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 | 17 | |
cd174e67 HFS |
18 | ip_no_pmtu_disc - INTEGER |
19 | Disable Path MTU Discovery. If enabled in mode 1 and a | |
188b04d5 HFS |
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. | |
cd174e67 HFS |
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 | ||
8ed1dc44 HFS |
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 | |
188b04d5 | 41 | Default: FALSE |
1da177e4 LT |
42 | |
43 | min_pmtu - INTEGER | |
20db93c3 | 44 | default 552 - minimum discovered Path MTU |
1da177e4 | 45 | |
f87c10a8 HFS |
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 | ||
cbaf087a BG |
59 | route/max_size - INTEGER |
60 | Maximum number of routes allowed in the kernel. Increase | |
61 | this when using large numbers of interfaces and/or routes. | |
62 | ||
2724680b YH |
63 | neigh/default/gc_thresh1 - INTEGER |
64 | Minimum number of entries to keep. Garbage collector will not | |
65 | purge entries if there are fewer than this number. | |
b66c66dc | 66 | Default: 128 |
2724680b | 67 | |
cbaf087a BG |
68 | neigh/default/gc_thresh3 - INTEGER |
69 | Maximum number of neighbor entries allowed. Increase this | |
70 | when using large numbers of interfaces and when communicating | |
71 | with large numbers of directly-connected peers. | |
cc868028 | 72 | Default: 1024 |
cbaf087a | 73 | |
8b5c171b ED |
74 | neigh/default/unres_qlen_bytes - INTEGER |
75 | The maximum number of bytes which may be used by packets | |
76 | queued for each unresolved address by other network layers. | |
77 | (added in linux 3.3) | |
3b09adcb | 78 | Setting negative value is meaningless and will return error. |
cc868028 | 79 | Default: 65536 Bytes(64KB) |
8b5c171b ED |
80 | |
81 | neigh/default/unres_qlen - INTEGER | |
82 | The maximum number of packets which may be queued for each | |
83 | unresolved address by other network layers. | |
84 | (deprecated in linux 3.3) : use unres_qlen_bytes instead. | |
cc868028 | 85 | Prior to linux 3.3, the default value is 3 which may cause |
5d248c49 | 86 | unexpected packet loss. The current default value is calculated |
cc868028 SW |
87 | according to default value of unres_qlen_bytes and true size of |
88 | packet. | |
89 | Default: 31 | |
8b5c171b | 90 | |
1da177e4 LT |
91 | mtu_expires - INTEGER |
92 | Time, in seconds, that cached PMTU information is kept. | |
93 | ||
94 | min_adv_mss - INTEGER | |
95 | The advertised MSS depends on the first hop route MTU, but will | |
96 | never be lower than this setting. | |
97 | ||
98 | IP Fragmentation: | |
99 | ||
100 | ipfrag_high_thresh - INTEGER | |
e18f5feb | 101 | Maximum memory used to reassemble IP fragments. When |
1da177e4 LT |
102 | ipfrag_high_thresh bytes of memory is allocated for this purpose, |
103 | the fragment handler will toss packets until ipfrag_low_thresh | |
104 | is reached. | |
e18f5feb | 105 | |
1da177e4 | 106 | ipfrag_low_thresh - INTEGER |
e18f5feb | 107 | See ipfrag_high_thresh |
1da177e4 LT |
108 | |
109 | ipfrag_time - INTEGER | |
e18f5feb | 110 | Time in seconds to keep an IP fragment in memory. |
1da177e4 LT |
111 | |
112 | ipfrag_secret_interval - INTEGER | |
e18f5feb | 113 | Regeneration interval (in seconds) of the hash secret (or lifetime |
1da177e4 LT |
114 | for the hash secret) for IP fragments. |
115 | Default: 600 | |
116 | ||
89cee8b1 | 117 | ipfrag_max_dist - INTEGER |
e18f5feb JDB |
118 | ipfrag_max_dist is a non-negative integer value which defines the |
119 | maximum "disorder" which is allowed among fragments which share a | |
120 | common IP source address. Note that reordering of packets is | |
121 | not unusual, but if a large number of fragments arrive from a source | |
122 | IP address while a particular fragment queue remains incomplete, it | |
123 | probably indicates that one or more fragments belonging to that queue | |
124 | have been lost. When ipfrag_max_dist is positive, an additional check | |
125 | is done on fragments before they are added to a reassembly queue - if | |
126 | ipfrag_max_dist (or more) fragments have arrived from a particular IP | |
127 | address between additions to any IP fragment queue using that source | |
128 | address, it's presumed that one or more fragments in the queue are | |
129 | lost. The existing fragment queue will be dropped, and a new one | |
89cee8b1 HX |
130 | started. An ipfrag_max_dist value of zero disables this check. |
131 | ||
132 | Using a very small value, e.g. 1 or 2, for ipfrag_max_dist can | |
133 | result in unnecessarily dropping fragment queues when normal | |
e18f5feb JDB |
134 | reordering of packets occurs, which could lead to poor application |
135 | performance. Using a very large value, e.g. 50000, increases the | |
136 | likelihood of incorrectly reassembling IP fragments that originate | |
89cee8b1 HX |
137 | from different IP datagrams, which could result in data corruption. |
138 | Default: 64 | |
139 | ||
1da177e4 LT |
140 | INET peer storage: |
141 | ||
142 | inet_peer_threshold - INTEGER | |
e18f5feb | 143 | The approximate size of the storage. Starting from this threshold |
1da177e4 LT |
144 | entries will be thrown aggressively. This threshold also determines |
145 | entries' time-to-live and time intervals between garbage collection | |
146 | passes. More entries, less time-to-live, less GC interval. | |
147 | ||
148 | inet_peer_minttl - INTEGER | |
149 | Minimum time-to-live of entries. Should be enough to cover fragment | |
150 | time-to-live on the reassembling side. This minimum time-to-live is | |
151 | guaranteed if the pool size is less than inet_peer_threshold. | |
77a538d5 | 152 | Measured in seconds. |
1da177e4 LT |
153 | |
154 | inet_peer_maxttl - INTEGER | |
155 | Maximum time-to-live of entries. Unused entries will expire after | |
156 | this period of time if there is no memory pressure on the pool (i.e. | |
157 | when the number of entries in the pool is very small). | |
77a538d5 | 158 | Measured in seconds. |
1da177e4 | 159 | |
e18f5feb | 160 | TCP variables: |
1da177e4 | 161 | |
ef56e622 SH |
162 | somaxconn - INTEGER |
163 | Limit of socket listen() backlog, known in userspace as SOMAXCONN. | |
164 | Defaults to 128. See also tcp_max_syn_backlog for additional tuning | |
165 | for TCP sockets. | |
166 | ||
ef56e622 SH |
167 | tcp_abort_on_overflow - BOOLEAN |
168 | If listening service is too slow to accept new connections, | |
169 | reset them. Default state is FALSE. It means that if overflow | |
170 | occurred due to a burst, connection will recover. Enable this | |
171 | option _only_ if you are really sure that listening daemon | |
172 | cannot be tuned to accept connections faster. Enabling this | |
173 | option can harm clients of your server. | |
1da177e4 | 174 | |
ef56e622 SH |
175 | tcp_adv_win_scale - INTEGER |
176 | Count buffering overhead as bytes/2^tcp_adv_win_scale | |
177 | (if tcp_adv_win_scale > 0) or bytes-bytes/2^(-tcp_adv_win_scale), | |
178 | if it is <= 0. | |
0147fc05 | 179 | Possible values are [-31, 31], inclusive. |
b49960a0 | 180 | Default: 1 |
1da177e4 | 181 | |
ef56e622 SH |
182 | tcp_allowed_congestion_control - STRING |
183 | Show/set the congestion control choices available to non-privileged | |
184 | processes. The list is a subset of those listed in | |
185 | tcp_available_congestion_control. | |
186 | Default is "reno" and the default setting (tcp_congestion_control). | |
1da177e4 | 187 | |
ef56e622 SH |
188 | tcp_app_win - INTEGER |
189 | Reserve max(window/2^tcp_app_win, mss) of window for application | |
190 | buffer. Value 0 is special, it means that nothing is reserved. | |
191 | Default: 31 | |
1da177e4 | 192 | |
f54b3111 ED |
193 | tcp_autocorking - BOOLEAN |
194 | Enable TCP auto corking : | |
195 | When applications do consecutive small write()/sendmsg() system calls, | |
196 | we try to coalesce these small writes as much as possible, to lower | |
197 | total amount of sent packets. This is done if at least one prior | |
198 | packet for the flow is waiting in Qdisc queues or device transmit | |
199 | queue. Applications can still use TCP_CORK for optimal behavior | |
200 | when they know how/when to uncork their sockets. | |
201 | Default : 1 | |
202 | ||
ef56e622 SH |
203 | tcp_available_congestion_control - STRING |
204 | Shows the available congestion control choices that are registered. | |
205 | More congestion control algorithms may be available as modules, | |
206 | but not loaded. | |
1da177e4 | 207 | |
71599cd1 | 208 | tcp_base_mss - INTEGER |
4edc2f34 SH |
209 | The initial value of search_low to be used by the packetization layer |
210 | Path MTU discovery (MTU probing). If MTU probing is enabled, | |
211 | this is the initial MSS used by the connection. | |
71599cd1 | 212 | |
ef56e622 SH |
213 | tcp_congestion_control - STRING |
214 | Set the congestion control algorithm to be used for new | |
215 | connections. The algorithm "reno" is always available, but | |
216 | additional choices may be available based on kernel configuration. | |
217 | Default is set as part of kernel configuration. | |
d8a6e65f ED |
218 | For passive connections, the listener congestion control choice |
219 | is inherited. | |
220 | [see setsockopt(listenfd, SOL_TCP, TCP_CONGESTION, "name" ...) ] | |
1da177e4 | 221 | |
ef56e622 SH |
222 | tcp_dsack - BOOLEAN |
223 | Allows TCP to send "duplicate" SACKs. | |
1da177e4 | 224 | |
eed530b6 YC |
225 | tcp_early_retrans - INTEGER |
226 | Enable Early Retransmit (ER), per RFC 5827. ER lowers the threshold | |
227 | for triggering fast retransmit when the amount of outstanding data is | |
228 | small and when no previously unsent data can be transmitted (such | |
6ba8a3b1 | 229 | that limited transmit could be used). Also controls the use of |
3dd17ede | 230 | Tail loss probe (TLP) that converts RTOs occurring due to tail |
6ba8a3b1 | 231 | losses into fast recovery (draft-dukkipati-tcpm-tcp-loss-probe-01). |
eed530b6 YC |
232 | Possible values: |
233 | 0 disables ER | |
234 | 1 enables ER | |
235 | 2 enables ER but delays fast recovery and fast retransmit | |
236 | by a fourth of RTT. This mitigates connection falsely | |
237 | recovers when network has a small degree of reordering | |
238 | (less than 3 packets). | |
6ba8a3b1 ND |
239 | 3 enables delayed ER and TLP. |
240 | 4 enables TLP only. | |
241 | Default: 3 | |
eed530b6 | 242 | |
34a6ef38 | 243 | tcp_ecn - INTEGER |
7e3a2dc5 RJ |
244 | Control use of Explicit Congestion Notification (ECN) by TCP. |
245 | ECN is used only when both ends of the TCP connection indicate | |
246 | support for it. This feature is useful in avoiding losses due | |
247 | to congestion by allowing supporting routers to signal | |
248 | congestion before having to drop packets. | |
255cac91 | 249 | Possible values are: |
7e3a2dc5 | 250 | 0 Disable ECN. Neither initiate nor accept ECN. |
3d55b323 VS |
251 | 1 Enable ECN when requested by incoming connections and |
252 | also request ECN on outgoing connection attempts. | |
253 | 2 Enable ECN when requested by incoming connections | |
7e3a2dc5 | 254 | but do not request ECN on outgoing connections. |
255cac91 | 255 | Default: 2 |
ef56e622 SH |
256 | |
257 | tcp_fack - BOOLEAN | |
258 | Enable FACK congestion avoidance and fast retransmission. | |
259 | The value is not used, if tcp_sack is not enabled. | |
1da177e4 LT |
260 | |
261 | tcp_fin_timeout - INTEGER | |
d825da2e RJ |
262 | The length of time an orphaned (no longer referenced by any |
263 | application) connection will remain in the FIN_WAIT_2 state | |
264 | before it is aborted at the local end. While a perfectly | |
265 | valid "receive only" state for an un-orphaned connection, an | |
266 | orphaned connection in FIN_WAIT_2 state could otherwise wait | |
267 | forever for the remote to close its end of the connection. | |
268 | Cf. tcp_max_orphans | |
269 | Default: 60 seconds | |
1da177e4 | 270 | |
89808060 | 271 | tcp_frto - INTEGER |
e33099f9 | 272 | Enables Forward RTO-Recovery (F-RTO) defined in RFC5682. |
cd99889c | 273 | F-RTO is an enhanced recovery algorithm for TCP retransmission |
e33099f9 YC |
274 | timeouts. It is particularly beneficial in networks where the |
275 | RTT fluctuates (e.g., wireless). F-RTO is sender-side only | |
276 | modification. It does not require any support from the peer. | |
277 | ||
278 | By default it's enabled with a non-zero value. 0 disables F-RTO. | |
1da177e4 | 279 | |
ef56e622 SH |
280 | tcp_keepalive_time - INTEGER |
281 | How often TCP sends out keepalive messages when keepalive is enabled. | |
282 | Default: 2hours. | |
1da177e4 | 283 | |
ef56e622 SH |
284 | tcp_keepalive_probes - INTEGER |
285 | How many keepalive probes TCP sends out, until it decides that the | |
286 | connection is broken. Default value: 9. | |
287 | ||
288 | tcp_keepalive_intvl - INTEGER | |
289 | How frequently the probes are send out. Multiplied by | |
290 | tcp_keepalive_probes it is time to kill not responding connection, | |
291 | after probes started. Default value: 75sec i.e. connection | |
292 | will be aborted after ~11 minutes of retries. | |
293 | ||
294 | tcp_low_latency - BOOLEAN | |
295 | If set, the TCP stack makes decisions that prefer lower | |
296 | latency as opposed to higher throughput. By default, this | |
297 | option is not set meaning that higher throughput is preferred. | |
298 | An example of an application where this default should be | |
299 | changed would be a Beowulf compute cluster. | |
300 | Default: 0 | |
1da177e4 LT |
301 | |
302 | tcp_max_orphans - INTEGER | |
303 | Maximal number of TCP sockets not attached to any user file handle, | |
304 | held by system. If this number is exceeded orphaned connections are | |
305 | reset immediately and warning is printed. This limit exists | |
306 | only to prevent simple DoS attacks, you _must_ not rely on this | |
307 | or lower the limit artificially, but rather increase it | |
308 | (probably, after increasing installed memory), | |
309 | if network conditions require more than default value, | |
310 | and tune network services to linger and kill such states | |
311 | more aggressively. Let me to remind again: each orphan eats | |
312 | up to ~64K of unswappable memory. | |
313 | ||
1da177e4 | 314 | tcp_max_syn_backlog - INTEGER |
99b53bdd PP |
315 | Maximal number of remembered connection requests, which have not |
316 | received an acknowledgment from connecting client. | |
317 | The minimal value is 128 for low memory machines, and it will | |
318 | increase in proportion to the memory of machine. | |
319 | If server suffers from overload, try increasing this number. | |
1da177e4 | 320 | |
ef56e622 SH |
321 | tcp_max_tw_buckets - INTEGER |
322 | Maximal number of timewait sockets held by system simultaneously. | |
323 | If this number is exceeded time-wait socket is immediately destroyed | |
324 | and warning is printed. This limit exists only to prevent | |
325 | simple DoS attacks, you _must_ not lower the limit artificially, | |
326 | but rather increase it (probably, after increasing installed memory), | |
327 | if network conditions require more than default value. | |
1da177e4 | 328 | |
ef56e622 SH |
329 | tcp_mem - vector of 3 INTEGERs: min, pressure, max |
330 | min: below this number of pages TCP is not bothered about its | |
331 | memory appetite. | |
1da177e4 | 332 | |
ef56e622 SH |
333 | pressure: when amount of memory allocated by TCP exceeds this number |
334 | of pages, TCP moderates its memory consumption and enters memory | |
335 | pressure mode, which is exited when memory consumption falls | |
336 | under "min". | |
1da177e4 | 337 | |
ef56e622 | 338 | max: number of pages allowed for queueing by all TCP sockets. |
1da177e4 | 339 | |
ef56e622 SH |
340 | Defaults are calculated at boot time from amount of available |
341 | memory. | |
1da177e4 | 342 | |
71599cd1 | 343 | tcp_moderate_rcvbuf - BOOLEAN |
4edc2f34 | 344 | If set, TCP performs receive buffer auto-tuning, attempting to |
71599cd1 JH |
345 | automatically size the buffer (no greater than tcp_rmem[2]) to |
346 | match the size required by the path for full throughput. Enabled by | |
347 | default. | |
348 | ||
349 | tcp_mtu_probing - INTEGER | |
350 | Controls TCP Packetization-Layer Path MTU Discovery. Takes three | |
351 | values: | |
352 | 0 - Disabled | |
353 | 1 - Disabled by default, enabled when an ICMP black hole detected | |
354 | 2 - Always enabled, use initial MSS of tcp_base_mss. | |
355 | ||
356 | tcp_no_metrics_save - BOOLEAN | |
357 | By default, TCP saves various connection metrics in the route cache | |
358 | when the connection closes, so that connections established in the | |
359 | near future can use these to set initial conditions. Usually, this | |
360 | increases overall performance, but may sometimes cause performance | |
0f035b8e | 361 | degradation. If set, TCP will not cache metrics on closing |
71599cd1 JH |
362 | connections. |
363 | ||
ef56e622 | 364 | tcp_orphan_retries - INTEGER |
5d789229 DL |
365 | This value influences the timeout of a locally closed TCP connection, |
366 | when RTO retransmissions remain unacknowledged. | |
367 | See tcp_retries2 for more details. | |
368 | ||
06b8fc5d | 369 | The default value is 8. |
5d789229 | 370 | If your machine is a loaded WEB server, |
ef56e622 SH |
371 | you should think about lowering this value, such sockets |
372 | may consume significant resources. Cf. tcp_max_orphans. | |
1da177e4 LT |
373 | |
374 | tcp_reordering - INTEGER | |
375 | Maximal reordering of packets in a TCP stream. | |
e18f5feb | 376 | Default: 3 |
1da177e4 LT |
377 | |
378 | tcp_retrans_collapse - BOOLEAN | |
379 | Bug-to-bug compatibility with some broken printers. | |
380 | On retransmit try to send bigger packets to work around bugs in | |
381 | certain TCP stacks. | |
382 | ||
ef56e622 | 383 | tcp_retries1 - INTEGER |
5d789229 DL |
384 | This value influences the time, after which TCP decides, that |
385 | something is wrong due to unacknowledged RTO retransmissions, | |
386 | and reports this suspicion to the network layer. | |
387 | See tcp_retries2 for more details. | |
388 | ||
389 | RFC 1122 recommends at least 3 retransmissions, which is the | |
390 | default. | |
1da177e4 | 391 | |
ef56e622 | 392 | tcp_retries2 - INTEGER |
5d789229 DL |
393 | This value influences the timeout of an alive TCP connection, |
394 | when RTO retransmissions remain unacknowledged. | |
395 | Given a value of N, a hypothetical TCP connection following | |
396 | exponential backoff with an initial RTO of TCP_RTO_MIN would | |
397 | retransmit N times before killing the connection at the (N+1)th RTO. | |
398 | ||
399 | The default value of 15 yields a hypothetical timeout of 924.6 | |
400 | seconds and is a lower bound for the effective timeout. | |
401 | TCP will effectively time out at the first RTO which exceeds the | |
402 | hypothetical timeout. | |
403 | ||
404 | RFC 1122 recommends at least 100 seconds for the timeout, | |
405 | which corresponds to a value of at least 8. | |
1da177e4 | 406 | |
ef56e622 SH |
407 | tcp_rfc1337 - BOOLEAN |
408 | If set, the TCP stack behaves conforming to RFC1337. If unset, | |
409 | we are not conforming to RFC, but prevent TCP TIME_WAIT | |
410 | assassination. | |
411 | Default: 0 | |
1da177e4 LT |
412 | |
413 | tcp_rmem - vector of 3 INTEGERs: min, default, max | |
414 | min: Minimal size of receive buffer used by TCP sockets. | |
415 | It is guaranteed to each TCP socket, even under moderate memory | |
416 | pressure. | |
6539fefd | 417 | Default: 1 page |
1da177e4 | 418 | |
53025f5e | 419 | default: initial size of receive buffer used by TCP sockets. |
1da177e4 LT |
420 | This value overrides net.core.rmem_default used by other protocols. |
421 | Default: 87380 bytes. This value results in window of 65535 with | |
422 | default setting of tcp_adv_win_scale and tcp_app_win:0 and a bit | |
423 | less for default tcp_app_win. See below about these variables. | |
424 | ||
425 | max: maximal size of receive buffer allowed for automatically | |
426 | selected receiver buffers for TCP socket. This value does not override | |
53025f5e BF |
427 | net.core.rmem_max. Calling setsockopt() with SO_RCVBUF disables |
428 | automatic tuning of that socket's receive buffer size, in which | |
429 | case this value is ignored. | |
b49960a0 | 430 | Default: between 87380B and 6MB, depending on RAM size. |
1da177e4 | 431 | |
ef56e622 SH |
432 | tcp_sack - BOOLEAN |
433 | Enable select acknowledgments (SACKS). | |
1da177e4 | 434 | |
ef56e622 SH |
435 | tcp_slow_start_after_idle - BOOLEAN |
436 | If set, provide RFC2861 behavior and time out the congestion | |
437 | window after an idle period. An idle period is defined at | |
438 | the current RTO. If unset, the congestion window will not | |
439 | be timed out after an idle period. | |
440 | Default: 1 | |
1da177e4 | 441 | |
ef56e622 | 442 | tcp_stdurg - BOOLEAN |
4edc2f34 | 443 | Use the Host requirements interpretation of the TCP urgent pointer field. |
ef56e622 SH |
444 | Most hosts use the older BSD interpretation, so if you turn this on |
445 | Linux might not communicate correctly with them. | |
446 | Default: FALSE | |
1da177e4 | 447 | |
ef56e622 SH |
448 | tcp_synack_retries - INTEGER |
449 | Number of times SYNACKs for a passive TCP connection attempt will | |
450 | be retransmitted. Should not be higher than 255. Default value | |
6c9ff979 AB |
451 | is 5, which corresponds to 31seconds till the last retransmission |
452 | with the current initial RTO of 1second. With this the final timeout | |
453 | for a passive TCP connection will happen after 63seconds. | |
1da177e4 | 454 | |
ef56e622 | 455 | tcp_syncookies - BOOLEAN |
a3c910d2 | 456 | Only valid when the kernel was compiled with CONFIG_SYN_COOKIES |
ef56e622 | 457 | Send out syncookies when the syn backlog queue of a socket |
4edc2f34 | 458 | overflows. This is to prevent against the common 'SYN flood attack' |
a3c910d2 | 459 | Default: 1 |
1da177e4 | 460 | |
ef56e622 SH |
461 | Note, that syncookies is fallback facility. |
462 | It MUST NOT be used to help highly loaded servers to stand | |
4edc2f34 | 463 | against legal connection rate. If you see SYN flood warnings |
ef56e622 SH |
464 | in your logs, but investigation shows that they occur |
465 | because of overload with legal connections, you should tune | |
466 | another parameters until this warning disappear. | |
467 | See: tcp_max_syn_backlog, tcp_synack_retries, tcp_abort_on_overflow. | |
1da177e4 | 468 | |
ef56e622 SH |
469 | syncookies seriously violate TCP protocol, do not allow |
470 | to use TCP extensions, can result in serious degradation | |
471 | of some services (f.e. SMTP relaying), visible not by you, | |
472 | but your clients and relays, contacting you. While you see | |
4edc2f34 | 473 | SYN flood warnings in logs not being really flooded, your server |
ef56e622 | 474 | is seriously misconfigured. |
1da177e4 | 475 | |
5ad37d5d HFS |
476 | If you want to test which effects syncookies have to your |
477 | network connections you can set this knob to 2 to enable | |
478 | unconditionally generation of syncookies. | |
479 | ||
cf60af03 YC |
480 | tcp_fastopen - INTEGER |
481 | Enable TCP Fast Open feature (draft-ietf-tcpm-fastopen) to send data | |
482 | in the opening SYN packet. To use this feature, the client application | |
10467163 JC |
483 | must use sendmsg() or sendto() with MSG_FASTOPEN flag rather than |
484 | connect() to perform a TCP handshake automatically. | |
485 | ||
486 | The values (bitmap) are | |
0d41cca4 | 487 | 1: Enables sending data in the opening SYN on the client w/ MSG_FASTOPEN. |
10467163 JC |
488 | 2: Enables TCP Fast Open on the server side, i.e., allowing data in |
489 | a SYN packet to be accepted and passed to the application before | |
490 | 3-way hand shake finishes. | |
491 | 4: Send data in the opening SYN regardless of cookie availability and | |
492 | without a cookie option. | |
493 | 0x100: Accept SYN data w/o validating the cookie. | |
494 | 0x200: Accept data-in-SYN w/o any cookie option present. | |
495 | 0x400/0x800: Enable Fast Open on all listeners regardless of the | |
496 | TCP_FASTOPEN socket option. The two different flags designate two | |
497 | different ways of setting max_qlen without the TCP_FASTOPEN socket | |
498 | option. | |
cf60af03 | 499 | |
0d41cca4 | 500 | Default: 1 |
cf60af03 | 501 | |
10467163 JC |
502 | Note that the client & server side Fast Open flags (1 and 2 |
503 | respectively) must be also enabled before the rest of flags can take | |
504 | effect. | |
505 | ||
506 | See include/net/tcp.h and the code for more details. | |
507 | ||
ef56e622 SH |
508 | tcp_syn_retries - INTEGER |
509 | Number of times initial SYNs for an active TCP connection attempt | |
510 | will be retransmitted. Should not be higher than 255. Default value | |
3b09adcb | 511 | is 6, which corresponds to 63seconds till the last retransmission |
6c9ff979 AB |
512 | with the current initial RTO of 1second. With this the final timeout |
513 | for an active TCP connection attempt will happen after 127seconds. | |
ef56e622 SH |
514 | |
515 | tcp_timestamps - BOOLEAN | |
516 | Enable timestamps as defined in RFC1323. | |
1da177e4 | 517 | |
95bd09eb ED |
518 | tcp_min_tso_segs - INTEGER |
519 | Minimal number of segments per TSO frame. | |
520 | Since linux-3.12, TCP does an automatic sizing of TSO frames, | |
521 | depending on flow rate, instead of filling 64Kbytes packets. | |
522 | For specific usages, it's possible to force TCP to build big | |
523 | TSO frames. Note that TCP stack might split too big TSO packets | |
524 | if available window is too small. | |
525 | Default: 2 | |
526 | ||
1da177e4 | 527 | tcp_tso_win_divisor - INTEGER |
ef56e622 SH |
528 | This allows control over what percentage of the congestion window |
529 | can be consumed by a single TSO frame. | |
530 | The setting of this parameter is a choice between burstiness and | |
531 | building larger TSO frames. | |
532 | Default: 3 | |
1da177e4 | 533 | |
ef56e622 SH |
534 | tcp_tw_recycle - BOOLEAN |
535 | Enable fast recycling TIME-WAIT sockets. Default value is 0. | |
536 | It should not be changed without advice/request of technical | |
537 | experts. | |
1da177e4 | 538 | |
ef56e622 SH |
539 | tcp_tw_reuse - BOOLEAN |
540 | Allow to reuse TIME-WAIT sockets for new connections when it is | |
541 | safe from protocol viewpoint. Default value is 0. | |
542 | It should not be changed without advice/request of technical | |
543 | experts. | |
ce7bc3bf | 544 | |
ef56e622 SH |
545 | tcp_window_scaling - BOOLEAN |
546 | Enable window scaling as defined in RFC1323. | |
3ff825b2 | 547 | |
ef56e622 | 548 | tcp_wmem - vector of 3 INTEGERs: min, default, max |
53025f5e | 549 | min: Amount of memory reserved for send buffers for TCP sockets. |
ef56e622 | 550 | Each TCP socket has rights to use it due to fact of its birth. |
6539fefd | 551 | Default: 1 page |
9d7bcfc6 | 552 | |
53025f5e BF |
553 | default: initial size of send buffer used by TCP sockets. This |
554 | value overrides net.core.wmem_default used by other protocols. | |
555 | It is usually lower than net.core.wmem_default. | |
ef56e622 SH |
556 | Default: 16K |
557 | ||
53025f5e BF |
558 | max: Maximal amount of memory allowed for automatically tuned |
559 | send buffers for TCP sockets. This value does not override | |
560 | net.core.wmem_max. Calling setsockopt() with SO_SNDBUF disables | |
561 | automatic tuning of that socket's send buffer size, in which case | |
562 | this value is ignored. | |
563 | Default: between 64K and 4MB, depending on RAM size. | |
1da177e4 | 564 | |
c9bee3b7 ED |
565 | tcp_notsent_lowat - UNSIGNED INTEGER |
566 | A TCP socket can control the amount of unsent bytes in its write queue, | |
567 | thanks to TCP_NOTSENT_LOWAT socket option. poll()/select()/epoll() | |
568 | reports POLLOUT events if the amount of unsent bytes is below a per | |
569 | socket value, and if the write queue is not full. sendmsg() will | |
570 | also not add new buffers if the limit is hit. | |
571 | ||
572 | This global variable controls the amount of unsent data for | |
573 | sockets not using TCP_NOTSENT_LOWAT. For these sockets, a change | |
574 | to the global variable has immediate effect. | |
575 | ||
576 | Default: UINT_MAX (0xFFFFFFFF) | |
577 | ||
15d99e02 RJ |
578 | tcp_workaround_signed_windows - BOOLEAN |
579 | If set, assume no receipt of a window scaling option means the | |
580 | remote TCP is broken and treats the window as a signed quantity. | |
581 | If unset, assume the remote TCP is not broken even if we do | |
582 | not receive a window scaling option from them. | |
583 | Default: 0 | |
584 | ||
36e31b0a AP |
585 | tcp_thin_linear_timeouts - BOOLEAN |
586 | Enable dynamic triggering of linear timeouts for thin streams. | |
587 | If set, a check is performed upon retransmission by timeout to | |
588 | determine if the stream is thin (less than 4 packets in flight). | |
589 | As long as the stream is found to be thin, up to 6 linear | |
590 | timeouts may be performed before exponential backoff mode is | |
591 | initiated. This improves retransmission latency for | |
592 | non-aggressive thin streams, often found to be time-dependent. | |
593 | For more information on thin streams, see | |
594 | Documentation/networking/tcp-thin.txt | |
595 | Default: 0 | |
596 | ||
7e380175 AP |
597 | tcp_thin_dupack - BOOLEAN |
598 | Enable dynamic triggering of retransmissions after one dupACK | |
599 | for thin streams. If set, a check is performed upon reception | |
600 | of a dupACK to determine if the stream is thin (less than 4 | |
601 | packets in flight). As long as the stream is found to be thin, | |
602 | data is retransmitted on the first received dupACK. This | |
603 | improves retransmission latency for non-aggressive thin | |
604 | streams, often found to be time-dependent. | |
605 | For more information on thin streams, see | |
606 | Documentation/networking/tcp-thin.txt | |
607 | Default: 0 | |
608 | ||
46d3ceab ED |
609 | tcp_limit_output_bytes - INTEGER |
610 | Controls TCP Small Queue limit per tcp socket. | |
611 | TCP bulk sender tends to increase packets in flight until it | |
612 | gets losses notifications. With SNDBUF autotuning, this can | |
613 | result in a large amount of packets queued in qdisc/device | |
614 | on the local machine, hurting latency of other flows, for | |
615 | typical pfifo_fast qdiscs. | |
616 | tcp_limit_output_bytes limits the number of bytes on qdisc | |
617 | or device to reduce artificial RTT/cwnd and reduce bufferbloat. | |
46d3ceab ED |
618 | Default: 131072 |
619 | ||
282f23c6 ED |
620 | tcp_challenge_ack_limit - INTEGER |
621 | Limits number of Challenge ACK sent per second, as recommended | |
622 | in RFC 5961 (Improving TCP's Robustness to Blind In-Window Attacks) | |
623 | Default: 100 | |
624 | ||
95766fff HA |
625 | UDP variables: |
626 | ||
627 | udp_mem - vector of 3 INTEGERs: min, pressure, max | |
628 | Number of pages allowed for queueing by all UDP sockets. | |
629 | ||
630 | min: Below this number of pages UDP is not bothered about its | |
631 | memory appetite. When amount of memory allocated by UDP exceeds | |
632 | this number, UDP starts to moderate memory usage. | |
633 | ||
634 | pressure: This value was introduced to follow format of tcp_mem. | |
635 | ||
636 | max: Number of pages allowed for queueing by all UDP sockets. | |
637 | ||
638 | Default is calculated at boot time from amount of available memory. | |
639 | ||
640 | udp_rmem_min - INTEGER | |
641 | Minimal size of receive buffer used by UDP sockets in moderation. | |
642 | Each UDP socket is able to use the size for receiving data, even if | |
643 | total pages of UDP sockets exceed udp_mem pressure. The unit is byte. | |
6539fefd | 644 | Default: 1 page |
95766fff HA |
645 | |
646 | udp_wmem_min - INTEGER | |
647 | Minimal size of send buffer used by UDP sockets in moderation. | |
648 | Each UDP socket is able to use the size for sending data, even if | |
649 | total pages of UDP sockets exceed udp_mem pressure. The unit is byte. | |
6539fefd | 650 | Default: 1 page |
95766fff | 651 | |
8802f616 PM |
652 | CIPSOv4 Variables: |
653 | ||
654 | cipso_cache_enable - BOOLEAN | |
655 | If set, enable additions to and lookups from the CIPSO label mapping | |
656 | cache. If unset, additions are ignored and lookups always result in a | |
657 | miss. However, regardless of the setting the cache is still | |
658 | invalidated when required when means you can safely toggle this on and | |
659 | off and the cache will always be "safe". | |
660 | Default: 1 | |
661 | ||
662 | cipso_cache_bucket_size - INTEGER | |
663 | The CIPSO label cache consists of a fixed size hash table with each | |
664 | hash bucket containing a number of cache entries. This variable limits | |
665 | the number of entries in each hash bucket; the larger the value the | |
666 | more CIPSO label mappings that can be cached. When the number of | |
667 | entries in a given hash bucket reaches this limit adding new entries | |
668 | causes the oldest entry in the bucket to be removed to make room. | |
669 | Default: 10 | |
670 | ||
671 | cipso_rbm_optfmt - BOOLEAN | |
672 | Enable the "Optimized Tag 1 Format" as defined in section 3.4.2.6 of | |
673 | the CIPSO draft specification (see Documentation/netlabel for details). | |
674 | This means that when set the CIPSO tag will be padded with empty | |
675 | categories in order to make the packet data 32-bit aligned. | |
676 | Default: 0 | |
677 | ||
678 | cipso_rbm_structvalid - BOOLEAN | |
679 | If set, do a very strict check of the CIPSO option when | |
680 | ip_options_compile() is called. If unset, relax the checks done during | |
681 | ip_options_compile(). Either way is "safe" as errors are caught else | |
682 | where in the CIPSO processing code but setting this to 0 (False) should | |
683 | result in less work (i.e. it should be faster) but could cause problems | |
684 | with other implementations that require strict checking. | |
685 | Default: 0 | |
686 | ||
1da177e4 LT |
687 | IP Variables: |
688 | ||
689 | ip_local_port_range - 2 INTEGERS | |
690 | Defines the local port range that is used by TCP and UDP to | |
e18f5feb | 691 | choose the local port. The first number is the first, the |
5d6bd861 FLVC |
692 | second the last local port number. The default values are |
693 | 32768 and 61000 respectively. | |
1da177e4 | 694 | |
e3826f1e AW |
695 | ip_local_reserved_ports - list of comma separated ranges |
696 | Specify the ports which are reserved for known third-party | |
697 | applications. These ports will not be used by automatic port | |
698 | assignments (e.g. when calling connect() or bind() with port | |
699 | number 0). Explicit port allocation behavior is unchanged. | |
700 | ||
701 | The format used for both input and output is a comma separated | |
702 | list of ranges (e.g. "1,2-4,10-10" for ports 1, 2, 3, 4 and | |
703 | 10). Writing to the file will clear all previously reserved | |
704 | ports and update the current list with the one given in the | |
705 | input. | |
706 | ||
707 | Note that ip_local_port_range and ip_local_reserved_ports | |
708 | settings are independent and both are considered by the kernel | |
709 | when determining which ports are available for automatic port | |
710 | assignments. | |
711 | ||
712 | You can reserve ports which are not in the current | |
713 | ip_local_port_range, e.g.: | |
714 | ||
715 | $ cat /proc/sys/net/ipv4/ip_local_port_range | |
716 | 32000 61000 | |
717 | $ cat /proc/sys/net/ipv4/ip_local_reserved_ports | |
718 | 8080,9148 | |
719 | ||
720 | although this is redundant. However such a setting is useful | |
721 | if later the port range is changed to a value that will | |
722 | include the reserved ports. | |
723 | ||
724 | Default: Empty | |
725 | ||
1da177e4 LT |
726 | ip_nonlocal_bind - BOOLEAN |
727 | If set, allows processes to bind() to non-local IP addresses, | |
728 | which can be quite useful - but may break some applications. | |
729 | Default: 0 | |
730 | ||
731 | ip_dynaddr - BOOLEAN | |
732 | If set non-zero, enables support for dynamic addresses. | |
733 | If set to a non-zero value larger than 1, a kernel log | |
734 | message will be printed when dynamic address rewriting | |
735 | occurs. | |
736 | Default: 0 | |
737 | ||
e3d73bce CW |
738 | ip_early_demux - BOOLEAN |
739 | Optimize input packet processing down to one demux for | |
740 | certain kinds of local sockets. Currently we only do this | |
741 | for established TCP sockets. | |
742 | ||
743 | It may add an additional cost for pure routing workloads that | |
744 | reduces overall throughput, in such case you should disable it. | |
745 | Default: 1 | |
746 | ||
1da177e4 | 747 | icmp_echo_ignore_all - BOOLEAN |
7ce31246 DM |
748 | If set non-zero, then the kernel will ignore all ICMP ECHO |
749 | requests sent to it. | |
750 | Default: 0 | |
751 | ||
1da177e4 | 752 | icmp_echo_ignore_broadcasts - BOOLEAN |
7ce31246 DM |
753 | If set non-zero, then the kernel will ignore all ICMP ECHO and |
754 | TIMESTAMP requests sent to it via broadcast/multicast. | |
755 | Default: 1 | |
1da177e4 LT |
756 | |
757 | icmp_ratelimit - INTEGER | |
758 | Limit the maximal rates for sending ICMP packets whose type matches | |
759 | icmp_ratemask (see below) to specific targets. | |
6dbf4bca SH |
760 | 0 to disable any limiting, |
761 | otherwise the minimal space between responses in milliseconds. | |
762 | Default: 1000 | |
1da177e4 LT |
763 | |
764 | icmp_ratemask - INTEGER | |
765 | Mask made of ICMP types for which rates are being limited. | |
766 | Significant bits: IHGFEDCBA9876543210 | |
767 | Default mask: 0000001100000011000 (6168) | |
768 | ||
769 | Bit definitions (see include/linux/icmp.h): | |
770 | 0 Echo Reply | |
771 | 3 Destination Unreachable * | |
772 | 4 Source Quench * | |
773 | 5 Redirect | |
774 | 8 Echo Request | |
775 | B Time Exceeded * | |
776 | C Parameter Problem * | |
777 | D Timestamp Request | |
778 | E Timestamp Reply | |
779 | F Info Request | |
780 | G Info Reply | |
781 | H Address Mask Request | |
782 | I Address Mask Reply | |
783 | ||
784 | * These are rate limited by default (see default mask above) | |
785 | ||
786 | icmp_ignore_bogus_error_responses - BOOLEAN | |
787 | Some routers violate RFC1122 by sending bogus responses to broadcast | |
788 | frames. Such violations are normally logged via a kernel warning. | |
789 | If this is set to TRUE, the kernel will not give such warnings, which | |
790 | will avoid log file clutter. | |
e8b265e8 | 791 | Default: 1 |
1da177e4 | 792 | |
95f7daf1 H |
793 | icmp_errors_use_inbound_ifaddr - BOOLEAN |
794 | ||
795 | If zero, icmp error messages are sent with the primary address of | |
796 | the exiting interface. | |
e18f5feb | 797 | |
95f7daf1 H |
798 | If non-zero, the message will be sent with the primary address of |
799 | the interface that received the packet that caused the icmp error. | |
800 | This is the behaviour network many administrators will expect from | |
801 | a router. And it can make debugging complicated network layouts | |
e18f5feb | 802 | much easier. |
95f7daf1 H |
803 | |
804 | Note that if no primary address exists for the interface selected, | |
805 | then the primary address of the first non-loopback interface that | |
d6bc8ac9 | 806 | has one will be used regardless of this setting. |
95f7daf1 H |
807 | |
808 | Default: 0 | |
809 | ||
1da177e4 LT |
810 | igmp_max_memberships - INTEGER |
811 | Change the maximum number of multicast groups we can subscribe to. | |
812 | Default: 20 | |
813 | ||
d67ef35f JE |
814 | Theoretical maximum value is bounded by having to send a membership |
815 | report in a single datagram (i.e. the report can't span multiple | |
816 | datagrams, or risk confusing the switch and leaving groups you don't | |
817 | intend to). | |
1da177e4 | 818 | |
d67ef35f JE |
819 | The number of supported groups 'M' is bounded by the number of group |
820 | report entries you can fit into a single datagram of 65535 bytes. | |
821 | ||
822 | M = 65536-sizeof (ip header)/(sizeof(Group record)) | |
823 | ||
824 | Group records are variable length, with a minimum of 12 bytes. | |
825 | So net.ipv4.igmp_max_memberships should not be set higher than: | |
826 | ||
827 | (65536-24) / 12 = 5459 | |
828 | ||
829 | The value 5459 assumes no IP header options, so in practice | |
830 | this number may be lower. | |
831 | ||
832 | conf/interface/* changes special settings per interface (where | |
833 | "interface" is the name of your network interface) | |
834 | ||
835 | conf/all/* is special, changes the settings for all interfaces | |
1da177e4 LT |
836 | |
837 | log_martians - BOOLEAN | |
838 | Log packets with impossible addresses to kernel log. | |
839 | log_martians for the interface will be enabled if at least one of | |
840 | conf/{all,interface}/log_martians is set to TRUE, | |
841 | it will be disabled otherwise | |
842 | ||
843 | accept_redirects - BOOLEAN | |
844 | Accept ICMP redirect messages. | |
845 | accept_redirects for the interface will be enabled if: | |
e18f5feb JDB |
846 | - both conf/{all,interface}/accept_redirects are TRUE in the case |
847 | forwarding for the interface is enabled | |
1da177e4 | 848 | or |
e18f5feb JDB |
849 | - at least one of conf/{all,interface}/accept_redirects is TRUE in the |
850 | case forwarding for the interface is disabled | |
1da177e4 LT |
851 | accept_redirects for the interface will be disabled otherwise |
852 | default TRUE (host) | |
853 | FALSE (router) | |
854 | ||
855 | forwarding - BOOLEAN | |
856 | Enable IP forwarding on this interface. | |
857 | ||
858 | mc_forwarding - BOOLEAN | |
859 | Do multicast routing. The kernel needs to be compiled with CONFIG_MROUTE | |
860 | and a multicast routing daemon is required. | |
e18f5feb JDB |
861 | conf/all/mc_forwarding must also be set to TRUE to enable multicast |
862 | routing for the interface | |
1da177e4 LT |
863 | |
864 | medium_id - INTEGER | |
865 | Integer value used to differentiate the devices by the medium they | |
866 | are attached to. Two devices can have different id values when | |
867 | the broadcast packets are received only on one of them. | |
868 | The default value 0 means that the device is the only interface | |
869 | to its medium, value of -1 means that medium is not known. | |
e18f5feb | 870 | |
1da177e4 LT |
871 | Currently, it is used to change the proxy_arp behavior: |
872 | the proxy_arp feature is enabled for packets forwarded between | |
873 | two devices attached to different media. | |
874 | ||
875 | proxy_arp - BOOLEAN | |
876 | Do proxy arp. | |
877 | proxy_arp for the interface will be enabled if at least one of | |
878 | conf/{all,interface}/proxy_arp is set to TRUE, | |
879 | it will be disabled otherwise | |
880 | ||
65324144 JDB |
881 | proxy_arp_pvlan - BOOLEAN |
882 | Private VLAN proxy arp. | |
883 | Basically allow proxy arp replies back to the same interface | |
884 | (from which the ARP request/solicitation was received). | |
885 | ||
886 | This is done to support (ethernet) switch features, like RFC | |
887 | 3069, where the individual ports are NOT allowed to | |
888 | communicate with each other, but they are allowed to talk to | |
889 | the upstream router. As described in RFC 3069, it is possible | |
890 | to allow these hosts to communicate through the upstream | |
891 | router by proxy_arp'ing. Don't need to be used together with | |
892 | proxy_arp. | |
893 | ||
894 | This technology is known by different names: | |
895 | In RFC 3069 it is called VLAN Aggregation. | |
896 | Cisco and Allied Telesyn call it Private VLAN. | |
897 | Hewlett-Packard call it Source-Port filtering or port-isolation. | |
898 | Ericsson call it MAC-Forced Forwarding (RFC Draft). | |
899 | ||
1da177e4 LT |
900 | shared_media - BOOLEAN |
901 | Send(router) or accept(host) RFC1620 shared media redirects. | |
902 | Overrides ip_secure_redirects. | |
903 | shared_media for the interface will be enabled if at least one of | |
904 | conf/{all,interface}/shared_media is set to TRUE, | |
905 | it will be disabled otherwise | |
906 | default TRUE | |
907 | ||
908 | secure_redirects - BOOLEAN | |
909 | Accept ICMP redirect messages only for gateways, | |
910 | listed in default gateway list. | |
911 | secure_redirects for the interface will be enabled if at least one of | |
912 | conf/{all,interface}/secure_redirects is set to TRUE, | |
913 | it will be disabled otherwise | |
914 | default TRUE | |
915 | ||
916 | send_redirects - BOOLEAN | |
917 | Send redirects, if router. | |
918 | send_redirects for the interface will be enabled if at least one of | |
919 | conf/{all,interface}/send_redirects is set to TRUE, | |
920 | it will be disabled otherwise | |
921 | Default: TRUE | |
922 | ||
923 | bootp_relay - BOOLEAN | |
924 | Accept packets with source address 0.b.c.d destined | |
925 | not to this host as local ones. It is supposed, that | |
926 | BOOTP relay daemon will catch and forward such packets. | |
927 | conf/all/bootp_relay must also be set to TRUE to enable BOOTP relay | |
928 | for the interface | |
929 | default FALSE | |
930 | Not Implemented Yet. | |
931 | ||
932 | accept_source_route - BOOLEAN | |
933 | Accept packets with SRR option. | |
934 | conf/all/accept_source_route must also be set to TRUE to accept packets | |
935 | with SRR option on the interface | |
936 | default TRUE (router) | |
937 | FALSE (host) | |
938 | ||
8153a10c | 939 | accept_local - BOOLEAN |
c801e3cc DM |
940 | Accept packets with local source addresses. In combination |
941 | with suitable routing, this can be used to direct packets | |
942 | between two local interfaces over the wire and have them | |
943 | accepted properly. | |
944 | ||
945 | rp_filter must be set to a non-zero value in order for | |
946 | accept_local to have an effect. | |
947 | ||
8153a10c PM |
948 | default FALSE |
949 | ||
d0daebc3 TG |
950 | route_localnet - BOOLEAN |
951 | Do not consider loopback addresses as martian source or destination | |
952 | while routing. This enables the use of 127/8 for local routing purposes. | |
953 | default FALSE | |
954 | ||
c1cf8422 | 955 | rp_filter - INTEGER |
1da177e4 | 956 | 0 - No source validation. |
c1cf8422 SH |
957 | 1 - Strict mode as defined in RFC3704 Strict Reverse Path |
958 | Each incoming packet is tested against the FIB and if the interface | |
959 | is not the best reverse path the packet check will fail. | |
960 | By default failed packets are discarded. | |
961 | 2 - Loose mode as defined in RFC3704 Loose Reverse Path | |
962 | Each incoming packet's source address is also tested against the FIB | |
963 | and if the source address is not reachable via any interface | |
964 | the packet check will fail. | |
965 | ||
e18f5feb | 966 | Current recommended practice in RFC3704 is to enable strict mode |
bf869c30 | 967 | to prevent IP spoofing from DDos attacks. If using asymmetric routing |
e18f5feb | 968 | or other complicated routing, then loose mode is recommended. |
c1cf8422 | 969 | |
1f5865e7 SW |
970 | The max value from conf/{all,interface}/rp_filter is used |
971 | when doing source validation on the {interface}. | |
1da177e4 LT |
972 | |
973 | Default value is 0. Note that some distributions enable it | |
974 | in startup scripts. | |
975 | ||
976 | arp_filter - BOOLEAN | |
977 | 1 - Allows you to have multiple network interfaces on the same | |
978 | subnet, and have the ARPs for each interface be answered | |
979 | based on whether or not the kernel would route a packet from | |
980 | the ARP'd IP out that interface (therefore you must use source | |
981 | based routing for this to work). In other words it allows control | |
982 | of which cards (usually 1) will respond to an arp request. | |
983 | ||
984 | 0 - (default) The kernel can respond to arp requests with addresses | |
985 | from other interfaces. This may seem wrong but it usually makes | |
986 | sense, because it increases the chance of successful communication. | |
987 | IP addresses are owned by the complete host on Linux, not by | |
988 | particular interfaces. Only for more complex setups like load- | |
989 | balancing, does this behaviour cause problems. | |
990 | ||
991 | arp_filter for the interface will be enabled if at least one of | |
992 | conf/{all,interface}/arp_filter is set to TRUE, | |
993 | it will be disabled otherwise | |
994 | ||
995 | arp_announce - INTEGER | |
996 | Define different restriction levels for announcing the local | |
997 | source IP address from IP packets in ARP requests sent on | |
998 | interface: | |
999 | 0 - (default) Use any local address, configured on any interface | |
1000 | 1 - Try to avoid local addresses that are not in the target's | |
1001 | subnet for this interface. This mode is useful when target | |
1002 | hosts reachable via this interface require the source IP | |
1003 | address in ARP requests to be part of their logical network | |
1004 | configured on the receiving interface. When we generate the | |
1005 | request we will check all our subnets that include the | |
1006 | target IP and will preserve the source address if it is from | |
1007 | such subnet. If there is no such subnet we select source | |
1008 | address according to the rules for level 2. | |
1009 | 2 - Always use the best local address for this target. | |
1010 | In this mode we ignore the source address in the IP packet | |
1011 | and try to select local address that we prefer for talks with | |
1012 | the target host. Such local address is selected by looking | |
1013 | for primary IP addresses on all our subnets on the outgoing | |
1014 | interface that include the target IP address. If no suitable | |
1015 | local address is found we select the first local address | |
1016 | we have on the outgoing interface or on all other interfaces, | |
1017 | with the hope we will receive reply for our request and | |
1018 | even sometimes no matter the source IP address we announce. | |
1019 | ||
1020 | The max value from conf/{all,interface}/arp_announce is used. | |
1021 | ||
1022 | Increasing the restriction level gives more chance for | |
1023 | receiving answer from the resolved target while decreasing | |
1024 | the level announces more valid sender's information. | |
1025 | ||
1026 | arp_ignore - INTEGER | |
1027 | Define different modes for sending replies in response to | |
1028 | received ARP requests that resolve local target IP addresses: | |
1029 | 0 - (default): reply for any local target IP address, configured | |
1030 | on any interface | |
1031 | 1 - reply only if the target IP address is local address | |
1032 | configured on the incoming interface | |
1033 | 2 - reply only if the target IP address is local address | |
1034 | configured on the incoming interface and both with the | |
1035 | sender's IP address are part from same subnet on this interface | |
1036 | 3 - do not reply for local addresses configured with scope host, | |
1037 | only resolutions for global and link addresses are replied | |
1038 | 4-7 - reserved | |
1039 | 8 - do not reply for all local addresses | |
1040 | ||
1041 | The max value from conf/{all,interface}/arp_ignore is used | |
1042 | when ARP request is received on the {interface} | |
1043 | ||
eefef1cf SH |
1044 | arp_notify - BOOLEAN |
1045 | Define mode for notification of address and device changes. | |
1046 | 0 - (default): do nothing | |
3f8dc236 | 1047 | 1 - Generate gratuitous arp requests when device is brought up |
eefef1cf SH |
1048 | or hardware address changes. |
1049 | ||
c1b1bce8 | 1050 | arp_accept - BOOLEAN |
6d955180 OP |
1051 | Define behavior for gratuitous ARP frames who's IP is not |
1052 | already present in the ARP table: | |
1053 | 0 - don't create new entries in the ARP table | |
1054 | 1 - create new entries in the ARP table | |
1055 | ||
1056 | Both replies and requests type gratuitous arp will trigger the | |
1057 | ARP table to be updated, if this setting is on. | |
1058 | ||
1059 | If the ARP table already contains the IP address of the | |
1060 | gratuitous arp frame, the arp table will be updated regardless | |
1061 | if this setting is on or off. | |
1062 | ||
c1b1bce8 | 1063 | |
1da177e4 LT |
1064 | app_solicit - INTEGER |
1065 | The maximum number of probes to send to the user space ARP daemon | |
1066 | via netlink before dropping back to multicast probes (see | |
1067 | mcast_solicit). Defaults to 0. | |
1068 | ||
1069 | disable_policy - BOOLEAN | |
1070 | Disable IPSEC policy (SPD) for this interface | |
1071 | ||
1072 | disable_xfrm - BOOLEAN | |
1073 | Disable IPSEC encryption on this interface, whatever the policy | |
1074 | ||
fc4eba58 HFS |
1075 | igmpv2_unsolicited_report_interval - INTEGER |
1076 | The interval in milliseconds in which the next unsolicited | |
1077 | IGMPv1 or IGMPv2 report retransmit will take place. | |
1078 | Default: 10000 (10 seconds) | |
1da177e4 | 1079 | |
fc4eba58 HFS |
1080 | igmpv3_unsolicited_report_interval - INTEGER |
1081 | The interval in milliseconds in which the next unsolicited | |
1082 | IGMPv3 report retransmit will take place. | |
1083 | Default: 1000 (1 seconds) | |
1da177e4 | 1084 | |
d922e1cb MS |
1085 | promote_secondaries - BOOLEAN |
1086 | When a primary IP address is removed from this interface | |
1087 | promote a corresponding secondary IP address instead of | |
1088 | removing all the corresponding secondary IP addresses. | |
1089 | ||
1090 | ||
1da177e4 LT |
1091 | tag - INTEGER |
1092 | Allows you to write a number, which can be used as required. | |
1093 | Default value is 0. | |
1094 | ||
1da177e4 LT |
1095 | Alexey Kuznetsov. |
1096 | kuznet@ms2.inr.ac.ru | |
1097 | ||
1098 | Updated by: | |
1099 | Andi Kleen | |
1100 | ak@muc.de | |
1101 | Nicolas Delon | |
1102 | delon.nicolas@wanadoo.fr | |
1103 | ||
1104 | ||
1105 | ||
1106 | ||
1107 | /proc/sys/net/ipv6/* Variables: | |
1108 | ||
1109 | IPv6 has no global variables such as tcp_*. tcp_* settings under ipv4/ also | |
1110 | apply to IPv6 [XXX?]. | |
1111 | ||
1112 | bindv6only - BOOLEAN | |
1113 | Default value for IPV6_V6ONLY socket option, | |
e18f5feb | 1114 | which restricts use of the IPv6 socket to IPv6 communication |
1da177e4 LT |
1115 | only. |
1116 | TRUE: disable IPv4-mapped address feature | |
1117 | FALSE: enable IPv4-mapped address feature | |
1118 | ||
d5c073ca | 1119 | Default: FALSE (as specified in RFC3493) |
1da177e4 | 1120 | |
6444f72b FF |
1121 | flowlabel_consistency - BOOLEAN |
1122 | Protect the consistency (and unicity) of flow label. | |
1123 | You have to disable it to use IPV6_FL_F_REFLECT flag on the | |
1124 | flow label manager. | |
1125 | TRUE: enabled | |
1126 | FALSE: disabled | |
1127 | Default: TRUE | |
1128 | ||
509aba3b FLB |
1129 | anycast_src_echo_reply - BOOLEAN |
1130 | Controls the use of anycast addresses as source addresses for ICMPv6 | |
1131 | echo reply | |
1132 | TRUE: enabled | |
1133 | FALSE: disabled | |
1134 | Default: FALSE | |
1135 | ||
1da177e4 LT |
1136 | IPv6 Fragmentation: |
1137 | ||
1138 | ip6frag_high_thresh - INTEGER | |
e18f5feb | 1139 | Maximum memory used to reassemble IPv6 fragments. When |
1da177e4 LT |
1140 | ip6frag_high_thresh bytes of memory is allocated for this purpose, |
1141 | the fragment handler will toss packets until ip6frag_low_thresh | |
1142 | is reached. | |
e18f5feb | 1143 | |
1da177e4 | 1144 | ip6frag_low_thresh - INTEGER |
e18f5feb | 1145 | See ip6frag_high_thresh |
1da177e4 LT |
1146 | |
1147 | ip6frag_time - INTEGER | |
1148 | Time in seconds to keep an IPv6 fragment in memory. | |
1149 | ||
1150 | ip6frag_secret_interval - INTEGER | |
e18f5feb | 1151 | Regeneration interval (in seconds) of the hash secret (or lifetime |
1da177e4 LT |
1152 | for the hash secret) for IPv6 fragments. |
1153 | Default: 600 | |
1154 | ||
1155 | conf/default/*: | |
1156 | Change the interface-specific default settings. | |
1157 | ||
1158 | ||
1159 | conf/all/*: | |
e18f5feb | 1160 | Change all the interface-specific settings. |
1da177e4 LT |
1161 | |
1162 | [XXX: Other special features than forwarding?] | |
1163 | ||
1164 | conf/all/forwarding - BOOLEAN | |
e18f5feb | 1165 | Enable global IPv6 forwarding between all interfaces. |
1da177e4 | 1166 | |
e18f5feb | 1167 | IPv4 and IPv6 work differently here; e.g. netfilter must be used |
1da177e4 LT |
1168 | to control which interfaces may forward packets and which not. |
1169 | ||
e18f5feb | 1170 | This also sets all interfaces' Host/Router setting |
1da177e4 LT |
1171 | 'forwarding' to the specified value. See below for details. |
1172 | ||
1173 | This referred to as global forwarding. | |
1174 | ||
fbea49e1 YH |
1175 | proxy_ndp - BOOLEAN |
1176 | Do proxy ndp. | |
1177 | ||
1da177e4 LT |
1178 | conf/interface/*: |
1179 | Change special settings per interface. | |
1180 | ||
e18f5feb | 1181 | The functional behaviour for certain settings is different |
1da177e4 LT |
1182 | depending on whether local forwarding is enabled or not. |
1183 | ||
605b91c8 | 1184 | accept_ra - INTEGER |
1da177e4 | 1185 | Accept Router Advertisements; autoconfigure using them. |
e18f5feb | 1186 | |
026359bc TA |
1187 | It also determines whether or not to transmit Router |
1188 | Solicitations. If and only if the functional setting is to | |
1189 | accept Router Advertisements, Router Solicitations will be | |
1190 | transmitted. | |
1191 | ||
ae8abfa0 TG |
1192 | Possible values are: |
1193 | 0 Do not accept Router Advertisements. | |
1194 | 1 Accept Router Advertisements if forwarding is disabled. | |
1195 | 2 Overrule forwarding behaviour. Accept Router Advertisements | |
1196 | even if forwarding is enabled. | |
1197 | ||
1da177e4 LT |
1198 | Functional default: enabled if local forwarding is disabled. |
1199 | disabled if local forwarding is enabled. | |
1200 | ||
65f5c7c1 YH |
1201 | accept_ra_defrtr - BOOLEAN |
1202 | Learn default router in Router Advertisement. | |
1203 | ||
1204 | Functional default: enabled if accept_ra is enabled. | |
1205 | disabled if accept_ra is disabled. | |
1206 | ||
c4fd30eb | 1207 | accept_ra_pinfo - BOOLEAN |
2fe0ae78 | 1208 | Learn Prefix Information in Router Advertisement. |
c4fd30eb YH |
1209 | |
1210 | Functional default: enabled if accept_ra is enabled. | |
1211 | disabled if accept_ra is disabled. | |
1212 | ||
09c884d4 YH |
1213 | accept_ra_rt_info_max_plen - INTEGER |
1214 | Maximum prefix length of Route Information in RA. | |
1215 | ||
1216 | Route Information w/ prefix larger than or equal to this | |
1217 | variable shall be ignored. | |
1218 | ||
1219 | Functional default: 0 if accept_ra_rtr_pref is enabled. | |
1220 | -1 if accept_ra_rtr_pref is disabled. | |
1221 | ||
930d6ff2 YH |
1222 | accept_ra_rtr_pref - BOOLEAN |
1223 | Accept Router Preference in RA. | |
1224 | ||
1225 | Functional default: enabled if accept_ra is enabled. | |
1226 | disabled if accept_ra is disabled. | |
1227 | ||
1da177e4 LT |
1228 | accept_redirects - BOOLEAN |
1229 | Accept Redirects. | |
1230 | ||
1231 | Functional default: enabled if local forwarding is disabled. | |
1232 | disabled if local forwarding is enabled. | |
1233 | ||
0bcbc926 YH |
1234 | accept_source_route - INTEGER |
1235 | Accept source routing (routing extension header). | |
1236 | ||
bb4dbf9e | 1237 | >= 0: Accept only routing header type 2. |
0bcbc926 YH |
1238 | < 0: Do not accept routing header. |
1239 | ||
1240 | Default: 0 | |
1241 | ||
1da177e4 | 1242 | autoconf - BOOLEAN |
e18f5feb | 1243 | Autoconfigure addresses using Prefix Information in Router |
1da177e4 LT |
1244 | Advertisements. |
1245 | ||
c4fd30eb YH |
1246 | Functional default: enabled if accept_ra_pinfo is enabled. |
1247 | disabled if accept_ra_pinfo is disabled. | |
1da177e4 LT |
1248 | |
1249 | dad_transmits - INTEGER | |
1250 | The amount of Duplicate Address Detection probes to send. | |
1251 | Default: 1 | |
e18f5feb | 1252 | |
605b91c8 | 1253 | forwarding - INTEGER |
e18f5feb | 1254 | Configure interface-specific Host/Router behaviour. |
1da177e4 | 1255 | |
e18f5feb | 1256 | Note: It is recommended to have the same setting on all |
1da177e4 LT |
1257 | interfaces; mixed router/host scenarios are rather uncommon. |
1258 | ||
ae8abfa0 TG |
1259 | Possible values are: |
1260 | 0 Forwarding disabled | |
1261 | 1 Forwarding enabled | |
ae8abfa0 TG |
1262 | |
1263 | FALSE (0): | |
1da177e4 LT |
1264 | |
1265 | By default, Host behaviour is assumed. This means: | |
1266 | ||
1267 | 1. IsRouter flag is not set in Neighbour Advertisements. | |
026359bc TA |
1268 | 2. If accept_ra is TRUE (default), transmit Router |
1269 | Solicitations. | |
e18f5feb | 1270 | 3. If accept_ra is TRUE (default), accept Router |
1da177e4 LT |
1271 | Advertisements (and do autoconfiguration). |
1272 | 4. If accept_redirects is TRUE (default), accept Redirects. | |
1273 | ||
ae8abfa0 | 1274 | TRUE (1): |
1da177e4 | 1275 | |
e18f5feb | 1276 | If local forwarding is enabled, Router behaviour is assumed. |
1da177e4 LT |
1277 | This means exactly the reverse from the above: |
1278 | ||
1279 | 1. IsRouter flag is set in Neighbour Advertisements. | |
026359bc | 1280 | 2. Router Solicitations are not sent unless accept_ra is 2. |
ae8abfa0 | 1281 | 3. Router Advertisements are ignored unless accept_ra is 2. |
1da177e4 LT |
1282 | 4. Redirects are ignored. |
1283 | ||
ae8abfa0 TG |
1284 | Default: 0 (disabled) if global forwarding is disabled (default), |
1285 | otherwise 1 (enabled). | |
1da177e4 LT |
1286 | |
1287 | hop_limit - INTEGER | |
1288 | Default Hop Limit to set. | |
1289 | Default: 64 | |
1290 | ||
1291 | mtu - INTEGER | |
1292 | Default Maximum Transfer Unit | |
1293 | Default: 1280 (IPv6 required minimum) | |
1294 | ||
52e16356 YH |
1295 | router_probe_interval - INTEGER |
1296 | Minimum interval (in seconds) between Router Probing described | |
1297 | in RFC4191. | |
1298 | ||
1299 | Default: 60 | |
1300 | ||
1da177e4 LT |
1301 | router_solicitation_delay - INTEGER |
1302 | Number of seconds to wait after interface is brought up | |
1303 | before sending Router Solicitations. | |
1304 | Default: 1 | |
1305 | ||
1306 | router_solicitation_interval - INTEGER | |
1307 | Number of seconds to wait between Router Solicitations. | |
1308 | Default: 4 | |
1309 | ||
1310 | router_solicitations - INTEGER | |
e18f5feb | 1311 | Number of Router Solicitations to send until assuming no |
1da177e4 LT |
1312 | routers are present. |
1313 | Default: 3 | |
1314 | ||
1315 | use_tempaddr - INTEGER | |
1316 | Preference for Privacy Extensions (RFC3041). | |
1317 | <= 0 : disable Privacy Extensions | |
1318 | == 1 : enable Privacy Extensions, but prefer public | |
1319 | addresses over temporary addresses. | |
1320 | > 1 : enable Privacy Extensions and prefer temporary | |
1321 | addresses over public addresses. | |
1322 | Default: 0 (for most devices) | |
1323 | -1 (for point-to-point devices and loopback devices) | |
1324 | ||
1325 | temp_valid_lft - INTEGER | |
1326 | valid lifetime (in seconds) for temporary addresses. | |
1327 | Default: 604800 (7 days) | |
1328 | ||
1329 | temp_prefered_lft - INTEGER | |
1330 | Preferred lifetime (in seconds) for temporary addresses. | |
1331 | Default: 86400 (1 day) | |
1332 | ||
1333 | max_desync_factor - INTEGER | |
1334 | Maximum value for DESYNC_FACTOR, which is a random value | |
e18f5feb | 1335 | that ensures that clients don't synchronize with each |
1da177e4 LT |
1336 | other and generate new addresses at exactly the same time. |
1337 | value is in seconds. | |
1338 | Default: 600 | |
e18f5feb | 1339 | |
1da177e4 LT |
1340 | regen_max_retry - INTEGER |
1341 | Number of attempts before give up attempting to generate | |
1342 | valid temporary addresses. | |
1343 | Default: 5 | |
1344 | ||
1345 | max_addresses - INTEGER | |
e79dc484 BH |
1346 | Maximum number of autoconfigured addresses per interface. Setting |
1347 | to zero disables the limitation. It is not recommended to set this | |
1348 | value too large (or to zero) because it would be an easy way to | |
1349 | crash the kernel by allowing too many addresses to be created. | |
1da177e4 LT |
1350 | Default: 16 |
1351 | ||
778d80be | 1352 | disable_ipv6 - BOOLEAN |
9bdd8d40 BH |
1353 | Disable IPv6 operation. If accept_dad is set to 2, this value |
1354 | will be dynamically set to TRUE if DAD fails for the link-local | |
1355 | address. | |
778d80be YH |
1356 | Default: FALSE (enable IPv6 operation) |
1357 | ||
56d417b1 BH |
1358 | When this value is changed from 1 to 0 (IPv6 is being enabled), |
1359 | it will dynamically create a link-local address on the given | |
1360 | interface and start Duplicate Address Detection, if necessary. | |
1361 | ||
1362 | When this value is changed from 0 to 1 (IPv6 is being disabled), | |
1363 | it will dynamically delete all address on the given interface. | |
1364 | ||
1b34be74 YH |
1365 | accept_dad - INTEGER |
1366 | Whether to accept DAD (Duplicate Address Detection). | |
1367 | 0: Disable DAD | |
1368 | 1: Enable DAD (default) | |
1369 | 2: Enable DAD, and disable IPv6 operation if MAC-based duplicate | |
1370 | link-local address has been found. | |
1371 | ||
f7734fdf OP |
1372 | force_tllao - BOOLEAN |
1373 | Enable sending the target link-layer address option even when | |
1374 | responding to a unicast neighbor solicitation. | |
1375 | Default: FALSE | |
1376 | ||
1377 | Quoting from RFC 2461, section 4.4, Target link-layer address: | |
1378 | ||
1379 | "The option MUST be included for multicast solicitations in order to | |
1380 | avoid infinite Neighbor Solicitation "recursion" when the peer node | |
1381 | does not have a cache entry to return a Neighbor Advertisements | |
1382 | message. When responding to unicast solicitations, the option can be | |
1383 | omitted since the sender of the solicitation has the correct link- | |
1384 | layer address; otherwise it would not have be able to send the unicast | |
1385 | solicitation in the first place. However, including the link-layer | |
1386 | address in this case adds little overhead and eliminates a potential | |
1387 | race condition where the sender deletes the cached link-layer address | |
1388 | prior to receiving a response to a previous solicitation." | |
1389 | ||
db2b620a HFS |
1390 | ndisc_notify - BOOLEAN |
1391 | Define mode for notification of address and device changes. | |
1392 | 0 - (default): do nothing | |
1393 | 1 - Generate unsolicited neighbour advertisements when device is brought | |
1394 | up or hardware address changes. | |
1395 | ||
fc4eba58 HFS |
1396 | mldv1_unsolicited_report_interval - INTEGER |
1397 | The interval in milliseconds in which the next unsolicited | |
1398 | MLDv1 report retransmit will take place. | |
1399 | Default: 10000 (10 seconds) | |
1400 | ||
1401 | mldv2_unsolicited_report_interval - INTEGER | |
1402 | The interval in milliseconds in which the next unsolicited | |
1403 | MLDv2 report retransmit will take place. | |
1404 | Default: 1000 (1 second) | |
1405 | ||
f2127810 DB |
1406 | force_mld_version - INTEGER |
1407 | 0 - (default) No enforcement of a MLD version, MLDv1 fallback allowed | |
1408 | 1 - Enforce to use MLD version 1 | |
1409 | 2 - Enforce to use MLD version 2 | |
1410 | ||
b800c3b9 HFS |
1411 | suppress_frag_ndisc - INTEGER |
1412 | Control RFC 6980 (Security Implications of IPv6 Fragmentation | |
1413 | with IPv6 Neighbor Discovery) behavior: | |
1414 | 1 - (default) discard fragmented neighbor discovery packets | |
1415 | 0 - allow fragmented neighbor discovery packets | |
1416 | ||
1da177e4 LT |
1417 | icmp/*: |
1418 | ratelimit - INTEGER | |
1419 | Limit the maximal rates for sending ICMPv6 packets. | |
6dbf4bca SH |
1420 | 0 to disable any limiting, |
1421 | otherwise the minimal space between responses in milliseconds. | |
1422 | Default: 1000 | |
1da177e4 LT |
1423 | |
1424 | ||
1425 | IPv6 Update by: | |
1426 | Pekka Savola <pekkas@netcore.fi> | |
1427 | YOSHIFUJI Hideaki / USAGI Project <yoshfuji@linux-ipv6.org> | |
1428 | ||
1429 | ||
1430 | /proc/sys/net/bridge/* Variables: | |
1431 | ||
1432 | bridge-nf-call-arptables - BOOLEAN | |
1433 | 1 : pass bridged ARP traffic to arptables' FORWARD chain. | |
1434 | 0 : disable this. | |
1435 | Default: 1 | |
1436 | ||
1437 | bridge-nf-call-iptables - BOOLEAN | |
1438 | 1 : pass bridged IPv4 traffic to iptables' chains. | |
1439 | 0 : disable this. | |
1440 | Default: 1 | |
1441 | ||
1442 | bridge-nf-call-ip6tables - BOOLEAN | |
1443 | 1 : pass bridged IPv6 traffic to ip6tables' chains. | |
1444 | 0 : disable this. | |
1445 | Default: 1 | |
1446 | ||
1447 | bridge-nf-filter-vlan-tagged - BOOLEAN | |
516299d2 MM |
1448 | 1 : pass bridged vlan-tagged ARP/IP/IPv6 traffic to {arp,ip,ip6}tables. |
1449 | 0 : disable this. | |
4981682c | 1450 | Default: 0 |
516299d2 MM |
1451 | |
1452 | bridge-nf-filter-pppoe-tagged - BOOLEAN | |
1453 | 1 : pass bridged pppoe-tagged IP/IPv6 traffic to {ip,ip6}tables. | |
1da177e4 | 1454 | 0 : disable this. |
4981682c | 1455 | Default: 0 |
1da177e4 | 1456 | |
4981682c PNA |
1457 | bridge-nf-pass-vlan-input-dev - BOOLEAN |
1458 | 1: if bridge-nf-filter-vlan-tagged is enabled, try to find a vlan | |
1459 | interface on the bridge and set the netfilter input device to the vlan. | |
1460 | This allows use of e.g. "iptables -i br0.1" and makes the REDIRECT | |
1461 | target work with vlan-on-top-of-bridge interfaces. When no matching | |
1462 | vlan interface is found, or this switch is off, the input device is | |
1463 | set to the bridge interface. | |
1464 | 0: disable bridge netfilter vlan interface lookup. | |
1465 | Default: 0 | |
1da177e4 | 1466 | |
32e8d494 VY |
1467 | proc/sys/net/sctp/* Variables: |
1468 | ||
1469 | addip_enable - BOOLEAN | |
1470 | Enable or disable extension of Dynamic Address Reconfiguration | |
1471 | (ADD-IP) functionality specified in RFC5061. This extension provides | |
1472 | the ability to dynamically add and remove new addresses for the SCTP | |
1473 | associations. | |
1474 | ||
1475 | 1: Enable extension. | |
1476 | ||
1477 | 0: Disable extension. | |
1478 | ||
1479 | Default: 0 | |
1480 | ||
1481 | addip_noauth_enable - BOOLEAN | |
1482 | Dynamic Address Reconfiguration (ADD-IP) requires the use of | |
1483 | authentication to protect the operations of adding or removing new | |
1484 | addresses. This requirement is mandated so that unauthorized hosts | |
1485 | would not be able to hijack associations. However, older | |
1486 | implementations may not have implemented this requirement while | |
1487 | allowing the ADD-IP extension. For reasons of interoperability, | |
1488 | we provide this variable to control the enforcement of the | |
1489 | authentication requirement. | |
1490 | ||
1491 | 1: Allow ADD-IP extension to be used without authentication. This | |
1492 | should only be set in a closed environment for interoperability | |
1493 | with older implementations. | |
1494 | ||
1495 | 0: Enforce the authentication requirement | |
1496 | ||
1497 | Default: 0 | |
1498 | ||
1499 | auth_enable - BOOLEAN | |
1500 | Enable or disable Authenticated Chunks extension. This extension | |
1501 | provides the ability to send and receive authenticated chunks and is | |
1502 | required for secure operation of Dynamic Address Reconfiguration | |
1503 | (ADD-IP) extension. | |
1504 | ||
1505 | 1: Enable this extension. | |
1506 | 0: Disable this extension. | |
1507 | ||
1508 | Default: 0 | |
1509 | ||
1510 | prsctp_enable - BOOLEAN | |
1511 | Enable or disable the Partial Reliability extension (RFC3758) which | |
1512 | is used to notify peers that a given DATA should no longer be expected. | |
1513 | ||
1514 | 1: Enable extension | |
1515 | 0: Disable | |
1516 | ||
1517 | Default: 1 | |
1518 | ||
1519 | max_burst - INTEGER | |
1520 | The limit of the number of new packets that can be initially sent. It | |
1521 | controls how bursty the generated traffic can be. | |
1522 | ||
1523 | Default: 4 | |
1524 | ||
1525 | association_max_retrans - INTEGER | |
1526 | Set the maximum number for retransmissions that an association can | |
1527 | attempt deciding that the remote end is unreachable. If this value | |
1528 | is exceeded, the association is terminated. | |
1529 | ||
1530 | Default: 10 | |
1531 | ||
1532 | max_init_retransmits - INTEGER | |
1533 | The maximum number of retransmissions of INIT and COOKIE-ECHO chunks | |
1534 | that an association will attempt before declaring the destination | |
1535 | unreachable and terminating. | |
1536 | ||
1537 | Default: 8 | |
1538 | ||
1539 | path_max_retrans - INTEGER | |
1540 | The maximum number of retransmissions that will be attempted on a given | |
1541 | path. Once this threshold is exceeded, the path is considered | |
1542 | unreachable, and new traffic will use a different path when the | |
1543 | association is multihomed. | |
1544 | ||
1545 | Default: 5 | |
1546 | ||
5aa93bcf NH |
1547 | pf_retrans - INTEGER |
1548 | The number of retransmissions that will be attempted on a given path | |
1549 | before traffic is redirected to an alternate transport (should one | |
1550 | exist). Note this is distinct from path_max_retrans, as a path that | |
1551 | passes the pf_retrans threshold can still be used. Its only | |
1552 | deprioritized when a transmission path is selected by the stack. This | |
1553 | setting is primarily used to enable fast failover mechanisms without | |
1554 | having to reduce path_max_retrans to a very low value. See: | |
1555 | http://www.ietf.org/id/draft-nishida-tsvwg-sctp-failover-05.txt | |
1556 | for details. Note also that a value of pf_retrans > path_max_retrans | |
1557 | disables this feature | |
1558 | ||
1559 | Default: 0 | |
1560 | ||
32e8d494 VY |
1561 | rto_initial - INTEGER |
1562 | The initial round trip timeout value in milliseconds that will be used | |
1563 | in calculating round trip times. This is the initial time interval | |
1564 | for retransmissions. | |
1565 | ||
1566 | Default: 3000 | |
1da177e4 | 1567 | |
32e8d494 VY |
1568 | rto_max - INTEGER |
1569 | The maximum value (in milliseconds) of the round trip timeout. This | |
1570 | is the largest time interval that can elapse between retransmissions. | |
1571 | ||
1572 | Default: 60000 | |
1573 | ||
1574 | rto_min - INTEGER | |
1575 | The minimum value (in milliseconds) of the round trip timeout. This | |
1576 | is the smallest time interval the can elapse between retransmissions. | |
1577 | ||
1578 | Default: 1000 | |
1579 | ||
1580 | hb_interval - INTEGER | |
1581 | The interval (in milliseconds) between HEARTBEAT chunks. These chunks | |
1582 | are sent at the specified interval on idle paths to probe the state of | |
1583 | a given path between 2 associations. | |
1584 | ||
1585 | Default: 30000 | |
1586 | ||
1587 | sack_timeout - INTEGER | |
1588 | The amount of time (in milliseconds) that the implementation will wait | |
1589 | to send a SACK. | |
1590 | ||
1591 | Default: 200 | |
1592 | ||
1593 | valid_cookie_life - INTEGER | |
1594 | The default lifetime of the SCTP cookie (in milliseconds). The cookie | |
1595 | is used during association establishment. | |
1596 | ||
1597 | Default: 60000 | |
1598 | ||
1599 | cookie_preserve_enable - BOOLEAN | |
1600 | Enable or disable the ability to extend the lifetime of the SCTP cookie | |
1601 | that is used during the establishment phase of SCTP association | |
1602 | ||
1603 | 1: Enable cookie lifetime extension. | |
1604 | 0: Disable | |
1605 | ||
1606 | Default: 1 | |
1607 | ||
3c68198e NH |
1608 | cookie_hmac_alg - STRING |
1609 | Select the hmac algorithm used when generating the cookie value sent by | |
1610 | a listening sctp socket to a connecting client in the INIT-ACK chunk. | |
1611 | Valid values are: | |
1612 | * md5 | |
1613 | * sha1 | |
1614 | * none | |
1615 | Ability to assign md5 or sha1 as the selected alg is predicated on the | |
3b09adcb | 1616 | configuration of those algorithms at build time (CONFIG_CRYPTO_MD5 and |
3c68198e NH |
1617 | CONFIG_CRYPTO_SHA1). |
1618 | ||
1619 | Default: Dependent on configuration. MD5 if available, else SHA1 if | |
1620 | available, else none. | |
1621 | ||
32e8d494 VY |
1622 | rcvbuf_policy - INTEGER |
1623 | Determines if the receive buffer is attributed to the socket or to | |
1624 | association. SCTP supports the capability to create multiple | |
1625 | associations on a single socket. When using this capability, it is | |
1626 | possible that a single stalled association that's buffering a lot | |
1627 | of data may block other associations from delivering their data by | |
1628 | consuming all of the receive buffer space. To work around this, | |
1629 | the rcvbuf_policy could be set to attribute the receiver buffer space | |
1630 | to each association instead of the socket. This prevents the described | |
1631 | blocking. | |
1632 | ||
1633 | 1: rcvbuf space is per association | |
3b09adcb | 1634 | 0: rcvbuf space is per socket |
32e8d494 VY |
1635 | |
1636 | Default: 0 | |
1637 | ||
1638 | sndbuf_policy - INTEGER | |
1639 | Similar to rcvbuf_policy above, this applies to send buffer space. | |
1640 | ||
1641 | 1: Send buffer is tracked per association | |
1642 | 0: Send buffer is tracked per socket. | |
1643 | ||
1644 | Default: 0 | |
1645 | ||
1646 | sctp_mem - vector of 3 INTEGERs: min, pressure, max | |
1647 | Number of pages allowed for queueing by all SCTP sockets. | |
1648 | ||
1649 | min: Below this number of pages SCTP is not bothered about its | |
1650 | memory appetite. When amount of memory allocated by SCTP exceeds | |
1651 | this number, SCTP starts to moderate memory usage. | |
1652 | ||
1653 | pressure: This value was introduced to follow format of tcp_mem. | |
1654 | ||
1655 | max: Number of pages allowed for queueing by all SCTP sockets. | |
1656 | ||
1657 | Default is calculated at boot time from amount of available memory. | |
1658 | ||
1659 | sctp_rmem - vector of 3 INTEGERs: min, default, max | |
a6e1204b MM |
1660 | Only the first value ("min") is used, "default" and "max" are |
1661 | ignored. | |
1662 | ||
1663 | min: Minimal size of receive buffer used by SCTP socket. | |
1664 | It is guaranteed to each SCTP socket (but not association) even | |
1665 | under moderate memory pressure. | |
1666 | ||
1667 | Default: 1 page | |
32e8d494 VY |
1668 | |
1669 | sctp_wmem - vector of 3 INTEGERs: min, default, max | |
a6e1204b | 1670 | Currently this tunable has no effect. |
32e8d494 | 1671 | |
72388433 BD |
1672 | addr_scope_policy - INTEGER |
1673 | Control IPv4 address scoping - draft-stewart-tsvwg-sctp-ipv4-00 | |
1674 | ||
1675 | 0 - Disable IPv4 address scoping | |
1676 | 1 - Enable IPv4 address scoping | |
1677 | 2 - Follow draft but allow IPv4 private addresses | |
1678 | 3 - Follow draft but allow IPv4 link local addresses | |
1679 | ||
1680 | Default: 1 | |
1681 | ||
1da177e4 | 1682 | |
4edc2f34 | 1683 | /proc/sys/net/core/* |
c60f6aa8 | 1684 | Please see: Documentation/sysctl/net.txt for descriptions of these entries. |
705efc3b | 1685 | |
4edc2f34 SH |
1686 | |
1687 | /proc/sys/net/unix/* | |
705efc3b WT |
1688 | max_dgram_qlen - INTEGER |
1689 | The maximum length of dgram socket receive queue | |
1690 | ||
1691 | Default: 10 | |
1692 | ||
1693 | ||
1694 | UNDOCUMENTED: | |
4edc2f34 SH |
1695 | |
1696 | /proc/sys/net/irda/* | |
1697 | fast_poll_increase FIXME | |
1698 | warn_noreply_time FIXME | |
1699 | discovery_slots FIXME | |
1700 | slot_timeout FIXME | |
1701 | max_baud_rate FIXME | |
1702 | discovery_timeout FIXME | |
1703 | lap_keepalive_time FIXME | |
1704 | max_noreply_time FIXME | |
1705 | max_tx_data_size FIXME | |
1706 | max_tx_window FIXME | |
1707 | min_tx_turn_time FIXME |