Outline ! Socket basics ! TCP sockets ! Socket details ! Socket - PDF document

Outline ! Socket basics ! TCP sockets ! Socket details ! Socket options Computer Networks ! Final notes Sockets Socket Basics Ports ! An end-point for a IP network connection ! Numbers (vary in BSD, Solaris): – what the application layer “plugs into” – 0-1023 “reserved”, must be root – programmer cares about Application – 1024 - 5000 “ephemeral” Programming Interface (API) – however, many systems allow > 3977 ports ! End point determined by two things: N (50,000 is correct number) – Host address: IP address is Network Layer ! /etc/services: – Port number: is Transport Layer ftp 21/tcp ! Two end-points determine a connection: telnet 23/tcp socket pair finger 79/tcp – ex: 206.62.226.35,p21 + 198.69.10.2,p1500 snmp 161/udp – ex: 206.62.226.35,p21 + 198.69.10.2,p1499 Transport Layer Sockets and the OS ! UDP: User Datagram Protocol User – no acknowledgements Socket – no retransmissions Operating System – out of order, duplicate possible (Transport Layer) – connectionless ! TCP: Transmission Control Protocol – reliable (in order, all arrive, no duplicates) ! User sees “descriptor”, integer index – flow control – like: FILE * , or file index – connection – returned by socket() call (more later) – duplex – (proj 2) 1

Socket Details Addresses and Sockets ! Structure to hold address information Unix Network Programming , W. Richard Stevens, 2nd edition,  1998, Prentice Hall ! Functions pass address from user to OS – bind() – connect() ! Socket address structure – sendto() ! TCP client-server ! Functions pass address from OS to user ! Misc stuff – accept() – setsockopt(), getsockopt() – recvfrom() – fcntl() Server TCP Client-Server Socket Address Structure socket() “well-known” bind() port struct in_addr { in_addr_t s_addr; /* 32-bit IPv4 addresses */ listen() Client }; accept() struct sock_addr_in { socket() (Block until connection ) “Handshake” unit8_t sin_len; /* length of structure */ connect() sa_family_t sin_family; /* AF_INET */ Data (request) send() in_port_t sin_port; /* TCP/UDP Port num */ recv() struct in_addr sin_addr; /* IPv4 address */ Data (reply) send() char sin_zero[8]; /* unused */ recv() } End-of-File close() recv() ! Are also “generic” and “IPv6” socket structures close() socket() bind() int socket(int family , int type , int protocol ); int bind(int sockfd , const struct sockaddr * myaddr , socklen_t addrlen ); Create a socket, giving access to transport layer service. Assign a local protocol address (“name”) to a socket . ! family is one of ! sockfd is socket descriptor from socket() – AF_INET (IPv4), AF_INET6 (IPv6), AF_LOCAL (local Unix), ! myaddr is a pointer to address struct with: – AF_ROUTE (access to routing tables), AF_KEY (new, for encryption) – port number and IP address ! type is one of – SOCK_STREAM (TCP), SOCK_DGRAM (UDP) – if port is 0, then host will pick ephemeral port – SOCK_RAW (for special IP packets, PING, etc. Must be root) N not usually for server (exception RPC port-map) N setuid bit ( -rws--x--x root 1997 /sbin/ping*) – IP address != INADDR_ANY (multiple net cards) ! protocol is 0 (used for some raw socket options) ! addrlen is length of structure ! upon success returns socket descriptor ! returns 0 if ok, -1 on error – like file descriptor – EADDRINUSE (“Address already in use”) – -1 if failure 2

accept() listen() int accept(int sockfd , struct sockaddr cliaddr , int listen(int sockfd , int backlog ); socklen_t * addrlen ); Change socket state for TCP server. Return next completed connection. ! sockfd is socket descriptor from socket() ! sockfd is socket descriptor from socket() ! backlog is maximum number of incomplete ! cliaddr and addrlen return protocol address connections from client – historically 5 ! returns brand new descriptor, created by OS – rarely above 15 on a even moderate web server! ! if used with fork(), can create ! Sockets default to active (for client) concurrent server (more later) – change to passive to OS will accept connection connect() close() int connect(int sockfd , const struct sockaddr int close(int sockfd) ; * servaddr , socklen_t addrlen ); Connect to server. Close socket for use. ! sockfd is socket descriptor from socket() ! sockfd is socket descriptor from socket() ! servaddr is a pointer to a structure with: ! closes socket for reading/writing – port number and IP address – returns (doesn’t block) – must be specified (unlike bind() ) – attempts to send any unsent data ! addrlen is length of structure – socket option SO_LINGER ! client doesn’t need bind() N block until data sent – OS will pick ephemeral port N or discard any remaining data – -1 if error ! returns socket descriptor if ok, -1 on error Sending and Receiving Socket Options ! setsockopt (), getsockopt () int recv(int sockfd , void * buff , ! SO_LINGER size_t mbytes , int flags ); – upon close, discard data or block until sent int send(int sockfd , void * buff , ! SO_RCVBUF, SO_SNDBUF size_t mbytes , int flags ); – change buffer sizes ! Same as read() and write() but for flags – for TCP is “pipeline”, for UDP is “discard” – MSG_DONTWAIT (this send non-blocking) ! SO_RCVLOWAT, SO_SNDLOWAT – MSG_OOB (out of band data, 1 byte sent ahead) – how much data before “readable” via select () – MSG_PEEK (look, but don’t remove) ! SO_RCVTIMEO, SO_SNDTIMEO – MSG_WAITALL (don’t give me less than max) – timeouts – MSG_DONTROUTE (bypass routing table) 3

fcntl() Socket Options (TCP) ! ‘File control’ but used for sockets, too ! TCP_KEEPALIVE ! Signal driven sockets – idle time before close (2 hours, default) ! Set socket owner ! TCP_MAXRT ! Get socket owner – set timeout value ! Set socket non-blocking ! TCP_NODELAY flags = fcntl(sockfd, F_GETFL, 0); – disable Nagle Algorithm flags |= O_NONBLOCK; fcntl(sockfd, F_SETFL, flags); ! Beware not getting flags before setting! Concurrent TCP Server UDP Client-Server Server Text segment socket() Parent sock = socket() /* setup socket */ “well-known” int sock; bind() Client port while (1) { int newsock; newsock = accept(sock) recvfrom() socket() fork() (Block until receive datagram) Data (request) sendto() if child Child read(newsock) int sock; until exit sendto() int newsock; recvfrom() Data (reply) } close() - No “handshake” ! Close sock in child, newsock in parent - No simultaneous close() - Note, usually fork() for concurrent servers! ! Reference count for socket descriptor Called iterative server Sending and Receiving connect() with UDP int recvfrom(int sockfd , void * buff , size_t mbytes , int flags, struct sockaddr *from, socklen _ t *addrlen ); int sendto(int sockfd , void * buff , size_t mbytes , int ! Record address and port of peer flags, const struct sockaddr * to , socklen_t addrlen ); – datagrams to/from others are not allowed ! Same as recv() and send() but for addr – does not do three way handshake, or connection – recvfrom fills in address of where packet came – connect a misnomer, here. Should be from setpeername () – sento requires address of where sending packet to ! Use send () instead of sendto () ! Use recv () instead of recvfrom () ! Can change connect (or unconnect) by repeating connect () call 4

Why else about connected UDP? Why use connected UDP? ! Consider client: ! Send two datagrams ! Send two datagrams sendto() unconnected: connected: recvfrom() – connect the socket – connect the socket – NULL in recvfrom args could come from other – output first dgram – output first dgram – could fill in, or done in connect() – unconnect the socket – output second dgram ! Asynchronous errors not returned – connect the socket – talk.udp reno // server down, unconncected – output second dgram – unconnect the socket – hey! // no response, error not returned to process vs. – talk.udp reno // server down, conncected – port unreachable // error returned Mcast is Group and Port Mcast Extensions to UDP Client Server join Application Application socket() Sendto socket() host = 224.0.1.1 setsockopt() port = 123 port=123 Frames UDP UDP sendto() recvfrom() host=224.0.1.1 IP IP Other pause play clients Data Link Data Link LAN - sendto() in server to multicast group to router - setsockopt() in client to join group IP_ADD_MEMBERSHIP sendto() and recvfrom() specify port and group Scope of IPv4 Addresses Scope TTL Addr 0 node 1 224.0.0.0 to 224.0.0.255 link <32 239.255.0.0 to 239.255.255.255 site <32 239.192.0.0 to 239.195.255.255 org <255 224.0.1.0 to 238.255.255.255 global ! Use “link-local” addresses – only 255, but port numbers also unique ! On garden , ernie and bert 5