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 This enables hardening for the Berkeley Packet Filter Just in Time compiler.
50 Supported are eBPF JIT backends. Enabling hardening trades off performance,
51 but can mitigate JIT spraying.
53 0 - disable JIT hardening (default value)
54 1 - enable JIT hardening for unprivileged users only
55 2 - enable JIT hardening for all users
60 The maximum number of packets that kernel can handle on a NAPI interrupt,
61 it's a Per-CPU variable.
67 The default queuing discipline to use for network devices. This allows
68 overriding the default of pfifo_fast with an alternative. Since the default
69 queuing discipline is created without additional parameters so is best suited
70 to queuing disciplines that work well without configuration like stochastic
71 fair queue (sfq), CoDel (codel) or fair queue CoDel (fq_codel). Don't use
72 queuing disciplines like Hierarchical Token Bucket or Deficit Round Robin
73 which require setting up classes and bandwidths. Note that physical multiqueue
74 interfaces still use mq as root qdisc, which in turn uses this default for its
75 leaves. Virtual devices (like e.g. lo or veth) ignore this setting and instead
81 Low latency busy poll timeout for socket reads. (needs CONFIG_NET_RX_BUSY_POLL)
82 Approximate time in us to busy loop waiting for packets on the device queue.
83 This sets the default value of the SO_BUSY_POLL socket option.
84 Can be set or overridden per socket by setting socket option SO_BUSY_POLL,
85 which is the preferred method of enabling. If you need to enable the feature
86 globally via sysctl, a value of 50 is recommended.
87 Will increase power usage.
92 Low latency busy poll timeout for poll and select. (needs CONFIG_NET_RX_BUSY_POLL)
93 Approximate time in us to busy loop waiting for events.
94 Recommended value depends on the number of sockets you poll on.
95 For several sockets 50, for several hundreds 100.
96 For more than that you probably want to use epoll.
97 Note that only sockets with SO_BUSY_POLL set will be busy polled,
98 so you want to either selectively set SO_BUSY_POLL on those sockets or set
99 sysctl.net.busy_read globally.
100 Will increase power usage.
106 The default setting of the socket receive buffer in bytes.
111 The maximum receive socket buffer size in bytes.
115 Allow processes to receive tx timestamps looped together with the original
116 packet contents. If disabled, transmit timestamp requests from unprivileged
117 processes are dropped unless socket option SOF_TIMESTAMPING_OPT_TSONLY is set.
124 The default setting (in bytes) of the socket send buffer.
129 The maximum send socket buffer size in bytes.
131 message_burst and message_cost
132 ------------------------------
134 These parameters are used to limit the warning messages written to the kernel
135 log from the networking code. They enforce a rate limit to make a
136 denial-of-service attack impossible. A higher message_cost factor, results in
137 fewer messages that will be written. Message_burst controls when messages will
138 be dropped. The default settings limit warning messages to one every five
144 This sysctl is now unused.
146 This was used to control console messages from the networking stack that
147 occur because of problems on the network like duplicate address or bad
150 These messages are now emitted at KERN_DEBUG and can generally be enabled
151 and controlled by the dynamic_debug facility.
156 Maximum number of packets taken from all interfaces in one polling cycle (NAPI
157 poll). In one polling cycle interfaces which are registered to polling are
158 probed in a round-robin manner.
163 Maximum number of packets, queued on the INPUT side, when the interface
164 receives packets faster than kernel can process them.
169 RSS (Receive Side Scaling) enabled drivers use a 40 bytes host key that is
171 Some user space might need to gather its content even if drivers do not
172 provide ethtool -x support yet.
174 myhost:~# cat /proc/sys/net/core/netdev_rss_key
175 84:50:f4:00:a8:15:d1:a7:e9:7f:1d:60:35:c7:47:25:42:97:74:ca:56:bb:b6:a1:d8: ... (52 bytes total)
177 File contains nul bytes if no driver ever called netdev_rss_key_fill() function.
179 /proc/sys/net/core/netdev_rss_key contains 52 bytes of key,
180 but most drivers only use 40 bytes of it.
182 myhost:~# ethtool -x eth0
183 RX flow hash indirection table for eth0 with 8 RX ring(s):
186 84:50:f4:00:a8:15:d1:a7:e9:7f:1d:60:35:c7:47:25:42:97:74:ca:56:bb:b6:a1:d8:43:e3:c9:0c:fd:17:55:c2:3a:4d:69:ed:f1:42:89
188 netdev_tstamp_prequeue
189 ----------------------
191 If set to 0, RX packet timestamps can be sampled after RPS processing, when
192 the target CPU processes packets. It might give some delay on timestamps, but
193 permit to distribute the load on several cpus.
195 If set to 1 (default), timestamps are sampled as soon as possible, before
201 Maximum ancillary buffer size allowed per socket. Ancillary data is a sequence
202 of struct cmsghdr structures with appended data.
204 2. /proc/sys/net/unix - Parameters for Unix domain sockets
205 -------------------------------------------------------
207 There is only one file in this directory.
208 unix_dgram_qlen limits the max number of datagrams queued in Unix domain
209 socket's buffer. It will not take effect unless PF_UNIX flag is specified.
212 3. /proc/sys/net/ipv4 - IPV4 settings
213 -------------------------------------------------------
214 Please see: Documentation/networking/ip-sysctl.txt and ipvs-sysctl.txt for
215 descriptions of these entries.
219 -------------------------------------------------------
221 The /proc/sys/net/appletalk directory holds the Appletalk configuration data
222 when Appletalk is loaded. The configurable parameters are:
227 The amount of time we keep an ARP entry before expiring it. Used to age out
233 The amount of time we will spend trying to resolve an Appletalk address.
235 aarp-retransmit-limit
236 ---------------------
238 The number of times we will retransmit a query before giving up.
243 Controls the rate at which expires are checked.
245 The directory /proc/net/appletalk holds the list of active Appletalk sockets
248 The fields indicate the DDP type, the local address (in network:node format)
249 the remote address, the size of the transmit pending queue, the size of the
250 received queue (bytes waiting for applications to read) the state and the uid
253 /proc/net/atalk_iface lists all the interfaces configured for appletalk.It
254 shows the name of the interface, its Appletalk address, the network range on
255 that address (or network number for phase 1 networks), and the status of the
258 /proc/net/atalk_route lists each known network route. It lists the target
259 (network) that the route leads to, the router (may be directly connected), the
260 route flags, and the device the route is using.
264 -------------------------------------------------------
266 The IPX protocol has no tunable values in proc/sys/net.
268 The IPX protocol does, however, provide proc/net/ipx. This lists each IPX
269 socket giving the local and remote addresses in Novell format (that is
270 network:node:port). In accordance with the strange Novell tradition,
271 everything but the port is in hex. Not_Connected is displayed for sockets that
272 are not tied to a specific remote address. The Tx and Rx queue sizes indicate
273 the number of bytes pending for transmission and reception. The state
274 indicates the state the socket is in and the uid is the owning uid of the
277 The /proc/net/ipx_interface file lists all IPX interfaces. For each interface
278 it gives the network number, the node number, and indicates if the network is
279 the primary network. It also indicates which device it is bound to (or
280 Internal for internal networks) and the Frame Type if appropriate. Linux
281 supports 802.3, 802.2, 802.2 SNAP and DIX (Blue Book) ethernet framing for
284 The /proc/net/ipx_route table holds a list of IPX routes. For each route it
285 gives the destination network, the router node (or Directly) and the network
286 address of the router (or Connected) for internal networks.
289 -------------------------------------------------------
294 The TIPC protocol now has a tunable for the receive memory, similar to the
295 tcp_rmem - i.e. a vector of 3 INTEGERs: (min, default, max)
297 # cat /proc/sys/net/tipc/tipc_rmem
298 4252725 34021800 68043600
301 The max value is set to CONN_OVERLOAD_LIMIT, and the default and min values
302 are scaled (shifted) versions of that same value. Note that the min value
303 is not at this point in time used in any meaningful way, but the triplet is
304 preserved in order to be consistent with things like tcp_rmem.
309 TIPC name table updates are distributed asynchronously in a cluster, without
310 any form of transaction handling. This means that different race scenarios are
311 possible. One such is that a name withdrawal sent out by one node and received
312 by another node may arrive after a second, overlapping name publication already
313 has been accepted from a third node, although the conflicting updates
314 originally may have been issued in the correct sequential order.
315 If named_timeout is nonzero, failed topology updates will be placed on a defer
316 queue until another event arrives that clears the error, or until the timeout
317 expires. Value is in milliseconds.