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
16 ip_no_pmtu_disc - BOOLEAN
17 Disable Path MTU Discovery.
21 default 562 - minimum discovered Path MTU
24 Time, in seconds, that cached PMTU information is kept.
27 The advertised MSS depends on the first hop route MTU, but will
28 never be lower than this setting.
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
38 ipfrag_low_thresh - INTEGER
39 See ipfrag_high_thresh
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.
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.
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).
105 Limit of socket listen() backlog, known in userspace as SOMAXCONN.
106 Defaults to 128. See also tcp_max_syn_backlog for additional tuning
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.
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.
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),
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.
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,
150 tcp_base_mss - INTEGER
151 The initial value of search_low to be used by Packetization Layer
152 Path MTU Discovery (MTU probing). If MTU probing is enabled,
153 this is the inital MSS used by the connection.
155 tcp_congestion_control - STRING
156 Set the congestion control algorithm to be used for new
157 connections. The algorithm "reno" is always available, but
158 additional choices may be available based on kernel configuration.
159 Default is set as part of kernel configuration.
162 Allows TCP to send "duplicate" SACKs.
165 Enable Explicit Congestion Notification in TCP.
168 Enable FACK congestion avoidance and fast retransmission.
169 The value is not used, if tcp_sack is not enabled.
171 tcp_fin_timeout - INTEGER
172 Time to hold socket in state FIN-WAIT-2, if it was closed
173 by our side. Peer can be broken and never close its side,
174 or even died unexpectedly. Default value is 60sec.
175 Usual value used in 2.2 was 180 seconds, you may restore
176 it, but remember that if your machine is even underloaded WEB server,
177 you risk to overflow memory with kilotons of dead sockets,
178 FIN-WAIT-2 sockets are less dangerous than FIN-WAIT-1,
179 because they eat maximum 1.5K of memory, but they tend
180 to live longer. Cf. tcp_max_orphans.
183 Enables F-RTO, an enhanced recovery algorithm for TCP retransmission
184 timeouts. It is particularly beneficial in wireless environments
185 where packet loss is typically due to random radio interference
186 rather than intermediate router congestion. If set to 1, basic
187 version is enabled. 2 enables SACK enhanced FRTO, which is
188 EXPERIMENTAL. The basic version can be used also when SACK is
189 enabled for a flow through tcp_sack sysctl.
191 tcp_keepalive_time - INTEGER
192 How often TCP sends out keepalive messages when keepalive is enabled.
195 tcp_keepalive_probes - INTEGER
196 How many keepalive probes TCP sends out, until it decides that the
197 connection is broken. Default value: 9.
199 tcp_keepalive_intvl - INTEGER
200 How frequently the probes are send out. Multiplied by
201 tcp_keepalive_probes it is time to kill not responding connection,
202 after probes started. Default value: 75sec i.e. connection
203 will be aborted after ~11 minutes of retries.
205 tcp_low_latency - BOOLEAN
206 If set, the TCP stack makes decisions that prefer lower
207 latency as opposed to higher throughput. By default, this
208 option is not set meaning that higher throughput is preferred.
209 An example of an application where this default should be
210 changed would be a Beowulf compute cluster.
213 tcp_max_orphans - INTEGER
214 Maximal number of TCP sockets not attached to any user file handle,
215 held by system. If this number is exceeded orphaned connections are
216 reset immediately and warning is printed. This limit exists
217 only to prevent simple DoS attacks, you _must_ not rely on this
218 or lower the limit artificially, but rather increase it
219 (probably, after increasing installed memory),
220 if network conditions require more than default value,
221 and tune network services to linger and kill such states
222 more aggressively. Let me to remind again: each orphan eats
223 up to ~64K of unswappable memory.
225 tcp_max_syn_backlog - INTEGER
226 Maximal number of remembered connection requests, which are
227 still did not receive an acknowledgment from connecting client.
228 Default value is 1024 for systems with more than 128Mb of memory,
229 and 128 for low memory machines. If server suffers of overload,
230 try to increase this number.
232 tcp_max_tw_buckets - INTEGER
233 Maximal number of timewait sockets held by system simultaneously.
234 If this number is exceeded time-wait socket is immediately destroyed
235 and warning is printed. This limit exists only to prevent
236 simple DoS attacks, you _must_ not lower the limit artificially,
237 but rather increase it (probably, after increasing installed memory),
238 if network conditions require more than default value.
240 tcp_mem - vector of 3 INTEGERs: min, pressure, max
241 min: below this number of pages TCP is not bothered about its
244 pressure: when amount of memory allocated by TCP exceeds this number
245 of pages, TCP moderates its memory consumption and enters memory
246 pressure mode, which is exited when memory consumption falls
249 max: number of pages allowed for queueing by all TCP sockets.
251 Defaults are calculated at boot time from amount of available
254 tcp_moderate_rcvbuf - BOOLEAN
255 If set, TCP performs receive buffer autotuning, attempting to
256 automatically size the buffer (no greater than tcp_rmem[2]) to
257 match the size required by the path for full throughput. Enabled by
260 tcp_mtu_probing - INTEGER
261 Controls TCP Packetization-Layer Path MTU Discovery. Takes three
264 1 - Disabled by default, enabled when an ICMP black hole detected
265 2 - Always enabled, use initial MSS of tcp_base_mss.
267 tcp_no_metrics_save - BOOLEAN
268 By default, TCP saves various connection metrics in the route cache
269 when the connection closes, so that connections established in the
270 near future can use these to set initial conditions. Usually, this
271 increases overall performance, but may sometimes cause performance
272 degredation. If set, TCP will not cache metrics on closing
275 tcp_orphan_retries - INTEGER
276 How may times to retry before killing TCP connection, closed
277 by our side. Default value 7 corresponds to ~50sec-16min
278 depending on RTO. If you machine is loaded WEB server,
279 you should think about lowering this value, such sockets
280 may consume significant resources. Cf. tcp_max_orphans.
282 tcp_reordering - INTEGER
283 Maximal reordering of packets in a TCP stream.
286 tcp_retrans_collapse - BOOLEAN
287 Bug-to-bug compatibility with some broken printers.
288 On retransmit try to send bigger packets to work around bugs in
291 tcp_retries1 - INTEGER
292 How many times to retry before deciding that something is wrong
293 and it is necessary to report this suspicion to network layer.
294 Minimal RFC value is 3, it is default, which corresponds
295 to ~3sec-8min depending on RTO.
297 tcp_retries2 - INTEGER
298 How may times to retry before killing alive TCP connection.
299 RFC1122 says that the limit should be longer than 100 sec.
300 It is too small number. Default value 15 corresponds to ~13-30min
303 tcp_rfc1337 - BOOLEAN
304 If set, the TCP stack behaves conforming to RFC1337. If unset,
305 we are not conforming to RFC, but prevent TCP TIME_WAIT
309 tcp_rmem - vector of 3 INTEGERs: min, default, max
310 min: Minimal size of receive buffer used by TCP sockets.
311 It is guaranteed to each TCP socket, even under moderate memory
315 default: default size of receive buffer used by TCP sockets.
316 This value overrides net.core.rmem_default used by other protocols.
317 Default: 87380 bytes. This value results in window of 65535 with
318 default setting of tcp_adv_win_scale and tcp_app_win:0 and a bit
319 less for default tcp_app_win. See below about these variables.
321 max: maximal size of receive buffer allowed for automatically
322 selected receiver buffers for TCP socket. This value does not override
323 net.core.rmem_max, "static" selection via SO_RCVBUF does not use this.
324 Default: 87380*2 bytes.
327 Enable select acknowledgments (SACKS).
329 tcp_slow_start_after_idle - BOOLEAN
330 If set, provide RFC2861 behavior and time out the congestion
331 window after an idle period. An idle period is defined at
332 the current RTO. If unset, the congestion window will not
333 be timed out after an idle period.
337 Use the Host requirements interpretation of the TCP urg pointer field.
338 Most hosts use the older BSD interpretation, so if you turn this on
339 Linux might not communicate correctly with them.
342 tcp_synack_retries - INTEGER
343 Number of times SYNACKs for a passive TCP connection attempt will
344 be retransmitted. Should not be higher than 255. Default value
345 is 5, which corresponds to ~180seconds.
347 tcp_syncookies - BOOLEAN
348 Only valid when the kernel was compiled with CONFIG_SYNCOOKIES
349 Send out syncookies when the syn backlog queue of a socket
350 overflows. This is to prevent against the common 'syn flood attack'
353 Note, that syncookies is fallback facility.
354 It MUST NOT be used to help highly loaded servers to stand
355 against legal connection rate. If you see synflood warnings
356 in your logs, but investigation shows that they occur
357 because of overload with legal connections, you should tune
358 another parameters until this warning disappear.
359 See: tcp_max_syn_backlog, tcp_synack_retries, tcp_abort_on_overflow.
361 syncookies seriously violate TCP protocol, do not allow
362 to use TCP extensions, can result in serious degradation
363 of some services (f.e. SMTP relaying), visible not by you,
364 but your clients and relays, contacting you. While you see
365 synflood warnings in logs not being really flooded, your server
366 is seriously misconfigured.
368 tcp_syn_retries - INTEGER
369 Number of times initial SYNs for an active TCP connection attempt
370 will be retransmitted. Should not be higher than 255. Default value
371 is 5, which corresponds to ~180seconds.
373 tcp_timestamps - BOOLEAN
374 Enable timestamps as defined in RFC1323.
376 tcp_tso_win_divisor - INTEGER
377 This allows control over what percentage of the congestion window
378 can be consumed by a single TSO frame.
379 The setting of this parameter is a choice between burstiness and
380 building larger TSO frames.
383 tcp_tw_recycle - BOOLEAN
384 Enable fast recycling TIME-WAIT sockets. Default value is 0.
385 It should not be changed without advice/request of technical
388 tcp_tw_reuse - BOOLEAN
389 Allow to reuse TIME-WAIT sockets for new connections when it is
390 safe from protocol viewpoint. Default value is 0.
391 It should not be changed without advice/request of technical
394 tcp_window_scaling - BOOLEAN
395 Enable window scaling as defined in RFC1323.
397 tcp_wmem - vector of 3 INTEGERs: min, default, max
398 min: Amount of memory reserved for send buffers for TCP socket.
399 Each TCP socket has rights to use it due to fact of its birth.
402 default: Amount of memory allowed for send buffers for TCP socket
403 by default. This value overrides net.core.wmem_default used
404 by other protocols, it is usually lower than net.core.wmem_default.
407 max: Maximal amount of memory allowed for automatically selected
408 send buffers for TCP socket. This value does not override
409 net.core.wmem_max, "static" selection via SO_SNDBUF does not use this.
412 tcp_workaround_signed_windows - BOOLEAN
413 If set, assume no receipt of a window scaling option means the
414 remote TCP is broken and treats the window as a signed quantity.
415 If unset, assume the remote TCP is not broken even if we do
416 not receive a window scaling option from them.
421 cipso_cache_enable - BOOLEAN
422 If set, enable additions to and lookups from the CIPSO label mapping
423 cache. If unset, additions are ignored and lookups always result in a
424 miss. However, regardless of the setting the cache is still
425 invalidated when required when means you can safely toggle this on and
426 off and the cache will always be "safe".
429 cipso_cache_bucket_size - INTEGER
430 The CIPSO label cache consists of a fixed size hash table with each
431 hash bucket containing a number of cache entries. This variable limits
432 the number of entries in each hash bucket; the larger the value the
433 more CIPSO label mappings that can be cached. When the number of
434 entries in a given hash bucket reaches this limit adding new entries
435 causes the oldest entry in the bucket to be removed to make room.
438 cipso_rbm_optfmt - BOOLEAN
439 Enable the "Optimized Tag 1 Format" as defined in section 3.4.2.6 of
440 the CIPSO draft specification (see Documentation/netlabel for details).
441 This means that when set the CIPSO tag will be padded with empty
442 categories in order to make the packet data 32-bit aligned.
445 cipso_rbm_structvalid - BOOLEAN
446 If set, do a very strict check of the CIPSO option when
447 ip_options_compile() is called. If unset, relax the checks done during
448 ip_options_compile(). Either way is "safe" as errors are caught else
449 where in the CIPSO processing code but setting this to 0 (False) should
450 result in less work (i.e. it should be faster) but could cause problems
451 with other implementations that require strict checking.
456 ip_local_port_range - 2 INTEGERS
457 Defines the local port range that is used by TCP and UDP to
458 choose the local port. The first number is the first, the
459 second the last local port number. Default value depends on
460 amount of memory available on the system:
462 < 128Mb 1024-4999 or even less.
463 This number defines number of active connections, which this
464 system can issue simultaneously to systems not supporting
465 TCP extensions (timestamps). With tcp_tw_recycle enabled
466 (i.e. by default) range 1024-4999 is enough to issue up to
467 2000 connections per second to systems supporting timestamps.
469 ip_nonlocal_bind - BOOLEAN
470 If set, allows processes to bind() to non-local IP addresses,
471 which can be quite useful - but may break some applications.
475 If set non-zero, enables support for dynamic addresses.
476 If set to a non-zero value larger than 1, a kernel log
477 message will be printed when dynamic address rewriting
481 icmp_echo_ignore_all - BOOLEAN
482 If set non-zero, then the kernel will ignore all ICMP ECHO
486 icmp_echo_ignore_broadcasts - BOOLEAN
487 If set non-zero, then the kernel will ignore all ICMP ECHO and
488 TIMESTAMP requests sent to it via broadcast/multicast.
491 icmp_ratelimit - INTEGER
492 Limit the maximal rates for sending ICMP packets whose type matches
493 icmp_ratemask (see below) to specific targets.
494 0 to disable any limiting, otherwise the maximal rate in jiffies(1)
497 icmp_ratemask - INTEGER
498 Mask made of ICMP types for which rates are being limited.
499 Significant bits: IHGFEDCBA9876543210
500 Default mask: 0000001100000011000 (6168)
502 Bit definitions (see include/linux/icmp.h):
504 3 Destination Unreachable *
509 C Parameter Problem *
514 H Address Mask Request
517 * These are rate limited by default (see default mask above)
519 icmp_ignore_bogus_error_responses - BOOLEAN
520 Some routers violate RFC1122 by sending bogus responses to broadcast
521 frames. Such violations are normally logged via a kernel warning.
522 If this is set to TRUE, the kernel will not give such warnings, which
523 will avoid log file clutter.
526 icmp_errors_use_inbound_ifaddr - BOOLEAN
528 If zero, icmp error messages are sent with the primary address of
529 the exiting interface.
531 If non-zero, the message will be sent with the primary address of
532 the interface that received the packet that caused the icmp error.
533 This is the behaviour network many administrators will expect from
534 a router. And it can make debugging complicated network layouts
537 Note that if no primary address exists for the interface selected,
538 then the primary address of the first non-loopback interface that
539 has one will be used regardless of this setting.
543 igmp_max_memberships - INTEGER
544 Change the maximum number of multicast groups we can subscribe to.
547 conf/interface/* changes special settings per interface (where "interface" is
548 the name of your network interface)
549 conf/all/* is special, changes the settings for all interfaces
552 log_martians - BOOLEAN
553 Log packets with impossible addresses to kernel log.
554 log_martians for the interface will be enabled if at least one of
555 conf/{all,interface}/log_martians is set to TRUE,
556 it will be disabled otherwise
558 accept_redirects - BOOLEAN
559 Accept ICMP redirect messages.
560 accept_redirects for the interface will be enabled if:
561 - both conf/{all,interface}/accept_redirects are TRUE in the case forwarding
562 for the interface is enabled
564 - at least one of conf/{all,interface}/accept_redirects is TRUE in the case
565 forwarding for the interface is disabled
566 accept_redirects for the interface will be disabled otherwise
571 Enable IP forwarding on this interface.
573 mc_forwarding - BOOLEAN
574 Do multicast routing. The kernel needs to be compiled with CONFIG_MROUTE
575 and a multicast routing daemon is required.
576 conf/all/mc_forwarding must also be set to TRUE to enable multicast routing
580 Integer value used to differentiate the devices by the medium they
581 are attached to. Two devices can have different id values when
582 the broadcast packets are received only on one of them.
583 The default value 0 means that the device is the only interface
584 to its medium, value of -1 means that medium is not known.
586 Currently, it is used to change the proxy_arp behavior:
587 the proxy_arp feature is enabled for packets forwarded between
588 two devices attached to different media.
592 proxy_arp for the interface will be enabled if at least one of
593 conf/{all,interface}/proxy_arp is set to TRUE,
594 it will be disabled otherwise
596 shared_media - BOOLEAN
597 Send(router) or accept(host) RFC1620 shared media redirects.
598 Overrides ip_secure_redirects.
599 shared_media for the interface will be enabled if at least one of
600 conf/{all,interface}/shared_media is set to TRUE,
601 it will be disabled otherwise
604 secure_redirects - BOOLEAN
605 Accept ICMP redirect messages only for gateways,
606 listed in default gateway list.
607 secure_redirects for the interface will be enabled if at least one of
608 conf/{all,interface}/secure_redirects is set to TRUE,
609 it will be disabled otherwise
612 send_redirects - BOOLEAN
613 Send redirects, if router.
614 send_redirects for the interface will be enabled if at least one of
615 conf/{all,interface}/send_redirects is set to TRUE,
616 it will be disabled otherwise
619 bootp_relay - BOOLEAN
620 Accept packets with source address 0.b.c.d destined
621 not to this host as local ones. It is supposed, that
622 BOOTP relay daemon will catch and forward such packets.
623 conf/all/bootp_relay must also be set to TRUE to enable BOOTP relay
628 accept_source_route - BOOLEAN
629 Accept packets with SRR option.
630 conf/all/accept_source_route must also be set to TRUE to accept packets
631 with SRR option on the interface
632 default TRUE (router)
636 1 - do source validation by reversed path, as specified in RFC1812
637 Recommended option for single homed hosts and stub network
638 routers. Could cause troubles for complicated (not loop free)
639 networks running a slow unreliable protocol (sort of RIP),
640 or using static routes.
642 0 - No source validation.
644 conf/all/rp_filter must also be set to TRUE to do source validation
647 Default value is 0. Note that some distributions enable it
651 1 - Allows you to have multiple network interfaces on the same
652 subnet, and have the ARPs for each interface be answered
653 based on whether or not the kernel would route a packet from
654 the ARP'd IP out that interface (therefore you must use source
655 based routing for this to work). In other words it allows control
656 of which cards (usually 1) will respond to an arp request.
658 0 - (default) The kernel can respond to arp requests with addresses
659 from other interfaces. This may seem wrong but it usually makes
660 sense, because it increases the chance of successful communication.
661 IP addresses are owned by the complete host on Linux, not by
662 particular interfaces. Only for more complex setups like load-
663 balancing, does this behaviour cause problems.
665 arp_filter for the interface will be enabled if at least one of
666 conf/{all,interface}/arp_filter is set to TRUE,
667 it will be disabled otherwise
669 arp_announce - INTEGER
670 Define different restriction levels for announcing the local
671 source IP address from IP packets in ARP requests sent on
673 0 - (default) Use any local address, configured on any interface
674 1 - Try to avoid local addresses that are not in the target's
675 subnet for this interface. This mode is useful when target
676 hosts reachable via this interface require the source IP
677 address in ARP requests to be part of their logical network
678 configured on the receiving interface. When we generate the
679 request we will check all our subnets that include the
680 target IP and will preserve the source address if it is from
681 such subnet. If there is no such subnet we select source
682 address according to the rules for level 2.
683 2 - Always use the best local address for this target.
684 In this mode we ignore the source address in the IP packet
685 and try to select local address that we prefer for talks with
686 the target host. Such local address is selected by looking
687 for primary IP addresses on all our subnets on the outgoing
688 interface that include the target IP address. If no suitable
689 local address is found we select the first local address
690 we have on the outgoing interface or on all other interfaces,
691 with the hope we will receive reply for our request and
692 even sometimes no matter the source IP address we announce.
694 The max value from conf/{all,interface}/arp_announce is used.
696 Increasing the restriction level gives more chance for
697 receiving answer from the resolved target while decreasing
698 the level announces more valid sender's information.
701 Define different modes for sending replies in response to
702 received ARP requests that resolve local target IP addresses:
703 0 - (default): reply for any local target IP address, configured
705 1 - reply only if the target IP address is local address
706 configured on the incoming interface
707 2 - reply only if the target IP address is local address
708 configured on the incoming interface and both with the
709 sender's IP address are part from same subnet on this interface
710 3 - do not reply for local addresses configured with scope host,
711 only resolutions for global and link addresses are replied
713 8 - do not reply for all local addresses
715 The max value from conf/{all,interface}/arp_ignore is used
716 when ARP request is received on the {interface}
719 Define behavior when gratuitous arp replies are received:
720 0 - drop gratuitous arp frames
721 1 - accept gratuitous arp frames
723 app_solicit - INTEGER
724 The maximum number of probes to send to the user space ARP daemon
725 via netlink before dropping back to multicast probes (see
726 mcast_solicit). Defaults to 0.
728 disable_policy - BOOLEAN
729 Disable IPSEC policy (SPD) for this interface
731 disable_xfrm - BOOLEAN
732 Disable IPSEC encryption on this interface, whatever the policy
737 Allows you to write a number, which can be used as required.
740 (1) Jiffie: internal timeunit for the kernel. On the i386 1/100s, on the
741 Alpha 1/1024s. See the HZ define in /usr/include/asm/param.h for the exact
742 value on your system.
751 delon.nicolas@wanadoo.fr
756 /proc/sys/net/ipv6/* Variables:
758 IPv6 has no global variables such as tcp_*. tcp_* settings under ipv4/ also
759 apply to IPv6 [XXX?].
762 Default value for IPV6_V6ONLY socket option,
763 which restricts use of the IPv6 socket to IPv6 communication
765 TRUE: disable IPv4-mapped address feature
766 FALSE: enable IPv4-mapped address feature
768 Default: FALSE (as specified in RFC2553bis)
772 ip6frag_high_thresh - INTEGER
773 Maximum memory used to reassemble IPv6 fragments. When
774 ip6frag_high_thresh bytes of memory is allocated for this purpose,
775 the fragment handler will toss packets until ip6frag_low_thresh
778 ip6frag_low_thresh - INTEGER
779 See ip6frag_high_thresh
781 ip6frag_time - INTEGER
782 Time in seconds to keep an IPv6 fragment in memory.
784 ip6frag_secret_interval - INTEGER
785 Regeneration interval (in seconds) of the hash secret (or lifetime
786 for the hash secret) for IPv6 fragments.
790 Change the interface-specific default settings.
794 Change all the interface-specific settings.
796 [XXX: Other special features than forwarding?]
798 conf/all/forwarding - BOOLEAN
799 Enable global IPv6 forwarding between all interfaces.
801 IPv4 and IPv6 work differently here; e.g. netfilter must be used
802 to control which interfaces may forward packets and which not.
804 This also sets all interfaces' Host/Router setting
805 'forwarding' to the specified value. See below for details.
807 This referred to as global forwarding.
813 Change special settings per interface.
815 The functional behaviour for certain settings is different
816 depending on whether local forwarding is enabled or not.
819 Accept Router Advertisements; autoconfigure using them.
821 Functional default: enabled if local forwarding is disabled.
822 disabled if local forwarding is enabled.
824 accept_ra_defrtr - BOOLEAN
825 Learn default router in Router Advertisement.
827 Functional default: enabled if accept_ra is enabled.
828 disabled if accept_ra is disabled.
830 accept_ra_pinfo - BOOLEAN
831 Learn Prefix Information in Router Advertisement.
833 Functional default: enabled if accept_ra is enabled.
834 disabled if accept_ra is disabled.
836 accept_ra_rt_info_max_plen - INTEGER
837 Maximum prefix length of Route Information in RA.
839 Route Information w/ prefix larger than or equal to this
840 variable shall be ignored.
842 Functional default: 0 if accept_ra_rtr_pref is enabled.
843 -1 if accept_ra_rtr_pref is disabled.
845 accept_ra_rtr_pref - BOOLEAN
846 Accept Router Preference in RA.
848 Functional default: enabled if accept_ra is enabled.
849 disabled if accept_ra is disabled.
851 accept_redirects - BOOLEAN
854 Functional default: enabled if local forwarding is disabled.
855 disabled if local forwarding is enabled.
857 accept_source_route - INTEGER
858 Accept source routing (routing extension header).
860 > 0: Accept routing header.
861 = 0: Accept only routing header type 2.
862 < 0: Do not accept routing header.
867 Autoconfigure addresses using Prefix Information in Router
870 Functional default: enabled if accept_ra_pinfo is enabled.
871 disabled if accept_ra_pinfo is disabled.
873 dad_transmits - INTEGER
874 The amount of Duplicate Address Detection probes to send.
878 Configure interface-specific Host/Router behaviour.
880 Note: It is recommended to have the same setting on all
881 interfaces; mixed router/host scenarios are rather uncommon.
885 By default, Host behaviour is assumed. This means:
887 1. IsRouter flag is not set in Neighbour Advertisements.
888 2. Router Solicitations are being sent when necessary.
889 3. If accept_ra is TRUE (default), accept Router
890 Advertisements (and do autoconfiguration).
891 4. If accept_redirects is TRUE (default), accept Redirects.
895 If local forwarding is enabled, Router behaviour is assumed.
896 This means exactly the reverse from the above:
898 1. IsRouter flag is set in Neighbour Advertisements.
899 2. Router Solicitations are not sent.
900 3. Router Advertisements are ignored.
901 4. Redirects are ignored.
903 Default: FALSE if global forwarding is disabled (default),
907 Default Hop Limit to set.
911 Default Maximum Transfer Unit
912 Default: 1280 (IPv6 required minimum)
914 router_probe_interval - INTEGER
915 Minimum interval (in seconds) between Router Probing described
920 router_solicitation_delay - INTEGER
921 Number of seconds to wait after interface is brought up
922 before sending Router Solicitations.
925 router_solicitation_interval - INTEGER
926 Number of seconds to wait between Router Solicitations.
929 router_solicitations - INTEGER
930 Number of Router Solicitations to send until assuming no
934 use_tempaddr - INTEGER
935 Preference for Privacy Extensions (RFC3041).
936 <= 0 : disable Privacy Extensions
937 == 1 : enable Privacy Extensions, but prefer public
938 addresses over temporary addresses.
939 > 1 : enable Privacy Extensions and prefer temporary
940 addresses over public addresses.
941 Default: 0 (for most devices)
942 -1 (for point-to-point devices and loopback devices)
944 temp_valid_lft - INTEGER
945 valid lifetime (in seconds) for temporary addresses.
946 Default: 604800 (7 days)
948 temp_prefered_lft - INTEGER
949 Preferred lifetime (in seconds) for temporary addresses.
950 Default: 86400 (1 day)
952 max_desync_factor - INTEGER
953 Maximum value for DESYNC_FACTOR, which is a random value
954 that ensures that clients don't synchronize with each
955 other and generate new addresses at exactly the same time.
959 regen_max_retry - INTEGER
960 Number of attempts before give up attempting to generate
961 valid temporary addresses.
964 max_addresses - INTEGER
965 Number of maximum addresses per interface. 0 disables limitation.
966 It is recommended not set too large value (or 0) because it would
967 be too easy way to crash kernel to allow to create too much of
968 autoconfigured addresses.
973 Limit the maximal rates for sending ICMPv6 packets.
974 0 to disable any limiting, otherwise the maximal rate in jiffies(1)
979 Pekka Savola <pekkas@netcore.fi>
980 YOSHIFUJI Hideaki / USAGI Project <yoshfuji@linux-ipv6.org>
983 /proc/sys/net/bridge/* Variables:
985 bridge-nf-call-arptables - BOOLEAN
986 1 : pass bridged ARP traffic to arptables' FORWARD chain.
990 bridge-nf-call-iptables - BOOLEAN
991 1 : pass bridged IPv4 traffic to iptables' chains.
995 bridge-nf-call-ip6tables - BOOLEAN
996 1 : pass bridged IPv6 traffic to ip6tables' chains.
1000 bridge-nf-filter-vlan-tagged - BOOLEAN
1001 1 : pass bridged vlan-tagged ARP/IP traffic to arptables/iptables.
1009 discovery_slots FIXME
1010 discovery_timeout FIXME
1011 fast_poll_increase FIXME
1012 ip6_queue_maxlen FIXME
1013 lap_keepalive_time FIXME
1016 max_dgram_qlen FIXME
1017 max_noreply_time FIXME
1018 max_tx_data_size FIXME
1020 min_tx_turn_time FIXME
1023 no_cong_thresh FIXME
1025 warn_noreply_time FIXME