blob: ebe94f2cab9811e62f0ea1d64c0d6c75838d0367 [file] [log] [blame]
Linus Torvalds1da177e2005-04-16 15:20:36 -07001/proc/sys/net/ipv4/* Variables:
2
3ip_forward - BOOLEAN
4 0 - disabled (default)
Jesper Dangaard Brouere18f5fe2009-02-23 04:39:04 +00005 not 0 - enabled
Linus Torvalds1da177e2005-04-16 15:20:36 -07006
7 Forward Packets between interfaces.
8
9 This variable is special, its change resets all configuration
10 parameters to their default state (RFC1122 for hosts, RFC1812
11 for routers)
12
13ip_default_ttl - INTEGER
Eric Dumazetcc6f02d2010-12-13 12:50:49 -080014 Default value of TTL field (Time To Live) for outgoing (but not
15 forwarded) IP packets. Should be between 1 and 255 inclusive.
16 Default: 64 (as recommended by RFC1700)
Linus Torvalds1da177e2005-04-16 15:20:36 -070017
Hannes Frederic Sowacd174e62013-12-14 05:13:45 +010018ip_no_pmtu_disc - INTEGER
19 Disable Path MTU Discovery. If enabled in mode 1 and a
Hannes Frederic Sowa188b04d2013-12-14 04:42:13 +010020 fragmentation-required ICMP is received, the PMTU to this
21 destination will be set to min_pmtu (see below). You will need
22 to raise min_pmtu to the smallest interface MTU on your system
23 manually if you want to avoid locally generated fragments.
Hannes Frederic Sowacd174e62013-12-14 05:13:45 +010024
25 In mode 2 incoming Path MTU Discovery messages will be
26 discarded. Outgoing frames are handled the same as in mode 1,
27 implicitly setting IP_PMTUDISC_DONT on every created socket.
28
Hannes Frederic Sowa8ed1dc42014-01-09 10:01:17 +010029 Mode 3 is a hardend pmtu discover mode. The kernel will only
30 accept fragmentation-needed errors if the underlying protocol
31 can verify them besides a plain socket lookup. Current
32 protocols for which pmtu events will be honored are TCP, SCTP
33 and DCCP as they verify e.g. the sequence number or the
34 association. This mode should not be enabled globally but is
35 only intended to secure e.g. name servers in namespaces where
36 TCP path mtu must still work but path MTU information of other
37 protocols should be discarded. If enabled globally this mode
38 could break other protocols.
39
40 Possible values: 0-3
Hannes Frederic Sowa188b04d2013-12-14 04:42:13 +010041 Default: FALSE
Linus Torvalds1da177e2005-04-16 15:20:36 -070042
43min_pmtu - INTEGER
Eric Dumazet20db93c2011-11-08 14:21:44 -050044 default 552 - minimum discovered Path MTU
Linus Torvalds1da177e2005-04-16 15:20:36 -070045
Hannes Frederic Sowaf87c10a2014-01-09 10:01:15 +010046ip_forward_use_pmtu - BOOLEAN
47 By default we don't trust protocol path MTUs while forwarding
48 because they could be easily forged and can lead to unwanted
49 fragmentation by the router.
50 You only need to enable this if you have user-space software
51 which tries to discover path mtus by itself and depends on the
52 kernel honoring this information. This is normally not the
53 case.
54 Default: 0 (disabled)
55 Possible values:
56 0 - disabled
57 1 - enabled
58
Loganaden Velvindron219b5f22014-11-04 03:02:49 -080059fwmark_reflect - BOOLEAN
60 Controls the fwmark of kernel-generated IPv4 reply packets that are not
61 associated with a socket for example, TCP RSTs or ICMP echo replies).
62 If unset, these packets have a fwmark of zero. If set, they have the
63 fwmark of the packet they are replying to.
64 Default: 0
65
Ben Greearcbaf0872010-11-08 09:13:48 +000066route/max_size - INTEGER
67 Maximum number of routes allowed in the kernel. Increase
68 this when using large numbers of interfaces and/or routes.
Ani Sinha25050c62015-01-07 15:45:56 -080069 From linux kernel 3.6 onwards, this is deprecated for ipv4
70 as route cache is no longer used.
Ben Greearcbaf0872010-11-08 09:13:48 +000071
YOSHIFUJI Hideaki / 吉藤英明27246802013-01-22 05:20:05 +000072neigh/default/gc_thresh1 - INTEGER
73 Minimum number of entries to keep. Garbage collector will not
74 purge entries if there are fewer than this number.
Li RongQingb66c66d2013-03-14 22:49:47 +000075 Default: 128
YOSHIFUJI Hideaki / 吉藤英明27246802013-01-22 05:20:05 +000076
stephen hemmingera3d12142014-08-25 15:05:30 -070077neigh/default/gc_thresh2 - INTEGER
78 Threshold when garbage collector becomes more aggressive about
79 purging entries. Entries older than 5 seconds will be cleared
80 when over this number.
81 Default: 512
82
Ben Greearcbaf0872010-11-08 09:13:48 +000083neigh/default/gc_thresh3 - INTEGER
84 Maximum number of neighbor entries allowed. Increase this
85 when using large numbers of interfaces and when communicating
86 with large numbers of directly-connected peers.
Shan Weicc868022012-12-04 18:50:35 +000087 Default: 1024
Ben Greearcbaf0872010-11-08 09:13:48 +000088
Eric Dumazet8b5c1712011-11-09 12:07:14 +000089neigh/default/unres_qlen_bytes - INTEGER
90 The maximum number of bytes which may be used by packets
91 queued for each unresolved address by other network layers.
92 (added in linux 3.3)
stephen hemminger3b09adc2013-01-03 07:50:29 +000093 Setting negative value is meaningless and will return error.
Shan Weicc868022012-12-04 18:50:35 +000094 Default: 65536 Bytes(64KB)
Eric Dumazet8b5c1712011-11-09 12:07:14 +000095
96neigh/default/unres_qlen - INTEGER
97 The maximum number of packets which may be queued for each
98 unresolved address by other network layers.
99 (deprecated in linux 3.3) : use unres_qlen_bytes instead.
Shan Weicc868022012-12-04 18:50:35 +0000100 Prior to linux 3.3, the default value is 3 which may cause
Shan Wei5d248c42012-12-06 16:27:51 +0000101 unexpected packet loss. The current default value is calculated
Shan Weicc868022012-12-04 18:50:35 +0000102 according to default value of unres_qlen_bytes and true size of
103 packet.
104 Default: 31
Eric Dumazet8b5c1712011-11-09 12:07:14 +0000105
Linus Torvalds1da177e2005-04-16 15:20:36 -0700106mtu_expires - INTEGER
107 Time, in seconds, that cached PMTU information is kept.
108
109min_adv_mss - INTEGER
110 The advertised MSS depends on the first hop route MTU, but will
111 never be lower than this setting.
112
113IP Fragmentation:
114
115ipfrag_high_thresh - INTEGER
Jesper Dangaard Brouere18f5fe2009-02-23 04:39:04 +0000116 Maximum memory used to reassemble IP fragments. When
Linus Torvalds1da177e2005-04-16 15:20:36 -0700117 ipfrag_high_thresh bytes of memory is allocated for this purpose,
118 the fragment handler will toss packets until ipfrag_low_thresh
Nikolay Aleksandrov1bab4c72014-07-24 16:50:37 +0200119 is reached. This also serves as a maximum limit to namespaces
120 different from the initial one.
Jesper Dangaard Brouere18f5fe2009-02-23 04:39:04 +0000121
Linus Torvalds1da177e2005-04-16 15:20:36 -0700122ipfrag_low_thresh - INTEGER
Florian Westphalb13d3cb2014-07-24 16:50:32 +0200123 Maximum memory used to reassemble IP fragments before the kernel
124 begins to remove incomplete fragment queues to free up resources.
125 The kernel still accepts new fragments for defragmentation.
Linus Torvalds1da177e2005-04-16 15:20:36 -0700126
127ipfrag_time - INTEGER
Jesper Dangaard Brouere18f5fe2009-02-23 04:39:04 +0000128 Time in seconds to keep an IP fragment in memory.
Linus Torvalds1da177e2005-04-16 15:20:36 -0700129
Herbert Xu89cee8b2005-12-13 23:14:27 -0800130ipfrag_max_dist - INTEGER
Jesper Dangaard Brouere18f5fe2009-02-23 04:39:04 +0000131 ipfrag_max_dist is a non-negative integer value which defines the
132 maximum "disorder" which is allowed among fragments which share a
133 common IP source address. Note that reordering of packets is
134 not unusual, but if a large number of fragments arrive from a source
135 IP address while a particular fragment queue remains incomplete, it
136 probably indicates that one or more fragments belonging to that queue
137 have been lost. When ipfrag_max_dist is positive, an additional check
138 is done on fragments before they are added to a reassembly queue - if
139 ipfrag_max_dist (or more) fragments have arrived from a particular IP
140 address between additions to any IP fragment queue using that source
141 address, it's presumed that one or more fragments in the queue are
142 lost. The existing fragment queue will be dropped, and a new one
Herbert Xu89cee8b2005-12-13 23:14:27 -0800143 started. An ipfrag_max_dist value of zero disables this check.
144
145 Using a very small value, e.g. 1 or 2, for ipfrag_max_dist can
146 result in unnecessarily dropping fragment queues when normal
Jesper Dangaard Brouere18f5fe2009-02-23 04:39:04 +0000147 reordering of packets occurs, which could lead to poor application
148 performance. Using a very large value, e.g. 50000, increases the
149 likelihood of incorrectly reassembling IP fragments that originate
Herbert Xu89cee8b2005-12-13 23:14:27 -0800150 from different IP datagrams, which could result in data corruption.
151 Default: 64
152
Linus Torvalds1da177e2005-04-16 15:20:36 -0700153INET peer storage:
154
155inet_peer_threshold - INTEGER
Jesper Dangaard Brouere18f5fe2009-02-23 04:39:04 +0000156 The approximate size of the storage. Starting from this threshold
Linus Torvalds1da177e2005-04-16 15:20:36 -0700157 entries will be thrown aggressively. This threshold also determines
158 entries' time-to-live and time intervals between garbage collection
159 passes. More entries, less time-to-live, less GC interval.
160
161inet_peer_minttl - INTEGER
162 Minimum time-to-live of entries. Should be enough to cover fragment
163 time-to-live on the reassembling side. This minimum time-to-live is
164 guaranteed if the pool size is less than inet_peer_threshold.
Stephen Hemminger77a538d2008-07-01 17:22:48 -0700165 Measured in seconds.
Linus Torvalds1da177e2005-04-16 15:20:36 -0700166
167inet_peer_maxttl - INTEGER
168 Maximum time-to-live of entries. Unused entries will expire after
169 this period of time if there is no memory pressure on the pool (i.e.
170 when the number of entries in the pool is very small).
Stephen Hemminger77a538d2008-07-01 17:22:48 -0700171 Measured in seconds.
Linus Torvalds1da177e2005-04-16 15:20:36 -0700172
Jesper Dangaard Brouere18f5fe2009-02-23 04:39:04 +0000173TCP variables:
Linus Torvalds1da177e2005-04-16 15:20:36 -0700174
Stephen Hemmingeref56e622006-11-09 16:37:26 -0800175somaxconn - INTEGER
176 Limit of socket listen() backlog, known in userspace as SOMAXCONN.
177 Defaults to 128. See also tcp_max_syn_backlog for additional tuning
178 for TCP sockets.
179
Stephen Hemmingeref56e622006-11-09 16:37:26 -0800180tcp_abort_on_overflow - BOOLEAN
181 If listening service is too slow to accept new connections,
182 reset them. Default state is FALSE. It means that if overflow
183 occurred due to a burst, connection will recover. Enable this
184 option _only_ if you are really sure that listening daemon
185 cannot be tuned to accept connections faster. Enabling this
186 option can harm clients of your server.
Linus Torvalds1da177e2005-04-16 15:20:36 -0700187
Stephen Hemmingeref56e622006-11-09 16:37:26 -0800188tcp_adv_win_scale - INTEGER
189 Count buffering overhead as bytes/2^tcp_adv_win_scale
190 (if tcp_adv_win_scale > 0) or bytes-bytes/2^(-tcp_adv_win_scale),
191 if it is <= 0.
Alexey Dobriyan0147fc02010-11-22 12:54:21 +0000192 Possible values are [-31, 31], inclusive.
Eric Dumazetb49960a2012-05-02 02:28:41 +0000193 Default: 1
Stephen Hemmingeref56e622006-11-09 16:37:26 -0800194
195tcp_allowed_congestion_control - STRING
196 Show/set the congestion control choices available to non-privileged
197 processes. The list is a subset of those listed in
198 tcp_available_congestion_control.
199 Default is "reno" and the default setting (tcp_congestion_control).
200
201tcp_app_win - INTEGER
202 Reserve max(window/2^tcp_app_win, mss) of window for application
203 buffer. Value 0 is special, it means that nothing is reserved.
204 Default: 31
205
Eric Dumazetf54b3112013-12-05 22:36:05 -0800206tcp_autocorking - BOOLEAN
207 Enable TCP auto corking :
208 When applications do consecutive small write()/sendmsg() system calls,
209 we try to coalesce these small writes as much as possible, to lower
210 total amount of sent packets. This is done if at least one prior
211 packet for the flow is waiting in Qdisc queues or device transmit
212 queue. Applications can still use TCP_CORK for optimal behavior
213 when they know how/when to uncork their sockets.
214 Default : 1
215
Stephen Hemmingeref56e622006-11-09 16:37:26 -0800216tcp_available_congestion_control - STRING
217 Shows the available congestion control choices that are registered.
218 More congestion control algorithms may be available as modules,
219 but not loaded.
220
John Heffner71599cd2007-02-27 10:03:56 -0800221tcp_base_mss - INTEGER
Stephen Hemminger4edc2f32008-07-10 16:50:26 -0700222 The initial value of search_low to be used by the packetization layer
223 Path MTU discovery (MTU probing). If MTU probing is enabled,
224 this is the initial MSS used by the connection.
John Heffner71599cd2007-02-27 10:03:56 -0800225
Stephen Hemmingeref56e622006-11-09 16:37:26 -0800226tcp_congestion_control - STRING
227 Set the congestion control algorithm to be used for new
228 connections. The algorithm "reno" is always available, but
229 additional choices may be available based on kernel configuration.
230 Default is set as part of kernel configuration.
Eric Dumazetd8a6e652011-11-30 01:02:41 +0000231 For passive connections, the listener congestion control choice
232 is inherited.
233 [see setsockopt(listenfd, SOL_TCP, TCP_CONGESTION, "name" ...) ]
Stephen Hemmingeref56e622006-11-09 16:37:26 -0800234
235tcp_dsack - BOOLEAN
236 Allows TCP to send "duplicate" SACKs.
237
Yuchung Chengeed530b2012-05-02 13:30:03 +0000238tcp_early_retrans - INTEGER
239 Enable Early Retransmit (ER), per RFC 5827. ER lowers the threshold
240 for triggering fast retransmit when the amount of outstanding data is
241 small and when no previously unsent data can be transmitted (such
Nandita Dukkipati6ba8a3b2013-03-11 10:00:43 +0000242 that limited transmit could be used). Also controls the use of
Masanari Iida3dd17ed2013-05-24 07:05:59 +0000243 Tail loss probe (TLP) that converts RTOs occurring due to tail
Nandita Dukkipati6ba8a3b2013-03-11 10:00:43 +0000244 losses into fast recovery (draft-dukkipati-tcpm-tcp-loss-probe-01).
Yuchung Chengeed530b2012-05-02 13:30:03 +0000245 Possible values:
246 0 disables ER
247 1 enables ER
248 2 enables ER but delays fast recovery and fast retransmit
249 by a fourth of RTT. This mitigates connection falsely
250 recovers when network has a small degree of reordering
251 (less than 3 packets).
Nandita Dukkipati6ba8a3b2013-03-11 10:00:43 +0000252 3 enables delayed ER and TLP.
253 4 enables TLP only.
254 Default: 3
Yuchung Chengeed530b2012-05-02 13:30:03 +0000255
Peter Chubb34a6ef32011-02-02 15:39:58 -0800256tcp_ecn - INTEGER
Rick Jones7e3a2dc2012-11-28 09:53:10 +0000257 Control use of Explicit Congestion Notification (ECN) by TCP.
258 ECN is used only when both ends of the TCP connection indicate
259 support for it. This feature is useful in avoiding losses due
260 to congestion by allowing supporting routers to signal
261 congestion before having to drop packets.
Ilpo Järvinen255cac92009-05-04 11:07:36 -0700262 Possible values are:
Rick Jones7e3a2dc2012-11-28 09:53:10 +0000263 0 Disable ECN. Neither initiate nor accept ECN.
Vijay Subramanian3d55b322013-01-09 12:21:30 +0000264 1 Enable ECN when requested by incoming connections and
265 also request ECN on outgoing connection attempts.
266 2 Enable ECN when requested by incoming connections
Rick Jones7e3a2dc2012-11-28 09:53:10 +0000267 but do not request ECN on outgoing connections.
Ilpo Järvinen255cac92009-05-04 11:07:36 -0700268 Default: 2
Stephen Hemmingeref56e622006-11-09 16:37:26 -0800269
Daniel Borkmann49213552015-05-19 21:04:22 +0200270tcp_ecn_fallback - BOOLEAN
271 If the kernel detects that ECN connection misbehaves, enable fall
272 back to non-ECN. Currently, this knob implements the fallback
273 from RFC3168, section 6.1.1.1., but we reserve that in future,
274 additional detection mechanisms could be implemented under this
275 knob. The value is not used, if tcp_ecn or per route (or congestion
276 control) ECN settings are disabled.
277 Default: 1 (fallback enabled)
278
Stephen Hemmingeref56e622006-11-09 16:37:26 -0800279tcp_fack - BOOLEAN
280 Enable FACK congestion avoidance and fast retransmission.
281 The value is not used, if tcp_sack is not enabled.
282
283tcp_fin_timeout - INTEGER
Rick Jonesd825da22012-12-10 11:33:00 +0000284 The length of time an orphaned (no longer referenced by any
285 application) connection will remain in the FIN_WAIT_2 state
286 before it is aborted at the local end. While a perfectly
287 valid "receive only" state for an un-orphaned connection, an
288 orphaned connection in FIN_WAIT_2 state could otherwise wait
289 forever for the remote to close its end of the connection.
290 Cf. tcp_max_orphans
291 Default: 60 seconds
Stephen Hemmingeref56e622006-11-09 16:37:26 -0800292
Ilpo Järvinen89808062007-02-27 10:10:55 -0800293tcp_frto - INTEGER
Yuchung Chenge33099f2013-03-20 13:33:00 +0000294 Enables Forward RTO-Recovery (F-RTO) defined in RFC5682.
Ilpo Järvinencd998892007-09-20 11:35:26 -0700295 F-RTO is an enhanced recovery algorithm for TCP retransmission
Yuchung Chenge33099f2013-03-20 13:33:00 +0000296 timeouts. It is particularly beneficial in networks where the
297 RTT fluctuates (e.g., wireless). F-RTO is sender-side only
298 modification. It does not require any support from the peer.
Stephen Hemminger4edc2f32008-07-10 16:50:26 -0700299
Yuchung Chenge33099f2013-03-20 13:33:00 +0000300 By default it's enabled with a non-zero value. 0 disables F-RTO.
Linus Torvalds1da177e2005-04-16 15:20:36 -0700301
Neal Cardwell032ee422015-02-06 16:04:38 -0500302tcp_invalid_ratelimit - INTEGER
303 Limit the maximal rate for sending duplicate acknowledgments
304 in response to incoming TCP packets that are for an existing
305 connection but that are invalid due to any of these reasons:
306
307 (a) out-of-window sequence number,
308 (b) out-of-window acknowledgment number, or
309 (c) PAWS (Protection Against Wrapped Sequence numbers) check failure
310
311 This can help mitigate simple "ack loop" DoS attacks, wherein
312 a buggy or malicious middlebox or man-in-the-middle can
313 rewrite TCP header fields in manner that causes each endpoint
314 to think that the other is sending invalid TCP segments, thus
315 causing each side to send an unterminating stream of duplicate
316 acknowledgments for invalid segments.
317
318 Using 0 disables rate-limiting of dupacks in response to
319 invalid segments; otherwise this value specifies the minimal
320 space between sending such dupacks, in milliseconds.
321
322 Default: 500 (milliseconds).
323
Linus Torvalds1da177e2005-04-16 15:20:36 -0700324tcp_keepalive_time - INTEGER
325 How often TCP sends out keepalive messages when keepalive is enabled.
326 Default: 2hours.
327
328tcp_keepalive_probes - INTEGER
329 How many keepalive probes TCP sends out, until it decides that the
330 connection is broken. Default value: 9.
331
332tcp_keepalive_intvl - INTEGER
333 How frequently the probes are send out. Multiplied by
334 tcp_keepalive_probes it is time to kill not responding connection,
335 after probes started. Default value: 75sec i.e. connection
336 will be aborted after ~11 minutes of retries.
337
Stephen Hemmingeref56e622006-11-09 16:37:26 -0800338tcp_low_latency - BOOLEAN
339 If set, the TCP stack makes decisions that prefer lower
340 latency as opposed to higher throughput. By default, this
341 option is not set meaning that higher throughput is preferred.
342 An example of an application where this default should be
343 changed would be a Beowulf compute cluster.
344 Default: 0
Linus Torvalds1da177e2005-04-16 15:20:36 -0700345
346tcp_max_orphans - INTEGER
347 Maximal number of TCP sockets not attached to any user file handle,
348 held by system. If this number is exceeded orphaned connections are
349 reset immediately and warning is printed. This limit exists
350 only to prevent simple DoS attacks, you _must_ not rely on this
351 or lower the limit artificially, but rather increase it
352 (probably, after increasing installed memory),
353 if network conditions require more than default value,
354 and tune network services to linger and kill such states
355 more aggressively. Let me to remind again: each orphan eats
356 up to ~64K of unswappable memory.
357
Linus Torvalds1da177e2005-04-16 15:20:36 -0700358tcp_max_syn_backlog - INTEGER
Peter Pan(潘卫平)99b53bd2011-12-05 21:39:41 +0000359 Maximal number of remembered connection requests, which have not
360 received an acknowledgment from connecting client.
361 The minimal value is 128 for low memory machines, and it will
362 increase in proportion to the memory of machine.
363 If server suffers from overload, try increasing this number.
Linus Torvalds1da177e2005-04-16 15:20:36 -0700364
Stephen Hemmingeref56e622006-11-09 16:37:26 -0800365tcp_max_tw_buckets - INTEGER
366 Maximal number of timewait sockets held by system simultaneously.
367 If this number is exceeded time-wait socket is immediately destroyed
368 and warning is printed. This limit exists only to prevent
369 simple DoS attacks, you _must_ not lower the limit artificially,
370 but rather increase it (probably, after increasing installed memory),
371 if network conditions require more than default value.
Linus Torvalds1da177e2005-04-16 15:20:36 -0700372
Stephen Hemmingeref56e622006-11-09 16:37:26 -0800373tcp_mem - vector of 3 INTEGERs: min, pressure, max
374 min: below this number of pages TCP is not bothered about its
375 memory appetite.
Linus Torvalds1da177e2005-04-16 15:20:36 -0700376
Stephen Hemmingeref56e622006-11-09 16:37:26 -0800377 pressure: when amount of memory allocated by TCP exceeds this number
378 of pages, TCP moderates its memory consumption and enters memory
379 pressure mode, which is exited when memory consumption falls
380 under "min".
Linus Torvalds1da177e2005-04-16 15:20:36 -0700381
Stephen Hemmingeref56e622006-11-09 16:37:26 -0800382 max: number of pages allowed for queueing by all TCP sockets.
Linus Torvalds1da177e2005-04-16 15:20:36 -0700383
Stephen Hemmingeref56e622006-11-09 16:37:26 -0800384 Defaults are calculated at boot time from amount of available
385 memory.
Linus Torvalds1da177e2005-04-16 15:20:36 -0700386
John Heffner71599cd2007-02-27 10:03:56 -0800387tcp_moderate_rcvbuf - BOOLEAN
Stephen Hemminger4edc2f32008-07-10 16:50:26 -0700388 If set, TCP performs receive buffer auto-tuning, attempting to
John Heffner71599cd2007-02-27 10:03:56 -0800389 automatically size the buffer (no greater than tcp_rmem[2]) to
390 match the size required by the path for full throughput. Enabled by
391 default.
392
393tcp_mtu_probing - INTEGER
394 Controls TCP Packetization-Layer Path MTU Discovery. Takes three
395 values:
396 0 - Disabled
397 1 - Disabled by default, enabled when an ICMP black hole detected
398 2 - Always enabled, use initial MSS of tcp_base_mss.
399
Fan Dufab42762015-03-06 11:18:25 +0800400tcp_probe_interval - INTEGER
401 Controls how often to start TCP Packetization-Layer Path MTU
402 Discovery reprobe. The default is reprobing every 10 minutes as
403 per RFC4821.
404
405tcp_probe_threshold - INTEGER
406 Controls when TCP Packetization-Layer Path MTU Discovery probing
407 will stop in respect to the width of search range in bytes. Default
408 is 8 bytes.
409
John Heffner71599cd2007-02-27 10:03:56 -0800410tcp_no_metrics_save - BOOLEAN
411 By default, TCP saves various connection metrics in the route cache
412 when the connection closes, so that connections established in the
413 near future can use these to set initial conditions. Usually, this
414 increases overall performance, but may sometimes cause performance
Simon Arlott0f035b82007-10-20 01:30:25 +0200415 degradation. If set, TCP will not cache metrics on closing
John Heffner71599cd2007-02-27 10:03:56 -0800416 connections.
417
Stephen Hemmingeref56e622006-11-09 16:37:26 -0800418tcp_orphan_retries - INTEGER
Damian Lukowski5d789222009-09-01 10:24:04 +0000419 This value influences the timeout of a locally closed TCP connection,
420 when RTO retransmissions remain unacknowledged.
421 See tcp_retries2 for more details.
422
David S. Miller06b8fc52011-07-08 09:31:31 -0700423 The default value is 8.
Damian Lukowski5d789222009-09-01 10:24:04 +0000424 If your machine is a loaded WEB server,
Stephen Hemmingeref56e622006-11-09 16:37:26 -0800425 you should think about lowering this value, such sockets
426 may consume significant resources. Cf. tcp_max_orphans.
Linus Torvalds1da177e2005-04-16 15:20:36 -0700427
428tcp_reordering - INTEGER
Eric Dumazetdca145f2014-10-27 21:45:24 -0700429 Initial reordering level of packets in a TCP stream.
430 TCP stack can then dynamically adjust flow reordering level
431 between this initial value and tcp_max_reordering
Jesper Dangaard Brouere18f5fe2009-02-23 04:39:04 +0000432 Default: 3
Linus Torvalds1da177e2005-04-16 15:20:36 -0700433
Eric Dumazetdca145f2014-10-27 21:45:24 -0700434tcp_max_reordering - INTEGER
435 Maximal reordering level of packets in a TCP stream.
436 300 is a fairly conservative value, but you might increase it
437 if paths are using per packet load balancing (like bonding rr mode)
438 Default: 300
439
Linus Torvalds1da177e2005-04-16 15:20:36 -0700440tcp_retrans_collapse - BOOLEAN
441 Bug-to-bug compatibility with some broken printers.
442 On retransmit try to send bigger packets to work around bugs in
443 certain TCP stacks.
444
Stephen Hemmingeref56e622006-11-09 16:37:26 -0800445tcp_retries1 - INTEGER
Damian Lukowski5d789222009-09-01 10:24:04 +0000446 This value influences the time, after which TCP decides, that
447 something is wrong due to unacknowledged RTO retransmissions,
448 and reports this suspicion to the network layer.
449 See tcp_retries2 for more details.
450
451 RFC 1122 recommends at least 3 retransmissions, which is the
452 default.
Linus Torvalds1da177e2005-04-16 15:20:36 -0700453
Stephen Hemmingeref56e622006-11-09 16:37:26 -0800454tcp_retries2 - INTEGER
Damian Lukowski5d789222009-09-01 10:24:04 +0000455 This value influences the timeout of an alive TCP connection,
456 when RTO retransmissions remain unacknowledged.
457 Given a value of N, a hypothetical TCP connection following
458 exponential backoff with an initial RTO of TCP_RTO_MIN would
459 retransmit N times before killing the connection at the (N+1)th RTO.
460
461 The default value of 15 yields a hypothetical timeout of 924.6
462 seconds and is a lower bound for the effective timeout.
463 TCP will effectively time out at the first RTO which exceeds the
464 hypothetical timeout.
465
466 RFC 1122 recommends at least 100 seconds for the timeout,
467 which corresponds to a value of at least 8.
Linus Torvalds1da177e2005-04-16 15:20:36 -0700468
Stephen Hemmingeref56e622006-11-09 16:37:26 -0800469tcp_rfc1337 - BOOLEAN
470 If set, the TCP stack behaves conforming to RFC1337. If unset,
471 we are not conforming to RFC, but prevent TCP TIME_WAIT
472 assassination.
473 Default: 0
Linus Torvalds1da177e2005-04-16 15:20:36 -0700474
475tcp_rmem - vector of 3 INTEGERs: min, default, max
476 min: Minimal size of receive buffer used by TCP sockets.
477 It is guaranteed to each TCP socket, even under moderate memory
478 pressure.
Max Matveev6539fef2011-06-21 21:18:13 +0000479 Default: 1 page
Linus Torvalds1da177e2005-04-16 15:20:36 -0700480
J. Bruce Fields53025f52008-07-10 16:47:41 -0700481 default: initial size of receive buffer used by TCP sockets.
Linus Torvalds1da177e2005-04-16 15:20:36 -0700482 This value overrides net.core.rmem_default used by other protocols.
483 Default: 87380 bytes. This value results in window of 65535 with
484 default setting of tcp_adv_win_scale and tcp_app_win:0 and a bit
485 less for default tcp_app_win. See below about these variables.
486
487 max: maximal size of receive buffer allowed for automatically
488 selected receiver buffers for TCP socket. This value does not override
J. Bruce Fields53025f52008-07-10 16:47:41 -0700489 net.core.rmem_max. Calling setsockopt() with SO_RCVBUF disables
490 automatic tuning of that socket's receive buffer size, in which
491 case this value is ignored.
Eric Dumazetb49960a2012-05-02 02:28:41 +0000492 Default: between 87380B and 6MB, depending on RAM size.
Linus Torvalds1da177e2005-04-16 15:20:36 -0700493
Stephen Hemmingeref56e622006-11-09 16:37:26 -0800494tcp_sack - BOOLEAN
495 Enable select acknowledgments (SACKS).
Rick Jones15d99e02006-03-20 22:40:29 -0800496
David S. Miller35089bb2006-06-13 22:33:04 -0700497tcp_slow_start_after_idle - BOOLEAN
498 If set, provide RFC2861 behavior and time out the congestion
499 window after an idle period. An idle period is defined at
500 the current RTO. If unset, the congestion window will not
501 be timed out after an idle period.
502 Default: 1
503
Stephen Hemmingeref56e622006-11-09 16:37:26 -0800504tcp_stdurg - BOOLEAN
Stephen Hemminger4edc2f32008-07-10 16:50:26 -0700505 Use the Host requirements interpretation of the TCP urgent pointer field.
Stephen Hemmingeref56e622006-11-09 16:37:26 -0800506 Most hosts use the older BSD interpretation, so if you turn this on
507 Linux might not communicate correctly with them.
508 Default: FALSE
509
510tcp_synack_retries - INTEGER
511 Number of times SYNACKs for a passive TCP connection attempt will
512 be retransmitted. Should not be higher than 255. Default value
Alex Bergmann6c9ff972012-08-31 02:48:31 +0000513 is 5, which corresponds to 31seconds till the last retransmission
514 with the current initial RTO of 1second. With this the final timeout
515 for a passive TCP connection will happen after 63seconds.
Stephen Hemmingeref56e622006-11-09 16:37:26 -0800516
517tcp_syncookies - BOOLEAN
Shan Weia3c910d2013-06-21 15:18:32 +0800518 Only valid when the kernel was compiled with CONFIG_SYN_COOKIES
Stephen Hemmingeref56e622006-11-09 16:37:26 -0800519 Send out syncookies when the syn backlog queue of a socket
Stephen Hemminger4edc2f32008-07-10 16:50:26 -0700520 overflows. This is to prevent against the common 'SYN flood attack'
Shan Weia3c910d2013-06-21 15:18:32 +0800521 Default: 1
Stephen Hemmingeref56e622006-11-09 16:37:26 -0800522
523 Note, that syncookies is fallback facility.
524 It MUST NOT be used to help highly loaded servers to stand
Stephen Hemminger4edc2f32008-07-10 16:50:26 -0700525 against legal connection rate. If you see SYN flood warnings
Stephen Hemmingeref56e622006-11-09 16:37:26 -0800526 in your logs, but investigation shows that they occur
527 because of overload with legal connections, you should tune
528 another parameters until this warning disappear.
529 See: tcp_max_syn_backlog, tcp_synack_retries, tcp_abort_on_overflow.
530
531 syncookies seriously violate TCP protocol, do not allow
532 to use TCP extensions, can result in serious degradation
533 of some services (f.e. SMTP relaying), visible not by you,
534 but your clients and relays, contacting you. While you see
Stephen Hemminger4edc2f32008-07-10 16:50:26 -0700535 SYN flood warnings in logs not being really flooded, your server
Stephen Hemmingeref56e622006-11-09 16:37:26 -0800536 is seriously misconfigured.
537
Hannes Frederic Sowa5ad37d52013-07-26 17:43:23 +0200538 If you want to test which effects syncookies have to your
539 network connections you can set this knob to 2 to enable
540 unconditionally generation of syncookies.
541
Yuchung Chengcf60af02012-07-19 06:43:09 +0000542tcp_fastopen - INTEGER
543 Enable TCP Fast Open feature (draft-ietf-tcpm-fastopen) to send data
544 in the opening SYN packet. To use this feature, the client application
Jerry Chu10467162012-08-31 12:29:11 +0000545 must use sendmsg() or sendto() with MSG_FASTOPEN flag rather than
546 connect() to perform a TCP handshake automatically.
Yuchung Chengcf60af02012-07-19 06:43:09 +0000547
Jerry Chu10467162012-08-31 12:29:11 +0000548 The values (bitmap) are
Yuchung Cheng0d41cca2013-10-31 09:19:32 -0700549 1: Enables sending data in the opening SYN on the client w/ MSG_FASTOPEN.
Jerry Chu10467162012-08-31 12:29:11 +0000550 2: Enables TCP Fast Open on the server side, i.e., allowing data in
551 a SYN packet to be accepted and passed to the application before
552 3-way hand shake finishes.
553 4: Send data in the opening SYN regardless of cookie availability and
554 without a cookie option.
555 0x100: Accept SYN data w/o validating the cookie.
556 0x200: Accept data-in-SYN w/o any cookie option present.
557 0x400/0x800: Enable Fast Open on all listeners regardless of the
558 TCP_FASTOPEN socket option. The two different flags designate two
559 different ways of setting max_qlen without the TCP_FASTOPEN socket
560 option.
Yuchung Chengcf60af02012-07-19 06:43:09 +0000561
Yuchung Cheng0d41cca2013-10-31 09:19:32 -0700562 Default: 1
Yuchung Chengcf60af02012-07-19 06:43:09 +0000563
Jerry Chu10467162012-08-31 12:29:11 +0000564 Note that the client & server side Fast Open flags (1 and 2
565 respectively) must be also enabled before the rest of flags can take
566 effect.
567
568 See include/net/tcp.h and the code for more details.
569
Stephen Hemmingeref56e622006-11-09 16:37:26 -0800570tcp_syn_retries - INTEGER
571 Number of times initial SYNs for an active TCP connection attempt
572 will be retransmitted. Should not be higher than 255. Default value
stephen hemminger3b09adc2013-01-03 07:50:29 +0000573 is 6, which corresponds to 63seconds till the last retransmission
Alex Bergmann6c9ff972012-08-31 02:48:31 +0000574 with the current initial RTO of 1second. With this the final timeout
575 for an active TCP connection attempt will happen after 127seconds.
Stephen Hemmingeref56e622006-11-09 16:37:26 -0800576
577tcp_timestamps - BOOLEAN
578 Enable timestamps as defined in RFC1323.
579
Eric Dumazet95bd09e2013-08-27 05:46:32 -0700580tcp_min_tso_segs - INTEGER
581 Minimal number of segments per TSO frame.
582 Since linux-3.12, TCP does an automatic sizing of TSO frames,
583 depending on flow rate, instead of filling 64Kbytes packets.
584 For specific usages, it's possible to force TCP to build big
585 TSO frames. Note that TCP stack might split too big TSO packets
586 if available window is too small.
587 Default: 2
588
Eric Dumazet43e122b2015-08-21 17:38:02 -0700589tcp_pacing_ss_ratio - INTEGER
590 sk->sk_pacing_rate is set by TCP stack using a ratio applied
591 to current rate. (current_rate = cwnd * mss / srtt)
592 If TCP is in slow start, tcp_pacing_ss_ratio is applied
593 to let TCP probe for bigger speeds, assuming cwnd can be
594 doubled every other RTT.
595 Default: 200
596
597tcp_pacing_ca_ratio - INTEGER
598 sk->sk_pacing_rate is set by TCP stack using a ratio applied
599 to current rate. (current_rate = cwnd * mss / srtt)
600 If TCP is in congestion avoidance phase, tcp_pacing_ca_ratio
601 is applied to conservatively probe for bigger throughput.
602 Default: 120
603
Stephen Hemmingeref56e622006-11-09 16:37:26 -0800604tcp_tso_win_divisor - INTEGER
605 This allows control over what percentage of the congestion window
606 can be consumed by a single TSO frame.
607 The setting of this parameter is a choice between burstiness and
608 building larger TSO frames.
609 Default: 3
610
611tcp_tw_recycle - BOOLEAN
612 Enable fast recycling TIME-WAIT sockets. Default value is 0.
613 It should not be changed without advice/request of technical
614 experts.
615
616tcp_tw_reuse - BOOLEAN
617 Allow to reuse TIME-WAIT sockets for new connections when it is
618 safe from protocol viewpoint. Default value is 0.
619 It should not be changed without advice/request of technical
620 experts.
621
622tcp_window_scaling - BOOLEAN
623 Enable window scaling as defined in RFC1323.
624
625tcp_wmem - vector of 3 INTEGERs: min, default, max
J. Bruce Fields53025f52008-07-10 16:47:41 -0700626 min: Amount of memory reserved for send buffers for TCP sockets.
Stephen Hemmingeref56e622006-11-09 16:37:26 -0800627 Each TCP socket has rights to use it due to fact of its birth.
Max Matveev6539fef2011-06-21 21:18:13 +0000628 Default: 1 page
Stephen Hemmingeref56e622006-11-09 16:37:26 -0800629
J. Bruce Fields53025f52008-07-10 16:47:41 -0700630 default: initial size of send buffer used by TCP sockets. This
631 value overrides net.core.wmem_default used by other protocols.
632 It is usually lower than net.core.wmem_default.
Stephen Hemmingeref56e622006-11-09 16:37:26 -0800633 Default: 16K
634
J. Bruce Fields53025f52008-07-10 16:47:41 -0700635 max: Maximal amount of memory allowed for automatically tuned
636 send buffers for TCP sockets. This value does not override
637 net.core.wmem_max. Calling setsockopt() with SO_SNDBUF disables
638 automatic tuning of that socket's send buffer size, in which case
639 this value is ignored.
640 Default: between 64K and 4MB, depending on RAM size.
Stephen Hemmingeref56e622006-11-09 16:37:26 -0800641
Eric Dumazetc9bee3b72013-07-22 20:27:07 -0700642tcp_notsent_lowat - UNSIGNED INTEGER
643 A TCP socket can control the amount of unsent bytes in its write queue,
644 thanks to TCP_NOTSENT_LOWAT socket option. poll()/select()/epoll()
645 reports POLLOUT events if the amount of unsent bytes is below a per
646 socket value, and if the write queue is not full. sendmsg() will
647 also not add new buffers if the limit is hit.
648
649 This global variable controls the amount of unsent data for
650 sockets not using TCP_NOTSENT_LOWAT. For these sockets, a change
651 to the global variable has immediate effect.
652
653 Default: UINT_MAX (0xFFFFFFFF)
654
Stephen Hemmingeref56e622006-11-09 16:37:26 -0800655tcp_workaround_signed_windows - BOOLEAN
656 If set, assume no receipt of a window scaling option means the
657 remote TCP is broken and treats the window as a signed quantity.
658 If unset, assume the remote TCP is not broken even if we do
659 not receive a window scaling option from them.
660 Default: 0
661
Andreas Petlund36e31b0a2010-02-18 02:47:01 +0000662tcp_thin_linear_timeouts - BOOLEAN
663 Enable dynamic triggering of linear timeouts for thin streams.
664 If set, a check is performed upon retransmission by timeout to
665 determine if the stream is thin (less than 4 packets in flight).
666 As long as the stream is found to be thin, up to 6 linear
667 timeouts may be performed before exponential backoff mode is
668 initiated. This improves retransmission latency for
669 non-aggressive thin streams, often found to be time-dependent.
670 For more information on thin streams, see
671 Documentation/networking/tcp-thin.txt
672 Default: 0
673
Andreas Petlund7e380172010-02-18 04:48:19 +0000674tcp_thin_dupack - BOOLEAN
675 Enable dynamic triggering of retransmissions after one dupACK
676 for thin streams. If set, a check is performed upon reception
677 of a dupACK to determine if the stream is thin (less than 4
678 packets in flight). As long as the stream is found to be thin,
679 data is retransmitted on the first received dupACK. This
680 improves retransmission latency for non-aggressive thin
681 streams, often found to be time-dependent.
682 For more information on thin streams, see
683 Documentation/networking/tcp-thin.txt
684 Default: 0
685
Eric Dumazet46d3cea2012-07-11 05:50:31 +0000686tcp_limit_output_bytes - INTEGER
687 Controls TCP Small Queue limit per tcp socket.
688 TCP bulk sender tends to increase packets in flight until it
689 gets losses notifications. With SNDBUF autotuning, this can
690 result in a large amount of packets queued in qdisc/device
691 on the local machine, hurting latency of other flows, for
692 typical pfifo_fast qdiscs.
693 tcp_limit_output_bytes limits the number of bytes on qdisc
694 or device to reduce artificial RTT/cwnd and reduce bufferbloat.
Eric Dumazet46d3cea2012-07-11 05:50:31 +0000695 Default: 131072
696
Eric Dumazet282f23c2012-07-17 10:13:05 +0200697tcp_challenge_ack_limit - INTEGER
698 Limits number of Challenge ACK sent per second, as recommended
699 in RFC 5961 (Improving TCP's Robustness to Blind In-Window Attacks)
700 Default: 100
701
Hideo Aoki95766ff2007-12-31 00:29:24 -0800702UDP variables:
703
704udp_mem - vector of 3 INTEGERs: min, pressure, max
705 Number of pages allowed for queueing by all UDP sockets.
706
707 min: Below this number of pages UDP is not bothered about its
708 memory appetite. When amount of memory allocated by UDP exceeds
709 this number, UDP starts to moderate memory usage.
710
711 pressure: This value was introduced to follow format of tcp_mem.
712
713 max: Number of pages allowed for queueing by all UDP sockets.
714
715 Default is calculated at boot time from amount of available memory.
716
717udp_rmem_min - INTEGER
718 Minimal size of receive buffer used by UDP sockets in moderation.
719 Each UDP socket is able to use the size for receiving data, even if
720 total pages of UDP sockets exceed udp_mem pressure. The unit is byte.
Max Matveev6539fef2011-06-21 21:18:13 +0000721 Default: 1 page
Hideo Aoki95766ff2007-12-31 00:29:24 -0800722
723udp_wmem_min - INTEGER
724 Minimal size of send buffer used by UDP sockets in moderation.
725 Each UDP socket is able to use the size for sending data, even if
726 total pages of UDP sockets exceed udp_mem pressure. The unit is byte.
Max Matveev6539fef2011-06-21 21:18:13 +0000727 Default: 1 page
Hideo Aoki95766ff2007-12-31 00:29:24 -0800728
Paul Moore8802f612006-08-03 16:45:49 -0700729CIPSOv4 Variables:
730
731cipso_cache_enable - BOOLEAN
732 If set, enable additions to and lookups from the CIPSO label mapping
733 cache. If unset, additions are ignored and lookups always result in a
734 miss. However, regardless of the setting the cache is still
735 invalidated when required when means you can safely toggle this on and
736 off and the cache will always be "safe".
737 Default: 1
738
739cipso_cache_bucket_size - INTEGER
740 The CIPSO label cache consists of a fixed size hash table with each
741 hash bucket containing a number of cache entries. This variable limits
742 the number of entries in each hash bucket; the larger the value the
743 more CIPSO label mappings that can be cached. When the number of
744 entries in a given hash bucket reaches this limit adding new entries
745 causes the oldest entry in the bucket to be removed to make room.
746 Default: 10
747
748cipso_rbm_optfmt - BOOLEAN
749 Enable the "Optimized Tag 1 Format" as defined in section 3.4.2.6 of
750 the CIPSO draft specification (see Documentation/netlabel for details).
751 This means that when set the CIPSO tag will be padded with empty
752 categories in order to make the packet data 32-bit aligned.
753 Default: 0
754
755cipso_rbm_structvalid - BOOLEAN
756 If set, do a very strict check of the CIPSO option when
757 ip_options_compile() is called. If unset, relax the checks done during
758 ip_options_compile(). Either way is "safe" as errors are caught else
759 where in the CIPSO processing code but setting this to 0 (False) should
760 result in less work (i.e. it should be faster) but could cause problems
761 with other implementations that require strict checking.
762 Default: 0
763
Linus Torvalds1da177e2005-04-16 15:20:36 -0700764IP Variables:
765
766ip_local_port_range - 2 INTEGERS
767 Defines the local port range that is used by TCP and UDP to
Jesper Dangaard Brouere18f5fe2009-02-23 04:39:04 +0000768 choose the local port. The first number is the first, the
Eric Dumazet07f4c902015-05-24 14:49:35 -0700769 second the last local port number.
770 If possible, it is better these numbers have different parity.
771 (one even and one odd values)
772 The default values are 32768 and 60999 respectively.
Linus Torvalds1da177e2005-04-16 15:20:36 -0700773
Amerigo Wange3826f12010-05-05 00:27:06 +0000774ip_local_reserved_ports - list of comma separated ranges
775 Specify the ports which are reserved for known third-party
776 applications. These ports will not be used by automatic port
777 assignments (e.g. when calling connect() or bind() with port
778 number 0). Explicit port allocation behavior is unchanged.
779
780 The format used for both input and output is a comma separated
781 list of ranges (e.g. "1,2-4,10-10" for ports 1, 2, 3, 4 and
782 10). Writing to the file will clear all previously reserved
783 ports and update the current list with the one given in the
784 input.
785
786 Note that ip_local_port_range and ip_local_reserved_ports
787 settings are independent and both are considered by the kernel
788 when determining which ports are available for automatic port
789 assignments.
790
791 You can reserve ports which are not in the current
792 ip_local_port_range, e.g.:
793
794 $ cat /proc/sys/net/ipv4/ip_local_port_range
Eric Dumazet07f4c902015-05-24 14:49:35 -0700795 32000 60999
Amerigo Wange3826f12010-05-05 00:27:06 +0000796 $ cat /proc/sys/net/ipv4/ip_local_reserved_ports
797 8080,9148
798
799 although this is redundant. However such a setting is useful
800 if later the port range is changed to a value that will
801 include the reserved ports.
802
803 Default: Empty
804
Linus Torvalds1da177e2005-04-16 15:20:36 -0700805ip_nonlocal_bind - BOOLEAN
806 If set, allows processes to bind() to non-local IP addresses,
807 which can be quite useful - but may break some applications.
808 Default: 0
809
810ip_dynaddr - BOOLEAN
811 If set non-zero, enables support for dynamic addresses.
812 If set to a non-zero value larger than 1, a kernel log
813 message will be printed when dynamic address rewriting
814 occurs.
815 Default: 0
816
Cong Wange3d73bc2013-06-11 18:54:39 +0800817ip_early_demux - BOOLEAN
818 Optimize input packet processing down to one demux for
819 certain kinds of local sockets. Currently we only do this
820 for established TCP sockets.
821
822 It may add an additional cost for pure routing workloads that
823 reduces overall throughput, in such case you should disable it.
824 Default: 1
825
Linus Torvalds1da177e2005-04-16 15:20:36 -0700826icmp_echo_ignore_all - BOOLEAN
David S. Miller7ce312462005-10-03 16:07:30 -0700827 If set non-zero, then the kernel will ignore all ICMP ECHO
828 requests sent to it.
829 Default: 0
830
Linus Torvalds1da177e2005-04-16 15:20:36 -0700831icmp_echo_ignore_broadcasts - BOOLEAN
David S. Miller7ce312462005-10-03 16:07:30 -0700832 If set non-zero, then the kernel will ignore all ICMP ECHO and
833 TIMESTAMP requests sent to it via broadcast/multicast.
834 Default: 1
Linus Torvalds1da177e2005-04-16 15:20:36 -0700835
836icmp_ratelimit - INTEGER
837 Limit the maximal rates for sending ICMP packets whose type matches
838 icmp_ratemask (see below) to specific targets.
Stephen Hemminger6dbf4bc2008-07-01 19:29:07 -0700839 0 to disable any limiting,
840 otherwise the minimal space between responses in milliseconds.
Eric Dumazet4cdf5072014-09-19 07:38:40 -0700841 Note that another sysctl, icmp_msgs_per_sec limits the number
842 of ICMP packets sent on all targets.
Stephen Hemminger6dbf4bc2008-07-01 19:29:07 -0700843 Default: 1000
Linus Torvalds1da177e2005-04-16 15:20:36 -0700844
Eric Dumazet4cdf5072014-09-19 07:38:40 -0700845icmp_msgs_per_sec - INTEGER
846 Limit maximal number of ICMP packets sent per second from this host.
847 Only messages whose type matches icmp_ratemask (see below) are
848 controlled by this limit.
849 Default: 1000
850
851icmp_msgs_burst - INTEGER
852 icmp_msgs_per_sec controls number of ICMP packets sent per second,
853 while icmp_msgs_burst controls the burst size of these packets.
854 Default: 50
855
Linus Torvalds1da177e2005-04-16 15:20:36 -0700856icmp_ratemask - INTEGER
857 Mask made of ICMP types for which rates are being limited.
858 Significant bits: IHGFEDCBA9876543210
859 Default mask: 0000001100000011000 (6168)
860
861 Bit definitions (see include/linux/icmp.h):
862 0 Echo Reply
863 3 Destination Unreachable *
864 4 Source Quench *
865 5 Redirect
866 8 Echo Request
867 B Time Exceeded *
868 C Parameter Problem *
869 D Timestamp Request
870 E Timestamp Reply
871 F Info Request
872 G Info Reply
873 H Address Mask Request
874 I Address Mask Reply
875
876 * These are rate limited by default (see default mask above)
877
878icmp_ignore_bogus_error_responses - BOOLEAN
879 Some routers violate RFC1122 by sending bogus responses to broadcast
880 frames. Such violations are normally logged via a kernel warning.
881 If this is set to TRUE, the kernel will not give such warnings, which
882 will avoid log file clutter.
Rami Rosene8b265e2013-06-07 20:16:19 +0000883 Default: 1
Linus Torvalds1da177e2005-04-16 15:20:36 -0700884
Horms95f7daf2006-02-02 17:02:25 -0800885icmp_errors_use_inbound_ifaddr - BOOLEAN
886
887 If zero, icmp error messages are sent with the primary address of
888 the exiting interface.
Jesper Dangaard Brouere18f5fe2009-02-23 04:39:04 +0000889
Horms95f7daf2006-02-02 17:02:25 -0800890 If non-zero, the message will be sent with the primary address of
891 the interface that received the packet that caused the icmp error.
892 This is the behaviour network many administrators will expect from
893 a router. And it can make debugging complicated network layouts
Jesper Dangaard Brouere18f5fe2009-02-23 04:39:04 +0000894 much easier.
Horms95f7daf2006-02-02 17:02:25 -0800895
896 Note that if no primary address exists for the interface selected,
897 then the primary address of the first non-loopback interface that
Matt LaPlanted6bc8ac2006-10-03 22:54:15 +0200898 has one will be used regardless of this setting.
Horms95f7daf2006-02-02 17:02:25 -0800899
900 Default: 0
901
Linus Torvalds1da177e2005-04-16 15:20:36 -0700902igmp_max_memberships - INTEGER
903 Change the maximum number of multicast groups we can subscribe to.
904 Default: 20
905
Jeremy Ederd67ef352010-11-15 05:41:31 +0000906 Theoretical maximum value is bounded by having to send a membership
907 report in a single datagram (i.e. the report can't span multiple
908 datagrams, or risk confusing the switch and leaving groups you don't
909 intend to).
Linus Torvalds1da177e2005-04-16 15:20:36 -0700910
Jeremy Ederd67ef352010-11-15 05:41:31 +0000911 The number of supported groups 'M' is bounded by the number of group
912 report entries you can fit into a single datagram of 65535 bytes.
913
914 M = 65536-sizeof (ip header)/(sizeof(Group record))
915
916 Group records are variable length, with a minimum of 12 bytes.
917 So net.ipv4.igmp_max_memberships should not be set higher than:
918
919 (65536-24) / 12 = 5459
920
921 The value 5459 assumes no IP header options, so in practice
922 this number may be lower.
923
924 conf/interface/* changes special settings per interface (where
925 "interface" is the name of your network interface)
926
927 conf/all/* is special, changes the settings for all interfaces
Linus Torvalds1da177e2005-04-16 15:20:36 -0700928
Hannes Frederic Sowaa9fe8e22014-09-02 15:49:26 +0200929igmp_qrv - INTEGER
930 Controls the IGMP query robustness variable (see RFC2236 8.1).
931 Default: 2 (as specified by RFC2236 8.1)
932 Minimum: 1 (as specified by RFC6636 4.5)
933
Linus Torvalds1da177e2005-04-16 15:20:36 -0700934log_martians - BOOLEAN
935 Log packets with impossible addresses to kernel log.
936 log_martians for the interface will be enabled if at least one of
937 conf/{all,interface}/log_martians is set to TRUE,
938 it will be disabled otherwise
939
940accept_redirects - BOOLEAN
941 Accept ICMP redirect messages.
942 accept_redirects for the interface will be enabled if:
Jesper Dangaard Brouere18f5fe2009-02-23 04:39:04 +0000943 - both conf/{all,interface}/accept_redirects are TRUE in the case
944 forwarding for the interface is enabled
Linus Torvalds1da177e2005-04-16 15:20:36 -0700945 or
Jesper Dangaard Brouere18f5fe2009-02-23 04:39:04 +0000946 - at least one of conf/{all,interface}/accept_redirects is TRUE in the
947 case forwarding for the interface is disabled
Linus Torvalds1da177e2005-04-16 15:20:36 -0700948 accept_redirects for the interface will be disabled otherwise
949 default TRUE (host)
950 FALSE (router)
951
952forwarding - BOOLEAN
953 Enable IP forwarding on this interface.
954
955mc_forwarding - BOOLEAN
956 Do multicast routing. The kernel needs to be compiled with CONFIG_MROUTE
957 and a multicast routing daemon is required.
Jesper Dangaard Brouere18f5fe2009-02-23 04:39:04 +0000958 conf/all/mc_forwarding must also be set to TRUE to enable multicast
959 routing for the interface
Linus Torvalds1da177e2005-04-16 15:20:36 -0700960
961medium_id - INTEGER
962 Integer value used to differentiate the devices by the medium they
963 are attached to. Two devices can have different id values when
964 the broadcast packets are received only on one of them.
965 The default value 0 means that the device is the only interface
966 to its medium, value of -1 means that medium is not known.
Jesper Dangaard Brouere18f5fe2009-02-23 04:39:04 +0000967
Linus Torvalds1da177e2005-04-16 15:20:36 -0700968 Currently, it is used to change the proxy_arp behavior:
969 the proxy_arp feature is enabled for packets forwarded between
970 two devices attached to different media.
971
972proxy_arp - BOOLEAN
973 Do proxy arp.
974 proxy_arp for the interface will be enabled if at least one of
975 conf/{all,interface}/proxy_arp is set to TRUE,
976 it will be disabled otherwise
977
Jesper Dangaard Brouer65324142010-01-05 05:50:47 +0000978proxy_arp_pvlan - BOOLEAN
979 Private VLAN proxy arp.
980 Basically allow proxy arp replies back to the same interface
981 (from which the ARP request/solicitation was received).
982
983 This is done to support (ethernet) switch features, like RFC
984 3069, where the individual ports are NOT allowed to
985 communicate with each other, but they are allowed to talk to
986 the upstream router. As described in RFC 3069, it is possible
987 to allow these hosts to communicate through the upstream
988 router by proxy_arp'ing. Don't need to be used together with
989 proxy_arp.
990
991 This technology is known by different names:
992 In RFC 3069 it is called VLAN Aggregation.
993 Cisco and Allied Telesyn call it Private VLAN.
994 Hewlett-Packard call it Source-Port filtering or port-isolation.
995 Ericsson call it MAC-Forced Forwarding (RFC Draft).
996
Linus Torvalds1da177e2005-04-16 15:20:36 -0700997shared_media - BOOLEAN
998 Send(router) or accept(host) RFC1620 shared media redirects.
999 Overrides ip_secure_redirects.
1000 shared_media for the interface will be enabled if at least one of
1001 conf/{all,interface}/shared_media is set to TRUE,
1002 it will be disabled otherwise
1003 default TRUE
1004
1005secure_redirects - BOOLEAN
1006 Accept ICMP redirect messages only for gateways,
1007 listed in default gateway list.
1008 secure_redirects for the interface will be enabled if at least one of
1009 conf/{all,interface}/secure_redirects is set to TRUE,
1010 it will be disabled otherwise
1011 default TRUE
1012
1013send_redirects - BOOLEAN
1014 Send redirects, if router.
1015 send_redirects for the interface will be enabled if at least one of
1016 conf/{all,interface}/send_redirects is set to TRUE,
1017 it will be disabled otherwise
1018 Default: TRUE
1019
1020bootp_relay - BOOLEAN
1021 Accept packets with source address 0.b.c.d destined
1022 not to this host as local ones. It is supposed, that
1023 BOOTP relay daemon will catch and forward such packets.
1024 conf/all/bootp_relay must also be set to TRUE to enable BOOTP relay
1025 for the interface
1026 default FALSE
1027 Not Implemented Yet.
1028
1029accept_source_route - BOOLEAN
1030 Accept packets with SRR option.
1031 conf/all/accept_source_route must also be set to TRUE to accept packets
1032 with SRR option on the interface
1033 default TRUE (router)
1034 FALSE (host)
1035
Patrick McHardy8153a102009-12-03 01:25:58 +00001036accept_local - BOOLEAN
Sébastien Barré72b126a2014-09-10 18:20:23 +02001037 Accept packets with local source addresses. In combination with
1038 suitable routing, this can be used to direct packets between two
1039 local interfaces over the wire and have them accepted properly.
Patrick McHardy8153a102009-12-03 01:25:58 +00001040 default FALSE
1041
Thomas Grafd0daebc32012-06-12 00:44:01 +00001042route_localnet - BOOLEAN
1043 Do not consider loopback addresses as martian source or destination
1044 while routing. This enables the use of 127/8 for local routing purposes.
1045 default FALSE
1046
Stephen Hemmingerc1cf8422009-02-20 08:25:36 +00001047rp_filter - INTEGER
Linus Torvalds1da177e2005-04-16 15:20:36 -07001048 0 - No source validation.
Stephen Hemmingerc1cf8422009-02-20 08:25:36 +00001049 1 - Strict mode as defined in RFC3704 Strict Reverse Path
1050 Each incoming packet is tested against the FIB and if the interface
1051 is not the best reverse path the packet check will fail.
1052 By default failed packets are discarded.
1053 2 - Loose mode as defined in RFC3704 Loose Reverse Path
1054 Each incoming packet's source address is also tested against the FIB
1055 and if the source address is not reachable via any interface
1056 the packet check will fail.
Linus Torvalds1da177e2005-04-16 15:20:36 -07001057
Jesper Dangaard Brouere18f5fe2009-02-23 04:39:04 +00001058 Current recommended practice in RFC3704 is to enable strict mode
Jesper Dangaard Brouerbf869c32009-02-23 04:37:55 +00001059 to prevent IP spoofing from DDos attacks. If using asymmetric routing
Jesper Dangaard Brouere18f5fe2009-02-23 04:39:04 +00001060 or other complicated routing, then loose mode is recommended.
Stephen Hemmingerc1cf8422009-02-20 08:25:36 +00001061
Shan Wei1f5865e2009-12-02 15:39:04 -08001062 The max value from conf/{all,interface}/rp_filter is used
1063 when doing source validation on the {interface}.
Linus Torvalds1da177e2005-04-16 15:20:36 -07001064
1065 Default value is 0. Note that some distributions enable it
1066 in startup scripts.
1067
1068arp_filter - BOOLEAN
1069 1 - Allows you to have multiple network interfaces on the same
1070 subnet, and have the ARPs for each interface be answered
1071 based on whether or not the kernel would route a packet from
1072 the ARP'd IP out that interface (therefore you must use source
1073 based routing for this to work). In other words it allows control
1074 of which cards (usually 1) will respond to an arp request.
1075
1076 0 - (default) The kernel can respond to arp requests with addresses
1077 from other interfaces. This may seem wrong but it usually makes
1078 sense, because it increases the chance of successful communication.
1079 IP addresses are owned by the complete host on Linux, not by
1080 particular interfaces. Only for more complex setups like load-
1081 balancing, does this behaviour cause problems.
1082
1083 arp_filter for the interface will be enabled if at least one of
1084 conf/{all,interface}/arp_filter is set to TRUE,
1085 it will be disabled otherwise
1086
1087arp_announce - INTEGER
1088 Define different restriction levels for announcing the local
1089 source IP address from IP packets in ARP requests sent on
1090 interface:
1091 0 - (default) Use any local address, configured on any interface
1092 1 - Try to avoid local addresses that are not in the target's
1093 subnet for this interface. This mode is useful when target
1094 hosts reachable via this interface require the source IP
1095 address in ARP requests to be part of their logical network
1096 configured on the receiving interface. When we generate the
1097 request we will check all our subnets that include the
1098 target IP and will preserve the source address if it is from
1099 such subnet. If there is no such subnet we select source
1100 address according to the rules for level 2.
1101 2 - Always use the best local address for this target.
1102 In this mode we ignore the source address in the IP packet
1103 and try to select local address that we prefer for talks with
1104 the target host. Such local address is selected by looking
1105 for primary IP addresses on all our subnets on the outgoing
1106 interface that include the target IP address. If no suitable
1107 local address is found we select the first local address
1108 we have on the outgoing interface or on all other interfaces,
1109 with the hope we will receive reply for our request and
1110 even sometimes no matter the source IP address we announce.
1111
1112 The max value from conf/{all,interface}/arp_announce is used.
1113
1114 Increasing the restriction level gives more chance for
1115 receiving answer from the resolved target while decreasing
1116 the level announces more valid sender's information.
1117
1118arp_ignore - INTEGER
1119 Define different modes for sending replies in response to
1120 received ARP requests that resolve local target IP addresses:
1121 0 - (default): reply for any local target IP address, configured
1122 on any interface
1123 1 - reply only if the target IP address is local address
1124 configured on the incoming interface
1125 2 - reply only if the target IP address is local address
1126 configured on the incoming interface and both with the
1127 sender's IP address are part from same subnet on this interface
1128 3 - do not reply for local addresses configured with scope host,
1129 only resolutions for global and link addresses are replied
1130 4-7 - reserved
1131 8 - do not reply for all local addresses
1132
1133 The max value from conf/{all,interface}/arp_ignore is used
1134 when ARP request is received on the {interface}
1135
Stephen Hemmingereefef1c2009-02-01 01:04:33 -08001136arp_notify - BOOLEAN
1137 Define mode for notification of address and device changes.
1138 0 - (default): do nothing
Ian Campbell3f8dc232010-05-26 00:09:41 +00001139 1 - Generate gratuitous arp requests when device is brought up
Stephen Hemmingereefef1c2009-02-01 01:04:33 -08001140 or hardware address changes.
1141
Neil Hormanc1b1bce2006-03-20 22:40:03 -08001142arp_accept - BOOLEAN
Octavian Purdila6d955182010-01-18 12:58:44 +00001143 Define behavior for gratuitous ARP frames who's IP is not
1144 already present in the ARP table:
1145 0 - don't create new entries in the ARP table
1146 1 - create new entries in the ARP table
1147
1148 Both replies and requests type gratuitous arp will trigger the
1149 ARP table to be updated, if this setting is on.
1150
1151 If the ARP table already contains the IP address of the
1152 gratuitous arp frame, the arp table will be updated regardless
1153 if this setting is on or off.
1154
YOSHIFUJI Hideaki/吉藤英明89c69d32015-03-19 22:42:04 +09001155mcast_solicit - INTEGER
1156 The maximum number of multicast probes in INCOMPLETE state,
1157 when the associated hardware address is unknown. Defaults
1158 to 3.
1159
1160ucast_solicit - INTEGER
1161 The maximum number of unicast probes in PROBE state, when
1162 the hardware address is being reconfirmed. Defaults to 3.
Neil Hormanc1b1bce2006-03-20 22:40:03 -08001163
Linus Torvalds1da177e2005-04-16 15:20:36 -07001164app_solicit - INTEGER
1165 The maximum number of probes to send to the user space ARP daemon
1166 via netlink before dropping back to multicast probes (see
YOSHIFUJI Hideaki/吉藤英明89c69d32015-03-19 22:42:04 +09001167 mcast_resolicit). Defaults to 0.
1168
1169mcast_resolicit - INTEGER
1170 The maximum number of multicast probes after unicast and
1171 app probes in PROBE state. Defaults to 0.
Linus Torvalds1da177e2005-04-16 15:20:36 -07001172
1173disable_policy - BOOLEAN
1174 Disable IPSEC policy (SPD) for this interface
1175
1176disable_xfrm - BOOLEAN
1177 Disable IPSEC encryption on this interface, whatever the policy
1178
Hannes Frederic Sowafc4eba52013-08-14 01:03:46 +02001179igmpv2_unsolicited_report_interval - INTEGER
1180 The interval in milliseconds in which the next unsolicited
1181 IGMPv1 or IGMPv2 report retransmit will take place.
1182 Default: 10000 (10 seconds)
Linus Torvalds1da177e2005-04-16 15:20:36 -07001183
Hannes Frederic Sowafc4eba52013-08-14 01:03:46 +02001184igmpv3_unsolicited_report_interval - INTEGER
1185 The interval in milliseconds in which the next unsolicited
1186 IGMPv3 report retransmit will take place.
1187 Default: 1000 (1 seconds)
Linus Torvalds1da177e2005-04-16 15:20:36 -07001188
Martin Schwenked922e1c2014-01-28 15:26:42 +11001189promote_secondaries - BOOLEAN
1190 When a primary IP address is removed from this interface
1191 promote a corresponding secondary IP address instead of
1192 removing all the corresponding secondary IP addresses.
1193
1194
Linus Torvalds1da177e2005-04-16 15:20:36 -07001195tag - INTEGER
1196 Allows you to write a number, which can be used as required.
1197 Default value is 0.
1198
Alexander Duycke69948a2015-08-11 13:35:01 -07001199xfrm4_gc_thresh - INTEGER
1200 The threshold at which we will start garbage collecting for IPv4
1201 destination cache entries. At twice this value the system will
1202 refuse new allocations.
1203
Philip Downey87583eb2015-08-31 11:30:38 +01001204igmp_link_local_mcast_reports - BOOLEAN
1205 Enable IGMP reports for link local multicast groups in the
1206 224.0.0.X range.
1207 Default TRUE
1208
Linus Torvalds1da177e2005-04-16 15:20:36 -07001209Alexey Kuznetsov.
1210kuznet@ms2.inr.ac.ru
1211
1212Updated by:
1213Andi Kleen
1214ak@muc.de
1215Nicolas Delon
1216delon.nicolas@wanadoo.fr
1217
1218
1219
1220
1221/proc/sys/net/ipv6/* Variables:
1222
1223IPv6 has no global variables such as tcp_*. tcp_* settings under ipv4/ also
1224apply to IPv6 [XXX?].
1225
1226bindv6only - BOOLEAN
1227 Default value for IPV6_V6ONLY socket option,
Jesper Dangaard Brouere18f5fe2009-02-23 04:39:04 +00001228 which restricts use of the IPv6 socket to IPv6 communication
Linus Torvalds1da177e2005-04-16 15:20:36 -07001229 only.
1230 TRUE: disable IPv4-mapped address feature
1231 FALSE: enable IPv4-mapped address feature
1232
Geoffrey Thomasd5c073ca2011-08-22 11:28:57 -07001233 Default: FALSE (as specified in RFC3493)
Linus Torvalds1da177e2005-04-16 15:20:36 -07001234
Florent Fourcot6444f722014-01-17 17:15:05 +01001235flowlabel_consistency - BOOLEAN
1236 Protect the consistency (and unicity) of flow label.
1237 You have to disable it to use IPV6_FL_F_REFLECT flag on the
1238 flow label manager.
1239 TRUE: enabled
1240 FALSE: disabled
1241 Default: TRUE
1242
Tom Herbert42240902015-07-31 16:52:12 -07001243auto_flowlabels - INTEGER
1244 Automatically generate flow labels based on a flow hash of the
1245 packet. This allows intermediate devices, such as routers, to
1246 identify packet flows for mechanisms like Equal Cost Multipath
Tom Herbertcb1ce2e2014-07-01 21:33:10 -07001247 Routing (see RFC 6438).
Tom Herbert42240902015-07-31 16:52:12 -07001248 0: automatic flow labels are completely disabled
1249 1: automatic flow labels are enabled by default, they can be
1250 disabled on a per socket basis using the IPV6_AUTOFLOWLABEL
1251 socket option
1252 2: automatic flow labels are allowed, they may be enabled on a
1253 per socket basis using the IPV6_AUTOFLOWLABEL socket option
1254 3: automatic flow labels are enabled and enforced, they cannot
1255 be disabled by the socket option
Tom Herbertb5677412015-07-31 16:52:14 -07001256 Default: 1
Tom Herbertcb1ce2e2014-07-01 21:33:10 -07001257
Tom Herbert82a584b2015-04-29 15:33:21 -07001258flowlabel_state_ranges - BOOLEAN
1259 Split the flow label number space into two ranges. 0-0x7FFFF is
1260 reserved for the IPv6 flow manager facility, 0x80000-0xFFFFF
1261 is reserved for stateless flow labels as described in RFC6437.
1262 TRUE: enabled
1263 FALSE: disabled
1264 Default: true
1265
FX Le Bail509aba32014-01-07 14:57:27 +01001266anycast_src_echo_reply - BOOLEAN
1267 Controls the use of anycast addresses as source addresses for ICMPv6
1268 echo reply
1269 TRUE: enabled
1270 FALSE: disabled
1271 Default: FALSE
1272
Hannes Frederic Sowa9f0761c2015-03-23 23:36:06 +01001273idgen_delay - INTEGER
1274 Controls the delay in seconds after which time to retry
1275 privacy stable address generation if a DAD conflict is
1276 detected.
1277 Default: 1 (as specified in RFC7217)
1278
1279idgen_retries - INTEGER
1280 Controls the number of retries to generate a stable privacy
1281 address if a DAD conflict is detected.
1282 Default: 3 (as specified in RFC7217)
1283
Hannes Frederic Sowa2f711932014-09-02 15:49:25 +02001284mld_qrv - INTEGER
1285 Controls the MLD query robustness variable (see RFC3810 9.1).
1286 Default: 2 (as specified by RFC3810 9.1)
1287 Minimum: 1 (as specified by RFC6636 4.5)
1288
Linus Torvalds1da177e2005-04-16 15:20:36 -07001289IPv6 Fragmentation:
1290
1291ip6frag_high_thresh - INTEGER
Jesper Dangaard Brouere18f5fe2009-02-23 04:39:04 +00001292 Maximum memory used to reassemble IPv6 fragments. When
Linus Torvalds1da177e2005-04-16 15:20:36 -07001293 ip6frag_high_thresh bytes of memory is allocated for this purpose,
1294 the fragment handler will toss packets until ip6frag_low_thresh
1295 is reached.
Jesper Dangaard Brouere18f5fe2009-02-23 04:39:04 +00001296
Linus Torvalds1da177e2005-04-16 15:20:36 -07001297ip6frag_low_thresh - INTEGER
Jesper Dangaard Brouere18f5fe2009-02-23 04:39:04 +00001298 See ip6frag_high_thresh
Linus Torvalds1da177e2005-04-16 15:20:36 -07001299
1300ip6frag_time - INTEGER
1301 Time in seconds to keep an IPv6 fragment in memory.
1302
Linus Torvalds1da177e2005-04-16 15:20:36 -07001303conf/default/*:
1304 Change the interface-specific default settings.
1305
1306
1307conf/all/*:
Jesper Dangaard Brouere18f5fe2009-02-23 04:39:04 +00001308 Change all the interface-specific settings.
Linus Torvalds1da177e2005-04-16 15:20:36 -07001309
1310 [XXX: Other special features than forwarding?]
1311
1312conf/all/forwarding - BOOLEAN
Jesper Dangaard Brouere18f5fe2009-02-23 04:39:04 +00001313 Enable global IPv6 forwarding between all interfaces.
Linus Torvalds1da177e2005-04-16 15:20:36 -07001314
Jesper Dangaard Brouere18f5fe2009-02-23 04:39:04 +00001315 IPv4 and IPv6 work differently here; e.g. netfilter must be used
Linus Torvalds1da177e2005-04-16 15:20:36 -07001316 to control which interfaces may forward packets and which not.
1317
Jesper Dangaard Brouere18f5fe2009-02-23 04:39:04 +00001318 This also sets all interfaces' Host/Router setting
Linus Torvalds1da177e2005-04-16 15:20:36 -07001319 'forwarding' to the specified value. See below for details.
1320
1321 This referred to as global forwarding.
1322
YOSHIFUJI Hideakifbea49e2006-09-22 14:43:49 -07001323proxy_ndp - BOOLEAN
1324 Do proxy ndp.
1325
Loganaden Velvindron219b5f22014-11-04 03:02:49 -08001326fwmark_reflect - BOOLEAN
1327 Controls the fwmark of kernel-generated IPv6 reply packets that are not
1328 associated with a socket for example, TCP RSTs or ICMPv6 echo replies).
1329 If unset, these packets have a fwmark of zero. If set, they have the
1330 fwmark of the packet they are replying to.
1331 Default: 0
1332
Linus Torvalds1da177e2005-04-16 15:20:36 -07001333conf/interface/*:
1334 Change special settings per interface.
1335
Jesper Dangaard Brouere18f5fe2009-02-23 04:39:04 +00001336 The functional behaviour for certain settings is different
Linus Torvalds1da177e2005-04-16 15:20:36 -07001337 depending on whether local forwarding is enabled or not.
1338
Roy.Li605b91c2011-09-28 19:51:54 +00001339accept_ra - INTEGER
Linus Torvalds1da177e2005-04-16 15:20:36 -07001340 Accept Router Advertisements; autoconfigure using them.
Jesper Dangaard Brouere18f5fe2009-02-23 04:39:04 +00001341
Tore Anderson026359b2011-08-28 23:47:33 +00001342 It also determines whether or not to transmit Router
1343 Solicitations. If and only if the functional setting is to
1344 accept Router Advertisements, Router Solicitations will be
1345 transmitted.
1346
Thomas Grafae8abfa2010-09-03 05:47:30 +00001347 Possible values are:
1348 0 Do not accept Router Advertisements.
1349 1 Accept Router Advertisements if forwarding is disabled.
1350 2 Overrule forwarding behaviour. Accept Router Advertisements
1351 even if forwarding is enabled.
1352
Linus Torvalds1da177e2005-04-16 15:20:36 -07001353 Functional default: enabled if local forwarding is disabled.
1354 disabled if local forwarding is enabled.
1355
YOSHIFUJI Hideaki65f5c7c2006-03-20 16:55:08 -08001356accept_ra_defrtr - BOOLEAN
1357 Learn default router in Router Advertisement.
1358
1359 Functional default: enabled if accept_ra is enabled.
1360 disabled if accept_ra is disabled.
1361
Ben Greeard9333192014-06-25 14:44:53 -07001362accept_ra_from_local - BOOLEAN
1363 Accept RA with source-address that is found on local machine
1364 if the RA is otherwise proper and able to be accepted.
1365 Default is to NOT accept these as it may be an un-intended
1366 network loop.
1367
1368 Functional default:
1369 enabled if accept_ra_from_local is enabled
1370 on a specific interface.
1371 disabled if accept_ra_from_local is disabled
1372 on a specific interface.
1373
Hangbin Liu8013d1d2015-07-30 14:28:42 +08001374accept_ra_min_hop_limit - INTEGER
1375 Minimum hop limit Information in Router Advertisement.
1376
1377 Hop limit Information in Router Advertisement less than this
1378 variable shall be ignored.
1379
1380 Default: 1
1381
YOSHIFUJI Hideakic4fd30e2006-03-20 16:55:26 -08001382accept_ra_pinfo - BOOLEAN
Matt LaPlante2fe0ae72006-10-03 22:50:39 +02001383 Learn Prefix Information in Router Advertisement.
YOSHIFUJI Hideakic4fd30e2006-03-20 16:55:26 -08001384
1385 Functional default: enabled if accept_ra is enabled.
1386 disabled if accept_ra is disabled.
1387
YOSHIFUJI Hideaki09c884d2006-03-20 17:07:03 -08001388accept_ra_rt_info_max_plen - INTEGER
1389 Maximum prefix length of Route Information in RA.
1390
1391 Route Information w/ prefix larger than or equal to this
1392 variable shall be ignored.
1393
1394 Functional default: 0 if accept_ra_rtr_pref is enabled.
1395 -1 if accept_ra_rtr_pref is disabled.
1396
YOSHIFUJI Hideaki930d6ff2006-03-20 17:05:30 -08001397accept_ra_rtr_pref - BOOLEAN
1398 Accept Router Preference in RA.
1399
1400 Functional default: enabled if accept_ra is enabled.
1401 disabled if accept_ra is disabled.
1402
Harout Hedeshianc2943f12015-01-20 10:06:05 -07001403accept_ra_mtu - BOOLEAN
1404 Apply the MTU value specified in RA option 5 (RFC4861). If
1405 disabled, the MTU specified in the RA will be ignored.
1406
1407 Functional default: enabled if accept_ra is enabled.
1408 disabled if accept_ra is disabled.
1409
Linus Torvalds1da177e2005-04-16 15:20:36 -07001410accept_redirects - BOOLEAN
1411 Accept Redirects.
1412
1413 Functional default: enabled if local forwarding is disabled.
1414 disabled if local forwarding is enabled.
1415
YOSHIFUJI Hideaki0bcbc922007-04-24 14:58:30 -07001416accept_source_route - INTEGER
1417 Accept source routing (routing extension header).
1418
YOSHIFUJI Hideakibb4dbf92007-07-10 22:55:49 -07001419 >= 0: Accept only routing header type 2.
YOSHIFUJI Hideaki0bcbc922007-04-24 14:58:30 -07001420 < 0: Do not accept routing header.
1421
1422 Default: 0
1423
Linus Torvalds1da177e2005-04-16 15:20:36 -07001424autoconf - BOOLEAN
Jesper Dangaard Brouere18f5fe2009-02-23 04:39:04 +00001425 Autoconfigure addresses using Prefix Information in Router
Linus Torvalds1da177e2005-04-16 15:20:36 -07001426 Advertisements.
1427
YOSHIFUJI Hideakic4fd30e2006-03-20 16:55:26 -08001428 Functional default: enabled if accept_ra_pinfo is enabled.
1429 disabled if accept_ra_pinfo is disabled.
Linus Torvalds1da177e2005-04-16 15:20:36 -07001430
1431dad_transmits - INTEGER
1432 The amount of Duplicate Address Detection probes to send.
1433 Default: 1
Linus Torvalds1da177e2005-04-16 15:20:36 -07001434
Roy.Li605b91c2011-09-28 19:51:54 +00001435forwarding - INTEGER
Jesper Dangaard Brouere18f5fe2009-02-23 04:39:04 +00001436 Configure interface-specific Host/Router behaviour.
1437
1438 Note: It is recommended to have the same setting on all
Linus Torvalds1da177e2005-04-16 15:20:36 -07001439 interfaces; mixed router/host scenarios are rather uncommon.
1440
Thomas Grafae8abfa2010-09-03 05:47:30 +00001441 Possible values are:
1442 0 Forwarding disabled
1443 1 Forwarding enabled
Thomas Grafae8abfa2010-09-03 05:47:30 +00001444
1445 FALSE (0):
Linus Torvalds1da177e2005-04-16 15:20:36 -07001446
1447 By default, Host behaviour is assumed. This means:
1448
1449 1. IsRouter flag is not set in Neighbour Advertisements.
Tore Anderson026359b2011-08-28 23:47:33 +00001450 2. If accept_ra is TRUE (default), transmit Router
1451 Solicitations.
Jesper Dangaard Brouere18f5fe2009-02-23 04:39:04 +00001452 3. If accept_ra is TRUE (default), accept Router
Linus Torvalds1da177e2005-04-16 15:20:36 -07001453 Advertisements (and do autoconfiguration).
1454 4. If accept_redirects is TRUE (default), accept Redirects.
1455
Thomas Grafae8abfa2010-09-03 05:47:30 +00001456 TRUE (1):
Linus Torvalds1da177e2005-04-16 15:20:36 -07001457
Jesper Dangaard Brouere18f5fe2009-02-23 04:39:04 +00001458 If local forwarding is enabled, Router behaviour is assumed.
Linus Torvalds1da177e2005-04-16 15:20:36 -07001459 This means exactly the reverse from the above:
1460
1461 1. IsRouter flag is set in Neighbour Advertisements.
Tore Anderson026359b2011-08-28 23:47:33 +00001462 2. Router Solicitations are not sent unless accept_ra is 2.
Thomas Grafae8abfa2010-09-03 05:47:30 +00001463 3. Router Advertisements are ignored unless accept_ra is 2.
Linus Torvalds1da177e2005-04-16 15:20:36 -07001464 4. Redirects are ignored.
1465
Thomas Grafae8abfa2010-09-03 05:47:30 +00001466 Default: 0 (disabled) if global forwarding is disabled (default),
1467 otherwise 1 (enabled).
Linus Torvalds1da177e2005-04-16 15:20:36 -07001468
1469hop_limit - INTEGER
1470 Default Hop Limit to set.
1471 Default: 64
1472
1473mtu - INTEGER
1474 Default Maximum Transfer Unit
1475 Default: 1280 (IPv6 required minimum)
1476
Tom Herbert35a256f2015-07-08 16:58:22 -07001477ip_nonlocal_bind - BOOLEAN
1478 If set, allows processes to bind() to non-local IPv6 addresses,
1479 which can be quite useful - but may break some applications.
1480 Default: 0
1481
YOSHIFUJI Hideaki52e163562006-03-20 17:05:47 -08001482router_probe_interval - INTEGER
1483 Minimum interval (in seconds) between Router Probing described
1484 in RFC4191.
1485
1486 Default: 60
1487
Linus Torvalds1da177e2005-04-16 15:20:36 -07001488router_solicitation_delay - INTEGER
1489 Number of seconds to wait after interface is brought up
1490 before sending Router Solicitations.
1491 Default: 1
1492
1493router_solicitation_interval - INTEGER
1494 Number of seconds to wait between Router Solicitations.
1495 Default: 4
1496
1497router_solicitations - INTEGER
Jesper Dangaard Brouere18f5fe2009-02-23 04:39:04 +00001498 Number of Router Solicitations to send until assuming no
Linus Torvalds1da177e2005-04-16 15:20:36 -07001499 routers are present.
1500 Default: 3
1501
Erik Kline3985e8a2015-07-22 16:38:25 +09001502use_oif_addrs_only - BOOLEAN
1503 When enabled, the candidate source addresses for destinations
1504 routed via this interface are restricted to the set of addresses
1505 configured on this interface (vis. RFC 6724, section 4).
1506
1507 Default: false
1508
Linus Torvalds1da177e2005-04-16 15:20:36 -07001509use_tempaddr - INTEGER
1510 Preference for Privacy Extensions (RFC3041).
1511 <= 0 : disable Privacy Extensions
1512 == 1 : enable Privacy Extensions, but prefer public
1513 addresses over temporary addresses.
1514 > 1 : enable Privacy Extensions and prefer temporary
1515 addresses over public addresses.
1516 Default: 0 (for most devices)
1517 -1 (for point-to-point devices and loopback devices)
1518
1519temp_valid_lft - INTEGER
1520 valid lifetime (in seconds) for temporary addresses.
1521 Default: 604800 (7 days)
1522
1523temp_prefered_lft - INTEGER
1524 Preferred lifetime (in seconds) for temporary addresses.
1525 Default: 86400 (1 day)
1526
1527max_desync_factor - INTEGER
1528 Maximum value for DESYNC_FACTOR, which is a random value
Jesper Dangaard Brouere18f5fe2009-02-23 04:39:04 +00001529 that ensures that clients don't synchronize with each
Linus Torvalds1da177e2005-04-16 15:20:36 -07001530 other and generate new addresses at exactly the same time.
1531 value is in seconds.
1532 Default: 600
Jesper Dangaard Brouere18f5fe2009-02-23 04:39:04 +00001533
Linus Torvalds1da177e2005-04-16 15:20:36 -07001534regen_max_retry - INTEGER
1535 Number of attempts before give up attempting to generate
1536 valid temporary addresses.
1537 Default: 5
1538
1539max_addresses - INTEGER
Brian Haleye79dc482010-02-22 12:27:21 +00001540 Maximum number of autoconfigured addresses per interface. Setting
1541 to zero disables the limitation. It is not recommended to set this
1542 value too large (or to zero) because it would be an easy way to
1543 crash the kernel by allowing too many addresses to be created.
Linus Torvalds1da177e2005-04-16 15:20:36 -07001544 Default: 16
1545
YOSHIFUJI Hideaki778d80b2008-06-28 14:17:11 +09001546disable_ipv6 - BOOLEAN
Brian Haley9bdd8d42009-03-18 18:22:48 -07001547 Disable IPv6 operation. If accept_dad is set to 2, this value
1548 will be dynamically set to TRUE if DAD fails for the link-local
1549 address.
YOSHIFUJI Hideaki778d80b2008-06-28 14:17:11 +09001550 Default: FALSE (enable IPv6 operation)
1551
Brian Haley56d417b2009-06-01 03:07:33 -07001552 When this value is changed from 1 to 0 (IPv6 is being enabled),
1553 it will dynamically create a link-local address on the given
1554 interface and start Duplicate Address Detection, if necessary.
1555
1556 When this value is changed from 0 to 1 (IPv6 is being disabled),
1557 it will dynamically delete all address on the given interface.
1558
YOSHIFUJI Hideaki1b34be72008-06-28 14:18:38 +09001559accept_dad - INTEGER
1560 Whether to accept DAD (Duplicate Address Detection).
1561 0: Disable DAD
1562 1: Enable DAD (default)
1563 2: Enable DAD, and disable IPv6 operation if MAC-based duplicate
1564 link-local address has been found.
1565
Octavian Purdilaf7734fd2009-10-02 11:39:15 +00001566force_tllao - BOOLEAN
1567 Enable sending the target link-layer address option even when
1568 responding to a unicast neighbor solicitation.
1569 Default: FALSE
1570
1571 Quoting from RFC 2461, section 4.4, Target link-layer address:
1572
1573 "The option MUST be included for multicast solicitations in order to
1574 avoid infinite Neighbor Solicitation "recursion" when the peer node
1575 does not have a cache entry to return a Neighbor Advertisements
1576 message. When responding to unicast solicitations, the option can be
1577 omitted since the sender of the solicitation has the correct link-
1578 layer address; otherwise it would not have be able to send the unicast
1579 solicitation in the first place. However, including the link-layer
1580 address in this case adds little overhead and eliminates a potential
1581 race condition where the sender deletes the cached link-layer address
1582 prior to receiving a response to a previous solicitation."
1583
Hannes Frederic Sowadb2b6202013-01-01 00:35:31 +00001584ndisc_notify - BOOLEAN
1585 Define mode for notification of address and device changes.
1586 0 - (default): do nothing
1587 1 - Generate unsolicited neighbour advertisements when device is brought
1588 up or hardware address changes.
1589
Hannes Frederic Sowafc4eba52013-08-14 01:03:46 +02001590mldv1_unsolicited_report_interval - INTEGER
1591 The interval in milliseconds in which the next unsolicited
1592 MLDv1 report retransmit will take place.
1593 Default: 10000 (10 seconds)
1594
1595mldv2_unsolicited_report_interval - INTEGER
1596 The interval in milliseconds in which the next unsolicited
1597 MLDv2 report retransmit will take place.
1598 Default: 1000 (1 second)
1599
Daniel Borkmannf2127812013-09-04 00:19:44 +02001600force_mld_version - INTEGER
1601 0 - (default) No enforcement of a MLD version, MLDv1 fallback allowed
1602 1 - Enforce to use MLD version 1
1603 2 - Enforce to use MLD version 2
1604
Hannes Frederic Sowab800c3b2013-08-27 01:36:51 +02001605suppress_frag_ndisc - INTEGER
1606 Control RFC 6980 (Security Implications of IPv6 Fragmentation
1607 with IPv6 Neighbor Discovery) behavior:
1608 1 - (default) discard fragmented neighbor discovery packets
1609 0 - allow fragmented neighbor discovery packets
1610
Erik Kline7fd25612014-10-28 18:11:14 +09001611optimistic_dad - BOOLEAN
1612 Whether to perform Optimistic Duplicate Address Detection (RFC 4429).
1613 0: disabled (default)
1614 1: enabled
1615
1616use_optimistic - BOOLEAN
1617 If enabled, do not classify optimistic addresses as deprecated during
1618 source address selection. Preferred addresses will still be chosen
1619 before optimistic addresses, subject to other ranking in the source
1620 address selection algorithm.
1621 0: disabled (default)
1622 1: enabled
1623
Hannes Frederic Sowa9f0761c2015-03-23 23:36:06 +01001624stable_secret - IPv6 address
1625 This IPv6 address will be used as a secret to generate IPv6
1626 addresses for link-local addresses and autoconfigured
1627 ones. All addresses generated after setting this secret will
1628 be stable privacy ones by default. This can be changed via the
1629 addrgenmode ip-link. conf/default/stable_secret is used as the
1630 secret for the namespace, the interface specific ones can
1631 overwrite that. Writes to conf/all/stable_secret are refused.
1632
1633 It is recommended to generate this secret during installation
1634 of a system and keep it stable after that.
1635
1636 By default the stable secret is unset.
1637
Linus Torvalds1da177e2005-04-16 15:20:36 -07001638icmp/*:
1639ratelimit - INTEGER
1640 Limit the maximal rates for sending ICMPv6 packets.
Stephen Hemminger6dbf4bc2008-07-01 19:29:07 -07001641 0 to disable any limiting,
1642 otherwise the minimal space between responses in milliseconds.
1643 Default: 1000
Linus Torvalds1da177e2005-04-16 15:20:36 -07001644
Alexander Duycke69948a2015-08-11 13:35:01 -07001645xfrm6_gc_thresh - INTEGER
1646 The threshold at which we will start garbage collecting for IPv6
1647 destination cache entries. At twice this value the system will
1648 refuse new allocations.
1649
Linus Torvalds1da177e2005-04-16 15:20:36 -07001650
1651IPv6 Update by:
1652Pekka Savola <pekkas@netcore.fi>
1653YOSHIFUJI Hideaki / USAGI Project <yoshfuji@linux-ipv6.org>
1654
1655
1656/proc/sys/net/bridge/* Variables:
1657
1658bridge-nf-call-arptables - BOOLEAN
1659 1 : pass bridged ARP traffic to arptables' FORWARD chain.
1660 0 : disable this.
1661 Default: 1
1662
1663bridge-nf-call-iptables - BOOLEAN
1664 1 : pass bridged IPv4 traffic to iptables' chains.
1665 0 : disable this.
1666 Default: 1
1667
1668bridge-nf-call-ip6tables - BOOLEAN
1669 1 : pass bridged IPv6 traffic to ip6tables' chains.
1670 0 : disable this.
1671 Default: 1
1672
1673bridge-nf-filter-vlan-tagged - BOOLEAN
Michael Milner516299d2007-04-12 22:14:23 -07001674 1 : pass bridged vlan-tagged ARP/IP/IPv6 traffic to {arp,ip,ip6}tables.
1675 0 : disable this.
Pablo Neira Ayuso49816822012-05-08 19:36:44 +02001676 Default: 0
Michael Milner516299d2007-04-12 22:14:23 -07001677
1678bridge-nf-filter-pppoe-tagged - BOOLEAN
1679 1 : pass bridged pppoe-tagged IP/IPv6 traffic to {ip,ip6}tables.
Linus Torvalds1da177e2005-04-16 15:20:36 -07001680 0 : disable this.
Pablo Neira Ayuso49816822012-05-08 19:36:44 +02001681 Default: 0
Linus Torvalds1da177e2005-04-16 15:20:36 -07001682
Pablo Neira Ayuso49816822012-05-08 19:36:44 +02001683bridge-nf-pass-vlan-input-dev - BOOLEAN
1684 1: if bridge-nf-filter-vlan-tagged is enabled, try to find a vlan
1685 interface on the bridge and set the netfilter input device to the vlan.
1686 This allows use of e.g. "iptables -i br0.1" and makes the REDIRECT
1687 target work with vlan-on-top-of-bridge interfaces. When no matching
1688 vlan interface is found, or this switch is off, the input device is
1689 set to the bridge interface.
1690 0: disable bridge netfilter vlan interface lookup.
1691 Default: 0
Linus Torvalds1da177e2005-04-16 15:20:36 -07001692
Vlad Yasevich32e8d492008-07-08 16:43:29 -07001693proc/sys/net/sctp/* Variables:
1694
1695addip_enable - BOOLEAN
1696 Enable or disable extension of Dynamic Address Reconfiguration
1697 (ADD-IP) functionality specified in RFC5061. This extension provides
1698 the ability to dynamically add and remove new addresses for the SCTP
1699 associations.
1700
1701 1: Enable extension.
1702
1703 0: Disable extension.
1704
1705 Default: 0
1706
1707addip_noauth_enable - BOOLEAN
1708 Dynamic Address Reconfiguration (ADD-IP) requires the use of
1709 authentication to protect the operations of adding or removing new
1710 addresses. This requirement is mandated so that unauthorized hosts
1711 would not be able to hijack associations. However, older
1712 implementations may not have implemented this requirement while
1713 allowing the ADD-IP extension. For reasons of interoperability,
1714 we provide this variable to control the enforcement of the
1715 authentication requirement.
1716
1717 1: Allow ADD-IP extension to be used without authentication. This
1718 should only be set in a closed environment for interoperability
1719 with older implementations.
1720
1721 0: Enforce the authentication requirement
1722
1723 Default: 0
1724
1725auth_enable - BOOLEAN
1726 Enable or disable Authenticated Chunks extension. This extension
1727 provides the ability to send and receive authenticated chunks and is
1728 required for secure operation of Dynamic Address Reconfiguration
1729 (ADD-IP) extension.
1730
1731 1: Enable this extension.
1732 0: Disable this extension.
1733
1734 Default: 0
1735
1736prsctp_enable - BOOLEAN
1737 Enable or disable the Partial Reliability extension (RFC3758) which
1738 is used to notify peers that a given DATA should no longer be expected.
1739
1740 1: Enable extension
1741 0: Disable
1742
1743 Default: 1
1744
1745max_burst - INTEGER
1746 The limit of the number of new packets that can be initially sent. It
1747 controls how bursty the generated traffic can be.
1748
1749 Default: 4
1750
1751association_max_retrans - INTEGER
1752 Set the maximum number for retransmissions that an association can
1753 attempt deciding that the remote end is unreachable. If this value
1754 is exceeded, the association is terminated.
1755
1756 Default: 10
1757
1758max_init_retransmits - INTEGER
1759 The maximum number of retransmissions of INIT and COOKIE-ECHO chunks
1760 that an association will attempt before declaring the destination
1761 unreachable and terminating.
1762
1763 Default: 8
1764
1765path_max_retrans - INTEGER
1766 The maximum number of retransmissions that will be attempted on a given
1767 path. Once this threshold is exceeded, the path is considered
1768 unreachable, and new traffic will use a different path when the
1769 association is multihomed.
1770
1771 Default: 5
1772
Neil Horman5aa93bc2012-07-21 07:56:07 +00001773pf_retrans - INTEGER
1774 The number of retransmissions that will be attempted on a given path
1775 before traffic is redirected to an alternate transport (should one
1776 exist). Note this is distinct from path_max_retrans, as a path that
1777 passes the pf_retrans threshold can still be used. Its only
1778 deprioritized when a transmission path is selected by the stack. This
1779 setting is primarily used to enable fast failover mechanisms without
1780 having to reduce path_max_retrans to a very low value. See:
1781 http://www.ietf.org/id/draft-nishida-tsvwg-sctp-failover-05.txt
1782 for details. Note also that a value of pf_retrans > path_max_retrans
1783 disables this feature
1784
1785 Default: 0
1786
Vlad Yasevich32e8d492008-07-08 16:43:29 -07001787rto_initial - INTEGER
1788 The initial round trip timeout value in milliseconds that will be used
1789 in calculating round trip times. This is the initial time interval
1790 for retransmissions.
1791
1792 Default: 3000
1793
1794rto_max - INTEGER
1795 The maximum value (in milliseconds) of the round trip timeout. This
1796 is the largest time interval that can elapse between retransmissions.
1797
1798 Default: 60000
1799
1800rto_min - INTEGER
1801 The minimum value (in milliseconds) of the round trip timeout. This
1802 is the smallest time interval the can elapse between retransmissions.
1803
1804 Default: 1000
1805
1806hb_interval - INTEGER
1807 The interval (in milliseconds) between HEARTBEAT chunks. These chunks
1808 are sent at the specified interval on idle paths to probe the state of
1809 a given path between 2 associations.
1810
1811 Default: 30000
1812
1813sack_timeout - INTEGER
1814 The amount of time (in milliseconds) that the implementation will wait
1815 to send a SACK.
1816
1817 Default: 200
1818
1819valid_cookie_life - INTEGER
1820 The default lifetime of the SCTP cookie (in milliseconds). The cookie
1821 is used during association establishment.
1822
1823 Default: 60000
1824
1825cookie_preserve_enable - BOOLEAN
1826 Enable or disable the ability to extend the lifetime of the SCTP cookie
1827 that is used during the establishment phase of SCTP association
1828
1829 1: Enable cookie lifetime extension.
1830 0: Disable
1831
1832 Default: 1
1833
Neil Horman3c681982012-10-24 09:20:03 +00001834cookie_hmac_alg - STRING
1835 Select the hmac algorithm used when generating the cookie value sent by
1836 a listening sctp socket to a connecting client in the INIT-ACK chunk.
1837 Valid values are:
1838 * md5
1839 * sha1
1840 * none
1841 Ability to assign md5 or sha1 as the selected alg is predicated on the
stephen hemminger3b09adc2013-01-03 07:50:29 +00001842 configuration of those algorithms at build time (CONFIG_CRYPTO_MD5 and
Neil Horman3c681982012-10-24 09:20:03 +00001843 CONFIG_CRYPTO_SHA1).
1844
1845 Default: Dependent on configuration. MD5 if available, else SHA1 if
1846 available, else none.
1847
Vlad Yasevich32e8d492008-07-08 16:43:29 -07001848rcvbuf_policy - INTEGER
1849 Determines if the receive buffer is attributed to the socket or to
1850 association. SCTP supports the capability to create multiple
1851 associations on a single socket. When using this capability, it is
1852 possible that a single stalled association that's buffering a lot
1853 of data may block other associations from delivering their data by
1854 consuming all of the receive buffer space. To work around this,
1855 the rcvbuf_policy could be set to attribute the receiver buffer space
1856 to each association instead of the socket. This prevents the described
1857 blocking.
1858
1859 1: rcvbuf space is per association
stephen hemminger3b09adc2013-01-03 07:50:29 +00001860 0: rcvbuf space is per socket
Vlad Yasevich32e8d492008-07-08 16:43:29 -07001861
1862 Default: 0
1863
1864sndbuf_policy - INTEGER
1865 Similar to rcvbuf_policy above, this applies to send buffer space.
1866
1867 1: Send buffer is tracked per association
1868 0: Send buffer is tracked per socket.
1869
1870 Default: 0
1871
1872sctp_mem - vector of 3 INTEGERs: min, pressure, max
1873 Number of pages allowed for queueing by all SCTP sockets.
1874
1875 min: Below this number of pages SCTP is not bothered about its
1876 memory appetite. When amount of memory allocated by SCTP exceeds
1877 this number, SCTP starts to moderate memory usage.
1878
1879 pressure: This value was introduced to follow format of tcp_mem.
1880
1881 max: Number of pages allowed for queueing by all SCTP sockets.
1882
1883 Default is calculated at boot time from amount of available memory.
1884
1885sctp_rmem - vector of 3 INTEGERs: min, default, max
Max Matveeva6e12042011-06-19 22:08:10 +00001886 Only the first value ("min") is used, "default" and "max" are
1887 ignored.
1888
1889 min: Minimal size of receive buffer used by SCTP socket.
1890 It is guaranteed to each SCTP socket (but not association) even
1891 under moderate memory pressure.
1892
1893 Default: 1 page
Vlad Yasevich32e8d492008-07-08 16:43:29 -07001894
1895sctp_wmem - vector of 3 INTEGERs: min, default, max
Max Matveeva6e12042011-06-19 22:08:10 +00001896 Currently this tunable has no effect.
Vlad Yasevich32e8d492008-07-08 16:43:29 -07001897
Bhaskar Dutta72388432009-09-03 17:25:47 +05301898addr_scope_policy - INTEGER
1899 Control IPv4 address scoping - draft-stewart-tsvwg-sctp-ipv4-00
1900
1901 0 - Disable IPv4 address scoping
1902 1 - Enable IPv4 address scoping
1903 2 - Follow draft but allow IPv4 private addresses
1904 3 - Follow draft but allow IPv4 link local addresses
1905
1906 Default: 1
1907
Linus Torvalds1da177e2005-04-16 15:20:36 -07001908
Stephen Hemminger4edc2f32008-07-10 16:50:26 -07001909/proc/sys/net/core/*
Shan Weic60f6aa2012-04-26 16:52:52 +00001910 Please see: Documentation/sysctl/net.txt for descriptions of these entries.
Wang Tinggong705efc32009-05-14 22:49:36 +00001911
Linus Torvalds1da177e2005-04-16 15:20:36 -07001912
Stephen Hemminger4edc2f32008-07-10 16:50:26 -07001913/proc/sys/net/unix/*
Wang Tinggong705efc32009-05-14 22:49:36 +00001914max_dgram_qlen - INTEGER
1915 The maximum length of dgram socket receive queue
1916
1917 Default: 10
1918
1919
1920UNDOCUMENTED:
Stephen Hemminger4edc2f32008-07-10 16:50:26 -07001921
1922/proc/sys/net/irda/*
1923 fast_poll_increase FIXME
1924 warn_noreply_time FIXME
1925 discovery_slots FIXME
1926 slot_timeout FIXME
1927 max_baud_rate FIXME
1928 discovery_timeout FIXME
1929 lap_keepalive_time FIXME
1930 max_noreply_time FIXME
1931 max_tx_data_size FIXME
1932 max_tx_window FIXME
1933 min_tx_turn_time FIXME