1 /proc/sys/net/ipv4/* Variables:
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
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.
23 default 552 - 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_thresh1 - INTEGER
30 Minimum number of entries to keep. Garbage collector will not
31 purge entries if there are fewer than this number.
34 neigh/default/gc_thresh3 - INTEGER
35 Maximum number of neighbor entries allowed. Increase this
36 when using large numbers of interfaces and when communicating
37 with large numbers of directly-connected peers.
40 neigh/default/unres_qlen_bytes - INTEGER
41 The maximum number of bytes which may be used by packets
42 queued for each unresolved address by other network layers.
44 Setting negative value is meaningless and will return error.
45 Default: 65536 Bytes(64KB)
47 neigh/default/unres_qlen - INTEGER
48 The maximum number of packets which may be queued for each
49 unresolved address by other network layers.
50 (deprecated in linux 3.3) : use unres_qlen_bytes instead.
51 Prior to linux 3.3, the default value is 3 which may cause
52 unexpected packet loss. The current default value is calculated
53 according to default value of unres_qlen_bytes and true size of
58 Time, in seconds, that cached PMTU information is kept.
61 The advertised MSS depends on the first hop route MTU, but will
62 never be lower than this setting.
66 ipfrag_high_thresh - INTEGER
67 Maximum memory used to reassemble IP fragments. When
68 ipfrag_high_thresh bytes of memory is allocated for this purpose,
69 the fragment handler will toss packets until ipfrag_low_thresh
72 ipfrag_low_thresh - INTEGER
73 See ipfrag_high_thresh
76 Time in seconds to keep an IP fragment in memory.
78 ipfrag_secret_interval - INTEGER
79 Regeneration interval (in seconds) of the hash secret (or lifetime
80 for the hash secret) for IP fragments.
83 ipfrag_max_dist - INTEGER
84 ipfrag_max_dist is a non-negative integer value which defines the
85 maximum "disorder" which is allowed among fragments which share a
86 common IP source address. Note that reordering of packets is
87 not unusual, but if a large number of fragments arrive from a source
88 IP address while a particular fragment queue remains incomplete, it
89 probably indicates that one or more fragments belonging to that queue
90 have been lost. When ipfrag_max_dist is positive, an additional check
91 is done on fragments before they are added to a reassembly queue - if
92 ipfrag_max_dist (or more) fragments have arrived from a particular IP
93 address between additions to any IP fragment queue using that source
94 address, it's presumed that one or more fragments in the queue are
95 lost. The existing fragment queue will be dropped, and a new one
96 started. An ipfrag_max_dist value of zero disables this check.
98 Using a very small value, e.g. 1 or 2, for ipfrag_max_dist can
99 result in unnecessarily dropping fragment queues when normal
100 reordering of packets occurs, which could lead to poor application
101 performance. Using a very large value, e.g. 50000, increases the
102 likelihood of incorrectly reassembling IP fragments that originate
103 from different IP datagrams, which could result in data corruption.
108 inet_peer_threshold - INTEGER
109 The approximate size of the storage. Starting from this threshold
110 entries will be thrown aggressively. This threshold also determines
111 entries' time-to-live and time intervals between garbage collection
112 passes. More entries, less time-to-live, less GC interval.
114 inet_peer_minttl - INTEGER
115 Minimum time-to-live of entries. Should be enough to cover fragment
116 time-to-live on the reassembling side. This minimum time-to-live is
117 guaranteed if the pool size is less than inet_peer_threshold.
120 inet_peer_maxttl - INTEGER
121 Maximum time-to-live of entries. Unused entries will expire after
122 this period of time if there is no memory pressure on the pool (i.e.
123 when the number of entries in the pool is very small).
129 Limit of socket listen() backlog, known in userspace as SOMAXCONN.
130 Defaults to 128. See also tcp_max_syn_backlog for additional tuning
133 tcp_abort_on_overflow - BOOLEAN
134 If listening service is too slow to accept new connections,
135 reset them. Default state is FALSE. It means that if overflow
136 occurred due to a burst, connection will recover. Enable this
137 option _only_ if you are really sure that listening daemon
138 cannot be tuned to accept connections faster. Enabling this
139 option can harm clients of your server.
141 tcp_adv_win_scale - INTEGER
142 Count buffering overhead as bytes/2^tcp_adv_win_scale
143 (if tcp_adv_win_scale > 0) or bytes-bytes/2^(-tcp_adv_win_scale),
145 Possible values are [-31, 31], inclusive.
148 tcp_allowed_congestion_control - STRING
149 Show/set the congestion control choices available to non-privileged
150 processes. The list is a subset of those listed in
151 tcp_available_congestion_control.
152 Default is "reno" and the default setting (tcp_congestion_control).
154 tcp_app_win - INTEGER
155 Reserve max(window/2^tcp_app_win, mss) of window for application
156 buffer. Value 0 is special, it means that nothing is reserved.
159 tcp_available_congestion_control - STRING
160 Shows the available congestion control choices that are registered.
161 More congestion control algorithms may be available as modules,
164 tcp_base_mss - INTEGER
165 The initial value of search_low to be used by the packetization layer
166 Path MTU discovery (MTU probing). If MTU probing is enabled,
167 this is the initial MSS used by the connection.
169 tcp_congestion_control - STRING
170 Set the congestion control algorithm to be used for new
171 connections. The algorithm "reno" is always available, but
172 additional choices may be available based on kernel configuration.
173 Default is set as part of kernel configuration.
174 For passive connections, the listener congestion control choice
176 [see setsockopt(listenfd, SOL_TCP, TCP_CONGESTION, "name" ...) ]
179 Allows TCP to send "duplicate" SACKs.
181 tcp_early_retrans - INTEGER
182 Enable Early Retransmit (ER), per RFC 5827. ER lowers the threshold
183 for triggering fast retransmit when the amount of outstanding data is
184 small and when no previously unsent data can be transmitted (such
185 that limited transmit could be used). Also controls the use of
186 Tail loss probe (TLP) that converts RTOs occuring due to tail
187 losses into fast recovery (draft-dukkipati-tcpm-tcp-loss-probe-01).
191 2 enables ER but delays fast recovery and fast retransmit
192 by a fourth of RTT. This mitigates connection falsely
193 recovers when network has a small degree of reordering
194 (less than 3 packets).
195 3 enables delayed ER and TLP.
200 Control use of Explicit Congestion Notification (ECN) by TCP.
201 ECN is used only when both ends of the TCP connection indicate
202 support for it. This feature is useful in avoiding losses due
203 to congestion by allowing supporting routers to signal
204 congestion before having to drop packets.
206 0 Disable ECN. Neither initiate nor accept ECN.
207 1 Enable ECN when requested by incoming connections and
208 also request ECN on outgoing connection attempts.
209 2 Enable ECN when requested by incoming connections
210 but do not request ECN on outgoing connections.
214 Enable FACK congestion avoidance and fast retransmission.
215 The value is not used, if tcp_sack is not enabled.
217 tcp_fin_timeout - INTEGER
218 The length of time an orphaned (no longer referenced by any
219 application) connection will remain in the FIN_WAIT_2 state
220 before it is aborted at the local end. While a perfectly
221 valid "receive only" state for an un-orphaned connection, an
222 orphaned connection in FIN_WAIT_2 state could otherwise wait
223 forever for the remote to close its end of the connection.
228 Enables Forward RTO-Recovery (F-RTO) defined in RFC4138.
229 F-RTO is an enhanced recovery algorithm for TCP retransmission
230 timeouts. It is particularly beneficial in wireless environments
231 where packet loss is typically due to random radio interference
232 rather than intermediate router congestion. F-RTO is sender-side
233 only modification. Therefore it does not require any support from
236 If set to 1, basic version is enabled. 2 enables SACK enhanced
237 F-RTO if flow uses SACK. The basic version can be used also when
238 SACK is in use though scenario(s) with it exists where F-RTO
239 interacts badly with the packet counting of the SACK enabled TCP
242 tcp_keepalive_time - INTEGER
243 How often TCP sends out keepalive messages when keepalive is enabled.
246 tcp_keepalive_probes - INTEGER
247 How many keepalive probes TCP sends out, until it decides that the
248 connection is broken. Default value: 9.
250 tcp_keepalive_intvl - INTEGER
251 How frequently the probes are send out. Multiplied by
252 tcp_keepalive_probes it is time to kill not responding connection,
253 after probes started. Default value: 75sec i.e. connection
254 will be aborted after ~11 minutes of retries.
256 tcp_low_latency - BOOLEAN
257 If set, the TCP stack makes decisions that prefer lower
258 latency as opposed to higher throughput. By default, this
259 option is not set meaning that higher throughput is preferred.
260 An example of an application where this default should be
261 changed would be a Beowulf compute cluster.
264 tcp_max_orphans - INTEGER
265 Maximal number of TCP sockets not attached to any user file handle,
266 held by system. If this number is exceeded orphaned connections are
267 reset immediately and warning is printed. This limit exists
268 only to prevent simple DoS attacks, you _must_ not rely on this
269 or lower the limit artificially, but rather increase it
270 (probably, after increasing installed memory),
271 if network conditions require more than default value,
272 and tune network services to linger and kill such states
273 more aggressively. Let me to remind again: each orphan eats
274 up to ~64K of unswappable memory.
276 tcp_max_ssthresh - INTEGER
277 Limited Slow-Start for TCP with large congestion windows (cwnd) defined in
278 RFC3742. Limited slow-start is a mechanism to limit growth of the cwnd
279 on the region where cwnd is larger than tcp_max_ssthresh. TCP increases cwnd
280 by at most tcp_max_ssthresh segments, and by at least tcp_max_ssthresh/2
281 segments per RTT when the cwnd is above tcp_max_ssthresh.
282 If TCP connection increased cwnd to thousands (or tens of thousands) segments,
283 and thousands of packets were being dropped during slow-start, you can set
284 tcp_max_ssthresh to improve performance for new TCP connection.
287 tcp_max_syn_backlog - INTEGER
288 Maximal number of remembered connection requests, which have not
289 received an acknowledgment from connecting client.
290 The minimal value is 128 for low memory machines, and it will
291 increase in proportion to the memory of machine.
292 If server suffers from overload, try increasing this number.
294 tcp_max_tw_buckets - INTEGER
295 Maximal number of timewait sockets held by system simultaneously.
296 If this number is exceeded time-wait socket is immediately destroyed
297 and warning is printed. This limit exists only to prevent
298 simple DoS attacks, you _must_ not lower the limit artificially,
299 but rather increase it (probably, after increasing installed memory),
300 if network conditions require more than default value.
302 tcp_mem - vector of 3 INTEGERs: min, pressure, max
303 min: below this number of pages TCP is not bothered about its
306 pressure: when amount of memory allocated by TCP exceeds this number
307 of pages, TCP moderates its memory consumption and enters memory
308 pressure mode, which is exited when memory consumption falls
311 max: number of pages allowed for queueing by all TCP sockets.
313 Defaults are calculated at boot time from amount of available
316 tcp_moderate_rcvbuf - BOOLEAN
317 If set, TCP performs receive buffer auto-tuning, attempting to
318 automatically size the buffer (no greater than tcp_rmem[2]) to
319 match the size required by the path for full throughput. Enabled by
322 tcp_mtu_probing - INTEGER
323 Controls TCP Packetization-Layer Path MTU Discovery. Takes three
326 1 - Disabled by default, enabled when an ICMP black hole detected
327 2 - Always enabled, use initial MSS of tcp_base_mss.
329 tcp_no_metrics_save - BOOLEAN
330 By default, TCP saves various connection metrics in the route cache
331 when the connection closes, so that connections established in the
332 near future can use these to set initial conditions. Usually, this
333 increases overall performance, but may sometimes cause performance
334 degradation. If set, TCP will not cache metrics on closing
337 tcp_orphan_retries - INTEGER
338 This value influences the timeout of a locally closed TCP connection,
339 when RTO retransmissions remain unacknowledged.
340 See tcp_retries2 for more details.
342 The default value is 8.
343 If your machine is a loaded WEB server,
344 you should think about lowering this value, such sockets
345 may consume significant resources. Cf. tcp_max_orphans.
347 tcp_reordering - INTEGER
348 Maximal reordering of packets in a TCP stream.
351 tcp_retrans_collapse - BOOLEAN
352 Bug-to-bug compatibility with some broken printers.
353 On retransmit try to send bigger packets to work around bugs in
356 tcp_retries1 - INTEGER
357 This value influences the time, after which TCP decides, that
358 something is wrong due to unacknowledged RTO retransmissions,
359 and reports this suspicion to the network layer.
360 See tcp_retries2 for more details.
362 RFC 1122 recommends at least 3 retransmissions, which is the
365 tcp_retries2 - INTEGER
366 This value influences the timeout of an alive TCP connection,
367 when RTO retransmissions remain unacknowledged.
368 Given a value of N, a hypothetical TCP connection following
369 exponential backoff with an initial RTO of TCP_RTO_MIN would
370 retransmit N times before killing the connection at the (N+1)th RTO.
372 The default value of 15 yields a hypothetical timeout of 924.6
373 seconds and is a lower bound for the effective timeout.
374 TCP will effectively time out at the first RTO which exceeds the
375 hypothetical timeout.
377 RFC 1122 recommends at least 100 seconds for the timeout,
378 which corresponds to a value of at least 8.
380 tcp_rfc1337 - BOOLEAN
381 If set, the TCP stack behaves conforming to RFC1337. If unset,
382 we are not conforming to RFC, but prevent TCP TIME_WAIT
386 tcp_rmem - vector of 3 INTEGERs: min, default, max
387 min: Minimal size of receive buffer used by TCP sockets.
388 It is guaranteed to each TCP socket, even under moderate memory
392 default: initial size of receive buffer used by TCP sockets.
393 This value overrides net.core.rmem_default used by other protocols.
394 Default: 87380 bytes. This value results in window of 65535 with
395 default setting of tcp_adv_win_scale and tcp_app_win:0 and a bit
396 less for default tcp_app_win. See below about these variables.
398 max: maximal size of receive buffer allowed for automatically
399 selected receiver buffers for TCP socket. This value does not override
400 net.core.rmem_max. Calling setsockopt() with SO_RCVBUF disables
401 automatic tuning of that socket's receive buffer size, in which
402 case this value is ignored.
403 Default: between 87380B and 6MB, depending on RAM size.
406 Enable select acknowledgments (SACKS).
408 tcp_slow_start_after_idle - BOOLEAN
409 If set, provide RFC2861 behavior and time out the congestion
410 window after an idle period. An idle period is defined at
411 the current RTO. If unset, the congestion window will not
412 be timed out after an idle period.
416 Use the Host requirements interpretation of the TCP urgent pointer field.
417 Most hosts use the older BSD interpretation, so if you turn this on
418 Linux might not communicate correctly with them.
421 tcp_synack_retries - INTEGER
422 Number of times SYNACKs for a passive TCP connection attempt will
423 be retransmitted. Should not be higher than 255. Default value
424 is 5, which corresponds to 31seconds till the last retransmission
425 with the current initial RTO of 1second. With this the final timeout
426 for a passive TCP connection will happen after 63seconds.
428 tcp_syncookies - BOOLEAN
429 Only valid when the kernel was compiled with CONFIG_SYNCOOKIES
430 Send out syncookies when the syn backlog queue of a socket
431 overflows. This is to prevent against the common 'SYN flood attack'
434 Note, that syncookies is fallback facility.
435 It MUST NOT be used to help highly loaded servers to stand
436 against legal connection rate. If you see SYN flood warnings
437 in your logs, but investigation shows that they occur
438 because of overload with legal connections, you should tune
439 another parameters until this warning disappear.
440 See: tcp_max_syn_backlog, tcp_synack_retries, tcp_abort_on_overflow.
442 syncookies seriously violate TCP protocol, do not allow
443 to use TCP extensions, can result in serious degradation
444 of some services (f.e. SMTP relaying), visible not by you,
445 but your clients and relays, contacting you. While you see
446 SYN flood warnings in logs not being really flooded, your server
447 is seriously misconfigured.
449 tcp_fastopen - INTEGER
450 Enable TCP Fast Open feature (draft-ietf-tcpm-fastopen) to send data
451 in the opening SYN packet. To use this feature, the client application
452 must use sendmsg() or sendto() with MSG_FASTOPEN flag rather than
453 connect() to perform a TCP handshake automatically.
455 The values (bitmap) are
456 1: Enables sending data in the opening SYN on the client.
457 2: Enables TCP Fast Open on the server side, i.e., allowing data in
458 a SYN packet to be accepted and passed to the application before
459 3-way hand shake finishes.
460 4: Send data in the opening SYN regardless of cookie availability and
461 without a cookie option.
462 0x100: Accept SYN data w/o validating the cookie.
463 0x200: Accept data-in-SYN w/o any cookie option present.
464 0x400/0x800: Enable Fast Open on all listeners regardless of the
465 TCP_FASTOPEN socket option. The two different flags designate two
466 different ways of setting max_qlen without the TCP_FASTOPEN socket
471 Note that the client & server side Fast Open flags (1 and 2
472 respectively) must be also enabled before the rest of flags can take
475 See include/net/tcp.h and the code for more details.
477 tcp_syn_retries - INTEGER
478 Number of times initial SYNs for an active TCP connection attempt
479 will be retransmitted. Should not be higher than 255. Default value
480 is 6, which corresponds to 63seconds till the last retransmission
481 with the current initial RTO of 1second. With this the final timeout
482 for an active TCP connection attempt will happen after 127seconds.
484 tcp_timestamps - BOOLEAN
485 Enable timestamps as defined in RFC1323.
487 tcp_tso_win_divisor - INTEGER
488 This allows control over what percentage of the congestion window
489 can be consumed by a single TSO frame.
490 The setting of this parameter is a choice between burstiness and
491 building larger TSO frames.
494 tcp_tw_recycle - BOOLEAN
495 Enable fast recycling TIME-WAIT sockets. Default value is 0.
496 It should not be changed without advice/request of technical
499 tcp_tw_reuse - BOOLEAN
500 Allow to reuse TIME-WAIT sockets for new connections when it is
501 safe from protocol viewpoint. Default value is 0.
502 It should not be changed without advice/request of technical
505 tcp_window_scaling - BOOLEAN
506 Enable window scaling as defined in RFC1323.
508 tcp_wmem - vector of 3 INTEGERs: min, default, max
509 min: Amount of memory reserved for send buffers for TCP sockets.
510 Each TCP socket has rights to use it due to fact of its birth.
513 default: initial size of send buffer used by TCP sockets. This
514 value overrides net.core.wmem_default used by other protocols.
515 It is usually lower than net.core.wmem_default.
518 max: Maximal amount of memory allowed for automatically tuned
519 send buffers for TCP sockets. This value does not override
520 net.core.wmem_max. Calling setsockopt() with SO_SNDBUF disables
521 automatic tuning of that socket's send buffer size, in which case
522 this value is ignored.
523 Default: between 64K and 4MB, depending on RAM size.
525 tcp_workaround_signed_windows - BOOLEAN
526 If set, assume no receipt of a window scaling option means the
527 remote TCP is broken and treats the window as a signed quantity.
528 If unset, assume the remote TCP is not broken even if we do
529 not receive a window scaling option from them.
532 tcp_dma_copybreak - INTEGER
533 Lower limit, in bytes, of the size of socket reads that will be
534 offloaded to a DMA copy engine, if one is present in the system
535 and CONFIG_NET_DMA is enabled.
538 tcp_thin_linear_timeouts - BOOLEAN
539 Enable dynamic triggering of linear timeouts for thin streams.
540 If set, a check is performed upon retransmission by timeout to
541 determine if the stream is thin (less than 4 packets in flight).
542 As long as the stream is found to be thin, up to 6 linear
543 timeouts may be performed before exponential backoff mode is
544 initiated. This improves retransmission latency for
545 non-aggressive thin streams, often found to be time-dependent.
546 For more information on thin streams, see
547 Documentation/networking/tcp-thin.txt
550 tcp_thin_dupack - BOOLEAN
551 Enable dynamic triggering of retransmissions after one dupACK
552 for thin streams. If set, a check is performed upon reception
553 of a dupACK to determine if the stream is thin (less than 4
554 packets in flight). As long as the stream is found to be thin,
555 data is retransmitted on the first received dupACK. This
556 improves retransmission latency for non-aggressive thin
557 streams, often found to be time-dependent.
558 For more information on thin streams, see
559 Documentation/networking/tcp-thin.txt
562 tcp_limit_output_bytes - INTEGER
563 Controls TCP Small Queue limit per tcp socket.
564 TCP bulk sender tends to increase packets in flight until it
565 gets losses notifications. With SNDBUF autotuning, this can
566 result in a large amount of packets queued in qdisc/device
567 on the local machine, hurting latency of other flows, for
568 typical pfifo_fast qdiscs.
569 tcp_limit_output_bytes limits the number of bytes on qdisc
570 or device to reduce artificial RTT/cwnd and reduce bufferbloat.
571 Note: For GSO/TSO enabled flows, we try to have at least two
572 packets in flight. Reducing tcp_limit_output_bytes might also
573 reduce the size of individual GSO packet (64KB being the max)
576 tcp_challenge_ack_limit - INTEGER
577 Limits number of Challenge ACK sent per second, as recommended
578 in RFC 5961 (Improving TCP's Robustness to Blind In-Window Attacks)
583 udp_mem - vector of 3 INTEGERs: min, pressure, max
584 Number of pages allowed for queueing by all UDP sockets.
586 min: Below this number of pages UDP is not bothered about its
587 memory appetite. When amount of memory allocated by UDP exceeds
588 this number, UDP starts to moderate memory usage.
590 pressure: This value was introduced to follow format of tcp_mem.
592 max: Number of pages allowed for queueing by all UDP sockets.
594 Default is calculated at boot time from amount of available memory.
596 udp_rmem_min - INTEGER
597 Minimal size of receive buffer used by UDP sockets in moderation.
598 Each UDP socket is able to use the size for receiving data, even if
599 total pages of UDP sockets exceed udp_mem pressure. The unit is byte.
602 udp_wmem_min - INTEGER
603 Minimal size of send buffer used by UDP sockets in moderation.
604 Each UDP socket is able to use the size for sending data, even if
605 total pages of UDP sockets exceed udp_mem pressure. The unit is byte.
610 cipso_cache_enable - BOOLEAN
611 If set, enable additions to and lookups from the CIPSO label mapping
612 cache. If unset, additions are ignored and lookups always result in a
613 miss. However, regardless of the setting the cache is still
614 invalidated when required when means you can safely toggle this on and
615 off and the cache will always be "safe".
618 cipso_cache_bucket_size - INTEGER
619 The CIPSO label cache consists of a fixed size hash table with each
620 hash bucket containing a number of cache entries. This variable limits
621 the number of entries in each hash bucket; the larger the value the
622 more CIPSO label mappings that can be cached. When the number of
623 entries in a given hash bucket reaches this limit adding new entries
624 causes the oldest entry in the bucket to be removed to make room.
627 cipso_rbm_optfmt - BOOLEAN
628 Enable the "Optimized Tag 1 Format" as defined in section 3.4.2.6 of
629 the CIPSO draft specification (see Documentation/netlabel for details).
630 This means that when set the CIPSO tag will be padded with empty
631 categories in order to make the packet data 32-bit aligned.
634 cipso_rbm_structvalid - BOOLEAN
635 If set, do a very strict check of the CIPSO option when
636 ip_options_compile() is called. If unset, relax the checks done during
637 ip_options_compile(). Either way is "safe" as errors are caught else
638 where in the CIPSO processing code but setting this to 0 (False) should
639 result in less work (i.e. it should be faster) but could cause problems
640 with other implementations that require strict checking.
645 ip_local_port_range - 2 INTEGERS
646 Defines the local port range that is used by TCP and UDP to
647 choose the local port. The first number is the first, the
648 second the last local port number. The default values are
649 32768 and 61000 respectively.
651 ip_local_reserved_ports - list of comma separated ranges
652 Specify the ports which are reserved for known third-party
653 applications. These ports will not be used by automatic port
654 assignments (e.g. when calling connect() or bind() with port
655 number 0). Explicit port allocation behavior is unchanged.
657 The format used for both input and output is a comma separated
658 list of ranges (e.g. "1,2-4,10-10" for ports 1, 2, 3, 4 and
659 10). Writing to the file will clear all previously reserved
660 ports and update the current list with the one given in the
663 Note that ip_local_port_range and ip_local_reserved_ports
664 settings are independent and both are considered by the kernel
665 when determining which ports are available for automatic port
668 You can reserve ports which are not in the current
669 ip_local_port_range, e.g.:
671 $ cat /proc/sys/net/ipv4/ip_local_port_range
673 $ cat /proc/sys/net/ipv4/ip_local_reserved_ports
676 although this is redundant. However such a setting is useful
677 if later the port range is changed to a value that will
678 include the reserved ports.
682 ip_nonlocal_bind - BOOLEAN
683 If set, allows processes to bind() to non-local IP addresses,
684 which can be quite useful - but may break some applications.
688 If set non-zero, enables support for dynamic addresses.
689 If set to a non-zero value larger than 1, a kernel log
690 message will be printed when dynamic address rewriting
694 icmp_echo_ignore_all - BOOLEAN
695 If set non-zero, then the kernel will ignore all ICMP ECHO
699 icmp_echo_ignore_broadcasts - BOOLEAN
700 If set non-zero, then the kernel will ignore all ICMP ECHO and
701 TIMESTAMP requests sent to it via broadcast/multicast.
704 icmp_ratelimit - INTEGER
705 Limit the maximal rates for sending ICMP packets whose type matches
706 icmp_ratemask (see below) to specific targets.
707 0 to disable any limiting,
708 otherwise the minimal space between responses in milliseconds.
711 icmp_ratemask - INTEGER
712 Mask made of ICMP types for which rates are being limited.
713 Significant bits: IHGFEDCBA9876543210
714 Default mask: 0000001100000011000 (6168)
716 Bit definitions (see include/linux/icmp.h):
718 3 Destination Unreachable *
723 C Parameter Problem *
728 H Address Mask Request
731 * These are rate limited by default (see default mask above)
733 icmp_ignore_bogus_error_responses - BOOLEAN
734 Some routers violate RFC1122 by sending bogus responses to broadcast
735 frames. Such violations are normally logged via a kernel warning.
736 If this is set to TRUE, the kernel will not give such warnings, which
737 will avoid log file clutter.
740 icmp_errors_use_inbound_ifaddr - BOOLEAN
742 If zero, icmp error messages are sent with the primary address of
743 the exiting interface.
745 If non-zero, the message will be sent with the primary address of
746 the interface that received the packet that caused the icmp error.
747 This is the behaviour network many administrators will expect from
748 a router. And it can make debugging complicated network layouts
751 Note that if no primary address exists for the interface selected,
752 then the primary address of the first non-loopback interface that
753 has one will be used regardless of this setting.
757 igmp_max_memberships - INTEGER
758 Change the maximum number of multicast groups we can subscribe to.
761 Theoretical maximum value is bounded by having to send a membership
762 report in a single datagram (i.e. the report can't span multiple
763 datagrams, or risk confusing the switch and leaving groups you don't
766 The number of supported groups 'M' is bounded by the number of group
767 report entries you can fit into a single datagram of 65535 bytes.
769 M = 65536-sizeof (ip header)/(sizeof(Group record))
771 Group records are variable length, with a minimum of 12 bytes.
772 So net.ipv4.igmp_max_memberships should not be set higher than:
774 (65536-24) / 12 = 5459
776 The value 5459 assumes no IP header options, so in practice
777 this number may be lower.
779 conf/interface/* changes special settings per interface (where
780 "interface" is the name of your network interface)
782 conf/all/* is special, changes the settings for all interfaces
784 log_martians - BOOLEAN
785 Log packets with impossible addresses to kernel log.
786 log_martians for the interface will be enabled if at least one of
787 conf/{all,interface}/log_martians is set to TRUE,
788 it will be disabled otherwise
790 accept_redirects - BOOLEAN
791 Accept ICMP redirect messages.
792 accept_redirects for the interface will be enabled if:
793 - both conf/{all,interface}/accept_redirects are TRUE in the case
794 forwarding for the interface is enabled
796 - at least one of conf/{all,interface}/accept_redirects is TRUE in the
797 case forwarding for the interface is disabled
798 accept_redirects for the interface will be disabled otherwise
803 Enable IP forwarding on this interface.
805 mc_forwarding - BOOLEAN
806 Do multicast routing. The kernel needs to be compiled with CONFIG_MROUTE
807 and a multicast routing daemon is required.
808 conf/all/mc_forwarding must also be set to TRUE to enable multicast
809 routing for the interface
812 Integer value used to differentiate the devices by the medium they
813 are attached to. Two devices can have different id values when
814 the broadcast packets are received only on one of them.
815 The default value 0 means that the device is the only interface
816 to its medium, value of -1 means that medium is not known.
818 Currently, it is used to change the proxy_arp behavior:
819 the proxy_arp feature is enabled for packets forwarded between
820 two devices attached to different media.
824 proxy_arp for the interface will be enabled if at least one of
825 conf/{all,interface}/proxy_arp is set to TRUE,
826 it will be disabled otherwise
828 proxy_arp_pvlan - BOOLEAN
829 Private VLAN proxy arp.
830 Basically allow proxy arp replies back to the same interface
831 (from which the ARP request/solicitation was received).
833 This is done to support (ethernet) switch features, like RFC
834 3069, where the individual ports are NOT allowed to
835 communicate with each other, but they are allowed to talk to
836 the upstream router. As described in RFC 3069, it is possible
837 to allow these hosts to communicate through the upstream
838 router by proxy_arp'ing. Don't need to be used together with
841 This technology is known by different names:
842 In RFC 3069 it is called VLAN Aggregation.
843 Cisco and Allied Telesyn call it Private VLAN.
844 Hewlett-Packard call it Source-Port filtering or port-isolation.
845 Ericsson call it MAC-Forced Forwarding (RFC Draft).
847 shared_media - BOOLEAN
848 Send(router) or accept(host) RFC1620 shared media redirects.
849 Overrides ip_secure_redirects.
850 shared_media for the interface will be enabled if at least one of
851 conf/{all,interface}/shared_media is set to TRUE,
852 it will be disabled otherwise
855 secure_redirects - BOOLEAN
856 Accept ICMP redirect messages only for gateways,
857 listed in default gateway list.
858 secure_redirects for the interface will be enabled if at least one of
859 conf/{all,interface}/secure_redirects is set to TRUE,
860 it will be disabled otherwise
863 send_redirects - BOOLEAN
864 Send redirects, if router.
865 send_redirects for the interface will be enabled if at least one of
866 conf/{all,interface}/send_redirects is set to TRUE,
867 it will be disabled otherwise
870 bootp_relay - BOOLEAN
871 Accept packets with source address 0.b.c.d destined
872 not to this host as local ones. It is supposed, that
873 BOOTP relay daemon will catch and forward such packets.
874 conf/all/bootp_relay must also be set to TRUE to enable BOOTP relay
879 accept_source_route - BOOLEAN
880 Accept packets with SRR option.
881 conf/all/accept_source_route must also be set to TRUE to accept packets
882 with SRR option on the interface
883 default TRUE (router)
886 accept_local - BOOLEAN
887 Accept packets with local source addresses. In combination
888 with suitable routing, this can be used to direct packets
889 between two local interfaces over the wire and have them
892 rp_filter must be set to a non-zero value in order for
893 accept_local to have an effect.
897 route_localnet - BOOLEAN
898 Do not consider loopback addresses as martian source or destination
899 while routing. This enables the use of 127/8 for local routing purposes.
903 0 - No source validation.
904 1 - Strict mode as defined in RFC3704 Strict Reverse Path
905 Each incoming packet is tested against the FIB and if the interface
906 is not the best reverse path the packet check will fail.
907 By default failed packets are discarded.
908 2 - Loose mode as defined in RFC3704 Loose Reverse Path
909 Each incoming packet's source address is also tested against the FIB
910 and if the source address is not reachable via any interface
911 the packet check will fail.
913 Current recommended practice in RFC3704 is to enable strict mode
914 to prevent IP spoofing from DDos attacks. If using asymmetric routing
915 or other complicated routing, then loose mode is recommended.
917 The max value from conf/{all,interface}/rp_filter is used
918 when doing source validation on the {interface}.
920 Default value is 0. Note that some distributions enable it
924 1 - Allows you to have multiple network interfaces on the same
925 subnet, and have the ARPs for each interface be answered
926 based on whether or not the kernel would route a packet from
927 the ARP'd IP out that interface (therefore you must use source
928 based routing for this to work). In other words it allows control
929 of which cards (usually 1) will respond to an arp request.
931 0 - (default) The kernel can respond to arp requests with addresses
932 from other interfaces. This may seem wrong but it usually makes
933 sense, because it increases the chance of successful communication.
934 IP addresses are owned by the complete host on Linux, not by
935 particular interfaces. Only for more complex setups like load-
936 balancing, does this behaviour cause problems.
938 arp_filter for the interface will be enabled if at least one of
939 conf/{all,interface}/arp_filter is set to TRUE,
940 it will be disabled otherwise
942 arp_announce - INTEGER
943 Define different restriction levels for announcing the local
944 source IP address from IP packets in ARP requests sent on
946 0 - (default) Use any local address, configured on any interface
947 1 - Try to avoid local addresses that are not in the target's
948 subnet for this interface. This mode is useful when target
949 hosts reachable via this interface require the source IP
950 address in ARP requests to be part of their logical network
951 configured on the receiving interface. When we generate the
952 request we will check all our subnets that include the
953 target IP and will preserve the source address if it is from
954 such subnet. If there is no such subnet we select source
955 address according to the rules for level 2.
956 2 - Always use the best local address for this target.
957 In this mode we ignore the source address in the IP packet
958 and try to select local address that we prefer for talks with
959 the target host. Such local address is selected by looking
960 for primary IP addresses on all our subnets on the outgoing
961 interface that include the target IP address. If no suitable
962 local address is found we select the first local address
963 we have on the outgoing interface or on all other interfaces,
964 with the hope we will receive reply for our request and
965 even sometimes no matter the source IP address we announce.
967 The max value from conf/{all,interface}/arp_announce is used.
969 Increasing the restriction level gives more chance for
970 receiving answer from the resolved target while decreasing
971 the level announces more valid sender's information.
974 Define different modes for sending replies in response to
975 received ARP requests that resolve local target IP addresses:
976 0 - (default): reply for any local target IP address, configured
978 1 - reply only if the target IP address is local address
979 configured on the incoming interface
980 2 - reply only if the target IP address is local address
981 configured on the incoming interface and both with the
982 sender's IP address are part from same subnet on this interface
983 3 - do not reply for local addresses configured with scope host,
984 only resolutions for global and link addresses are replied
986 8 - do not reply for all local addresses
988 The max value from conf/{all,interface}/arp_ignore is used
989 when ARP request is received on the {interface}
992 Define mode for notification of address and device changes.
993 0 - (default): do nothing
994 1 - Generate gratuitous arp requests when device is brought up
995 or hardware address changes.
998 Define behavior for gratuitous ARP frames who's IP is not
999 already present in the ARP table:
1000 0 - don't create new entries in the ARP table
1001 1 - create new entries in the ARP table
1003 Both replies and requests type gratuitous arp will trigger the
1004 ARP table to be updated, if this setting is on.
1006 If the ARP table already contains the IP address of the
1007 gratuitous arp frame, the arp table will be updated regardless
1008 if this setting is on or off.
1011 app_solicit - INTEGER
1012 The maximum number of probes to send to the user space ARP daemon
1013 via netlink before dropping back to multicast probes (see
1014 mcast_solicit). Defaults to 0.
1016 disable_policy - BOOLEAN
1017 Disable IPSEC policy (SPD) for this interface
1019 disable_xfrm - BOOLEAN
1020 Disable IPSEC encryption on this interface, whatever the policy
1025 Allows you to write a number, which can be used as required.
1029 kuznet@ms2.inr.ac.ru
1035 delon.nicolas@wanadoo.fr
1040 /proc/sys/net/ipv6/* Variables:
1042 IPv6 has no global variables such as tcp_*. tcp_* settings under ipv4/ also
1043 apply to IPv6 [XXX?].
1045 bindv6only - BOOLEAN
1046 Default value for IPV6_V6ONLY socket option,
1047 which restricts use of the IPv6 socket to IPv6 communication
1049 TRUE: disable IPv4-mapped address feature
1050 FALSE: enable IPv4-mapped address feature
1052 Default: FALSE (as specified in RFC3493)
1056 ip6frag_high_thresh - INTEGER
1057 Maximum memory used to reassemble IPv6 fragments. When
1058 ip6frag_high_thresh bytes of memory is allocated for this purpose,
1059 the fragment handler will toss packets until ip6frag_low_thresh
1062 ip6frag_low_thresh - INTEGER
1063 See ip6frag_high_thresh
1065 ip6frag_time - INTEGER
1066 Time in seconds to keep an IPv6 fragment in memory.
1068 ip6frag_secret_interval - INTEGER
1069 Regeneration interval (in seconds) of the hash secret (or lifetime
1070 for the hash secret) for IPv6 fragments.
1074 Change the interface-specific default settings.
1078 Change all the interface-specific settings.
1080 [XXX: Other special features than forwarding?]
1082 conf/all/forwarding - BOOLEAN
1083 Enable global IPv6 forwarding between all interfaces.
1085 IPv4 and IPv6 work differently here; e.g. netfilter must be used
1086 to control which interfaces may forward packets and which not.
1088 This also sets all interfaces' Host/Router setting
1089 'forwarding' to the specified value. See below for details.
1091 This referred to as global forwarding.
1097 Change special settings per interface.
1099 The functional behaviour for certain settings is different
1100 depending on whether local forwarding is enabled or not.
1103 Accept Router Advertisements; autoconfigure using them.
1105 It also determines whether or not to transmit Router
1106 Solicitations. If and only if the functional setting is to
1107 accept Router Advertisements, Router Solicitations will be
1110 Possible values are:
1111 0 Do not accept Router Advertisements.
1112 1 Accept Router Advertisements if forwarding is disabled.
1113 2 Overrule forwarding behaviour. Accept Router Advertisements
1114 even if forwarding is enabled.
1116 Functional default: enabled if local forwarding is disabled.
1117 disabled if local forwarding is enabled.
1119 accept_ra_defrtr - BOOLEAN
1120 Learn default router in Router Advertisement.
1122 Functional default: enabled if accept_ra is enabled.
1123 disabled if accept_ra is disabled.
1125 accept_ra_pinfo - BOOLEAN
1126 Learn Prefix Information in Router Advertisement.
1128 Functional default: enabled if accept_ra is enabled.
1129 disabled if accept_ra is disabled.
1131 accept_ra_rt_info_max_plen - INTEGER
1132 Maximum prefix length of Route Information in RA.
1134 Route Information w/ prefix larger than or equal to this
1135 variable shall be ignored.
1137 Functional default: 0 if accept_ra_rtr_pref is enabled.
1138 -1 if accept_ra_rtr_pref is disabled.
1140 accept_ra_rtr_pref - BOOLEAN
1141 Accept Router Preference in RA.
1143 Functional default: enabled if accept_ra is enabled.
1144 disabled if accept_ra is disabled.
1146 accept_redirects - BOOLEAN
1149 Functional default: enabled if local forwarding is disabled.
1150 disabled if local forwarding is enabled.
1152 accept_source_route - INTEGER
1153 Accept source routing (routing extension header).
1155 >= 0: Accept only routing header type 2.
1156 < 0: Do not accept routing header.
1161 Autoconfigure addresses using Prefix Information in Router
1164 Functional default: enabled if accept_ra_pinfo is enabled.
1165 disabled if accept_ra_pinfo is disabled.
1167 dad_transmits - INTEGER
1168 The amount of Duplicate Address Detection probes to send.
1171 forwarding - INTEGER
1172 Configure interface-specific Host/Router behaviour.
1174 Note: It is recommended to have the same setting on all
1175 interfaces; mixed router/host scenarios are rather uncommon.
1177 Possible values are:
1178 0 Forwarding disabled
1179 1 Forwarding enabled
1183 By default, Host behaviour is assumed. This means:
1185 1. IsRouter flag is not set in Neighbour Advertisements.
1186 2. If accept_ra is TRUE (default), transmit Router
1188 3. If accept_ra is TRUE (default), accept Router
1189 Advertisements (and do autoconfiguration).
1190 4. If accept_redirects is TRUE (default), accept Redirects.
1194 If local forwarding is enabled, Router behaviour is assumed.
1195 This means exactly the reverse from the above:
1197 1. IsRouter flag is set in Neighbour Advertisements.
1198 2. Router Solicitations are not sent unless accept_ra is 2.
1199 3. Router Advertisements are ignored unless accept_ra is 2.
1200 4. Redirects are ignored.
1202 Default: 0 (disabled) if global forwarding is disabled (default),
1203 otherwise 1 (enabled).
1206 Default Hop Limit to set.
1210 Default Maximum Transfer Unit
1211 Default: 1280 (IPv6 required minimum)
1213 router_probe_interval - INTEGER
1214 Minimum interval (in seconds) between Router Probing described
1219 router_solicitation_delay - INTEGER
1220 Number of seconds to wait after interface is brought up
1221 before sending Router Solicitations.
1224 router_solicitation_interval - INTEGER
1225 Number of seconds to wait between Router Solicitations.
1228 router_solicitations - INTEGER
1229 Number of Router Solicitations to send until assuming no
1230 routers are present.
1233 use_tempaddr - INTEGER
1234 Preference for Privacy Extensions (RFC3041).
1235 <= 0 : disable Privacy Extensions
1236 == 1 : enable Privacy Extensions, but prefer public
1237 addresses over temporary addresses.
1238 > 1 : enable Privacy Extensions and prefer temporary
1239 addresses over public addresses.
1240 Default: 0 (for most devices)
1241 -1 (for point-to-point devices and loopback devices)
1243 temp_valid_lft - INTEGER
1244 valid lifetime (in seconds) for temporary addresses.
1245 Default: 604800 (7 days)
1247 temp_prefered_lft - INTEGER
1248 Preferred lifetime (in seconds) for temporary addresses.
1249 Default: 86400 (1 day)
1251 max_desync_factor - INTEGER
1252 Maximum value for DESYNC_FACTOR, which is a random value
1253 that ensures that clients don't synchronize with each
1254 other and generate new addresses at exactly the same time.
1255 value is in seconds.
1258 regen_max_retry - INTEGER
1259 Number of attempts before give up attempting to generate
1260 valid temporary addresses.
1263 max_addresses - INTEGER
1264 Maximum number of autoconfigured addresses per interface. Setting
1265 to zero disables the limitation. It is not recommended to set this
1266 value too large (or to zero) because it would be an easy way to
1267 crash the kernel by allowing too many addresses to be created.
1270 disable_ipv6 - BOOLEAN
1271 Disable IPv6 operation. If accept_dad is set to 2, this value
1272 will be dynamically set to TRUE if DAD fails for the link-local
1274 Default: FALSE (enable IPv6 operation)
1276 When this value is changed from 1 to 0 (IPv6 is being enabled),
1277 it will dynamically create a link-local address on the given
1278 interface and start Duplicate Address Detection, if necessary.
1280 When this value is changed from 0 to 1 (IPv6 is being disabled),
1281 it will dynamically delete all address on the given interface.
1283 accept_dad - INTEGER
1284 Whether to accept DAD (Duplicate Address Detection).
1286 1: Enable DAD (default)
1287 2: Enable DAD, and disable IPv6 operation if MAC-based duplicate
1288 link-local address has been found.
1290 force_tllao - BOOLEAN
1291 Enable sending the target link-layer address option even when
1292 responding to a unicast neighbor solicitation.
1295 Quoting from RFC 2461, section 4.4, Target link-layer address:
1297 "The option MUST be included for multicast solicitations in order to
1298 avoid infinite Neighbor Solicitation "recursion" when the peer node
1299 does not have a cache entry to return a Neighbor Advertisements
1300 message. When responding to unicast solicitations, the option can be
1301 omitted since the sender of the solicitation has the correct link-
1302 layer address; otherwise it would not have be able to send the unicast
1303 solicitation in the first place. However, including the link-layer
1304 address in this case adds little overhead and eliminates a potential
1305 race condition where the sender deletes the cached link-layer address
1306 prior to receiving a response to a previous solicitation."
1308 ndisc_notify - BOOLEAN
1309 Define mode for notification of address and device changes.
1310 0 - (default): do nothing
1311 1 - Generate unsolicited neighbour advertisements when device is brought
1312 up or hardware address changes.
1316 Limit the maximal rates for sending ICMPv6 packets.
1317 0 to disable any limiting,
1318 otherwise the minimal space between responses in milliseconds.
1323 Pekka Savola <pekkas@netcore.fi>
1324 YOSHIFUJI Hideaki / USAGI Project <yoshfuji@linux-ipv6.org>
1327 /proc/sys/net/bridge/* Variables:
1329 bridge-nf-call-arptables - BOOLEAN
1330 1 : pass bridged ARP traffic to arptables' FORWARD chain.
1334 bridge-nf-call-iptables - BOOLEAN
1335 1 : pass bridged IPv4 traffic to iptables' chains.
1339 bridge-nf-call-ip6tables - BOOLEAN
1340 1 : pass bridged IPv6 traffic to ip6tables' chains.
1344 bridge-nf-filter-vlan-tagged - BOOLEAN
1345 1 : pass bridged vlan-tagged ARP/IP/IPv6 traffic to {arp,ip,ip6}tables.
1349 bridge-nf-filter-pppoe-tagged - BOOLEAN
1350 1 : pass bridged pppoe-tagged IP/IPv6 traffic to {ip,ip6}tables.
1354 bridge-nf-pass-vlan-input-dev - BOOLEAN
1355 1: if bridge-nf-filter-vlan-tagged is enabled, try to find a vlan
1356 interface on the bridge and set the netfilter input device to the vlan.
1357 This allows use of e.g. "iptables -i br0.1" and makes the REDIRECT
1358 target work with vlan-on-top-of-bridge interfaces. When no matching
1359 vlan interface is found, or this switch is off, the input device is
1360 set to the bridge interface.
1361 0: disable bridge netfilter vlan interface lookup.
1364 proc/sys/net/sctp/* Variables:
1366 addip_enable - BOOLEAN
1367 Enable or disable extension of Dynamic Address Reconfiguration
1368 (ADD-IP) functionality specified in RFC5061. This extension provides
1369 the ability to dynamically add and remove new addresses for the SCTP
1372 1: Enable extension.
1374 0: Disable extension.
1378 addip_noauth_enable - BOOLEAN
1379 Dynamic Address Reconfiguration (ADD-IP) requires the use of
1380 authentication to protect the operations of adding or removing new
1381 addresses. This requirement is mandated so that unauthorized hosts
1382 would not be able to hijack associations. However, older
1383 implementations may not have implemented this requirement while
1384 allowing the ADD-IP extension. For reasons of interoperability,
1385 we provide this variable to control the enforcement of the
1386 authentication requirement.
1388 1: Allow ADD-IP extension to be used without authentication. This
1389 should only be set in a closed environment for interoperability
1390 with older implementations.
1392 0: Enforce the authentication requirement
1396 auth_enable - BOOLEAN
1397 Enable or disable Authenticated Chunks extension. This extension
1398 provides the ability to send and receive authenticated chunks and is
1399 required for secure operation of Dynamic Address Reconfiguration
1402 1: Enable this extension.
1403 0: Disable this extension.
1407 prsctp_enable - BOOLEAN
1408 Enable or disable the Partial Reliability extension (RFC3758) which
1409 is used to notify peers that a given DATA should no longer be expected.
1417 The limit of the number of new packets that can be initially sent. It
1418 controls how bursty the generated traffic can be.
1422 association_max_retrans - INTEGER
1423 Set the maximum number for retransmissions that an association can
1424 attempt deciding that the remote end is unreachable. If this value
1425 is exceeded, the association is terminated.
1429 max_init_retransmits - INTEGER
1430 The maximum number of retransmissions of INIT and COOKIE-ECHO chunks
1431 that an association will attempt before declaring the destination
1432 unreachable and terminating.
1436 path_max_retrans - INTEGER
1437 The maximum number of retransmissions that will be attempted on a given
1438 path. Once this threshold is exceeded, the path is considered
1439 unreachable, and new traffic will use a different path when the
1440 association is multihomed.
1444 pf_retrans - INTEGER
1445 The number of retransmissions that will be attempted on a given path
1446 before traffic is redirected to an alternate transport (should one
1447 exist). Note this is distinct from path_max_retrans, as a path that
1448 passes the pf_retrans threshold can still be used. Its only
1449 deprioritized when a transmission path is selected by the stack. This
1450 setting is primarily used to enable fast failover mechanisms without
1451 having to reduce path_max_retrans to a very low value. See:
1452 http://www.ietf.org/id/draft-nishida-tsvwg-sctp-failover-05.txt
1453 for details. Note also that a value of pf_retrans > path_max_retrans
1454 disables this feature
1458 rto_initial - INTEGER
1459 The initial round trip timeout value in milliseconds that will be used
1460 in calculating round trip times. This is the initial time interval
1461 for retransmissions.
1466 The maximum value (in milliseconds) of the round trip timeout. This
1467 is the largest time interval that can elapse between retransmissions.
1472 The minimum value (in milliseconds) of the round trip timeout. This
1473 is the smallest time interval the can elapse between retransmissions.
1477 hb_interval - INTEGER
1478 The interval (in milliseconds) between HEARTBEAT chunks. These chunks
1479 are sent at the specified interval on idle paths to probe the state of
1480 a given path between 2 associations.
1484 sack_timeout - INTEGER
1485 The amount of time (in milliseconds) that the implementation will wait
1490 valid_cookie_life - INTEGER
1491 The default lifetime of the SCTP cookie (in milliseconds). The cookie
1492 is used during association establishment.
1496 cookie_preserve_enable - BOOLEAN
1497 Enable or disable the ability to extend the lifetime of the SCTP cookie
1498 that is used during the establishment phase of SCTP association
1500 1: Enable cookie lifetime extension.
1505 cookie_hmac_alg - STRING
1506 Select the hmac algorithm used when generating the cookie value sent by
1507 a listening sctp socket to a connecting client in the INIT-ACK chunk.
1512 Ability to assign md5 or sha1 as the selected alg is predicated on the
1513 configuration of those algorithms at build time (CONFIG_CRYPTO_MD5 and
1514 CONFIG_CRYPTO_SHA1).
1516 Default: Dependent on configuration. MD5 if available, else SHA1 if
1517 available, else none.
1519 rcvbuf_policy - INTEGER
1520 Determines if the receive buffer is attributed to the socket or to
1521 association. SCTP supports the capability to create multiple
1522 associations on a single socket. When using this capability, it is
1523 possible that a single stalled association that's buffering a lot
1524 of data may block other associations from delivering their data by
1525 consuming all of the receive buffer space. To work around this,
1526 the rcvbuf_policy could be set to attribute the receiver buffer space
1527 to each association instead of the socket. This prevents the described
1530 1: rcvbuf space is per association
1531 0: rcvbuf space is per socket
1535 sndbuf_policy - INTEGER
1536 Similar to rcvbuf_policy above, this applies to send buffer space.
1538 1: Send buffer is tracked per association
1539 0: Send buffer is tracked per socket.
1543 sctp_mem - vector of 3 INTEGERs: min, pressure, max
1544 Number of pages allowed for queueing by all SCTP sockets.
1546 min: Below this number of pages SCTP is not bothered about its
1547 memory appetite. When amount of memory allocated by SCTP exceeds
1548 this number, SCTP starts to moderate memory usage.
1550 pressure: This value was introduced to follow format of tcp_mem.
1552 max: Number of pages allowed for queueing by all SCTP sockets.
1554 Default is calculated at boot time from amount of available memory.
1556 sctp_rmem - vector of 3 INTEGERs: min, default, max
1557 Only the first value ("min") is used, "default" and "max" are
1560 min: Minimal size of receive buffer used by SCTP socket.
1561 It is guaranteed to each SCTP socket (but not association) even
1562 under moderate memory pressure.
1566 sctp_wmem - vector of 3 INTEGERs: min, default, max
1567 Currently this tunable has no effect.
1569 addr_scope_policy - INTEGER
1570 Control IPv4 address scoping - draft-stewart-tsvwg-sctp-ipv4-00
1572 0 - Disable IPv4 address scoping
1573 1 - Enable IPv4 address scoping
1574 2 - Follow draft but allow IPv4 private addresses
1575 3 - Follow draft but allow IPv4 link local addresses
1580 /proc/sys/net/core/*
1581 Please see: Documentation/sysctl/net.txt for descriptions of these entries.
1584 /proc/sys/net/unix/*
1585 max_dgram_qlen - INTEGER
1586 The maximum length of dgram socket receive queue
1593 /proc/sys/net/irda/*
1594 fast_poll_increase FIXME
1595 warn_noreply_time FIXME
1596 discovery_slots FIXME
1599 discovery_timeout FIXME
1600 lap_keepalive_time FIXME
1601 max_noreply_time FIXME
1602 max_tx_data_size FIXME
1604 min_tx_turn_time FIXME