1 ================================
2 Documentation for /proc/sys/net/
3 ================================
20 For general info and legal blurb, please look in index.rst.
22 ------------------------------------------------------------------------------
24 This file contains the documentation for the sysctl files in
27 The interface to the networking parts of the kernel is located in
28 /proc/sys/net. The following table shows all possible subdirectories. You may
29 see only some of them, depending on your kernel's configuration.
32 Table : Subdirectories in /proc/sys/net
34 ========= =================== = ========== ===================
35 Directory Content Directory Content
36 ========= =================== = ========== ===================
37 802 E802 protocol mptcp Multipath TCP
38 appletalk Appletalk protocol netfilter Network Filter
39 ax25 AX25 netrom NET/ROM
40 bridge Bridging rose X.25 PLP layer
41 core General parameter tipc TIPC
42 ethernet Ethernet protocol unix Unix domain sockets
43 ipv4 IP version 4 x25 X.25 protocol
45 ========= =================== = ========== ===================
47 1. /proc/sys/net/core - Network core options
48 ============================================
53 This enables the BPF Just in Time (JIT) compiler. BPF is a flexible
54 and efficient infrastructure allowing to execute bytecode at various
55 hook points. It is used in a number of Linux kernel subsystems such
56 as networking (e.g. XDP, tc), tracing (e.g. kprobes, uprobes, tracepoints)
57 and security (e.g. seccomp). LLVM has a BPF back end that can compile
58 restricted C into a sequence of BPF instructions. After program load
59 through bpf(2) and passing a verifier in the kernel, a JIT will then
60 translate these BPF proglets into native CPU instructions. There are
61 two flavors of JITs, the newer eBPF JIT currently supported on:
75 And the older cBPF JIT supported on the following archs:
80 eBPF JITs are a superset of cBPF JITs, meaning the kernel will
81 migrate cBPF instructions into eBPF instructions and then JIT
82 compile them transparently. Older cBPF JITs can only translate
83 tcpdump filters, seccomp rules, etc, but not mentioned eBPF
84 programs loaded through bpf(2).
88 - 0 - disable the JIT (default value)
90 - 2 - enable the JIT and ask the compiler to emit traces on kernel log.
95 This enables hardening for the BPF JIT compiler. Supported are eBPF
96 JIT backends. Enabling hardening trades off performance, but can
97 mitigate JIT spraying.
101 - 0 - disable JIT hardening (default value)
102 - 1 - enable JIT hardening for unprivileged users only
103 - 2 - enable JIT hardening for all users
105 where "privileged user" in this context means a process having
106 CAP_BPF or CAP_SYS_ADMIN in the root user name space.
111 When BPF JIT compiler is enabled, then compiled images are unknown
112 addresses to the kernel, meaning they neither show up in traces nor
113 in /proc/kallsyms. This enables export of these addresses, which can
114 be used for debugging/tracing. If bpf_jit_harden is enabled, this
119 - 0 - disable JIT kallsyms export (default value)
120 - 1 - enable JIT kallsyms export for privileged users only
125 This enforces a global limit for memory allocations to the BPF JIT
126 compiler in order to reject unprivileged JIT requests once it has
127 been surpassed. bpf_jit_limit contains the value of the global limit
133 The maximum number of packets that kernel can handle on a NAPI interrupt,
134 it's a Per-CPU variable. For drivers that support LRO or GRO_HW, a hardware
135 aggregated packet is counted as one packet in this context.
142 RPS (e.g. RFS, aRFS) processing is competing with the registered NAPI poll function
143 of the driver for the per softirq cycle netdev_budget. This parameter influences
144 the proportion of the configured netdev_budget that is spent on RPS based packet
145 processing during RX softirq cycles. It is further meant for making current
146 dev_weight adaptable for asymmetric CPU needs on RX/TX side of the network stack.
147 (see dev_weight_tx_bias) It is effective on a per CPU basis. Determination is based
148 on dev_weight and is calculated multiplicative (dev_weight * dev_weight_rx_bias).
155 Scales the maximum number of packets that can be processed during a TX softirq cycle.
156 Effective on a per CPU basis. Allows scaling of current dev_weight for asymmetric
157 net stack processing needs. Be careful to avoid making TX softirq processing a CPU hog.
159 Calculation is based on dev_weight (dev_weight * dev_weight_tx_bias).
166 The default queuing discipline to use for network devices. This allows
167 overriding the default of pfifo_fast with an alternative. Since the default
168 queuing discipline is created without additional parameters so is best suited
169 to queuing disciplines that work well without configuration like stochastic
170 fair queue (sfq), CoDel (codel) or fair queue CoDel (fq_codel). Don't use
171 queuing disciplines like Hierarchical Token Bucket or Deficit Round Robin
172 which require setting up classes and bandwidths. Note that physical multiqueue
173 interfaces still use mq as root qdisc, which in turn uses this default for its
174 leaves. Virtual devices (like e.g. lo or veth) ignore this setting and instead
182 Low latency busy poll timeout for socket reads. (needs CONFIG_NET_RX_BUSY_POLL)
183 Approximate time in us to busy loop waiting for packets on the device queue.
184 This sets the default value of the SO_BUSY_POLL socket option.
185 Can be set or overridden per socket by setting socket option SO_BUSY_POLL,
186 which is the preferred method of enabling. If you need to enable the feature
187 globally via sysctl, a value of 50 is recommended.
189 Will increase power usage.
195 Low latency busy poll timeout for poll and select. (needs CONFIG_NET_RX_BUSY_POLL)
196 Approximate time in us to busy loop waiting for events.
197 Recommended value depends on the number of sockets you poll on.
198 For several sockets 50, for several hundreds 100.
199 For more than that you probably want to use epoll.
200 Note that only sockets with SO_BUSY_POLL set will be busy polled,
201 so you want to either selectively set SO_BUSY_POLL on those sockets or set
202 sysctl.net.busy_read globally.
204 Will increase power usage.
211 The default setting of the socket receive buffer in bytes.
216 The maximum receive socket buffer size in bytes.
220 Allow processes to receive tx timestamps looped together with the original
221 packet contents. If disabled, transmit timestamp requests from unprivileged
222 processes are dropped unless socket option SOF_TIMESTAMPING_OPT_TSONLY is set.
230 The default setting (in bytes) of the socket send buffer.
235 The maximum send socket buffer size in bytes.
237 message_burst and message_cost
238 ------------------------------
240 These parameters are used to limit the warning messages written to the kernel
241 log from the networking code. They enforce a rate limit to make a
242 denial-of-service attack impossible. A higher message_cost factor, results in
243 fewer messages that will be written. Message_burst controls when messages will
244 be dropped. The default settings limit warning messages to one every five
250 This sysctl is now unused.
252 This was used to control console messages from the networking stack that
253 occur because of problems on the network like duplicate address or bad
256 These messages are now emitted at KERN_DEBUG and can generally be enabled
257 and controlled by the dynamic_debug facility.
262 Maximum number of packets taken from all interfaces in one polling cycle (NAPI
263 poll). In one polling cycle interfaces which are registered to polling are
264 probed in a round-robin manner. Also, a polling cycle may not exceed
265 netdev_budget_usecs microseconds, even if netdev_budget has not been
269 ---------------------
271 Maximum number of microseconds in one NAPI polling cycle. Polling
272 will exit when either netdev_budget_usecs have elapsed during the
273 poll cycle or the number of packets processed reaches netdev_budget.
278 Maximum number of packets, queued on the INPUT side, when the interface
279 receives packets faster than kernel can process them.
284 RSS (Receive Side Scaling) enabled drivers use a 40 bytes host key that is
286 Some user space might need to gather its content even if drivers do not
287 provide ethtool -x support yet.
291 myhost:~# cat /proc/sys/net/core/netdev_rss_key
292 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)
294 File contains nul bytes if no driver ever called netdev_rss_key_fill() function.
297 /proc/sys/net/core/netdev_rss_key contains 52 bytes of key,
298 but most drivers only use 40 bytes of it.
302 myhost:~# ethtool -x eth0
303 RX flow hash indirection table for eth0 with 8 RX ring(s):
306 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
308 netdev_tstamp_prequeue
309 ----------------------
311 If set to 0, RX packet timestamps can be sampled after RPS processing, when
312 the target CPU processes packets. It might give some delay on timestamps, but
313 permit to distribute the load on several cpus.
315 If set to 1 (default), timestamps are sampled as soon as possible, before
318 netdev_unregister_timeout_secs
319 ------------------------------
321 Unregister network device timeout in seconds.
322 This option controls the timeout (in seconds) used to issue a warning while
323 waiting for a network device refcount to drop to 0 during device
324 unregistration. A lower value may be useful during bisection to detect
325 a leaked reference faster. A larger value may be useful to prevent false
326 warnings on slow/loaded systems.
327 Default value is 10, minimum 1, maximum 3600.
332 Max size (in skbs) of the per-cpu list of skbs being freed
333 by the cpu which allocated them. Used by TCP stack so far.
340 Maximum ancillary buffer size allowed per socket. Ancillary data is a sequence
341 of struct cmsghdr structures with appended data.
343 fb_tunnels_only_for_init_net
344 ----------------------------
346 Controls if fallback tunnels (like tunl0, gre0, gretap0, erspan0,
347 sit0, ip6tnl0, ip6gre0) are automatically created. There are 3 possibilities
348 (a) value = 0; respective fallback tunnels are created when module is
349 loaded in every net namespaces (backward compatible behavior).
350 (b) value = 1; [kcmd value: initns] respective fallback tunnels are
351 created only in init net namespace and every other net namespace will
353 (c) value = 2; [kcmd value: none] fallback tunnels are not created
354 when a module is loaded in any of the net namespace. Setting value to
355 "2" is pointless after boot if these modules are built-in, so there is
356 a kernel command-line option that can change this default. Please refer to
357 Documentation/admin-guide/kernel-parameters.txt for additional details.
359 Not creating fallback tunnels gives control to userspace to create
360 whatever is needed only and avoid creating devices which are redundant.
362 Default : 0 (for compatibility reasons)
364 devconf_inherit_init_net
365 ------------------------
367 Controls if a new network namespace should inherit all current
368 settings under /proc/sys/net/{ipv4,ipv6}/conf/{all,default}/. By
369 default, we keep the current behavior: for IPv4 we inherit all current
370 settings from init_net and for IPv6 we reset all settings to default.
372 If set to 1, both IPv4 and IPv6 settings are forced to inherit from
373 current ones in init_net. If set to 2, both IPv4 and IPv6 settings are
374 forced to reset to their default values. If set to 3, both IPv4 and IPv6
375 settings are forced to inherit from current ones in the netns where this
376 new netns has been created.
378 Default : 0 (for compatibility reasons)
383 Controls default hash rethink behaviour on listening socket when SO_TXREHASH
384 option is set to SOCK_TXREHASH_DEFAULT (i. e. not overridden by setsockopt).
386 If set to 1 (default), hash rethink is performed on listening socket.
387 If set to 0, hash rethink is not performed.
392 Maximum number of the segments to batch up on output of GRO. When a packet
393 exits GRO, either as a coalesced superframe or as an original packet which
394 GRO has decided not to coalesce, it is placed on a per-NAPI list. This
395 list is then passed to the stack when the number of segments reaches the
396 gro_normal_batch limit.
398 high_order_alloc_disable
399 ------------------------
401 By default the allocator for page frags tries to use high order pages (order-3
402 on x86). While the default behavior gives good results in most cases, some users
403 might have hit a contention in page allocations/freeing. This was especially
404 true on older kernels (< 5.14) when high-order pages were not stored on per-cpu
405 lists. This allows to opt-in for order-0 allocation instead but is now mostly of
406 historical importance.
410 2. /proc/sys/net/unix - Parameters for Unix domain sockets
411 ----------------------------------------------------------
413 There is only one file in this directory.
414 unix_dgram_qlen limits the max number of datagrams queued in Unix domain
415 socket's buffer. It will not take effect unless PF_UNIX flag is specified.
418 3. /proc/sys/net/ipv4 - IPV4 settings
419 -------------------------------------
420 Please see: Documentation/networking/ip-sysctl.rst and
421 Documentation/admin-guide/sysctl/net.rst for descriptions of these entries.
427 The /proc/sys/net/appletalk directory holds the Appletalk configuration data
428 when Appletalk is loaded. The configurable parameters are:
433 The amount of time we keep an ARP entry before expiring it. Used to age out
439 The amount of time we will spend trying to resolve an Appletalk address.
441 aarp-retransmit-limit
442 ---------------------
444 The number of times we will retransmit a query before giving up.
449 Controls the rate at which expires are checked.
451 The directory /proc/net/appletalk holds the list of active Appletalk sockets
454 The fields indicate the DDP type, the local address (in network:node format)
455 the remote address, the size of the transmit pending queue, the size of the
456 received queue (bytes waiting for applications to read) the state and the uid
459 /proc/net/atalk_iface lists all the interfaces configured for appletalk.It
460 shows the name of the interface, its Appletalk address, the network range on
461 that address (or network number for phase 1 networks), and the status of the
464 /proc/net/atalk_route lists each known network route. It lists the target
465 (network) that the route leads to, the router (may be directly connected), the
466 route flags, and the device the route is using.
474 The TIPC protocol now has a tunable for the receive memory, similar to the
475 tcp_rmem - i.e. a vector of 3 INTEGERs: (min, default, max)
479 # cat /proc/sys/net/tipc/tipc_rmem
480 4252725 34021800 68043600
483 The max value is set to CONN_OVERLOAD_LIMIT, and the default and min values
484 are scaled (shifted) versions of that same value. Note that the min value
485 is not at this point in time used in any meaningful way, but the triplet is
486 preserved in order to be consistent with things like tcp_rmem.
491 TIPC name table updates are distributed asynchronously in a cluster, without
492 any form of transaction handling. This means that different race scenarios are
493 possible. One such is that a name withdrawal sent out by one node and received
494 by another node may arrive after a second, overlapping name publication already
495 has been accepted from a third node, although the conflicting updates
496 originally may have been issued in the correct sequential order.
497 If named_timeout is nonzero, failed topology updates will be placed on a defer
498 queue until another event arrives that clears the error, or until the timeout
499 expires. Value is in milliseconds.
projects / J-linux.git / blob