Commit | Line | Data |
---|---|---|
1da177e4 LT |
1 | /proc/sys/net/ipv4/* Variables: |
2 | ||
3 | ip_forward - BOOLEAN | |
4 | 0 - disabled (default) | |
e18f5feb | 5 | not 0 - enabled |
1da177e4 LT |
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 | |
cc6f02dd ED |
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) | |
1da177e4 LT |
17 | |
18 | ip_no_pmtu_disc - BOOLEAN | |
19 | Disable Path MTU Discovery. | |
20 | default FALSE | |
21 | ||
22 | min_pmtu - INTEGER | |
20db93c3 | 23 | default 552 - minimum discovered Path MTU |
1da177e4 | 24 | |
cbaf087a BG |
25 | route/max_size - INTEGER |
26 | Maximum number of routes allowed in the kernel. Increase | |
27 | this when using large numbers of interfaces and/or routes. | |
28 | ||
29 | neigh/default/gc_thresh3 - INTEGER | |
30 | Maximum number of neighbor entries allowed. Increase this | |
31 | when using large numbers of interfaces and when communicating | |
32 | with large numbers of directly-connected peers. | |
33 | ||
8b5c171b ED |
34 | neigh/default/unres_qlen_bytes - INTEGER |
35 | The maximum number of bytes which may be used by packets | |
36 | queued for each unresolved address by other network layers. | |
37 | (added in linux 3.3) | |
38 | ||
39 | neigh/default/unres_qlen - INTEGER | |
40 | The maximum number of packets which may be queued for each | |
41 | unresolved address by other network layers. | |
42 | (deprecated in linux 3.3) : use unres_qlen_bytes instead. | |
43 | ||
1da177e4 LT |
44 | mtu_expires - INTEGER |
45 | Time, in seconds, that cached PMTU information is kept. | |
46 | ||
47 | min_adv_mss - INTEGER | |
48 | The advertised MSS depends on the first hop route MTU, but will | |
49 | never be lower than this setting. | |
50 | ||
1080d709 NH |
51 | rt_cache_rebuild_count - INTEGER |
52 | The per net-namespace route cache emergency rebuild threshold. | |
53 | Any net-namespace having its route cache rebuilt due to | |
54 | a hash bucket chain being too long more than this many times | |
55 | will have its route caching disabled | |
56 | ||
1da177e4 LT |
57 | IP Fragmentation: |
58 | ||
59 | ipfrag_high_thresh - INTEGER | |
e18f5feb | 60 | Maximum memory used to reassemble IP fragments. When |
1da177e4 LT |
61 | ipfrag_high_thresh bytes of memory is allocated for this purpose, |
62 | the fragment handler will toss packets until ipfrag_low_thresh | |
63 | is reached. | |
e18f5feb | 64 | |
1da177e4 | 65 | ipfrag_low_thresh - INTEGER |
e18f5feb | 66 | See ipfrag_high_thresh |
1da177e4 LT |
67 | |
68 | ipfrag_time - INTEGER | |
e18f5feb | 69 | Time in seconds to keep an IP fragment in memory. |
1da177e4 LT |
70 | |
71 | ipfrag_secret_interval - INTEGER | |
e18f5feb | 72 | Regeneration interval (in seconds) of the hash secret (or lifetime |
1da177e4 LT |
73 | for the hash secret) for IP fragments. |
74 | Default: 600 | |
75 | ||
89cee8b1 | 76 | ipfrag_max_dist - INTEGER |
e18f5feb JDB |
77 | ipfrag_max_dist is a non-negative integer value which defines the |
78 | maximum "disorder" which is allowed among fragments which share a | |
79 | common IP source address. Note that reordering of packets is | |
80 | not unusual, but if a large number of fragments arrive from a source | |
81 | IP address while a particular fragment queue remains incomplete, it | |
82 | probably indicates that one or more fragments belonging to that queue | |
83 | have been lost. When ipfrag_max_dist is positive, an additional check | |
84 | is done on fragments before they are added to a reassembly queue - if | |
85 | ipfrag_max_dist (or more) fragments have arrived from a particular IP | |
86 | address between additions to any IP fragment queue using that source | |
87 | address, it's presumed that one or more fragments in the queue are | |
88 | lost. The existing fragment queue will be dropped, and a new one | |
89cee8b1 HX |
89 | started. An ipfrag_max_dist value of zero disables this check. |
90 | ||
91 | Using a very small value, e.g. 1 or 2, for ipfrag_max_dist can | |
92 | result in unnecessarily dropping fragment queues when normal | |
e18f5feb JDB |
93 | reordering of packets occurs, which could lead to poor application |
94 | performance. Using a very large value, e.g. 50000, increases the | |
95 | likelihood of incorrectly reassembling IP fragments that originate | |
89cee8b1 HX |
96 | from different IP datagrams, which could result in data corruption. |
97 | Default: 64 | |
98 | ||
1da177e4 LT |
99 | INET peer storage: |
100 | ||
101 | inet_peer_threshold - INTEGER | |
e18f5feb | 102 | The approximate size of the storage. Starting from this threshold |
1da177e4 LT |
103 | entries will be thrown aggressively. This threshold also determines |
104 | entries' time-to-live and time intervals between garbage collection | |
105 | passes. More entries, less time-to-live, less GC interval. | |
106 | ||
107 | inet_peer_minttl - INTEGER | |
108 | Minimum time-to-live of entries. Should be enough to cover fragment | |
109 | time-to-live on the reassembling side. This minimum time-to-live is | |
110 | guaranteed if the pool size is less than inet_peer_threshold. | |
77a538d5 | 111 | Measured in seconds. |
1da177e4 LT |
112 | |
113 | inet_peer_maxttl - INTEGER | |
114 | Maximum time-to-live of entries. Unused entries will expire after | |
115 | this period of time if there is no memory pressure on the pool (i.e. | |
116 | when the number of entries in the pool is very small). | |
77a538d5 | 117 | Measured in seconds. |
1da177e4 | 118 | |
e18f5feb | 119 | TCP variables: |
1da177e4 | 120 | |
ef56e622 SH |
121 | somaxconn - INTEGER |
122 | Limit of socket listen() backlog, known in userspace as SOMAXCONN. | |
123 | Defaults to 128. See also tcp_max_syn_backlog for additional tuning | |
124 | for TCP sockets. | |
125 | ||
9772efb9 | 126 | tcp_abc - INTEGER |
b3a8a40d SH |
127 | Controls Appropriate Byte Count (ABC) defined in RFC3465. |
128 | ABC is a way of increasing congestion window (cwnd) more slowly | |
129 | in response to partial acknowledgments. | |
130 | Possible values are: | |
131 | 0 increase cwnd once per acknowledgment (no ABC) | |
132 | 1 increase cwnd once per acknowledgment of full sized segment | |
133 | 2 allow increase cwnd by two if acknowledgment is | |
134 | of two segments to compensate for delayed acknowledgments. | |
135 | Default: 0 (off) | |
9772efb9 | 136 | |
ef56e622 SH |
137 | tcp_abort_on_overflow - BOOLEAN |
138 | If listening service is too slow to accept new connections, | |
139 | reset them. Default state is FALSE. It means that if overflow | |
140 | occurred due to a burst, connection will recover. Enable this | |
141 | option _only_ if you are really sure that listening daemon | |
142 | cannot be tuned to accept connections faster. Enabling this | |
143 | option can harm clients of your server. | |
1da177e4 | 144 | |
ef56e622 SH |
145 | tcp_adv_win_scale - INTEGER |
146 | Count buffering overhead as bytes/2^tcp_adv_win_scale | |
147 | (if tcp_adv_win_scale > 0) or bytes-bytes/2^(-tcp_adv_win_scale), | |
148 | if it is <= 0. | |
0147fc05 | 149 | Possible values are [-31, 31], inclusive. |
b49960a0 | 150 | Default: 1 |
1da177e4 | 151 | |
ef56e622 SH |
152 | tcp_allowed_congestion_control - STRING |
153 | Show/set the congestion control choices available to non-privileged | |
154 | processes. The list is a subset of those listed in | |
155 | tcp_available_congestion_control. | |
156 | Default is "reno" and the default setting (tcp_congestion_control). | |
1da177e4 | 157 | |
ef56e622 SH |
158 | tcp_app_win - INTEGER |
159 | Reserve max(window/2^tcp_app_win, mss) of window for application | |
160 | buffer. Value 0 is special, it means that nothing is reserved. | |
161 | Default: 31 | |
1da177e4 | 162 | |
ef56e622 SH |
163 | tcp_available_congestion_control - STRING |
164 | Shows the available congestion control choices that are registered. | |
165 | More congestion control algorithms may be available as modules, | |
166 | but not loaded. | |
1da177e4 | 167 | |
71599cd1 | 168 | tcp_base_mss - INTEGER |
4edc2f34 SH |
169 | The initial value of search_low to be used by the packetization layer |
170 | Path MTU discovery (MTU probing). If MTU probing is enabled, | |
171 | this is the initial MSS used by the connection. | |
71599cd1 | 172 | |
ef56e622 SH |
173 | tcp_congestion_control - STRING |
174 | Set the congestion control algorithm to be used for new | |
175 | connections. The algorithm "reno" is always available, but | |
176 | additional choices may be available based on kernel configuration. | |
177 | Default is set as part of kernel configuration. | |
d8a6e65f ED |
178 | For passive connections, the listener congestion control choice |
179 | is inherited. | |
180 | [see setsockopt(listenfd, SOL_TCP, TCP_CONGESTION, "name" ...) ] | |
1da177e4 | 181 | |
519855c5 WAS |
182 | tcp_cookie_size - INTEGER |
183 | Default size of TCP Cookie Transactions (TCPCT) option, that may be | |
184 | overridden on a per socket basis by the TCPCT socket option. | |
185 | Values greater than the maximum (16) are interpreted as the maximum. | |
186 | Values greater than zero and less than the minimum (8) are interpreted | |
187 | as the minimum. Odd values are interpreted as the next even value. | |
188 | Default: 0 (off). | |
189 | ||
ef56e622 SH |
190 | tcp_dsack - BOOLEAN |
191 | Allows TCP to send "duplicate" SACKs. | |
1da177e4 | 192 | |
eed530b6 YC |
193 | tcp_early_retrans - INTEGER |
194 | Enable Early Retransmit (ER), per RFC 5827. ER lowers the threshold | |
195 | for triggering fast retransmit when the amount of outstanding data is | |
196 | small and when no previously unsent data can be transmitted (such | |
197 | that limited transmit could be used). | |
198 | Possible values: | |
199 | 0 disables ER | |
200 | 1 enables ER | |
201 | 2 enables ER but delays fast recovery and fast retransmit | |
202 | by a fourth of RTT. This mitigates connection falsely | |
203 | recovers when network has a small degree of reordering | |
204 | (less than 3 packets). | |
205 | Default: 2 | |
206 | ||
34a6ef38 | 207 | tcp_ecn - INTEGER |
255cac91 IJ |
208 | Enable Explicit Congestion Notification (ECN) in TCP. ECN is only |
209 | used when both ends of the TCP flow support it. It is useful to | |
210 | avoid losses due to congestion (when the bottleneck router supports | |
211 | ECN). | |
212 | Possible values are: | |
213 | 0 disable ECN | |
214 | 1 ECN enabled | |
215 | 2 Only server-side ECN enabled. If the other end does | |
216 | not support ECN, behavior is like with ECN disabled. | |
217 | Default: 2 | |
ef56e622 SH |
218 | |
219 | tcp_fack - BOOLEAN | |
220 | Enable FACK congestion avoidance and fast retransmission. | |
221 | The value is not used, if tcp_sack is not enabled. | |
1da177e4 LT |
222 | |
223 | tcp_fin_timeout - INTEGER | |
224 | Time to hold socket in state FIN-WAIT-2, if it was closed | |
225 | by our side. Peer can be broken and never close its side, | |
226 | or even died unexpectedly. Default value is 60sec. | |
227 | Usual value used in 2.2 was 180 seconds, you may restore | |
228 | it, but remember that if your machine is even underloaded WEB server, | |
229 | you risk to overflow memory with kilotons of dead sockets, | |
230 | FIN-WAIT-2 sockets are less dangerous than FIN-WAIT-1, | |
231 | because they eat maximum 1.5K of memory, but they tend | |
232 | to live longer. Cf. tcp_max_orphans. | |
233 | ||
89808060 | 234 | tcp_frto - INTEGER |
cd99889c IJ |
235 | Enables Forward RTO-Recovery (F-RTO) defined in RFC4138. |
236 | F-RTO is an enhanced recovery algorithm for TCP retransmission | |
ef56e622 SH |
237 | timeouts. It is particularly beneficial in wireless environments |
238 | where packet loss is typically due to random radio interference | |
564262c1 | 239 | rather than intermediate router congestion. F-RTO is sender-side |
4edc2f34 SH |
240 | only modification. Therefore it does not require any support from |
241 | the peer. | |
242 | ||
cd99889c IJ |
243 | If set to 1, basic version is enabled. 2 enables SACK enhanced |
244 | F-RTO if flow uses SACK. The basic version can be used also when | |
564262c1 | 245 | SACK is in use though scenario(s) with it exists where F-RTO |
cd99889c IJ |
246 | interacts badly with the packet counting of the SACK enabled TCP |
247 | flow. | |
1da177e4 | 248 | |
89808060 IJ |
249 | tcp_frto_response - INTEGER |
250 | When F-RTO has detected that a TCP retransmission timeout was | |
251 | spurious (i.e, the timeout would have been avoided had TCP set a | |
252 | longer retransmission timeout), TCP has several options what to do | |
253 | next. Possible values are: | |
254 | 0 Rate halving based; a smooth and conservative response, | |
255 | results in halved cwnd and ssthresh after one RTT | |
256 | 1 Very conservative response; not recommended because even | |
257 | though being valid, it interacts poorly with the rest of | |
258 | Linux TCP, halves cwnd and ssthresh immediately | |
259 | 2 Aggressive response; undoes congestion control measures | |
260 | that are now known to be unnecessary (ignoring the | |
261 | possibility of a lost retransmission that would require | |
262 | TCP to be more cautious), cwnd and ssthresh are restored | |
263 | to the values prior timeout | |
264 | Default: 0 (rate halving based) | |
265 | ||
ef56e622 SH |
266 | tcp_keepalive_time - INTEGER |
267 | How often TCP sends out keepalive messages when keepalive is enabled. | |
268 | Default: 2hours. | |
1da177e4 | 269 | |
ef56e622 SH |
270 | tcp_keepalive_probes - INTEGER |
271 | How many keepalive probes TCP sends out, until it decides that the | |
272 | connection is broken. Default value: 9. | |
273 | ||
274 | tcp_keepalive_intvl - INTEGER | |
275 | How frequently the probes are send out. Multiplied by | |
276 | tcp_keepalive_probes it is time to kill not responding connection, | |
277 | after probes started. Default value: 75sec i.e. connection | |
278 | will be aborted after ~11 minutes of retries. | |
279 | ||
280 | tcp_low_latency - BOOLEAN | |
281 | If set, the TCP stack makes decisions that prefer lower | |
282 | latency as opposed to higher throughput. By default, this | |
283 | option is not set meaning that higher throughput is preferred. | |
284 | An example of an application where this default should be | |
285 | changed would be a Beowulf compute cluster. | |
286 | Default: 0 | |
1da177e4 LT |
287 | |
288 | tcp_max_orphans - INTEGER | |
289 | Maximal number of TCP sockets not attached to any user file handle, | |
290 | held by system. If this number is exceeded orphaned connections are | |
291 | reset immediately and warning is printed. This limit exists | |
292 | only to prevent simple DoS attacks, you _must_ not rely on this | |
293 | or lower the limit artificially, but rather increase it | |
294 | (probably, after increasing installed memory), | |
295 | if network conditions require more than default value, | |
296 | and tune network services to linger and kill such states | |
297 | more aggressively. Let me to remind again: each orphan eats | |
298 | up to ~64K of unswappable memory. | |
299 | ||
81146ec1 IJ |
300 | tcp_max_ssthresh - INTEGER |
301 | Limited Slow-Start for TCP with large congestion windows (cwnd) defined in | |
302 | RFC3742. Limited slow-start is a mechanism to limit growth of the cwnd | |
303 | on the region where cwnd is larger than tcp_max_ssthresh. TCP increases cwnd | |
304 | by at most tcp_max_ssthresh segments, and by at least tcp_max_ssthresh/2 | |
305 | segments per RTT when the cwnd is above tcp_max_ssthresh. | |
306 | If TCP connection increased cwnd to thousands (or tens of thousands) segments, | |
307 | and thousands of packets were being dropped during slow-start, you can set | |
308 | tcp_max_ssthresh to improve performance for new TCP connection. | |
309 | Default: 0 (off) | |
310 | ||
1da177e4 | 311 | tcp_max_syn_backlog - INTEGER |
99b53bdd PP |
312 | Maximal number of remembered connection requests, which have not |
313 | received an acknowledgment from connecting client. | |
314 | The minimal value is 128 for low memory machines, and it will | |
315 | increase in proportion to the memory of machine. | |
316 | If server suffers from overload, try increasing this number. | |
1da177e4 | 317 | |
ef56e622 SH |
318 | tcp_max_tw_buckets - INTEGER |
319 | Maximal number of timewait sockets held by system simultaneously. | |
320 | If this number is exceeded time-wait socket is immediately destroyed | |
321 | and warning is printed. This limit exists only to prevent | |
322 | simple DoS attacks, you _must_ not lower the limit artificially, | |
323 | but rather increase it (probably, after increasing installed memory), | |
324 | if network conditions require more than default value. | |
1da177e4 | 325 | |
ef56e622 SH |
326 | tcp_mem - vector of 3 INTEGERs: min, pressure, max |
327 | min: below this number of pages TCP is not bothered about its | |
328 | memory appetite. | |
1da177e4 | 329 | |
ef56e622 SH |
330 | pressure: when amount of memory allocated by TCP exceeds this number |
331 | of pages, TCP moderates its memory consumption and enters memory | |
332 | pressure mode, which is exited when memory consumption falls | |
333 | under "min". | |
1da177e4 | 334 | |
ef56e622 | 335 | max: number of pages allowed for queueing by all TCP sockets. |
1da177e4 | 336 | |
ef56e622 SH |
337 | Defaults are calculated at boot time from amount of available |
338 | memory. | |
1da177e4 | 339 | |
71599cd1 | 340 | tcp_moderate_rcvbuf - BOOLEAN |
4edc2f34 | 341 | If set, TCP performs receive buffer auto-tuning, attempting to |
71599cd1 JH |
342 | automatically size the buffer (no greater than tcp_rmem[2]) to |
343 | match the size required by the path for full throughput. Enabled by | |
344 | default. | |
345 | ||
346 | tcp_mtu_probing - INTEGER | |
347 | Controls TCP Packetization-Layer Path MTU Discovery. Takes three | |
348 | values: | |
349 | 0 - Disabled | |
350 | 1 - Disabled by default, enabled when an ICMP black hole detected | |
351 | 2 - Always enabled, use initial MSS of tcp_base_mss. | |
352 | ||
353 | tcp_no_metrics_save - BOOLEAN | |
354 | By default, TCP saves various connection metrics in the route cache | |
355 | when the connection closes, so that connections established in the | |
356 | near future can use these to set initial conditions. Usually, this | |
357 | increases overall performance, but may sometimes cause performance | |
0f035b8e | 358 | degradation. If set, TCP will not cache metrics on closing |
71599cd1 JH |
359 | connections. |
360 | ||
ef56e622 | 361 | tcp_orphan_retries - INTEGER |
5d789229 DL |
362 | This value influences the timeout of a locally closed TCP connection, |
363 | when RTO retransmissions remain unacknowledged. | |
364 | See tcp_retries2 for more details. | |
365 | ||
06b8fc5d | 366 | The default value is 8. |
5d789229 | 367 | If your machine is a loaded WEB server, |
ef56e622 SH |
368 | you should think about lowering this value, such sockets |
369 | may consume significant resources. Cf. tcp_max_orphans. | |
1da177e4 LT |
370 | |
371 | tcp_reordering - INTEGER | |
372 | Maximal reordering of packets in a TCP stream. | |
e18f5feb | 373 | Default: 3 |
1da177e4 LT |
374 | |
375 | tcp_retrans_collapse - BOOLEAN | |
376 | Bug-to-bug compatibility with some broken printers. | |
377 | On retransmit try to send bigger packets to work around bugs in | |
378 | certain TCP stacks. | |
379 | ||
ef56e622 | 380 | tcp_retries1 - INTEGER |
5d789229 DL |
381 | This value influences the time, after which TCP decides, that |
382 | something is wrong due to unacknowledged RTO retransmissions, | |
383 | and reports this suspicion to the network layer. | |
384 | See tcp_retries2 for more details. | |
385 | ||
386 | RFC 1122 recommends at least 3 retransmissions, which is the | |
387 | default. | |
1da177e4 | 388 | |
ef56e622 | 389 | tcp_retries2 - INTEGER |
5d789229 DL |
390 | This value influences the timeout of an alive TCP connection, |
391 | when RTO retransmissions remain unacknowledged. | |
392 | Given a value of N, a hypothetical TCP connection following | |
393 | exponential backoff with an initial RTO of TCP_RTO_MIN would | |
394 | retransmit N times before killing the connection at the (N+1)th RTO. | |
395 | ||
396 | The default value of 15 yields a hypothetical timeout of 924.6 | |
397 | seconds and is a lower bound for the effective timeout. | |
398 | TCP will effectively time out at the first RTO which exceeds the | |
399 | hypothetical timeout. | |
400 | ||
401 | RFC 1122 recommends at least 100 seconds for the timeout, | |
402 | which corresponds to a value of at least 8. | |
1da177e4 | 403 | |
ef56e622 SH |
404 | tcp_rfc1337 - BOOLEAN |
405 | If set, the TCP stack behaves conforming to RFC1337. If unset, | |
406 | we are not conforming to RFC, but prevent TCP TIME_WAIT | |
407 | assassination. | |
408 | Default: 0 | |
1da177e4 LT |
409 | |
410 | tcp_rmem - vector of 3 INTEGERs: min, default, max | |
411 | min: Minimal size of receive buffer used by TCP sockets. | |
412 | It is guaranteed to each TCP socket, even under moderate memory | |
413 | pressure. | |
6539fefd | 414 | Default: 1 page |
1da177e4 | 415 | |
53025f5e | 416 | default: initial size of receive buffer used by TCP sockets. |
1da177e4 LT |
417 | This value overrides net.core.rmem_default used by other protocols. |
418 | Default: 87380 bytes. This value results in window of 65535 with | |
419 | default setting of tcp_adv_win_scale and tcp_app_win:0 and a bit | |
420 | less for default tcp_app_win. See below about these variables. | |
421 | ||
422 | max: maximal size of receive buffer allowed for automatically | |
423 | selected receiver buffers for TCP socket. This value does not override | |
53025f5e BF |
424 | net.core.rmem_max. Calling setsockopt() with SO_RCVBUF disables |
425 | automatic tuning of that socket's receive buffer size, in which | |
426 | case this value is ignored. | |
b49960a0 | 427 | Default: between 87380B and 6MB, depending on RAM size. |
1da177e4 | 428 | |
ef56e622 SH |
429 | tcp_sack - BOOLEAN |
430 | Enable select acknowledgments (SACKS). | |
1da177e4 | 431 | |
ef56e622 SH |
432 | tcp_slow_start_after_idle - BOOLEAN |
433 | If set, provide RFC2861 behavior and time out the congestion | |
434 | window after an idle period. An idle period is defined at | |
435 | the current RTO. If unset, the congestion window will not | |
436 | be timed out after an idle period. | |
437 | Default: 1 | |
1da177e4 | 438 | |
ef56e622 | 439 | tcp_stdurg - BOOLEAN |
4edc2f34 | 440 | Use the Host requirements interpretation of the TCP urgent pointer field. |
ef56e622 SH |
441 | Most hosts use the older BSD interpretation, so if you turn this on |
442 | Linux might not communicate correctly with them. | |
443 | Default: FALSE | |
1da177e4 | 444 | |
ef56e622 SH |
445 | tcp_synack_retries - INTEGER |
446 | Number of times SYNACKs for a passive TCP connection attempt will | |
447 | be retransmitted. Should not be higher than 255. Default value | |
448 | is 5, which corresponds to ~180seconds. | |
1da177e4 | 449 | |
ef56e622 SH |
450 | tcp_syncookies - BOOLEAN |
451 | Only valid when the kernel was compiled with CONFIG_SYNCOOKIES | |
452 | Send out syncookies when the syn backlog queue of a socket | |
4edc2f34 | 453 | overflows. This is to prevent against the common 'SYN flood attack' |
ef56e622 | 454 | Default: FALSE |
1da177e4 | 455 | |
ef56e622 SH |
456 | Note, that syncookies is fallback facility. |
457 | It MUST NOT be used to help highly loaded servers to stand | |
4edc2f34 | 458 | against legal connection rate. If you see SYN flood warnings |
ef56e622 SH |
459 | in your logs, but investigation shows that they occur |
460 | because of overload with legal connections, you should tune | |
461 | another parameters until this warning disappear. | |
462 | See: tcp_max_syn_backlog, tcp_synack_retries, tcp_abort_on_overflow. | |
1da177e4 | 463 | |
ef56e622 SH |
464 | syncookies seriously violate TCP protocol, do not allow |
465 | to use TCP extensions, can result in serious degradation | |
466 | of some services (f.e. SMTP relaying), visible not by you, | |
467 | but your clients and relays, contacting you. While you see | |
4edc2f34 | 468 | SYN flood warnings in logs not being really flooded, your server |
ef56e622 | 469 | is seriously misconfigured. |
1da177e4 | 470 | |
ef56e622 SH |
471 | tcp_syn_retries - INTEGER |
472 | Number of times initial SYNs for an active TCP connection attempt | |
473 | will be retransmitted. Should not be higher than 255. Default value | |
474 | is 5, which corresponds to ~180seconds. | |
475 | ||
476 | tcp_timestamps - BOOLEAN | |
477 | Enable timestamps as defined in RFC1323. | |
1da177e4 | 478 | |
1da177e4 | 479 | tcp_tso_win_divisor - INTEGER |
ef56e622 SH |
480 | This allows control over what percentage of the congestion window |
481 | can be consumed by a single TSO frame. | |
482 | The setting of this parameter is a choice between burstiness and | |
483 | building larger TSO frames. | |
484 | Default: 3 | |
1da177e4 | 485 | |
ef56e622 SH |
486 | tcp_tw_recycle - BOOLEAN |
487 | Enable fast recycling TIME-WAIT sockets. Default value is 0. | |
488 | It should not be changed without advice/request of technical | |
489 | experts. | |
1da177e4 | 490 | |
ef56e622 SH |
491 | tcp_tw_reuse - BOOLEAN |
492 | Allow to reuse TIME-WAIT sockets for new connections when it is | |
493 | safe from protocol viewpoint. Default value is 0. | |
494 | It should not be changed without advice/request of technical | |
495 | experts. | |
ce7bc3bf | 496 | |
ef56e622 SH |
497 | tcp_window_scaling - BOOLEAN |
498 | Enable window scaling as defined in RFC1323. | |
3ff825b2 | 499 | |
ef56e622 | 500 | tcp_wmem - vector of 3 INTEGERs: min, default, max |
53025f5e | 501 | min: Amount of memory reserved for send buffers for TCP sockets. |
ef56e622 | 502 | Each TCP socket has rights to use it due to fact of its birth. |
6539fefd | 503 | Default: 1 page |
9d7bcfc6 | 504 | |
53025f5e BF |
505 | default: initial size of send buffer used by TCP sockets. This |
506 | value overrides net.core.wmem_default used by other protocols. | |
507 | It is usually lower than net.core.wmem_default. | |
ef56e622 SH |
508 | Default: 16K |
509 | ||
53025f5e BF |
510 | max: Maximal amount of memory allowed for automatically tuned |
511 | send buffers for TCP sockets. This value does not override | |
512 | net.core.wmem_max. Calling setsockopt() with SO_SNDBUF disables | |
513 | automatic tuning of that socket's send buffer size, in which case | |
514 | this value is ignored. | |
515 | Default: between 64K and 4MB, depending on RAM size. | |
1da177e4 | 516 | |
15d99e02 RJ |
517 | tcp_workaround_signed_windows - BOOLEAN |
518 | If set, assume no receipt of a window scaling option means the | |
519 | remote TCP is broken and treats the window as a signed quantity. | |
520 | If unset, assume the remote TCP is not broken even if we do | |
521 | not receive a window scaling option from them. | |
522 | Default: 0 | |
523 | ||
72d0b7a8 CL |
524 | tcp_dma_copybreak - INTEGER |
525 | Lower limit, in bytes, of the size of socket reads that will be | |
526 | offloaded to a DMA copy engine, if one is present in the system | |
527 | and CONFIG_NET_DMA is enabled. | |
528 | Default: 4096 | |
529 | ||
36e31b0a AP |
530 | tcp_thin_linear_timeouts - BOOLEAN |
531 | Enable dynamic triggering of linear timeouts for thin streams. | |
532 | If set, a check is performed upon retransmission by timeout to | |
533 | determine if the stream is thin (less than 4 packets in flight). | |
534 | As long as the stream is found to be thin, up to 6 linear | |
535 | timeouts may be performed before exponential backoff mode is | |
536 | initiated. This improves retransmission latency for | |
537 | non-aggressive thin streams, often found to be time-dependent. | |
538 | For more information on thin streams, see | |
539 | Documentation/networking/tcp-thin.txt | |
540 | Default: 0 | |
541 | ||
7e380175 AP |
542 | tcp_thin_dupack - BOOLEAN |
543 | Enable dynamic triggering of retransmissions after one dupACK | |
544 | for thin streams. If set, a check is performed upon reception | |
545 | of a dupACK to determine if the stream is thin (less than 4 | |
546 | packets in flight). As long as the stream is found to be thin, | |
547 | data is retransmitted on the first received dupACK. This | |
548 | improves retransmission latency for non-aggressive thin | |
549 | streams, often found to be time-dependent. | |
550 | For more information on thin streams, see | |
551 | Documentation/networking/tcp-thin.txt | |
552 | Default: 0 | |
553 | ||
95766fff HA |
554 | UDP variables: |
555 | ||
556 | udp_mem - vector of 3 INTEGERs: min, pressure, max | |
557 | Number of pages allowed for queueing by all UDP sockets. | |
558 | ||
559 | min: Below this number of pages UDP is not bothered about its | |
560 | memory appetite. When amount of memory allocated by UDP exceeds | |
561 | this number, UDP starts to moderate memory usage. | |
562 | ||
563 | pressure: This value was introduced to follow format of tcp_mem. | |
564 | ||
565 | max: Number of pages allowed for queueing by all UDP sockets. | |
566 | ||
567 | Default is calculated at boot time from amount of available memory. | |
568 | ||
569 | udp_rmem_min - INTEGER | |
570 | Minimal size of receive buffer used by UDP sockets in moderation. | |
571 | Each UDP socket is able to use the size for receiving data, even if | |
572 | total pages of UDP sockets exceed udp_mem pressure. The unit is byte. | |
6539fefd | 573 | Default: 1 page |
95766fff HA |
574 | |
575 | udp_wmem_min - INTEGER | |
576 | Minimal size of send buffer used by UDP sockets in moderation. | |
577 | Each UDP socket is able to use the size for sending data, even if | |
578 | total pages of UDP sockets exceed udp_mem pressure. The unit is byte. | |
6539fefd | 579 | Default: 1 page |
95766fff | 580 | |
8802f616 PM |
581 | CIPSOv4 Variables: |
582 | ||
583 | cipso_cache_enable - BOOLEAN | |
584 | If set, enable additions to and lookups from the CIPSO label mapping | |
585 | cache. If unset, additions are ignored and lookups always result in a | |
586 | miss. However, regardless of the setting the cache is still | |
587 | invalidated when required when means you can safely toggle this on and | |
588 | off and the cache will always be "safe". | |
589 | Default: 1 | |
590 | ||
591 | cipso_cache_bucket_size - INTEGER | |
592 | The CIPSO label cache consists of a fixed size hash table with each | |
593 | hash bucket containing a number of cache entries. This variable limits | |
594 | the number of entries in each hash bucket; the larger the value the | |
595 | more CIPSO label mappings that can be cached. When the number of | |
596 | entries in a given hash bucket reaches this limit adding new entries | |
597 | causes the oldest entry in the bucket to be removed to make room. | |
598 | Default: 10 | |
599 | ||
600 | cipso_rbm_optfmt - BOOLEAN | |
601 | Enable the "Optimized Tag 1 Format" as defined in section 3.4.2.6 of | |
602 | the CIPSO draft specification (see Documentation/netlabel for details). | |
603 | This means that when set the CIPSO tag will be padded with empty | |
604 | categories in order to make the packet data 32-bit aligned. | |
605 | Default: 0 | |
606 | ||
607 | cipso_rbm_structvalid - BOOLEAN | |
608 | If set, do a very strict check of the CIPSO option when | |
609 | ip_options_compile() is called. If unset, relax the checks done during | |
610 | ip_options_compile(). Either way is "safe" as errors are caught else | |
611 | where in the CIPSO processing code but setting this to 0 (False) should | |
612 | result in less work (i.e. it should be faster) but could cause problems | |
613 | with other implementations that require strict checking. | |
614 | Default: 0 | |
615 | ||
1da177e4 LT |
616 | IP Variables: |
617 | ||
618 | ip_local_port_range - 2 INTEGERS | |
619 | Defines the local port range that is used by TCP and UDP to | |
e18f5feb | 620 | choose the local port. The first number is the first, the |
5d6bd861 FLVC |
621 | second the last local port number. The default values are |
622 | 32768 and 61000 respectively. | |
1da177e4 | 623 | |
e3826f1e AW |
624 | ip_local_reserved_ports - list of comma separated ranges |
625 | Specify the ports which are reserved for known third-party | |
626 | applications. These ports will not be used by automatic port | |
627 | assignments (e.g. when calling connect() or bind() with port | |
628 | number 0). Explicit port allocation behavior is unchanged. | |
629 | ||
630 | The format used for both input and output is a comma separated | |
631 | list of ranges (e.g. "1,2-4,10-10" for ports 1, 2, 3, 4 and | |
632 | 10). Writing to the file will clear all previously reserved | |
633 | ports and update the current list with the one given in the | |
634 | input. | |
635 | ||
636 | Note that ip_local_port_range and ip_local_reserved_ports | |
637 | settings are independent and both are considered by the kernel | |
638 | when determining which ports are available for automatic port | |
639 | assignments. | |
640 | ||
641 | You can reserve ports which are not in the current | |
642 | ip_local_port_range, e.g.: | |
643 | ||
644 | $ cat /proc/sys/net/ipv4/ip_local_port_range | |
645 | 32000 61000 | |
646 | $ cat /proc/sys/net/ipv4/ip_local_reserved_ports | |
647 | 8080,9148 | |
648 | ||
649 | although this is redundant. However such a setting is useful | |
650 | if later the port range is changed to a value that will | |
651 | include the reserved ports. | |
652 | ||
653 | Default: Empty | |
654 | ||
1da177e4 LT |
655 | ip_nonlocal_bind - BOOLEAN |
656 | If set, allows processes to bind() to non-local IP addresses, | |
657 | which can be quite useful - but may break some applications. | |
658 | Default: 0 | |
659 | ||
660 | ip_dynaddr - BOOLEAN | |
661 | If set non-zero, enables support for dynamic addresses. | |
662 | If set to a non-zero value larger than 1, a kernel log | |
663 | message will be printed when dynamic address rewriting | |
664 | occurs. | |
665 | Default: 0 | |
666 | ||
667 | icmp_echo_ignore_all - BOOLEAN | |
7ce31246 DM |
668 | If set non-zero, then the kernel will ignore all ICMP ECHO |
669 | requests sent to it. | |
670 | Default: 0 | |
671 | ||
1da177e4 | 672 | icmp_echo_ignore_broadcasts - BOOLEAN |
7ce31246 DM |
673 | If set non-zero, then the kernel will ignore all ICMP ECHO and |
674 | TIMESTAMP requests sent to it via broadcast/multicast. | |
675 | Default: 1 | |
1da177e4 LT |
676 | |
677 | icmp_ratelimit - INTEGER | |
678 | Limit the maximal rates for sending ICMP packets whose type matches | |
679 | icmp_ratemask (see below) to specific targets. | |
6dbf4bca SH |
680 | 0 to disable any limiting, |
681 | otherwise the minimal space between responses in milliseconds. | |
682 | Default: 1000 | |
1da177e4 LT |
683 | |
684 | icmp_ratemask - INTEGER | |
685 | Mask made of ICMP types for which rates are being limited. | |
686 | Significant bits: IHGFEDCBA9876543210 | |
687 | Default mask: 0000001100000011000 (6168) | |
688 | ||
689 | Bit definitions (see include/linux/icmp.h): | |
690 | 0 Echo Reply | |
691 | 3 Destination Unreachable * | |
692 | 4 Source Quench * | |
693 | 5 Redirect | |
694 | 8 Echo Request | |
695 | B Time Exceeded * | |
696 | C Parameter Problem * | |
697 | D Timestamp Request | |
698 | E Timestamp Reply | |
699 | F Info Request | |
700 | G Info Reply | |
701 | H Address Mask Request | |
702 | I Address Mask Reply | |
703 | ||
704 | * These are rate limited by default (see default mask above) | |
705 | ||
706 | icmp_ignore_bogus_error_responses - BOOLEAN | |
707 | Some routers violate RFC1122 by sending bogus responses to broadcast | |
708 | frames. Such violations are normally logged via a kernel warning. | |
709 | If this is set to TRUE, the kernel will not give such warnings, which | |
710 | will avoid log file clutter. | |
711 | Default: FALSE | |
712 | ||
95f7daf1 H |
713 | icmp_errors_use_inbound_ifaddr - BOOLEAN |
714 | ||
715 | If zero, icmp error messages are sent with the primary address of | |
716 | the exiting interface. | |
e18f5feb | 717 | |
95f7daf1 H |
718 | If non-zero, the message will be sent with the primary address of |
719 | the interface that received the packet that caused the icmp error. | |
720 | This is the behaviour network many administrators will expect from | |
721 | a router. And it can make debugging complicated network layouts | |
e18f5feb | 722 | much easier. |
95f7daf1 H |
723 | |
724 | Note that if no primary address exists for the interface selected, | |
725 | then the primary address of the first non-loopback interface that | |
d6bc8ac9 | 726 | has one will be used regardless of this setting. |
95f7daf1 H |
727 | |
728 | Default: 0 | |
729 | ||
1da177e4 LT |
730 | igmp_max_memberships - INTEGER |
731 | Change the maximum number of multicast groups we can subscribe to. | |
732 | Default: 20 | |
733 | ||
d67ef35f JE |
734 | Theoretical maximum value is bounded by having to send a membership |
735 | report in a single datagram (i.e. the report can't span multiple | |
736 | datagrams, or risk confusing the switch and leaving groups you don't | |
737 | intend to). | |
1da177e4 | 738 | |
d67ef35f JE |
739 | The number of supported groups 'M' is bounded by the number of group |
740 | report entries you can fit into a single datagram of 65535 bytes. | |
741 | ||
742 | M = 65536-sizeof (ip header)/(sizeof(Group record)) | |
743 | ||
744 | Group records are variable length, with a minimum of 12 bytes. | |
745 | So net.ipv4.igmp_max_memberships should not be set higher than: | |
746 | ||
747 | (65536-24) / 12 = 5459 | |
748 | ||
749 | The value 5459 assumes no IP header options, so in practice | |
750 | this number may be lower. | |
751 | ||
752 | conf/interface/* changes special settings per interface (where | |
753 | "interface" is the name of your network interface) | |
754 | ||
755 | conf/all/* is special, changes the settings for all interfaces | |
1da177e4 LT |
756 | |
757 | log_martians - BOOLEAN | |
758 | Log packets with impossible addresses to kernel log. | |
759 | log_martians for the interface will be enabled if at least one of | |
760 | conf/{all,interface}/log_martians is set to TRUE, | |
761 | it will be disabled otherwise | |
762 | ||
763 | accept_redirects - BOOLEAN | |
764 | Accept ICMP redirect messages. | |
765 | accept_redirects for the interface will be enabled if: | |
e18f5feb JDB |
766 | - both conf/{all,interface}/accept_redirects are TRUE in the case |
767 | forwarding for the interface is enabled | |
1da177e4 | 768 | or |
e18f5feb JDB |
769 | - at least one of conf/{all,interface}/accept_redirects is TRUE in the |
770 | case forwarding for the interface is disabled | |
1da177e4 LT |
771 | accept_redirects for the interface will be disabled otherwise |
772 | default TRUE (host) | |
773 | FALSE (router) | |
774 | ||
775 | forwarding - BOOLEAN | |
776 | Enable IP forwarding on this interface. | |
777 | ||
778 | mc_forwarding - BOOLEAN | |
779 | Do multicast routing. The kernel needs to be compiled with CONFIG_MROUTE | |
780 | and a multicast routing daemon is required. | |
e18f5feb JDB |
781 | conf/all/mc_forwarding must also be set to TRUE to enable multicast |
782 | routing for the interface | |
1da177e4 LT |
783 | |
784 | medium_id - INTEGER | |
785 | Integer value used to differentiate the devices by the medium they | |
786 | are attached to. Two devices can have different id values when | |
787 | the broadcast packets are received only on one of them. | |
788 | The default value 0 means that the device is the only interface | |
789 | to its medium, value of -1 means that medium is not known. | |
e18f5feb | 790 | |
1da177e4 LT |
791 | Currently, it is used to change the proxy_arp behavior: |
792 | the proxy_arp feature is enabled for packets forwarded between | |
793 | two devices attached to different media. | |
794 | ||
795 | proxy_arp - BOOLEAN | |
796 | Do proxy arp. | |
797 | proxy_arp for the interface will be enabled if at least one of | |
798 | conf/{all,interface}/proxy_arp is set to TRUE, | |
799 | it will be disabled otherwise | |
800 | ||
65324144 JDB |
801 | proxy_arp_pvlan - BOOLEAN |
802 | Private VLAN proxy arp. | |
803 | Basically allow proxy arp replies back to the same interface | |
804 | (from which the ARP request/solicitation was received). | |
805 | ||
806 | This is done to support (ethernet) switch features, like RFC | |
807 | 3069, where the individual ports are NOT allowed to | |
808 | communicate with each other, but they are allowed to talk to | |
809 | the upstream router. As described in RFC 3069, it is possible | |
810 | to allow these hosts to communicate through the upstream | |
811 | router by proxy_arp'ing. Don't need to be used together with | |
812 | proxy_arp. | |
813 | ||
814 | This technology is known by different names: | |
815 | In RFC 3069 it is called VLAN Aggregation. | |
816 | Cisco and Allied Telesyn call it Private VLAN. | |
817 | Hewlett-Packard call it Source-Port filtering or port-isolation. | |
818 | Ericsson call it MAC-Forced Forwarding (RFC Draft). | |
819 | ||
1da177e4 LT |
820 | shared_media - BOOLEAN |
821 | Send(router) or accept(host) RFC1620 shared media redirects. | |
822 | Overrides ip_secure_redirects. | |
823 | shared_media for the interface will be enabled if at least one of | |
824 | conf/{all,interface}/shared_media is set to TRUE, | |
825 | it will be disabled otherwise | |
826 | default TRUE | |
827 | ||
828 | secure_redirects - BOOLEAN | |
829 | Accept ICMP redirect messages only for gateways, | |
830 | listed in default gateway list. | |
831 | secure_redirects for the interface will be enabled if at least one of | |
832 | conf/{all,interface}/secure_redirects is set to TRUE, | |
833 | it will be disabled otherwise | |
834 | default TRUE | |
835 | ||
836 | send_redirects - BOOLEAN | |
837 | Send redirects, if router. | |
838 | send_redirects for the interface will be enabled if at least one of | |
839 | conf/{all,interface}/send_redirects is set to TRUE, | |
840 | it will be disabled otherwise | |
841 | Default: TRUE | |
842 | ||
843 | bootp_relay - BOOLEAN | |
844 | Accept packets with source address 0.b.c.d destined | |
845 | not to this host as local ones. It is supposed, that | |
846 | BOOTP relay daemon will catch and forward such packets. | |
847 | conf/all/bootp_relay must also be set to TRUE to enable BOOTP relay | |
848 | for the interface | |
849 | default FALSE | |
850 | Not Implemented Yet. | |
851 | ||
852 | accept_source_route - BOOLEAN | |
853 | Accept packets with SRR option. | |
854 | conf/all/accept_source_route must also be set to TRUE to accept packets | |
855 | with SRR option on the interface | |
856 | default TRUE (router) | |
857 | FALSE (host) | |
858 | ||
8153a10c PM |
859 | accept_local - BOOLEAN |
860 | Accept packets with local source addresses. In combination with | |
861 | suitable routing, this can be used to direct packets between two | |
862 | local interfaces over the wire and have them accepted properly. | |
863 | default FALSE | |
864 | ||
c1cf8422 | 865 | rp_filter - INTEGER |
1da177e4 | 866 | 0 - No source validation. |
c1cf8422 SH |
867 | 1 - Strict mode as defined in RFC3704 Strict Reverse Path |
868 | Each incoming packet is tested against the FIB and if the interface | |
869 | is not the best reverse path the packet check will fail. | |
870 | By default failed packets are discarded. | |
871 | 2 - Loose mode as defined in RFC3704 Loose Reverse Path | |
872 | Each incoming packet's source address is also tested against the FIB | |
873 | and if the source address is not reachable via any interface | |
874 | the packet check will fail. | |
875 | ||
e18f5feb | 876 | Current recommended practice in RFC3704 is to enable strict mode |
bf869c30 | 877 | to prevent IP spoofing from DDos attacks. If using asymmetric routing |
e18f5feb | 878 | or other complicated routing, then loose mode is recommended. |
c1cf8422 | 879 | |
1f5865e7 SW |
880 | The max value from conf/{all,interface}/rp_filter is used |
881 | when doing source validation on the {interface}. | |
1da177e4 LT |
882 | |
883 | Default value is 0. Note that some distributions enable it | |
884 | in startup scripts. | |
885 | ||
886 | arp_filter - BOOLEAN | |
887 | 1 - Allows you to have multiple network interfaces on the same | |
888 | subnet, and have the ARPs for each interface be answered | |
889 | based on whether or not the kernel would route a packet from | |
890 | the ARP'd IP out that interface (therefore you must use source | |
891 | based routing for this to work). In other words it allows control | |
892 | of which cards (usually 1) will respond to an arp request. | |
893 | ||
894 | 0 - (default) The kernel can respond to arp requests with addresses | |
895 | from other interfaces. This may seem wrong but it usually makes | |
896 | sense, because it increases the chance of successful communication. | |
897 | IP addresses are owned by the complete host on Linux, not by | |
898 | particular interfaces. Only for more complex setups like load- | |
899 | balancing, does this behaviour cause problems. | |
900 | ||
901 | arp_filter for the interface will be enabled if at least one of | |
902 | conf/{all,interface}/arp_filter is set to TRUE, | |
903 | it will be disabled otherwise | |
904 | ||
905 | arp_announce - INTEGER | |
906 | Define different restriction levels for announcing the local | |
907 | source IP address from IP packets in ARP requests sent on | |
908 | interface: | |
909 | 0 - (default) Use any local address, configured on any interface | |
910 | 1 - Try to avoid local addresses that are not in the target's | |
911 | subnet for this interface. This mode is useful when target | |
912 | hosts reachable via this interface require the source IP | |
913 | address in ARP requests to be part of their logical network | |
914 | configured on the receiving interface. When we generate the | |
915 | request we will check all our subnets that include the | |
916 | target IP and will preserve the source address if it is from | |
917 | such subnet. If there is no such subnet we select source | |
918 | address according to the rules for level 2. | |
919 | 2 - Always use the best local address for this target. | |
920 | In this mode we ignore the source address in the IP packet | |
921 | and try to select local address that we prefer for talks with | |
922 | the target host. Such local address is selected by looking | |
923 | for primary IP addresses on all our subnets on the outgoing | |
924 | interface that include the target IP address. If no suitable | |
925 | local address is found we select the first local address | |
926 | we have on the outgoing interface or on all other interfaces, | |
927 | with the hope we will receive reply for our request and | |
928 | even sometimes no matter the source IP address we announce. | |
929 | ||
930 | The max value from conf/{all,interface}/arp_announce is used. | |
931 | ||
932 | Increasing the restriction level gives more chance for | |
933 | receiving answer from the resolved target while decreasing | |
934 | the level announces more valid sender's information. | |
935 | ||
936 | arp_ignore - INTEGER | |
937 | Define different modes for sending replies in response to | |
938 | received ARP requests that resolve local target IP addresses: | |
939 | 0 - (default): reply for any local target IP address, configured | |
940 | on any interface | |
941 | 1 - reply only if the target IP address is local address | |
942 | configured on the incoming interface | |
943 | 2 - reply only if the target IP address is local address | |
944 | configured on the incoming interface and both with the | |
945 | sender's IP address are part from same subnet on this interface | |
946 | 3 - do not reply for local addresses configured with scope host, | |
947 | only resolutions for global and link addresses are replied | |
948 | 4-7 - reserved | |
949 | 8 - do not reply for all local addresses | |
950 | ||
951 | The max value from conf/{all,interface}/arp_ignore is used | |
952 | when ARP request is received on the {interface} | |
953 | ||
eefef1cf SH |
954 | arp_notify - BOOLEAN |
955 | Define mode for notification of address and device changes. | |
956 | 0 - (default): do nothing | |
3f8dc236 | 957 | 1 - Generate gratuitous arp requests when device is brought up |
eefef1cf SH |
958 | or hardware address changes. |
959 | ||
c1b1bce8 | 960 | arp_accept - BOOLEAN |
6d955180 OP |
961 | Define behavior for gratuitous ARP frames who's IP is not |
962 | already present in the ARP table: | |
963 | 0 - don't create new entries in the ARP table | |
964 | 1 - create new entries in the ARP table | |
965 | ||
966 | Both replies and requests type gratuitous arp will trigger the | |
967 | ARP table to be updated, if this setting is on. | |
968 | ||
969 | If the ARP table already contains the IP address of the | |
970 | gratuitous arp frame, the arp table will be updated regardless | |
971 | if this setting is on or off. | |
972 | ||
c1b1bce8 | 973 | |
1da177e4 LT |
974 | app_solicit - INTEGER |
975 | The maximum number of probes to send to the user space ARP daemon | |
976 | via netlink before dropping back to multicast probes (see | |
977 | mcast_solicit). Defaults to 0. | |
978 | ||
979 | disable_policy - BOOLEAN | |
980 | Disable IPSEC policy (SPD) for this interface | |
981 | ||
982 | disable_xfrm - BOOLEAN | |
983 | Disable IPSEC encryption on this interface, whatever the policy | |
984 | ||
985 | ||
986 | ||
987 | tag - INTEGER | |
988 | Allows you to write a number, which can be used as required. | |
989 | Default value is 0. | |
990 | ||
1da177e4 LT |
991 | Alexey Kuznetsov. |
992 | kuznet@ms2.inr.ac.ru | |
993 | ||
994 | Updated by: | |
995 | Andi Kleen | |
996 | ak@muc.de | |
997 | Nicolas Delon | |
998 | delon.nicolas@wanadoo.fr | |
999 | ||
1000 | ||
1001 | ||
1002 | ||
1003 | /proc/sys/net/ipv6/* Variables: | |
1004 | ||
1005 | IPv6 has no global variables such as tcp_*. tcp_* settings under ipv4/ also | |
1006 | apply to IPv6 [XXX?]. | |
1007 | ||
1008 | bindv6only - BOOLEAN | |
1009 | Default value for IPV6_V6ONLY socket option, | |
e18f5feb | 1010 | which restricts use of the IPv6 socket to IPv6 communication |
1da177e4 LT |
1011 | only. |
1012 | TRUE: disable IPv4-mapped address feature | |
1013 | FALSE: enable IPv4-mapped address feature | |
1014 | ||
d5c073ca | 1015 | Default: FALSE (as specified in RFC3493) |
1da177e4 LT |
1016 | |
1017 | IPv6 Fragmentation: | |
1018 | ||
1019 | ip6frag_high_thresh - INTEGER | |
e18f5feb | 1020 | Maximum memory used to reassemble IPv6 fragments. When |
1da177e4 LT |
1021 | ip6frag_high_thresh bytes of memory is allocated for this purpose, |
1022 | the fragment handler will toss packets until ip6frag_low_thresh | |
1023 | is reached. | |
e18f5feb | 1024 | |
1da177e4 | 1025 | ip6frag_low_thresh - INTEGER |
e18f5feb | 1026 | See ip6frag_high_thresh |
1da177e4 LT |
1027 | |
1028 | ip6frag_time - INTEGER | |
1029 | Time in seconds to keep an IPv6 fragment in memory. | |
1030 | ||
1031 | ip6frag_secret_interval - INTEGER | |
e18f5feb | 1032 | Regeneration interval (in seconds) of the hash secret (or lifetime |
1da177e4 LT |
1033 | for the hash secret) for IPv6 fragments. |
1034 | Default: 600 | |
1035 | ||
1036 | conf/default/*: | |
1037 | Change the interface-specific default settings. | |
1038 | ||
1039 | ||
1040 | conf/all/*: | |
e18f5feb | 1041 | Change all the interface-specific settings. |
1da177e4 LT |
1042 | |
1043 | [XXX: Other special features than forwarding?] | |
1044 | ||
1045 | conf/all/forwarding - BOOLEAN | |
e18f5feb | 1046 | Enable global IPv6 forwarding between all interfaces. |
1da177e4 | 1047 | |
e18f5feb | 1048 | IPv4 and IPv6 work differently here; e.g. netfilter must be used |
1da177e4 LT |
1049 | to control which interfaces may forward packets and which not. |
1050 | ||
e18f5feb | 1051 | This also sets all interfaces' Host/Router setting |
1da177e4 LT |
1052 | 'forwarding' to the specified value. See below for details. |
1053 | ||
1054 | This referred to as global forwarding. | |
1055 | ||
fbea49e1 YH |
1056 | proxy_ndp - BOOLEAN |
1057 | Do proxy ndp. | |
1058 | ||
1da177e4 LT |
1059 | conf/interface/*: |
1060 | Change special settings per interface. | |
1061 | ||
e18f5feb | 1062 | The functional behaviour for certain settings is different |
1da177e4 LT |
1063 | depending on whether local forwarding is enabled or not. |
1064 | ||
605b91c8 | 1065 | accept_ra - INTEGER |
1da177e4 | 1066 | Accept Router Advertisements; autoconfigure using them. |
e18f5feb | 1067 | |
026359bc TA |
1068 | It also determines whether or not to transmit Router |
1069 | Solicitations. If and only if the functional setting is to | |
1070 | accept Router Advertisements, Router Solicitations will be | |
1071 | transmitted. | |
1072 | ||
ae8abfa0 TG |
1073 | Possible values are: |
1074 | 0 Do not accept Router Advertisements. | |
1075 | 1 Accept Router Advertisements if forwarding is disabled. | |
1076 | 2 Overrule forwarding behaviour. Accept Router Advertisements | |
1077 | even if forwarding is enabled. | |
1078 | ||
1da177e4 LT |
1079 | Functional default: enabled if local forwarding is disabled. |
1080 | disabled if local forwarding is enabled. | |
1081 | ||
65f5c7c1 YH |
1082 | accept_ra_defrtr - BOOLEAN |
1083 | Learn default router in Router Advertisement. | |
1084 | ||
1085 | Functional default: enabled if accept_ra is enabled. | |
1086 | disabled if accept_ra is disabled. | |
1087 | ||
c4fd30eb | 1088 | accept_ra_pinfo - BOOLEAN |
2fe0ae78 | 1089 | Learn Prefix Information in Router Advertisement. |
c4fd30eb YH |
1090 | |
1091 | Functional default: enabled if accept_ra is enabled. | |
1092 | disabled if accept_ra is disabled. | |
1093 | ||
09c884d4 YH |
1094 | accept_ra_rt_info_max_plen - INTEGER |
1095 | Maximum prefix length of Route Information in RA. | |
1096 | ||
1097 | Route Information w/ prefix larger than or equal to this | |
1098 | variable shall be ignored. | |
1099 | ||
1100 | Functional default: 0 if accept_ra_rtr_pref is enabled. | |
1101 | -1 if accept_ra_rtr_pref is disabled. | |
1102 | ||
930d6ff2 YH |
1103 | accept_ra_rtr_pref - BOOLEAN |
1104 | Accept Router Preference in RA. | |
1105 | ||
1106 | Functional default: enabled if accept_ra is enabled. | |
1107 | disabled if accept_ra is disabled. | |
1108 | ||
1da177e4 LT |
1109 | accept_redirects - BOOLEAN |
1110 | Accept Redirects. | |
1111 | ||
1112 | Functional default: enabled if local forwarding is disabled. | |
1113 | disabled if local forwarding is enabled. | |
1114 | ||
0bcbc926 YH |
1115 | accept_source_route - INTEGER |
1116 | Accept source routing (routing extension header). | |
1117 | ||
bb4dbf9e | 1118 | >= 0: Accept only routing header type 2. |
0bcbc926 YH |
1119 | < 0: Do not accept routing header. |
1120 | ||
1121 | Default: 0 | |
1122 | ||
1da177e4 | 1123 | autoconf - BOOLEAN |
e18f5feb | 1124 | Autoconfigure addresses using Prefix Information in Router |
1da177e4 LT |
1125 | Advertisements. |
1126 | ||
c4fd30eb YH |
1127 | Functional default: enabled if accept_ra_pinfo is enabled. |
1128 | disabled if accept_ra_pinfo is disabled. | |
1da177e4 LT |
1129 | |
1130 | dad_transmits - INTEGER | |
1131 | The amount of Duplicate Address Detection probes to send. | |
1132 | Default: 1 | |
e18f5feb | 1133 | |
605b91c8 | 1134 | forwarding - INTEGER |
e18f5feb | 1135 | Configure interface-specific Host/Router behaviour. |
1da177e4 | 1136 | |
e18f5feb | 1137 | Note: It is recommended to have the same setting on all |
1da177e4 LT |
1138 | interfaces; mixed router/host scenarios are rather uncommon. |
1139 | ||
ae8abfa0 TG |
1140 | Possible values are: |
1141 | 0 Forwarding disabled | |
1142 | 1 Forwarding enabled | |
ae8abfa0 TG |
1143 | |
1144 | FALSE (0): | |
1da177e4 LT |
1145 | |
1146 | By default, Host behaviour is assumed. This means: | |
1147 | ||
1148 | 1. IsRouter flag is not set in Neighbour Advertisements. | |
026359bc TA |
1149 | 2. If accept_ra is TRUE (default), transmit Router |
1150 | Solicitations. | |
e18f5feb | 1151 | 3. If accept_ra is TRUE (default), accept Router |
1da177e4 LT |
1152 | Advertisements (and do autoconfiguration). |
1153 | 4. If accept_redirects is TRUE (default), accept Redirects. | |
1154 | ||
ae8abfa0 | 1155 | TRUE (1): |
1da177e4 | 1156 | |
e18f5feb | 1157 | If local forwarding is enabled, Router behaviour is assumed. |
1da177e4 LT |
1158 | This means exactly the reverse from the above: |
1159 | ||
1160 | 1. IsRouter flag is set in Neighbour Advertisements. | |
026359bc | 1161 | 2. Router Solicitations are not sent unless accept_ra is 2. |
ae8abfa0 | 1162 | 3. Router Advertisements are ignored unless accept_ra is 2. |
1da177e4 LT |
1163 | 4. Redirects are ignored. |
1164 | ||
ae8abfa0 TG |
1165 | Default: 0 (disabled) if global forwarding is disabled (default), |
1166 | otherwise 1 (enabled). | |
1da177e4 LT |
1167 | |
1168 | hop_limit - INTEGER | |
1169 | Default Hop Limit to set. | |
1170 | Default: 64 | |
1171 | ||
1172 | mtu - INTEGER | |
1173 | Default Maximum Transfer Unit | |
1174 | Default: 1280 (IPv6 required minimum) | |
1175 | ||
52e16356 YH |
1176 | router_probe_interval - INTEGER |
1177 | Minimum interval (in seconds) between Router Probing described | |
1178 | in RFC4191. | |
1179 | ||
1180 | Default: 60 | |
1181 | ||
1da177e4 LT |
1182 | router_solicitation_delay - INTEGER |
1183 | Number of seconds to wait after interface is brought up | |
1184 | before sending Router Solicitations. | |
1185 | Default: 1 | |
1186 | ||
1187 | router_solicitation_interval - INTEGER | |
1188 | Number of seconds to wait between Router Solicitations. | |
1189 | Default: 4 | |
1190 | ||
1191 | router_solicitations - INTEGER | |
e18f5feb | 1192 | Number of Router Solicitations to send until assuming no |
1da177e4 LT |
1193 | routers are present. |
1194 | Default: 3 | |
1195 | ||
1196 | use_tempaddr - INTEGER | |
1197 | Preference for Privacy Extensions (RFC3041). | |
1198 | <= 0 : disable Privacy Extensions | |
1199 | == 1 : enable Privacy Extensions, but prefer public | |
1200 | addresses over temporary addresses. | |
1201 | > 1 : enable Privacy Extensions and prefer temporary | |
1202 | addresses over public addresses. | |
1203 | Default: 0 (for most devices) | |
1204 | -1 (for point-to-point devices and loopback devices) | |
1205 | ||
1206 | temp_valid_lft - INTEGER | |
1207 | valid lifetime (in seconds) for temporary addresses. | |
1208 | Default: 604800 (7 days) | |
1209 | ||
1210 | temp_prefered_lft - INTEGER | |
1211 | Preferred lifetime (in seconds) for temporary addresses. | |
1212 | Default: 86400 (1 day) | |
1213 | ||
1214 | max_desync_factor - INTEGER | |
1215 | Maximum value for DESYNC_FACTOR, which is a random value | |
e18f5feb | 1216 | that ensures that clients don't synchronize with each |
1da177e4 LT |
1217 | other and generate new addresses at exactly the same time. |
1218 | value is in seconds. | |
1219 | Default: 600 | |
e18f5feb | 1220 | |
1da177e4 LT |
1221 | regen_max_retry - INTEGER |
1222 | Number of attempts before give up attempting to generate | |
1223 | valid temporary addresses. | |
1224 | Default: 5 | |
1225 | ||
1226 | max_addresses - INTEGER | |
e79dc484 BH |
1227 | Maximum number of autoconfigured addresses per interface. Setting |
1228 | to zero disables the limitation. It is not recommended to set this | |
1229 | value too large (or to zero) because it would be an easy way to | |
1230 | crash the kernel by allowing too many addresses to be created. | |
1da177e4 LT |
1231 | Default: 16 |
1232 | ||
778d80be | 1233 | disable_ipv6 - BOOLEAN |
9bdd8d40 BH |
1234 | Disable IPv6 operation. If accept_dad is set to 2, this value |
1235 | will be dynamically set to TRUE if DAD fails for the link-local | |
1236 | address. | |
778d80be YH |
1237 | Default: FALSE (enable IPv6 operation) |
1238 | ||
56d417b1 BH |
1239 | When this value is changed from 1 to 0 (IPv6 is being enabled), |
1240 | it will dynamically create a link-local address on the given | |
1241 | interface and start Duplicate Address Detection, if necessary. | |
1242 | ||
1243 | When this value is changed from 0 to 1 (IPv6 is being disabled), | |
1244 | it will dynamically delete all address on the given interface. | |
1245 | ||
1b34be74 YH |
1246 | accept_dad - INTEGER |
1247 | Whether to accept DAD (Duplicate Address Detection). | |
1248 | 0: Disable DAD | |
1249 | 1: Enable DAD (default) | |
1250 | 2: Enable DAD, and disable IPv6 operation if MAC-based duplicate | |
1251 | link-local address has been found. | |
1252 | ||
f7734fdf OP |
1253 | force_tllao - BOOLEAN |
1254 | Enable sending the target link-layer address option even when | |
1255 | responding to a unicast neighbor solicitation. | |
1256 | Default: FALSE | |
1257 | ||
1258 | Quoting from RFC 2461, section 4.4, Target link-layer address: | |
1259 | ||
1260 | "The option MUST be included for multicast solicitations in order to | |
1261 | avoid infinite Neighbor Solicitation "recursion" when the peer node | |
1262 | does not have a cache entry to return a Neighbor Advertisements | |
1263 | message. When responding to unicast solicitations, the option can be | |
1264 | omitted since the sender of the solicitation has the correct link- | |
1265 | layer address; otherwise it would not have be able to send the unicast | |
1266 | solicitation in the first place. However, including the link-layer | |
1267 | address in this case adds little overhead and eliminates a potential | |
1268 | race condition where the sender deletes the cached link-layer address | |
1269 | prior to receiving a response to a previous solicitation." | |
1270 | ||
1da177e4 LT |
1271 | icmp/*: |
1272 | ratelimit - INTEGER | |
1273 | Limit the maximal rates for sending ICMPv6 packets. | |
6dbf4bca SH |
1274 | 0 to disable any limiting, |
1275 | otherwise the minimal space between responses in milliseconds. | |
1276 | Default: 1000 | |
1da177e4 LT |
1277 | |
1278 | ||
1279 | IPv6 Update by: | |
1280 | Pekka Savola <pekkas@netcore.fi> | |
1281 | YOSHIFUJI Hideaki / USAGI Project <yoshfuji@linux-ipv6.org> | |
1282 | ||
1283 | ||
1284 | /proc/sys/net/bridge/* Variables: | |
1285 | ||
1286 | bridge-nf-call-arptables - BOOLEAN | |
1287 | 1 : pass bridged ARP traffic to arptables' FORWARD chain. | |
1288 | 0 : disable this. | |
1289 | Default: 1 | |
1290 | ||
1291 | bridge-nf-call-iptables - BOOLEAN | |
1292 | 1 : pass bridged IPv4 traffic to iptables' chains. | |
1293 | 0 : disable this. | |
1294 | Default: 1 | |
1295 | ||
1296 | bridge-nf-call-ip6tables - BOOLEAN | |
1297 | 1 : pass bridged IPv6 traffic to ip6tables' chains. | |
1298 | 0 : disable this. | |
1299 | Default: 1 | |
1300 | ||
1301 | bridge-nf-filter-vlan-tagged - BOOLEAN | |
516299d2 MM |
1302 | 1 : pass bridged vlan-tagged ARP/IP/IPv6 traffic to {arp,ip,ip6}tables. |
1303 | 0 : disable this. | |
4981682c | 1304 | Default: 0 |
516299d2 MM |
1305 | |
1306 | bridge-nf-filter-pppoe-tagged - BOOLEAN | |
1307 | 1 : pass bridged pppoe-tagged IP/IPv6 traffic to {ip,ip6}tables. | |
1da177e4 | 1308 | 0 : disable this. |
4981682c | 1309 | Default: 0 |
1da177e4 | 1310 | |
4981682c PNA |
1311 | bridge-nf-pass-vlan-input-dev - BOOLEAN |
1312 | 1: if bridge-nf-filter-vlan-tagged is enabled, try to find a vlan | |
1313 | interface on the bridge and set the netfilter input device to the vlan. | |
1314 | This allows use of e.g. "iptables -i br0.1" and makes the REDIRECT | |
1315 | target work with vlan-on-top-of-bridge interfaces. When no matching | |
1316 | vlan interface is found, or this switch is off, the input device is | |
1317 | set to the bridge interface. | |
1318 | 0: disable bridge netfilter vlan interface lookup. | |
1319 | Default: 0 | |
1da177e4 | 1320 | |
32e8d494 VY |
1321 | proc/sys/net/sctp/* Variables: |
1322 | ||
1323 | addip_enable - BOOLEAN | |
1324 | Enable or disable extension of Dynamic Address Reconfiguration | |
1325 | (ADD-IP) functionality specified in RFC5061. This extension provides | |
1326 | the ability to dynamically add and remove new addresses for the SCTP | |
1327 | associations. | |
1328 | ||
1329 | 1: Enable extension. | |
1330 | ||
1331 | 0: Disable extension. | |
1332 | ||
1333 | Default: 0 | |
1334 | ||
1335 | addip_noauth_enable - BOOLEAN | |
1336 | Dynamic Address Reconfiguration (ADD-IP) requires the use of | |
1337 | authentication to protect the operations of adding or removing new | |
1338 | addresses. This requirement is mandated so that unauthorized hosts | |
1339 | would not be able to hijack associations. However, older | |
1340 | implementations may not have implemented this requirement while | |
1341 | allowing the ADD-IP extension. For reasons of interoperability, | |
1342 | we provide this variable to control the enforcement of the | |
1343 | authentication requirement. | |
1344 | ||
1345 | 1: Allow ADD-IP extension to be used without authentication. This | |
1346 | should only be set in a closed environment for interoperability | |
1347 | with older implementations. | |
1348 | ||
1349 | 0: Enforce the authentication requirement | |
1350 | ||
1351 | Default: 0 | |
1352 | ||
1353 | auth_enable - BOOLEAN | |
1354 | Enable or disable Authenticated Chunks extension. This extension | |
1355 | provides the ability to send and receive authenticated chunks and is | |
1356 | required for secure operation of Dynamic Address Reconfiguration | |
1357 | (ADD-IP) extension. | |
1358 | ||
1359 | 1: Enable this extension. | |
1360 | 0: Disable this extension. | |
1361 | ||
1362 | Default: 0 | |
1363 | ||
1364 | prsctp_enable - BOOLEAN | |
1365 | Enable or disable the Partial Reliability extension (RFC3758) which | |
1366 | is used to notify peers that a given DATA should no longer be expected. | |
1367 | ||
1368 | 1: Enable extension | |
1369 | 0: Disable | |
1370 | ||
1371 | Default: 1 | |
1372 | ||
1373 | max_burst - INTEGER | |
1374 | The limit of the number of new packets that can be initially sent. It | |
1375 | controls how bursty the generated traffic can be. | |
1376 | ||
1377 | Default: 4 | |
1378 | ||
1379 | association_max_retrans - INTEGER | |
1380 | Set the maximum number for retransmissions that an association can | |
1381 | attempt deciding that the remote end is unreachable. If this value | |
1382 | is exceeded, the association is terminated. | |
1383 | ||
1384 | Default: 10 | |
1385 | ||
1386 | max_init_retransmits - INTEGER | |
1387 | The maximum number of retransmissions of INIT and COOKIE-ECHO chunks | |
1388 | that an association will attempt before declaring the destination | |
1389 | unreachable and terminating. | |
1390 | ||
1391 | Default: 8 | |
1392 | ||
1393 | path_max_retrans - INTEGER | |
1394 | The maximum number of retransmissions that will be attempted on a given | |
1395 | path. Once this threshold is exceeded, the path is considered | |
1396 | unreachable, and new traffic will use a different path when the | |
1397 | association is multihomed. | |
1398 | ||
1399 | Default: 5 | |
1400 | ||
1401 | rto_initial - INTEGER | |
1402 | The initial round trip timeout value in milliseconds that will be used | |
1403 | in calculating round trip times. This is the initial time interval | |
1404 | for retransmissions. | |
1405 | ||
1406 | Default: 3000 | |
1da177e4 | 1407 | |
32e8d494 VY |
1408 | rto_max - INTEGER |
1409 | The maximum value (in milliseconds) of the round trip timeout. This | |
1410 | is the largest time interval that can elapse between retransmissions. | |
1411 | ||
1412 | Default: 60000 | |
1413 | ||
1414 | rto_min - INTEGER | |
1415 | The minimum value (in milliseconds) of the round trip timeout. This | |
1416 | is the smallest time interval the can elapse between retransmissions. | |
1417 | ||
1418 | Default: 1000 | |
1419 | ||
1420 | hb_interval - INTEGER | |
1421 | The interval (in milliseconds) between HEARTBEAT chunks. These chunks | |
1422 | are sent at the specified interval on idle paths to probe the state of | |
1423 | a given path between 2 associations. | |
1424 | ||
1425 | Default: 30000 | |
1426 | ||
1427 | sack_timeout - INTEGER | |
1428 | The amount of time (in milliseconds) that the implementation will wait | |
1429 | to send a SACK. | |
1430 | ||
1431 | Default: 200 | |
1432 | ||
1433 | valid_cookie_life - INTEGER | |
1434 | The default lifetime of the SCTP cookie (in milliseconds). The cookie | |
1435 | is used during association establishment. | |
1436 | ||
1437 | Default: 60000 | |
1438 | ||
1439 | cookie_preserve_enable - BOOLEAN | |
1440 | Enable or disable the ability to extend the lifetime of the SCTP cookie | |
1441 | that is used during the establishment phase of SCTP association | |
1442 | ||
1443 | 1: Enable cookie lifetime extension. | |
1444 | 0: Disable | |
1445 | ||
1446 | Default: 1 | |
1447 | ||
1448 | rcvbuf_policy - INTEGER | |
1449 | Determines if the receive buffer is attributed to the socket or to | |
1450 | association. SCTP supports the capability to create multiple | |
1451 | associations on a single socket. When using this capability, it is | |
1452 | possible that a single stalled association that's buffering a lot | |
1453 | of data may block other associations from delivering their data by | |
1454 | consuming all of the receive buffer space. To work around this, | |
1455 | the rcvbuf_policy could be set to attribute the receiver buffer space | |
1456 | to each association instead of the socket. This prevents the described | |
1457 | blocking. | |
1458 | ||
1459 | 1: rcvbuf space is per association | |
1460 | 0: recbuf space is per socket | |
1461 | ||
1462 | Default: 0 | |
1463 | ||
1464 | sndbuf_policy - INTEGER | |
1465 | Similar to rcvbuf_policy above, this applies to send buffer space. | |
1466 | ||
1467 | 1: Send buffer is tracked per association | |
1468 | 0: Send buffer is tracked per socket. | |
1469 | ||
1470 | Default: 0 | |
1471 | ||
1472 | sctp_mem - vector of 3 INTEGERs: min, pressure, max | |
1473 | Number of pages allowed for queueing by all SCTP sockets. | |
1474 | ||
1475 | min: Below this number of pages SCTP is not bothered about its | |
1476 | memory appetite. When amount of memory allocated by SCTP exceeds | |
1477 | this number, SCTP starts to moderate memory usage. | |
1478 | ||
1479 | pressure: This value was introduced to follow format of tcp_mem. | |
1480 | ||
1481 | max: Number of pages allowed for queueing by all SCTP sockets. | |
1482 | ||
1483 | Default is calculated at boot time from amount of available memory. | |
1484 | ||
1485 | sctp_rmem - vector of 3 INTEGERs: min, default, max | |
a6e1204b MM |
1486 | Only the first value ("min") is used, "default" and "max" are |
1487 | ignored. | |
1488 | ||
1489 | min: Minimal size of receive buffer used by SCTP socket. | |
1490 | It is guaranteed to each SCTP socket (but not association) even | |
1491 | under moderate memory pressure. | |
1492 | ||
1493 | Default: 1 page | |
32e8d494 VY |
1494 | |
1495 | sctp_wmem - vector of 3 INTEGERs: min, default, max | |
a6e1204b | 1496 | Currently this tunable has no effect. |
32e8d494 | 1497 | |
72388433 BD |
1498 | addr_scope_policy - INTEGER |
1499 | Control IPv4 address scoping - draft-stewart-tsvwg-sctp-ipv4-00 | |
1500 | ||
1501 | 0 - Disable IPv4 address scoping | |
1502 | 1 - Enable IPv4 address scoping | |
1503 | 2 - Follow draft but allow IPv4 private addresses | |
1504 | 3 - Follow draft but allow IPv4 link local addresses | |
1505 | ||
1506 | Default: 1 | |
1507 | ||
1da177e4 | 1508 | |
4edc2f34 | 1509 | /proc/sys/net/core/* |
c60f6aa8 | 1510 | Please see: Documentation/sysctl/net.txt for descriptions of these entries. |
705efc3b | 1511 | |
4edc2f34 SH |
1512 | |
1513 | /proc/sys/net/unix/* | |
705efc3b WT |
1514 | max_dgram_qlen - INTEGER |
1515 | The maximum length of dgram socket receive queue | |
1516 | ||
1517 | Default: 10 | |
1518 | ||
1519 | ||
1520 | UNDOCUMENTED: | |
4edc2f34 SH |
1521 | |
1522 | /proc/sys/net/irda/* | |
1523 | fast_poll_increase FIXME | |
1524 | warn_noreply_time FIXME | |
1525 | discovery_slots FIXME | |
1526 | slot_timeout FIXME | |
1527 | max_baud_rate FIXME | |
1528 | discovery_timeout FIXME | |
1529 | lap_keepalive_time FIXME | |
1530 | max_noreply_time FIXME | |
1531 | max_tx_data_size FIXME | |
1532 | max_tx_window FIXME | |
1533 | min_tx_turn_time FIXME |