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. If enabled and a
20 fragmentation-required ICMP is received, the PMTU to this
21 destination will be set to min_pmtu (see below). You will need
22 to raise min_pmtu to the smallest interface MTU on your system
23 manually if you want to avoid locally generated fragments.
27 default 552 - minimum discovered Path MTU
29 route/max_size - INTEGER
30 Maximum number of routes allowed in the kernel. Increase
31 this when using large numbers of interfaces and/or routes.
33 neigh/default/gc_thresh1 - INTEGER
34 Minimum number of entries to keep. Garbage collector will not
35 purge entries if there are fewer than this number.
38 neigh/default/gc_thresh3 - INTEGER
39 Maximum number of neighbor entries allowed. Increase this
40 when using large numbers of interfaces and when communicating
41 with large numbers of directly-connected peers.
44 neigh/default/unres_qlen_bytes - INTEGER
45 The maximum number of bytes which may be used by packets
46 queued for each unresolved address by other network layers.
48 Setting negative value is meaningless and will return error.
49 Default: 65536 Bytes(64KB)
51 neigh/default/unres_qlen - INTEGER
52 The maximum number of packets which may be queued for each
53 unresolved address by other network layers.
54 (deprecated in linux 3.3) : use unres_qlen_bytes instead.
55 Prior to linux 3.3, the default value is 3 which may cause
56 unexpected packet loss. The current default value is calculated
57 according to default value of unres_qlen_bytes and true size of
62 Time, in seconds, that cached PMTU information is kept.
65 The advertised MSS depends on the first hop route MTU, but will
66 never be lower than this setting.
70 ipfrag_high_thresh - INTEGER
71 Maximum memory used to reassemble IP fragments. When
72 ipfrag_high_thresh bytes of memory is allocated for this purpose,
73 the fragment handler will toss packets until ipfrag_low_thresh
76 ipfrag_low_thresh - INTEGER
77 See ipfrag_high_thresh
80 Time in seconds to keep an IP fragment in memory.
82 ipfrag_secret_interval - INTEGER
83 Regeneration interval (in seconds) of the hash secret (or lifetime
84 for the hash secret) for IP fragments.
87 ipfrag_max_dist - INTEGER
88 ipfrag_max_dist is a non-negative integer value which defines the
89 maximum "disorder" which is allowed among fragments which share a
90 common IP source address. Note that reordering of packets is
91 not unusual, but if a large number of fragments arrive from a source
92 IP address while a particular fragment queue remains incomplete, it
93 probably indicates that one or more fragments belonging to that queue
94 have been lost. When ipfrag_max_dist is positive, an additional check
95 is done on fragments before they are added to a reassembly queue - if
96 ipfrag_max_dist (or more) fragments have arrived from a particular IP
97 address between additions to any IP fragment queue using that source
98 address, it's presumed that one or more fragments in the queue are
99 lost. The existing fragment queue will be dropped, and a new one
100 started. An ipfrag_max_dist value of zero disables this check.
102 Using a very small value, e.g. 1 or 2, for ipfrag_max_dist can
103 result in unnecessarily dropping fragment queues when normal
104 reordering of packets occurs, which could lead to poor application
105 performance. Using a very large value, e.g. 50000, increases the
106 likelihood of incorrectly reassembling IP fragments that originate
107 from different IP datagrams, which could result in data corruption.
112 inet_peer_threshold - INTEGER
113 The approximate size of the storage. Starting from this threshold
114 entries will be thrown aggressively. This threshold also determines
115 entries' time-to-live and time intervals between garbage collection
116 passes. More entries, less time-to-live, less GC interval.
118 inet_peer_minttl - INTEGER
119 Minimum time-to-live of entries. Should be enough to cover fragment
120 time-to-live on the reassembling side. This minimum time-to-live is
121 guaranteed if the pool size is less than inet_peer_threshold.
124 inet_peer_maxttl - INTEGER
125 Maximum time-to-live of entries. Unused entries will expire after
126 this period of time if there is no memory pressure on the pool (i.e.
127 when the number of entries in the pool is very small).
133 Limit of socket listen() backlog, known in userspace as SOMAXCONN.
134 Defaults to 128. See also tcp_max_syn_backlog for additional tuning
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),
149 Possible values are [-31, 31], inclusive.
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.
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,
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.
178 For passive connections, the listener congestion control choice
180 [see setsockopt(listenfd, SOL_TCP, TCP_CONGESTION, "name" ...) ]
183 Allows TCP to send "duplicate" SACKs.
185 tcp_early_retrans - INTEGER
186 Enable Early Retransmit (ER), per RFC 5827. ER lowers the threshold
187 for triggering fast retransmit when the amount of outstanding data is
188 small and when no previously unsent data can be transmitted (such
189 that limited transmit could be used). Also controls the use of
190 Tail loss probe (TLP) that converts RTOs occurring due to tail
191 losses into fast recovery (draft-dukkipati-tcpm-tcp-loss-probe-01).
195 2 enables ER but delays fast recovery and fast retransmit
196 by a fourth of RTT. This mitigates connection falsely
197 recovers when network has a small degree of reordering
198 (less than 3 packets).
199 3 enables delayed ER and TLP.
204 Control use of Explicit Congestion Notification (ECN) by TCP.
205 ECN is used only when both ends of the TCP connection indicate
206 support for it. This feature is useful in avoiding losses due
207 to congestion by allowing supporting routers to signal
208 congestion before having to drop packets.
210 0 Disable ECN. Neither initiate nor accept ECN.
211 1 Enable ECN when requested by incoming connections and
212 also request ECN on outgoing connection attempts.
213 2 Enable ECN when requested by incoming connections
214 but do not request ECN on outgoing connections.
218 Enable FACK congestion avoidance and fast retransmission.
219 The value is not used, if tcp_sack is not enabled.
221 tcp_fin_timeout - INTEGER
222 The length of time an orphaned (no longer referenced by any
223 application) connection will remain in the FIN_WAIT_2 state
224 before it is aborted at the local end. While a perfectly
225 valid "receive only" state for an un-orphaned connection, an
226 orphaned connection in FIN_WAIT_2 state could otherwise wait
227 forever for the remote to close its end of the connection.
232 Enables Forward RTO-Recovery (F-RTO) defined in RFC5682.
233 F-RTO is an enhanced recovery algorithm for TCP retransmission
234 timeouts. It is particularly beneficial in networks where the
235 RTT fluctuates (e.g., wireless). F-RTO is sender-side only
236 modification. It does not require any support from the peer.
238 By default it's enabled with a non-zero value. 0 disables F-RTO.
240 tcp_keepalive_time - INTEGER
241 How often TCP sends out keepalive messages when keepalive is enabled.
244 tcp_keepalive_probes - INTEGER
245 How many keepalive probes TCP sends out, until it decides that the
246 connection is broken. Default value: 9.
248 tcp_keepalive_intvl - INTEGER
249 How frequently the probes are send out. Multiplied by
250 tcp_keepalive_probes it is time to kill not responding connection,
251 after probes started. Default value: 75sec i.e. connection
252 will be aborted after ~11 minutes of retries.
254 tcp_low_latency - BOOLEAN
255 If set, the TCP stack makes decisions that prefer lower
256 latency as opposed to higher throughput. By default, this
257 option is not set meaning that higher throughput is preferred.
258 An example of an application where this default should be
259 changed would be a Beowulf compute cluster.
262 tcp_max_orphans - INTEGER
263 Maximal number of TCP sockets not attached to any user file handle,
264 held by system. If this number is exceeded orphaned connections are
265 reset immediately and warning is printed. This limit exists
266 only to prevent simple DoS attacks, you _must_ not rely on this
267 or lower the limit artificially, but rather increase it
268 (probably, after increasing installed memory),
269 if network conditions require more than default value,
270 and tune network services to linger and kill such states
271 more aggressively. Let me to remind again: each orphan eats
272 up to ~64K of unswappable memory.
274 tcp_max_syn_backlog - INTEGER
275 Maximal number of remembered connection requests, which have not
276 received an acknowledgment from connecting client.
277 The minimal value is 128 for low memory machines, and it will
278 increase in proportion to the memory of machine.
279 If server suffers from overload, try increasing this number.
281 tcp_max_tw_buckets - INTEGER
282 Maximal number of timewait sockets held by system simultaneously.
283 If this number is exceeded time-wait socket is immediately destroyed
284 and warning is printed. This limit exists only to prevent
285 simple DoS attacks, you _must_ not lower the limit artificially,
286 but rather increase it (probably, after increasing installed memory),
287 if network conditions require more than default value.
289 tcp_mem - vector of 3 INTEGERs: min, pressure, max
290 min: below this number of pages TCP is not bothered about its
293 pressure: when amount of memory allocated by TCP exceeds this number
294 of pages, TCP moderates its memory consumption and enters memory
295 pressure mode, which is exited when memory consumption falls
298 max: number of pages allowed for queueing by all TCP sockets.
300 Defaults are calculated at boot time from amount of available
303 tcp_moderate_rcvbuf - BOOLEAN
304 If set, TCP performs receive buffer auto-tuning, attempting to
305 automatically size the buffer (no greater than tcp_rmem[2]) to
306 match the size required by the path for full throughput. Enabled by
309 tcp_mtu_probing - INTEGER
310 Controls TCP Packetization-Layer Path MTU Discovery. Takes three
313 1 - Disabled by default, enabled when an ICMP black hole detected
314 2 - Always enabled, use initial MSS of tcp_base_mss.
316 tcp_no_metrics_save - BOOLEAN
317 By default, TCP saves various connection metrics in the route cache
318 when the connection closes, so that connections established in the
319 near future can use these to set initial conditions. Usually, this
320 increases overall performance, but may sometimes cause performance
321 degradation. If set, TCP will not cache metrics on closing
324 tcp_orphan_retries - INTEGER
325 This value influences the timeout of a locally closed TCP connection,
326 when RTO retransmissions remain unacknowledged.
327 See tcp_retries2 for more details.
329 The default value is 8.
330 If your machine is a loaded WEB server,
331 you should think about lowering this value, such sockets
332 may consume significant resources. Cf. tcp_max_orphans.
334 tcp_reordering - INTEGER
335 Maximal reordering of packets in a TCP stream.
338 tcp_retrans_collapse - BOOLEAN
339 Bug-to-bug compatibility with some broken printers.
340 On retransmit try to send bigger packets to work around bugs in
343 tcp_retries1 - INTEGER
344 This value influences the time, after which TCP decides, that
345 something is wrong due to unacknowledged RTO retransmissions,
346 and reports this suspicion to the network layer.
347 See tcp_retries2 for more details.
349 RFC 1122 recommends at least 3 retransmissions, which is the
352 tcp_retries2 - INTEGER
353 This value influences the timeout of an alive TCP connection,
354 when RTO retransmissions remain unacknowledged.
355 Given a value of N, a hypothetical TCP connection following
356 exponential backoff with an initial RTO of TCP_RTO_MIN would
357 retransmit N times before killing the connection at the (N+1)th RTO.
359 The default value of 15 yields a hypothetical timeout of 924.6
360 seconds and is a lower bound for the effective timeout.
361 TCP will effectively time out at the first RTO which exceeds the
362 hypothetical timeout.
364 RFC 1122 recommends at least 100 seconds for the timeout,
365 which corresponds to a value of at least 8.
367 tcp_rfc1337 - BOOLEAN
368 If set, the TCP stack behaves conforming to RFC1337. If unset,
369 we are not conforming to RFC, but prevent TCP TIME_WAIT
373 tcp_rmem - vector of 3 INTEGERs: min, default, max
374 min: Minimal size of receive buffer used by TCP sockets.
375 It is guaranteed to each TCP socket, even under moderate memory
379 default: initial size of receive buffer used by TCP sockets.
380 This value overrides net.core.rmem_default used by other protocols.
381 Default: 87380 bytes. This value results in window of 65535 with
382 default setting of tcp_adv_win_scale and tcp_app_win:0 and a bit
383 less for default tcp_app_win. See below about these variables.
385 max: maximal size of receive buffer allowed for automatically
386 selected receiver buffers for TCP socket. This value does not override
387 net.core.rmem_max. Calling setsockopt() with SO_RCVBUF disables
388 automatic tuning of that socket's receive buffer size, in which
389 case this value is ignored.
390 Default: between 87380B and 6MB, depending on RAM size.
393 Enable select acknowledgments (SACKS).
395 tcp_slow_start_after_idle - BOOLEAN
396 If set, provide RFC2861 behavior and time out the congestion
397 window after an idle period. An idle period is defined at
398 the current RTO. If unset, the congestion window will not
399 be timed out after an idle period.
403 Use the Host requirements interpretation of the TCP urgent pointer field.
404 Most hosts use the older BSD interpretation, so if you turn this on
405 Linux might not communicate correctly with them.
408 tcp_synack_retries - INTEGER
409 Number of times SYNACKs for a passive TCP connection attempt will
410 be retransmitted. Should not be higher than 255. Default value
411 is 5, which corresponds to 31seconds till the last retransmission
412 with the current initial RTO of 1second. With this the final timeout
413 for a passive TCP connection will happen after 63seconds.
415 tcp_syncookies - BOOLEAN
416 Only valid when the kernel was compiled with CONFIG_SYN_COOKIES
417 Send out syncookies when the syn backlog queue of a socket
418 overflows. This is to prevent against the common 'SYN flood attack'
421 Note, that syncookies is fallback facility.
422 It MUST NOT be used to help highly loaded servers to stand
423 against legal connection rate. If you see SYN flood warnings
424 in your logs, but investigation shows that they occur
425 because of overload with legal connections, you should tune
426 another parameters until this warning disappear.
427 See: tcp_max_syn_backlog, tcp_synack_retries, tcp_abort_on_overflow.
429 syncookies seriously violate TCP protocol, do not allow
430 to use TCP extensions, can result in serious degradation
431 of some services (f.e. SMTP relaying), visible not by you,
432 but your clients and relays, contacting you. While you see
433 SYN flood warnings in logs not being really flooded, your server
434 is seriously misconfigured.
436 If you want to test which effects syncookies have to your
437 network connections you can set this knob to 2 to enable
438 unconditionally generation of syncookies.
440 tcp_fastopen - INTEGER
441 Enable TCP Fast Open feature (draft-ietf-tcpm-fastopen) to send data
442 in the opening SYN packet. To use this feature, the client application
443 must use sendmsg() or sendto() with MSG_FASTOPEN flag rather than
444 connect() to perform a TCP handshake automatically.
446 The values (bitmap) are
447 1: Enables sending data in the opening SYN on the client w/ MSG_FASTOPEN.
448 2: Enables TCP Fast Open on the server side, i.e., allowing data in
449 a SYN packet to be accepted and passed to the application before
450 3-way hand shake finishes.
451 4: Send data in the opening SYN regardless of cookie availability and
452 without a cookie option.
453 0x100: Accept SYN data w/o validating the cookie.
454 0x200: Accept data-in-SYN w/o any cookie option present.
455 0x400/0x800: Enable Fast Open on all listeners regardless of the
456 TCP_FASTOPEN socket option. The two different flags designate two
457 different ways of setting max_qlen without the TCP_FASTOPEN socket
462 Note that the client & server side Fast Open flags (1 and 2
463 respectively) must be also enabled before the rest of flags can take
466 See include/net/tcp.h and the code for more details.
468 tcp_syn_retries - INTEGER
469 Number of times initial SYNs for an active TCP connection attempt
470 will be retransmitted. Should not be higher than 255. Default value
471 is 6, which corresponds to 63seconds till the last retransmission
472 with the current initial RTO of 1second. With this the final timeout
473 for an active TCP connection attempt will happen after 127seconds.
475 tcp_timestamps - BOOLEAN
476 Enable timestamps as defined in RFC1323.
478 tcp_min_tso_segs - INTEGER
479 Minimal number of segments per TSO frame.
480 Since linux-3.12, TCP does an automatic sizing of TSO frames,
481 depending on flow rate, instead of filling 64Kbytes packets.
482 For specific usages, it's possible to force TCP to build big
483 TSO frames. Note that TCP stack might split too big TSO packets
484 if available window is too small.
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_notsent_lowat - UNSIGNED INTEGER
526 A TCP socket can control the amount of unsent bytes in its write queue,
527 thanks to TCP_NOTSENT_LOWAT socket option. poll()/select()/epoll()
528 reports POLLOUT events if the amount of unsent bytes is below a per
529 socket value, and if the write queue is not full. sendmsg() will
530 also not add new buffers if the limit is hit.
532 This global variable controls the amount of unsent data for
533 sockets not using TCP_NOTSENT_LOWAT. For these sockets, a change
534 to the global variable has immediate effect.
536 Default: UINT_MAX (0xFFFFFFFF)
538 tcp_workaround_signed_windows - BOOLEAN
539 If set, assume no receipt of a window scaling option means the
540 remote TCP is broken and treats the window as a signed quantity.
541 If unset, assume the remote TCP is not broken even if we do
542 not receive a window scaling option from them.
545 tcp_dma_copybreak - INTEGER
546 Lower limit, in bytes, of the size of socket reads that will be
547 offloaded to a DMA copy engine, if one is present in the system
548 and CONFIG_NET_DMA is enabled.
551 tcp_thin_linear_timeouts - BOOLEAN
552 Enable dynamic triggering of linear timeouts for thin streams.
553 If set, a check is performed upon retransmission by timeout to
554 determine if the stream is thin (less than 4 packets in flight).
555 As long as the stream is found to be thin, up to 6 linear
556 timeouts may be performed before exponential backoff mode is
557 initiated. This improves retransmission latency for
558 non-aggressive thin streams, often found to be time-dependent.
559 For more information on thin streams, see
560 Documentation/networking/tcp-thin.txt
563 tcp_thin_dupack - BOOLEAN
564 Enable dynamic triggering of retransmissions after one dupACK
565 for thin streams. If set, a check is performed upon reception
566 of a dupACK to determine if the stream is thin (less than 4
567 packets in flight). As long as the stream is found to be thin,
568 data is retransmitted on the first received dupACK. This
569 improves retransmission latency for non-aggressive thin
570 streams, often found to be time-dependent.
571 For more information on thin streams, see
572 Documentation/networking/tcp-thin.txt
575 tcp_limit_output_bytes - INTEGER
576 Controls TCP Small Queue limit per tcp socket.
577 TCP bulk sender tends to increase packets in flight until it
578 gets losses notifications. With SNDBUF autotuning, this can
579 result in a large amount of packets queued in qdisc/device
580 on the local machine, hurting latency of other flows, for
581 typical pfifo_fast qdiscs.
582 tcp_limit_output_bytes limits the number of bytes on qdisc
583 or device to reduce artificial RTT/cwnd and reduce bufferbloat.
586 tcp_challenge_ack_limit - INTEGER
587 Limits number of Challenge ACK sent per second, as recommended
588 in RFC 5961 (Improving TCP's Robustness to Blind In-Window Attacks)
593 udp_mem - vector of 3 INTEGERs: min, pressure, max
594 Number of pages allowed for queueing by all UDP sockets.
596 min: Below this number of pages UDP is not bothered about its
597 memory appetite. When amount of memory allocated by UDP exceeds
598 this number, UDP starts to moderate memory usage.
600 pressure: This value was introduced to follow format of tcp_mem.
602 max: Number of pages allowed for queueing by all UDP sockets.
604 Default is calculated at boot time from amount of available memory.
606 udp_rmem_min - INTEGER
607 Minimal size of receive buffer used by UDP sockets in moderation.
608 Each UDP socket is able to use the size for receiving data, even if
609 total pages of UDP sockets exceed udp_mem pressure. The unit is byte.
612 udp_wmem_min - INTEGER
613 Minimal size of send buffer used by UDP sockets in moderation.
614 Each UDP socket is able to use the size for sending data, even if
615 total pages of UDP sockets exceed udp_mem pressure. The unit is byte.
620 cipso_cache_enable - BOOLEAN
621 If set, enable additions to and lookups from the CIPSO label mapping
622 cache. If unset, additions are ignored and lookups always result in a
623 miss. However, regardless of the setting the cache is still
624 invalidated when required when means you can safely toggle this on and
625 off and the cache will always be "safe".
628 cipso_cache_bucket_size - INTEGER
629 The CIPSO label cache consists of a fixed size hash table with each
630 hash bucket containing a number of cache entries. This variable limits
631 the number of entries in each hash bucket; the larger the value the
632 more CIPSO label mappings that can be cached. When the number of
633 entries in a given hash bucket reaches this limit adding new entries
634 causes the oldest entry in the bucket to be removed to make room.
637 cipso_rbm_optfmt - BOOLEAN
638 Enable the "Optimized Tag 1 Format" as defined in section 3.4.2.6 of
639 the CIPSO draft specification (see Documentation/netlabel for details).
640 This means that when set the CIPSO tag will be padded with empty
641 categories in order to make the packet data 32-bit aligned.
644 cipso_rbm_structvalid - BOOLEAN
645 If set, do a very strict check of the CIPSO option when
646 ip_options_compile() is called. If unset, relax the checks done during
647 ip_options_compile(). Either way is "safe" as errors are caught else
648 where in the CIPSO processing code but setting this to 0 (False) should
649 result in less work (i.e. it should be faster) but could cause problems
650 with other implementations that require strict checking.
655 ip_local_port_range - 2 INTEGERS
656 Defines the local port range that is used by TCP and UDP to
657 choose the local port. The first number is the first, the
658 second the last local port number. The default values are
659 32768 and 61000 respectively.
661 ip_local_reserved_ports - list of comma separated ranges
662 Specify the ports which are reserved for known third-party
663 applications. These ports will not be used by automatic port
664 assignments (e.g. when calling connect() or bind() with port
665 number 0). Explicit port allocation behavior is unchanged.
667 The format used for both input and output is a comma separated
668 list of ranges (e.g. "1,2-4,10-10" for ports 1, 2, 3, 4 and
669 10). Writing to the file will clear all previously reserved
670 ports and update the current list with the one given in the
673 Note that ip_local_port_range and ip_local_reserved_ports
674 settings are independent and both are considered by the kernel
675 when determining which ports are available for automatic port
678 You can reserve ports which are not in the current
679 ip_local_port_range, e.g.:
681 $ cat /proc/sys/net/ipv4/ip_local_port_range
683 $ cat /proc/sys/net/ipv4/ip_local_reserved_ports
686 although this is redundant. However such a setting is useful
687 if later the port range is changed to a value that will
688 include the reserved ports.
692 ip_nonlocal_bind - BOOLEAN
693 If set, allows processes to bind() to non-local IP addresses,
694 which can be quite useful - but may break some applications.
698 If set non-zero, enables support for dynamic addresses.
699 If set to a non-zero value larger than 1, a kernel log
700 message will be printed when dynamic address rewriting
704 ip_early_demux - BOOLEAN
705 Optimize input packet processing down to one demux for
706 certain kinds of local sockets. Currently we only do this
707 for established TCP sockets.
709 It may add an additional cost for pure routing workloads that
710 reduces overall throughput, in such case you should disable it.
713 icmp_echo_ignore_all - BOOLEAN
714 If set non-zero, then the kernel will ignore all ICMP ECHO
718 icmp_echo_ignore_broadcasts - BOOLEAN
719 If set non-zero, then the kernel will ignore all ICMP ECHO and
720 TIMESTAMP requests sent to it via broadcast/multicast.
723 icmp_ratelimit - INTEGER
724 Limit the maximal rates for sending ICMP packets whose type matches
725 icmp_ratemask (see below) to specific targets.
726 0 to disable any limiting,
727 otherwise the minimal space between responses in milliseconds.
730 icmp_ratemask - INTEGER
731 Mask made of ICMP types for which rates are being limited.
732 Significant bits: IHGFEDCBA9876543210
733 Default mask: 0000001100000011000 (6168)
735 Bit definitions (see include/linux/icmp.h):
737 3 Destination Unreachable *
742 C Parameter Problem *
747 H Address Mask Request
750 * These are rate limited by default (see default mask above)
752 icmp_ignore_bogus_error_responses - BOOLEAN
753 Some routers violate RFC1122 by sending bogus responses to broadcast
754 frames. Such violations are normally logged via a kernel warning.
755 If this is set to TRUE, the kernel will not give such warnings, which
756 will avoid log file clutter.
759 icmp_errors_use_inbound_ifaddr - BOOLEAN
761 If zero, icmp error messages are sent with the primary address of
762 the exiting interface.
764 If non-zero, the message will be sent with the primary address of
765 the interface that received the packet that caused the icmp error.
766 This is the behaviour network many administrators will expect from
767 a router. And it can make debugging complicated network layouts
770 Note that if no primary address exists for the interface selected,
771 then the primary address of the first non-loopback interface that
772 has one will be used regardless of this setting.
776 igmp_max_memberships - INTEGER
777 Change the maximum number of multicast groups we can subscribe to.
780 Theoretical maximum value is bounded by having to send a membership
781 report in a single datagram (i.e. the report can't span multiple
782 datagrams, or risk confusing the switch and leaving groups you don't
785 The number of supported groups 'M' is bounded by the number of group
786 report entries you can fit into a single datagram of 65535 bytes.
788 M = 65536-sizeof (ip header)/(sizeof(Group record))
790 Group records are variable length, with a minimum of 12 bytes.
791 So net.ipv4.igmp_max_memberships should not be set higher than:
793 (65536-24) / 12 = 5459
795 The value 5459 assumes no IP header options, so in practice
796 this number may be lower.
798 conf/interface/* changes special settings per interface (where
799 "interface" is the name of your network interface)
801 conf/all/* is special, changes the settings for all interfaces
803 log_martians - BOOLEAN
804 Log packets with impossible addresses to kernel log.
805 log_martians for the interface will be enabled if at least one of
806 conf/{all,interface}/log_martians is set to TRUE,
807 it will be disabled otherwise
809 accept_redirects - BOOLEAN
810 Accept ICMP redirect messages.
811 accept_redirects for the interface will be enabled if:
812 - both conf/{all,interface}/accept_redirects are TRUE in the case
813 forwarding for the interface is enabled
815 - at least one of conf/{all,interface}/accept_redirects is TRUE in the
816 case forwarding for the interface is disabled
817 accept_redirects for the interface will be disabled otherwise
822 Enable IP forwarding on this interface.
824 mc_forwarding - BOOLEAN
825 Do multicast routing. The kernel needs to be compiled with CONFIG_MROUTE
826 and a multicast routing daemon is required.
827 conf/all/mc_forwarding must also be set to TRUE to enable multicast
828 routing for the interface
831 Integer value used to differentiate the devices by the medium they
832 are attached to. Two devices can have different id values when
833 the broadcast packets are received only on one of them.
834 The default value 0 means that the device is the only interface
835 to its medium, value of -1 means that medium is not known.
837 Currently, it is used to change the proxy_arp behavior:
838 the proxy_arp feature is enabled for packets forwarded between
839 two devices attached to different media.
843 proxy_arp for the interface will be enabled if at least one of
844 conf/{all,interface}/proxy_arp is set to TRUE,
845 it will be disabled otherwise
847 proxy_arp_pvlan - BOOLEAN
848 Private VLAN proxy arp.
849 Basically allow proxy arp replies back to the same interface
850 (from which the ARP request/solicitation was received).
852 This is done to support (ethernet) switch features, like RFC
853 3069, where the individual ports are NOT allowed to
854 communicate with each other, but they are allowed to talk to
855 the upstream router. As described in RFC 3069, it is possible
856 to allow these hosts to communicate through the upstream
857 router by proxy_arp'ing. Don't need to be used together with
860 This technology is known by different names:
861 In RFC 3069 it is called VLAN Aggregation.
862 Cisco and Allied Telesyn call it Private VLAN.
863 Hewlett-Packard call it Source-Port filtering or port-isolation.
864 Ericsson call it MAC-Forced Forwarding (RFC Draft).
866 shared_media - BOOLEAN
867 Send(router) or accept(host) RFC1620 shared media redirects.
868 Overrides ip_secure_redirects.
869 shared_media for the interface will be enabled if at least one of
870 conf/{all,interface}/shared_media is set to TRUE,
871 it will be disabled otherwise
874 secure_redirects - BOOLEAN
875 Accept ICMP redirect messages only for gateways,
876 listed in default gateway list.
877 secure_redirects for the interface will be enabled if at least one of
878 conf/{all,interface}/secure_redirects is set to TRUE,
879 it will be disabled otherwise
882 send_redirects - BOOLEAN
883 Send redirects, if router.
884 send_redirects for the interface will be enabled if at least one of
885 conf/{all,interface}/send_redirects is set to TRUE,
886 it will be disabled otherwise
889 bootp_relay - BOOLEAN
890 Accept packets with source address 0.b.c.d destined
891 not to this host as local ones. It is supposed, that
892 BOOTP relay daemon will catch and forward such packets.
893 conf/all/bootp_relay must also be set to TRUE to enable BOOTP relay
898 accept_source_route - BOOLEAN
899 Accept packets with SRR option.
900 conf/all/accept_source_route must also be set to TRUE to accept packets
901 with SRR option on the interface
902 default TRUE (router)
905 accept_local - BOOLEAN
906 Accept packets with local source addresses. In combination
907 with suitable routing, this can be used to direct packets
908 between two local interfaces over the wire and have them
911 rp_filter must be set to a non-zero value in order for
912 accept_local to have an effect.
916 route_localnet - BOOLEAN
917 Do not consider loopback addresses as martian source or destination
918 while routing. This enables the use of 127/8 for local routing purposes.
922 0 - No source validation.
923 1 - Strict mode as defined in RFC3704 Strict Reverse Path
924 Each incoming packet is tested against the FIB and if the interface
925 is not the best reverse path the packet check will fail.
926 By default failed packets are discarded.
927 2 - Loose mode as defined in RFC3704 Loose Reverse Path
928 Each incoming packet's source address is also tested against the FIB
929 and if the source address is not reachable via any interface
930 the packet check will fail.
932 Current recommended practice in RFC3704 is to enable strict mode
933 to prevent IP spoofing from DDos attacks. If using asymmetric routing
934 or other complicated routing, then loose mode is recommended.
936 The max value from conf/{all,interface}/rp_filter is used
937 when doing source validation on the {interface}.
939 Default value is 0. Note that some distributions enable it
943 1 - Allows you to have multiple network interfaces on the same
944 subnet, and have the ARPs for each interface be answered
945 based on whether or not the kernel would route a packet from
946 the ARP'd IP out that interface (therefore you must use source
947 based routing for this to work). In other words it allows control
948 of which cards (usually 1) will respond to an arp request.
950 0 - (default) The kernel can respond to arp requests with addresses
951 from other interfaces. This may seem wrong but it usually makes
952 sense, because it increases the chance of successful communication.
953 IP addresses are owned by the complete host on Linux, not by
954 particular interfaces. Only for more complex setups like load-
955 balancing, does this behaviour cause problems.
957 arp_filter for the interface will be enabled if at least one of
958 conf/{all,interface}/arp_filter is set to TRUE,
959 it will be disabled otherwise
961 arp_announce - INTEGER
962 Define different restriction levels for announcing the local
963 source IP address from IP packets in ARP requests sent on
965 0 - (default) Use any local address, configured on any interface
966 1 - Try to avoid local addresses that are not in the target's
967 subnet for this interface. This mode is useful when target
968 hosts reachable via this interface require the source IP
969 address in ARP requests to be part of their logical network
970 configured on the receiving interface. When we generate the
971 request we will check all our subnets that include the
972 target IP and will preserve the source address if it is from
973 such subnet. If there is no such subnet we select source
974 address according to the rules for level 2.
975 2 - Always use the best local address for this target.
976 In this mode we ignore the source address in the IP packet
977 and try to select local address that we prefer for talks with
978 the target host. Such local address is selected by looking
979 for primary IP addresses on all our subnets on the outgoing
980 interface that include the target IP address. If no suitable
981 local address is found we select the first local address
982 we have on the outgoing interface or on all other interfaces,
983 with the hope we will receive reply for our request and
984 even sometimes no matter the source IP address we announce.
986 The max value from conf/{all,interface}/arp_announce is used.
988 Increasing the restriction level gives more chance for
989 receiving answer from the resolved target while decreasing
990 the level announces more valid sender's information.
993 Define different modes for sending replies in response to
994 received ARP requests that resolve local target IP addresses:
995 0 - (default): reply for any local target IP address, configured
997 1 - reply only if the target IP address is local address
998 configured on the incoming interface
999 2 - reply only if the target IP address is local address
1000 configured on the incoming interface and both with the
1001 sender's IP address are part from same subnet on this interface
1002 3 - do not reply for local addresses configured with scope host,
1003 only resolutions for global and link addresses are replied
1005 8 - do not reply for all local addresses
1007 The max value from conf/{all,interface}/arp_ignore is used
1008 when ARP request is received on the {interface}
1010 arp_notify - BOOLEAN
1011 Define mode for notification of address and device changes.
1012 0 - (default): do nothing
1013 1 - Generate gratuitous arp requests when device is brought up
1014 or hardware address changes.
1016 arp_accept - BOOLEAN
1017 Define behavior for gratuitous ARP frames who's IP is not
1018 already present in the ARP table:
1019 0 - don't create new entries in the ARP table
1020 1 - create new entries in the ARP table
1022 Both replies and requests type gratuitous arp will trigger the
1023 ARP table to be updated, if this setting is on.
1025 If the ARP table already contains the IP address of the
1026 gratuitous arp frame, the arp table will be updated regardless
1027 if this setting is on or off.
1030 app_solicit - INTEGER
1031 The maximum number of probes to send to the user space ARP daemon
1032 via netlink before dropping back to multicast probes (see
1033 mcast_solicit). Defaults to 0.
1035 disable_policy - BOOLEAN
1036 Disable IPSEC policy (SPD) for this interface
1038 disable_xfrm - BOOLEAN
1039 Disable IPSEC encryption on this interface, whatever the policy
1041 igmpv2_unsolicited_report_interval - INTEGER
1042 The interval in milliseconds in which the next unsolicited
1043 IGMPv1 or IGMPv2 report retransmit will take place.
1044 Default: 10000 (10 seconds)
1046 igmpv3_unsolicited_report_interval - INTEGER
1047 The interval in milliseconds in which the next unsolicited
1048 IGMPv3 report retransmit will take place.
1049 Default: 1000 (1 seconds)
1052 Allows you to write a number, which can be used as required.
1056 kuznet@ms2.inr.ac.ru
1062 delon.nicolas@wanadoo.fr
1067 /proc/sys/net/ipv6/* Variables:
1069 IPv6 has no global variables such as tcp_*. tcp_* settings under ipv4/ also
1070 apply to IPv6 [XXX?].
1072 bindv6only - BOOLEAN
1073 Default value for IPV6_V6ONLY socket option,
1074 which restricts use of the IPv6 socket to IPv6 communication
1076 TRUE: disable IPv4-mapped address feature
1077 FALSE: enable IPv4-mapped address feature
1079 Default: FALSE (as specified in RFC3493)
1083 ip6frag_high_thresh - INTEGER
1084 Maximum memory used to reassemble IPv6 fragments. When
1085 ip6frag_high_thresh bytes of memory is allocated for this purpose,
1086 the fragment handler will toss packets until ip6frag_low_thresh
1089 ip6frag_low_thresh - INTEGER
1090 See ip6frag_high_thresh
1092 ip6frag_time - INTEGER
1093 Time in seconds to keep an IPv6 fragment in memory.
1095 ip6frag_secret_interval - INTEGER
1096 Regeneration interval (in seconds) of the hash secret (or lifetime
1097 for the hash secret) for IPv6 fragments.
1101 Change the interface-specific default settings.
1105 Change all the interface-specific settings.
1107 [XXX: Other special features than forwarding?]
1109 conf/all/forwarding - BOOLEAN
1110 Enable global IPv6 forwarding between all interfaces.
1112 IPv4 and IPv6 work differently here; e.g. netfilter must be used
1113 to control which interfaces may forward packets and which not.
1115 This also sets all interfaces' Host/Router setting
1116 'forwarding' to the specified value. See below for details.
1118 This referred to as global forwarding.
1124 Change special settings per interface.
1126 The functional behaviour for certain settings is different
1127 depending on whether local forwarding is enabled or not.
1130 Accept Router Advertisements; autoconfigure using them.
1132 It also determines whether or not to transmit Router
1133 Solicitations. If and only if the functional setting is to
1134 accept Router Advertisements, Router Solicitations will be
1137 Possible values are:
1138 0 Do not accept Router Advertisements.
1139 1 Accept Router Advertisements if forwarding is disabled.
1140 2 Overrule forwarding behaviour. Accept Router Advertisements
1141 even if forwarding is enabled.
1143 Functional default: enabled if local forwarding is disabled.
1144 disabled if local forwarding is enabled.
1146 accept_ra_defrtr - BOOLEAN
1147 Learn default router in Router Advertisement.
1149 Functional default: enabled if accept_ra is enabled.
1150 disabled if accept_ra is disabled.
1152 accept_ra_pinfo - BOOLEAN
1153 Learn Prefix Information in Router Advertisement.
1155 Functional default: enabled if accept_ra is enabled.
1156 disabled if accept_ra is disabled.
1158 accept_ra_rt_info_max_plen - INTEGER
1159 Maximum prefix length of Route Information in RA.
1161 Route Information w/ prefix larger than or equal to this
1162 variable shall be ignored.
1164 Functional default: 0 if accept_ra_rtr_pref is enabled.
1165 -1 if accept_ra_rtr_pref is disabled.
1167 accept_ra_rtr_pref - BOOLEAN
1168 Accept Router Preference in RA.
1170 Functional default: enabled if accept_ra is enabled.
1171 disabled if accept_ra is disabled.
1173 accept_redirects - BOOLEAN
1176 Functional default: enabled if local forwarding is disabled.
1177 disabled if local forwarding is enabled.
1179 accept_source_route - INTEGER
1180 Accept source routing (routing extension header).
1182 >= 0: Accept only routing header type 2.
1183 < 0: Do not accept routing header.
1188 Autoconfigure addresses using Prefix Information in Router
1191 Functional default: enabled if accept_ra_pinfo is enabled.
1192 disabled if accept_ra_pinfo is disabled.
1194 dad_transmits - INTEGER
1195 The amount of Duplicate Address Detection probes to send.
1198 forwarding - INTEGER
1199 Configure interface-specific Host/Router behaviour.
1201 Note: It is recommended to have the same setting on all
1202 interfaces; mixed router/host scenarios are rather uncommon.
1204 Possible values are:
1205 0 Forwarding disabled
1206 1 Forwarding enabled
1210 By default, Host behaviour is assumed. This means:
1212 1. IsRouter flag is not set in Neighbour Advertisements.
1213 2. If accept_ra is TRUE (default), transmit Router
1215 3. If accept_ra is TRUE (default), accept Router
1216 Advertisements (and do autoconfiguration).
1217 4. If accept_redirects is TRUE (default), accept Redirects.
1221 If local forwarding is enabled, Router behaviour is assumed.
1222 This means exactly the reverse from the above:
1224 1. IsRouter flag is set in Neighbour Advertisements.
1225 2. Router Solicitations are not sent unless accept_ra is 2.
1226 3. Router Advertisements are ignored unless accept_ra is 2.
1227 4. Redirects are ignored.
1229 Default: 0 (disabled) if global forwarding is disabled (default),
1230 otherwise 1 (enabled).
1233 Default Hop Limit to set.
1237 Default Maximum Transfer Unit
1238 Default: 1280 (IPv6 required minimum)
1240 router_probe_interval - INTEGER
1241 Minimum interval (in seconds) between Router Probing described
1246 router_solicitation_delay - INTEGER
1247 Number of seconds to wait after interface is brought up
1248 before sending Router Solicitations.
1251 router_solicitation_interval - INTEGER
1252 Number of seconds to wait between Router Solicitations.
1255 router_solicitations - INTEGER
1256 Number of Router Solicitations to send until assuming no
1257 routers are present.
1260 use_tempaddr - INTEGER
1261 Preference for Privacy Extensions (RFC3041).
1262 <= 0 : disable Privacy Extensions
1263 == 1 : enable Privacy Extensions, but prefer public
1264 addresses over temporary addresses.
1265 > 1 : enable Privacy Extensions and prefer temporary
1266 addresses over public addresses.
1267 Default: 0 (for most devices)
1268 -1 (for point-to-point devices and loopback devices)
1270 temp_valid_lft - INTEGER
1271 valid lifetime (in seconds) for temporary addresses.
1272 Default: 604800 (7 days)
1274 temp_prefered_lft - INTEGER
1275 Preferred lifetime (in seconds) for temporary addresses.
1276 Default: 86400 (1 day)
1278 max_desync_factor - INTEGER
1279 Maximum value for DESYNC_FACTOR, which is a random value
1280 that ensures that clients don't synchronize with each
1281 other and generate new addresses at exactly the same time.
1282 value is in seconds.
1285 regen_max_retry - INTEGER
1286 Number of attempts before give up attempting to generate
1287 valid temporary addresses.
1290 max_addresses - INTEGER
1291 Maximum number of autoconfigured addresses per interface. Setting
1292 to zero disables the limitation. It is not recommended to set this
1293 value too large (or to zero) because it would be an easy way to
1294 crash the kernel by allowing too many addresses to be created.
1297 disable_ipv6 - BOOLEAN
1298 Disable IPv6 operation. If accept_dad is set to 2, this value
1299 will be dynamically set to TRUE if DAD fails for the link-local
1301 Default: FALSE (enable IPv6 operation)
1303 When this value is changed from 1 to 0 (IPv6 is being enabled),
1304 it will dynamically create a link-local address on the given
1305 interface and start Duplicate Address Detection, if necessary.
1307 When this value is changed from 0 to 1 (IPv6 is being disabled),
1308 it will dynamically delete all address on the given interface.
1310 accept_dad - INTEGER
1311 Whether to accept DAD (Duplicate Address Detection).
1313 1: Enable DAD (default)
1314 2: Enable DAD, and disable IPv6 operation if MAC-based duplicate
1315 link-local address has been found.
1317 force_tllao - BOOLEAN
1318 Enable sending the target link-layer address option even when
1319 responding to a unicast neighbor solicitation.
1322 Quoting from RFC 2461, section 4.4, Target link-layer address:
1324 "The option MUST be included for multicast solicitations in order to
1325 avoid infinite Neighbor Solicitation "recursion" when the peer node
1326 does not have a cache entry to return a Neighbor Advertisements
1327 message. When responding to unicast solicitations, the option can be
1328 omitted since the sender of the solicitation has the correct link-
1329 layer address; otherwise it would not have be able to send the unicast
1330 solicitation in the first place. However, including the link-layer
1331 address in this case adds little overhead and eliminates a potential
1332 race condition where the sender deletes the cached link-layer address
1333 prior to receiving a response to a previous solicitation."
1335 ndisc_notify - BOOLEAN
1336 Define mode for notification of address and device changes.
1337 0 - (default): do nothing
1338 1 - Generate unsolicited neighbour advertisements when device is brought
1339 up or hardware address changes.
1341 mldv1_unsolicited_report_interval - INTEGER
1342 The interval in milliseconds in which the next unsolicited
1343 MLDv1 report retransmit will take place.
1344 Default: 10000 (10 seconds)
1346 mldv2_unsolicited_report_interval - INTEGER
1347 The interval in milliseconds in which the next unsolicited
1348 MLDv2 report retransmit will take place.
1349 Default: 1000 (1 second)
1351 force_mld_version - INTEGER
1352 0 - (default) No enforcement of a MLD version, MLDv1 fallback allowed
1353 1 - Enforce to use MLD version 1
1354 2 - Enforce to use MLD version 2
1356 suppress_frag_ndisc - INTEGER
1357 Control RFC 6980 (Security Implications of IPv6 Fragmentation
1358 with IPv6 Neighbor Discovery) behavior:
1359 1 - (default) discard fragmented neighbor discovery packets
1360 0 - allow fragmented neighbor discovery packets
1364 Limit the maximal rates for sending ICMPv6 packets.
1365 0 to disable any limiting,
1366 otherwise the minimal space between responses in milliseconds.
1371 Pekka Savola <pekkas@netcore.fi>
1372 YOSHIFUJI Hideaki / USAGI Project <yoshfuji@linux-ipv6.org>
1375 /proc/sys/net/bridge/* Variables:
1377 bridge-nf-call-arptables - BOOLEAN
1378 1 : pass bridged ARP traffic to arptables' FORWARD chain.
1382 bridge-nf-call-iptables - BOOLEAN
1383 1 : pass bridged IPv4 traffic to iptables' chains.
1387 bridge-nf-call-ip6tables - BOOLEAN
1388 1 : pass bridged IPv6 traffic to ip6tables' chains.
1392 bridge-nf-filter-vlan-tagged - BOOLEAN
1393 1 : pass bridged vlan-tagged ARP/IP/IPv6 traffic to {arp,ip,ip6}tables.
1397 bridge-nf-filter-pppoe-tagged - BOOLEAN
1398 1 : pass bridged pppoe-tagged IP/IPv6 traffic to {ip,ip6}tables.
1402 bridge-nf-pass-vlan-input-dev - BOOLEAN
1403 1: if bridge-nf-filter-vlan-tagged is enabled, try to find a vlan
1404 interface on the bridge and set the netfilter input device to the vlan.
1405 This allows use of e.g. "iptables -i br0.1" and makes the REDIRECT
1406 target work with vlan-on-top-of-bridge interfaces. When no matching
1407 vlan interface is found, or this switch is off, the input device is
1408 set to the bridge interface.
1409 0: disable bridge netfilter vlan interface lookup.
1412 proc/sys/net/sctp/* Variables:
1414 addip_enable - BOOLEAN
1415 Enable or disable extension of Dynamic Address Reconfiguration
1416 (ADD-IP) functionality specified in RFC5061. This extension provides
1417 the ability to dynamically add and remove new addresses for the SCTP
1420 1: Enable extension.
1422 0: Disable extension.
1426 addip_noauth_enable - BOOLEAN
1427 Dynamic Address Reconfiguration (ADD-IP) requires the use of
1428 authentication to protect the operations of adding or removing new
1429 addresses. This requirement is mandated so that unauthorized hosts
1430 would not be able to hijack associations. However, older
1431 implementations may not have implemented this requirement while
1432 allowing the ADD-IP extension. For reasons of interoperability,
1433 we provide this variable to control the enforcement of the
1434 authentication requirement.
1436 1: Allow ADD-IP extension to be used without authentication. This
1437 should only be set in a closed environment for interoperability
1438 with older implementations.
1440 0: Enforce the authentication requirement
1444 auth_enable - BOOLEAN
1445 Enable or disable Authenticated Chunks extension. This extension
1446 provides the ability to send and receive authenticated chunks and is
1447 required for secure operation of Dynamic Address Reconfiguration
1450 1: Enable this extension.
1451 0: Disable this extension.
1455 prsctp_enable - BOOLEAN
1456 Enable or disable the Partial Reliability extension (RFC3758) which
1457 is used to notify peers that a given DATA should no longer be expected.
1465 The limit of the number of new packets that can be initially sent. It
1466 controls how bursty the generated traffic can be.
1470 association_max_retrans - INTEGER
1471 Set the maximum number for retransmissions that an association can
1472 attempt deciding that the remote end is unreachable. If this value
1473 is exceeded, the association is terminated.
1477 max_init_retransmits - INTEGER
1478 The maximum number of retransmissions of INIT and COOKIE-ECHO chunks
1479 that an association will attempt before declaring the destination
1480 unreachable and terminating.
1484 path_max_retrans - INTEGER
1485 The maximum number of retransmissions that will be attempted on a given
1486 path. Once this threshold is exceeded, the path is considered
1487 unreachable, and new traffic will use a different path when the
1488 association is multihomed.
1492 pf_retrans - INTEGER
1493 The number of retransmissions that will be attempted on a given path
1494 before traffic is redirected to an alternate transport (should one
1495 exist). Note this is distinct from path_max_retrans, as a path that
1496 passes the pf_retrans threshold can still be used. Its only
1497 deprioritized when a transmission path is selected by the stack. This
1498 setting is primarily used to enable fast failover mechanisms without
1499 having to reduce path_max_retrans to a very low value. See:
1500 http://www.ietf.org/id/draft-nishida-tsvwg-sctp-failover-05.txt
1501 for details. Note also that a value of pf_retrans > path_max_retrans
1502 disables this feature
1506 rto_initial - INTEGER
1507 The initial round trip timeout value in milliseconds that will be used
1508 in calculating round trip times. This is the initial time interval
1509 for retransmissions.
1514 The maximum value (in milliseconds) of the round trip timeout. This
1515 is the largest time interval that can elapse between retransmissions.
1520 The minimum value (in milliseconds) of the round trip timeout. This
1521 is the smallest time interval the can elapse between retransmissions.
1525 hb_interval - INTEGER
1526 The interval (in milliseconds) between HEARTBEAT chunks. These chunks
1527 are sent at the specified interval on idle paths to probe the state of
1528 a given path between 2 associations.
1532 sack_timeout - INTEGER
1533 The amount of time (in milliseconds) that the implementation will wait
1538 valid_cookie_life - INTEGER
1539 The default lifetime of the SCTP cookie (in milliseconds). The cookie
1540 is used during association establishment.
1544 cookie_preserve_enable - BOOLEAN
1545 Enable or disable the ability to extend the lifetime of the SCTP cookie
1546 that is used during the establishment phase of SCTP association
1548 1: Enable cookie lifetime extension.
1553 cookie_hmac_alg - STRING
1554 Select the hmac algorithm used when generating the cookie value sent by
1555 a listening sctp socket to a connecting client in the INIT-ACK chunk.
1560 Ability to assign md5 or sha1 as the selected alg is predicated on the
1561 configuration of those algorithms at build time (CONFIG_CRYPTO_MD5 and
1562 CONFIG_CRYPTO_SHA1).
1564 Default: Dependent on configuration. MD5 if available, else SHA1 if
1565 available, else none.
1567 rcvbuf_policy - INTEGER
1568 Determines if the receive buffer is attributed to the socket or to
1569 association. SCTP supports the capability to create multiple
1570 associations on a single socket. When using this capability, it is
1571 possible that a single stalled association that's buffering a lot
1572 of data may block other associations from delivering their data by
1573 consuming all of the receive buffer space. To work around this,
1574 the rcvbuf_policy could be set to attribute the receiver buffer space
1575 to each association instead of the socket. This prevents the described
1578 1: rcvbuf space is per association
1579 0: rcvbuf space is per socket
1583 sndbuf_policy - INTEGER
1584 Similar to rcvbuf_policy above, this applies to send buffer space.
1586 1: Send buffer is tracked per association
1587 0: Send buffer is tracked per socket.
1591 sctp_mem - vector of 3 INTEGERs: min, pressure, max
1592 Number of pages allowed for queueing by all SCTP sockets.
1594 min: Below this number of pages SCTP is not bothered about its
1595 memory appetite. When amount of memory allocated by SCTP exceeds
1596 this number, SCTP starts to moderate memory usage.
1598 pressure: This value was introduced to follow format of tcp_mem.
1600 max: Number of pages allowed for queueing by all SCTP sockets.
1602 Default is calculated at boot time from amount of available memory.
1604 sctp_rmem - vector of 3 INTEGERs: min, default, max
1605 Only the first value ("min") is used, "default" and "max" are
1608 min: Minimal size of receive buffer used by SCTP socket.
1609 It is guaranteed to each SCTP socket (but not association) even
1610 under moderate memory pressure.
1614 sctp_wmem - vector of 3 INTEGERs: min, default, max
1615 Currently this tunable has no effect.
1617 addr_scope_policy - INTEGER
1618 Control IPv4 address scoping - draft-stewart-tsvwg-sctp-ipv4-00
1620 0 - Disable IPv4 address scoping
1621 1 - Enable IPv4 address scoping
1622 2 - Follow draft but allow IPv4 private addresses
1623 3 - Follow draft but allow IPv4 link local addresses
1628 /proc/sys/net/core/*
1629 Please see: Documentation/sysctl/net.txt for descriptions of these entries.
1632 /proc/sys/net/unix/*
1633 max_dgram_qlen - INTEGER
1634 The maximum length of dgram socket receive queue
1641 /proc/sys/net/irda/*
1642 fast_poll_increase FIXME
1643 warn_noreply_time FIXME
1644 discovery_slots FIXME
1647 discovery_timeout FIXME
1648 lap_keepalive_time FIXME
1649 max_noreply_time FIXME
1650 max_tx_data_size FIXME
1652 min_tx_turn_time FIXME