inet
Access to TCP/IP Protocols
Provides access to TCP/IP protocols.
See also ERTS User's Guide, Inet configuration for more information on how to configure an Erlang runtime system for IP communication.
Two Kernel configuration parameters affect the behaviour of all
sockets opened on an Erlang node:
inet_default_connect_options
can contain a list of default
options used for all sockets returned when doing connect
,
and inet_default_listen_options
can contain a list of
default options used when issuing a listen
call. When
accept
is issued, the values of the listensocket options
are inherited, why no such application variable is needed for
accept
.
Using the Kernel configuration parameters mentioned above, one
can set default options for all TCP sockets on a node. This should
be used with care, but options like {delay_send,true}
might be specified in this way. An example of starting an Erlang
node with all sockets using delayed send could look like this:
$erl -sname test -kernel \
inet_default_connect_options '[{delay_send,true}]' \
inet_default_listen_options '[{delay_send,true}]'
Note that the default option {active, true}
currently
cannot be changed, for internal reasons.
Addresses as inputs to functions can be either a string or a
tuple. For instance, the IP address 150.236.20.73 can be passed to
gethostbyaddr/1
either as the string "150.236.20.73"
or as the tuple {150, 236, 20, 73}
.
IPv4 address examples:
Address ip_address() ------- ------------ 127.0.0.1 {127,0,0,1} 192.168.42.2 {192,168,42,2}
IPv6 address examples:
Address ip_address() ------- ------------ ::1 {0,0,0,0,0,0,0,1} ::192.168.42.2 {0,0,0,0,0,0,(192 bsl 8) bor 168,(42 bsl 8) bor 2} FFFF::192.168.42.2 {16#FFFF,0,0,0,0,0,(192 bsl 8) bor 168,(42 bsl 8) bor 2} 3ffe:b80:1f8d:2:204:acff:fe17:bf38 {16#3ffe,16#b80,16#1f8d,16#2,16#204,16#acff,16#fe17,16#bf38} fe80::204:acff:fe17:bf38 {16#fe80,0,0,0,0,16#204,16#acff,16#fe17,16#bf38}
A function that may be useful is parse_address/1:
1>inet:parse_address("192.168.42.2").
{ok,{192,168,42,2}} 2>inet:parse_address("FFFF::192.168.42.2").
{ok,{65535,0,0,0,0,0,49320,10754}}
Types
hostent() =
#hostent{h_name = undefined | inet:hostname(),
h_aliases = [inet:hostname()],
h_addrtype = undefined | inet | inet6,
h_length = undefined | integer() >= 0,
h_addr_list = [inet:ip_address()]}
The record is defined in the Kernel include file "inet.hrl". Add the following directive to the module:
-include_lib("kernel/include/inet.hrl").
hostname() = atom() | string()
ip_address() = ip4_address() | ip6_address()
ip4_address() = {0..255, 0..255, 0..255, 0..255}
ip6_address() =
{0..65535,
0..65535,
0..65535,
0..65535,
0..65535,
0..65535,
0..65535,
0..65535}
port_number() = 0..65535
posix() = exbadport | exbadseq | file:posix()
An atom which is named from the Posix error codes used in Unix, and in the runtime libraries of most C compilers. See POSIX Error Codes.
socket()
See gen_tcp(3) and gen_udp(3).
address_family() = inet | inet6
Functions
get_rc() -> [{Par :: any(), Val :: any()}]
Returns the state of the Inet configuration database in form of a list of recorded configuration parameters. (See the ERTS User's Guide, Inet configuration, for more information). Only parameters with other than default values are returned.
format_error(Reason) -> string()
Reason = posix() | system_limit
Returns a diagnostic error string. See the section below for possible Posix values and the corresponding strings.
getaddr(Host, Family) -> {ok, Address} | {error, posix()}
Host = ip_address() | hostname()
Family = address_family()
Address = ip_address()
Returns the IP-address for
as a tuple of
integers.
can be an IP-address, a single hostname
or a fully qualified hostname.
getaddrs(Host, Family) -> {ok, Addresses} | {error, posix()}
Host = ip_address() | hostname()
Family = address_family()
Addresses = [ip_address()]
Returns a list of all IP-addresses for
.
can be an IP-address, a single hostname or a fully
qualified hostname.
gethostbyaddr(Address) -> {ok, Hostent} | {error, posix()}
Address = string() | ip_address()
Hostent = hostent()
Returns a hostent
record given an address.
gethostbyname(Hostname) -> {ok, Hostent} | {error, posix()}
Hostname = hostname()
Hostent = hostent()
Returns a hostent
record given a hostname.
gethostbyname(Hostname, Family) ->
{ok, Hostent} | {error, posix()}
Hostname = hostname()
Family = address_family()
Hostent = hostent()
Returns a hostent
record given a hostname, restricted
to the given address family.
gethostname() -> {ok, Hostname}
Hostname = string()
Returns the local hostname. Will never fail.
getifaddrs() -> {ok, Iflist} | {error, posix()}
Iflist = [{Ifname, [Ifopt]}]
Ifname = string()
Ifopt =
{flag, [Flag]} |
{addr, Addr} |
{netmask, Netmask} |
{broadaddr, Broadaddr} |
{dstaddr, Dstaddr} |
{hwaddr, Hwaddr}Flag =
up | broadcast | loopback | pointtopoint | running | multicastAddr = Netmask = Broadaddr = Dstaddr = ip_address()
Hwaddr = [byte()]
Returns a list of 2-tuples containing interface names and the
interface's addresses.
is a Unicode string.
is hardware dependent, e.g on Ethernet interfaces
it is the 6-byte Ethernet address (MAC address (EUI-48 address)).
The {addr,
, {netmask,_}
and {broadaddr,_}
tuples are repeated in the result list iff the interface has multiple
addresses. If you come across an interface that has
multiple {flag,_}
or {hwaddr,_}
tuples you have
a really strange interface or possibly a bug in this function.
The {flag,_}
tuple is mandatory, all other optional.
Do not rely too much on the order of
atoms or
tuples. There are some rules, though:
{addr,_}
follows {netmask,_}
{broadaddr,_}
if
the broadcast
flag is not set and the
pointtopoint
flag is set.
{netmask,_}
, {broadaddr,_}
or
{dstaddr,_}
tuples that follow an {addr,_}
tuple concerns that address.
The {hwaddr,_}
tuple is not returned on Solaris since the
hardware address historically belongs to the link layer and only
the superuser can read such addresses.
On Windows, the data is fetched from quite different OS API
functions, so the
and
values may be calculated, just as some
values.
You have been warned. Report flagrant bugs.
getopts(Socket, Options) -> {ok, OptionValues} | {error, posix()}
Socket = socket()
Options = [socket_getopt()]
OptionValues = [socket_setopt()]
socket_getopt() =
gen_sctp:option_name() |
gen_tcp:option_name() |
gen_udp:option_name()
socket_setopt() =
gen_sctp:option() | gen_tcp:option() | gen_udp:option()
Gets one or more options for a socket. See setopts/2 for a list of available options.
The number of elements in the returned
list does not necessarily correspond to the number of options
asked for. If the operating system fails to support an option,
it is simply left out in the returned list. An error tuple is only
returned when getting options for the socket is impossible
(i.e. the socket is closed or the buffer size in a raw request
is too large). This behavior is kept for backward
compatibility reasons.
A raw option request RawOptReq = {raw, Protocol, OptionNum, ValueSpec}
can be used to get information about
socket options not (explicitly) supported by the emulator. The
use of raw socket options makes the code non portable, but
allows the Erlang programmer to take advantage of unusual features
present on the current platform.
The RawOptReq
consists of the tag raw
followed
by the protocol level, the option number and either a binary
or the size, in bytes, of the
buffer in which the option value is to be stored. A binary
should be used when the underlying getsockopt
requires
input
in the argument field, in which case the size of the binary
should correspond to the required buffer
size of the return value. The supplied values in a RawOptReq
correspond to the second, third and fourth/fifth parameters to the
getsockopt
call in the C socket API. The value stored
in the buffer is returned as a binary ValueBin
where all values are coded in the native endianess.
Asking for and inspecting raw socket options require low level information about the current operating system and TCP stack.
As an example, consider a Linux machine where the
TCP_INFO
option could be used to collect TCP statistics
for a socket. Lets say we're interested in the
tcpi_sacked
field of the struct tcp_info
filled in when asking for TCP_INFO
. To
be able to access this information, we need to know both the
numeric value of the protocol level IPPROTO_TCP
, the
numeric value of the option TCP_INFO
, the size of the
struct tcp_info
and the size and offset of
the specific field. By inspecting the headers or writing a small C
program, we found IPPROTO_TCP
to be 6,
TCP_INFO
to be 11, the structure size to be 92 (bytes),
the offset of tcpi_sacked
to be 28 bytes and the actual
value to be a 32 bit integer. We could use the following
code to retrieve the value:
get_tcpi_sacked(Sock) -> {ok,[{raw,_,_,Info}]} = inet:getopts(Sock,[{raw,6,11,92}]), <<_:28/binary,TcpiSacked:32/native,_/binary>> = Info, TcpiSacked.
Preferably, you would check the machine type, the OS and the kernel version prior to executing anything similar to the code above.
getstat(Socket) -> {ok, OptionValues} | {error, posix()}
Socket = socket()
OptionValues = [{stat_option(), integer()}]
getstat(Socket, Options) -> {ok, OptionValues} | {error, posix()}
Socket = socket()
Options = [stat_option()]
OptionValues = [{stat_option(), integer()}]
stat_option() =
recv_cnt |
recv_max |
recv_avg |
recv_oct |
recv_dvi |
send_cnt |
send_max |
send_avg |
send_oct |
send_pend
Gets one or more statistic options for a socket.
getstat(
is equivalent to
getstat(
.
The following options are available:
recv_avg
Average size of packets in bytes received to the socket.
recv_cnt
Number of packets received to the socket.
recv_dvi
Average packet size deviation in bytes received to the socket.
recv_max
The size of the largest packet in bytes received to the socket.
recv_oct
Number of bytes received to the socket.
send_avg
Average size of packets in bytes sent from the socket.
send_cnt
Number of packets sent from the socket.
send_dvi
Average packet size deviation in bytes received sent from the socket.
send_max
The size of the largest packet in bytes sent from the socket.
send_oct
Number of bytes sent from the socket.
ntoa(IpAddress) -> Address | {error, einval}
Address = string()
IpAddress = ip_address()
Parses an ip_address() and returns an IPv4 or IPv6 address string.
parse_ipv4_address(Address) -> {ok, IPv4Address} | {error, einval}
Address = string()
IPv4Address = ip_address()
Parses an IPv4 address string and returns an ip4_address(). Accepts a shortened IPv4 shortened address string.
parse_ipv4strict_address(Address) ->
{ok, IPv4Address} | {error, einval}
Address = string()
IPv4Address = ip_address()
Parses an IPv4 address string containing four fields, i.e not shortened, and returns an ip4_address().
parse_ipv6_address(Address) -> {ok, IPv6Address} | {error, einval}
Address = string()
IPv6Address = ip_address()
Parses an IPv6 address string and returns an ip6_address(). If an IPv4 address string is passed, an IPv4-mapped IPv6 address is returned.
parse_ipv6strict_address(Address) ->
{ok, IPv6Address} | {error, einval}
Address = string()
IPv6Address = ip_address()
Parses an IPv6 address string and returns an ip6_address(). Does not accept IPv4 adresses.
parse_address(Address) -> {ok, IPAddress} | {error, einval}
Address = string()
IPAddress = ip_address()
Parses an IPv4 or IPv6 address string and returns an ip4_address() or ip6_address(). Accepts a shortened IPv4 address string.
parse_strict_address(Address) -> {ok, IPAddress} | {error, einval}
Address = string()
IPAddress = ip_address()
Parses an IPv4 or IPv6 address string and returns an ip4_address() or ip6_address(). Does not accept a shortened IPv4 address string.
peername(Socket) -> {ok, {Address, Port}} | {error, posix()}
Socket = socket()
Address = ip_address()
Port = integer() >= 0
Returns the address and port for the other end of a connection.
Note that for SCTP sockets this function only returns one of the socket's peer addresses. The function peernames/1,2 returns all.
peernames(Socket) -> {ok, [{Address, Port}]} | {error, posix()}
Socket = socket()
Address = ip_address()
Port = integer() >= 0
Equivalent to
peernames(
.
Note that this function's behaviour for an SCTP
one-to-many style socket is not defined by the
peernames(Socket, Assoc) ->
{ok, [{Address, Port}]} | {error, posix()}
Socket = socket()
Assoc = #sctp_assoc_change{} | gen_sctp:assoc_id()
Address = ip_address()
Port = integer() >= 0
Returns a list of all address/port number pairs for the other end
of a socket's association
.
This function can return multiple addresses for multihomed sockets such as SCTP sockets. For other sockets it returns a one element list.
Note that the
parameter is by the
0
means hence its behaviour for one-to-many style sockets
is unfortunately not defined.
port(Socket) -> {ok, Port} | {error, any()}
Socket = socket()
Port = port_number()
Returns the local port number for a socket.
sockname(Socket) -> {ok, {Address, Port}} | {error, posix()}
Socket = socket()
Address = ip_address()
Port = integer() >= 0
Returns the local address and port number for a socket.
Note that for SCTP sockets this function only returns one of the socket addresses. The function socknames/1,2 returns all.
socknames(Socket) -> {ok, [{Address, Port}]} | {error, posix()}
Socket = socket()
Address = ip_address()
Port = integer() >= 0
Equivalent to
socknames(
.
socknames(Socket, Assoc) ->
{ok, [{Address, Port}]} | {error, posix()}
Socket = socket()
Assoc = #sctp_assoc_change{} | gen_sctp:assoc_id()
Address = ip_address()
Port = integer() >= 0
Returns a list of all local address/port number pairs for a socket
for the given association
.
This function can return multiple addresses for multihomed sockets such as SCTP sockets. For other sockets it returns a one element list.
Note that the
parameter is by the
0
is defined to mean that the returned addresses shall be
without regard to any particular association.
How different SCTP implementations interprets this varies somewhat.
setopts(Socket, Options) -> ok | {error, posix()}
Socket = socket()
Options = [socket_setopt()]
socket_setopt() =
gen_sctp:option() | gen_tcp:option() | gen_udp:option()
Sets one or more options for a socket. The following options are available:
{active, true | false | once | N}
If the value is true
, which is the default,
everything received from the socket will be sent as
messages to the receiving process. If the value is
false
(passive mode), the process must explicitly
receive incoming data by calling
gen_tcp:recv/2,3
,
gen_udp:recv/2,3
or gen_sctp:recv/1,2
(depending on the type of socket).
If the value is once
({active, once}
),
one data message from the socket will be sent
to the process. To receive one more message,
setopts/2
must be called again with the
{active, once}
option.
If the value is an integer N
in the range -32768 to 32767
(inclusive), the value is added to the socket's count of data
messages sent to the controlling process. A socket's default
message count is 0. If a negative value is specified and its
magnitude is equal to or greater than the socket's current
message count, the socket's message count is set to 0. Once
the socket's message count reaches 0, either due to sending
received data messages to the process or by being explicitly set,
the process is then notified by a special message, specific to
the type of socket, that the socket has entered passive
mode. Once the socket enters passive mode, to receive more
messages setopts/2
must be called again to set the
socket back into an active mode.
When using {active, once}
or {active, N}
, the
socket changes behaviour automatically when data is received.
This can sometimes be confusing in combination with
connection-oriented sockets (i.e. gen_tcp
) as a socket
with {active, false}
behaviour reports closing
differently than a socket with {active, true}
behaviour. To make programming easier, a socket where
the peer closed and this was detected while in
{active, false}
mode, will still generate the
message
{tcp_closed,Socket}
when set to {active, once}
,
{active, true}
or {active, N}
mode. It is therefore
safe to assume that the message
{tcp_closed,Socket}
, possibly followed by socket
port termination (depending on the exit_on_close
option) will eventually appear when a socket changes
back and forth between {active, true}
and
{active, false}
mode. However,
when peer closing is detected is all up to the
underlying TCP/IP stack and protocol.
Note that {active, true}
mode provides no flow
control; a fast sender could easily overflow the
receiver with incoming messages. The same is true of
{active, N}
mode while the message count is greater
than zero. Use active mode only if
your high-level protocol provides its own flow control
(for instance, acknowledging received messages) or the
amount of data exchanged is small. {active, false}
mode, use of the {active, once}
mode or {active, N}
mode with values of N
appropriate for the application
provides flow control; the other side will not be able send
faster than the receiver can read.
{broadcast, Boolean}
(UDP sockets)Enable/disable permission to send broadcasts.
{buffer, Size}
Determines the size of the user-level software buffer used by
the driver. Not to be confused with sndbuf
and recbuf
options which correspond to
the kernel socket buffers. It is recommended
to have val(buffer) >= max(val(sndbuf),val(recbuf))
.
In fact, the val(buffer)
is automatically set to
the above maximum when sndbuf
or recbuf
values are set.
{delay_send, Boolean}
Normally, when an Erlang process sends to a socket,
the driver will try to immediately send the data. If that
fails, the driver will use any means available to queue
up the message to be sent whenever the operating system
says it can handle it. Setting {delay_send, true}
will make all messages queue up. This makes
the messages actually sent onto the network be larger but
fewer. The option actually affects the scheduling of send
requests versus Erlang processes instead of changing any
real property of the socket. Needless to say it is an
implementation specific option. Default is false
.
{deliver, port | term}
When {active, true}
delivers data on the forms
port
: {S, {data, [H1,..Hsz | Data]}}
or
term
: {tcp, S, [H1..Hsz | Data]}
.
{dontroute, Boolean}
Enable/disable routing bypass for outgoing messages.
{exit_on_close, Boolean}
By default this option is set to true
.
The only reason to set it to false
is if you want
to continue sending data to the socket after a close has
been detected, for instance if the peer has used
gen_tcp:shutdown/2
to shutdown the write side.
{header, Size}
This option is only meaningful if the binary
option was specified when the socket was created. If
the header
option is specified, the first
Size
number bytes of data received from the socket
will be elements of a list, and the rest of the data will
be a binary given as the tail of the same list. If for
example Size == 2
, the data received will match
[Byte1,Byte2|Binary]
.
{high_msgq_watermark, Size}
The socket message queue will be set into a busy state when the amount of data queued on the message queue reaches this limit. Note that this limit only concerns data that have not yet reached the ERTS internal socket implementation. Default value used is 8 kB.
Senders of data to the socket will be suspended if either the socket message queue is busy, or the socket itself is busy.
For more information see the low_msgq_watermark
,
high_watermark
, and low_watermark
options.
Note that distribution sockets will disable the use of
high_msgq_watermark
and low_msgq_watermark
,
and will instead use the
distribution
buffer busy limit which is a similar feature.
{high_watermark, Size}
(TCP/IP sockets)The socket will be set into a busy state when the amount of data queued internally by the ERTS socket implementation reaches this limit. Default value used is 8 kB.
Senders of data to the socket will be suspended if either the socket message queue is busy, or the socket itself is busy.
For more information see the low_watermark
,
high_msgq_watermark
, and low_msqg_watermark
options.
{ipv6_v6only, Boolean}
Restricts the socket to only use IPv6, prohibiting any
IPv4 connections. This is only applicable for
IPv6 sockets (option inet6
).
On most platforms this option has to be set on the socket before associating it to an address. Therefore it is only reasonable to give it when creating the socket and not to use it when calling the function (setopts/2) containing this description.
The behaviour of a socket with this socket option set to
true
is becoming the only portable one. The original
idea when IPv6 was new of using IPv6 for all traffic
is now not recommended by FreeBSD (you can use
{ipv6_v6only,false}
to override the recommended
system default value),
forbidden by OpenBSD (the supported GENERIC kernel)
and impossible on Windows (that has separate
IPv4 and IPv6 protocol stacks). Most Linux distros
still have a system default value of false
.
This policy shift among operating systems towards
separating IPv6 from IPv4 traffic has evolved since
it gradually proved hard and complicated to get
a dual stack implementation correct and secure.
On some platforms the only allowed value for this option
is true
, e.g. OpenBSD and Windows. Trying to set
this option to false
when creating the socket
will in this case fail.
Setting this option on platforms where it does not exist
is ignored and getting this option with
getopts/2
returns no value i.e the returned list will not contain an
{ipv6_v6only,_}
tuple. On Windows the option acually
does not exist, but it is emulated as being a
read-only option with the value true
.
So it boils down to that setting this option to true
when creating a socket will never fail except possibly
(at the time of this writing) on a platform where you
have customized the kernel to only allow false
,
which might be doable (but weird) on e.g. OpenBSD.
If you read back the option value using getopts/2 and get no value the option does not exist in the host OS and all bets are off regarding the behaviour of both an IPv6 and an IPv4 socket listening on the same port as well as for an IPv6 socket getting IPv4 traffic.
{keepalive, Boolean}
(TCP/IP sockets)Enables/disables periodic transmission on a connected socket, when no other data is being exchanged. If the other end does not respond, the connection is considered broken and an error message will be sent to the controlling process. Default disabled.
{linger, {true|false, Seconds}}
Determines the timeout in seconds for flushing unsent data in the
close/1
socket call. If the 1st component of the value
tuple is false
, the 2nd one is ignored, which means that
close/1
returns immediately not waiting
for data to be flushed. Otherwise, the 2nd component is
the flushing time-out in seconds.
{low_msgq_watermark, Size}
If the socket message queue is in a busy state, the socket message queue will be set in a not busy state when the amount of data queued in the message queue falls below this limit. Note that this limit only concerns data that have not yet reached the ERTS internal socket implementation. Default value used is 4 kB.
Senders that have been suspended due to either a busy message queue or a busy socket, will be resumed when neither the socket message queue, nor the socket are busy.
For more information see the high_msgq_watermark
,
high_watermark
, and low_watermark
options.
Note that distribution sockets will disable the use of
high_msgq_watermark
and low_msgq_watermark
,
and will instead use the
distribution
buffer busy limit which is a similar feature.
{low_watermark, Size}
(TCP/IP sockets)If the socket is in a busy state, the socket will be set in a not busy state when the amount of data queued internally by the ERTS socket implementation falls below this limit. Default value used is 4 kB.
Senders that have been suspended due to either a busy message queue or a busy socket, will be resumed when neither the socket message queue, nor the socket are busy.
For more information see the high_watermark
,
high_msgq_watermark
, and low_msgq_watermark
options.
{mode, Mode :: binary | list}
Received Packet
is delivered as defined by Mode.
{netns, Namespace :: file:filename_all()}
Set a network namespace for the socket. The Namespace
parameter is a filename defining the namespace for example
"/var/run/netns/example"
typically created by the command
ip netns add example
. This option must be used in a
function call that creates a socket i.e
gen_tcp:connect/3,4,
gen_tcp:listen/2,
gen_udp:open/1,2 or
gen_sctp:open/0-2.
This option uses the Linux specific syscall
setns()
such as in Linux kernel 3.0 or later
and therefore only exists when the runtime system
has been compiled for such an operating system.
The virtual machine also needs elevated privileges either
running as superuser or (for Linux) having the capability
CAP_SYS_ADMIN
according to the documentation for setns(2).
However, during testing also CAP_SYS_PTRACE
and CAP_DAC_READ_SEARCH
has proven to be necessary.
Example:
setcap cap_sys_admin,cap_sys_ptrace,cap_dac_read_search+epi beam.smp
Note also that the filesystem containing the virtual machine
executable (beam.smp
in the example above) has to be local,
mounted without the nosetuid
flag,
support extended attributes and that
the kernel has to support file capabilities.
All this runs out of the box on at least Ubuntu 12.04 LTS,
except that SCTP sockets appears to not support
network namespaces.
The Namespace
is a file name and is encoded
and decoded as discussed in
file
except that the emulator flag +fnu
is ignored and
getopts/2
for this option will return a binary for the filename
if the stored filename can not be decoded,
which should only happen if you set the option using a binary
that can not be decoded with the emulator's filename encoding:
file:native_name_encoding/0.
list
Received Packet
is delivered as a list.
binary
Received Packet
is delivered as a binary.
{nodelay, Boolean}
(TCP/IP sockets)If Boolean == true
, the TCP_NODELAY
option
is turned on for the socket, which means that even small
amounts of data will be sent immediately.
{packet, PacketType}
(TCP/IP sockets)Defines the type of packets to use for a socket. The following values are valid:
raw | 0
No packaging is done.
1 | 2 | 4
Packets consist of a header specifying the number of bytes in the packet, followed by that number of bytes. The length of header can be one, two, or four bytes; containing an unsigned integer in big-endian byte order. Each send operation will generate the header, and the header will be stripped off on each receive operation.
In current implementation the 4-byte header is limited to 2Gb.
asn1 | cdr | sunrm | fcgi | tpkt | line
These packet types only have effect on receiving.
When sending a packet, it is the responsibility of
the application to supply a correct header. On
receiving, however, there will be one message sent to
the controlling process for each complete packet
received, and, similarly, each call to
gen_tcp:recv/2,3
returns one complete packet.
The header is not stripped off.
The meanings of the packet types are as follows:
asn1
- ASN.1 BER,
sunrm
- Sun's RPC encoding,
cdr
- CORBA (GIOP 1.1),
fcgi
- Fast CGI,
tpkt
- TPKT format [RFC1006],
line
- Line mode, a packet is a line
terminated with newline, lines longer than
the receive buffer are truncated.
http | http_bin
The Hypertext Transfer Protocol. The packets
are returned with the format according to HttpPacket
described in
erlang:decode_packet/3. A socket in passive
mode will return {ok, HttpPacket}
from gen_tcp:recv
while an active socket will send messages like {http,
Socket, HttpPacket}
.
httph | httph_bin
These two types are often not needed as the socket will
automatically switch from http
/http_bin
to
httph
/httph_bin
internally after the first line
has been read. There might be occasions however when they are
useful, such as parsing trailers from chunked encoding.
{packet_size, Integer}
(TCP/IP sockets)Sets the max allowed length of the packet body. If the packet header indicates that the length of the packet is longer than the max allowed length, the packet is considered invalid. The same happens if the packet header is too big for the socket receive buffer.
For line oriented protocols (line
,http*
),
option packet_size
also guarantees that lines up to the
indicated length are accepted and not considered invalid due
to internal buffer limitations.
{priority, Priority}
Set the protocol-defined priority for all packets to be sent on this socket.
{raw, Protocol, OptionNum, ValueBin}
See below.
{read_packets, Integer}
(UDP sockets)Sets the max number of UDP packets to read without intervention from the socket when data is available. When this many packets have been read and delivered to the destination process, new packets are not read until a new notification of available data has arrived. The default is 5, and if this parameter is set too high the system can become unresponsive due to UDP packet flooding.
{recbuf, Size}
Gives the size of the receive buffer to use for the socket.
{reuseaddr, Boolean}
Allows or disallows local reuse of port numbers. By default, reuse is disallowed.
{send_timeout, Integer}
Only allowed for connection oriented sockets.
Specifies a longest time to wait for a send operation to
be accepted by the underlying TCP stack. When the limit is
exceeded, the send operation will return
{error,timeout}
. How much of a packet that actually
got sent is unknown, why the socket should be closed
whenever a timeout has occurred (see send_timeout_close
).
Default is infinity
.
{send_timeout_close, Boolean}
Only allowed for connection oriented sockets.
Used together with send_timeout
to specify whether
the socket will be automatically closed when the send operation
returns {error,timeout}
. The recommended setting is
true
which will automatically close the socket.
Default is false
due to backward compatibility.
{sndbuf, Size}
Gives the size of the send buffer to use for the socket.
{priority, Integer}
Sets the SO_PRIORITY socket level option on platforms where this is implemented. The behaviour and allowed range varies on different systems. The option is ignored on platforms where the option is not implemented. Use with caution.
{tos, Integer}
Sets IP_TOS IP level options on platforms where this is implemented. The behaviour and allowed range varies on different systems. The option is ignored on platforms where the option is not implemented. Use with caution.
In addition to the options mentioned above, raw
option specifications can be used. The raw options are
specified as a tuple of arity four, beginning with the tag
raw
, followed by the protocol level, the option number
and the actual option value specified as a binary. This
corresponds to the second, third and fourth argument to the
setsockopt
call in the C socket API. The option value
needs to be coded in the native endianess of the platform and,
if a structure is required, needs to follow the struct
alignment conventions on the specific platform.
Using raw socket options require detailed knowledge about the current operating system and TCP stack.
As an example of the usage of raw options, consider a Linux
system where you want to set the TCP_LINGER2
option on
the IPPROTO_TCP
protocol level in the stack. You know
that on this particular system it defaults to 60 (seconds),
but you would like to lower it to 30 for a particular
socket. The TCP_LINGER2
option is not explicitly
supported by inet, but you know that the protocol level
translates to the number 6, the option number to the number 8
and the value is to be given as a 32 bit integer. You can use
this line of code to set the option for the socket named
Sock
:
inet:setopts(Sock,[{raw,6,8,<<30:32/native>>}]),
As many options are silently discarded by the stack if they are given out of range, it could be a good idea to check that a raw option really got accepted. This code places the value in the variable TcpLinger2:
{ok,[{raw,6,8,<<TcpLinger2:32/native>>}]}=inet:getopts(Sock,[{raw,6,8,4}]),
Code such as the examples above is inherently non portable, even different versions of the same OS on the same platform may respond differently to this kind of option manipulation. Use with care.
Note that the default options for TCP/IP sockets can be changed with the Kernel configuration parameters mentioned in the beginning of this document.
POSIX Error Codes
e2big
- argument list too longeacces
- permission deniedeaddrinuse
- address already in useeaddrnotavail
- cannot assign requested addresseadv
- advertise erroreafnosupport
- address family not supported by protocol familyeagain
- resource temporarily unavailableealign
- EALIGNealready
- operation already in progressebade
- bad exchange descriptorebadf
- bad file numberebadfd
- file descriptor in bad stateebadmsg
- not a data messageebadr
- bad request descriptorebadrpc
- RPC structure is badebadrqc
- bad request codeebadslt
- invalid slotebfont
- bad font file formatebusy
- file busyechild
- no childrenechrng
- channel number out of rangeecomm
- communication error on sendeconnaborted
- software caused connection aborteconnrefused
- connection refusedeconnreset
- connection reset by peeredeadlk
- resource deadlock avoidededeadlock
- resource deadlock avoidededestaddrreq
- destination address requirededirty
- mounting a dirty fs w/o forceedom
- math argument out of rangeedotdot
- cross mount pointedquot
- disk quota exceedededuppkg
- duplicate package nameeexist
- file already existsefault
- bad address in system call argumentefbig
- file too largeehostdown
- host is downehostunreach
- host is unreachableeidrm
- identifier removedeinit
- initialization erroreinprogress
- operation now in progresseintr
- interrupted system calleinval
- invalid argumenteio
- I/O erroreisconn
- socket is already connectedeisdir
- illegal operation on a directoryeisnam
- is a named fileel2hlt
- level 2 haltedel2nsync
- level 2 not synchronizedel3hlt
- level 3 haltedel3rst
- level 3 resetelbin
- ELBINelibacc
- cannot access a needed shared libraryelibbad
- accessing a corrupted shared libraryelibexec
- cannot exec a shared library directlyelibmax
- attempting to link in more shared libraries than system limitelibscn
- .lib section in a.out corruptedelnrng
- link number out of rangeeloop
- too many levels of symbolic linksemfile
- too many open filesemlink
- too many linksemsgsize
- message too longemultihop
- multihop attemptedenametoolong
- file name too longenavail
- not availableenet
- ENETenetdown
- network is downenetreset
- network dropped connection on resetenetunreach
- network is unreachableenfile
- file table overflowenoano
- anode table overflowenobufs
- no buffer space availableenocsi
- no CSI structure availableenodata
- no data availableenodev
- no such deviceenoent
- no such file or directoryenoexec
- exec format errorenolck
- no locks availableenolink
- link has be severedenomem
- not enough memoryenomsg
- no message of desired typeenonet
- machine is not on the networkenopkg
- package not installedenoprotoopt
- bad protocol optionenospc
- no space left on deviceenosr
- out of stream resources or not a stream deviceenosym
- unresolved symbol nameenosys
- function not implementedenotblk
- block device requiredenotconn
- socket is not connectedenotdir
- not a directoryenotempty
- directory not emptyenotnam
- not a named fileenotsock
- socket operation on non-socketenotsup
- operation not supportedenotty
- inappropriate device for ioctlenotuniq
- name not unique on networkenxio
- no such device or addresseopnotsupp
- operation not supported on socketeperm
- not ownerepfnosupport
- protocol family not supportedepipe
- broken pipeeproclim
- too many processeseprocunavail
- bad procedure for programeprogmismatch
- program version wrongeprogunavail
- RPC program not availableeproto
- protocol erroreprotonosupport
- protocol not supportedeprototype
- protocol wrong type for socketerange
- math result unrepresentableerefused
- EREFUSEDeremchg
- remote address changederemdev
- remote deviceeremote
- pathname hit remote file systemeremoteio
- remote i/o erroreremoterelease
- EREMOTERELEASEerofs
- read-only file systemerpcmismatch
- RPC version is wrongerremote
- object is remoteeshutdown
- cannot send after socket shutdownesocktnosupport
- socket type not supportedespipe
- invalid seekesrch
- no such processesrmnt
- srmount errorestale
- stale remote file handleesuccess
- Error 0etime
- timer expiredetimedout
- connection timed outetoomanyrefs
- too many referencesetxtbsy
- text file or pseudo-device busyeuclean
- structure needs cleaningeunatch
- protocol driver not attachedeusers
- too many userseversion
- version mismatchewouldblock
- operation would blockexdev
- cross-domain linkexfull
- message tables fullnxdomain
- the hostname or domain name could not be found