| blob | a0f6842368c37b77b571c5817029adaf45db1769 |
1 /proc/sys/net/ipv4/* Variables:
3 ip_forward - BOOLEAN
4 0 - disabled (default)
5 not 0 - enabled
7 Forward Packets between interfaces.
9 This variable is special, its change resets all configuration
10 parameters to their default state (RFC1122 for hosts, RFC1812
11 for routers)
13 ip_default_ttl - INTEGER
14 default 64
16 ip_no_pmtu_disc - BOOLEAN
17 Disable Path MTU Discovery.
18 default FALSE
20 min_pmtu - INTEGER
21 default 562 - minimum discovered Path MTU
23 mtu_expires - INTEGER
24 Time, in seconds, that cached PMTU information is kept.
26 min_adv_mss - INTEGER
27 The advertised MSS depends on the first hop route MTU, but will
28 never be lower than this setting.
30 IP Fragmentation:
32 ipfrag_high_thresh - INTEGER
33 Maximum memory used to reassemble IP fragments. When
34 ipfrag_high_thresh bytes of memory is allocated for this purpose,
35 the fragment handler will toss packets until ipfrag_low_thresh
36 is reached.
38 ipfrag_low_thresh - INTEGER
39 See ipfrag_high_thresh
41 ipfrag_time - INTEGER
42 Time in seconds to keep an IP fragment in memory.
44 ipfrag_secret_interval - INTEGER
45 Regeneration interval (in seconds) of the hash secret (or lifetime
46 for the hash secret) for IP fragments.
47 Default: 600
49 ipfrag_max_dist - INTEGER
50 ipfrag_max_dist is a non-negative integer value which defines the
51 maximum "disorder" which is allowed among fragments which share a
52 common IP source address. Note that reordering of packets is
53 not unusual, but if a large number of fragments arrive from a source
54 IP address while a particular fragment queue remains incomplete, it
55 probably indicates that one or more fragments belonging to that queue
56 have been lost. When ipfrag_max_dist is positive, an additional check
57 is done on fragments before they are added to a reassembly queue - if
58 ipfrag_max_dist (or more) fragments have arrived from a particular IP
59 address between additions to any IP fragment queue using that source
60 address, it's presumed that one or more fragments in the queue are
61 lost. The existing fragment queue will be dropped, and a new one
62 started. An ipfrag_max_dist value of zero disables this check.
64 Using a very small value, e.g. 1 or 2, for ipfrag_max_dist can
65 result in unnecessarily dropping fragment queues when normal
66 reordering of packets occurs, which could lead to poor application
67 performance. Using a very large value, e.g. 50000, increases the
68 likelihood of incorrectly reassembling IP fragments that originate
69 from different IP datagrams, which could result in data corruption.
70 Default: 64
72 INET peer storage:
74 inet_peer_threshold - INTEGER
75 The approximate size of the storage. Starting from this threshold
76 entries will be thrown aggressively. This threshold also determines
77 entries' time-to-live and time intervals between garbage collection
78 passes. More entries, less time-to-live, less GC interval.
80 inet_peer_minttl - INTEGER
81 Minimum time-to-live of entries. Should be enough to cover fragment
82 time-to-live on the reassembling side. This minimum time-to-live is
83 guaranteed if the pool size is less than inet_peer_threshold.
84 Measured in jiffies(1).
86 inet_peer_maxttl - INTEGER
87 Maximum time-to-live of entries. Unused entries will expire after
88 this period of time if there is no memory pressure on the pool (i.e.
89 when the number of entries in the pool is very small).
90 Measured in jiffies(1).
92 inet_peer_gc_mintime - INTEGER
93 Minimum interval between garbage collection passes. This interval is
94 in effect under high memory pressure on the pool.
95 Measured in jiffies(1).
97 inet_peer_gc_maxtime - INTEGER
98 Minimum interval between garbage collection passes. This interval is
99 in effect under low (or absent) memory pressure on the pool.
100 Measured in jiffies(1).
102 TCP variables:
104 somaxconn - INTEGER
105 Limit of socket listen() backlog, known in userspace as SOMAXCONN.
106 Defaults to 128. See also tcp_max_syn_backlog for additional tuning
107 for TCP sockets.
109 tcp_abc - INTEGER
110 Controls Appropriate Byte Count (ABC) defined in RFC3465.
111 ABC is a way of increasing congestion window (cwnd) more slowly
112 in response to partial acknowledgments.
113 Possible values are:
114 0 increase cwnd once per acknowledgment (no ABC)
115 1 increase cwnd once per acknowledgment of full sized segment
116 2 allow increase cwnd by two if acknowledgment is
117 of two segments to compensate for delayed acknowledgments.
118 Default: 0 (off)
120 tcp_abort_on_overflow - BOOLEAN
121 If listening service is too slow to accept new connections,
122 reset them. Default state is FALSE. It means that if overflow
123 occurred due to a burst, connection will recover. Enable this
124 option _only_ if you are really sure that listening daemon
125 cannot be tuned to accept connections faster. Enabling this
126 option can harm clients of your server.
128 tcp_adv_win_scale - INTEGER
129 Count buffering overhead as bytes/2^tcp_adv_win_scale
130 (if tcp_adv_win_scale > 0) or bytes-bytes/2^(-tcp_adv_win_scale),
131 if it is <= 0.
132 Default: 2
134 tcp_allowed_congestion_control - STRING
135 Show/set the congestion control choices available to non-privileged
136 processes. The list is a subset of those listed in
137 tcp_available_congestion_control.
138 Default is "reno" and the default setting (tcp_congestion_control).
140 tcp_app_win - INTEGER
141 Reserve max(window/2^tcp_app_win, mss) of window for application
142 buffer. Value 0 is special, it means that nothing is reserved.
143 Default: 31
145 tcp_available_congestion_control - STRING
146 Shows the available congestion control choices that are registered.
147 More congestion control algorithms may be available as modules,
148 but not loaded.
150 tcp_congestion_control - STRING
151 Set the congestion control algorithm to be used for new
152 connections. The algorithm "reno" is always available, but
153 additional choices may be available based on kernel configuration.
154 Default is set as part of kernel configuration.
156 tcp_dsack - BOOLEAN
157 Allows TCP to send "duplicate" SACKs.
159 tcp_ecn - BOOLEAN
160 Enable Explicit Congestion Notification in TCP.
162 tcp_fack - BOOLEAN
163 Enable FACK congestion avoidance and fast retransmission.
164 The value is not used, if tcp_sack is not enabled.
166 tcp_fin_timeout - INTEGER
167 Time to hold socket in state FIN-WAIT-2, if it was closed
168 by our side. Peer can be broken and never close its side,
169 or even died unexpectedly. Default value is 60sec.
170 Usual value used in 2.2 was 180 seconds, you may restore
171 it, but remember that if your machine is even underloaded WEB server,
172 you risk to overflow memory with kilotons of dead sockets,
173 FIN-WAIT-2 sockets are less dangerous than FIN-WAIT-1,
174 because they eat maximum 1.5K of memory, but they tend
175 to live longer. Cf. tcp_max_orphans.
177 tcp_frto - BOOLEAN
178 Enables F-RTO, an enhanced recovery algorithm for TCP retransmission
179 timeouts. It is particularly beneficial in wireless environments
180 where packet loss is typically due to random radio interference
181 rather than intermediate router congestion.
183 tcp_keepalive_time - INTEGER
184 How often TCP sends out keepalive messages when keepalive is enabled.
185 Default: 2hours.
187 tcp_keepalive_probes - INTEGER
188 How many keepalive probes TCP sends out, until it decides that the
189 connection is broken. Default value: 9.
191 tcp_keepalive_intvl - INTEGER
192 How frequently the probes are send out. Multiplied by
193 tcp_keepalive_probes it is time to kill not responding connection,
194 after probes started. Default value: 75sec i.e. connection
195 will be aborted after ~11 minutes of retries.
197 tcp_low_latency - BOOLEAN
198 If set, the TCP stack makes decisions that prefer lower
199 latency as opposed to higher throughput. By default, this
200 option is not set meaning that higher throughput is preferred.
201 An example of an application where this default should be
202 changed would be a Beowulf compute cluster.
203 Default: 0
205 tcp_max_orphans - INTEGER
206 Maximal number of TCP sockets not attached to any user file handle,
207 held by system. If this number is exceeded orphaned connections are
208 reset immediately and warning is printed. This limit exists
209 only to prevent simple DoS attacks, you _must_ not rely on this
210 or lower the limit artificially, but rather increase it
211 (probably, after increasing installed memory),
212 if network conditions require more than default value,
213 and tune network services to linger and kill such states
214 more aggressively. Let me to remind again: each orphan eats
215 up to ~64K of unswappable memory.
217 tcp_max_syn_backlog - INTEGER
218 Maximal number of remembered connection requests, which are
219 still did not receive an acknowledgment from connecting client.
220 Default value is 1024 for systems with more than 128Mb of memory,
221 and 128 for low memory machines. If server suffers of overload,
222 try to increase this number.
224 tcp_max_tw_buckets - INTEGER
225 Maximal number of timewait sockets held by system simultaneously.
226 If this number is exceeded time-wait socket is immediately destroyed
227 and warning is printed. This limit exists only to prevent
228 simple DoS attacks, you _must_ not lower the limit artificially,
229 but rather increase it (probably, after increasing installed memory),
230 if network conditions require more than default value.
232 tcp_mem - vector of 3 INTEGERs: min, pressure, max
233 min: below this number of pages TCP is not bothered about its
234 memory appetite.
236 pressure: when amount of memory allocated by TCP exceeds this number
237 of pages, TCP moderates its memory consumption and enters memory
238 pressure mode, which is exited when memory consumption falls
239 under "min".
241 max: number of pages allowed for queueing by all TCP sockets.
243 Defaults are calculated at boot time from amount of available
244 memory.
246 tcp_orphan_retries - INTEGER
247 How may times to retry before killing TCP connection, closed
248 by our side. Default value 7 corresponds to ~50sec-16min
249 depending on RTO. If you machine is loaded WEB server,
250 you should think about lowering this value, such sockets
251 may consume significant resources. Cf. tcp_max_orphans.
253 tcp_reordering - INTEGER
254 Maximal reordering of packets in a TCP stream.
255 Default: 3
257 tcp_retrans_collapse - BOOLEAN
258 Bug-to-bug compatibility with some broken printers.
259 On retransmit try to send bigger packets to work around bugs in
260 certain TCP stacks.
262 tcp_retries1 - INTEGER
263 How many times to retry before deciding that something is wrong
264 and it is necessary to report this suspicion to network layer.
265 Minimal RFC value is 3, it is default, which corresponds
266 to ~3sec-8min depending on RTO.
268 tcp_retries2 - INTEGER
269 How may times to retry before killing alive TCP connection.
270 RFC1122 says that the limit should be longer than 100 sec.
271 It is too small number. Default value 15 corresponds to ~13-30min
272 depending on RTO.
274 tcp_rfc1337 - BOOLEAN
275 If set, the TCP stack behaves conforming to RFC1337. If unset,
276 we are not conforming to RFC, but prevent TCP TIME_WAIT
277 assassination.
278 Default: 0
280 tcp_rmem - vector of 3 INTEGERs: min, default, max
281 min: Minimal size of receive buffer used by TCP sockets.
282 It is guaranteed to each TCP socket, even under moderate memory
283 pressure.
284 Default: 8K
286 default: default size of receive buffer used by TCP sockets.
287 This value overrides net.core.rmem_default used by other protocols.
288 Default: 87380 bytes. This value results in window of 65535 with
289 default setting of tcp_adv_win_scale and tcp_app_win:0 and a bit
290 less for default tcp_app_win. See below about these variables.
292 max: maximal size of receive buffer allowed for automatically
293 selected receiver buffers for TCP socket. This value does not override
294 net.core.rmem_max, "static" selection via SO_RCVBUF does not use this.
295 Default: 87380*2 bytes.
297 tcp_sack - BOOLEAN
298 Enable select acknowledgments (SACKS).
300 tcp_slow_start_after_idle - BOOLEAN
301 If set, provide RFC2861 behavior and time out the congestion
302 window after an idle period. An idle period is defined at
303 the current RTO. If unset, the congestion window will not
304 be timed out after an idle period.
305 Default: 1
307 tcp_stdurg - BOOLEAN
308 Use the Host requirements interpretation of the TCP urg pointer field.
309 Most hosts use the older BSD interpretation, so if you turn this on
310 Linux might not communicate correctly with them.
311 Default: FALSE
313 tcp_synack_retries - INTEGER
314 Number of times SYNACKs for a passive TCP connection attempt will
315 be retransmitted. Should not be higher than 255. Default value
316 is 5, which corresponds to ~180seconds.
318 tcp_syncookies - BOOLEAN
319 Only valid when the kernel was compiled with CONFIG_SYNCOOKIES
320 Send out syncookies when the syn backlog queue of a socket
321 overflows. This is to prevent against the common 'syn flood attack'
322 Default: FALSE
324 Note, that syncookies is fallback facility.
325 It MUST NOT be used to help highly loaded servers to stand
326 against legal connection rate. If you see synflood warnings
327 in your logs, but investigation shows that they occur
328 because of overload with legal connections, you should tune
329 another parameters until this warning disappear.
330 See: tcp_max_syn_backlog, tcp_synack_retries, tcp_abort_on_overflow.
332 syncookies seriously violate TCP protocol, do not allow
333 to use TCP extensions, can result in serious degradation
334 of some services (f.e. SMTP relaying), visible not by you,
335 but your clients and relays, contacting you. While you see
336 synflood warnings in logs not being really flooded, your server
337 is seriously misconfigured.
339 tcp_syn_retries - INTEGER
340 Number of times initial SYNs for an active TCP connection attempt
341 will be retransmitted. Should not be higher than 255. Default value
342 is 5, which corresponds to ~180seconds.
344 tcp_timestamps - BOOLEAN
345 Enable timestamps as defined in RFC1323.
347 tcp_tso_win_divisor - INTEGER
348 This allows control over what percentage of the congestion window
349 can be consumed by a single TSO frame.
350 The setting of this parameter is a choice between burstiness and
351 building larger TSO frames.
352 Default: 3
354 tcp_tw_recycle - BOOLEAN
355 Enable fast recycling TIME-WAIT sockets. Default value is 0.
356 It should not be changed without advice/request of technical
357 experts.
359 tcp_tw_reuse - BOOLEAN
360 Allow to reuse TIME-WAIT sockets for new connections when it is
361 safe from protocol viewpoint. Default value is 0.
362 It should not be changed without advice/request of technical
363 experts.
365 tcp_window_scaling - BOOLEAN
366 Enable window scaling as defined in RFC1323.
368 tcp_wmem - vector of 3 INTEGERs: min, default, max
369 min: Amount of memory reserved for send buffers for TCP socket.
370 Each TCP socket has rights to use it due to fact of its birth.
371 Default: 4K
373 default: Amount of memory allowed for send buffers for TCP socket
374 by default. This value overrides net.core.wmem_default used
375 by other protocols, it is usually lower than net.core.wmem_default.
376 Default: 16K
378 max: Maximal amount of memory allowed for automatically selected
379 send buffers for TCP socket. This value does not override
380 net.core.wmem_max, "static" selection via SO_SNDBUF does not use this.
381 Default: 128K
383 tcp_workaround_signed_windows - BOOLEAN
384 If set, assume no receipt of a window scaling option means the
385 remote TCP is broken and treats the window as a signed quantity.
386 If unset, assume the remote TCP is not broken even if we do
387 not receive a window scaling option from them.
388 Default: 0
390 CIPSOv4 Variables:
392 cipso_cache_enable - BOOLEAN
393 If set, enable additions to and lookups from the CIPSO label mapping
394 cache. If unset, additions are ignored and lookups always result in a
395 miss. However, regardless of the setting the cache is still
396 invalidated when required when means you can safely toggle this on and
397 off and the cache will always be "safe".
398 Default: 1
400 cipso_cache_bucket_size - INTEGER
401 The CIPSO label cache consists of a fixed size hash table with each
402 hash bucket containing a number of cache entries. This variable limits
403 the number of entries in each hash bucket; the larger the value the
404 more CIPSO label mappings that can be cached. When the number of
405 entries in a given hash bucket reaches this limit adding new entries
406 causes the oldest entry in the bucket to be removed to make room.
407 Default: 10
409 cipso_rbm_optfmt - BOOLEAN
410 Enable the "Optimized Tag 1 Format" as defined in section 3.4.2.6 of
411 the CIPSO draft specification (see Documentation/netlabel for details).
412 This means that when set the CIPSO tag will be padded with empty
413 categories in order to make the packet data 32-bit aligned.
414 Default: 0
416 cipso_rbm_structvalid - BOOLEAN
417 If set, do a very strict check of the CIPSO option when
418 ip_options_compile() is called. If unset, relax the checks done during
419 ip_options_compile(). Either way is "safe" as errors are caught else
420 where in the CIPSO processing code but setting this to 0 (False) should
421 result in less work (i.e. it should be faster) but could cause problems
422 with other implementations that require strict checking.
423 Default: 0
425 IP Variables:
427 ip_local_port_range - 2 INTEGERS
428 Defines the local port range that is used by TCP and UDP to
429 choose the local port. The first number is the first, the
430 second the last local port number. Default value depends on
431 amount of memory available on the system:
432 > 128Mb 32768-61000
433 < 128Mb 1024-4999 or even less.
434 This number defines number of active connections, which this
435 system can issue simultaneously to systems not supporting
436 TCP extensions (timestamps). With tcp_tw_recycle enabled
437 (i.e. by default) range 1024-4999 is enough to issue up to
438 2000 connections per second to systems supporting timestamps.
440 ip_nonlocal_bind - BOOLEAN
441 If set, allows processes to bind() to non-local IP addresses,
442 which can be quite useful - but may break some applications.
443 Default: 0
445 ip_dynaddr - BOOLEAN
446 If set non-zero, enables support for dynamic addresses.
447 If set to a non-zero value larger than 1, a kernel log
448 message will be printed when dynamic address rewriting
449 occurs.
450 Default: 0
452 icmp_echo_ignore_all - BOOLEAN
453 If set non-zero, then the kernel will ignore all ICMP ECHO
454 requests sent to it.
455 Default: 0
457 icmp_echo_ignore_broadcasts - BOOLEAN
458 If set non-zero, then the kernel will ignore all ICMP ECHO and
459 TIMESTAMP requests sent to it via broadcast/multicast.
460 Default: 1
462 icmp_ratelimit - INTEGER
463 Limit the maximal rates for sending ICMP packets whose type matches
464 icmp_ratemask (see below) to specific targets.
465 0 to disable any limiting, otherwise the maximal rate in jiffies(1)
466 Default: 100
468 icmp_ratemask - INTEGER
469 Mask made of ICMP types for which rates are being limited.
470 Significant bits: IHGFEDCBA9876543210
471 Default mask: 0000001100000011000 (6168)
473 Bit definitions (see include/linux/icmp.h):
474 0 Echo Reply
475 3 Destination Unreachable *
476 4 Source Quench *
477 5 Redirect
478 8 Echo Request
479 B Time Exceeded *
480 C Parameter Problem *
481 D Timestamp Request
482 E Timestamp Reply
483 F Info Request
484 G Info Reply
485 H Address Mask Request
486 I Address Mask Reply
488 * These are rate limited by default (see default mask above)
490 icmp_ignore_bogus_error_responses - BOOLEAN
491 Some routers violate RFC1122 by sending bogus responses to broadcast
492 frames. Such violations are normally logged via a kernel warning.
493 If this is set to TRUE, the kernel will not give such warnings, which
494 will avoid log file clutter.
495 Default: FALSE
497 icmp_errors_use_inbound_ifaddr - BOOLEAN
499 If zero, icmp error messages are sent with the primary address of
500 the exiting interface.
502 If non-zero, the message will be sent with the primary address of
503 the interface that received the packet that caused the icmp error.
504 This is the behaviour network many administrators will expect from
505 a router. And it can make debugging complicated network layouts
506 much easier.
508 Note that if no primary address exists for the interface selected,
509 then the primary address of the first non-loopback interface that
510 has one will be used regardless of this setting.
512 Default: 0
514 igmp_max_memberships - INTEGER
515 Change the maximum number of multicast groups we can subscribe to.
516 Default: 20
518 conf/interface/* changes special settings per interface (where "interface" is
519 the name of your network interface)
520 conf/all/* is special, changes the settings for all interfaces
523 log_martians - BOOLEAN
524 Log packets with impossible addresses to kernel log.
525 log_martians for the interface will be enabled if at least one of
526 conf/{all,interface}/log_martians is set to TRUE,
527 it will be disabled otherwise
529 accept_redirects - BOOLEAN
530 Accept ICMP redirect messages.
531 accept_redirects for the interface will be enabled if:
532 - both conf/{all,interface}/accept_redirects are TRUE in the case forwarding
533 for the interface is enabled
534 or
535 - at least one of conf/{all,interface}/accept_redirects is TRUE in the case
536 forwarding for the interface is disabled
537 accept_redirects for the interface will be disabled otherwise
538 default TRUE (host)
539 FALSE (router)
541 forwarding - BOOLEAN
542 Enable IP forwarding on this interface.
544 mc_forwarding - BOOLEAN
545 Do multicast routing. The kernel needs to be compiled with CONFIG_MROUTE
546 and a multicast routing daemon is required.
547 conf/all/mc_forwarding must also be set to TRUE to enable multicast routing
548 for the interface
550 medium_id - INTEGER
551 Integer value used to differentiate the devices by the medium they
552 are attached to. Two devices can have different id values when
553 the broadcast packets are received only on one of them.
554 The default value 0 means that the device is the only interface
555 to its medium, value of -1 means that medium is not known.
557 Currently, it is used to change the proxy_arp behavior:
558 the proxy_arp feature is enabled for packets forwarded between
559 two devices attached to different media.
561 proxy_arp - BOOLEAN
562 Do proxy arp.
563 proxy_arp for the interface will be enabled if at least one of
564 conf/{all,interface}/proxy_arp is set to TRUE,
565 it will be disabled otherwise
567 shared_media - BOOLEAN
568 Send(router) or accept(host) RFC1620 shared media redirects.
569 Overrides ip_secure_redirects.
570 shared_media for the interface will be enabled if at least one of
571 conf/{all,interface}/shared_media is set to TRUE,
572 it will be disabled otherwise
573 default TRUE
575 secure_redirects - BOOLEAN
576 Accept ICMP redirect messages only for gateways,
577 listed in default gateway list.
578 secure_redirects for the interface will be enabled if at least one of
579 conf/{all,interface}/secure_redirects is set to TRUE,
580 it will be disabled otherwise
581 default TRUE
583 send_redirects - BOOLEAN
584 Send redirects, if router.
585 send_redirects for the interface will be enabled if at least one of
586 conf/{all,interface}/send_redirects is set to TRUE,
587 it will be disabled otherwise
588 Default: TRUE
590 bootp_relay - BOOLEAN
591 Accept packets with source address 0.b.c.d destined
592 not to this host as local ones. It is supposed, that
593 BOOTP relay daemon will catch and forward such packets.
594 conf/all/bootp_relay must also be set to TRUE to enable BOOTP relay
595 for the interface
596 default FALSE
597 Not Implemented Yet.
599 accept_source_route - BOOLEAN
600 Accept packets with SRR option.
601 conf/all/accept_source_route must also be set to TRUE to accept packets
602 with SRR option on the interface
603 default TRUE (router)
604 FALSE (host)
606 rp_filter - BOOLEAN
607 1 - do source validation by reversed path, as specified in RFC1812
608 Recommended option for single homed hosts and stub network
609 routers. Could cause troubles for complicated (not loop free)
610 networks running a slow unreliable protocol (sort of RIP),
611 or using static routes.
613 0 - No source validation.
615 conf/all/rp_filter must also be set to TRUE to do source validation
616 on the interface
618 Default value is 0. Note that some distributions enable it
619 in startup scripts.
621 arp_filter - BOOLEAN
622 1 - Allows you to have multiple network interfaces on the same
623 subnet, and have the ARPs for each interface be answered
624 based on whether or not the kernel would route a packet from
625 the ARP'd IP out that interface (therefore you must use source
626 based routing for this to work). In other words it allows control
627 of which cards (usually 1) will respond to an arp request.
629 0 - (default) The kernel can respond to arp requests with addresses
630 from other interfaces. This may seem wrong but it usually makes
631 sense, because it increases the chance of successful communication.
632 IP addresses are owned by the complete host on Linux, not by
633 particular interfaces. Only for more complex setups like load-
634 balancing, does this behaviour cause problems.
636 arp_filter for the interface will be enabled if at least one of
637 conf/{all,interface}/arp_filter is set to TRUE,
638 it will be disabled otherwise
640 arp_announce - INTEGER
641 Define different restriction levels for announcing the local
642 source IP address from IP packets in ARP requests sent on
643 interface:
644 0 - (default) Use any local address, configured on any interface
645 1 - Try to avoid local addresses that are not in the target's
646 subnet for this interface. This mode is useful when target
647 hosts reachable via this interface require the source IP
648 address in ARP requests to be part of their logical network
649 configured on the receiving interface. When we generate the
650 request we will check all our subnets that include the
651 target IP and will preserve the source address if it is from
652 such subnet. If there is no such subnet we select source
653 address according to the rules for level 2.
654 2 - Always use the best local address for this target.
655 In this mode we ignore the source address in the IP packet
656 and try to select local address that we prefer for talks with
657 the target host. Such local address is selected by looking
658 for primary IP addresses on all our subnets on the outgoing
659 interface that include the target IP address. If no suitable
660 local address is found we select the first local address
661 we have on the outgoing interface or on all other interfaces,
662 with the hope we will receive reply for our request and
663 even sometimes no matter the source IP address we announce.
665 The max value from conf/{all,interface}/arp_announce is used.
667 Increasing the restriction level gives more chance for
668 receiving answer from the resolved target while decreasing
669 the level announces more valid sender's information.
671 arp_ignore - INTEGER
672 Define different modes for sending replies in response to
673 received ARP requests that resolve local target IP addresses:
674 0 - (default): reply for any local target IP address, configured
675 on any interface
676 1 - reply only if the target IP address is local address
677 configured on the incoming interface
678 2 - reply only if the target IP address is local address
679 configured on the incoming interface and both with the
680 sender's IP address are part from same subnet on this interface
681 3 - do not reply for local addresses configured with scope host,
682 only resolutions for global and link addresses are replied
683 4-7 - reserved
684 8 - do not reply for all local addresses
686 The max value from conf/{all,interface}/arp_ignore is used
687 when ARP request is received on the {interface}
689 arp_accept - BOOLEAN
690 Define behavior when gratuitous arp replies are received:
691 0 - drop gratuitous arp frames
692 1 - accept gratuitous arp frames
694 app_solicit - INTEGER
695 The maximum number of probes to send to the user space ARP daemon
696 via netlink before dropping back to multicast probes (see
697 mcast_solicit). Defaults to 0.
699 disable_policy - BOOLEAN
700 Disable IPSEC policy (SPD) for this interface
702 disable_xfrm - BOOLEAN
703 Disable IPSEC encryption on this interface, whatever the policy
707 tag - INTEGER
708 Allows you to write a number, which can be used as required.
709 Default value is 0.
711 (1) Jiffie: internal timeunit for the kernel. On the i386 1/100s, on the
712 Alpha 1/1024s. See the HZ define in /usr/include/asm/param.h for the exact
713 value on your system.
715 Alexey Kuznetsov.
716 kuznet@ms2.inr.ac.ru
718 Updated by:
719 Andi Kleen
720 ak@muc.de
721 Nicolas Delon
722 delon.nicolas@wanadoo.fr
727 /proc/sys/net/ipv6/* Variables:
729 IPv6 has no global variables such as tcp_*. tcp_* settings under ipv4/ also
730 apply to IPv6 [XXX?].
732 bindv6only - BOOLEAN
733 Default value for IPV6_V6ONLY socket option,
734 which restricts use of the IPv6 socket to IPv6 communication
735 only.
736 TRUE: disable IPv4-mapped address feature
737 FALSE: enable IPv4-mapped address feature
739 Default: FALSE (as specified in RFC2553bis)
741 IPv6 Fragmentation:
743 ip6frag_high_thresh - INTEGER
744 Maximum memory used to reassemble IPv6 fragments. When
745 ip6frag_high_thresh bytes of memory is allocated for this purpose,
746 the fragment handler will toss packets until ip6frag_low_thresh
747 is reached.
749 ip6frag_low_thresh - INTEGER
750 See ip6frag_high_thresh
752 ip6frag_time - INTEGER
753 Time in seconds to keep an IPv6 fragment in memory.
755 ip6frag_secret_interval - INTEGER
756 Regeneration interval (in seconds) of the hash secret (or lifetime
757 for the hash secret) for IPv6 fragments.
758 Default: 600
760 conf/default/*:
761 Change the interface-specific default settings.
764 conf/all/*:
765 Change all the interface-specific settings.
767 [XXX: Other special features than forwarding?]
769 conf/all/forwarding - BOOLEAN
770 Enable global IPv6 forwarding between all interfaces.
772 IPv4 and IPv6 work differently here; e.g. netfilter must be used
773 to control which interfaces may forward packets and which not.
775 This also sets all interfaces' Host/Router setting
776 'forwarding' to the specified value. See below for details.
778 This referred to as global forwarding.
780 proxy_ndp - BOOLEAN
781 Do proxy ndp.
783 conf/interface/*:
784 Change special settings per interface.
786 The functional behaviour for certain settings is different
787 depending on whether local forwarding is enabled or not.
789 accept_ra - BOOLEAN
790 Accept Router Advertisements; autoconfigure using them.
792 Functional default: enabled if local forwarding is disabled.
793 disabled if local forwarding is enabled.
795 accept_ra_defrtr - BOOLEAN
796 Learn default router in Router Advertisement.
798 Functional default: enabled if accept_ra is enabled.
799 disabled if accept_ra is disabled.
801 accept_ra_pinfo - BOOLEAN
802 Learn Prefix Information in Router Advertisement.
804 Functional default: enabled if accept_ra is enabled.
805 disabled if accept_ra is disabled.
807 accept_ra_rt_info_max_plen - INTEGER
808 Maximum prefix length of Route Information in RA.
810 Route Information w/ prefix larger than or equal to this
811 variable shall be ignored.
813 Functional default: 0 if accept_ra_rtr_pref is enabled.
814 -1 if accept_ra_rtr_pref is disabled.
816 accept_ra_rtr_pref - BOOLEAN
817 Accept Router Preference in RA.
819 Functional default: enabled if accept_ra is enabled.
820 disabled if accept_ra is disabled.
822 accept_redirects - BOOLEAN
823 Accept Redirects.
825 Functional default: enabled if local forwarding is disabled.
826 disabled if local forwarding is enabled.
828 autoconf - BOOLEAN
829 Autoconfigure addresses using Prefix Information in Router
830 Advertisements.
832 Functional default: enabled if accept_ra_pinfo is enabled.
833 disabled if accept_ra_pinfo is disabled.
835 dad_transmits - INTEGER
836 The amount of Duplicate Address Detection probes to send.
837 Default: 1
839 forwarding - BOOLEAN
840 Configure interface-specific Host/Router behaviour.
842 Note: It is recommended to have the same setting on all
843 interfaces; mixed router/host scenarios are rather uncommon.
845 FALSE:
847 By default, Host behaviour is assumed. This means:
849 1. IsRouter flag is not set in Neighbour Advertisements.
850 2. Router Solicitations are being sent when necessary.
851 3. If accept_ra is TRUE (default), accept Router
852 Advertisements (and do autoconfiguration).
853 4. If accept_redirects is TRUE (default), accept Redirects.
855 TRUE:
857 If local forwarding is enabled, Router behaviour is assumed.
858 This means exactly the reverse from the above:
860 1. IsRouter flag is set in Neighbour Advertisements.
861 2. Router Solicitations are not sent.
862 3. Router Advertisements are ignored.
863 4. Redirects are ignored.
865 Default: FALSE if global forwarding is disabled (default),
866 otherwise TRUE.
868 hop_limit - INTEGER
869 Default Hop Limit to set.
870 Default: 64
872 mtu - INTEGER
873 Default Maximum Transfer Unit
874 Default: 1280 (IPv6 required minimum)
876 router_probe_interval - INTEGER
877 Minimum interval (in seconds) between Router Probing described
878 in RFC4191.
880 Default: 60
882 router_solicitation_delay - INTEGER
883 Number of seconds to wait after interface is brought up
884 before sending Router Solicitations.
885 Default: 1
887 router_solicitation_interval - INTEGER
888 Number of seconds to wait between Router Solicitations.
889 Default: 4
891 router_solicitations - INTEGER
892 Number of Router Solicitations to send until assuming no
893 routers are present.
894 Default: 3
896 use_tempaddr - INTEGER
897 Preference for Privacy Extensions (RFC3041).
898 <= 0 : disable Privacy Extensions
899 == 1 : enable Privacy Extensions, but prefer public
900 addresses over temporary addresses.
901 > 1 : enable Privacy Extensions and prefer temporary
902 addresses over public addresses.
903 Default: 0 (for most devices)
904 -1 (for point-to-point devices and loopback devices)
906 temp_valid_lft - INTEGER
907 valid lifetime (in seconds) for temporary addresses.
908 Default: 604800 (7 days)
910 temp_prefered_lft - INTEGER
911 Preferred lifetime (in seconds) for temporary addresses.
912 Default: 86400 (1 day)
914 max_desync_factor - INTEGER
915 Maximum value for DESYNC_FACTOR, which is a random value
916 that ensures that clients don't synchronize with each
917 other and generate new addresses at exactly the same time.
918 value is in seconds.
919 Default: 600
921 regen_max_retry - INTEGER
922 Number of attempts before give up attempting to generate
923 valid temporary addresses.
924 Default: 5
926 max_addresses - INTEGER
927 Number of maximum addresses per interface. 0 disables limitation.
928 It is recommended not set too large value (or 0) because it would
929 be too easy way to crash kernel to allow to create too much of
930 autoconfigured addresses.
931 Default: 16
933 icmp/*:
934 ratelimit - INTEGER
935 Limit the maximal rates for sending ICMPv6 packets.
936 0 to disable any limiting, otherwise the maximal rate in jiffies(1)
937 Default: 100
940 IPv6 Update by:
941 Pekka Savola <pekkas@netcore.fi>
942 YOSHIFUJI Hideaki / USAGI Project <yoshfuji@linux-ipv6.org>
945 /proc/sys/net/bridge/* Variables:
947 bridge-nf-call-arptables - BOOLEAN
948 1 : pass bridged ARP traffic to arptables' FORWARD chain.
949 0 : disable this.
950 Default: 1
952 bridge-nf-call-iptables - BOOLEAN
953 1 : pass bridged IPv4 traffic to iptables' chains.
954 0 : disable this.
955 Default: 1
957 bridge-nf-call-ip6tables - BOOLEAN
958 1 : pass bridged IPv6 traffic to ip6tables' chains.
959 0 : disable this.
960 Default: 1
962 bridge-nf-filter-vlan-tagged - BOOLEAN
963 1 : pass bridged vlan-tagged ARP/IP traffic to arptables/iptables.
964 0 : disable this.
965 Default: 1
968 UNDOCUMENTED:
970 dev_weight FIXME
971 discovery_slots FIXME
972 discovery_timeout FIXME
973 fast_poll_increase FIXME
974 ip6_queue_maxlen FIXME
975 lap_keepalive_time FIXME
976 lo_cong FIXME
977 max_baud_rate FIXME
978 max_dgram_qlen FIXME
979 max_noreply_time FIXME
980 max_tx_data_size FIXME
981 max_tx_window FIXME
982 min_tx_turn_time FIXME
983 mod_cong FIXME
984 no_cong FIXME
985 no_cong_thresh FIXME
986 slot_timeout FIXME
987 warn_noreply_time FIXME