Thread: unknown socket error

I am using the following piece of code for performing a connectionless packet sending to a remote server. The domain' s IP is found by pinging it with cmd. However, the sendto() function fails, whereas 0 socket error code is returned. As I cannot spot any mistake, could you please help me fix it?

Code:

IPHEADER ipHeader;
char szSendbuf[60]={0}; 
TCPHEADER tcpHeader; 
PSDHEADER psdHeader;
SOCKADDR_IN ssin;
memset(&ssin, 0, sizeof(ssin));	
ssin.sin_family=AF_INET; 
ssin.sin_port=htons(0);
ssin.sin_addr.s_addr=inet_addr(target_ip); 

SOCKET ssock;
ssock=socket(AF_INET,SOCK_RAW,IPPROTO_RAW);
BOOL flag = TRUE; 
setsockopt(ssock, IPPROTO_IP, IP_HDRINCL, (char *)&flag, sizeof(flag));

ipHeader.verlen=(4<<4 | sizeof(ipHeader)/sizeof(unsigned long)); 
ipHeader.total_len=htons(sizeof(ipHeader)+sizeof(tcpHeader)); 
ipHeader.ident=1; 
ipHeader.frag_and_flags=0; 
ipHeader.ttl=128; 
ipHeader.proto=IPPROTO_TCP; 
ipHeader.checksum=0; 
ipHeader.sourceIP=my_ip;
ipHeader.destIP=ssin.sin_addr.s_addr;
tcpHeader.sport=htons((unsigned short)(rand()%1025));  
tcpHeader.seq=htonl(0x12345678); 
tcpHeader.flags=SYN;
tcpHeader.lenres=(sizeof(tcpHeader)/4<<4|0); 
tcpHeader.window=htons(512); 
tcpHeader.urg_ptr=0; 
tcpHeader.checksum=0;	
psdHeader.saddr=ipHeader.sourceIP; 
psdHeader.daddr=ipHeader.destIP; 
psdHeader.zero=0; 
psdHeader.proto=IPPROTO_TCP; 
psdHeader.length=htons((unsigned short)(sizeof(tcpHeader))); 
memcpy(szSendBuf, &psdHeader, sizeof(psdHeader)); 
memcpy(szSendBuf+sizeof(psdHeader), &tcpHeader, sizeof(tcpHeader)); 
tcpHeader.checksum=checksum((USHORT *)szSendBuf,sizeof(psdHeader)+sizeof(tcpHeader)); 
memcpy(szSendBuf, &ipHeader, sizeof(ipHeader)); 
memcpy(szSendBuf+sizeof(ipHeader), &tcpHeader, sizeof(tcpHeader)); 
memset(szSendBuf+sizeof(ipHeader)+sizeof(tcpHeader), 0, 4); 
ipHeader.checksum=checksum((USHORT *)szSendBuf, sizeof( ipHeader)+sizeof(tcpHeader)); 
memcpy(szSendBuf, &ipHeader, sizeof(ipHeader)); 

if (sendto(ssock, (char *)&szSendBuf, sizeof(szSendBuf), 0, (LPSOCKADDR)&ssin, sizeof(ssin)) == SOCKET_ERROR) {
			closesocket(ssock);
			printf("Error: %d\n", WSAGetLastError());
}

first, tell us the platform your code is for. you said you get the ip using "cmd", which we could then assume your on windows, but its certainly possible your just using "cmd" to get the ip, then manually enter it into the program on a different computer which may or may not be windows. we might be able to find out the platform, from platform specific types or calls from your code, but thats basically telling us you dont want to help us help you ("here you go, fix it").

next:

Originally Posted by http://msdn.microsoft.com/en-us/library/ms741580%28VS.85%29.aspx

If a function call's return value indicates that error or other relevant data was returned in the error code, WSAGetLastError should be called immediately. This is necessary because some functions may reset the last extended error code to 0 if they succeed, overwriting the extended error code returned by a previously failed function

Code:

if (sendto(ssock, (char *)&szSendBuf, sizeof(szSendBuf), 0, (LPSOCKADDR)&ssin, sizeof(ssin)) == SOCKET_ERROR) {
			closesocket(ssock);
			printf("Error: %d\n", WSAGetLastError());
}

from the horses mouth in the quote above, its possible that the "closesocket" is resetting the error code to 0. so, as the document says, check the error code immediately after the failed call (i.e. before the "closesocket" line).

also, after taking a quick look through the possible socket error codes on msdn, there are none that have value "0". this seems to confirm that your second call is overwriting the error code.

Raw socket support in Windows is deliberately broken. Have you read the "Limitations on Raw Sockets" section of the following page:

TCP/IP Raw Sockets (Windows)