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 This enforces a global limit for memory allocations to the BPF JIT
61 compiler in order to reject unprivileged JIT requests once it has
62 been surpassed. bpf_jit_limit contains the value of the global limit
68 The maximum number of packets that kernel can handle on a NAPI interrupt,
69 it's a Per-CPU variable.
75 The default queuing discipline to use for network devices. This allows
76 overriding the default of pfifo_fast with an alternative. Since the default
77 queuing discipline is created without additional parameters so is best suited
78 to queuing disciplines that work well without configuration like stochastic
79 fair queue (sfq), CoDel (codel) or fair queue CoDel (fq_codel). Don't use
80 queuing disciplines like Hierarchical Token Bucket or Deficit Round Robin
81 which require setting up classes and bandwidths. Note that physical multiqueue
82 interfaces still use mq as root qdisc, which in turn uses this default for its
83 leaves. Virtual devices (like e.g. lo or veth) ignore this setting and instead
89 Low latency busy poll timeout for socket reads. (needs CONFIG_NET_RX_BUSY_POLL)
90 Approximate time in us to busy loop waiting for packets on the device queue.
91 This sets the default value of the SO_BUSY_POLL socket option.
92 Can be set or overridden per socket by setting socket option SO_BUSY_POLL,
93 which is the preferred method of enabling. If you need to enable the feature
94 globally via sysctl, a value of 50 is recommended.
95 Will increase power usage.
100 Low latency busy poll timeout for poll and select. (needs CONFIG_NET_RX_BUSY_POLL)
101 Approximate time in us to busy loop waiting for events.
102 Recommended value depends on the number of sockets you poll on.
103 For several sockets 50, for several hundreds 100.
104 For more than that you probably want to use epoll.
105 Note that only sockets with SO_BUSY_POLL set will be busy polled,
106 so you want to either selectively set SO_BUSY_POLL on those sockets or set
107 sysctl.net.busy_read globally.
108 Will increase power usage.
114 The default setting of the socket receive buffer in bytes.
119 The maximum receive socket buffer size in bytes.
123 Allow processes to receive tx timestamps looped together with the original
124 packet contents. If disabled, transmit timestamp requests from unprivileged
125 processes are dropped unless socket option SOF_TIMESTAMPING_OPT_TSONLY is set.
132 The default setting (in bytes) of the socket send buffer.
137 The maximum send socket buffer size in bytes.
139 message_burst and message_cost
140 ------------------------------
142 These parameters are used to limit the warning messages written to the kernel
143 log from the networking code. They enforce a rate limit to make a
144 denial-of-service attack impossible. A higher message_cost factor, results in
145 fewer messages that will be written. Message_burst controls when messages will
146 be dropped. The default settings limit warning messages to one every five
152 This sysctl is now unused.
154 This was used to control console messages from the networking stack that
155 occur because of problems on the network like duplicate address or bad
158 These messages are now emitted at KERN_DEBUG and can generally be enabled
159 and controlled by the dynamic_debug facility.
164 Maximum number of packets taken from all interfaces in one polling cycle (NAPI
165 poll). In one polling cycle interfaces which are registered to polling are
166 probed in a round-robin manner.
171 Maximum number of packets, queued on the INPUT side, when the interface
172 receives packets faster than kernel can process them.
177 RSS (Receive Side Scaling) enabled drivers use a 40 bytes host key that is
179 Some user space might need to gather its content even if drivers do not
180 provide ethtool -x support yet.
182 myhost:~# cat /proc/sys/net/core/netdev_rss_key
183 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)
185 File contains nul bytes if no driver ever called netdev_rss_key_fill() function.
187 /proc/sys/net/core/netdev_rss_key contains 52 bytes of key,
188 but most drivers only use 40 bytes of it.
190 myhost:~# ethtool -x eth0
191 RX flow hash indirection table for eth0 with 8 RX ring(s):
194 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
196 netdev_tstamp_prequeue
197 ----------------------
199 If set to 0, RX packet timestamps can be sampled after RPS processing, when
200 the target CPU processes packets. It might give some delay on timestamps, but
201 permit to distribute the load on several cpus.
203 If set to 1 (default), timestamps are sampled as soon as possible, before
209 Maximum ancillary buffer size allowed per socket. Ancillary data is a sequence
210 of struct cmsghdr structures with appended data.
212 2. /proc/sys/net/unix - Parameters for Unix domain sockets
213 -------------------------------------------------------
215 There is only one file in this directory.
216 unix_dgram_qlen limits the max number of datagrams queued in Unix domain
217 socket's buffer. It will not take effect unless PF_UNIX flag is specified.
220 3. /proc/sys/net/ipv4 - IPV4 settings
221 -------------------------------------------------------
222 Please see: Documentation/networking/ip-sysctl.txt and ipvs-sysctl.txt for
223 descriptions of these entries.
227 -------------------------------------------------------
229 The /proc/sys/net/appletalk directory holds the Appletalk configuration data
230 when Appletalk is loaded. The configurable parameters are:
235 The amount of time we keep an ARP entry before expiring it. Used to age out
241 The amount of time we will spend trying to resolve an Appletalk address.
243 aarp-retransmit-limit
244 ---------------------
246 The number of times we will retransmit a query before giving up.
251 Controls the rate at which expires are checked.
253 The directory /proc/net/appletalk holds the list of active Appletalk sockets
256 The fields indicate the DDP type, the local address (in network:node format)
257 the remote address, the size of the transmit pending queue, the size of the
258 received queue (bytes waiting for applications to read) the state and the uid
261 /proc/net/atalk_iface lists all the interfaces configured for appletalk.It
262 shows the name of the interface, its Appletalk address, the network range on
263 that address (or network number for phase 1 networks), and the status of the
266 /proc/net/atalk_route lists each known network route. It lists the target
267 (network) that the route leads to, the router (may be directly connected), the
268 route flags, and the device the route is using.
272 -------------------------------------------------------
274 The IPX protocol has no tunable values in proc/sys/net.
276 The IPX protocol does, however, provide proc/net/ipx. This lists each IPX
277 socket giving the local and remote addresses in Novell format (that is
278 network:node:port). In accordance with the strange Novell tradition,
279 everything but the port is in hex. Not_Connected is displayed for sockets that
280 are not tied to a specific remote address. The Tx and Rx queue sizes indicate
281 the number of bytes pending for transmission and reception. The state
282 indicates the state the socket is in and the uid is the owning uid of the
285 The /proc/net/ipx_interface file lists all IPX interfaces. For each interface
286 it gives the network number, the node number, and indicates if the network is
287 the primary network. It also indicates which device it is bound to (or
288 Internal for internal networks) and the Frame Type if appropriate. Linux
289 supports 802.3, 802.2, 802.2 SNAP and DIX (Blue Book) ethernet framing for
292 The /proc/net/ipx_route table holds a list of IPX routes. For each route it
293 gives the destination network, the router node (or Directly) and the network
294 address of the router (or Connected) for internal networks.
297 -------------------------------------------------------
302 The TIPC protocol now has a tunable for the receive memory, similar to the
303 tcp_rmem - i.e. a vector of 3 INTEGERs: (min, default, max)
305 # cat /proc/sys/net/tipc/tipc_rmem
306 4252725 34021800 68043600
309 The max value is set to CONN_OVERLOAD_LIMIT, and the default and min values
310 are scaled (shifted) versions of that same value. Note that the min value
311 is not at this point in time used in any meaningful way, but the triplet is
312 preserved in order to be consistent with things like tcp_rmem.
317 TIPC name table updates are distributed asynchronously in a cluster, without
318 any form of transaction handling. This means that different race scenarios are
319 possible. One such is that a name withdrawal sent out by one node and received
320 by another node may arrive after a second, overlapping name publication already
321 has been accepted from a third node, although the conflicting updates
322 originally may have been issued in the correct sequential order.
323 If named_timeout is nonzero, failed topology updates will be placed on a defer
324 queue until another event arrives that clears the error, or until the timeout
325 expires. Value is in milliseconds.