About Kernel Documentation Linux Kernel Contact Linux Resources Linux Blog

Documentation / networking / ip-sysctl.txt




Custom Search

Based on kernel version 4.16.1. Page generated on 2018-04-09 11:53 EST.

1	/proc/sys/net/ipv4/* Variables:
2	
3	ip_forward - BOOLEAN
4		0 - disabled (default)
5		not 0 - enabled
6	
7		Forward Packets between interfaces.
8	
9		This variable is special, its change resets all configuration
10		parameters to their default state (RFC1122 for hosts, RFC1812
11		for routers)
12	
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)
17	
18	ip_no_pmtu_disc - INTEGER
19		Disable Path MTU Discovery. If enabled in mode 1 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.
24	
25		In mode 2 incoming Path MTU Discovery messages will be
26		discarded. Outgoing frames are handled the same as in mode 1,
27		implicitly setting IP_PMTUDISC_DONT on every created socket.
28	
29		Mode 3 is a hardend pmtu discover mode. The kernel will only
30		accept fragmentation-needed errors if the underlying protocol
31		can verify them besides a plain socket lookup. Current
32		protocols for which pmtu events will be honored are TCP, SCTP
33		and DCCP as they verify e.g. the sequence number or the
34		association. This mode should not be enabled globally but is
35		only intended to secure e.g. name servers in namespaces where
36		TCP path mtu must still work but path MTU information of other
37		protocols should be discarded. If enabled globally this mode
38		could break other protocols.
39	
40		Possible values: 0-3
41		Default: FALSE
42	
43	min_pmtu - INTEGER
44		default 552 - minimum discovered Path MTU
45	
46	ip_forward_use_pmtu - BOOLEAN
47		By default we don't trust protocol path MTUs while forwarding
48		because they could be easily forged and can lead to unwanted
49		fragmentation by the router.
50		You only need to enable this if you have user-space software
51		which tries to discover path mtus by itself and depends on the
52		kernel honoring this information. This is normally not the
53		case.
54		Default: 0 (disabled)
55		Possible values:
56		0 - disabled
57		1 - enabled
58	
59	fwmark_reflect - BOOLEAN
60		Controls the fwmark of kernel-generated IPv4 reply packets that are not
61		associated with a socket for example, TCP RSTs or ICMP echo replies).
62		If unset, these packets have a fwmark of zero. If set, they have the
63		fwmark of the packet they are replying to.
64		Default: 0
65	
66	fib_multipath_use_neigh - BOOLEAN
67		Use status of existing neighbor entry when determining nexthop for
68		multipath routes. If disabled, neighbor information is not used and
69		packets could be directed to a failed nexthop. Only valid for kernels
70		built with CONFIG_IP_ROUTE_MULTIPATH enabled.
71		Default: 0 (disabled)
72		Possible values:
73		0 - disabled
74		1 - enabled
75	
76	fib_multipath_hash_policy - INTEGER
77		Controls which hash policy to use for multipath routes. Only valid
78		for kernels built with CONFIG_IP_ROUTE_MULTIPATH enabled.
79		Default: 0 (Layer 3)
80		Possible values:
81		0 - Layer 3
82		1 - Layer 4
83	
84	route/max_size - INTEGER
85		Maximum number of routes allowed in the kernel.  Increase
86		this when using large numbers of interfaces and/or routes.
87		From linux kernel 3.6 onwards, this is deprecated for ipv4
88		as route cache is no longer used.
89	
90	neigh/default/gc_thresh1 - INTEGER
91		Minimum number of entries to keep.  Garbage collector will not
92		purge entries if there are fewer than this number.
93		Default: 128
94	
95	neigh/default/gc_thresh2 - INTEGER
96		Threshold when garbage collector becomes more aggressive about
97		purging entries. Entries older than 5 seconds will be cleared
98		when over this number.
99		Default: 512
100	
101	neigh/default/gc_thresh3 - INTEGER
102		Maximum number of neighbor entries allowed.  Increase this
103		when using large numbers of interfaces and when communicating
104		with large numbers of directly-connected peers.
105		Default: 1024
106	
107	neigh/default/unres_qlen_bytes - INTEGER
108		The maximum number of bytes which may be used by packets
109		queued for each	unresolved address by other network layers.
110		(added in linux 3.3)
111		Setting negative value is meaningless and will return error.
112		Default: SK_WMEM_MAX, (same as net.core.wmem_default).
113			Exact value depends on architecture and kernel options,
114			but should be enough to allow queuing 256 packets
115			of medium size.
116	
117	neigh/default/unres_qlen - INTEGER
118		The maximum number of packets which may be queued for each
119		unresolved address by other network layers.
120		(deprecated in linux 3.3) : use unres_qlen_bytes instead.
121		Prior to linux 3.3, the default value is 3 which may cause
122		unexpected packet loss. The current default value is calculated
123		according to default value of unres_qlen_bytes and true size of
124		packet.
125		Default: 101
126	
127	mtu_expires - INTEGER
128		Time, in seconds, that cached PMTU information is kept.
129	
130	min_adv_mss - INTEGER
131		The advertised MSS depends on the first hop route MTU, but will
132		never be lower than this setting.
133	
134	IP Fragmentation:
135	
136	ipfrag_high_thresh - INTEGER
137		Maximum memory used to reassemble IP fragments. When
138		ipfrag_high_thresh bytes of memory is allocated for this purpose,
139		the fragment handler will toss packets until ipfrag_low_thresh
140		is reached. This also serves as a maximum limit to namespaces
141		different from the initial one.
142	
143	ipfrag_low_thresh - INTEGER
144		Maximum memory used to reassemble IP fragments before the kernel
145		begins to remove incomplete fragment queues to free up resources.
146		The kernel still accepts new fragments for defragmentation.
147	
148	ipfrag_time - INTEGER
149		Time in seconds to keep an IP fragment in memory.
150	
151	ipfrag_max_dist - INTEGER
152		ipfrag_max_dist is a non-negative integer value which defines the
153		maximum "disorder" which is allowed among fragments which share a
154		common IP source address. Note that reordering of packets is
155		not unusual, but if a large number of fragments arrive from a source
156		IP address while a particular fragment queue remains incomplete, it
157		probably indicates that one or more fragments belonging to that queue
158		have been lost. When ipfrag_max_dist is positive, an additional check
159		is done on fragments before they are added to a reassembly queue - if
160		ipfrag_max_dist (or more) fragments have arrived from a particular IP
161		address between additions to any IP fragment queue using that source
162		address, it's presumed that one or more fragments in the queue are
163		lost. The existing fragment queue will be dropped, and a new one
164		started. An ipfrag_max_dist value of zero disables this check.
165	
166		Using a very small value, e.g. 1 or 2, for ipfrag_max_dist can
167		result in unnecessarily dropping fragment queues when normal
168		reordering of packets occurs, which could lead to poor application
169		performance. Using a very large value, e.g. 50000, increases the
170		likelihood of incorrectly reassembling IP fragments that originate
171		from different IP datagrams, which could result in data corruption.
172		Default: 64
173	
174	INET peer storage:
175	
176	inet_peer_threshold - INTEGER
177		The approximate size of the storage.  Starting from this threshold
178		entries will be thrown aggressively.  This threshold also determines
179		entries' time-to-live and time intervals between garbage collection
180		passes.  More entries, less time-to-live, less GC interval.
181	
182	inet_peer_minttl - INTEGER
183		Minimum time-to-live of entries.  Should be enough to cover fragment
184		time-to-live on the reassembling side.  This minimum time-to-live  is
185		guaranteed if the pool size is less than inet_peer_threshold.
186		Measured in seconds.
187	
188	inet_peer_maxttl - INTEGER
189		Maximum time-to-live of entries.  Unused entries will expire after
190		this period of time if there is no memory pressure on the pool (i.e.
191		when the number of entries in the pool is very small).
192		Measured in seconds.
193	
194	TCP variables:
195	
196	somaxconn - INTEGER
197		Limit of socket listen() backlog, known in userspace as SOMAXCONN.
198		Defaults to 128.  See also tcp_max_syn_backlog for additional tuning
199		for TCP sockets.
200	
201	tcp_abort_on_overflow - BOOLEAN
202		If listening service is too slow to accept new connections,
203		reset them. Default state is FALSE. It means that if overflow
204		occurred due to a burst, connection will recover. Enable this
205		option _only_ if you are really sure that listening daemon
206		cannot be tuned to accept connections faster. Enabling this
207		option can harm clients of your server.
208	
209	tcp_adv_win_scale - INTEGER
210		Count buffering overhead as bytes/2^tcp_adv_win_scale
211		(if tcp_adv_win_scale > 0) or bytes-bytes/2^(-tcp_adv_win_scale),
212		if it is <= 0.
213		Possible values are [-31, 31], inclusive.
214		Default: 1
215	
216	tcp_allowed_congestion_control - STRING
217		Show/set the congestion control choices available to non-privileged
218		processes. The list is a subset of those listed in
219		tcp_available_congestion_control.
220		Default is "reno" and the default setting (tcp_congestion_control).
221	
222	tcp_app_win - INTEGER
223		Reserve max(window/2^tcp_app_win, mss) of window for application
224		buffer. Value 0 is special, it means that nothing is reserved.
225		Default: 31
226	
227	tcp_autocorking - BOOLEAN
228		Enable TCP auto corking :
229		When applications do consecutive small write()/sendmsg() system calls,
230		we try to coalesce these small writes as much as possible, to lower
231		total amount of sent packets. This is done if at least one prior
232		packet for the flow is waiting in Qdisc queues or device transmit
233		queue. Applications can still use TCP_CORK for optimal behavior
234		when they know how/when to uncork their sockets.
235		Default : 1
236	
237	tcp_available_congestion_control - STRING
238		Shows the available congestion control choices that are registered.
239		More congestion control algorithms may be available as modules,
240		but not loaded.
241	
242	tcp_base_mss - INTEGER
243		The initial value of search_low to be used by the packetization layer
244		Path MTU discovery (MTU probing).  If MTU probing is enabled,
245		this is the initial MSS used by the connection.
246	
247	tcp_congestion_control - STRING
248		Set the congestion control algorithm to be used for new
249		connections. The algorithm "reno" is always available, but
250		additional choices may be available based on kernel configuration.
251		Default is set as part of kernel configuration.
252		For passive connections, the listener congestion control choice
253		is inherited.
254		[see setsockopt(listenfd, SOL_TCP, TCP_CONGESTION, "name" ...) ]
255	
256	tcp_dsack - BOOLEAN
257		Allows TCP to send "duplicate" SACKs.
258	
259	tcp_early_retrans - INTEGER
260		Tail loss probe (TLP) converts RTOs occurring due to tail
261		losses into fast recovery (draft-ietf-tcpm-rack). Note that
262		TLP requires RACK to function properly (see tcp_recovery below)
263		Possible values:
264			0 disables TLP
265			3 or 4 enables TLP
266		Default: 3
267	
268	tcp_ecn - INTEGER
269		Control use of Explicit Congestion Notification (ECN) by TCP.
270		ECN is used only when both ends of the TCP connection indicate
271		support for it.  This feature is useful in avoiding losses due
272		to congestion by allowing supporting routers to signal
273		congestion before having to drop packets.
274		Possible values are:
275			0 Disable ECN.  Neither initiate nor accept ECN.
276			1 Enable ECN when requested by incoming connections and
277			  also request ECN on outgoing connection attempts.
278			2 Enable ECN when requested by incoming connections
279			  but do not request ECN on outgoing connections.
280		Default: 2
281	
282	tcp_ecn_fallback - BOOLEAN
283		If the kernel detects that ECN connection misbehaves, enable fall
284		back to non-ECN. Currently, this knob implements the fallback
285		from RFC3168, section 6.1.1.1., but we reserve that in future,
286		additional detection mechanisms could be implemented under this
287		knob. The value	is not used, if tcp_ecn or per route (or congestion
288		control) ECN settings are disabled.
289		Default: 1 (fallback enabled)
290	
291	tcp_fack - BOOLEAN
292		This is a legacy option, it has no effect anymore.
293	
294	tcp_fin_timeout - INTEGER
295		The length of time an orphaned (no longer referenced by any
296		application) connection will remain in the FIN_WAIT_2 state
297		before it is aborted at the local end.  While a perfectly
298		valid "receive only" state for an un-orphaned connection, an
299		orphaned connection in FIN_WAIT_2 state could otherwise wait
300		forever for the remote to close its end of the connection.
301		Cf. tcp_max_orphans
302		Default: 60 seconds
303	
304	tcp_frto - INTEGER
305		Enables Forward RTO-Recovery (F-RTO) defined in RFC5682.
306		F-RTO is an enhanced recovery algorithm for TCP retransmission
307		timeouts.  It is particularly beneficial in networks where the
308		RTT fluctuates (e.g., wireless). F-RTO is sender-side only
309		modification. It does not require any support from the peer.
310	
311		By default it's enabled with a non-zero value. 0 disables F-RTO.
312	
313	tcp_invalid_ratelimit - INTEGER
314		Limit the maximal rate for sending duplicate acknowledgments
315		in response to incoming TCP packets that are for an existing
316		connection but that are invalid due to any of these reasons:
317	
318		  (a) out-of-window sequence number,
319		  (b) out-of-window acknowledgment number, or
320		  (c) PAWS (Protection Against Wrapped Sequence numbers) check failure
321	
322		This can help mitigate simple "ack loop" DoS attacks, wherein
323		a buggy or malicious middlebox or man-in-the-middle can
324		rewrite TCP header fields in manner that causes each endpoint
325		to think that the other is sending invalid TCP segments, thus
326		causing each side to send an unterminating stream of duplicate
327		acknowledgments for invalid segments.
328	
329		Using 0 disables rate-limiting of dupacks in response to
330		invalid segments; otherwise this value specifies the minimal
331		space between sending such dupacks, in milliseconds.
332	
333		Default: 500 (milliseconds).
334	
335	tcp_keepalive_time - INTEGER
336		How often TCP sends out keepalive messages when keepalive is enabled.
337		Default: 2hours.
338	
339	tcp_keepalive_probes - INTEGER
340		How many keepalive probes TCP sends out, until it decides that the
341		connection is broken. Default value: 9.
342	
343	tcp_keepalive_intvl - INTEGER
344		How frequently the probes are send out. Multiplied by
345		tcp_keepalive_probes it is time to kill not responding connection,
346		after probes started. Default value: 75sec i.e. connection
347		will be aborted after ~11 minutes of retries.
348	
349	tcp_l3mdev_accept - BOOLEAN
350		Enables child sockets to inherit the L3 master device index.
351		Enabling this option allows a "global" listen socket to work
352		across L3 master domains (e.g., VRFs) with connected sockets
353		derived from the listen socket to be bound to the L3 domain in
354		which the packets originated. Only valid when the kernel was
355		compiled with CONFIG_NET_L3_MASTER_DEV.
356	
357	tcp_low_latency - BOOLEAN
358		This is a legacy option, it has no effect anymore.
359	
360	tcp_max_orphans - INTEGER
361		Maximal number of TCP sockets not attached to any user file handle,
362		held by system.	If this number is exceeded orphaned connections are
363		reset immediately and warning is printed. This limit exists
364		only to prevent simple DoS attacks, you _must_ not rely on this
365		or lower the limit artificially, but rather increase it
366		(probably, after increasing installed memory),
367		if network conditions require more than default value,
368		and tune network services to linger and kill such states
369		more aggressively. Let me to remind again: each orphan eats
370		up to ~64K of unswappable memory.
371	
372	tcp_max_syn_backlog - INTEGER
373		Maximal number of remembered connection requests, which have not
374		received an acknowledgment from connecting client.
375		The minimal value is 128 for low memory machines, and it will
376		increase in proportion to the memory of machine.
377		If server suffers from overload, try increasing this number.
378	
379	tcp_max_tw_buckets - INTEGER
380		Maximal number of timewait sockets held by system simultaneously.
381		If this number is exceeded time-wait socket is immediately destroyed
382		and warning is printed. This limit exists only to prevent
383		simple DoS attacks, you _must_ not lower the limit artificially,
384		but rather increase it (probably, after increasing installed memory),
385		if network conditions require more than default value.
386	
387	tcp_mem - vector of 3 INTEGERs: min, pressure, max
388		min: below this number of pages TCP is not bothered about its
389		memory appetite.
390	
391		pressure: when amount of memory allocated by TCP exceeds this number
392		of pages, TCP moderates its memory consumption and enters memory
393		pressure mode, which is exited when memory consumption falls
394		under "min".
395	
396		max: number of pages allowed for queueing by all TCP sockets.
397	
398		Defaults are calculated at boot time from amount of available
399		memory.
400	
401	tcp_min_rtt_wlen - INTEGER
402		The window length of the windowed min filter to track the minimum RTT.
403		A shorter window lets a flow more quickly pick up new (higher)
404		minimum RTT when it is moved to a longer path (e.g., due to traffic
405		engineering). A longer window makes the filter more resistant to RTT
406		inflations such as transient congestion. The unit is seconds.
407		Default: 300
408	
409	tcp_moderate_rcvbuf - BOOLEAN
410		If set, TCP performs receive buffer auto-tuning, attempting to
411		automatically size the buffer (no greater than tcp_rmem[2]) to
412		match the size required by the path for full throughput.  Enabled by
413		default.
414	
415	tcp_mtu_probing - INTEGER
416		Controls TCP Packetization-Layer Path MTU Discovery.  Takes three
417		values:
418		  0 - Disabled
419		  1 - Disabled by default, enabled when an ICMP black hole detected
420		  2 - Always enabled, use initial MSS of tcp_base_mss.
421	
422	tcp_probe_interval - INTEGER
423		Controls how often to start TCP Packetization-Layer Path MTU
424		Discovery reprobe. The default is reprobing every 10 minutes as
425		per RFC4821.
426	
427	tcp_probe_threshold - INTEGER
428		Controls when TCP Packetization-Layer Path MTU Discovery probing
429		will stop in respect to the width of search range in bytes. Default
430		is 8 bytes.
431	
432	tcp_no_metrics_save - BOOLEAN
433		By default, TCP saves various connection metrics in the route cache
434		when the connection closes, so that connections established in the
435		near future can use these to set initial conditions.  Usually, this
436		increases overall performance, but may sometimes cause performance
437		degradation.  If set, TCP will not cache metrics on closing
438		connections.
439	
440	tcp_orphan_retries - INTEGER
441		This value influences the timeout of a locally closed TCP connection,
442		when RTO retransmissions remain unacknowledged.
443		See tcp_retries2 for more details.
444	
445		The default value is 8.
446		If your machine is a loaded WEB server,
447		you should think about lowering this value, such sockets
448		may consume significant resources. Cf. tcp_max_orphans.
449	
450	tcp_recovery - INTEGER
451		This value is a bitmap to enable various experimental loss recovery
452		features.
453	
454		RACK: 0x1 enables the RACK loss detection for fast detection of lost
455		      retransmissions and tail drops.
456		RACK: 0x2 makes RACK's reordering window static (min_rtt/4).
457	
458		Default: 0x1
459	
460	tcp_reordering - INTEGER
461		Initial reordering level of packets in a TCP stream.
462		TCP stack can then dynamically adjust flow reordering level
463		between this initial value and tcp_max_reordering
464		Default: 3
465	
466	tcp_max_reordering - INTEGER
467		Maximal reordering level of packets in a TCP stream.
468		300 is a fairly conservative value, but you might increase it
469		if paths are using per packet load balancing (like bonding rr mode)
470		Default: 300
471	
472	tcp_retrans_collapse - BOOLEAN
473		Bug-to-bug compatibility with some broken printers.
474		On retransmit try to send bigger packets to work around bugs in
475		certain TCP stacks.
476	
477	tcp_retries1 - INTEGER
478		This value influences the time, after which TCP decides, that
479		something is wrong due to unacknowledged RTO retransmissions,
480		and reports this suspicion to the network layer.
481		See tcp_retries2 for more details.
482	
483		RFC 1122 recommends at least 3 retransmissions, which is the
484		default.
485	
486	tcp_retries2 - INTEGER
487		This value influences the timeout of an alive TCP connection,
488		when RTO retransmissions remain unacknowledged.
489		Given a value of N, a hypothetical TCP connection following
490		exponential backoff with an initial RTO of TCP_RTO_MIN would
491		retransmit N times before killing the connection at the (N+1)th RTO.
492	
493		The default value of 15 yields a hypothetical timeout of 924.6
494		seconds and is a lower bound for the effective timeout.
495		TCP will effectively time out at the first RTO which exceeds the
496		hypothetical timeout.
497	
498		RFC 1122 recommends at least 100 seconds for the timeout,
499		which corresponds to a value of at least 8.
500	
501	tcp_rfc1337 - BOOLEAN
502		If set, the TCP stack behaves conforming to RFC1337. If unset,
503		we are not conforming to RFC, but prevent TCP TIME_WAIT
504		assassination.
505		Default: 0
506	
507	tcp_rmem - vector of 3 INTEGERs: min, default, max
508		min: Minimal size of receive buffer used by TCP sockets.
509		It is guaranteed to each TCP socket, even under moderate memory
510		pressure.
511		Default: 4K
512	
513		default: initial size of receive buffer used by TCP sockets.
514		This value overrides net.core.rmem_default used by other protocols.
515		Default: 87380 bytes. This value results in window of 65535 with
516		default setting of tcp_adv_win_scale and tcp_app_win:0 and a bit
517		less for default tcp_app_win. See below about these variables.
518	
519		max: maximal size of receive buffer allowed for automatically
520		selected receiver buffers for TCP socket. This value does not override
521		net.core.rmem_max.  Calling setsockopt() with SO_RCVBUF disables
522		automatic tuning of that socket's receive buffer size, in which
523		case this value is ignored.
524		Default: between 87380B and 6MB, depending on RAM size.
525	
526	tcp_sack - BOOLEAN
527		Enable select acknowledgments (SACKS).
528	
529	tcp_slow_start_after_idle - BOOLEAN
530		If set, provide RFC2861 behavior and time out the congestion
531		window after an idle period.  An idle period is defined at
532		the current RTO.  If unset, the congestion window will not
533		be timed out after an idle period.
534		Default: 1
535	
536	tcp_stdurg - BOOLEAN
537		Use the Host requirements interpretation of the TCP urgent pointer field.
538		Most hosts use the older BSD interpretation, so if you turn this on
539		Linux might not communicate correctly with them.
540		Default: FALSE
541	
542	tcp_synack_retries - INTEGER
543		Number of times SYNACKs for a passive TCP connection attempt will
544		be retransmitted. Should not be higher than 255. Default value
545		is 5, which corresponds to 31seconds till the last retransmission
546		with the current initial RTO of 1second. With this the final timeout
547		for a passive TCP connection will happen after 63seconds.
548	
549	tcp_syncookies - BOOLEAN
550		Only valid when the kernel was compiled with CONFIG_SYN_COOKIES
551		Send out syncookies when the syn backlog queue of a socket
552		overflows. This is to prevent against the common 'SYN flood attack'
553		Default: 1
554	
555		Note, that syncookies is fallback facility.
556		It MUST NOT be used to help highly loaded servers to stand
557		against legal connection rate. If you see SYN flood warnings
558		in your logs, but investigation	shows that they occur
559		because of overload with legal connections, you should tune
560		another parameters until this warning disappear.
561		See: tcp_max_syn_backlog, tcp_synack_retries, tcp_abort_on_overflow.
562	
563		syncookies seriously violate TCP protocol, do not allow
564		to use TCP extensions, can result in serious degradation
565		of some services (f.e. SMTP relaying), visible not by you,
566		but your clients and relays, contacting you. While you see
567		SYN flood warnings in logs not being really flooded, your server
568		is seriously misconfigured.
569	
570		If you want to test which effects syncookies have to your
571		network connections you can set this knob to 2 to enable
572		unconditionally generation of syncookies.
573	
574	tcp_fastopen - INTEGER
575		Enable TCP Fast Open (RFC7413) to send and accept data in the opening
576		SYN packet.
577	
578		The client support is enabled by flag 0x1 (on by default). The client
579		then must use sendmsg() or sendto() with the MSG_FASTOPEN flag,
580		rather than connect() to send data in SYN.
581	
582		The server support is enabled by flag 0x2 (off by default). Then
583		either enable for all listeners with another flag (0x400) or
584		enable individual listeners via TCP_FASTOPEN socket option with
585		the option value being the length of the syn-data backlog.
586	
587		The values (bitmap) are
588		  0x1: (client) enables sending data in the opening SYN on the client.
589		  0x2: (server) enables the server support, i.e., allowing data in
590				a SYN packet to be accepted and passed to the
591				application before 3-way handshake finishes.
592		  0x4: (client) send data in the opening SYN regardless of cookie
593				availability and without a cookie option.
594		0x200: (server) accept data-in-SYN w/o any cookie option present.
595		0x400: (server) enable all listeners to support Fast Open by
596				default without explicit TCP_FASTOPEN socket option.
597	
598		Default: 0x1
599	
600		Note that that additional client or server features are only
601		effective if the basic support (0x1 and 0x2) are enabled respectively.
602	
603	tcp_fastopen_blackhole_timeout_sec - INTEGER
604		Initial time period in second to disable Fastopen on active TCP sockets
605		when a TFO firewall blackhole issue happens.
606		This time period will grow exponentially when more blackhole issues
607		get detected right after Fastopen is re-enabled and will reset to
608		initial value when the blackhole issue goes away.
609		0 to disable the blackhole detection.
610		By default, it is set to 1hr.
611	
612	tcp_syn_retries - INTEGER
613		Number of times initial SYNs for an active TCP connection attempt
614		will be retransmitted. Should not be higher than 127. Default value
615		is 6, which corresponds to 63seconds till the last retransmission
616		with the current initial RTO of 1second. With this the final timeout
617		for an active TCP connection attempt will happen after 127seconds.
618	
619	tcp_timestamps - INTEGER
620	Enable timestamps as defined in RFC1323.
621		0: Disabled.
622		1: Enable timestamps as defined in RFC1323 and use random offset for
623		each connection rather than only using the current time.
624		2: Like 1, but without random offsets.
625		Default: 1
626	
627	tcp_min_tso_segs - INTEGER
628		Minimal number of segments per TSO frame.
629		Since linux-3.12, TCP does an automatic sizing of TSO frames,
630		depending on flow rate, instead of filling 64Kbytes packets.
631		For specific usages, it's possible to force TCP to build big
632		TSO frames. Note that TCP stack might split too big TSO packets
633		if available window is too small.
634		Default: 2
635	
636	tcp_pacing_ss_ratio - INTEGER
637		sk->sk_pacing_rate is set by TCP stack using a ratio applied
638		to current rate. (current_rate = cwnd * mss / srtt)
639		If TCP is in slow start, tcp_pacing_ss_ratio is applied
640		to let TCP probe for bigger speeds, assuming cwnd can be
641		doubled every other RTT.
642		Default: 200
643	
644	tcp_pacing_ca_ratio - INTEGER
645		sk->sk_pacing_rate is set by TCP stack using a ratio applied
646		to current rate. (current_rate = cwnd * mss / srtt)
647		If TCP is in congestion avoidance phase, tcp_pacing_ca_ratio
648		is applied to conservatively probe for bigger throughput.
649		Default: 120
650	
651	tcp_tso_win_divisor - INTEGER
652		This allows control over what percentage of the congestion window
653		can be consumed by a single TSO frame.
654		The setting of this parameter is a choice between burstiness and
655		building larger TSO frames.
656		Default: 3
657	
658	tcp_tw_reuse - BOOLEAN
659		Allow to reuse TIME-WAIT sockets for new connections when it is
660		safe from protocol viewpoint. Default value is 0.
661		It should not be changed without advice/request of technical
662		experts.
663	
664	tcp_window_scaling - BOOLEAN
665		Enable window scaling as defined in RFC1323.
666	
667	tcp_wmem - vector of 3 INTEGERs: min, default, max
668		min: Amount of memory reserved for send buffers for TCP sockets.
669		Each TCP socket has rights to use it due to fact of its birth.
670		Default: 4K
671	
672		default: initial size of send buffer used by TCP sockets.  This
673		value overrides net.core.wmem_default used by other protocols.
674		It is usually lower than net.core.wmem_default.
675		Default: 16K
676	
677		max: Maximal amount of memory allowed for automatically tuned
678		send buffers for TCP sockets. This value does not override
679		net.core.wmem_max.  Calling setsockopt() with SO_SNDBUF disables
680		automatic tuning of that socket's send buffer size, in which case
681		this value is ignored.
682		Default: between 64K and 4MB, depending on RAM size.
683	
684	tcp_notsent_lowat - UNSIGNED INTEGER
685		A TCP socket can control the amount of unsent bytes in its write queue,
686		thanks to TCP_NOTSENT_LOWAT socket option. poll()/select()/epoll()
687		reports POLLOUT events if the amount of unsent bytes is below a per
688		socket value, and if the write queue is not full. sendmsg() will
689		also not add new buffers if the limit is hit.
690	
691		This global variable controls the amount of unsent data for
692		sockets not using TCP_NOTSENT_LOWAT. For these sockets, a change
693		to the global variable has immediate effect.
694	
695		Default: UINT_MAX (0xFFFFFFFF)
696	
697	tcp_workaround_signed_windows - BOOLEAN
698		If set, assume no receipt of a window scaling option means the
699		remote TCP is broken and treats the window as a signed quantity.
700		If unset, assume the remote TCP is not broken even if we do
701		not receive a window scaling option from them.
702		Default: 0
703	
704	tcp_thin_linear_timeouts - BOOLEAN
705		Enable dynamic triggering of linear timeouts for thin streams.
706		If set, a check is performed upon retransmission by timeout to
707		determine if the stream is thin (less than 4 packets in flight).
708		As long as the stream is found to be thin, up to 6 linear
709		timeouts may be performed before exponential backoff mode is
710		initiated. This improves retransmission latency for
711		non-aggressive thin streams, often found to be time-dependent.
712		For more information on thin streams, see
713		Documentation/networking/tcp-thin.txt
714		Default: 0
715	
716	tcp_limit_output_bytes - INTEGER
717		Controls TCP Small Queue limit per tcp socket.
718		TCP bulk sender tends to increase packets in flight until it
719		gets losses notifications. With SNDBUF autotuning, this can
720		result in a large amount of packets queued in qdisc/device
721		on the local machine, hurting latency of other flows, for
722		typical pfifo_fast qdiscs.
723		tcp_limit_output_bytes limits the number of bytes on qdisc
724		or device to reduce artificial RTT/cwnd and reduce bufferbloat.
725		Default: 262144
726	
727	tcp_challenge_ack_limit - INTEGER
728		Limits number of Challenge ACK sent per second, as recommended
729		in RFC 5961 (Improving TCP's Robustness to Blind In-Window Attacks)
730		Default: 100
731	
732	UDP variables:
733	
734	udp_l3mdev_accept - BOOLEAN
735		Enabling this option allows a "global" bound socket to work
736		across L3 master domains (e.g., VRFs) with packets capable of
737		being received regardless of the L3 domain in which they
738		originated. Only valid when the kernel was compiled with
739		CONFIG_NET_L3_MASTER_DEV.
740	
741	udp_mem - vector of 3 INTEGERs: min, pressure, max
742		Number of pages allowed for queueing by all UDP sockets.
743	
744		min: Below this number of pages UDP is not bothered about its
745		memory appetite. When amount of memory allocated by UDP exceeds
746		this number, UDP starts to moderate memory usage.
747	
748		pressure: This value was introduced to follow format of tcp_mem.
749	
750		max: Number of pages allowed for queueing by all UDP sockets.
751	
752		Default is calculated at boot time from amount of available memory.
753	
754	udp_rmem_min - INTEGER
755		Minimal size of receive buffer used by UDP sockets in moderation.
756		Each UDP socket is able to use the size for receiving data, even if
757		total pages of UDP sockets exceed udp_mem pressure. The unit is byte.
758		Default: 1 page
759	
760	udp_wmem_min - INTEGER
761		Minimal size of send buffer used by UDP sockets in moderation.
762		Each UDP socket is able to use the size for sending data, even if
763		total pages of UDP sockets exceed udp_mem pressure. The unit is byte.
764		Default: 1 page
765	
766	CIPSOv4 Variables:
767	
768	cipso_cache_enable - BOOLEAN
769		If set, enable additions to and lookups from the CIPSO label mapping
770		cache.  If unset, additions are ignored and lookups always result in a
771		miss.  However, regardless of the setting the cache is still
772		invalidated when required when means you can safely toggle this on and
773		off and the cache will always be "safe".
774		Default: 1
775	
776	cipso_cache_bucket_size - INTEGER
777		The CIPSO label cache consists of a fixed size hash table with each
778		hash bucket containing a number of cache entries.  This variable limits
779		the number of entries in each hash bucket; the larger the value the
780		more CIPSO label mappings that can be cached.  When the number of
781		entries in a given hash bucket reaches this limit adding new entries
782		causes the oldest entry in the bucket to be removed to make room.
783		Default: 10
784	
785	cipso_rbm_optfmt - BOOLEAN
786		Enable the "Optimized Tag 1 Format" as defined in section 3.4.2.6 of
787		the CIPSO draft specification (see Documentation/netlabel for details).
788		This means that when set the CIPSO tag will be padded with empty
789		categories in order to make the packet data 32-bit aligned.
790		Default: 0
791	
792	cipso_rbm_structvalid - BOOLEAN
793		If set, do a very strict check of the CIPSO option when
794		ip_options_compile() is called.  If unset, relax the checks done during
795		ip_options_compile().  Either way is "safe" as errors are caught else
796		where in the CIPSO processing code but setting this to 0 (False) should
797		result in less work (i.e. it should be faster) but could cause problems
798		with other implementations that require strict checking.
799		Default: 0
800	
801	IP Variables:
802	
803	ip_local_port_range - 2 INTEGERS
804		Defines the local port range that is used by TCP and UDP to
805		choose the local port. The first number is the first, the
806		second the last local port number.
807		If possible, it is better these numbers have different parity.
808		(one even and one odd values)
809		The default values are 32768 and 60999 respectively.
810	
811	ip_local_reserved_ports - list of comma separated ranges
812		Specify the ports which are reserved for known third-party
813		applications. These ports will not be used by automatic port
814		assignments (e.g. when calling connect() or bind() with port
815		number 0). Explicit port allocation behavior is unchanged.
816	
817		The format used for both input and output is a comma separated
818		list of ranges (e.g. "1,2-4,10-10" for ports 1, 2, 3, 4 and
819		10). Writing to the file will clear all previously reserved
820		ports and update the current list with the one given in the
821		input.
822	
823		Note that ip_local_port_range and ip_local_reserved_ports
824		settings are independent and both are considered by the kernel
825		when determining which ports are available for automatic port
826		assignments.
827	
828		You can reserve ports which are not in the current
829		ip_local_port_range, e.g.:
830	
831		$ cat /proc/sys/net/ipv4/ip_local_port_range
832		32000	60999
833		$ cat /proc/sys/net/ipv4/ip_local_reserved_ports
834		8080,9148
835	
836		although this is redundant. However such a setting is useful
837		if later the port range is changed to a value that will
838		include the reserved ports.
839	
840		Default: Empty
841	
842	ip_unprivileged_port_start - INTEGER
843		This is a per-namespace sysctl.  It defines the first
844		unprivileged port in the network namespace.  Privileged ports
845		require root or CAP_NET_BIND_SERVICE in order to bind to them.
846		To disable all privileged ports, set this to 0.  It may not
847		overlap with the ip_local_reserved_ports range.
848	
849		Default: 1024
850	
851	ip_nonlocal_bind - BOOLEAN
852		If set, allows processes to bind() to non-local IP addresses,
853		which can be quite useful - but may break some applications.
854		Default: 0
855	
856	ip_dynaddr - BOOLEAN
857		If set non-zero, enables support for dynamic addresses.
858		If set to a non-zero value larger than 1, a kernel log
859		message will be printed when dynamic address rewriting
860		occurs.
861		Default: 0
862	
863	ip_early_demux - BOOLEAN
864		Optimize input packet processing down to one demux for
865		certain kinds of local sockets.  Currently we only do this
866		for established TCP and connected UDP sockets.
867	
868		It may add an additional cost for pure routing workloads that
869		reduces overall throughput, in such case you should disable it.
870		Default: 1
871	
872	tcp_early_demux - BOOLEAN
873		Enable early demux for established TCP sockets.
874		Default: 1
875	
876	udp_early_demux - BOOLEAN
877		Enable early demux for connected UDP sockets. Disable this if
878		your system could experience more unconnected load.
879		Default: 1
880	
881	icmp_echo_ignore_all - BOOLEAN
882		If set non-zero, then the kernel will ignore all ICMP ECHO
883		requests sent to it.
884		Default: 0
885	
886	icmp_echo_ignore_broadcasts - BOOLEAN
887		If set non-zero, then the kernel will ignore all ICMP ECHO and
888		TIMESTAMP requests sent to it via broadcast/multicast.
889		Default: 1
890	
891	icmp_ratelimit - INTEGER
892		Limit the maximal rates for sending ICMP packets whose type matches
893		icmp_ratemask (see below) to specific targets.
894		0 to disable any limiting,
895		otherwise the minimal space between responses in milliseconds.
896		Note that another sysctl, icmp_msgs_per_sec limits the number
897		of ICMP packets	sent on all targets.
898		Default: 1000
899	
900	icmp_msgs_per_sec - INTEGER
901		Limit maximal number of ICMP packets sent per second from this host.
902		Only messages whose type matches icmp_ratemask (see below) are
903		controlled by this limit.
904		Default: 1000
905	
906	icmp_msgs_burst - INTEGER
907		icmp_msgs_per_sec controls number of ICMP packets sent per second,
908		while icmp_msgs_burst controls the burst size of these packets.
909		Default: 50
910	
911	icmp_ratemask - INTEGER
912		Mask made of ICMP types for which rates are being limited.
913		Significant bits: IHGFEDCBA9876543210
914		Default mask:     0000001100000011000 (6168)
915	
916		Bit definitions (see include/linux/icmp.h):
917			0 Echo Reply
918			3 Destination Unreachable *
919			4 Source Quench *
920			5 Redirect
921			8 Echo Request
922			B Time Exceeded *
923			C Parameter Problem *
924			D Timestamp Request
925			E Timestamp Reply
926			F Info Request
927			G Info Reply
928			H Address Mask Request
929			I Address Mask Reply
930	
931		* These are rate limited by default (see default mask above)
932	
933	icmp_ignore_bogus_error_responses - BOOLEAN
934		Some routers violate RFC1122 by sending bogus responses to broadcast
935		frames.  Such violations are normally logged via a kernel warning.
936		If this is set to TRUE, the kernel will not give such warnings, which
937		will avoid log file clutter.
938		Default: 1
939	
940	icmp_errors_use_inbound_ifaddr - BOOLEAN
941	
942		If zero, icmp error messages are sent with the primary address of
943		the exiting interface.
944	
945		If non-zero, the message will be sent with the primary address of
946		the interface that received the packet that caused the icmp error.
947		This is the behaviour network many administrators will expect from
948		a router. And it can make debugging complicated network layouts
949		much easier.
950	
951		Note that if no primary address exists for the interface selected,
952		then the primary address of the first non-loopback interface that
953		has one will be used regardless of this setting.
954	
955		Default: 0
956	
957	igmp_max_memberships - INTEGER
958		Change the maximum number of multicast groups we can subscribe to.
959		Default: 20
960	
961		Theoretical maximum value is bounded by having to send a membership
962		report in a single datagram (i.e. the report can't span multiple
963		datagrams, or risk confusing the switch and leaving groups you don't
964		intend to).
965	
966		The number of supported groups 'M' is bounded by the number of group
967		report entries you can fit into a single datagram of 65535 bytes.
968	
969		M = 65536-sizeof (ip header)/(sizeof(Group record))
970	
971		Group records are variable length, with a minimum of 12 bytes.
972		So net.ipv4.igmp_max_memberships should not be set higher than:
973	
974		(65536-24) / 12 = 5459
975	
976		The value 5459 assumes no IP header options, so in practice
977		this number may be lower.
978	
979	igmp_max_msf - INTEGER
980		Maximum number of addresses allowed in the source filter list for a
981		multicast group.
982		Default: 10
983	
984	igmp_qrv - INTEGER
985		Controls the IGMP query robustness variable (see RFC2236 8.1).
986		Default: 2 (as specified by RFC2236 8.1)
987		Minimum: 1 (as specified by RFC6636 4.5)
988	
989	force_igmp_version - INTEGER
990		0 - (default) No enforcement of a IGMP version, IGMPv1/v2 fallback
991		    allowed. Will back to IGMPv3 mode again if all IGMPv1/v2 Querier
992		    Present timer expires.
993		1 - Enforce to use IGMP version 1. Will also reply IGMPv1 report if
994		    receive IGMPv2/v3 query.
995		2 - Enforce to use IGMP version 2. Will fallback to IGMPv1 if receive
996		    IGMPv1 query message. Will reply report if receive IGMPv3 query.
997		3 - Enforce to use IGMP version 3. The same react with default 0.
998	
999		Note: this is not the same with force_mld_version because IGMPv3 RFC3376
1000		Security Considerations does not have clear description that we could
1001		ignore other version messages completely as MLDv2 RFC3810. So make
1002		this value as default 0 is recommended.
1003	
1004	conf/interface/*  changes special settings per interface (where
1005	"interface" is the name of your network interface)
1006	
1007	conf/all/*	  is special, changes the settings for all interfaces
1008	
1009	log_martians - BOOLEAN
1010		Log packets with impossible addresses to kernel log.
1011		log_martians for the interface will be enabled if at least one of
1012		conf/{all,interface}/log_martians is set to TRUE,
1013		it will be disabled otherwise
1014	
1015	accept_redirects - BOOLEAN
1016		Accept ICMP redirect messages.
1017		accept_redirects for the interface will be enabled if:
1018		- both conf/{all,interface}/accept_redirects are TRUE in the case
1019		  forwarding for the interface is enabled
1020		or
1021		- at least one of conf/{all,interface}/accept_redirects is TRUE in the
1022		  case forwarding for the interface is disabled
1023		accept_redirects for the interface will be disabled otherwise
1024		default TRUE (host)
1025			FALSE (router)
1026	
1027	forwarding - BOOLEAN
1028		Enable IP forwarding on this interface.  This controls whether packets
1029		received _on_ this interface can be forwarded.
1030	
1031	mc_forwarding - BOOLEAN
1032		Do multicast routing. The kernel needs to be compiled with CONFIG_MROUTE
1033		and a multicast routing daemon is required.
1034		conf/all/mc_forwarding must also be set to TRUE to enable multicast
1035		routing	for the interface
1036	
1037	medium_id - INTEGER
1038		Integer value used to differentiate the devices by the medium they
1039		are attached to. Two devices can have different id values when
1040		the broadcast packets are received only on one of them.
1041		The default value 0 means that the device is the only interface
1042		to its medium, value of -1 means that medium is not known.
1043	
1044		Currently, it is used to change the proxy_arp behavior:
1045		the proxy_arp feature is enabled for packets forwarded between
1046		two devices attached to different media.
1047	
1048	proxy_arp - BOOLEAN
1049		Do proxy arp.
1050		proxy_arp for the interface will be enabled if at least one of
1051		conf/{all,interface}/proxy_arp is set to TRUE,
1052		it will be disabled otherwise
1053	
1054	proxy_arp_pvlan - BOOLEAN
1055		Private VLAN proxy arp.
1056		Basically allow proxy arp replies back to the same interface
1057		(from which the ARP request/solicitation was received).
1058	
1059		This is done to support (ethernet) switch features, like RFC
1060		3069, where the individual ports are NOT allowed to
1061		communicate with each other, but they are allowed to talk to
1062		the upstream router.  As described in RFC 3069, it is possible
1063		to allow these hosts to communicate through the upstream
1064		router by proxy_arp'ing. Don't need to be used together with
1065		proxy_arp.
1066	
1067		This technology is known by different names:
1068		  In RFC 3069 it is called VLAN Aggregation.
1069		  Cisco and Allied Telesyn call it Private VLAN.
1070		  Hewlett-Packard call it Source-Port filtering or port-isolation.
1071		  Ericsson call it MAC-Forced Forwarding (RFC Draft).
1072	
1073	shared_media - BOOLEAN
1074		Send(router) or accept(host) RFC1620 shared media redirects.
1075		Overrides secure_redirects.
1076		shared_media for the interface will be enabled if at least one of
1077		conf/{all,interface}/shared_media is set to TRUE,
1078		it will be disabled otherwise
1079		default TRUE
1080	
1081	secure_redirects - BOOLEAN
1082		Accept ICMP redirect messages only to gateways listed in the
1083		interface's current gateway list. Even if disabled, RFC1122 redirect
1084		rules still apply.
1085		Overridden by shared_media.
1086		secure_redirects for the interface will be enabled if at least one of
1087		conf/{all,interface}/secure_redirects is set to TRUE,
1088		it will be disabled otherwise
1089		default TRUE
1090	
1091	send_redirects - BOOLEAN
1092		Send redirects, if router.
1093		send_redirects for the interface will be enabled if at least one of
1094		conf/{all,interface}/send_redirects is set to TRUE,
1095		it will be disabled otherwise
1096		Default: TRUE
1097	
1098	bootp_relay - BOOLEAN
1099		Accept packets with source address 0.b.c.d destined
1100		not to this host as local ones. It is supposed, that
1101		BOOTP relay daemon will catch and forward such packets.
1102		conf/all/bootp_relay must also be set to TRUE to enable BOOTP relay
1103		for the interface
1104		default FALSE
1105		Not Implemented Yet.
1106	
1107	accept_source_route - BOOLEAN
1108		Accept packets with SRR option.
1109		conf/all/accept_source_route must also be set to TRUE to accept packets
1110		with SRR option on the interface
1111		default TRUE (router)
1112			FALSE (host)
1113	
1114	accept_local - BOOLEAN
1115		Accept packets with local source addresses. In combination with
1116		suitable routing, this can be used to direct packets between two
1117		local interfaces over the wire and have them accepted properly.
1118		default FALSE
1119	
1120	route_localnet - BOOLEAN
1121		Do not consider loopback addresses as martian source or destination
1122		while routing. This enables the use of 127/8 for local routing purposes.
1123		default FALSE
1124	
1125	rp_filter - INTEGER
1126		0 - No source validation.
1127		1 - Strict mode as defined in RFC3704 Strict Reverse Path
1128		    Each incoming packet is tested against the FIB and if the interface
1129		    is not the best reverse path the packet check will fail.
1130		    By default failed packets are discarded.
1131		2 - Loose mode as defined in RFC3704 Loose Reverse Path
1132		    Each incoming packet's source address is also tested against the FIB
1133		    and if the source address is not reachable via any interface
1134		    the packet check will fail.
1135	
1136		Current recommended practice in RFC3704 is to enable strict mode
1137		to prevent IP spoofing from DDos attacks. If using asymmetric routing
1138		or other complicated routing, then loose mode is recommended.
1139	
1140		The max value from conf/{all,interface}/rp_filter is used
1141		when doing source validation on the {interface}.
1142	
1143		Default value is 0. Note that some distributions enable it
1144		in startup scripts.
1145	
1146	arp_filter - BOOLEAN
1147		1 - Allows you to have multiple network interfaces on the same
1148		subnet, and have the ARPs for each interface be answered
1149		based on whether or not the kernel would route a packet from
1150		the ARP'd IP out that interface (therefore you must use source
1151		based routing for this to work). In other words it allows control
1152		of which cards (usually 1) will respond to an arp request.
1153	
1154		0 - (default) The kernel can respond to arp requests with addresses
1155		from other interfaces. This may seem wrong but it usually makes
1156		sense, because it increases the chance of successful communication.
1157		IP addresses are owned by the complete host on Linux, not by
1158		particular interfaces. Only for more complex setups like load-
1159		balancing, does this behaviour cause problems.
1160	
1161		arp_filter for the interface will be enabled if at least one of
1162		conf/{all,interface}/arp_filter is set to TRUE,
1163		it will be disabled otherwise
1164	
1165	arp_announce - INTEGER
1166		Define different restriction levels for announcing the local
1167		source IP address from IP packets in ARP requests sent on
1168		interface:
1169		0 - (default) Use any local address, configured on any interface
1170		1 - Try to avoid local addresses that are not in the target's
1171		subnet for this interface. This mode is useful when target
1172		hosts reachable via this interface require the source IP
1173		address in ARP requests to be part of their logical network
1174		configured on the receiving interface. When we generate the
1175		request we will check all our subnets that include the
1176		target IP and will preserve the source address if it is from
1177		such subnet. If there is no such subnet we select source
1178		address according to the rules for level 2.
1179		2 - Always use the best local address for this target.
1180		In this mode we ignore the source address in the IP packet
1181		and try to select local address that we prefer for talks with
1182		the target host. Such local address is selected by looking
1183		for primary IP addresses on all our subnets on the outgoing
1184		interface that include the target IP address. If no suitable
1185		local address is found we select the first local address
1186		we have on the outgoing interface or on all other interfaces,
1187		with the hope we will receive reply for our request and
1188		even sometimes no matter the source IP address we announce.
1189	
1190		The max value from conf/{all,interface}/arp_announce is used.
1191	
1192		Increasing the restriction level gives more chance for
1193		receiving answer from the resolved target while decreasing
1194		the level announces more valid sender's information.
1195	
1196	arp_ignore - INTEGER
1197		Define different modes for sending replies in response to
1198		received ARP requests that resolve local target IP addresses:
1199		0 - (default): reply for any local target IP address, configured
1200		on any interface
1201		1 - reply only if the target IP address is local address
1202		configured on the incoming interface
1203		2 - reply only if the target IP address is local address
1204		configured on the incoming interface and both with the
1205		sender's IP address are part from same subnet on this interface
1206		3 - do not reply for local addresses configured with scope host,
1207		only resolutions for global and link addresses are replied
1208		4-7 - reserved
1209		8 - do not reply for all local addresses
1210	
1211		The max value from conf/{all,interface}/arp_ignore is used
1212		when ARP request is received on the {interface}
1213	
1214	arp_notify - BOOLEAN
1215		Define mode for notification of address and device changes.
1216		0 - (default): do nothing
1217		1 - Generate gratuitous arp requests when device is brought up
1218		    or hardware address changes.
1219	
1220	arp_accept - BOOLEAN
1221		Define behavior for gratuitous ARP frames who's IP is not
1222		already present in the ARP table:
1223		0 - don't create new entries in the ARP table
1224		1 - create new entries in the ARP table
1225	
1226		Both replies and requests type gratuitous arp will trigger the
1227		ARP table to be updated, if this setting is on.
1228	
1229		If the ARP table already contains the IP address of the
1230		gratuitous arp frame, the arp table will be updated regardless
1231		if this setting is on or off.
1232	
1233	mcast_solicit - INTEGER
1234		The maximum number of multicast probes in INCOMPLETE state,
1235		when the associated hardware address is unknown.  Defaults
1236		to 3.
1237	
1238	ucast_solicit - INTEGER
1239		The maximum number of unicast probes in PROBE state, when
1240		the hardware address is being reconfirmed.  Defaults to 3.
1241	
1242	app_solicit - INTEGER
1243		The maximum number of probes to send to the user space ARP daemon
1244		via netlink before dropping back to multicast probes (see
1245		mcast_resolicit).  Defaults to 0.
1246	
1247	mcast_resolicit - INTEGER
1248		The maximum number of multicast probes after unicast and
1249		app probes in PROBE state.  Defaults to 0.
1250	
1251	disable_policy - BOOLEAN
1252		Disable IPSEC policy (SPD) for this interface
1253	
1254	disable_xfrm - BOOLEAN
1255		Disable IPSEC encryption on this interface, whatever the policy
1256	
1257	igmpv2_unsolicited_report_interval - INTEGER
1258		The interval in milliseconds in which the next unsolicited
1259		IGMPv1 or IGMPv2 report retransmit will take place.
1260		Default: 10000 (10 seconds)
1261	
1262	igmpv3_unsolicited_report_interval - INTEGER
1263		The interval in milliseconds in which the next unsolicited
1264		IGMPv3 report retransmit will take place.
1265		Default: 1000 (1 seconds)
1266	
1267	promote_secondaries - BOOLEAN
1268		When a primary IP address is removed from this interface
1269		promote a corresponding secondary IP address instead of
1270		removing all the corresponding secondary IP addresses.
1271	
1272	drop_unicast_in_l2_multicast - BOOLEAN
1273		Drop any unicast IP packets that are received in link-layer
1274		multicast (or broadcast) frames.
1275		This behavior (for multicast) is actually a SHOULD in RFC
1276		1122, but is disabled by default for compatibility reasons.
1277		Default: off (0)
1278	
1279	drop_gratuitous_arp - BOOLEAN
1280		Drop all gratuitous ARP frames, for example if there's a known
1281		good ARP proxy on the network and such frames need not be used
1282		(or in the case of 802.11, must not be used to prevent attacks.)
1283		Default: off (0)
1284	
1285	
1286	tag - INTEGER
1287		Allows you to write a number, which can be used as required.
1288		Default value is 0.
1289	
1290	xfrm4_gc_thresh - INTEGER
1291		The threshold at which we will start garbage collecting for IPv4
1292		destination cache entries.  At twice this value the system will
1293		refuse new allocations.
1294	
1295	igmp_link_local_mcast_reports - BOOLEAN
1296		Enable IGMP reports for link local multicast groups in the
1297		224.0.0.X range.
1298		Default TRUE
1299	
1300	Alexey Kuznetsov.
1301	kuznet@ms2.inr.ac.ru
1302	
1303	Updated by:
1304	Andi Kleen
1305	ak@muc.de
1306	Nicolas Delon
1307	delon.nicolas@wanadoo.fr
1308	
1309	
1310	
1311	
1312	/proc/sys/net/ipv6/* Variables:
1313	
1314	IPv6 has no global variables such as tcp_*.  tcp_* settings under ipv4/ also
1315	apply to IPv6 [XXX?].
1316	
1317	bindv6only - BOOLEAN
1318		Default value for IPV6_V6ONLY socket option,
1319		which restricts use of the IPv6 socket to IPv6 communication
1320		only.
1321			TRUE: disable IPv4-mapped address feature
1322			FALSE: enable IPv4-mapped address feature
1323	
1324		Default: FALSE (as specified in RFC3493)
1325	
1326	flowlabel_consistency - BOOLEAN
1327		Protect the consistency (and unicity) of flow label.
1328		You have to disable it to use IPV6_FL_F_REFLECT flag on the
1329		flow label manager.
1330		TRUE: enabled
1331		FALSE: disabled
1332		Default: TRUE
1333	
1334	auto_flowlabels - INTEGER
1335		Automatically generate flow labels based on a flow hash of the
1336		packet. This allows intermediate devices, such as routers, to
1337		identify packet flows for mechanisms like Equal Cost Multipath
1338		Routing (see RFC 6438).
1339		0: automatic flow labels are completely disabled
1340		1: automatic flow labels are enabled by default, they can be
1341		   disabled on a per socket basis using the IPV6_AUTOFLOWLABEL
1342		   socket option
1343		2: automatic flow labels are allowed, they may be enabled on a
1344		   per socket basis using the IPV6_AUTOFLOWLABEL socket option
1345		3: automatic flow labels are enabled and enforced, they cannot
1346		   be disabled by the socket option
1347		Default: 1
1348	
1349	flowlabel_state_ranges - BOOLEAN
1350		Split the flow label number space into two ranges. 0-0x7FFFF is
1351		reserved for the IPv6 flow manager facility, 0x80000-0xFFFFF
1352		is reserved for stateless flow labels as described in RFC6437.
1353		TRUE: enabled
1354		FALSE: disabled
1355		Default: true
1356	
1357	flowlabel_reflect - BOOLEAN
1358		Automatically reflect the flow label. Needed for Path MTU
1359		Discovery to work with Equal Cost Multipath Routing in anycast
1360		environments. See RFC 7690 and:
1361		https://tools.ietf.org/html/draft-wang-6man-flow-label-reflection-01
1362		TRUE: enabled
1363		FALSE: disabled
1364		Default: FALSE
1365	
1366	anycast_src_echo_reply - BOOLEAN
1367		Controls the use of anycast addresses as source addresses for ICMPv6
1368		echo reply
1369		TRUE:  enabled
1370		FALSE: disabled
1371		Default: FALSE
1372	
1373	idgen_delay - INTEGER
1374		Controls the delay in seconds after which time to retry
1375		privacy stable address generation if a DAD conflict is
1376		detected.
1377		Default: 1 (as specified in RFC7217)
1378	
1379	idgen_retries - INTEGER
1380		Controls the number of retries to generate a stable privacy
1381		address if a DAD conflict is detected.
1382		Default: 3 (as specified in RFC7217)
1383	
1384	mld_qrv - INTEGER
1385		Controls the MLD query robustness variable (see RFC3810 9.1).
1386		Default: 2 (as specified by RFC3810 9.1)
1387		Minimum: 1 (as specified by RFC6636 4.5)
1388	
1389	max_dst_opts_cnt - INTEGER
1390		Maximum number of non-padding TLVs allowed in a Destination
1391		options extension header. If this value is less than zero
1392		then unknown options are disallowed and the number of known
1393		TLVs allowed is the absolute value of this number.
1394		Default: 8
1395	
1396	max_hbh_opts_cnt - INTEGER
1397		Maximum number of non-padding TLVs allowed in a Hop-by-Hop
1398		options extension header. If this value is less than zero
1399		then unknown options are disallowed and the number of known
1400		TLVs allowed is the absolute value of this number.
1401		Default: 8
1402	
1403	max dst_opts_len - INTEGER
1404		Maximum length allowed for a Destination options extension
1405		header.
1406		Default: INT_MAX (unlimited)
1407	
1408	max hbh_opts_len - INTEGER
1409		Maximum length allowed for a Hop-by-Hop options extension
1410		header.
1411		Default: INT_MAX (unlimited)
1412	
1413	IPv6 Fragmentation:
1414	
1415	ip6frag_high_thresh - INTEGER
1416		Maximum memory used to reassemble IPv6 fragments. When
1417		ip6frag_high_thresh bytes of memory is allocated for this purpose,
1418		the fragment handler will toss packets until ip6frag_low_thresh
1419		is reached.
1420	
1421	ip6frag_low_thresh - INTEGER
1422		See ip6frag_high_thresh
1423	
1424	ip6frag_time - INTEGER
1425		Time in seconds to keep an IPv6 fragment in memory.
1426	
1427	conf/default/*:
1428		Change the interface-specific default settings.
1429	
1430	
1431	conf/all/*:
1432		Change all the interface-specific settings.
1433	
1434		[XXX:  Other special features than forwarding?]
1435	
1436	conf/all/forwarding - BOOLEAN
1437		Enable global IPv6 forwarding between all interfaces.
1438	
1439		IPv4 and IPv6 work differently here; e.g. netfilter must be used
1440		to control which interfaces may forward packets and which not.
1441	
1442		This also sets all interfaces' Host/Router setting
1443		'forwarding' to the specified value.  See below for details.
1444	
1445		This referred to as global forwarding.
1446	
1447	proxy_ndp - BOOLEAN
1448		Do proxy ndp.
1449	
1450	fwmark_reflect - BOOLEAN
1451		Controls the fwmark of kernel-generated IPv6 reply packets that are not
1452		associated with a socket for example, TCP RSTs or ICMPv6 echo replies).
1453		If unset, these packets have a fwmark of zero. If set, they have the
1454		fwmark of the packet they are replying to.
1455		Default: 0
1456	
1457	conf/interface/*:
1458		Change special settings per interface.
1459	
1460		The functional behaviour for certain settings is different
1461		depending on whether local forwarding is enabled or not.
1462	
1463	accept_ra - INTEGER
1464		Accept Router Advertisements; autoconfigure using them.
1465	
1466		It also determines whether or not to transmit Router
1467		Solicitations. If and only if the functional setting is to
1468		accept Router Advertisements, Router Solicitations will be
1469		transmitted.
1470	
1471		Possible values are:
1472			0 Do not accept Router Advertisements.
1473			1 Accept Router Advertisements if forwarding is disabled.
1474			2 Overrule forwarding behaviour. Accept Router Advertisements
1475			  even if forwarding is enabled.
1476	
1477		Functional default: enabled if local forwarding is disabled.
1478				    disabled if local forwarding is enabled.
1479	
1480	accept_ra_defrtr - BOOLEAN
1481		Learn default router in Router Advertisement.
1482	
1483		Functional default: enabled if accept_ra is enabled.
1484				    disabled if accept_ra is disabled.
1485	
1486	accept_ra_from_local - BOOLEAN
1487		Accept RA with source-address that is found on local machine
1488	        if the RA is otherwise proper and able to be accepted.
1489	        Default is to NOT accept these as it may be an un-intended
1490	        network loop.
1491	
1492		Functional default:
1493	           enabled if accept_ra_from_local is enabled
1494	               on a specific interface.
1495		   disabled if accept_ra_from_local is disabled
1496	               on a specific interface.
1497	
1498	accept_ra_min_hop_limit - INTEGER
1499		Minimum hop limit Information in Router Advertisement.
1500	
1501		Hop limit Information in Router Advertisement less than this
1502		variable shall be ignored.
1503	
1504		Default: 1
1505	
1506	accept_ra_pinfo - BOOLEAN
1507		Learn Prefix Information in Router Advertisement.
1508	
1509		Functional default: enabled if accept_ra is enabled.
1510				    disabled if accept_ra is disabled.
1511	
1512	accept_ra_rt_info_min_plen - INTEGER
1513		Minimum prefix length of Route Information in RA.
1514	
1515		Route Information w/ prefix smaller than this variable shall
1516		be ignored.
1517	
1518		Functional default: 0 if accept_ra_rtr_pref is enabled.
1519				    -1 if accept_ra_rtr_pref is disabled.
1520	
1521	accept_ra_rt_info_max_plen - INTEGER
1522		Maximum prefix length of Route Information in RA.
1523	
1524		Route Information w/ prefix larger than this variable shall
1525		be ignored.
1526	
1527		Functional default: 0 if accept_ra_rtr_pref is enabled.
1528				    -1 if accept_ra_rtr_pref is disabled.
1529	
1530	accept_ra_rtr_pref - BOOLEAN
1531		Accept Router Preference in RA.
1532	
1533		Functional default: enabled if accept_ra is enabled.
1534				    disabled if accept_ra is disabled.
1535	
1536	accept_ra_mtu - BOOLEAN
1537		Apply the MTU value specified in RA option 5 (RFC4861). If
1538		disabled, the MTU specified in the RA will be ignored.
1539	
1540		Functional default: enabled if accept_ra is enabled.
1541				    disabled if accept_ra is disabled.
1542	
1543	accept_redirects - BOOLEAN
1544		Accept Redirects.
1545	
1546		Functional default: enabled if local forwarding is disabled.
1547				    disabled if local forwarding is enabled.
1548	
1549	accept_source_route - INTEGER
1550		Accept source routing (routing extension header).
1551	
1552		>= 0: Accept only routing header type 2.
1553		< 0: Do not accept routing header.
1554	
1555		Default: 0
1556	
1557	autoconf - BOOLEAN
1558		Autoconfigure addresses using Prefix Information in Router
1559		Advertisements.
1560	
1561		Functional default: enabled if accept_ra_pinfo is enabled.
1562				    disabled if accept_ra_pinfo is disabled.
1563	
1564	dad_transmits - INTEGER
1565		The amount of Duplicate Address Detection probes to send.
1566		Default: 1
1567	
1568	forwarding - INTEGER
1569		Configure interface-specific Host/Router behaviour.
1570	
1571		Note: It is recommended to have the same setting on all
1572		interfaces; mixed router/host scenarios are rather uncommon.
1573	
1574		Possible values are:
1575			0 Forwarding disabled
1576			1 Forwarding enabled
1577	
1578		FALSE (0):
1579	
1580		By default, Host behaviour is assumed.  This means:
1581	
1582		1. IsRouter flag is not set in Neighbour Advertisements.
1583		2. If accept_ra is TRUE (default), transmit Router
1584		   Solicitations.
1585		3. If accept_ra is TRUE (default), accept Router
1586		   Advertisements (and do autoconfiguration).
1587		4. If accept_redirects is TRUE (default), accept Redirects.
1588	
1589		TRUE (1):
1590	
1591		If local forwarding is enabled, Router behaviour is assumed.
1592		This means exactly the reverse from the above:
1593	
1594		1. IsRouter flag is set in Neighbour Advertisements.
1595		2. Router Solicitations are not sent unless accept_ra is 2.
1596		3. Router Advertisements are ignored unless accept_ra is 2.
1597		4. Redirects are ignored.
1598	
1599		Default: 0 (disabled) if global forwarding is disabled (default),
1600			 otherwise 1 (enabled).
1601	
1602	hop_limit - INTEGER
1603		Default Hop Limit to set.
1604		Default: 64
1605	
1606	mtu - INTEGER
1607		Default Maximum Transfer Unit
1608		Default: 1280 (IPv6 required minimum)
1609	
1610	ip_nonlocal_bind - BOOLEAN
1611		If set, allows processes to bind() to non-local IPv6 addresses,
1612		which can be quite useful - but may break some applications.
1613		Default: 0
1614	
1615	router_probe_interval - INTEGER
1616		Minimum interval (in seconds) between Router Probing described
1617		in RFC4191.
1618	
1619		Default: 60
1620	
1621	router_solicitation_delay - INTEGER
1622		Number of seconds to wait after interface is brought up
1623		before sending Router Solicitations.
1624		Default: 1
1625	
1626	router_solicitation_interval - INTEGER
1627		Number of seconds to wait between Router Solicitations.
1628		Default: 4
1629	
1630	router_solicitations - INTEGER
1631		Number of Router Solicitations to send until assuming no
1632		routers are present.
1633		Default: 3
1634	
1635	use_oif_addrs_only - BOOLEAN
1636		When enabled, the candidate source addresses for destinations
1637		routed via this interface are restricted to the set of addresses
1638		configured on this interface (vis. RFC 6724, section 4).
1639	
1640		Default: false
1641	
1642	use_tempaddr - INTEGER
1643		Preference for Privacy Extensions (RFC3041).
1644		  <= 0 : disable Privacy Extensions
1645		  == 1 : enable Privacy Extensions, but prefer public
1646		         addresses over temporary addresses.
1647		  >  1 : enable Privacy Extensions and prefer temporary
1648		         addresses over public addresses.
1649		Default:  0 (for most devices)
1650			 -1 (for point-to-point devices and loopback devices)
1651	
1652	temp_valid_lft - INTEGER
1653		valid lifetime (in seconds) for temporary addresses.
1654		Default: 604800 (7 days)
1655	
1656	temp_prefered_lft - INTEGER
1657		Preferred lifetime (in seconds) for temporary addresses.
1658		Default: 86400 (1 day)
1659	
1660	keep_addr_on_down - INTEGER
1661		Keep all IPv6 addresses on an interface down event. If set static
1662		global addresses with no expiration time are not flushed.
1663		  >0 : enabled
1664		   0 : system default
1665		  <0 : disabled
1666	
1667		Default: 0 (addresses are removed)
1668	
1669	max_desync_factor - INTEGER
1670		Maximum value for DESYNC_FACTOR, which is a random value
1671		that ensures that clients don't synchronize with each
1672		other and generate new addresses at exactly the same time.
1673		value is in seconds.
1674		Default: 600
1675	
1676	regen_max_retry - INTEGER
1677		Number of attempts before give up attempting to generate
1678		valid temporary addresses.
1679		Default: 5
1680	
1681	max_addresses - INTEGER
1682		Maximum number of autoconfigured addresses per interface.  Setting
1683		to zero disables the limitation.  It is not recommended to set this
1684		value too large (or to zero) because it would be an easy way to
1685		crash the kernel by allowing too many addresses to be created.
1686		Default: 16
1687	
1688	disable_ipv6 - BOOLEAN
1689		Disable IPv6 operation.  If accept_dad is set to 2, this value
1690		will be dynamically set to TRUE if DAD fails for the link-local
1691		address.
1692		Default: FALSE (enable IPv6 operation)
1693	
1694		When this value is changed from 1 to 0 (IPv6 is being enabled),
1695		it will dynamically create a link-local address on the given
1696		interface and start Duplicate Address Detection, if necessary.
1697	
1698		When this value is changed from 0 to 1 (IPv6 is being disabled),
1699		it will dynamically delete all address on the given interface.
1700	
1701	accept_dad - INTEGER
1702		Whether to accept DAD (Duplicate Address Detection).
1703		0: Disable DAD
1704		1: Enable DAD (default)
1705		2: Enable DAD, and disable IPv6 operation if MAC-based duplicate
1706		   link-local address has been found.
1707	
1708		DAD operation and mode on a given interface will be selected according
1709		to the maximum value of conf/{all,interface}/accept_dad.
1710	
1711	force_tllao - BOOLEAN
1712		Enable sending the target link-layer address option even when
1713		responding to a unicast neighbor solicitation.
1714		Default: FALSE
1715	
1716		Quoting from RFC 2461, section 4.4, Target link-layer address:
1717	
1718		"The option MUST be included for multicast solicitations in order to
1719		avoid infinite Neighbor Solicitation "recursion" when the peer node
1720		does not have a cache entry to return a Neighbor Advertisements
1721		message.  When responding to unicast solicitations, the option can be
1722		omitted since the sender of the solicitation has the correct link-
1723		layer address; otherwise it would not have be able to send the unicast
1724		solicitation in the first place. However, including the link-layer
1725		address in this case adds little overhead and eliminates a potential
1726		race condition where the sender deletes the cached link-layer address
1727		prior to receiving a response to a previous solicitation."
1728	
1729	ndisc_notify - BOOLEAN
1730		Define mode for notification of address and device changes.
1731		0 - (default): do nothing
1732		1 - Generate unsolicited neighbour advertisements when device is brought
1733		    up or hardware address changes.
1734	
1735	ndisc_tclass - INTEGER
1736		The IPv6 Traffic Class to use by default when sending IPv6 Neighbor
1737		Discovery (Router Solicitation, Router Advertisement, Neighbor
1738		Solicitation, Neighbor Advertisement, Redirect) messages.
1739		These 8 bits can be interpreted as 6 high order bits holding the DSCP
1740		value and 2 low order bits representing ECN (which you probably want
1741		to leave cleared).
1742		0 - (default)
1743	
1744	mldv1_unsolicited_report_interval - INTEGER
1745		The interval in milliseconds in which the next unsolicited
1746		MLDv1 report retransmit will take place.
1747		Default: 10000 (10 seconds)
1748	
1749	mldv2_unsolicited_report_interval - INTEGER
1750		The interval in milliseconds in which the next unsolicited
1751		MLDv2 report retransmit will take place.
1752		Default: 1000 (1 second)
1753	
1754	force_mld_version - INTEGER
1755		0 - (default) No enforcement of a MLD version, MLDv1 fallback allowed
1756		1 - Enforce to use MLD version 1
1757		2 - Enforce to use MLD version 2
1758	
1759	suppress_frag_ndisc - INTEGER
1760		Control RFC 6980 (Security Implications of IPv6 Fragmentation
1761		with IPv6 Neighbor Discovery) behavior:
1762		1 - (default) discard fragmented neighbor discovery packets
1763		0 - allow fragmented neighbor discovery packets
1764	
1765	optimistic_dad - BOOLEAN
1766		Whether to perform Optimistic Duplicate Address Detection (RFC 4429).
1767		0: disabled (default)
1768		1: enabled
1769	
1770		Optimistic Duplicate Address Detection for the interface will be enabled
1771		if at least one of conf/{all,interface}/optimistic_dad is set to 1,
1772		it will be disabled otherwise.
1773	
1774	use_optimistic - BOOLEAN
1775		If enabled, do not classify optimistic addresses as deprecated during
1776		source address selection.  Preferred addresses will still be chosen
1777		before optimistic addresses, subject to other ranking in the source
1778		address selection algorithm.
1779		0: disabled (default)
1780		1: enabled
1781	
1782		This will be enabled if at least one of
1783		conf/{all,interface}/use_optimistic is set to 1, disabled otherwise.
1784	
1785	stable_secret - IPv6 address
1786		This IPv6 address will be used as a secret to generate IPv6
1787		addresses for link-local addresses and autoconfigured
1788		ones. All addresses generated after setting this secret will
1789		be stable privacy ones by default. This can be changed via the
1790		addrgenmode ip-link. conf/default/stable_secret is used as the
1791		secret for the namespace, the interface specific ones can
1792		overwrite that. Writes to conf/all/stable_secret are refused.
1793	
1794		It is recommended to generate this secret during installation
1795		of a system and keep it stable after that.
1796	
1797		By default the stable secret is unset.
1798	
1799	drop_unicast_in_l2_multicast - BOOLEAN
1800		Drop any unicast IPv6 packets that are received in link-layer
1801		multicast (or broadcast) frames.
1802	
1803		By default this is turned off.
1804	
1805	drop_unsolicited_na - BOOLEAN
1806		Drop all unsolicited neighbor advertisements, for example if there's
1807		a known good NA proxy on the network and such frames need not be used
1808		(or in the case of 802.11, must not be used to prevent attacks.)
1809	
1810		By default this is turned off.
1811	
1812	enhanced_dad - BOOLEAN
1813		Include a nonce option in the IPv6 neighbor solicitation messages used for
1814		duplicate address detection per RFC7527. A received DAD NS will only signal
1815		a duplicate address if the nonce is different. This avoids any false
1816		detection of duplicates due to loopback of the NS messages that we send.
1817		The nonce option will be sent on an interface unless both of
1818		conf/{all,interface}/enhanced_dad are set to FALSE.
1819		Default: TRUE
1820	
1821	icmp/*:
1822	ratelimit - INTEGER
1823		Limit the maximal rates for sending ICMPv6 packets.
1824		0 to disable any limiting,
1825		otherwise the minimal space between responses in milliseconds.
1826		Default: 1000
1827	
1828	xfrm6_gc_thresh - INTEGER
1829		The threshold at which we will start garbage collecting for IPv6
1830		destination cache entries.  At twice this value the system will
1831		refuse new allocations.
1832	
1833	
1834	IPv6 Update by:
1835	Pekka Savola <pekkas@netcore.fi>
1836	YOSHIFUJI Hideaki / USAGI Project <yoshfuji@linux-ipv6.org>
1837	
1838	
1839	/proc/sys/net/bridge/* Variables:
1840	
1841	bridge-nf-call-arptables - BOOLEAN
1842		1 : pass bridged ARP traffic to arptables' FORWARD chain.
1843		0 : disable this.
1844		Default: 1
1845	
1846	bridge-nf-call-iptables - BOOLEAN
1847		1 : pass bridged IPv4 traffic to iptables' chains.
1848		0 : disable this.
1849		Default: 1
1850	
1851	bridge-nf-call-ip6tables - BOOLEAN
1852		1 : pass bridged IPv6 traffic to ip6tables' chains.
1853		0 : disable this.
1854		Default: 1
1855	
1856	bridge-nf-filter-vlan-tagged - BOOLEAN
1857		1 : pass bridged vlan-tagged ARP/IP/IPv6 traffic to {arp,ip,ip6}tables.
1858		0 : disable this.
1859		Default: 0
1860	
1861	bridge-nf-filter-pppoe-tagged - BOOLEAN
1862		1 : pass bridged pppoe-tagged IP/IPv6 traffic to {ip,ip6}tables.
1863		0 : disable this.
1864		Default: 0
1865	
1866	bridge-nf-pass-vlan-input-dev - BOOLEAN
1867		1: if bridge-nf-filter-vlan-tagged is enabled, try to find a vlan
1868		interface on the bridge and set the netfilter input device to the vlan.
1869		This allows use of e.g. "iptables -i br0.1" and makes the REDIRECT
1870		target work with vlan-on-top-of-bridge interfaces.  When no matching
1871		vlan interface is found, or this switch is off, the input device is
1872		set to the bridge interface.
1873		0: disable bridge netfilter vlan interface lookup.
1874		Default: 0
1875	
1876	proc/sys/net/sctp/* Variables:
1877	
1878	addip_enable - BOOLEAN
1879		Enable or disable extension of  Dynamic Address Reconfiguration
1880		(ADD-IP) functionality specified in RFC5061.  This extension provides
1881		the ability to dynamically add and remove new addresses for the SCTP
1882		associations.
1883	
1884		1: Enable extension.
1885	
1886		0: Disable extension.
1887	
1888		Default: 0
1889	
1890	pf_enable - INTEGER
1891		Enable or disable pf (pf is short for potentially failed) state. A value
1892		of pf_retrans > path_max_retrans also disables pf state. That is, one of
1893		both pf_enable and pf_retrans > path_max_retrans can disable pf state.
1894		Since pf_retrans and path_max_retrans can be changed by userspace
1895		application, sometimes user expects to disable pf state by the value of
1896		pf_retrans > path_max_retrans, but occasionally the value of pf_retrans
1897		or path_max_retrans is changed by the user application, this pf state is
1898		enabled. As such, it is necessary to add this to dynamically enable
1899		and disable pf state. See:
1900		https://datatracker.ietf.org/doc/draft-ietf-tsvwg-sctp-failover for
1901		details.
1902	
1903		1: Enable pf.
1904	
1905		0: Disable pf.
1906	
1907		Default: 1
1908	
1909	addip_noauth_enable - BOOLEAN
1910		Dynamic Address Reconfiguration (ADD-IP) requires the use of
1911		authentication to protect the operations of adding or removing new
1912		addresses.  This requirement is mandated so that unauthorized hosts
1913		would not be able to hijack associations.  However, older
1914		implementations may not have implemented this requirement while
1915		allowing the ADD-IP extension.  For reasons of interoperability,
1916		we provide this variable to control the enforcement of the
1917		authentication requirement.
1918	
1919		1: Allow ADD-IP extension to be used without authentication.  This
1920		   should only be set in a closed environment for interoperability
1921		   with older implementations.
1922	
1923		0: Enforce the authentication requirement
1924	
1925		Default: 0
1926	
1927	auth_enable - BOOLEAN
1928		Enable or disable Authenticated Chunks extension.  This extension
1929		provides the ability to send and receive authenticated chunks and is
1930		required for secure operation of Dynamic Address Reconfiguration
1931		(ADD-IP) extension.
1932	
1933		1: Enable this extension.
1934		0: Disable this extension.
1935	
1936		Default: 0
1937	
1938	prsctp_enable - BOOLEAN
1939		Enable or disable the Partial Reliability extension (RFC3758) which
1940		is used to notify peers that a given DATA should no longer be expected.
1941	
1942		1: Enable extension
1943		0: Disable
1944	
1945		Default: 1
1946	
1947	max_burst - INTEGER
1948		The limit of the number of new packets that can be initially sent.  It
1949		controls how bursty the generated traffic can be.
1950	
1951		Default: 4
1952	
1953	association_max_retrans - INTEGER
1954		Set the maximum number for retransmissions that an association can
1955		attempt deciding that the remote end is unreachable.  If this value
1956		is exceeded, the association is terminated.
1957	
1958		Default: 10
1959	
1960	max_init_retransmits - INTEGER
1961		The maximum number of retransmissions of INIT and COOKIE-ECHO chunks
1962		that an association will attempt before declaring the destination
1963		unreachable and terminating.
1964	
1965		Default: 8
1966	
1967	path_max_retrans - INTEGER
1968		The maximum number of retransmissions that will be attempted on a given
1969		path.  Once this threshold is exceeded, the path is considered
1970		unreachable, and new traffic will use a different path when the
1971		association is multihomed.
1972	
1973		Default: 5
1974	
1975	pf_retrans - INTEGER
1976		The number of retransmissions that will be attempted on a given path
1977		before traffic is redirected to an alternate transport (should one
1978		exist).  Note this is distinct from path_max_retrans, as a path that
1979		passes the pf_retrans threshold can still be used.  Its only
1980		deprioritized when a transmission path is selected by the stack.  This
1981		setting is primarily used to enable fast failover mechanisms without
1982		having to reduce path_max_retrans to a very low value.  See:
1983		http://www.ietf.org/id/draft-nishida-tsvwg-sctp-failover-05.txt
1984		for details.  Note also that a value of pf_retrans > path_max_retrans
1985		disables this feature. Since both pf_retrans and path_max_retrans can
1986		be changed by userspace application, a variable pf_enable is used to
1987		disable pf state.
1988	
1989		Default: 0
1990	
1991	rto_initial - INTEGER
1992		The initial round trip timeout value in milliseconds that will be used
1993		in calculating round trip times.  This is the initial time interval
1994		for retransmissions.
1995	
1996		Default: 3000
1997	
1998	rto_max - INTEGER
1999		The maximum value (in milliseconds) of the round trip timeout.  This
2000		is the largest time interval that can elapse between retransmissions.
2001	
2002		Default: 60000
2003	
2004	rto_min - INTEGER
2005		The minimum value (in milliseconds) of the round trip timeout.  This
2006		is the smallest time interval the can elapse between retransmissions.
2007	
2008		Default: 1000
2009	
2010	hb_interval - INTEGER
2011		The interval (in milliseconds) between HEARTBEAT chunks.  These chunks
2012		are sent at the specified interval on idle paths to probe the state of
2013		a given path between 2 associations.
2014	
2015		Default: 30000
2016	
2017	sack_timeout - INTEGER
2018		The amount of time (in milliseconds) that the implementation will wait
2019		to send a SACK.
2020	
2021		Default: 200
2022	
2023	valid_cookie_life - INTEGER
2024		The default lifetime of the SCTP cookie (in milliseconds).  The cookie
2025		is used during association establishment.
2026	
2027		Default: 60000
2028	
2029	cookie_preserve_enable - BOOLEAN
2030		Enable or disable the ability to extend the lifetime of the SCTP cookie
2031		that is used during the establishment phase of SCTP association
2032	
2033		1: Enable cookie lifetime extension.
2034		0: Disable
2035	
2036		Default: 1
2037	
2038	cookie_hmac_alg - STRING
2039		Select the hmac algorithm used when generating the cookie value sent by
2040		a listening sctp socket to a connecting client in the INIT-ACK chunk.
2041		Valid values are:
2042		* md5
2043		* sha1
2044		* none
2045		Ability to assign md5 or sha1 as the selected alg is predicated on the
2046		configuration of those algorithms at build time (CONFIG_CRYPTO_MD5 and
2047		CONFIG_CRYPTO_SHA1).
2048	
2049		Default: Dependent on configuration.  MD5 if available, else SHA1 if
2050		available, else none.
2051	
2052	rcvbuf_policy - INTEGER
2053		Determines if the receive buffer is attributed to the socket or to
2054		association.   SCTP supports the capability to create multiple
2055		associations on a single socket.  When using this capability, it is
2056		possible that a single stalled association that's buffering a lot
2057		of data may block other associations from delivering their data by
2058		consuming all of the receive buffer space.  To work around this,
2059		the rcvbuf_policy could be set to attribute the receiver buffer space
2060		to each association instead of the socket.  This prevents the described
2061		blocking.
2062	
2063		1: rcvbuf space is per association
2064		0: rcvbuf space is per socket
2065	
2066		Default: 0
2067	
2068	sndbuf_policy - INTEGER
2069		Similar to rcvbuf_policy above, this applies to send buffer space.
2070	
2071		1: Send buffer is tracked per association
2072		0: Send buffer is tracked per socket.
2073	
2074		Default: 0
2075	
2076	sctp_mem - vector of 3 INTEGERs: min, pressure, max
2077		Number of pages allowed for queueing by all SCTP sockets.
2078	
2079		min: Below this number of pages SCTP is not bothered about its
2080		memory appetite. When amount of memory allocated by SCTP exceeds
2081		this number, SCTP starts to moderate memory usage.
2082	
2083		pressure: This value was introduced to follow format of tcp_mem.
2084	
2085		max: Number of pages allowed for queueing by all SCTP sockets.
2086	
2087		Default is calculated at boot time from amount of available memory.
2088	
2089	sctp_rmem - vector of 3 INTEGERs: min, default, max
2090		Only the first value ("min") is used, "default" and "max" are
2091		ignored.
2092	
2093		min: Minimal size of receive buffer used by SCTP socket.
2094		It is guaranteed to each SCTP socket (but not association) even
2095		under moderate memory pressure.
2096	
2097		Default: 1 page
2098	
2099	sctp_wmem  - vector of 3 INTEGERs: min, default, max
2100		Currently this tunable has no effect.
2101	
2102	addr_scope_policy - INTEGER
2103		Control IPv4 address scoping - draft-stewart-tsvwg-sctp-ipv4-00
2104	
2105		0   - Disable IPv4 address scoping
2106		1   - Enable IPv4 address scoping
2107		2   - Follow draft but allow IPv4 private addresses
2108		3   - Follow draft but allow IPv4 link local addresses
2109	
2110		Default: 1
2111	
2112	
2113	/proc/sys/net/core/*
2114		Please see: Documentation/sysctl/net.txt for descriptions of these entries.
2115	
2116	
2117	/proc/sys/net/unix/*
2118	max_dgram_qlen - INTEGER
2119		The maximum length of dgram socket receive queue
2120	
2121		Default: 10
2122	
2123	
2124	UNDOCUMENTED:
2125	
2126	/proc/sys/net/irda/*
2127		fast_poll_increase FIXME
2128		warn_noreply_time FIXME
2129		discovery_slots FIXME
2130		slot_timeout FIXME
2131		max_baud_rate FIXME
2132		discovery_timeout FIXME
2133		lap_keepalive_time FIXME
2134		max_noreply_time FIXME
2135		max_tx_data_size FIXME
2136		max_tx_window FIXME
2137		min_tx_turn_time FIXME
Hide Line Numbers
About Kernel Documentation Linux Kernel Contact Linux Resources Linux Blog

Information is copyright its respective author. All material is available from the Linux Kernel Source distributed under a GPL License. This page is provided as a free service by mjmwired.net.