1 Documentation for /proc/sys/net/*
2 (c) 1999 Terrehon Bowden <terrehon@pacbell.net>
3 Bodo Bauer <bb@ricochet.net>
4 (c) 2000 Jorge Nerin <comandante@zaralinux.com>
5 (c) 2009 Shen Feng <shen@cn.fujitsu.com>
7 For general info and legal blurb, please look in README.
9 ==============================================================
11 This file contains the documentation for the sysctl files in
14 The interface to the networking parts of the kernel is located in
15 /proc/sys/net. The following table shows all possible subdirectories. You may
16 see only some of them, depending on your kernel's configuration.
19 Table : Subdirectories in /proc/sys/net
20 ..............................................................................
21 Directory Content Directory Content
22 core General parameter appletalk Appletalk protocol
23 unix Unix domain sockets netrom NET/ROM
24 802 E802 protocol ax25 AX25
25 ethernet Ethernet protocol rose X.25 PLP layer
26 ipv4 IP version 4 x25 X.25 protocol
27 ipx IPX token-ring IBM token ring
28 bridge Bridging decnet DEC net
29 ipv6 IP version 6 tipc TIPC
30 ..............................................................................
32 1. /proc/sys/net/core - Network core options
33 -------------------------------------------------------
38 This enables Berkeley Packet Filter Just in Time compiler.
39 Currently supported on x86_64 architecture, bpf_jit provides a framework
40 to speed packet filtering, the one used by tcpdump/libpcap for example.
42 0 - disable the JIT (default value)
44 2 - enable the JIT and ask the compiler to emit traces on kernel log.
49 The maximum number of packets that kernel can handle on a NAPI interrupt,
50 it's a Per-CPU variable.
55 Low latency busy poll timeout for socket reads. (needs CONFIG_NET_RX_BUSY_POLL)
56 Approximate time in us to busy loop waiting for packets on the device queue.
57 This sets the default value of the SO_BUSY_POLL socket option.
58 Can be set or overridden per socket by setting socket option SO_BUSY_POLL,
59 which is the preferred method of enabling. If you need to enable the feature
60 globally via sysctl, a value of 50 is recommended.
61 Will increase power usage.
66 Low latency busy poll timeout for poll and select. (needs CONFIG_NET_RX_BUSY_POLL)
67 Approximate time in us to busy loop waiting for events.
68 Recommended value depends on the number of sockets you poll on.
69 For several sockets 50, for several hundreds 100.
70 For more than that you probably want to use epoll.
71 Note that only sockets with SO_BUSY_POLL set will be busy polled,
72 so you want to either selectively set SO_BUSY_POLL on those sockets or set
73 sysctl.net.busy_read globally.
74 Will increase power usage.
80 The default setting of the socket receive buffer in bytes.
85 The maximum receive socket buffer size in bytes.
90 The default setting (in bytes) of the socket send buffer.
95 The maximum send socket buffer size in bytes.
97 message_burst and message_cost
98 ------------------------------
100 These parameters are used to limit the warning messages written to the kernel
101 log from the networking code. They enforce a rate limit to make a
102 denial-of-service attack impossible. A higher message_cost factor, results in
103 fewer messages that will be written. Message_burst controls when messages will
104 be dropped. The default settings limit warning messages to one every five
110 This controls console messages from the networking stack that can occur because
111 of problems on the network like duplicate address or bad checksums. Normally,
112 this should be enabled, but if the problem persists the messages can be
118 Maximum number of packets taken from all interfaces in one polling cycle (NAPI
119 poll). In one polling cycle interfaces which are registered to polling are
120 probed in a round-robin manner.
125 Maximum number of packets, queued on the INPUT side, when the interface
126 receives packets faster than kernel can process them.
128 netdev_tstamp_prequeue
129 ----------------------
131 If set to 0, RX packet timestamps can be sampled after RPS processing, when
132 the target CPU processes packets. It might give some delay on timestamps, but
133 permit to distribute the load on several cpus.
135 If set to 1 (default), timestamps are sampled as soon as possible, before
141 Maximum ancillary buffer size allowed per socket. Ancillary data is a sequence
142 of struct cmsghdr structures with appended data.
144 2. /proc/sys/net/unix - Parameters for Unix domain sockets
145 -------------------------------------------------------
147 There is only one file in this directory.
148 unix_dgram_qlen limits the max number of datagrams queued in Unix domain
149 socket's buffer. It will not take effect unless PF_UNIX flag is specified.
152 3. /proc/sys/net/ipv4 - IPV4 settings
153 -------------------------------------------------------
154 Please see: Documentation/networking/ip-sysctl.txt and ipvs-sysctl.txt for
155 descriptions of these entries.
159 -------------------------------------------------------
161 The /proc/sys/net/appletalk directory holds the Appletalk configuration data
162 when Appletalk is loaded. The configurable parameters are:
167 The amount of time we keep an ARP entry before expiring it. Used to age out
173 The amount of time we will spend trying to resolve an Appletalk address.
175 aarp-retransmit-limit
176 ---------------------
178 The number of times we will retransmit a query before giving up.
183 Controls the rate at which expires are checked.
185 The directory /proc/net/appletalk holds the list of active Appletalk sockets
188 The fields indicate the DDP type, the local address (in network:node format)
189 the remote address, the size of the transmit pending queue, the size of the
190 received queue (bytes waiting for applications to read) the state and the uid
193 /proc/net/atalk_iface lists all the interfaces configured for appletalk.It
194 shows the name of the interface, its Appletalk address, the network range on
195 that address (or network number for phase 1 networks), and the status of the
198 /proc/net/atalk_route lists each known network route. It lists the target
199 (network) that the route leads to, the router (may be directly connected), the
200 route flags, and the device the route is using.
204 -------------------------------------------------------
206 The IPX protocol has no tunable values in proc/sys/net.
208 The IPX protocol does, however, provide proc/net/ipx. This lists each IPX
209 socket giving the local and remote addresses in Novell format (that is
210 network:node:port). In accordance with the strange Novell tradition,
211 everything but the port is in hex. Not_Connected is displayed for sockets that
212 are not tied to a specific remote address. The Tx and Rx queue sizes indicate
213 the number of bytes pending for transmission and reception. The state
214 indicates the state the socket is in and the uid is the owning uid of the
217 The /proc/net/ipx_interface file lists all IPX interfaces. For each interface
218 it gives the network number, the node number, and indicates if the network is
219 the primary network. It also indicates which device it is bound to (or
220 Internal for internal networks) and the Frame Type if appropriate. Linux
221 supports 802.3, 802.2, 802.2 SNAP and DIX (Blue Book) ethernet framing for
224 The /proc/net/ipx_route table holds a list of IPX routes. For each route it
225 gives the destination network, the router node (or Directly) and the network
226 address of the router (or Connected) for internal networks.
229 -------------------------------------------------------
231 The TIPC protocol now has a tunable for the receive memory, similar to the
232 tcp_rmem - i.e. a vector of 3 INTEGERs: (min, default, max)
234 # cat /proc/sys/net/tipc/tipc_rmem
235 4252725 34021800 68043600
238 The max value is set to CONN_OVERLOAD_LIMIT, and the default and min values
239 are scaled (shifted) versions of that same value. Note that the min value
240 is not at this point in time used in any meaningful way, but the triplet is
241 preserved in order to be consistent with things like tcp_rmem.