| blob | bfe924217f246a8c0a10846e3a034201a9b9095a |
1 /proc/sys/net/ipv4/* Variables:
3 ip_forward - BOOLEAN
4 0 - disabled (default)
5 not 0 - enabled
7 Forward Packets between interfaces.
9 This variable is special, its change resets all configuration
10 parameters to their default state (RFC1122 for hosts, RFC1812
11 for routers)
13 ip_default_ttl - INTEGER
14 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)
18 ip_no_pmtu_disc - BOOLEAN
19 Disable Path MTU Discovery.
20 default FALSE
22 min_pmtu - INTEGER
23 default 562 - minimum discovered Path MTU
25 route/max_size - INTEGER
26 Maximum number of routes allowed in the kernel. Increase
27 this when using large numbers of interfaces and/or routes.
29 neigh/default/gc_thresh3 - INTEGER
30 Maximum number of neighbor entries allowed. Increase this
31 when using large numbers of interfaces and when communicating
32 with large numbers of directly-connected peers.
34 mtu_expires - INTEGER
35 Time, in seconds, that cached PMTU information is kept.
37 min_adv_mss - INTEGER
38 The advertised MSS depends on the first hop route MTU, but will
39 never be lower than this setting.
41 rt_cache_rebuild_count - INTEGER
42 The per net-namespace route cache emergency rebuild threshold.
43 Any net-namespace having its route cache rebuilt due to
44 a hash bucket chain being too long more than this many times
45 will have its route caching disabled
47 IP Fragmentation:
49 ipfrag_high_thresh - INTEGER
50 Maximum memory used to reassemble IP fragments. When
51 ipfrag_high_thresh bytes of memory is allocated for this purpose,
52 the fragment handler will toss packets until ipfrag_low_thresh
53 is reached.
55 ipfrag_low_thresh - INTEGER
56 See ipfrag_high_thresh
58 ipfrag_time - INTEGER
59 Time in seconds to keep an IP fragment in memory.
61 ipfrag_secret_interval - INTEGER
62 Regeneration interval (in seconds) of the hash secret (or lifetime
63 for the hash secret) for IP fragments.
64 Default: 600
66 ipfrag_max_dist - INTEGER
67 ipfrag_max_dist is a non-negative integer value which defines the
68 maximum "disorder" which is allowed among fragments which share a
69 common IP source address. Note that reordering of packets is
70 not unusual, but if a large number of fragments arrive from a source
71 IP address while a particular fragment queue remains incomplete, it
72 probably indicates that one or more fragments belonging to that queue
73 have been lost. When ipfrag_max_dist is positive, an additional check
74 is done on fragments before they are added to a reassembly queue - if
75 ipfrag_max_dist (or more) fragments have arrived from a particular IP
76 address between additions to any IP fragment queue using that source
77 address, it's presumed that one or more fragments in the queue are
78 lost. The existing fragment queue will be dropped, and a new one
79 started. An ipfrag_max_dist value of zero disables this check.
81 Using a very small value, e.g. 1 or 2, for ipfrag_max_dist can
82 result in unnecessarily dropping fragment queues when normal
83 reordering of packets occurs, which could lead to poor application
84 performance. Using a very large value, e.g. 50000, increases the
85 likelihood of incorrectly reassembling IP fragments that originate
86 from different IP datagrams, which could result in data corruption.
87 Default: 64
89 INET peer storage:
91 inet_peer_threshold - INTEGER
92 The approximate size of the storage. Starting from this threshold
93 entries will be thrown aggressively. This threshold also determines
94 entries' time-to-live and time intervals between garbage collection
95 passes. More entries, less time-to-live, less GC interval.
97 inet_peer_minttl - INTEGER
98 Minimum time-to-live of entries. Should be enough to cover fragment
99 time-to-live on the reassembling side. This minimum time-to-live is
100 guaranteed if the pool size is less than inet_peer_threshold.
101 Measured in seconds.
103 inet_peer_maxttl - INTEGER
104 Maximum time-to-live of entries. Unused entries will expire after
105 this period of time if there is no memory pressure on the pool (i.e.
106 when the number of entries in the pool is very small).
107 Measured in seconds.
109 inet_peer_gc_mintime - INTEGER
110 Minimum interval between garbage collection passes. This interval is
111 in effect under high memory pressure on the pool.
112 Measured in seconds.
114 inet_peer_gc_maxtime - INTEGER
115 Minimum interval between garbage collection passes. This interval is
116 in effect under low (or absent) memory pressure on the pool.
117 Measured in seconds.
119 TCP variables:
121 somaxconn - INTEGER
122 Limit of socket listen() backlog, known in userspace as SOMAXCONN.
123 Defaults to 128. See also tcp_max_syn_backlog for additional tuning
124 for TCP sockets.
126 tcp_abc - INTEGER
127 Controls Appropriate Byte Count (ABC) defined in RFC3465.
128 ABC is a way of increasing congestion window (cwnd) more slowly
129 in response to partial acknowledgments.
130 Possible values are:
131 0 increase cwnd once per acknowledgment (no ABC)
132 1 increase cwnd once per acknowledgment of full sized segment
133 2 allow increase cwnd by two if acknowledgment is
134 of two segments to compensate for delayed acknowledgments.
135 Default: 0 (off)
137 tcp_abort_on_overflow - BOOLEAN
138 If listening service is too slow to accept new connections,
139 reset them. Default state is FALSE. It means that if overflow
140 occurred due to a burst, connection will recover. Enable this
141 option _only_ if you are really sure that listening daemon
142 cannot be tuned to accept connections faster. Enabling this
143 option can harm clients of your server.
145 tcp_adv_win_scale - INTEGER
146 Count buffering overhead as bytes/2^tcp_adv_win_scale
147 (if tcp_adv_win_scale > 0) or bytes-bytes/2^(-tcp_adv_win_scale),
148 if it is <= 0.
149 Possible values are [-31, 31], inclusive.
150 Default: 2
152 tcp_allowed_congestion_control - STRING
153 Show/set the congestion control choices available to non-privileged
154 processes. The list is a subset of those listed in
155 tcp_available_congestion_control.
156 Default is "reno" and the default setting (tcp_congestion_control).
158 tcp_app_win - INTEGER
159 Reserve max(window/2^tcp_app_win, mss) of window for application
160 buffer. Value 0 is special, it means that nothing is reserved.
161 Default: 31
163 tcp_available_congestion_control - STRING
164 Shows the available congestion control choices that are registered.
165 More congestion control algorithms may be available as modules,
166 but not loaded.
168 tcp_base_mss - INTEGER
169 The initial value of search_low to be used by the packetization layer
170 Path MTU discovery (MTU probing). If MTU probing is enabled,
171 this is the initial MSS used by the connection.
173 tcp_congestion_control - STRING
174 Set the congestion control algorithm to be used for new
175 connections. The algorithm "reno" is always available, but
176 additional choices may be available based on kernel configuration.
177 Default is set as part of kernel configuration.
179 tcp_cookie_size - INTEGER
180 Default size of TCP Cookie Transactions (TCPCT) option, that may be
181 overridden on a per socket basis by the TCPCT socket option.
182 Values greater than the maximum (16) are interpreted as the maximum.
183 Values greater than zero and less than the minimum (8) are interpreted
184 as the minimum. Odd values are interpreted as the next even value.
185 Default: 0 (off).
187 tcp_dsack - BOOLEAN
188 Allows TCP to send "duplicate" SACKs.
190 tcp_ecn - INTEGER
191 Enable Explicit Congestion Notification (ECN) in TCP. ECN is only
192 used when both ends of the TCP flow support it. It is useful to
193 avoid losses due to congestion (when the bottleneck router supports
194 ECN).
195 Possible values are:
196 0 disable ECN
197 1 ECN enabled
198 2 Only server-side ECN enabled. If the other end does
199 not support ECN, behavior is like with ECN disabled.
200 Default: 2
202 tcp_fack - BOOLEAN
203 Enable FACK congestion avoidance and fast retransmission.
204 The value is not used, if tcp_sack is not enabled.
206 tcp_fin_timeout - INTEGER
207 Time to hold socket in state FIN-WAIT-2, if it was closed
208 by our side. Peer can be broken and never close its side,
209 or even died unexpectedly. Default value is 60sec.
210 Usual value used in 2.2 was 180 seconds, you may restore
211 it, but remember that if your machine is even underloaded WEB server,
212 you risk to overflow memory with kilotons of dead sockets,
213 FIN-WAIT-2 sockets are less dangerous than FIN-WAIT-1,
214 because they eat maximum 1.5K of memory, but they tend
215 to live longer. Cf. tcp_max_orphans.
217 tcp_frto - INTEGER
218 Enables Forward RTO-Recovery (F-RTO) defined in RFC4138.
219 F-RTO is an enhanced recovery algorithm for TCP retransmission
220 timeouts. It is particularly beneficial in wireless environments
221 where packet loss is typically due to random radio interference
222 rather than intermediate router congestion. F-RTO is sender-side
223 only modification. Therefore it does not require any support from
224 the peer.
226 If set to 1, basic version is enabled. 2 enables SACK enhanced
227 F-RTO if flow uses SACK. The basic version can be used also when
228 SACK is in use though scenario(s) with it exists where F-RTO
229 interacts badly with the packet counting of the SACK enabled TCP
230 flow.
232 tcp_frto_response - INTEGER
233 When F-RTO has detected that a TCP retransmission timeout was
234 spurious (i.e, the timeout would have been avoided had TCP set a
235 longer retransmission timeout), TCP has several options what to do
236 next. Possible values are:
237 0 Rate halving based; a smooth and conservative response,
238 results in halved cwnd and ssthresh after one RTT
239 1 Very conservative response; not recommended because even
240 though being valid, it interacts poorly with the rest of
241 Linux TCP, halves cwnd and ssthresh immediately
242 2 Aggressive response; undoes congestion control measures
243 that are now known to be unnecessary (ignoring the
244 possibility of a lost retransmission that would require
245 TCP to be more cautious), cwnd and ssthresh are restored
246 to the values prior timeout
247 Default: 0 (rate halving based)
249 tcp_keepalive_time - INTEGER
250 How often TCP sends out keepalive messages when keepalive is enabled.
251 Default: 2hours.
253 tcp_keepalive_probes - INTEGER
254 How many keepalive probes TCP sends out, until it decides that the
255 connection is broken. Default value: 9.
257 tcp_keepalive_intvl - INTEGER
258 How frequently the probes are send out. Multiplied by
259 tcp_keepalive_probes it is time to kill not responding connection,
260 after probes started. Default value: 75sec i.e. connection
261 will be aborted after ~11 minutes of retries.
263 tcp_low_latency - BOOLEAN
264 If set, the TCP stack makes decisions that prefer lower
265 latency as opposed to higher throughput. By default, this
266 option is not set meaning that higher throughput is preferred.
267 An example of an application where this default should be
268 changed would be a Beowulf compute cluster.
269 Default: 0
271 tcp_max_orphans - INTEGER
272 Maximal number of TCP sockets not attached to any user file handle,
273 held by system. If this number is exceeded orphaned connections are
274 reset immediately and warning is printed. This limit exists
275 only to prevent simple DoS attacks, you _must_ not rely on this
276 or lower the limit artificially, but rather increase it
277 (probably, after increasing installed memory),
278 if network conditions require more than default value,
279 and tune network services to linger and kill such states
280 more aggressively. Let me to remind again: each orphan eats
281 up to ~64K of unswappable memory.
283 tcp_max_ssthresh - INTEGER
284 Limited Slow-Start for TCP with large congestion windows (cwnd) defined in
285 RFC3742. Limited slow-start is a mechanism to limit growth of the cwnd
286 on the region where cwnd is larger than tcp_max_ssthresh. TCP increases cwnd
287 by at most tcp_max_ssthresh segments, and by at least tcp_max_ssthresh/2
288 segments per RTT when the cwnd is above tcp_max_ssthresh.
289 If TCP connection increased cwnd to thousands (or tens of thousands) segments,
290 and thousands of packets were being dropped during slow-start, you can set
291 tcp_max_ssthresh to improve performance for new TCP connection.
292 Default: 0 (off)
294 tcp_max_syn_backlog - INTEGER
295 Maximal number of remembered connection requests, which are
296 still did not receive an acknowledgment from connecting client.
297 Default value is 1024 for systems with more than 128Mb of memory,
298 and 128 for low memory machines. If server suffers of overload,
299 try to increase this number.
301 tcp_max_tw_buckets - INTEGER
302 Maximal number of timewait sockets held by system simultaneously.
303 If this number is exceeded time-wait socket is immediately destroyed
304 and warning is printed. This limit exists only to prevent
305 simple DoS attacks, you _must_ not lower the limit artificially,
306 but rather increase it (probably, after increasing installed memory),
307 if network conditions require more than default value.
309 tcp_mem - vector of 3 INTEGERs: min, pressure, max
310 min: below this number of pages TCP is not bothered about its
311 memory appetite.
313 pressure: when amount of memory allocated by TCP exceeds this number
314 of pages, TCP moderates its memory consumption and enters memory
315 pressure mode, which is exited when memory consumption falls
316 under "min".
318 max: number of pages allowed for queueing by all TCP sockets.
320 Defaults are calculated at boot time from amount of available
321 memory.
323 tcp_moderate_rcvbuf - BOOLEAN
324 If set, TCP performs receive buffer auto-tuning, attempting to
325 automatically size the buffer (no greater than tcp_rmem[2]) to
326 match the size required by the path for full throughput. Enabled by
327 default.
329 tcp_mtu_probing - INTEGER
330 Controls TCP Packetization-Layer Path MTU Discovery. Takes three
331 values:
332 0 - Disabled
333 1 - Disabled by default, enabled when an ICMP black hole detected
334 2 - Always enabled, use initial MSS of tcp_base_mss.
336 tcp_no_metrics_save - BOOLEAN
337 By default, TCP saves various connection metrics in the route cache
338 when the connection closes, so that connections established in the
339 near future can use these to set initial conditions. Usually, this
340 increases overall performance, but may sometimes cause performance
341 degradation. If set, TCP will not cache metrics on closing
342 connections.
344 tcp_orphan_retries - INTEGER
345 This value influences the timeout of a locally closed TCP connection,
346 when RTO retransmissions remain unacknowledged.
347 See tcp_retries2 for more details.
349 The default value is 8.
350 If your machine is a loaded WEB server,
351 you should think about lowering this value, such sockets
352 may consume significant resources. Cf. tcp_max_orphans.
354 tcp_reordering - INTEGER
355 Maximal reordering of packets in a TCP stream.
356 Default: 3
358 tcp_retrans_collapse - BOOLEAN
359 Bug-to-bug compatibility with some broken printers.
360 On retransmit try to send bigger packets to work around bugs in
361 certain TCP stacks.
363 tcp_retries1 - INTEGER
364 This value influences the time, after which TCP decides, that
365 something is wrong due to unacknowledged RTO retransmissions,
366 and reports this suspicion to the network layer.
367 See tcp_retries2 for more details.
369 RFC 1122 recommends at least 3 retransmissions, which is the
370 default.
372 tcp_retries2 - INTEGER
373 This value influences the timeout of an alive TCP connection,
374 when RTO retransmissions remain unacknowledged.
375 Given a value of N, a hypothetical TCP connection following
376 exponential backoff with an initial RTO of TCP_RTO_MIN would
377 retransmit N times before killing the connection at the (N+1)th RTO.
379 The default value of 15 yields a hypothetical timeout of 924.6
380 seconds and is a lower bound for the effective timeout.
381 TCP will effectively time out at the first RTO which exceeds the
382 hypothetical timeout.
384 RFC 1122 recommends at least 100 seconds for the timeout,
385 which corresponds to a value of at least 8.
387 tcp_rfc1337 - BOOLEAN
388 If set, the TCP stack behaves conforming to RFC1337. If unset,
389 we are not conforming to RFC, but prevent TCP TIME_WAIT
390 assassination.
391 Default: 0
393 tcp_rmem - vector of 3 INTEGERs: min, default, max
394 min: Minimal size of receive buffer used by TCP sockets.
395 It is guaranteed to each TCP socket, even under moderate memory
396 pressure.
397 Default: 8K
399 default: initial size of receive buffer used by TCP sockets.
400 This value overrides net.core.rmem_default used by other protocols.
401 Default: 87380 bytes. This value results in window of 65535 with
402 default setting of tcp_adv_win_scale and tcp_app_win:0 and a bit
403 less for default tcp_app_win. See below about these variables.
405 max: maximal size of receive buffer allowed for automatically
406 selected receiver buffers for TCP socket. This value does not override
407 net.core.rmem_max. Calling setsockopt() with SO_RCVBUF disables
408 automatic tuning of that socket's receive buffer size, in which
409 case this value is ignored.
410 Default: between 87380B and 4MB, depending on RAM size.
412 tcp_sack - BOOLEAN
413 Enable select acknowledgments (SACKS).
415 tcp_slow_start_after_idle - BOOLEAN
416 If set, provide RFC2861 behavior and time out the congestion
417 window after an idle period. An idle period is defined at
418 the current RTO. If unset, the congestion window will not
419 be timed out after an idle period.
420 Default: 1
422 tcp_stdurg - BOOLEAN
423 Use the Host requirements interpretation of the TCP urgent pointer field.
424 Most hosts use the older BSD interpretation, so if you turn this on
425 Linux might not communicate correctly with them.
426 Default: FALSE
428 tcp_synack_retries - INTEGER
429 Number of times SYNACKs for a passive TCP connection attempt will
430 be retransmitted. Should not be higher than 255. Default value
431 is 5, which corresponds to ~180seconds.
433 tcp_syncookies - BOOLEAN
434 Only valid when the kernel was compiled with CONFIG_SYNCOOKIES
435 Send out syncookies when the syn backlog queue of a socket
436 overflows. This is to prevent against the common 'SYN flood attack'
437 Default: FALSE
439 Note, that syncookies is fallback facility.
440 It MUST NOT be used to help highly loaded servers to stand
441 against legal connection rate. If you see SYN flood warnings
442 in your logs, but investigation shows that they occur
443 because of overload with legal connections, you should tune
444 another parameters until this warning disappear.
445 See: tcp_max_syn_backlog, tcp_synack_retries, tcp_abort_on_overflow.
447 syncookies seriously violate TCP protocol, do not allow
448 to use TCP extensions, can result in serious degradation
449 of some services (f.e. SMTP relaying), visible not by you,
450 but your clients and relays, contacting you. While you see
451 SYN flood warnings in logs not being really flooded, your server
452 is seriously misconfigured.
454 tcp_syn_retries - INTEGER
455 Number of times initial SYNs for an active TCP connection attempt
456 will be retransmitted. Should not be higher than 255. Default value
457 is 5, which corresponds to ~180seconds.
459 tcp_timestamps - BOOLEAN
460 Enable timestamps as defined in RFC1323.
462 tcp_tso_win_divisor - INTEGER
463 This allows control over what percentage of the congestion window
464 can be consumed by a single TSO frame.
465 The setting of this parameter is a choice between burstiness and
466 building larger TSO frames.
467 Default: 3
469 tcp_tw_recycle - BOOLEAN
470 Enable fast recycling TIME-WAIT sockets. Default value is 0.
471 It should not be changed without advice/request of technical
472 experts.
474 tcp_tw_reuse - BOOLEAN
475 Allow to reuse TIME-WAIT sockets for new connections when it is
476 safe from protocol viewpoint. Default value is 0.
477 It should not be changed without advice/request of technical
478 experts.
480 tcp_window_scaling - BOOLEAN
481 Enable window scaling as defined in RFC1323.
483 tcp_wmem - vector of 3 INTEGERs: min, default, max
484 min: Amount of memory reserved for send buffers for TCP sockets.
485 Each TCP socket has rights to use it due to fact of its birth.
486 Default: 4K
488 default: initial size of send buffer used by TCP sockets. This
489 value overrides net.core.wmem_default used by other protocols.
490 It is usually lower than net.core.wmem_default.
491 Default: 16K
493 max: Maximal amount of memory allowed for automatically tuned
494 send buffers for TCP sockets. This value does not override
495 net.core.wmem_max. Calling setsockopt() with SO_SNDBUF disables
496 automatic tuning of that socket's send buffer size, in which case
497 this value is ignored.
498 Default: between 64K and 4MB, depending on RAM size.
500 tcp_workaround_signed_windows - BOOLEAN
501 If set, assume no receipt of a window scaling option means the
502 remote TCP is broken and treats the window as a signed quantity.
503 If unset, assume the remote TCP is not broken even if we do
504 not receive a window scaling option from them.
505 Default: 0
507 tcp_dma_copybreak - INTEGER
508 Lower limit, in bytes, of the size of socket reads that will be
509 offloaded to a DMA copy engine, if one is present in the system
510 and CONFIG_NET_DMA is enabled.
511 Default: 4096
513 tcp_thin_linear_timeouts - BOOLEAN
514 Enable dynamic triggering of linear timeouts for thin streams.
515 If set, a check is performed upon retransmission by timeout to
516 determine if the stream is thin (less than 4 packets in flight).
517 As long as the stream is found to be thin, up to 6 linear
518 timeouts may be performed before exponential backoff mode is
519 initiated. This improves retransmission latency for
520 non-aggressive thin streams, often found to be time-dependent.
521 For more information on thin streams, see
522 Documentation/networking/tcp-thin.txt
523 Default: 0
525 tcp_thin_dupack - BOOLEAN
526 Enable dynamic triggering of retransmissions after one dupACK
527 for thin streams. If set, a check is performed upon reception
528 of a dupACK to determine if the stream is thin (less than 4
529 packets in flight). As long as the stream is found to be thin,
530 data is retransmitted on the first received dupACK. This
531 improves retransmission latency for non-aggressive thin
532 streams, often found to be time-dependent.
533 For more information on thin streams, see
534 Documentation/networking/tcp-thin.txt
535 Default: 0
537 UDP variables:
539 udp_mem - vector of 3 INTEGERs: min, pressure, max
540 Number of pages allowed for queueing by all UDP sockets.
542 min: Below this number of pages UDP is not bothered about its
543 memory appetite. When amount of memory allocated by UDP exceeds
544 this number, UDP starts to moderate memory usage.
546 pressure: This value was introduced to follow format of tcp_mem.
548 max: Number of pages allowed for queueing by all UDP sockets.
550 Default is calculated at boot time from amount of available memory.
552 udp_rmem_min - INTEGER
553 Minimal size of receive buffer used by UDP sockets in moderation.
554 Each UDP socket is able to use the size for receiving data, even if
555 total pages of UDP sockets exceed udp_mem pressure. The unit is byte.
556 Default: 4096
558 udp_wmem_min - INTEGER
559 Minimal size of send buffer used by UDP sockets in moderation.
560 Each UDP socket is able to use the size for sending data, even if
561 total pages of UDP sockets exceed udp_mem pressure. The unit is byte.
562 Default: 4096
564 CIPSOv4 Variables:
566 cipso_cache_enable - BOOLEAN
567 If set, enable additions to and lookups from the CIPSO label mapping
568 cache. If unset, additions are ignored and lookups always result in a
569 miss. However, regardless of the setting the cache is still
570 invalidated when required when means you can safely toggle this on and
571 off and the cache will always be "safe".
572 Default: 1
574 cipso_cache_bucket_size - INTEGER
575 The CIPSO label cache consists of a fixed size hash table with each
576 hash bucket containing a number of cache entries. This variable limits
577 the number of entries in each hash bucket; the larger the value the
578 more CIPSO label mappings that can be cached. When the number of
579 entries in a given hash bucket reaches this limit adding new entries
580 causes the oldest entry in the bucket to be removed to make room.
581 Default: 10
583 cipso_rbm_optfmt - BOOLEAN
584 Enable the "Optimized Tag 1 Format" as defined in section 3.4.2.6 of
585 the CIPSO draft specification (see Documentation/netlabel for details).
586 This means that when set the CIPSO tag will be padded with empty
587 categories in order to make the packet data 32-bit aligned.
588 Default: 0
590 cipso_rbm_structvalid - BOOLEAN
591 If set, do a very strict check of the CIPSO option when
592 ip_options_compile() is called. If unset, relax the checks done during
593 ip_options_compile(). Either way is "safe" as errors are caught else
594 where in the CIPSO processing code but setting this to 0 (False) should
595 result in less work (i.e. it should be faster) but could cause problems
596 with other implementations that require strict checking.
597 Default: 0
599 IP Variables:
601 ip_local_port_range - 2 INTEGERS
602 Defines the local port range that is used by TCP and UDP to
603 choose the local port. The first number is the first, the
604 second the last local port number. Default value depends on
605 amount of memory available on the system:
606 > 128Mb 32768-61000
607 < 128Mb 1024-4999 or even less.
608 This number defines number of active connections, which this
609 system can issue simultaneously to systems not supporting
610 TCP extensions (timestamps). With tcp_tw_recycle enabled
611 (i.e. by default) range 1024-4999 is enough to issue up to
612 2000 connections per second to systems supporting timestamps.
614 ip_local_reserved_ports - list of comma separated ranges
615 Specify the ports which are reserved for known third-party
616 applications. These ports will not be used by automatic port
617 assignments (e.g. when calling connect() or bind() with port
618 number 0). Explicit port allocation behavior is unchanged.
620 The format used for both input and output is a comma separated
621 list of ranges (e.g. "1,2-4,10-10" for ports 1, 2, 3, 4 and
622 10). Writing to the file will clear all previously reserved
623 ports and update the current list with the one given in the
624 input.
626 Note that ip_local_port_range and ip_local_reserved_ports
627 settings are independent and both are considered by the kernel
628 when determining which ports are available for automatic port
629 assignments.
631 You can reserve ports which are not in the current
632 ip_local_port_range, e.g.:
634 $ cat /proc/sys/net/ipv4/ip_local_port_range
635 32000 61000
636 $ cat /proc/sys/net/ipv4/ip_local_reserved_ports
637 8080,9148
639 although this is redundant. However such a setting is useful
640 if later the port range is changed to a value that will
641 include the reserved ports.
643 Default: Empty
645 ip_nonlocal_bind - BOOLEAN
646 If set, allows processes to bind() to non-local IP addresses,
647 which can be quite useful - but may break some applications.
648 Default: 0
650 ip_dynaddr - BOOLEAN
651 If set non-zero, enables support for dynamic addresses.
652 If set to a non-zero value larger than 1, a kernel log
653 message will be printed when dynamic address rewriting
654 occurs.
655 Default: 0
657 icmp_echo_ignore_all - BOOLEAN
658 If set non-zero, then the kernel will ignore all ICMP ECHO
659 requests sent to it.
660 Default: 0
662 icmp_echo_ignore_broadcasts - BOOLEAN
663 If set non-zero, then the kernel will ignore all ICMP ECHO and
664 TIMESTAMP requests sent to it via broadcast/multicast.
665 Default: 1
667 icmp_ratelimit - INTEGER
668 Limit the maximal rates for sending ICMP packets whose type matches
669 icmp_ratemask (see below) to specific targets.
670 0 to disable any limiting,
671 otherwise the minimal space between responses in milliseconds.
672 Default: 1000
674 icmp_ratemask - INTEGER
675 Mask made of ICMP types for which rates are being limited.
676 Significant bits: IHGFEDCBA9876543210
677 Default mask: 0000001100000011000 (6168)
679 Bit definitions (see include/linux/icmp.h):
680 0 Echo Reply
681 3 Destination Unreachable *
682 4 Source Quench *
683 5 Redirect
684 8 Echo Request
685 B Time Exceeded *
686 C Parameter Problem *
687 D Timestamp Request
688 E Timestamp Reply
689 F Info Request
690 G Info Reply
691 H Address Mask Request
692 I Address Mask Reply
694 * These are rate limited by default (see default mask above)
696 icmp_ignore_bogus_error_responses - BOOLEAN
697 Some routers violate RFC1122 by sending bogus responses to broadcast
698 frames. Such violations are normally logged via a kernel warning.
699 If this is set to TRUE, the kernel will not give such warnings, which
700 will avoid log file clutter.
701 Default: FALSE
703 icmp_errors_use_inbound_ifaddr - BOOLEAN
705 If zero, icmp error messages are sent with the primary address of
706 the exiting interface.
708 If non-zero, the message will be sent with the primary address of
709 the interface that received the packet that caused the icmp error.
710 This is the behaviour network many administrators will expect from
711 a router. And it can make debugging complicated network layouts
712 much easier.
714 Note that if no primary address exists for the interface selected,
715 then the primary address of the first non-loopback interface that
716 has one will be used regardless of this setting.
718 Default: 0
720 igmp_max_memberships - INTEGER
721 Change the maximum number of multicast groups we can subscribe to.
722 Default: 20
724 Theoretical maximum value is bounded by having to send a membership
725 report in a single datagram (i.e. the report can't span multiple
726 datagrams, or risk confusing the switch and leaving groups you don't
727 intend to).
729 The number of supported groups 'M' is bounded by the number of group
730 report entries you can fit into a single datagram of 65535 bytes.
732 M = 65536-sizeof (ip header)/(sizeof(Group record))
734 Group records are variable length, with a minimum of 12 bytes.
735 So net.ipv4.igmp_max_memberships should not be set higher than:
737 (65536-24) / 12 = 5459
739 The value 5459 assumes no IP header options, so in practice
740 this number may be lower.
742 conf/interface/* changes special settings per interface (where
743 "interface" is the name of your network interface)
745 conf/all/* is special, changes the settings for all interfaces
747 log_martians - BOOLEAN
748 Log packets with impossible addresses to kernel log.
749 log_martians for the interface will be enabled if at least one of
750 conf/{all,interface}/log_martians is set to TRUE,
751 it will be disabled otherwise
753 accept_redirects - BOOLEAN
754 Accept ICMP redirect messages.
755 accept_redirects for the interface will be enabled if:
756 - both conf/{all,interface}/accept_redirects are TRUE in the case
757 forwarding for the interface is enabled
758 or
759 - at least one of conf/{all,interface}/accept_redirects is TRUE in the
760 case forwarding for the interface is disabled
761 accept_redirects for the interface will be disabled otherwise
762 default TRUE (host)
763 FALSE (router)
765 forwarding - BOOLEAN
766 Enable IP forwarding on this interface.
768 mc_forwarding - BOOLEAN
769 Do multicast routing. The kernel needs to be compiled with CONFIG_MROUTE
770 and a multicast routing daemon is required.
771 conf/all/mc_forwarding must also be set to TRUE to enable multicast
772 routing for the interface
774 medium_id - INTEGER
775 Integer value used to differentiate the devices by the medium they
776 are attached to. Two devices can have different id values when
777 the broadcast packets are received only on one of them.
778 The default value 0 means that the device is the only interface
779 to its medium, value of -1 means that medium is not known.
781 Currently, it is used to change the proxy_arp behavior:
782 the proxy_arp feature is enabled for packets forwarded between
783 two devices attached to different media.
785 proxy_arp - BOOLEAN
786 Do proxy arp.
787 proxy_arp for the interface will be enabled if at least one of
788 conf/{all,interface}/proxy_arp is set to TRUE,
789 it will be disabled otherwise
791 proxy_arp_pvlan - BOOLEAN
792 Private VLAN proxy arp.
793 Basically allow proxy arp replies back to the same interface
794 (from which the ARP request/solicitation was received).
796 This is done to support (ethernet) switch features, like RFC
797 3069, where the individual ports are NOT allowed to
798 communicate with each other, but they are allowed to talk to
799 the upstream router. As described in RFC 3069, it is possible
800 to allow these hosts to communicate through the upstream
801 router by proxy_arp'ing. Don't need to be used together with
802 proxy_arp.
804 This technology is known by different names:
805 In RFC 3069 it is called VLAN Aggregation.
806 Cisco and Allied Telesyn call it Private VLAN.
807 Hewlett-Packard call it Source-Port filtering or port-isolation.
808 Ericsson call it MAC-Forced Forwarding (RFC Draft).
810 shared_media - BOOLEAN
811 Send(router) or accept(host) RFC1620 shared media redirects.
812 Overrides ip_secure_redirects.
813 shared_media for the interface will be enabled if at least one of
814 conf/{all,interface}/shared_media is set to TRUE,
815 it will be disabled otherwise
816 default TRUE
818 secure_redirects - BOOLEAN
819 Accept ICMP redirect messages only for gateways,
820 listed in default gateway list.
821 secure_redirects for the interface will be enabled if at least one of
822 conf/{all,interface}/secure_redirects is set to TRUE,
823 it will be disabled otherwise
824 default TRUE
826 send_redirects - BOOLEAN
827 Send redirects, if router.
828 send_redirects for the interface will be enabled if at least one of
829 conf/{all,interface}/send_redirects is set to TRUE,
830 it will be disabled otherwise
831 Default: TRUE
833 bootp_relay - BOOLEAN
834 Accept packets with source address 0.b.c.d destined
835 not to this host as local ones. It is supposed, that
836 BOOTP relay daemon will catch and forward such packets.
837 conf/all/bootp_relay must also be set to TRUE to enable BOOTP relay
838 for the interface
839 default FALSE
840 Not Implemented Yet.
842 accept_source_route - BOOLEAN
843 Accept packets with SRR option.
844 conf/all/accept_source_route must also be set to TRUE to accept packets
845 with SRR option on the interface
846 default TRUE (router)
847 FALSE (host)
849 accept_local - BOOLEAN
850 Accept packets with local source addresses. In combination with
851 suitable routing, this can be used to direct packets between two
852 local interfaces over the wire and have them accepted properly.
853 default FALSE
855 rp_filter - INTEGER
856 0 - No source validation.
857 1 - Strict mode as defined in RFC3704 Strict Reverse Path
858 Each incoming packet is tested against the FIB and if the interface
859 is not the best reverse path the packet check will fail.
860 By default failed packets are discarded.
861 2 - Loose mode as defined in RFC3704 Loose Reverse Path
862 Each incoming packet's source address is also tested against the FIB
863 and if the source address is not reachable via any interface
864 the packet check will fail.
866 Current recommended practice in RFC3704 is to enable strict mode
867 to prevent IP spoofing from DDos attacks. If using asymmetric routing
868 or other complicated routing, then loose mode is recommended.
870 The max value from conf/{all,interface}/rp_filter is used
871 when doing source validation on the {interface}.
873 Default value is 0. Note that some distributions enable it
874 in startup scripts.
876 arp_filter - BOOLEAN
877 1 - Allows you to have multiple network interfaces on the same
878 subnet, and have the ARPs for each interface be answered
879 based on whether or not the kernel would route a packet from
880 the ARP'd IP out that interface (therefore you must use source
881 based routing for this to work). In other words it allows control
882 of which cards (usually 1) will respond to an arp request.
884 0 - (default) The kernel can respond to arp requests with addresses
885 from other interfaces. This may seem wrong but it usually makes
886 sense, because it increases the chance of successful communication.
887 IP addresses are owned by the complete host on Linux, not by
888 particular interfaces. Only for more complex setups like load-
889 balancing, does this behaviour cause problems.
891 arp_filter for the interface will be enabled if at least one of
892 conf/{all,interface}/arp_filter is set to TRUE,
893 it will be disabled otherwise
895 arp_announce - INTEGER
896 Define different restriction levels for announcing the local
897 source IP address from IP packets in ARP requests sent on
898 interface:
899 0 - (default) Use any local address, configured on any interface
900 1 - Try to avoid local addresses that are not in the target's
901 subnet for this interface. This mode is useful when target
902 hosts reachable via this interface require the source IP
903 address in ARP requests to be part of their logical network
904 configured on the receiving interface. When we generate the
905 request we will check all our subnets that include the
906 target IP and will preserve the source address if it is from
907 such subnet. If there is no such subnet we select source
908 address according to the rules for level 2.
909 2 - Always use the best local address for this target.
910 In this mode we ignore the source address in the IP packet
911 and try to select local address that we prefer for talks with
912 the target host. Such local address is selected by looking
913 for primary IP addresses on all our subnets on the outgoing
914 interface that include the target IP address. If no suitable
915 local address is found we select the first local address
916 we have on the outgoing interface or on all other interfaces,
917 with the hope we will receive reply for our request and
918 even sometimes no matter the source IP address we announce.
920 The max value from conf/{all,interface}/arp_announce is used.
922 Increasing the restriction level gives more chance for
923 receiving answer from the resolved target while decreasing
924 the level announces more valid sender's information.
926 arp_ignore - INTEGER
927 Define different modes for sending replies in response to
928 received ARP requests that resolve local target IP addresses:
929 0 - (default): reply for any local target IP address, configured
930 on any interface
931 1 - reply only if the target IP address is local address
932 configured on the incoming interface
933 2 - reply only if the target IP address is local address
934 configured on the incoming interface and both with the
935 sender's IP address are part from same subnet on this interface
936 3 - do not reply for local addresses configured with scope host,
937 only resolutions for global and link addresses are replied
938 4-7 - reserved
939 8 - do not reply for all local addresses
941 The max value from conf/{all,interface}/arp_ignore is used
942 when ARP request is received on the {interface}
944 arp_notify - BOOLEAN
945 Define mode for notification of address and device changes.
946 0 - (default): do nothing
947 1 - Generate gratuitous arp requests when device is brought up
948 or hardware address changes.
950 arp_accept - BOOLEAN
951 Define behavior for gratuitous ARP frames who's IP is not
952 already present in the ARP table:
953 0 - don't create new entries in the ARP table
954 1 - create new entries in the ARP table
956 Both replies and requests type gratuitous arp will trigger the
957 ARP table to be updated, if this setting is on.
959 If the ARP table already contains the IP address of the
960 gratuitous arp frame, the arp table will be updated regardless
961 if this setting is on or off.
964 app_solicit - INTEGER
965 The maximum number of probes to send to the user space ARP daemon
966 via netlink before dropping back to multicast probes (see
967 mcast_solicit). Defaults to 0.
969 disable_policy - BOOLEAN
970 Disable IPSEC policy (SPD) for this interface
972 disable_xfrm - BOOLEAN
973 Disable IPSEC encryption on this interface, whatever the policy
977 tag - INTEGER
978 Allows you to write a number, which can be used as required.
979 Default value is 0.
981 Alexey Kuznetsov.
982 kuznet@ms2.inr.ac.ru
984 Updated by:
985 Andi Kleen
986 ak@muc.de
987 Nicolas Delon
988 delon.nicolas@wanadoo.fr
993 /proc/sys/net/ipv6/* Variables:
995 IPv6 has no global variables such as tcp_*. tcp_* settings under ipv4/ also
996 apply to IPv6 [XXX?].
998 bindv6only - BOOLEAN
999 Default value for IPV6_V6ONLY socket option,
1000 which restricts use of the IPv6 socket to IPv6 communication
1001 only.
1002 TRUE: disable IPv4-mapped address feature
1003 FALSE: enable IPv4-mapped address feature
1005 Default: FALSE (as specified in RFC2553bis)
1007 IPv6 Fragmentation:
1009 ip6frag_high_thresh - INTEGER
1010 Maximum memory used to reassemble IPv6 fragments. When
1011 ip6frag_high_thresh bytes of memory is allocated for this purpose,
1012 the fragment handler will toss packets until ip6frag_low_thresh
1013 is reached.
1015 ip6frag_low_thresh - INTEGER
1016 See ip6frag_high_thresh
1018 ip6frag_time - INTEGER
1019 Time in seconds to keep an IPv6 fragment in memory.
1021 ip6frag_secret_interval - INTEGER
1022 Regeneration interval (in seconds) of the hash secret (or lifetime
1023 for the hash secret) for IPv6 fragments.
1024 Default: 600
1026 conf/default/*:
1027 Change the interface-specific default settings.
1030 conf/all/*:
1031 Change all the interface-specific settings.
1033 [XXX: Other special features than forwarding?]
1035 conf/all/forwarding - BOOLEAN
1036 Enable global IPv6 forwarding between all interfaces.
1038 IPv4 and IPv6 work differently here; e.g. netfilter must be used
1039 to control which interfaces may forward packets and which not.
1041 This also sets all interfaces' Host/Router setting
1042 'forwarding' to the specified value. See below for details.
1044 This referred to as global forwarding.
1046 proxy_ndp - BOOLEAN
1047 Do proxy ndp.
1049 conf/interface/*:
1050 Change special settings per interface.
1052 The functional behaviour for certain settings is different
1053 depending on whether local forwarding is enabled or not.
1055 accept_ra - BOOLEAN
1056 Accept Router Advertisements; autoconfigure using them.
1058 Possible values are:
1059 0 Do not accept Router Advertisements.
1060 1 Accept Router Advertisements if forwarding is disabled.
1061 2 Overrule forwarding behaviour. Accept Router Advertisements
1062 even if forwarding is enabled.
1064 Functional default: enabled if local forwarding is disabled.
1065 disabled if local forwarding is enabled.
1067 accept_ra_defrtr - BOOLEAN
1068 Learn default router in Router Advertisement.
1070 Functional default: enabled if accept_ra is enabled.
1071 disabled if accept_ra is disabled.
1073 accept_ra_pinfo - BOOLEAN
1074 Learn Prefix Information in Router Advertisement.
1076 Functional default: enabled if accept_ra is enabled.
1077 disabled if accept_ra is disabled.
1079 accept_ra_rt_info_max_plen - INTEGER
1080 Maximum prefix length of Route Information in RA.
1082 Route Information w/ prefix larger than or equal to this
1083 variable shall be ignored.
1085 Functional default: 0 if accept_ra_rtr_pref is enabled.
1086 -1 if accept_ra_rtr_pref is disabled.
1088 accept_ra_rtr_pref - BOOLEAN
1089 Accept Router Preference in RA.
1091 Functional default: enabled if accept_ra is enabled.
1092 disabled if accept_ra is disabled.
1094 accept_redirects - BOOLEAN
1095 Accept Redirects.
1097 Functional default: enabled if local forwarding is disabled.
1098 disabled if local forwarding is enabled.
1100 accept_source_route - INTEGER
1101 Accept source routing (routing extension header).
1103 >= 0: Accept only routing header type 2.
1104 < 0: Do not accept routing header.
1106 Default: 0
1108 autoconf - BOOLEAN
1109 Autoconfigure addresses using Prefix Information in Router
1110 Advertisements.
1112 Functional default: enabled if accept_ra_pinfo is enabled.
1113 disabled if accept_ra_pinfo is disabled.
1115 dad_transmits - INTEGER
1116 The amount of Duplicate Address Detection probes to send.
1117 Default: 1
1119 forwarding - BOOLEAN
1120 Configure interface-specific Host/Router behaviour.
1122 Note: It is recommended to have the same setting on all
1123 interfaces; mixed router/host scenarios are rather uncommon.
1125 Possible values are:
1126 0 Forwarding disabled
1127 1 Forwarding enabled
1128 2 Forwarding enabled (Hybrid Mode)
1130 FALSE (0):
1132 By default, Host behaviour is assumed. This means:
1134 1. IsRouter flag is not set in Neighbour Advertisements.
1135 2. Router Solicitations are being sent when necessary.
1136 3. If accept_ra is TRUE (default), accept Router
1137 Advertisements (and do autoconfiguration).
1138 4. If accept_redirects is TRUE (default), accept Redirects.
1140 TRUE (1):
1142 If local forwarding is enabled, Router behaviour is assumed.
1143 This means exactly the reverse from the above:
1145 1. IsRouter flag is set in Neighbour Advertisements.
1146 2. Router Solicitations are not sent.
1147 3. Router Advertisements are ignored unless accept_ra is 2.
1148 4. Redirects are ignored.
1150 TRUE (2):
1152 Hybrid mode. Same behaviour as TRUE, except for:
1154 2. Router Solicitations are being sent when necessary.
1156 Default: 0 (disabled) if global forwarding is disabled (default),
1157 otherwise 1 (enabled).
1159 hop_limit - INTEGER
1160 Default Hop Limit to set.
1161 Default: 64
1163 mtu - INTEGER
1164 Default Maximum Transfer Unit
1165 Default: 1280 (IPv6 required minimum)
1167 router_probe_interval - INTEGER
1168 Minimum interval (in seconds) between Router Probing described
1169 in RFC4191.
1171 Default: 60
1173 router_solicitation_delay - INTEGER
1174 Number of seconds to wait after interface is brought up
1175 before sending Router Solicitations.
1176 Default: 1
1178 router_solicitation_interval - INTEGER
1179 Number of seconds to wait between Router Solicitations.
1180 Default: 4
1182 router_solicitations - INTEGER
1183 Number of Router Solicitations to send until assuming no
1184 routers are present.
1185 Default: 3
1187 use_tempaddr - INTEGER
1188 Preference for Privacy Extensions (RFC3041).
1189 <= 0 : disable Privacy Extensions
1190 == 1 : enable Privacy Extensions, but prefer public
1191 addresses over temporary addresses.
1192 > 1 : enable Privacy Extensions and prefer temporary
1193 addresses over public addresses.
1194 Default: 0 (for most devices)
1195 -1 (for point-to-point devices and loopback devices)
1197 temp_valid_lft - INTEGER
1198 valid lifetime (in seconds) for temporary addresses.
1199 Default: 604800 (7 days)
1201 temp_prefered_lft - INTEGER
1202 Preferred lifetime (in seconds) for temporary addresses.
1203 Default: 86400 (1 day)
1205 max_desync_factor - INTEGER
1206 Maximum value for DESYNC_FACTOR, which is a random value
1207 that ensures that clients don't synchronize with each
1208 other and generate new addresses at exactly the same time.
1209 value is in seconds.
1210 Default: 600
1212 regen_max_retry - INTEGER
1213 Number of attempts before give up attempting to generate
1214 valid temporary addresses.
1215 Default: 5
1217 max_addresses - INTEGER
1218 Maximum number of autoconfigured addresses per interface. Setting
1219 to zero disables the limitation. It is not recommended to set this
1220 value too large (or to zero) because it would be an easy way to
1221 crash the kernel by allowing too many addresses to be created.
1222 Default: 16
1224 disable_ipv6 - BOOLEAN
1225 Disable IPv6 operation. If accept_dad is set to 2, this value
1226 will be dynamically set to TRUE if DAD fails for the link-local
1227 address.
1228 Default: FALSE (enable IPv6 operation)
1230 When this value is changed from 1 to 0 (IPv6 is being enabled),
1231 it will dynamically create a link-local address on the given
1232 interface and start Duplicate Address Detection, if necessary.
1234 When this value is changed from 0 to 1 (IPv6 is being disabled),
1235 it will dynamically delete all address on the given interface.
1237 accept_dad - INTEGER
1238 Whether to accept DAD (Duplicate Address Detection).
1239 0: Disable DAD
1240 1: Enable DAD (default)
1241 2: Enable DAD, and disable IPv6 operation if MAC-based duplicate
1242 link-local address has been found.
1244 force_tllao - BOOLEAN
1245 Enable sending the target link-layer address option even when
1246 responding to a unicast neighbor solicitation.
1247 Default: FALSE
1249 Quoting from RFC 2461, section 4.4, Target link-layer address:
1251 "The option MUST be included for multicast solicitations in order to
1252 avoid infinite Neighbor Solicitation "recursion" when the peer node
1253 does not have a cache entry to return a Neighbor Advertisements
1254 message. When responding to unicast solicitations, the option can be
1255 omitted since the sender of the solicitation has the correct link-
1256 layer address; otherwise it would not have be able to send the unicast
1257 solicitation in the first place. However, including the link-layer
1258 address in this case adds little overhead and eliminates a potential
1259 race condition where the sender deletes the cached link-layer address
1260 prior to receiving a response to a previous solicitation."
1262 icmp/*:
1263 ratelimit - INTEGER
1264 Limit the maximal rates for sending ICMPv6 packets.
1265 0 to disable any limiting,
1266 otherwise the minimal space between responses in milliseconds.
1267 Default: 1000
1270 IPv6 Update by:
1271 Pekka Savola <pekkas@netcore.fi>
1272 YOSHIFUJI Hideaki / USAGI Project <yoshfuji@linux-ipv6.org>
1275 /proc/sys/net/bridge/* Variables:
1277 bridge-nf-call-arptables - BOOLEAN
1278 1 : pass bridged ARP traffic to arptables' FORWARD chain.
1279 0 : disable this.
1280 Default: 1
1282 bridge-nf-call-iptables - BOOLEAN
1283 1 : pass bridged IPv4 traffic to iptables' chains.
1284 0 : disable this.
1285 Default: 1
1287 bridge-nf-call-ip6tables - BOOLEAN
1288 1 : pass bridged IPv6 traffic to ip6tables' chains.
1289 0 : disable this.
1290 Default: 1
1292 bridge-nf-filter-vlan-tagged - BOOLEAN
1293 1 : pass bridged vlan-tagged ARP/IP/IPv6 traffic to {arp,ip,ip6}tables.
1294 0 : disable this.
1295 Default: 1
1297 bridge-nf-filter-pppoe-tagged - BOOLEAN
1298 1 : pass bridged pppoe-tagged IP/IPv6 traffic to {ip,ip6}tables.
1299 0 : disable this.
1300 Default: 1
1303 proc/sys/net/sctp/* Variables:
1305 addip_enable - BOOLEAN
1306 Enable or disable extension of Dynamic Address Reconfiguration
1307 (ADD-IP) functionality specified in RFC5061. This extension provides
1308 the ability to dynamically add and remove new addresses for the SCTP
1309 associations.
1311 1: Enable extension.
1313 0: Disable extension.
1315 Default: 0
1317 addip_noauth_enable - BOOLEAN
1318 Dynamic Address Reconfiguration (ADD-IP) requires the use of
1319 authentication to protect the operations of adding or removing new
1320 addresses. This requirement is mandated so that unauthorized hosts
1321 would not be able to hijack associations. However, older
1322 implementations may not have implemented this requirement while
1323 allowing the ADD-IP extension. For reasons of interoperability,
1324 we provide this variable to control the enforcement of the
1325 authentication requirement.
1327 1: Allow ADD-IP extension to be used without authentication. This
1328 should only be set in a closed environment for interoperability
1329 with older implementations.
1331 0: Enforce the authentication requirement
1333 Default: 0
1335 auth_enable - BOOLEAN
1336 Enable or disable Authenticated Chunks extension. This extension
1337 provides the ability to send and receive authenticated chunks and is
1338 required for secure operation of Dynamic Address Reconfiguration
1339 (ADD-IP) extension.
1341 1: Enable this extension.
1342 0: Disable this extension.
1344 Default: 0
1346 prsctp_enable - BOOLEAN
1347 Enable or disable the Partial Reliability extension (RFC3758) which
1348 is used to notify peers that a given DATA should no longer be expected.
1350 1: Enable extension
1351 0: Disable
1353 Default: 1
1355 max_burst - INTEGER
1356 The limit of the number of new packets that can be initially sent. It
1357 controls how bursty the generated traffic can be.
1359 Default: 4
1361 association_max_retrans - INTEGER
1362 Set the maximum number for retransmissions that an association can
1363 attempt deciding that the remote end is unreachable. If this value
1364 is exceeded, the association is terminated.
1366 Default: 10
1368 max_init_retransmits - INTEGER
1369 The maximum number of retransmissions of INIT and COOKIE-ECHO chunks
1370 that an association will attempt before declaring the destination
1371 unreachable and terminating.
1373 Default: 8
1375 path_max_retrans - INTEGER
1376 The maximum number of retransmissions that will be attempted on a given
1377 path. Once this threshold is exceeded, the path is considered
1378 unreachable, and new traffic will use a different path when the
1379 association is multihomed.
1381 Default: 5
1383 rto_initial - INTEGER
1384 The initial round trip timeout value in milliseconds that will be used
1385 in calculating round trip times. This is the initial time interval
1386 for retransmissions.
1388 Default: 3000
1390 rto_max - INTEGER
1391 The maximum value (in milliseconds) of the round trip timeout. This
1392 is the largest time interval that can elapse between retransmissions.
1394 Default: 60000
1396 rto_min - INTEGER
1397 The minimum value (in milliseconds) of the round trip timeout. This
1398 is the smallest time interval the can elapse between retransmissions.
1400 Default: 1000
1402 hb_interval - INTEGER
1403 The interval (in milliseconds) between HEARTBEAT chunks. These chunks
1404 are sent at the specified interval on idle paths to probe the state of
1405 a given path between 2 associations.
1407 Default: 30000
1409 sack_timeout - INTEGER
1410 The amount of time (in milliseconds) that the implementation will wait
1411 to send a SACK.
1413 Default: 200
1415 valid_cookie_life - INTEGER
1416 The default lifetime of the SCTP cookie (in milliseconds). The cookie
1417 is used during association establishment.
1419 Default: 60000
1421 cookie_preserve_enable - BOOLEAN
1422 Enable or disable the ability to extend the lifetime of the SCTP cookie
1423 that is used during the establishment phase of SCTP association
1425 1: Enable cookie lifetime extension.
1426 0: Disable
1428 Default: 1
1430 rcvbuf_policy - INTEGER
1431 Determines if the receive buffer is attributed to the socket or to
1432 association. SCTP supports the capability to create multiple
1433 associations on a single socket. When using this capability, it is
1434 possible that a single stalled association that's buffering a lot
1435 of data may block other associations from delivering their data by
1436 consuming all of the receive buffer space. To work around this,
1437 the rcvbuf_policy could be set to attribute the receiver buffer space
1438 to each association instead of the socket. This prevents the described
1439 blocking.
1441 1: rcvbuf space is per association
1442 0: recbuf space is per socket
1444 Default: 0
1446 sndbuf_policy - INTEGER
1447 Similar to rcvbuf_policy above, this applies to send buffer space.
1449 1: Send buffer is tracked per association
1450 0: Send buffer is tracked per socket.
1452 Default: 0
1454 sctp_mem - vector of 3 INTEGERs: min, pressure, max
1455 Number of pages allowed for queueing by all SCTP sockets.
1457 min: Below this number of pages SCTP is not bothered about its
1458 memory appetite. When amount of memory allocated by SCTP exceeds
1459 this number, SCTP starts to moderate memory usage.
1461 pressure: This value was introduced to follow format of tcp_mem.
1463 max: Number of pages allowed for queueing by all SCTP sockets.
1465 Default is calculated at boot time from amount of available memory.
1467 sctp_rmem - vector of 3 INTEGERs: min, default, max
1468 See tcp_rmem for a description.
1470 sctp_wmem - vector of 3 INTEGERs: min, default, max
1471 See tcp_wmem for a description.
1473 addr_scope_policy - INTEGER
1474 Control IPv4 address scoping - draft-stewart-tsvwg-sctp-ipv4-00
1476 0 - Disable IPv4 address scoping
1477 1 - Enable IPv4 address scoping
1478 2 - Follow draft but allow IPv4 private addresses
1479 3 - Follow draft but allow IPv4 link local addresses
1481 Default: 1
1484 /proc/sys/net/core/*
1485 dev_weight - INTEGER
1486 The maximum number of packets that kernel can handle on a NAPI
1487 interrupt, it's a Per-CPU variable.
1489 Default: 64
1491 /proc/sys/net/unix/*
1492 max_dgram_qlen - INTEGER
1493 The maximum length of dgram socket receive queue
1495 Default: 10
1498 UNDOCUMENTED:
1500 /proc/sys/net/irda/*
1501 fast_poll_increase FIXME
1502 warn_noreply_time FIXME
1503 discovery_slots FIXME
1504 slot_timeout FIXME
1505 max_baud_rate FIXME
1506 discovery_timeout FIXME
1507 lap_keepalive_time FIXME
1508 max_noreply_time FIXME
1509 max_tx_data_size FIXME
1510 max_tx_window FIXME
1511 min_tx_turn_time FIXME