{
recvBuffer[recvLen] = '\0';
ProcessData(recvBuffer);
}
else
{
closesocket(Socket);
}
}
I have never understood why communications protocols never support the one use case programmers expect to be able to do: exchange arbitrarily sized blobs with sends and recv's aligned on boundaries.
So theres no real shortcut here. You need to keep a persistent buffer that holds any data left over from the previous call to recv. Keep adding data to the end as you receive it, and return up to the terminating zero each time you find one. You'll probably have at least a partial following packet, so move that to the start of the buffer to serve as your initial state on the next call.
Create a buffer and extract your protocol messages from that. If the buffer does not contain a complete message, then recv() until it does. Here's a simple C implementation to buffer a socket (lightly tested, compiles on MS VS2008):
#include <winsock2.h>
#include <string.h>
typedef struct buffsock {
SOCKET s;
char* buf;
size_t maxlen;
size_t curlen;
} buffsock_t;
void buffsock_init(buffsock_t* bs,SOCKET s,size_t maxlen)
{
bs->s = s;
bs->buf = malloc(maxlen);
bs->maxlen = maxlen;
bs->curlen = 0;
}
void buffsock_free(buffsock_t* bs)
{
free(bs->buf);
bs->buf = NULL;
bs->maxlen = 0;
bs->curlen = 0;
bs->s = INVALID_SOCKET;
}
/* Attempt to fill internal buffer.
* Returns 0 if socket closed.
* Returns number of additional bytes in buffer otherwise.
*/
int buffsock_fill(buffsock_t* bs)
{
int bytes;
bytes = recv(bs->s,bs->buf + bs->curlen,bs->maxlen - bs->curlen,0);
if(bytes == SOCKET_ERROR)
return -1;
bs->curlen += bytes;
return bytes;
}
/* Return up to <bytes> from buffered socket.
* If return value 0 socket was closed.
* If return value >0 and <bytes socket received partial message.
*/
int buffsock_bytes(buffsock_t* bs,size_t bytes,void* msg)
{
while(bs->curlen < bytes)
{
int result;
result = buffsock_fill(bs);
if(result == -1)
return -1; /* error on socket */
if(result == 0)
break;
}
if(bytes > bs->curlen)
bytes = bs->curlen;
memcpy(msg,bs->buf,bytes);
bs->curlen -= bytes;
memmove(bs->buf,bs->buf + bytes,bs->curlen);
return bytes;
}
/* Implmementation of a protocol with two big-endian bytes indicating
* msg size followed by <size> bytes of message.
* Returns -1 if error on socket.
* Returns -2 if partial message recv'd (shouldn't happen as long as
* internal buffer is bigger than max message size).
* Returns -3 if user buffer not big enough to hold message.
* Returns size of message otherwise.
*/
int get_protocol_message(buffsock_t* bs,void* msg,size_t maxlen)
{
int bytes;
u_short len;
bytes = buffsock_bytes(bs,sizeof(u_short),&len);
if(bytes == 0)
return 0; /* socket closed, no more messages */
if(bytes == -1)
return -1; /* error on socket */
if(bytes < sizeof(u_short))
return -2; /* partial message */
len = ntohs(len);
if(len > maxlen)
return -3; /* message exceeds user buffer */
bytes = buffsock_bytes(bs,len,msg);
if(bytes < len)
return -2; /* partial message */
return bytes;
}
Use it like this:
int len;
char msg[256];
buffsock_t bs;
/* open a socket */
buffsock_init(&bs,sock,1024);
len = get_protocol_message(&bs,msg,sizeof(msg));
The key is TCP/IP has no concept of message boundaries, so recv() can return 1 to number of bytes requested. The received buffer could contain multiple or even partial messages.
This code just appends received data into a buffer. The protocol requests bytes from the buffer, and the buffer is filled from the socket. as bytes are removed the remaining buffered data is shifted to the beginning of the buffer.
In this case, two bytes are requested, converted to a length, then the remaining bytes are requested. If a request can't be satisfied, more data is recv'd.
Hope this helps.
There are several ways that you could do this.
Option #1: Before sending out any information, send out an int at the front of your packet which contains the size of the packet. Read this int, and then allocate a buffer which is the length of the int that you just received. Then you can recv() the entire packet at one time.
Option #2: Read in 1024 bytes at a time. recv() will give you back the number of bytes read. You can then use strlen() to figure out if you have more than one packet in your buffer. It would probably make the most sense to make this recursive(assuming that you could have several packets in 1024 bytes); so that you split the packets based on NULL bytes.