1.. SPDX-License-Identifier: GPL-2.0 2 3========= 4IP Sysctl 5========= 6 7/proc/sys/net/ipv4/* Variables 8============================== 9 10ip_forward - BOOLEAN 11 Forward Packets between interfaces. 12 13 This variable is special, its change resets all configuration 14 parameters to their default state (RFC1122 for hosts, RFC1812 15 for routers) 16 17 Possible values: 18 19 - 0 (disabled) 20 - 1 (enabled) 21 22 Default: 0 (disabled) 23 24ip_default_ttl - INTEGER 25 Default value of TTL field (Time To Live) for outgoing (but not 26 forwarded) IP packets. Should be between 1 and 255 inclusive. 27 Default: 64 (as recommended by RFC1700) 28 29ip_no_pmtu_disc - INTEGER 30 Disable Path MTU Discovery. If enabled in mode 1 and a 31 fragmentation-required ICMP is received, the PMTU to this 32 destination will be set to the smallest of the old MTU to 33 this destination and min_pmtu (see below). You will need 34 to raise min_pmtu to the smallest interface MTU on your system 35 manually if you want to avoid locally generated fragments. 36 37 In mode 2 incoming Path MTU Discovery messages will be 38 discarded. Outgoing frames are handled the same as in mode 1, 39 implicitly setting IP_PMTUDISC_DONT on every created socket. 40 41 Mode 3 is a hardened pmtu discover mode. The kernel will only 42 accept fragmentation-needed errors if the underlying protocol 43 can verify them besides a plain socket lookup. Current 44 protocols for which pmtu events will be honored are TCP and 45 SCTP as they verify e.g. the sequence number or the 46 association. This mode should not be enabled globally but is 47 only intended to secure e.g. name servers in namespaces where 48 TCP path mtu must still work but path MTU information of other 49 protocols should be discarded. If enabled globally this mode 50 could break other protocols. 51 52 Possible values: 0-3 53 54 Default: FALSE 55 56min_pmtu - INTEGER 57 default 552 - minimum Path MTU. Unless this is changed manually, 58 each cached pmtu will never be lower than this setting. 59 60ip_forward_use_pmtu - BOOLEAN 61 By default we don't trust protocol path MTUs while forwarding 62 because they could be easily forged and can lead to unwanted 63 fragmentation by the router. 64 You only need to enable this if you have user-space software 65 which tries to discover path mtus by itself and depends on the 66 kernel honoring this information. This is normally not the 67 case. 68 69 Possible values: 70 71 - 0 (disabled) 72 - 1 (enabled) 73 74 Default: 0 (disabled) 75 76fwmark_reflect - BOOLEAN 77 Controls the fwmark of kernel-generated IPv4 reply packets that are not 78 associated with a socket for example, TCP RSTs or ICMP echo replies). 79 If disabled, these packets have a fwmark of zero. If enabled, they have the 80 fwmark of the packet they are replying to. 81 82 Possible values: 83 84 - 0 (disabled) 85 - 1 (enabled) 86 87 Default: 0 (disabled) 88 89fib_multipath_use_neigh - BOOLEAN 90 Use status of existing neighbor entry when determining nexthop for 91 multipath routes. If disabled, neighbor information is not used and 92 packets could be directed to a failed nexthop. Only valid for kernels 93 built with CONFIG_IP_ROUTE_MULTIPATH enabled. 94 95 Possible values: 96 97 - 0 (disabled) 98 - 1 (enabled) 99 100 Default: 0 (disabled) 101 102fib_multipath_hash_policy - INTEGER 103 Controls which hash policy to use for multipath routes. Only valid 104 for kernels built with CONFIG_IP_ROUTE_MULTIPATH enabled. 105 106 Default: 0 (Layer 3) 107 108 Possible values: 109 110 - 0 - Layer 3 111 - 1 - Layer 4 112 - 2 - Layer 3 or inner Layer 3 if present 113 - 3 - Custom multipath hash. Fields used for multipath hash calculation 114 are determined by fib_multipath_hash_fields sysctl 115 116fib_multipath_hash_fields - UNSIGNED INTEGER 117 When fib_multipath_hash_policy is set to 3 (custom multipath hash), the 118 fields used for multipath hash calculation are determined by this 119 sysctl. 120 121 This value is a bitmask which enables various fields for multipath hash 122 calculation. 123 124 Possible fields are: 125 126 ====== ============================ 127 0x0001 Source IP address 128 0x0002 Destination IP address 129 0x0004 IP protocol 130 0x0008 Unused (Flow Label) 131 0x0010 Source port 132 0x0020 Destination port 133 0x0040 Inner source IP address 134 0x0080 Inner destination IP address 135 0x0100 Inner IP protocol 136 0x0200 Inner Flow Label 137 0x0400 Inner source port 138 0x0800 Inner destination port 139 ====== ============================ 140 141 Default: 0x0007 (source IP, destination IP and IP protocol) 142 143fib_multipath_hash_seed - UNSIGNED INTEGER 144 The seed value used when calculating hash for multipath routes. Applies 145 to both IPv4 and IPv6 datapath. Only present for kernels built with 146 CONFIG_IP_ROUTE_MULTIPATH enabled. 147 148 When set to 0, the seed value used for multipath routing defaults to an 149 internal random-generated one. 150 151 The actual hashing algorithm is not specified -- there is no guarantee 152 that a next hop distribution effected by a given seed will keep stable 153 across kernel versions. 154 155 Default: 0 (random) 156 157fib_sync_mem - UNSIGNED INTEGER 158 Amount of dirty memory from fib entries that can be backlogged before 159 synchronize_rcu is forced. 160 161 Default: 512kB Minimum: 64kB Maximum: 64MB 162 163ip_forward_update_priority - INTEGER 164 Whether to update SKB priority from "TOS" field in IPv4 header after it 165 is forwarded. The new SKB priority is mapped from TOS field value 166 according to an rt_tos2priority table (see e.g. man tc-prio). 167 168 Default: 1 (Update priority.) 169 170 Possible values: 171 172 - 0 - Do not update priority. 173 - 1 - Update priority. 174 175route/max_size - INTEGER 176 Maximum number of routes allowed in the kernel. Increase 177 this when using large numbers of interfaces and/or routes. 178 179 From linux kernel 3.6 onwards, this is deprecated for ipv4 180 as route cache is no longer used. 181 182 From linux kernel 6.3 onwards, this is deprecated for ipv6 183 as garbage collection manages cached route entries. 184 185neigh/default/gc_thresh1 - INTEGER 186 Minimum number of entries to keep. Garbage collector will not 187 purge entries if there are fewer than this number. 188 189 Default: 128 190 191neigh/default/gc_thresh2 - INTEGER 192 Threshold when garbage collector becomes more aggressive about 193 purging entries. Entries older than 5 seconds will be cleared 194 when over this number. 195 196 Default: 512 197 198neigh/default/gc_thresh3 - INTEGER 199 Maximum number of non-PERMANENT neighbor entries allowed. Increase 200 this when using large numbers of interfaces and when communicating 201 with large numbers of directly-connected peers. 202 203 Default: 1024 204 205neigh/default/gc_interval - INTEGER 206 Specifies how often the garbage collector for neighbor entries 207 should run. This value applies to the entire table, not 208 individual entries. Unused since kernel v2.6.8. 209 210 Default: 30 seconds 211 212neigh/default/gc_stale_time - INTEGER 213 Determines how long a neighbor entry can remain unused before it is 214 considered stale and eligible for garbage collection. Entries that have 215 not been used for longer than this time will be removed by the garbage 216 collector, unless they have active references, are marked as PERMANENT, 217 or carry the NTF_EXT_LEARNED or NTF_EXT_VALIDATED flag. Stale entries 218 are only removed by the periodic GC when there are at least gc_thresh1 219 neighbors in the table. 220 221 Default: 60 seconds 222 223neigh/default/unres_qlen_bytes - INTEGER 224 The maximum number of bytes which may be used by packets 225 queued for each unresolved address by other network layers. 226 (added in linux 3.3) 227 228 Setting negative value is meaningless and will return error. 229 230 Default: SK_WMEM_DEFAULT, (same as net.core.wmem_default). 231 232 Exact value depends on architecture and kernel options, 233 but should be enough to allow queuing 256 packets 234 of medium size. 235 236neigh/default/unres_qlen - INTEGER 237 The maximum number of packets which may be queued for each 238 unresolved address by other network layers. 239 240 (deprecated in linux 3.3) : use unres_qlen_bytes instead. 241 242 Prior to linux 3.3, the default value is 3 which may cause 243 unexpected packet loss. The current default value is calculated 244 according to default value of unres_qlen_bytes and true size of 245 packet. 246 247 Default: 101 248 249neigh/default/interval_probe_time_ms - INTEGER 250 The probe interval for neighbor entries with NTF_MANAGED flag, 251 the min value is 1000 (1s), and the max value is 86400000 (1 day). 252 253 Default: 5000 254 255mtu_expires - INTEGER 256 Time, in seconds, that cached PMTU information is kept. 257 258min_adv_mss - INTEGER 259 The advertised MSS depends on the first hop route MTU, but will 260 never be lower than this setting. 261 262fib_notify_on_flag_change - INTEGER 263 Whether to emit RTM_NEWROUTE notifications whenever RTM_F_OFFLOAD/ 264 RTM_F_TRAP/RTM_F_OFFLOAD_FAILED flags are changed. 265 266 After installing a route to the kernel, user space receives an 267 acknowledgment, which means the route was installed in the kernel, 268 but not necessarily in hardware. 269 It is also possible for a route already installed in hardware to change 270 its action and therefore its flags. For example, a host route that is 271 trapping packets can be "promoted" to perform decapsulation following 272 the installation of an IPinIP/VXLAN tunnel. 273 The notifications will indicate to user-space the state of the route. 274 275 Default: 0 (Do not emit notifications.) 276 277 Possible values: 278 279 - 0 - Do not emit notifications. 280 - 1 - Emit notifications. 281 - 2 - Emit notifications only for RTM_F_OFFLOAD_FAILED flag change. 282 283IP Fragmentation: 284 285ipfrag_high_thresh - LONG INTEGER 286 Maximum memory used to reassemble IP fragments. 287 288ipfrag_low_thresh - LONG INTEGER 289 (Obsolete since linux-4.17) 290 Maximum memory used to reassemble IP fragments before the kernel 291 begins to remove incomplete fragment queues to free up resources. 292 The kernel still accepts new fragments for defragmentation. 293 294ipfrag_time - INTEGER 295 Time in seconds to keep an IP fragment in memory. 296 297ipfrag_max_dist - INTEGER 298 ipfrag_max_dist is a non-negative integer value which defines the 299 maximum "disorder" which is allowed among fragments which share a 300 common IP source address. Note that reordering of packets is 301 not unusual, but if a large number of fragments arrive from a source 302 IP address while a particular fragment queue remains incomplete, it 303 probably indicates that one or more fragments belonging to that queue 304 have been lost. When ipfrag_max_dist is positive, an additional check 305 is done on fragments before they are added to a reassembly queue - if 306 ipfrag_max_dist (or more) fragments have arrived from a particular IP 307 address between additions to any IP fragment queue using that source 308 address, it's presumed that one or more fragments in the queue are 309 lost. The existing fragment queue will be dropped, and a new one 310 started. An ipfrag_max_dist value of zero disables this check. 311 312 Using a very small value, e.g. 1 or 2, for ipfrag_max_dist can 313 result in unnecessarily dropping fragment queues when normal 314 reordering of packets occurs, which could lead to poor application 315 performance. Using a very large value, e.g. 50000, increases the 316 likelihood of incorrectly reassembling IP fragments that originate 317 from different IP datagrams, which could result in data corruption. 318 Default: 64 319 320bc_forwarding - INTEGER 321 bc_forwarding enables the feature described in rfc1812#section-5.3.5.2 322 and rfc2644. It allows the router to forward directed broadcast. 323 To enable this feature, the 'all' entry and the input interface entry 324 should be set to 1. 325 Default: 0 326 327INET peer storage 328================= 329 330inet_peer_threshold - INTEGER 331 The approximate size of the storage. Starting from this threshold 332 entries will be thrown aggressively. This threshold also determines 333 entries' time-to-live and time intervals between garbage collection 334 passes. More entries, less time-to-live, less GC interval. 335 336inet_peer_minttl - INTEGER 337 Minimum time-to-live of entries. Should be enough to cover fragment 338 time-to-live on the reassembling side. This minimum time-to-live is 339 guaranteed if the pool size is less than inet_peer_threshold. 340 Measured in seconds. 341 342inet_peer_maxttl - INTEGER 343 Maximum time-to-live of entries. Unused entries will expire after 344 this period of time if there is no memory pressure on the pool (i.e. 345 when the number of entries in the pool is very small). 346 Measured in seconds. 347 348TCP variables 349============= 350 351somaxconn - INTEGER 352 Limit of socket listen() backlog, known in userspace as SOMAXCONN. 353 Defaults to 4096. (Was 128 before linux-5.4) 354 See also tcp_max_syn_backlog for additional tuning for TCP sockets. 355 356tcp_abort_on_overflow - BOOLEAN 357 If listening service is too slow to accept new connections, 358 reset them. Default state is FALSE. It means that if overflow 359 occurred due to a burst, connection will recover. Enable this 360 option _only_ if you are really sure that listening daemon 361 cannot be tuned to accept connections faster. Enabling this 362 option can harm clients of your server. 363 364tcp_adv_win_scale - INTEGER 365 Obsolete since linux-6.6 366 Count buffering overhead as bytes/2^tcp_adv_win_scale 367 (if tcp_adv_win_scale > 0) or bytes-bytes/2^(-tcp_adv_win_scale), 368 if it is <= 0. 369 370 Possible values are [-31, 31], inclusive. 371 372 Default: 1 373 374tcp_allowed_congestion_control - STRING 375 Show/set the congestion control choices available to non-privileged 376 processes. The list is a subset of those listed in 377 tcp_available_congestion_control. 378 379 Default is "reno" and the default setting (tcp_congestion_control). 380 381tcp_app_win - INTEGER 382 Reserve max(window/2^tcp_app_win, mss) of window for application 383 buffer. Value 0 is special, it means that nothing is reserved. 384 385 Possible values are [0, 31], inclusive. 386 387 Default: 31 388 389tcp_autocorking - BOOLEAN 390 Enable TCP auto corking : 391 When applications do consecutive small write()/sendmsg() system calls, 392 we try to coalesce these small writes as much as possible, to lower 393 total amount of sent packets. This is done if at least one prior 394 packet for the flow is waiting in Qdisc queues or device transmit 395 queue. Applications can still use TCP_CORK for optimal behavior 396 when they know how/when to uncork their sockets. 397 398 Possible values: 399 400 - 0 (disabled) 401 - 1 (enabled) 402 403 Default: 1 (enabled) 404 405tcp_available_congestion_control - STRING 406 Shows the available congestion control choices that are registered. 407 More congestion control algorithms may be available as modules, 408 but not loaded. 409 410tcp_base_mss - INTEGER 411 The initial value of search_low to be used by the packetization layer 412 Path MTU discovery (MTU probing). If MTU probing is enabled, 413 this is the initial MSS used by the connection. 414 415tcp_mtu_probe_floor - INTEGER 416 If MTU probing is enabled this caps the minimum MSS used for search_low 417 for the connection. 418 419 Default : 48 420 421tcp_min_snd_mss - INTEGER 422 TCP SYN and SYNACK messages usually advertise an ADVMSS option, 423 as described in RFC 1122 and RFC 6691. 424 425 If this ADVMSS option is smaller than tcp_min_snd_mss, 426 it is silently capped to tcp_min_snd_mss. 427 428 Default : 48 (at least 8 bytes of payload per segment) 429 430tcp_congestion_control - STRING 431 Set the congestion control algorithm to be used for new 432 connections. The algorithm "reno" is always available, but 433 additional choices may be available based on kernel configuration. 434 Default is set as part of kernel configuration. 435 For passive connections, the listener congestion control choice 436 is inherited. 437 438 [see setsockopt(listenfd, SOL_TCP, TCP_CONGESTION, "name" ...) ] 439 440tcp_dsack - BOOLEAN 441 Allows TCP to send "duplicate" SACKs. 442 443 Possible values: 444 445 - 0 (disabled) 446 - 1 (enabled) 447 448 Default: 1 (enabled) 449 450tcp_early_retrans - INTEGER 451 Tail loss probe (TLP) converts RTOs occurring due to tail 452 losses into fast recovery (RFC8985). Note that 453 TLP requires RACK to function properly (see tcp_recovery below) 454 455 Possible values: 456 457 - 0 disables TLP 458 - 3 or 4 enables TLP 459 460 Default: 3 461 462tcp_ecn - INTEGER 463 Control use of Explicit Congestion Notification (ECN) by TCP. 464 ECN is used only when both ends of the TCP connection indicate support 465 for it. This feature is useful in avoiding losses due to congestion by 466 allowing supporting routers to signal congestion before having to drop 467 packets. A host that supports ECN both sends ECN at the IP layer and 468 feeds back ECN at the TCP layer. The highest variant of ECN feedback 469 that both peers support is chosen by the ECN negotiation (Accurate ECN, 470 ECN, or no ECN). 471 472 The highest negotiated variant for incoming connection requests 473 and the highest variant requested by outgoing connection 474 attempts: 475 476 ===== ==================== ==================== 477 Value Incoming connections Outgoing connections 478 ===== ==================== ==================== 479 0 No ECN No ECN 480 1 ECN ECN 481 2 ECN No ECN 482 3 AccECN AccECN 483 4 AccECN ECN 484 5 AccECN No ECN 485 ===== ==================== ==================== 486 487 Default: 2 488 489tcp_ecn_option - INTEGER 490 Control Accurate ECN (AccECN) option sending when AccECN has been 491 successfully negotiated during handshake. Send logic inhibits 492 sending AccECN options regardless of this setting when no AccECN 493 option has been seen for the reverse direction. 494 495 Possible values are: 496 497 = ============================================================ 498 0 Never send AccECN option. This also disables sending AccECN 499 option in SYN/ACK during handshake. 500 1 Send AccECN option sparingly according to the minimum option 501 rules outlined in draft-ietf-tcpm-accurate-ecn. 502 2 Send AccECN option on every packet whenever it fits into TCP 503 option space except when AccECN fallback is triggered. 504 3 Send AccECN option on every packet whenever it fits into TCP 505 option space even when AccECN fallback is triggered. 506 = ============================================================ 507 508 Default: 2 509 510tcp_ecn_option_beacon - INTEGER 511 Control Accurate ECN (AccECN) option sending frequency per RTT and it 512 takes effect only when tcp_ecn_option is set to 2. 513 514 Default: 3 (AccECN will be send at least 3 times per RTT) 515 516tcp_ecn_fallback - BOOLEAN 517 If the kernel detects that ECN connection misbehaves, enable fall 518 back to non-ECN. Currently, this knob implements the fallback 519 from RFC3168, section 6.1.1.1., but we reserve that in future, 520 additional detection mechanisms could be implemented under this 521 knob. The value is not used, if tcp_ecn or per route (or congestion 522 control) ECN settings are disabled. 523 524 Possible values: 525 526 - 0 (disabled) 527 - 1 (enabled) 528 529 Default: 1 (enabled) 530 531tcp_fack - BOOLEAN 532 This is a legacy option, it has no effect anymore. 533 534tcp_fin_timeout - INTEGER 535 The length of time an orphaned (no longer referenced by any 536 application) connection will remain in the FIN_WAIT_2 state 537 before it is aborted at the local end. While a perfectly 538 valid "receive only" state for an un-orphaned connection, an 539 orphaned connection in FIN_WAIT_2 state could otherwise wait 540 forever for the remote to close its end of the connection. 541 542 Cf. tcp_max_orphans 543 544 Default: 60 seconds 545 546tcp_frto - INTEGER 547 Enables Forward RTO-Recovery (F-RTO) defined in RFC5682. 548 F-RTO is an enhanced recovery algorithm for TCP retransmission 549 timeouts. It is particularly beneficial in networks where the 550 RTT fluctuates (e.g., wireless). F-RTO is sender-side only 551 modification. It does not require any support from the peer. 552 553 By default it's enabled with a non-zero value. 0 disables F-RTO. 554 555tcp_fwmark_accept - BOOLEAN 556 If enabled, incoming connections to listening sockets that do not have a 557 socket mark will set the mark of the accepting socket to the fwmark of 558 the incoming SYN packet. This will cause all packets on that connection 559 (starting from the first SYNACK) to be sent with that fwmark. The 560 listening socket's mark is unchanged. Listening sockets that already 561 have a fwmark set via setsockopt(SOL_SOCKET, SO_MARK, ...) are 562 unaffected. 563 564 Possible values: 565 566 - 0 (disabled) 567 - 1 (enabled) 568 569 Default: 0 (disabled) 570 571tcp_invalid_ratelimit - INTEGER 572 Limit the maximal rate for sending duplicate acknowledgments 573 in response to incoming TCP packets that are for an existing 574 connection but that are invalid due to any of these reasons: 575 576 (a) out-of-window sequence number, 577 (b) out-of-window acknowledgment number, or 578 (c) PAWS (Protection Against Wrapped Sequence numbers) check failure 579 580 This can help mitigate simple "ack loop" DoS attacks, wherein 581 a buggy or malicious middlebox or man-in-the-middle can 582 rewrite TCP header fields in manner that causes each endpoint 583 to think that the other is sending invalid TCP segments, thus 584 causing each side to send an unterminating stream of duplicate 585 acknowledgments for invalid segments. 586 587 Using 0 disables rate-limiting of dupacks in response to 588 invalid segments; otherwise this value specifies the minimal 589 space between sending such dupacks, in milliseconds. 590 591 Default: 500 (milliseconds). 592 593tcp_keepalive_time - INTEGER 594 How often TCP sends out keepalive messages when keepalive is enabled. 595 Default: 2hours. 596 597tcp_keepalive_probes - INTEGER 598 How many keepalive probes TCP sends out, until it decides that the 599 connection is broken. Default value: 9. 600 601tcp_keepalive_intvl - INTEGER 602 How frequently the probes are send out. Multiplied by 603 tcp_keepalive_probes it is time to kill not responding connection, 604 after probes started. Default value: 75sec i.e. connection 605 will be aborted after ~11 minutes of retries. 606 607tcp_l3mdev_accept - BOOLEAN 608 Enables child sockets to inherit the L3 master device index. 609 Enabling this option allows a "global" listen socket to work 610 across L3 master domains (e.g., VRFs) with connected sockets 611 derived from the listen socket to be bound to the L3 domain in 612 which the packets originated. Only valid when the kernel was 613 compiled with CONFIG_NET_L3_MASTER_DEV. 614 615 Possible values: 616 617 - 0 (disabled) 618 - 1 (enabled) 619 620 Default: 0 (disabled) 621 622tcp_low_latency - BOOLEAN 623 This is a legacy option, it has no effect anymore. 624 625tcp_max_orphans - INTEGER 626 Maximal number of TCP sockets not attached to any user file handle, 627 held by system. If this number is exceeded orphaned connections are 628 reset immediately and warning is printed. This limit exists 629 only to prevent simple DoS attacks, you _must_ not rely on this 630 or lower the limit artificially, but rather increase it 631 (probably, after increasing installed memory), 632 if network conditions require more than default value, 633 and tune network services to linger and kill such states 634 more aggressively. Let me to remind again: each orphan eats 635 up to ~64K of unswappable memory. 636 637tcp_max_syn_backlog - INTEGER 638 Maximal number of remembered connection requests (SYN_RECV), 639 which have not received an acknowledgment from connecting client. 640 641 This is a per-listener limit. 642 643 The minimal value is 128 for low memory machines, and it will 644 increase in proportion to the memory of machine. 645 646 If server suffers from overload, try increasing this number. 647 648 Remember to also check /proc/sys/net/core/somaxconn 649 A SYN_RECV request socket consumes about 304 bytes of memory. 650 651tcp_max_tw_buckets - INTEGER 652 Maximal number of timewait sockets held by system simultaneously. 653 If this number is exceeded time-wait socket is immediately destroyed 654 and warning is printed. This limit exists only to prevent 655 simple DoS attacks, you _must_ not lower the limit artificially, 656 but rather increase it (probably, after increasing installed memory), 657 if network conditions require more than default value. 658 659tcp_mem - vector of 3 INTEGERs: min, pressure, max 660 min: below this number of pages TCP is not bothered about its 661 memory appetite. 662 663 pressure: when amount of memory allocated by TCP exceeds this number 664 of pages, TCP moderates its memory consumption and enters memory 665 pressure mode, which is exited when memory consumption falls 666 under "min". 667 668 max: number of pages allowed for queueing by all TCP sockets. 669 670 Defaults are calculated at boot time from amount of available 671 memory. 672 673tcp_min_rtt_wlen - INTEGER 674 The window length of the windowed min filter to track the minimum RTT. 675 A shorter window lets a flow more quickly pick up new (higher) 676 minimum RTT when it is moved to a longer path (e.g., due to traffic 677 engineering). A longer window makes the filter more resistant to RTT 678 inflations such as transient congestion. The unit is seconds. 679 680 Possible values: 0 - 86400 (1 day) 681 682 Default: 300 683 684tcp_moderate_rcvbuf - BOOLEAN 685 If enabled, TCP performs receive buffer auto-tuning, attempting to 686 automatically size the buffer (no greater than tcp_rmem[2]) to 687 match the size required by the path for full throughput. 688 689 Possible values: 690 691 - 0 (disabled) 692 - 1 (enabled) 693 694 Default: 1 (enabled) 695 696tcp_rcvbuf_low_rtt - INTEGER 697 rcvbuf autotuning can over estimate final socket rcvbuf, which 698 can lead to cache trashing for high throughput flows. 699 700 For small RTT flows (below tcp_rcvbuf_low_rtt usecs), we can relax 701 rcvbuf growth: Few additional ms to reach the final (and smaller) 702 rcvbuf is a good tradeoff. 703 704 Default : 1000 (1 ms) 705 706tcp_mtu_probing - INTEGER 707 Controls TCP Packetization-Layer Path MTU Discovery. Takes three 708 values: 709 710 - 0 - Disabled 711 - 1 - Disabled by default, enabled when an ICMP black hole detected 712 - 2 - Always enabled, use initial MSS of tcp_base_mss. 713 714tcp_probe_interval - UNSIGNED INTEGER 715 Controls how often to start TCP Packetization-Layer Path MTU 716 Discovery reprobe. The default is reprobing every 10 minutes as 717 per RFC4821. 718 719tcp_probe_threshold - INTEGER 720 Controls when TCP Packetization-Layer Path MTU Discovery probing 721 will stop in respect to the width of search range in bytes. Default 722 is 8 bytes. 723 724tcp_no_metrics_save - BOOLEAN 725 By default, TCP saves various connection metrics in the route cache 726 when the connection closes, so that connections established in the 727 near future can use these to set initial conditions. Usually, this 728 increases overall performance, but may sometimes cause performance 729 degradation. If enabled, TCP will not cache metrics on closing 730 connections. 731 732 Possible values: 733 734 - 0 (disabled) 735 - 1 (enabled) 736 737 Default: 0 (disabled) 738 739tcp_no_ssthresh_metrics_save - BOOLEAN 740 Controls whether TCP saves ssthresh metrics in the route cache. 741 If enabled, ssthresh metrics are disabled. 742 743 Possible values: 744 745 - 0 (disabled) 746 - 1 (enabled) 747 748 Default: 1 (enabled) 749 750tcp_orphan_retries - INTEGER 751 This value influences the timeout of a locally closed TCP connection, 752 when RTO retransmissions remain unacknowledged. 753 See tcp_retries2 for more details. 754 755 The default value is 8. 756 757 If your machine is a loaded WEB server, 758 you should think about lowering this value, such sockets 759 may consume significant resources. Cf. tcp_max_orphans. 760 761tcp_recovery - INTEGER 762 This value is a bitmap to enable various experimental loss recovery 763 features. 764 765 ========= ============================================================= 766 RACK: 0x1 enables RACK loss detection, for fast detection of lost 767 retransmissions and tail drops, and resilience to 768 reordering. currently, setting this bit to 0 has no 769 effect, since RACK is the only supported loss detection 770 algorithm. 771 772 RACK: 0x2 makes RACK's reordering window static (min_rtt/4). 773 774 RACK: 0x4 disables RACK's DUPACK threshold heuristic 775 ========= ============================================================= 776 777 Default: 0x1 778 779tcp_reflect_tos - BOOLEAN 780 For listening sockets, reuse the DSCP value of the initial SYN message 781 for outgoing packets. This allows to have both directions of a TCP 782 stream to use the same DSCP value, assuming DSCP remains unchanged for 783 the lifetime of the connection. 784 785 This options affects both IPv4 and IPv6. 786 787 Possible values: 788 789 - 0 (disabled) 790 - 1 (enabled) 791 792 Default: 0 (disabled) 793 794tcp_reordering - INTEGER 795 Initial reordering level of packets in a TCP stream. 796 TCP stack can then dynamically adjust flow reordering level 797 between this initial value and tcp_max_reordering 798 799 Default: 3 800 801tcp_max_reordering - INTEGER 802 Maximal reordering level of packets in a TCP stream. 803 300 is a fairly conservative value, but you might increase it 804 if paths are using per packet load balancing (like bonding rr mode) 805 806 Default: 300 807 808tcp_retrans_collapse - BOOLEAN 809 Bug-to-bug compatibility with some broken printers. 810 On retransmit try to send bigger packets to work around bugs in 811 certain TCP stacks. 812 813 Possible values: 814 815 - 0 (disabled) 816 - 1 (enabled) 817 818 Default: 1 (enabled) 819 820tcp_retries1 - INTEGER 821 This value influences the time, after which TCP decides, that 822 something is wrong due to unacknowledged RTO retransmissions, 823 and reports this suspicion to the network layer. 824 See tcp_retries2 for more details. 825 826 RFC 1122 recommends at least 3 retransmissions, which is the 827 default. 828 829tcp_retries2 - INTEGER 830 This value influences the timeout of an alive TCP connection, 831 when RTO retransmissions remain unacknowledged. 832 Given a value of N, a hypothetical TCP connection following 833 exponential backoff with an initial RTO of TCP_RTO_MIN would 834 retransmit N times before killing the connection at the (N+1)th RTO. 835 836 The default value of 15 yields a hypothetical timeout of 924.6 837 seconds and is a lower bound for the effective timeout. 838 TCP will effectively time out at the first RTO which exceeds the 839 hypothetical timeout. 840 If tcp_rto_max_ms is decreased, it is recommended to also 841 change tcp_retries2. 842 843 RFC 1122 recommends at least 100 seconds for the timeout, 844 which corresponds to a value of at least 8. 845 846tcp_rfc1337 - BOOLEAN 847 If enabled, the TCP stack behaves conforming to RFC1337. If unset, 848 we are not conforming to RFC, but prevent TCP TIME_WAIT 849 assassination. 850 851 Possible values: 852 853 - 0 (disabled) 854 - 1 (enabled) 855 856 Default: 0 (disabled) 857 858tcp_rmem - vector of 3 INTEGERs: min, default, max 859 min: Minimal size of receive buffer used by TCP sockets. 860 It is guaranteed to each TCP socket, even under moderate memory 861 pressure. 862 863 Default: 4K 864 865 default: initial size of receive buffer used by TCP sockets. 866 This value overrides net.core.rmem_default used by other protocols. 867 Default: 131072 bytes. 868 This value results in initial window of 65535. 869 870 max: maximal size of receive buffer allowed for automatically 871 selected receiver buffers for TCP socket. 872 Calling setsockopt() with SO_RCVBUF disables 873 automatic tuning of that socket's receive buffer size, in which 874 case this value is ignored. 875 Default: between 131072 and 32MB, depending on RAM size. 876 877 Each of the three values cannot be set below 4096. 878 879tcp_sack - BOOLEAN 880 Enable select acknowledgments (SACKS). 881 882 Possible values: 883 884 - 0 (disabled) 885 - 1 (enabled) 886 887 Default: 1 (enabled) 888 889tcp_comp_sack_rtt_percent - INTEGER 890 Percentage of SRTT used for the compressed SACK feature. 891 See tcp_comp_sack_nr, tcp_comp_sack_delay_ns, tcp_comp_sack_slack_ns. 892 893 Possible values : 1 - 1000 894 895 Default : 33 % 896 897tcp_comp_sack_delay_ns - LONG INTEGER 898 TCP tries to reduce number of SACK sent, using a timer based 899 on tcp_comp_sack_rtt_percent of SRTT, capped by this sysctl 900 in nano seconds. 901 The default is 1ms, based on TSO autosizing period. 902 903 Default : 1,000,000 ns (1 ms) 904 905tcp_comp_sack_slack_ns - LONG INTEGER 906 This sysctl control the slack used when arming the 907 timer used by SACK compression. This gives extra time 908 for small RTT flows, and reduces system overhead by allowing 909 opportunistic reduction of timer interrupts. 910 Too big values might reduce goodput. 911 912 Default : 10,000 ns (10 us) 913 914tcp_comp_sack_nr - INTEGER 915 Max number of SACK that can be compressed. 916 Using 0 disables SACK compression. 917 918 Default : 44 919 920tcp_backlog_ack_defer - BOOLEAN 921 If enabled, user thread processing socket backlog tries sending 922 one ACK for the whole queue. This helps to avoid potential 923 long latencies at end of a TCP socket syscall. 924 925 Possible values: 926 927 - 0 (disabled) 928 - 1 (enabled) 929 930 Default: 1 (enabled) 931 932tcp_slow_start_after_idle - BOOLEAN 933 If enabled, provide RFC2861 behavior and time out the congestion 934 window after an idle period. An idle period is defined at 935 the current RTO. If unset, the congestion window will not 936 be timed out after an idle period. 937 938 Possible values: 939 940 - 0 (disabled) 941 - 1 (enabled) 942 943 Default: 1 (enabled) 944 945tcp_stdurg - BOOLEAN 946 Use the Host requirements interpretation of the TCP urgent pointer field. 947 Most hosts use the older BSD interpretation, so if enabled, 948 Linux might not communicate correctly with them. 949 950 Possible values: 951 952 - 0 (disabled) 953 - 1 (enabled) 954 955 Default: 0 (disabled) 956 957tcp_synack_retries - INTEGER 958 Number of times SYNACKs for a passive TCP connection attempt will 959 be retransmitted. Should not be higher than 255. Default value 960 is 5, which corresponds to 31seconds till the last retransmission 961 with the current initial RTO of 1second. With this the final timeout 962 for a passive TCP connection will happen after 63seconds. 963 964tcp_syncookies - INTEGER 965 Only valid when the kernel was compiled with CONFIG_SYN_COOKIES 966 Send out syncookies when the syn backlog queue of a socket 967 overflows. This is to prevent against the common 'SYN flood attack' 968 Default: 1 969 970 Note, that syncookies is fallback facility. 971 It MUST NOT be used to help highly loaded servers to stand 972 against legal connection rate. If you see SYN flood warnings 973 in your logs, but investigation shows that they occur 974 because of overload with legal connections, you should tune 975 another parameters until this warning disappear. 976 See: tcp_max_syn_backlog, tcp_synack_retries, tcp_abort_on_overflow. 977 978 syncookies seriously violate TCP protocol, do not allow 979 to use TCP extensions, can result in serious degradation 980 of some services (f.e. SMTP relaying), visible not by you, 981 but your clients and relays, contacting you. While you see 982 SYN flood warnings in logs not being really flooded, your server 983 is seriously misconfigured. 984 985 If you want to test which effects syncookies have to your 986 network connections you can set this knob to 2 to enable 987 unconditionally generation of syncookies. 988 989tcp_migrate_req - BOOLEAN 990 The incoming connection is tied to a specific listening socket when 991 the initial SYN packet is received during the three-way handshake. 992 When a listener is closed, in-flight request sockets during the 993 handshake and established sockets in the accept queue are aborted. 994 995 If the listener has SO_REUSEPORT enabled, other listeners on the 996 same port should have been able to accept such connections. This 997 option makes it possible to migrate such child sockets to another 998 listener after close() or shutdown(). 999 1000 The BPF_SK_REUSEPORT_SELECT_OR_MIGRATE type of eBPF program should 1001 usually be used to define the policy to pick an alive listener. 1002 Otherwise, the kernel will randomly pick an alive listener only if 1003 this option is enabled. 1004 1005 Note that migration between listeners with different settings may 1006 crash applications. Let's say migration happens from listener A to 1007 B, and only B has TCP_SAVE_SYN enabled. B cannot read SYN data from 1008 the requests migrated from A. To avoid such a situation, cancel 1009 migration by returning SK_DROP in the type of eBPF program, or 1010 disable this option. 1011 1012 Possible values: 1013 1014 - 0 (disabled) 1015 - 1 (enabled) 1016 1017 Default: 0 (disabled) 1018 1019tcp_fastopen - INTEGER 1020 Enable TCP Fast Open (RFC7413) to send and accept data in the opening 1021 SYN packet. 1022 1023 The client support is enabled by flag 0x1 (on by default). The client 1024 then must use sendmsg() or sendto() with the MSG_FASTOPEN flag, 1025 rather than connect() to send data in SYN. 1026 1027 The server support is enabled by flag 0x2 (off by default). Then 1028 either enable for all listeners with another flag (0x400) or 1029 enable individual listeners via TCP_FASTOPEN socket option with 1030 the option value being the length of the syn-data backlog. 1031 1032 The values (bitmap) are 1033 1034 ===== ======== ====================================================== 1035 0x1 (client) enables sending data in the opening SYN on the client. 1036 0x2 (server) enables the server support, i.e., allowing data in 1037 a SYN packet to be accepted and passed to the 1038 application before 3-way handshake finishes. 1039 0x4 (client) send data in the opening SYN regardless of cookie 1040 availability and without a cookie option. 1041 0x200 (server) accept data-in-SYN w/o any cookie option present. 1042 0x400 (server) enable all listeners to support Fast Open by 1043 default without explicit TCP_FASTOPEN socket option. 1044 ===== ======== ====================================================== 1045 1046 Default: 0x1 1047 1048 Note that additional client or server features are only 1049 effective if the basic support (0x1 and 0x2) are enabled respectively. 1050 1051tcp_fastopen_blackhole_timeout_sec - INTEGER 1052 Initial time period in second to disable Fastopen on active TCP sockets 1053 when a TFO firewall blackhole issue happens. 1054 This time period will grow exponentially when more blackhole issues 1055 get detected right after Fastopen is re-enabled and will reset to 1056 initial value when the blackhole issue goes away. 1057 0 to disable the blackhole detection. 1058 1059 By default, it is set to 0 (feature is disabled). 1060 1061tcp_fastopen_key - list of comma separated 32-digit hexadecimal INTEGERs 1062 The list consists of a primary key and an optional backup key. The 1063 primary key is used for both creating and validating cookies, while the 1064 optional backup key is only used for validating cookies. The purpose of 1065 the backup key is to maximize TFO validation when keys are rotated. 1066 1067 A randomly chosen primary key may be configured by the kernel if 1068 the tcp_fastopen sysctl is set to 0x400 (see above), or if the 1069 TCP_FASTOPEN setsockopt() optname is set and a key has not been 1070 previously configured via sysctl. If keys are configured via 1071 setsockopt() by using the TCP_FASTOPEN_KEY optname, then those 1072 per-socket keys will be used instead of any keys that are specified via 1073 sysctl. 1074 1075 A key is specified as 4 8-digit hexadecimal integers which are separated 1076 by a '-' as: xxxxxxxx-xxxxxxxx-xxxxxxxx-xxxxxxxx. Leading zeros may be 1077 omitted. A primary and a backup key may be specified by separating them 1078 by a comma. If only one key is specified, it becomes the primary key and 1079 any previously configured backup keys are removed. 1080 1081tcp_syn_retries - INTEGER 1082 Number of times initial SYNs for an active TCP connection attempt 1083 will be retransmitted. Should not be higher than 127. Default value 1084 is 6, which corresponds to 67seconds (with tcp_syn_linear_timeouts = 4) 1085 till the last retransmission with the current initial RTO of 1second. 1086 With this the final timeout for an active TCP connection attempt 1087 will happen after 131seconds. 1088 1089tcp_timestamps - INTEGER 1090 Enable timestamps as defined in RFC1323. 1091 1092 - 0: Disabled. 1093 - 1: Enable timestamps as defined in RFC1323 and use random offset for 1094 each connection rather than only using the current time. 1095 - 2: Like 1, but without random offsets. 1096 1097 Default: 1 1098 1099tcp_min_tso_segs - INTEGER 1100 Minimal number of segments per TSO frame. 1101 1102 Since linux-3.12, TCP does an automatic sizing of TSO frames, 1103 depending on flow rate, instead of filling 64Kbytes packets. 1104 For specific usages, it's possible to force TCP to build big 1105 TSO frames. Note that TCP stack might split too big TSO packets 1106 if available window is too small. 1107 1108 Default: 2 1109 1110tcp_tso_rtt_log - INTEGER 1111 Adjustment of TSO packet sizes based on min_rtt 1112 1113 Starting from linux-5.18, TCP autosizing can be tweaked 1114 for flows having small RTT. 1115 1116 Old autosizing was splitting the pacing budget to send 1024 TSO 1117 per second. 1118 1119 tso_packet_size = sk->sk_pacing_rate / 1024; 1120 1121 With the new mechanism, we increase this TSO sizing using: 1122 1123 distance = min_rtt_usec / (2^tcp_tso_rtt_log) 1124 tso_packet_size += gso_max_size >> distance; 1125 1126 This means that flows between very close hosts can use bigger 1127 TSO packets, reducing their cpu costs. 1128 1129 If you want to use the old autosizing, set this sysctl to 0. 1130 1131 Possible values are [0, 31], inclusive. 1132 1133 Default: 9 (2^9 = 512 usec) 1134 1135tcp_pacing_ss_ratio - INTEGER 1136 sk->sk_pacing_rate is set by TCP stack using a ratio applied 1137 to current rate. (current_rate = cwnd * mss / srtt) 1138 If TCP is in slow start, tcp_pacing_ss_ratio is applied 1139 to let TCP probe for bigger speeds, assuming cwnd can be 1140 doubled every other RTT. 1141 1142 Default: 200 1143 1144tcp_pacing_ca_ratio - INTEGER 1145 sk->sk_pacing_rate is set by TCP stack using a ratio applied 1146 to current rate. (current_rate = cwnd * mss / srtt) 1147 If TCP is in congestion avoidance phase, tcp_pacing_ca_ratio 1148 is applied to conservatively probe for bigger throughput. 1149 1150 Default: 120 1151 1152tcp_syn_linear_timeouts - INTEGER 1153 The number of times for an active TCP connection to retransmit SYNs with 1154 a linear backoff timeout before defaulting to an exponential backoff 1155 timeout. This has no effect on SYNACK at the passive TCP side. 1156 1157 With an initial RTO of 1 and tcp_syn_linear_timeouts = 4 we would 1158 expect SYN RTOs to be: 1, 1, 1, 1, 1, 2, 4, ... (4 linear timeouts, 1159 and the first exponential backoff using 2^0 * initial_RTO). 1160 Default: 4 1161 1162tcp_tso_win_divisor - INTEGER 1163 This allows control over what percentage of the congestion window 1164 can be consumed by a single TSO frame. 1165 The setting of this parameter is a choice between burstiness and 1166 building larger TSO frames. 1167 1168 Default: 3 1169 1170tcp_tw_reuse - INTEGER 1171 Enable reuse of TIME-WAIT sockets for new connections when it is 1172 safe from protocol viewpoint. 1173 1174 - 0 - disable 1175 - 1 - global enable 1176 - 2 - enable for loopback traffic only 1177 1178 It should not be changed without advice/request of technical 1179 experts. 1180 1181 Default: 2 1182 1183tcp_tw_reuse_delay - UNSIGNED INTEGER 1184 The delay in milliseconds before a TIME-WAIT socket can be reused by a 1185 new connection, if TIME-WAIT socket reuse is enabled. The actual reuse 1186 threshold is within [N, N+1] range, where N is the requested delay in 1187 milliseconds, to ensure the delay interval is never shorter than the 1188 configured value. 1189 1190 This setting contains an assumption about the other TCP timestamp clock 1191 tick interval. It should not be set to a value lower than the peer's 1192 clock tick for PAWS (Protection Against Wrapped Sequence numbers) 1193 mechanism work correctly for the reused connection. 1194 1195 Default: 1000 (milliseconds) 1196 1197tcp_window_scaling - BOOLEAN 1198 Enable window scaling as defined in RFC1323. 1199 1200 Possible values: 1201 1202 - 0 (disabled) 1203 - 1 (enabled) 1204 1205 Default: 1 (enabled) 1206 1207tcp_shrink_window - BOOLEAN 1208 This changes how the TCP receive window is calculated. 1209 1210 RFC 7323, section 2.4, says there are instances when a retracted 1211 window can be offered, and that TCP implementations MUST ensure 1212 that they handle a shrinking window, as specified in RFC 1122. 1213 1214 Possible values: 1215 1216 - 0 (disabled) - The window is never shrunk. 1217 - 1 (enabled) - The window is shrunk when necessary to remain within 1218 the memory limit set by autotuning (sk_rcvbuf). 1219 This only occurs if a non-zero receive window 1220 scaling factor is also in effect. 1221 1222 Default: 0 (disabled) 1223 1224tcp_wmem - vector of 3 INTEGERs: min, default, max 1225 min: Amount of memory reserved for send buffers for TCP sockets. 1226 Each TCP socket has rights to use it due to fact of its birth. 1227 1228 Default: 4K 1229 1230 default: initial size of send buffer used by TCP sockets. This 1231 value overrides net.core.wmem_default used by other protocols. 1232 1233 It is usually lower than net.core.wmem_default. 1234 1235 Default: 16K 1236 1237 max: Maximal amount of memory allowed for automatically tuned 1238 send buffers for TCP sockets. This value does not override 1239 net.core.wmem_max. Calling setsockopt() with SO_SNDBUF disables 1240 automatic tuning of that socket's send buffer size, in which case 1241 this value is ignored. 1242 1243 Default: between 64K and 4MB, depending on RAM size. 1244 1245tcp_notsent_lowat - UNSIGNED INTEGER 1246 A TCP socket can control the amount of unsent bytes in its write queue, 1247 thanks to TCP_NOTSENT_LOWAT socket option. poll()/select()/epoll() 1248 reports POLLOUT events if the amount of unsent bytes is below a per 1249 socket value, and if the write queue is not full. sendmsg() will 1250 also not add new buffers if the limit is hit. 1251 1252 This global variable controls the amount of unsent data for 1253 sockets not using TCP_NOTSENT_LOWAT. For these sockets, a change 1254 to the global variable has immediate effect. 1255 1256 Default: UINT_MAX (0xFFFFFFFF) 1257 1258tcp_workaround_signed_windows - BOOLEAN 1259 If enabled, assume no receipt of a window scaling option means the 1260 remote TCP is broken and treats the window as a signed quantity. 1261 If disabled, assume the remote TCP is not broken even if we do 1262 not receive a window scaling option from them. 1263 1264 Possible values: 1265 1266 - 0 (disabled) 1267 - 1 (enabled) 1268 1269 Default: 0 (disabled) 1270 1271tcp_thin_linear_timeouts - BOOLEAN 1272 Enable dynamic triggering of linear timeouts for thin streams. 1273 If enabled, a check is performed upon retransmission by timeout to 1274 determine if the stream is thin (less than 4 packets in flight). 1275 As long as the stream is found to be thin, up to 6 linear 1276 timeouts may be performed before exponential backoff mode is 1277 initiated. This improves retransmission latency for 1278 non-aggressive thin streams, often found to be time-dependent. 1279 For more information on thin streams, see 1280 Documentation/networking/tcp-thin.rst 1281 1282 Possible values: 1283 1284 - 0 (disabled) 1285 - 1 (enabled) 1286 1287 Default: 0 (disabled) 1288 1289tcp_limit_output_bytes - INTEGER 1290 Controls TCP Small Queue limit per tcp socket. 1291 TCP bulk sender tends to increase packets in flight until it 1292 gets losses notifications. With SNDBUF autotuning, this can 1293 result in a large amount of packets queued on the local machine 1294 (e.g.: qdiscs, CPU backlog, or device) hurting latency of other 1295 flows, for typical pfifo_fast qdiscs. tcp_limit_output_bytes 1296 limits the number of bytes on qdisc or device to reduce artificial 1297 RTT/cwnd and reduce bufferbloat. 1298 1299 Default: 4194304 (4 MB) 1300 1301tcp_challenge_ack_limit - INTEGER 1302 Limits number of Challenge ACK sent per second, as recommended 1303 in RFC 5961 (Improving TCP's Robustness to Blind In-Window Attacks) 1304 Note that this per netns rate limit can allow some side channel 1305 attacks and probably should not be enabled. 1306 TCP stack implements per TCP socket limits anyway. 1307 Default: INT_MAX (unlimited) 1308 1309tcp_ehash_entries - INTEGER 1310 Show the number of hash buckets for TCP sockets in the current 1311 networking namespace. 1312 1313 A negative value means the networking namespace does not own its 1314 hash buckets and shares the initial networking namespace's one. 1315 1316tcp_child_ehash_entries - INTEGER 1317 Control the number of hash buckets for TCP sockets in the child 1318 networking namespace, which must be set before clone() or unshare(). 1319 1320 If the value is not 0, the kernel uses a value rounded up to 2^n 1321 as the actual hash bucket size. 0 is a special value, meaning 1322 the child networking namespace will share the initial networking 1323 namespace's hash buckets. 1324 1325 Note that the child will use the global one in case the kernel 1326 fails to allocate enough memory. In addition, the global hash 1327 buckets are spread over available NUMA nodes, but the allocation 1328 of the child hash table depends on the current process's NUMA 1329 policy, which could result in performance differences. 1330 1331 Note also that the default value of tcp_max_tw_buckets and 1332 tcp_max_syn_backlog depend on the hash bucket size. 1333 1334 Possible values: 0, 2^n (n: 0 - 24 (16Mi)) 1335 1336 Default: 0 1337 1338tcp_plb_enabled - BOOLEAN 1339 If enabled and the underlying congestion control (e.g. DCTCP) supports 1340 and enables PLB feature, TCP PLB (Protective Load Balancing) is 1341 enabled. PLB is described in the following paper: 1342 https://doi.org/10.1145/3544216.3544226. Based on PLB parameters, 1343 upon sensing sustained congestion, TCP triggers a change in 1344 flow label field for outgoing IPv6 packets. A change in flow label 1345 field potentially changes the path of outgoing packets for switches 1346 that use ECMP/WCMP for routing. 1347 1348 PLB changes socket txhash which results in a change in IPv6 Flow Label 1349 field, and currently no-op for IPv4 headers. It is possible 1350 to apply PLB for IPv4 with other network header fields (e.g. TCP 1351 or IPv4 options) or using encapsulation where outer header is used 1352 by switches to determine next hop. In either case, further host 1353 and switch side changes will be needed. 1354 1355 If enabled, PLB assumes that congestion signal (e.g. ECN) is made 1356 available and used by congestion control module to estimate a 1357 congestion measure (e.g. ce_ratio). PLB needs a congestion measure to 1358 make repathing decisions. 1359 1360 Possible values: 1361 1362 - 0 (disabled) 1363 - 1 (enabled) 1364 1365 Default: 0 (disabled) 1366 1367tcp_plb_idle_rehash_rounds - INTEGER 1368 Number of consecutive congested rounds (RTT) seen after which 1369 a rehash can be performed, given there are no packets in flight. 1370 This is referred to as M in PLB paper: 1371 https://doi.org/10.1145/3544216.3544226. 1372 1373 Possible Values: 0 - 31 1374 1375 Default: 3 1376 1377tcp_plb_rehash_rounds - INTEGER 1378 Number of consecutive congested rounds (RTT) seen after which 1379 a forced rehash can be performed. Be careful when setting this 1380 parameter, as a small value increases the risk of retransmissions. 1381 This is referred to as N in PLB paper: 1382 https://doi.org/10.1145/3544216.3544226. 1383 1384 Possible Values: 0 - 31 1385 1386 Default: 12 1387 1388tcp_plb_suspend_rto_sec - INTEGER 1389 Time, in seconds, to suspend PLB in event of an RTO. In order to avoid 1390 having PLB repath onto a connectivity "black hole", after an RTO a TCP 1391 connection suspends PLB repathing for a random duration between 1x and 1392 2x of this parameter. Randomness is added to avoid concurrent rehashing 1393 of multiple TCP connections. This should be set corresponding to the 1394 amount of time it takes to repair a failed link. 1395 1396 Possible Values: 0 - 255 1397 1398 Default: 60 1399 1400tcp_plb_cong_thresh - INTEGER 1401 Fraction of packets marked with congestion over a round (RTT) to 1402 tag that round as congested. This is referred to as K in the PLB paper: 1403 https://doi.org/10.1145/3544216.3544226. 1404 1405 The 0-1 fraction range is mapped to 0-256 range to avoid floating 1406 point operations. For example, 128 means that if at least 50% of 1407 the packets in a round were marked as congested then the round 1408 will be tagged as congested. 1409 1410 Setting threshold to 0 means that PLB repaths every RTT regardless 1411 of congestion. This is not intended behavior for PLB and should be 1412 used only for experimentation purpose. 1413 1414 Possible Values: 0 - 256 1415 1416 Default: 128 1417 1418tcp_pingpong_thresh - INTEGER 1419 The number of estimated data replies sent for estimated incoming data 1420 requests that must happen before TCP considers that a connection is a 1421 "ping-pong" (request-response) connection for which delayed 1422 acknowledgments can provide benefits. 1423 1424 This threshold is 1 by default, but some applications may need a higher 1425 threshold for optimal performance. 1426 1427 Possible Values: 1 - 255 1428 1429 Default: 1 1430 1431tcp_rto_min_us - INTEGER 1432 Minimal TCP retransmission timeout (in microseconds). Note that the 1433 rto_min route option has the highest precedence for configuring this 1434 setting, followed by the TCP_BPF_RTO_MIN and TCP_RTO_MIN_US socket 1435 options, followed by this tcp_rto_min_us sysctl. 1436 1437 The recommended practice is to use a value less or equal to 200000 1438 microseconds. 1439 1440 Possible Values: 1 - INT_MAX 1441 1442 Default: 200000 1443 1444tcp_rto_max_ms - INTEGER 1445 Maximal TCP retransmission timeout (in ms). 1446 Note that TCP_RTO_MAX_MS socket option has higher precedence. 1447 1448 When changing tcp_rto_max_ms, it is important to understand 1449 that tcp_retries2 might need a change. 1450 1451 Possible Values: 1000 - 120,000 1452 1453 Default: 120,000 1454 1455UDP variables 1456============= 1457 1458udp_l3mdev_accept - BOOLEAN 1459 Enabling this option allows a "global" bound socket to work 1460 across L3 master domains (e.g., VRFs) with packets capable of 1461 being received regardless of the L3 domain in which they 1462 originated. Only valid when the kernel was compiled with 1463 CONFIG_NET_L3_MASTER_DEV. 1464 1465 Possible values: 1466 1467 - 0 (disabled) 1468 - 1 (enabled) 1469 1470 Default: 0 (disabled) 1471 1472udp_mem - vector of 3 INTEGERs: min, pressure, max 1473 Number of pages allowed for queueing by all UDP sockets. 1474 1475 min: Number of pages allowed for queueing by all UDP sockets. 1476 1477 pressure: This value was introduced to follow format of tcp_mem. 1478 1479 max: This value was introduced to follow format of tcp_mem. 1480 1481 Default is calculated at boot time from amount of available memory. 1482 1483udp_rmem_min - INTEGER 1484 Minimal size of receive buffer used by UDP sockets in moderation. 1485 Each UDP socket is able to use the size for receiving data, even if 1486 total pages of UDP sockets exceed udp_mem pressure. The unit is byte. 1487 1488 Default: 4K 1489 1490udp_wmem_min - INTEGER 1491 UDP does not have tx memory accounting and this tunable has no effect. 1492 1493udp_hash_entries - INTEGER 1494 Show the number of hash buckets for UDP sockets in the current 1495 networking namespace. 1496 1497 A negative value means the networking namespace does not own its 1498 hash buckets and shares the initial networking namespace's one. 1499 1500udp_child_hash_entries - INTEGER 1501 Control the number of hash buckets for UDP sockets in the child 1502 networking namespace, which must be set before clone() or unshare(). 1503 1504 If the value is not 0, the kernel uses a value rounded up to 2^n 1505 as the actual hash bucket size. 0 is a special value, meaning 1506 the child networking namespace will share the initial networking 1507 namespace's hash buckets. 1508 1509 Note that the child will use the global one in case the kernel 1510 fails to allocate enough memory. In addition, the global hash 1511 buckets are spread over available NUMA nodes, but the allocation 1512 of the child hash table depends on the current process's NUMA 1513 policy, which could result in performance differences. 1514 1515 Possible values: 0, 2^n (n: 7 (128) - 16 (64K)) 1516 1517 Default: 0 1518 1519 1520RAW variables 1521============= 1522 1523raw_l3mdev_accept - BOOLEAN 1524 Enabling this option allows a "global" bound socket to work 1525 across L3 master domains (e.g., VRFs) with packets capable of 1526 being received regardless of the L3 domain in which they 1527 originated. Only valid when the kernel was compiled with 1528 CONFIG_NET_L3_MASTER_DEV. 1529 1530 Possible values: 1531 1532 - 0 (disabled) 1533 - 1 (enabled) 1534 1535 Default: 1 (enabled) 1536 1537CIPSOv4 Variables 1538================= 1539 1540cipso_cache_enable - BOOLEAN 1541 If enabled, enable additions to and lookups from the CIPSO label mapping 1542 cache. If disabled, additions are ignored and lookups always result in a 1543 miss. However, regardless of the setting the cache is still 1544 invalidated when required when means you can safely toggle this on and 1545 off and the cache will always be "safe". 1546 1547 Possible values: 1548 1549 - 0 (disabled) 1550 - 1 (enabled) 1551 1552 Default: 1 (enabled) 1553 1554cipso_cache_bucket_size - INTEGER 1555 The CIPSO label cache consists of a fixed size hash table with each 1556 hash bucket containing a number of cache entries. This variable limits 1557 the number of entries in each hash bucket; the larger the value is, the 1558 more CIPSO label mappings that can be cached. When the number of 1559 entries in a given hash bucket reaches this limit adding new entries 1560 causes the oldest entry in the bucket to be removed to make room. 1561 1562 Default: 10 1563 1564cipso_rbm_optfmt - BOOLEAN 1565 Enable the "Optimized Tag 1 Format" as defined in section 3.4.2.6 of 1566 the CIPSO draft specification (see Documentation/netlabel for details). 1567 This means that when set the CIPSO tag will be padded with empty 1568 categories in order to make the packet data 32-bit aligned. 1569 1570 Possible values: 1571 1572 - 0 (disabled) 1573 - 1 (enabled) 1574 1575 Default: 0 (disabled) 1576 1577cipso_rbm_strictvalid - BOOLEAN 1578 If enabled, do a very strict check of the CIPSO option when 1579 ip_options_compile() is called. If disabled, relax the checks done during 1580 ip_options_compile(). Either way is "safe" as errors are caught else 1581 where in the CIPSO processing code but setting this to 0 (False) should 1582 result in less work (i.e. it should be faster) but could cause problems 1583 with other implementations that require strict checking. 1584 1585 Possible values: 1586 1587 - 0 (disabled) 1588 - 1 (enabled) 1589 1590 Default: 0 (disabled) 1591 1592IP Variables 1593============ 1594 1595ip_local_port_range - 2 INTEGERS 1596 Defines the local port range that is used by TCP and UDP to 1597 choose the local port. The first number is the first, the 1598 second the last local port number. 1599 If possible, it is better these numbers have different parity 1600 (one even and one odd value). 1601 Must be greater than or equal to ip_unprivileged_port_start. 1602 The default values are 32768 and 60999 respectively. 1603 1604ip_local_reserved_ports - list of comma separated ranges 1605 Specify the ports which are reserved for known third-party 1606 applications. These ports will not be used by automatic port 1607 assignments (e.g. when calling connect() or bind() with port 1608 number 0). Explicit port allocation behavior is unchanged. 1609 1610 The format used for both input and output is a comma separated 1611 list of ranges (e.g. "1,2-4,10-10" for ports 1, 2, 3, 4 and 1612 10). Writing to the file will clear all previously reserved 1613 ports and update the current list with the one given in the 1614 input. 1615 1616 Note that ip_local_port_range and ip_local_reserved_ports 1617 settings are independent and both are considered by the kernel 1618 when determining which ports are available for automatic port 1619 assignments. 1620 1621 You can reserve ports which are not in the current 1622 ip_local_port_range, e.g.:: 1623 1624 $ cat /proc/sys/net/ipv4/ip_local_port_range 1625 32000 60999 1626 $ cat /proc/sys/net/ipv4/ip_local_reserved_ports 1627 8080,9148 1628 1629 although this is redundant. However such a setting is useful 1630 if later the port range is changed to a value that will 1631 include the reserved ports. Also keep in mind, that overlapping 1632 of these ranges may affect probability of selecting ephemeral 1633 ports which are right after block of reserved ports. 1634 1635 Default: Empty 1636 1637ip_local_port_step_width - INTEGER 1638 Defines the numerical maximum increment between successive port 1639 allocations within the ephemeral port range when an unavailable port is 1640 reached. This can be used to mitigate accumulated nodes in port 1641 distribution when reserved ports have been configured. Please note that 1642 port collisions may be more frequent in a system with a very high load. 1643 1644 It is recommended to set this value strictly larger than the largest 1645 contiguous block of ports configure in ip_local_reserved_ports. For 1646 large reserved port ranges, setting this to 3x or 4x the size of the 1647 largest block is advised. Using a value equal or greater than the local 1648 port range size completely solves the uneven port distribution problem, 1649 but it can degrade performance under port exhaustion situations. 1650 1651 Default: 0 (disabled) 1652 1653ip_unprivileged_port_start - INTEGER 1654 This is a per-namespace sysctl. It defines the first 1655 unprivileged port in the network namespace. Privileged ports 1656 require root or CAP_NET_BIND_SERVICE in order to bind to them. 1657 To disable all privileged ports, set this to 0. They must not 1658 overlap with the ip_local_port_range. 1659 1660 Default: 1024 1661 1662ip_nonlocal_bind - BOOLEAN 1663 If enabled, allows processes to bind() to non-local IP addresses, 1664 which can be quite useful - but may break some applications. 1665 1666 Possible values: 1667 1668 - 0 (disabled) 1669 - 1 (enabled) 1670 1671 Default: 0 (disabled) 1672 1673ip_autobind_reuse - BOOLEAN 1674 By default, bind() does not select the ports automatically even if 1675 the new socket and all sockets bound to the port have SO_REUSEADDR. 1676 ip_autobind_reuse allows bind() to reuse the port and this is useful 1677 when you use bind()+connect(), but may break some applications. 1678 The preferred solution is to use IP_BIND_ADDRESS_NO_PORT and this 1679 option should only be set by experts. 1680 1681 Possible values: 1682 1683 - 0 (disabled) 1684 - 1 (enabled) 1685 1686 Default: 0 (disabled) 1687 1688ip_dynaddr - INTEGER 1689 If set non-zero, enables support for dynamic addresses. 1690 If set to a non-zero value larger than 1, a kernel log 1691 message will be printed when dynamic address rewriting 1692 occurs. 1693 1694 Default: 0 1695 1696ip_early_demux - BOOLEAN 1697 Optimize input packet processing down to one demux for 1698 certain kinds of local sockets. Currently we only do this 1699 for established TCP and connected UDP sockets. 1700 1701 It may add an additional cost for pure routing workloads that 1702 reduces overall throughput, in such case you should disable it. 1703 1704 Possible values: 1705 1706 - 0 (disabled) 1707 - 1 (enabled) 1708 1709 Default: 1 (enabled) 1710 1711ping_group_range - 2 INTEGERS 1712 Restrict ICMP_PROTO datagram sockets to users in the group range. 1713 The default is "1 0", meaning, that nobody (not even root) may 1714 create ping sockets. Setting it to "100 100" would grant permissions 1715 to the single group. "0 4294967294" would enable it for the world, "100 1716 4294967294" would enable it for the users, but not daemons. 1717 1718tcp_early_demux - BOOLEAN 1719 Enable early demux for established TCP sockets. 1720 1721 Possible values: 1722 1723 - 0 (disabled) 1724 - 1 (enabled) 1725 1726 Default: 1 (enabled) 1727 1728udp_early_demux - BOOLEAN 1729 Enable early demux for connected UDP sockets. Disable this if 1730 your system could experience more unconnected load. 1731 1732 Possible values: 1733 1734 - 0 (disabled) 1735 - 1 (enabled) 1736 1737 Default: 1 (enabled) 1738 1739icmp_echo_ignore_all - BOOLEAN 1740 If enabled, then the kernel will ignore all ICMP ECHO 1741 requests sent to it. 1742 1743 Possible values: 1744 1745 - 0 (disabled) 1746 - 1 (enabled) 1747 1748 Default: 0 (disabled) 1749 1750icmp_echo_enable_probe - BOOLEAN 1751 If enabled, then the kernel will respond to RFC 8335 PROBE 1752 requests sent to it. 1753 1754 Possible values: 1755 1756 - 0 (disabled) 1757 - 1 (enabled) 1758 1759 Default: 0 (disabled) 1760 1761icmp_echo_ignore_broadcasts - BOOLEAN 1762 If enabled, then the kernel will ignore all ICMP ECHO and 1763 TIMESTAMP requests sent to it via broadcast/multicast. 1764 1765 Possible values: 1766 1767 - 0 (disabled) 1768 - 1 (enabled) 1769 1770 Default: 1 (enabled) 1771 1772icmp_ratelimit - INTEGER 1773 Limit the maximal rates for sending ICMP packets whose type matches 1774 icmp_ratemask (see below) to specific targets. 1775 0 to disable any limiting, 1776 otherwise the minimal space between responses in milliseconds. 1777 Note that another sysctl, icmp_msgs_per_sec limits the number 1778 of ICMP packets sent on all targets. 1779 1780 Default: 1000 1781 1782icmp_msgs_per_sec - INTEGER 1783 Limit maximal number of ICMP packets sent per second from this host. 1784 Only messages whose type matches icmp_ratemask (see below) are 1785 controlled by this limit. For security reasons, the precise count 1786 of messages per second is randomized. 1787 1788 Default: 10000 1789 1790icmp_msgs_burst - INTEGER 1791 icmp_msgs_per_sec controls number of ICMP packets sent per second, 1792 while icmp_msgs_burst controls the token bucket size. 1793 For security reasons, the precise burst size is randomized. 1794 1795 Default: 10000 1796 1797icmp_ratemask - INTEGER 1798 Mask made of ICMP types for which rates are being limited. 1799 1800 Significant bits: IHGFEDCBA9876543210 1801 1802 Default mask: 0000001100000011000 (6168) 1803 1804 Bit definitions (see include/linux/icmp.h): 1805 1806 = ========================= 1807 0 Echo Reply 1808 3 Destination Unreachable [1]_ 1809 4 Source Quench [1]_ 1810 5 Redirect 1811 8 Echo Request 1812 B Time Exceeded [1]_ 1813 C Parameter Problem [1]_ 1814 D Timestamp Request 1815 E Timestamp Reply 1816 F Info Request 1817 G Info Reply 1818 H Address Mask Request 1819 I Address Mask Reply 1820 = ========================= 1821 1822 .. [1] These are rate limited by default (see default mask above) 1823 1824icmp_ignore_bogus_error_responses - BOOLEAN 1825 Some routers violate RFC1122 by sending bogus responses to broadcast 1826 frames. Such violations are normally logged via a kernel warning. 1827 If enabled, the kernel will not give such warnings, which 1828 will avoid log file clutter. 1829 1830 Possible values: 1831 1832 - 0 (disabled) 1833 - 1 (enabled) 1834 1835 Default: 1 (enabled) 1836 1837icmp_errors_use_inbound_ifaddr - BOOLEAN 1838 1839 If disabled, icmp error messages are sent with the primary address of 1840 the exiting interface. 1841 1842 If enabled, the message will be sent with the primary address of 1843 the interface that received the packet that caused the icmp error. 1844 This is the behaviour many network administrators will expect from 1845 a router. And it can make debugging complicated network layouts 1846 much easier. 1847 1848 Note that if no primary address exists for the interface selected, 1849 then the primary address of the first non-loopback interface that 1850 has one will be used regardless of this setting. 1851 1852 Possible values: 1853 1854 - 0 (disabled) 1855 - 1 (enabled) 1856 1857 Default: 0 (disabled) 1858 1859icmp_errors_extension_mask - UNSIGNED INTEGER 1860 Bitmask of ICMP extensions to append to ICMPv4 error messages 1861 ("Destination Unreachable", "Time Exceeded" and "Parameter Problem"). 1862 The original datagram is trimmed / padded to 128 bytes in order to be 1863 compatible with applications that do not comply with RFC 4884. 1864 1865 Possible extensions are: 1866 1867 ==== ============================================================== 1868 0x01 Incoming IP interface information according to RFC 5837. 1869 Extension will include the index, IPv4 address (if present), 1870 name and MTU of the IP interface that received the datagram 1871 which elicited the ICMP error. 1872 ==== ============================================================== 1873 1874 Default: 0x00 (no extensions) 1875 1876igmp_max_memberships - INTEGER 1877 Change the maximum number of multicast groups we can subscribe to. 1878 Default: 20 1879 1880 Theoretical maximum value is bounded by having to send a membership 1881 report in a single datagram (i.e. the report can't span multiple 1882 datagrams, or risk confusing the switch and leaving groups you don't 1883 intend to). 1884 1885 The number of supported groups 'M' is bounded by the number of group 1886 report entries you can fit into a single datagram of 65535 bytes. 1887 1888 M = 65536-sizeof (ip header)/(sizeof(Group record)) 1889 1890 Group records are variable length, with a minimum of 12 bytes. 1891 So net.ipv4.igmp_max_memberships should not be set higher than: 1892 1893 (65536-24) / 12 = 5459 1894 1895 The value 5459 assumes no IP header options, so in practice 1896 this number may be lower. 1897 1898igmp_max_msf - INTEGER 1899 Maximum number of addresses allowed in the source filter list for a 1900 multicast group. 1901 1902 Default: 10 1903 1904igmp_qrv - INTEGER 1905 Controls the IGMP query robustness variable (see RFC2236 8.1). 1906 1907 Default: 2 (as specified by RFC2236 8.1) 1908 1909 Minimum: 1 (as specified by RFC6636 4.5) 1910 1911force_igmp_version - INTEGER 1912 - 0 - (default) No enforcement of a IGMP version, IGMPv1/v2 fallback 1913 allowed. Will back to IGMPv3 mode again if all IGMPv1/v2 Querier 1914 Present timer expires. 1915 - 1 - Enforce to use IGMP version 1. Will also reply IGMPv1 report if 1916 receive IGMPv2/v3 query. 1917 - 2 - Enforce to use IGMP version 2. Will fallback to IGMPv1 if receive 1918 IGMPv1 query message. Will reply report if receive IGMPv3 query. 1919 - 3 - Enforce to use IGMP version 3. The same react with default 0. 1920 1921 .. note:: 1922 1923 this is not the same with force_mld_version because IGMPv3 RFC3376 1924 Security Considerations does not have clear description that we could 1925 ignore other version messages completely as MLDv2 RFC3810. So make 1926 this value as default 0 is recommended. 1927 1928``conf/interface/*`` 1929 changes special settings per interface (where 1930 interface" is the name of your network interface) 1931 1932``conf/all/*`` 1933 is special, changes the settings for all interfaces 1934 1935log_martians - BOOLEAN 1936 Log packets with impossible addresses to kernel log. 1937 log_martians for the interface will be enabled if at least one of 1938 conf/{all,interface}/log_martians is set to TRUE, 1939 it will be disabled otherwise 1940 1941accept_redirects - BOOLEAN 1942 Accept ICMP redirect messages. 1943 accept_redirects for the interface will be enabled if: 1944 1945 - both conf/{all,interface}/accept_redirects are TRUE in the case 1946 forwarding for the interface is enabled 1947 1948 or 1949 1950 - at least one of conf/{all,interface}/accept_redirects is TRUE in the 1951 case forwarding for the interface is disabled 1952 1953 accept_redirects for the interface will be disabled otherwise 1954 1955 default: 1956 1957 - TRUE (host) 1958 - FALSE (router) 1959 1960forwarding - BOOLEAN 1961 Enable IP forwarding on this interface. This controls whether packets 1962 received _on_ this interface can be forwarded. 1963 1964mc_forwarding - BOOLEAN 1965 Do multicast routing. The kernel needs to be compiled with CONFIG_MROUTE 1966 and a multicast routing daemon is required. 1967 conf/all/mc_forwarding must also be set to TRUE to enable multicast 1968 routing for the interface 1969 1970medium_id - INTEGER 1971 Integer value used to differentiate the devices by the medium they 1972 are attached to. Two devices can have different id values when 1973 the broadcast packets are received only on one of them. 1974 The default value 0 means that the device is the only interface 1975 to its medium, value of -1 means that medium is not known. 1976 1977 Currently, it is used to change the proxy_arp behavior: 1978 the proxy_arp feature is enabled for packets forwarded between 1979 two devices attached to different media. 1980 1981proxy_arp - BOOLEAN 1982 Do proxy arp. 1983 1984 proxy_arp for the interface will be enabled if at least one of 1985 conf/{all,interface}/proxy_arp is set to TRUE, 1986 it will be disabled otherwise 1987 1988proxy_arp_pvlan - BOOLEAN 1989 Private VLAN proxy arp. 1990 1991 Basically allow proxy arp replies back to the same interface 1992 (from which the ARP request/solicitation was received). 1993 1994 This is done to support (ethernet) switch features, like RFC 1995 3069, where the individual ports are NOT allowed to 1996 communicate with each other, but they are allowed to talk to 1997 the upstream router. As described in RFC 3069, it is possible 1998 to allow these hosts to communicate through the upstream 1999 router by proxy_arp'ing. Don't need to be used together with 2000 proxy_arp. 2001 2002 This technology is known by different names: 2003 2004 - In RFC 3069 it is called VLAN Aggregation. 2005 - Cisco and Allied Telesyn call it Private VLAN. 2006 - Hewlett-Packard call it Source-Port filtering or port-isolation. 2007 - Ericsson call it MAC-Forced Forwarding (RFC Draft). 2008 2009proxy_delay - INTEGER 2010 Delay proxy response. 2011 2012 Delay response to a neighbor solicitation when proxy_arp 2013 or proxy_ndp is enabled. A random value between [0, proxy_delay) 2014 will be chosen, setting to zero means reply with no delay. 2015 Value in jiffies. Defaults to 80. 2016 2017shared_media - BOOLEAN 2018 Send(router) or accept(host) RFC1620 shared media redirects. 2019 Overrides secure_redirects. 2020 2021 shared_media for the interface will be enabled if at least one of 2022 conf/{all,interface}/shared_media is set to TRUE, 2023 it will be disabled otherwise 2024 2025 default TRUE 2026 2027secure_redirects - BOOLEAN 2028 Accept ICMP redirect messages only to gateways listed in the 2029 interface's current gateway list. Even if disabled, RFC1122 redirect 2030 rules still apply. 2031 2032 Overridden by shared_media. 2033 2034 secure_redirects for the interface will be enabled if at least one of 2035 conf/{all,interface}/secure_redirects is set to TRUE, 2036 it will be disabled otherwise 2037 2038 default TRUE 2039 2040send_redirects - BOOLEAN 2041 Send redirects, if router. 2042 2043 send_redirects for the interface will be enabled if at least one of 2044 conf/{all,interface}/send_redirects is set to TRUE, 2045 it will be disabled otherwise 2046 2047 Default: TRUE 2048 2049bootp_relay - BOOLEAN 2050 Accept packets with source address 0.b.c.d destined 2051 not to this host as local ones. It is supposed, that 2052 BOOTP relay daemon will catch and forward such packets. 2053 conf/all/bootp_relay must also be set to TRUE to enable BOOTP relay 2054 for the interface 2055 2056 default FALSE 2057 2058 Not Implemented Yet. 2059 2060accept_source_route - BOOLEAN 2061 Accept packets with SRR option. 2062 conf/all/accept_source_route must also be set to TRUE to accept packets 2063 with SRR option on the interface 2064 2065 default 2066 2067 - TRUE (router) 2068 - FALSE (host) 2069 2070accept_local - BOOLEAN 2071 Accept packets with local source addresses. In combination with 2072 suitable routing, this can be used to direct packets between two 2073 local interfaces over the wire and have them accepted properly. 2074 default FALSE 2075 2076route_localnet - BOOLEAN 2077 Do not consider loopback addresses as martian source or destination 2078 while routing. This enables the use of 127/8 for local routing purposes. 2079 2080 default FALSE 2081 2082rp_filter - INTEGER 2083 - 0 - No source validation. 2084 - 1 - Strict mode as defined in RFC3704 Strict Reverse Path 2085 Each incoming packet is tested against the FIB and if the interface 2086 is not the best reverse path the packet check will fail. 2087 By default failed packets are discarded. 2088 - 2 - Loose mode as defined in RFC3704 Loose Reverse Path 2089 Each incoming packet's source address is also tested against the FIB 2090 and if the source address is not reachable via any interface 2091 the packet check will fail. 2092 2093 Current recommended practice in RFC3704 is to enable strict mode 2094 to prevent IP spoofing from DDos attacks. If using asymmetric routing 2095 or other complicated routing, then loose mode is recommended. 2096 2097 The max value from conf/{all,interface}/rp_filter is used 2098 when doing source validation on the {interface}. 2099 2100 Default value is 0. Note that some distributions enable it 2101 in startup scripts. 2102 2103src_valid_mark - BOOLEAN 2104 - 0 - The fwmark of the packet is not included in reverse path 2105 route lookup. This allows for asymmetric routing configurations 2106 utilizing the fwmark in only one direction, e.g., transparent 2107 proxying. 2108 2109 - 1 - The fwmark of the packet is included in reverse path route 2110 lookup. This permits rp_filter to function when the fwmark is 2111 used for routing traffic in both directions. 2112 2113 This setting also affects the utilization of fmwark when 2114 performing source address selection for ICMP replies, or 2115 determining addresses stored for the IPOPT_TS_TSANDADDR and 2116 IPOPT_RR IP options. 2117 2118 The max value from conf/{all,interface}/src_valid_mark is used. 2119 2120 Default value is 0. 2121 2122arp_filter - BOOLEAN 2123 - 1 - Allows you to have multiple network interfaces on the same 2124 subnet, and have the ARPs for each interface be answered 2125 based on whether or not the kernel would route a packet from 2126 the ARP'd IP out that interface (therefore you must use source 2127 based routing for this to work). In other words it allows control 2128 of which cards (usually 1) will respond to an arp request. 2129 2130 - 0 - (default) The kernel can respond to arp requests with addresses 2131 from other interfaces. This may seem wrong but it usually makes 2132 sense, because it increases the chance of successful communication. 2133 IP addresses are owned by the complete host on Linux, not by 2134 particular interfaces. Only for more complex setups like load- 2135 balancing, does this behaviour cause problems. 2136 2137 arp_filter for the interface will be enabled if at least one of 2138 conf/{all,interface}/arp_filter is set to TRUE, 2139 it will be disabled otherwise 2140 2141arp_announce - INTEGER 2142 Define different restriction levels for announcing the local 2143 source IP address from IP packets in ARP requests sent on 2144 interface: 2145 2146 - 0 - (default) Use any local address, configured on any interface 2147 - 1 - Try to avoid local addresses that are not in the target's 2148 subnet for this interface. This mode is useful when target 2149 hosts reachable via this interface require the source IP 2150 address in ARP requests to be part of their logical network 2151 configured on the receiving interface. When we generate the 2152 request we will check all our subnets that include the 2153 target IP and will preserve the source address if it is from 2154 such subnet. If there is no such subnet we select source 2155 address according to the rules for level 2. 2156 - 2 - Always use the best local address for this target. 2157 In this mode we ignore the source address in the IP packet 2158 and try to select local address that we prefer for talks with 2159 the target host. Such local address is selected by looking 2160 for primary IP addresses on all our subnets on the outgoing 2161 interface that include the target IP address. If no suitable 2162 local address is found we select the first local address 2163 we have on the outgoing interface or on all other interfaces, 2164 with the hope we will receive reply for our request and 2165 even sometimes no matter the source IP address we announce. 2166 2167 The max value from conf/{all,interface}/arp_announce is used. 2168 2169 Increasing the restriction level gives more chance for 2170 receiving answer from the resolved target while decreasing 2171 the level announces more valid sender's information. 2172 2173arp_ignore - INTEGER 2174 Define different modes for sending replies in response to 2175 received ARP requests that resolve local target IP addresses: 2176 2177 - 0 - (default): reply for any local target IP address, configured 2178 on any interface 2179 - 1 - reply only if the target IP address is local address 2180 configured on the incoming interface 2181 - 2 - reply only if the target IP address is local address 2182 configured on the incoming interface and both with the 2183 sender's IP address are part from same subnet on this interface 2184 - 3 - do not reply for local addresses configured with scope host, 2185 only resolutions for global and link addresses are replied 2186 - 4-7 - reserved 2187 - 8 - do not reply for all local addresses 2188 2189 The max value from conf/{all,interface}/arp_ignore is used 2190 when ARP request is received on the {interface} 2191 2192arp_notify - BOOLEAN 2193 Define mode for notification of address and device changes. 2194 2195 == ========================================================== 2196 0 (default): do nothing 2197 1 Generate gratuitous arp requests when device is brought up 2198 or hardware address changes. 2199 == ========================================================== 2200 2201arp_accept - INTEGER 2202 Define behavior for accepting gratuitous ARP (garp) frames from devices 2203 that are not already present in the ARP table: 2204 2205 - 0 - don't create new entries in the ARP table 2206 - 1 - create new entries in the ARP table 2207 - 2 - create new entries only if the source IP address is in the same 2208 subnet as an address configured on the interface that received the 2209 garp message. 2210 2211 Both replies and requests type gratuitous arp will trigger the 2212 ARP table to be updated, if this setting is on. 2213 2214 If the ARP table already contains the IP address of the 2215 gratuitous arp frame, the arp table will be updated regardless 2216 if this setting is on or off. 2217 2218arp_evict_nocarrier - BOOLEAN 2219 Clears the ARP cache on NOCARRIER events. This option is important for 2220 wireless devices where the ARP cache should not be cleared when roaming 2221 between access points on the same network. In most cases this should 2222 remain as the default (1). 2223 2224 Possible values: 2225 2226 - 0 (disabled) - Do not clear ARP cache on NOCARRIER events 2227 - 1 (enabled) - Clear the ARP cache on NOCARRIER events 2228 2229 Default: 1 (enabled) 2230 2231mcast_solicit - INTEGER 2232 The maximum number of multicast probes in INCOMPLETE state, 2233 when the associated hardware address is unknown. Defaults 2234 to 3. 2235 2236ucast_solicit - INTEGER 2237 The maximum number of unicast probes in PROBE state, when 2238 the hardware address is being reconfirmed. Defaults to 3. 2239 2240app_solicit - INTEGER 2241 The maximum number of probes to send to the user space ARP daemon 2242 via netlink before dropping back to multicast probes (see 2243 mcast_resolicit). Defaults to 0. 2244 2245mcast_resolicit - INTEGER 2246 The maximum number of multicast probes after unicast and 2247 app probes in PROBE state. Defaults to 0. 2248 2249disable_policy - BOOLEAN 2250 Disable IPSEC policy (SPD) for this interface 2251 2252 Possible values: 2253 2254 - 0 (disabled) 2255 - 1 (enabled) 2256 2257 Default: 0 (disabled) 2258 2259disable_xfrm - BOOLEAN 2260 Disable IPSEC encryption on this interface, whatever the policy 2261 2262 Possible values: 2263 2264 - 0 (disabled) 2265 - 1 (enabled) 2266 2267 Default: 0 (disabled) 2268 2269igmpv2_unsolicited_report_interval - INTEGER 2270 The interval in milliseconds in which the next unsolicited 2271 IGMPv1 or IGMPv2 report retransmit will take place. 2272 2273 Default: 10000 (10 seconds) 2274 2275igmpv3_unsolicited_report_interval - INTEGER 2276 The interval in milliseconds in which the next unsolicited 2277 IGMPv3 report retransmit will take place. 2278 2279 Default: 1000 (1 seconds) 2280 2281ignore_routes_with_linkdown - BOOLEAN 2282 Ignore routes whose link is down when performing a FIB lookup. 2283 2284 Possible values: 2285 2286 - 0 (disabled) 2287 - 1 (enabled) 2288 2289 Default: 0 (disabled) 2290 2291promote_secondaries - BOOLEAN 2292 When a primary IP address is removed from this interface 2293 promote a corresponding secondary IP address instead of 2294 removing all the corresponding secondary IP addresses. 2295 2296 Possible values: 2297 2298 - 0 (disabled) 2299 - 1 (enabled) 2300 2301 Default: 0 (disabled) 2302 2303drop_unicast_in_l2_multicast - BOOLEAN 2304 Drop any unicast IP packets that are received in link-layer 2305 multicast (or broadcast) frames. 2306 2307 This behavior (for multicast) is actually a SHOULD in RFC 2308 1122, but is disabled by default for compatibility reasons. 2309 2310 Possible values: 2311 2312 - 0 (disabled) 2313 - 1 (enabled) 2314 2315 Default: 0 (disabled) 2316 2317drop_gratuitous_arp - BOOLEAN 2318 Drop all gratuitous ARP frames, for example if there's a known 2319 good ARP proxy on the network and such frames need not be used 2320 (or in the case of 802.11, must not be used to prevent attacks.) 2321 2322 Possible values: 2323 2324 - 0 (disabled) 2325 - 1 (enabled) 2326 2327 Default: 0 (disabled) 2328 2329 2330tag - INTEGER 2331 Allows you to write a number, which can be used as required. 2332 2333 Default value is 0. 2334 2335xfrm4_gc_thresh - INTEGER 2336 (Obsolete since linux-4.14) 2337 The threshold at which we will start garbage collecting for IPv4 2338 destination cache entries. At twice this value the system will 2339 refuse new allocations. 2340 2341igmp_link_local_mcast_reports - BOOLEAN 2342 Enable IGMP reports for link local multicast groups in the 2343 224.0.0.X range. 2344 2345 Default TRUE 2346 2347Alexey Kuznetsov. 2348kuznet@ms2.inr.ac.ru 2349 2350Updated by: 2351 2352- Andi Kleen 2353 ak@muc.de 2354- Nicolas Delon 2355 delon.nicolas@wanadoo.fr 2356 2357 2358 2359 2360/proc/sys/net/ipv6/* Variables 2361============================== 2362 2363IPv6 has no global variables such as tcp_*. tcp_* settings under ipv4/ also 2364apply to IPv6 [XXX?]. 2365 2366bindv6only - BOOLEAN 2367 Default value for IPV6_V6ONLY socket option, 2368 which restricts use of the IPv6 socket to IPv6 communication 2369 only. 2370 2371 Possible values: 2372 2373 - 0 (disabled) - enable IPv4-mapped address feature 2374 - 1 (enabled) - disable IPv4-mapped address feature 2375 2376 Default: 0 (disabled) 2377 2378flowlabel_consistency - BOOLEAN 2379 Protect the consistency (and unicity) of flow label. 2380 You have to disable it to use IPV6_FL_F_REFLECT flag on the 2381 flow label manager. 2382 2383 Possible values: 2384 2385 - 0 (disabled) 2386 - 1 (enabled) 2387 2388 Default: 1 (enabled) 2389 2390auto_flowlabels - INTEGER 2391 Automatically generate flow labels based on a flow hash of the 2392 packet. This allows intermediate devices, such as routers, to 2393 identify packet flows for mechanisms like Equal Cost Multipath 2394 Routing (see RFC 6438). 2395 2396 = =========================================================== 2397 0 automatic flow labels are completely disabled 2398 1 automatic flow labels are enabled by default, they can be 2399 disabled on a per socket basis using the IPV6_AUTOFLOWLABEL 2400 socket option 2401 2 automatic flow labels are allowed, they may be enabled on a 2402 per socket basis using the IPV6_AUTOFLOWLABEL socket option 2403 3 automatic flow labels are enabled and enforced, they cannot 2404 be disabled by the socket option 2405 = =========================================================== 2406 2407 Default: 1 2408 2409flowlabel_state_ranges - BOOLEAN 2410 Split the flow label number space into two ranges. 0-0x7FFFF is 2411 reserved for the IPv6 flow manager facility, 0x80000-0xFFFFF 2412 is reserved for stateless flow labels as described in RFC6437. 2413 2414 Possible values: 2415 2416 - 0 (disabled) 2417 - 1 (enabled) 2418 2419 Default: 1 (enabled) 2420 2421 2422flowlabel_reflect - INTEGER 2423 Control flow label reflection. Needed for Path MTU 2424 Discovery to work with Equal Cost Multipath Routing in anycast 2425 environments. See RFC 7690 and: 2426 https://tools.ietf.org/html/draft-wang-6man-flow-label-reflection-01 2427 2428 This is a bitmask. 2429 2430 - 1: enabled for established flows 2431 2432 Note that this prevents automatic flowlabel changes, as done 2433 in "tcp: change IPv6 flow-label upon receiving spurious retransmission" 2434 and "tcp: Change txhash on every SYN and RTO retransmit" 2435 2436 - 2: enabled for TCP RESET packets (no active listener) 2437 If set, a RST packet sent in response to a SYN packet on a closed 2438 port will reflect the incoming flow label. 2439 2440 - 4: enabled for ICMPv6 echo reply messages. 2441 2442 Default: 0 2443 2444fib_multipath_hash_policy - INTEGER 2445 Controls which hash policy to use for multipath routes. 2446 2447 Default: 0 (Layer 3) 2448 2449 Possible values: 2450 2451 - 0 - Layer 3 (source and destination addresses plus flow label). 2452 For IPv6 TCP, the local ECMP path is selected from the socket 2453 txhash rather than the flow label, and may change after a TCP 2454 rehash event (such as a retransmission timeout) to recover from 2455 path failure. The on-wire flow label is unaffected. 2456 - 1 - Layer 4 (standard 5-tuple) 2457 - 2 - Layer 3 or inner Layer 3 if present 2458 - 3 - Custom multipath hash. Fields used for multipath hash calculation 2459 are determined by fib_multipath_hash_fields sysctl 2460 2461fib_multipath_hash_fields - UNSIGNED INTEGER 2462 When fib_multipath_hash_policy is set to 3 (custom multipath hash), the 2463 fields used for multipath hash calculation are determined by this 2464 sysctl. 2465 2466 This value is a bitmask which enables various fields for multipath hash 2467 calculation. 2468 2469 Possible fields are: 2470 2471 ====== ============================ 2472 0x0001 Source IP address 2473 0x0002 Destination IP address 2474 0x0004 IP protocol 2475 0x0008 Flow Label 2476 0x0010 Source port 2477 0x0020 Destination port 2478 0x0040 Inner source IP address 2479 0x0080 Inner destination IP address 2480 0x0100 Inner IP protocol 2481 0x0200 Inner Flow Label 2482 0x0400 Inner source port 2483 0x0800 Inner destination port 2484 ====== ============================ 2485 2486 Default: 0x0007 (source IP, destination IP and IP protocol) 2487 2488anycast_src_echo_reply - BOOLEAN 2489 Controls the use of anycast addresses as source addresses for ICMPv6 2490 echo reply 2491 2492 Possible values: 2493 2494 - 0 (disabled) 2495 - 1 (enabled) 2496 2497 Default: 0 (disabled) 2498 2499 2500idgen_delay - INTEGER 2501 Controls the delay in seconds after which time to retry 2502 privacy stable address generation if a DAD conflict is 2503 detected. 2504 2505 Default: 1 (as specified in RFC7217) 2506 2507idgen_retries - INTEGER 2508 Controls the number of retries to generate a stable privacy 2509 address if a DAD conflict is detected. 2510 2511 Default: 3 (as specified in RFC7217) 2512 2513mld_qrv - INTEGER 2514 Controls the MLD query robustness variable (see RFC3810 9.1). 2515 2516 Default: 2 (as specified by RFC3810 9.1) 2517 2518 Minimum: 1 (as specified by RFC6636 4.5) 2519 2520max_dst_opts_number - INTEGER 2521 Maximum number of non-padding TLVs allowed in a Destination 2522 options extension header. If this value is less than zero 2523 then unknown options are disallowed and the number of known 2524 TLVs allowed is the absolute value of this number. 2525 2526 Default: 8 2527 2528max_hbh_opts_number - INTEGER 2529 Maximum number of non-padding TLVs allowed in a Hop-by-Hop 2530 options extension header. If this value is less than zero 2531 then unknown options are disallowed and the number of known 2532 TLVs allowed is the absolute value of this number. 2533 2534 Default: 8 2535 2536max_dst_opts_length - INTEGER 2537 Maximum length allowed for a Destination options extension 2538 header. 2539 2540 Default: INT_MAX (unlimited) 2541 2542max_hbh_length - INTEGER 2543 Maximum length allowed for a Hop-by-Hop options extension 2544 header. 2545 2546 Default: INT_MAX (unlimited) 2547 2548skip_notify_on_dev_down - BOOLEAN 2549 Controls whether an RTM_DELROUTE message is generated for routes 2550 removed when a device is taken down or deleted. IPv4 does not 2551 generate this message; IPv6 does by default. Setting this sysctl 2552 to true skips the message, making IPv4 and IPv6 on par in relying 2553 on userspace caches to track link events and evict routes. 2554 2555 Possible values: 2556 2557 - 0 (disabled) - generate the message 2558 - 1 (enabled) - skip generating the message 2559 2560 Default: 0 (disabled) 2561 2562nexthop_compat_mode - BOOLEAN 2563 New nexthop API provides a means for managing nexthops independent of 2564 prefixes. Backwards compatibility with old route format is enabled by 2565 default which means route dumps and notifications contain the new 2566 nexthop attribute but also the full, expanded nexthop definition. 2567 Further, updates or deletes of a nexthop configuration generate route 2568 notifications for each fib entry using the nexthop. Once a system 2569 understands the new API, this sysctl can be disabled to achieve full 2570 performance benefits of the new API by disabling the nexthop expansion 2571 and extraneous notifications. 2572 2573 Note that as a backward-compatible mode, dumping of modern features 2574 might be incomplete or wrong. For example, resilient groups will not be 2575 shown as such, but rather as just a list of next hops. Also weights that 2576 do not fit into 8 bits will show incorrectly. 2577 2578 Default: true (backward compat mode) 2579 2580fib_notify_on_flag_change - INTEGER 2581 Whether to emit RTM_NEWROUTE notifications whenever RTM_F_OFFLOAD/ 2582 RTM_F_TRAP/RTM_F_OFFLOAD_FAILED flags are changed. 2583 2584 After installing a route to the kernel, user space receives an 2585 acknowledgment, which means the route was installed in the kernel, 2586 but not necessarily in hardware. 2587 It is also possible for a route already installed in hardware to change 2588 its action and therefore its flags. For example, a host route that is 2589 trapping packets can be "promoted" to perform decapsulation following 2590 the installation of an IPinIP/VXLAN tunnel. 2591 The notifications will indicate to user-space the state of the route. 2592 2593 Default: 0 (Do not emit notifications.) 2594 2595 Possible values: 2596 2597 - 0 - Do not emit notifications. 2598 - 1 - Emit notifications. 2599 - 2 - Emit notifications only for RTM_F_OFFLOAD_FAILED flag change. 2600 2601ioam6_id - INTEGER 2602 Define the IOAM id of this node. Uses only 24 bits out of 32 in total. 2603 2604 Possible value range: 2605 2606 - Min: 0 2607 - Max: 0xFFFFFF 2608 2609 Default: 0xFFFFFF 2610 2611ioam6_id_wide - LONG INTEGER 2612 Define the wide IOAM id of this node. Uses only 56 bits out of 64 in 2613 total. Can be different from ioam6_id. 2614 2615 Possible value range: 2616 2617 - Min: 0 2618 - Max: 0xFFFFFFFFFFFFFF 2619 2620 Default: 0xFFFFFFFFFFFFFF 2621 2622IPv6 Fragmentation: 2623 2624ip6frag_high_thresh - INTEGER 2625 Maximum memory used to reassemble IPv6 fragments. When 2626 ip6frag_high_thresh bytes of memory is allocated for this purpose, 2627 the fragment handler will toss packets until ip6frag_low_thresh 2628 is reached. 2629 2630ip6frag_low_thresh - INTEGER 2631 See ip6frag_high_thresh 2632 2633ip6frag_time - INTEGER 2634 Time in seconds to keep an IPv6 fragment in memory. 2635 2636``conf/default/*``: 2637 Change the interface-specific default settings. 2638 2639 These settings would be used during creating new interfaces. 2640 2641 2642``conf/all/*``: 2643 Change all the interface-specific settings. 2644 2645 [XXX: Other special features than forwarding?] 2646 2647conf/all/disable_ipv6 - BOOLEAN 2648 Changing this value is same as changing ``conf/default/disable_ipv6`` 2649 setting and also all per-interface ``disable_ipv6`` settings to the same 2650 value. 2651 2652 Reading this value does not have any particular meaning. It does not say 2653 whether IPv6 support is enabled or disabled. Returned value can be 1 2654 also in the case when some interface has ``disable_ipv6`` set to 0 and 2655 has configured IPv6 addresses. 2656 2657conf/all/forwarding - BOOLEAN 2658 Enable global IPv6 forwarding between all interfaces. 2659 2660 IPv4 and IPv6 work differently here; the ``force_forwarding`` flag must 2661 be used to control which interfaces may forward packets. 2662 2663 This also sets all interfaces' Host/Router setting 2664 'forwarding' to the specified value. See below for details. 2665 2666 This referred to as global forwarding. 2667 2668proxy_ndp - BOOLEAN 2669 Do proxy ndp. 2670 2671 Possible values: 2672 2673 - 0 (disabled) 2674 - 1 (enabled) 2675 2676 Default: 0 (disabled) 2677 2678force_forwarding - BOOLEAN 2679 Enable forwarding on this interface only -- regardless of the setting on 2680 ``conf/all/forwarding``. When setting ``conf.all.forwarding`` to 0, 2681 the ``force_forwarding`` flag will be reset on all interfaces. 2682 2683fwmark_reflect - BOOLEAN 2684 Controls the fwmark of kernel-generated IPv6 reply packets that are not 2685 associated with a socket for example, TCP RSTs or ICMPv6 echo replies). 2686 If disabled, these packets have a fwmark of zero. If enabled, they have the 2687 fwmark of the packet they are replying to. 2688 2689 Possible values: 2690 2691 - 0 (disabled) 2692 - 1 (enabled) 2693 2694 Default: 0 (disabled) 2695 2696``conf/interface/*``: 2697 Change special settings per interface. 2698 2699 The functional behaviour for certain settings is different 2700 depending on whether local forwarding is enabled or not. 2701 2702accept_ra - INTEGER 2703 Accept Router Advertisements; autoconfigure using them. 2704 2705 It also determines whether or not to transmit Router 2706 Solicitations. If and only if the functional setting is to 2707 accept Router Advertisements, Router Solicitations will be 2708 transmitted. 2709 2710 Possible values are: 2711 2712 == =========================================================== 2713 0 Do not accept Router Advertisements. 2714 1 Accept Router Advertisements if forwarding is disabled. 2715 2 Overrule forwarding behaviour. Accept Router Advertisements 2716 even if forwarding is enabled. 2717 == =========================================================== 2718 2719 Functional default: 2720 2721 - enabled if local forwarding is disabled. 2722 - disabled if local forwarding is enabled. 2723 2724accept_ra_defrtr - BOOLEAN 2725 Learn default router in Router Advertisement. 2726 2727 Functional default: 2728 2729 - enabled if accept_ra is enabled. 2730 - disabled if accept_ra is disabled. 2731 2732ra_defrtr_metric - UNSIGNED INTEGER 2733 Route metric for default route learned in Router Advertisement. This value 2734 will be assigned as metric for the default route learned via IPv6 Router 2735 Advertisement. Takes affect only if accept_ra_defrtr is enabled. 2736 2737 Possible values: 2738 1 to 0xFFFFFFFF 2739 2740 Default: IP6_RT_PRIO_USER i.e. 1024. 2741 2742accept_ra_from_local - BOOLEAN 2743 Accept RA with source-address that is found on local machine 2744 if the RA is otherwise proper and able to be accepted. 2745 2746 Default is to NOT accept these as it may be an un-intended 2747 network loop. 2748 2749 Functional default: 2750 2751 - enabled if accept_ra_from_local is enabled 2752 on a specific interface. 2753 - disabled if accept_ra_from_local is disabled 2754 on a specific interface. 2755 2756accept_ra_min_hop_limit - INTEGER 2757 Minimum hop limit Information in Router Advertisement. 2758 2759 Hop limit Information in Router Advertisement less than this 2760 variable shall be ignored. 2761 2762 Default: 1 2763 2764accept_ra_min_lft - INTEGER 2765 Minimum acceptable lifetime value in Router Advertisement. 2766 2767 RA sections with a lifetime less than this value shall be 2768 ignored. Zero lifetimes stay unaffected. 2769 2770 Default: 0 2771 2772accept_ra_pinfo - BOOLEAN 2773 Learn Prefix Information in Router Advertisement. 2774 2775 Functional default: 2776 2777 - enabled if accept_ra is enabled. 2778 - disabled if accept_ra is disabled. 2779 2780ra_honor_pio_life - BOOLEAN 2781 Whether to use RFC4862 Section 5.5.3e to determine the valid 2782 lifetime of an address matching a prefix sent in a Router 2783 Advertisement Prefix Information Option. 2784 2785 Possible values: 2786 2787 - 0 (disabled) - RFC4862 section 5.5.3e is used to determine 2788 the valid lifetime of the address. 2789 - 1 (enabled) - the PIO valid lifetime will always be honored. 2790 2791 Default: 0 (disabled) 2792 2793ra_honor_pio_pflag - BOOLEAN 2794 The Prefix Information Option P-flag indicates the network can 2795 allocate a unique IPv6 prefix per client using DHCPv6-PD. 2796 This sysctl can be enabled when a userspace DHCPv6-PD client 2797 is running to cause the P-flag to take effect: i.e. the 2798 P-flag suppresses any effects of the A-flag within the same 2799 PIO. For a given PIO, P=1 and A=1 is treated as A=0. 2800 2801 Possible values: 2802 2803 - 0 (disabled) - the P-flag is ignored. 2804 - 1 (enabled) - the P-flag will disable SLAAC autoconfiguration 2805 for the given Prefix Information Option. 2806 2807 Default: 0 (disabled) 2808 2809accept_ra_rt_info_min_plen - INTEGER 2810 Minimum prefix length of Route Information in RA. 2811 2812 Route Information w/ prefix smaller than this variable shall 2813 be ignored. 2814 2815 Functional default: 2816 2817 * 0 if accept_ra_rtr_pref is enabled. 2818 * -1 if accept_ra_rtr_pref is disabled. 2819 2820accept_ra_rt_info_max_plen - INTEGER 2821 Maximum prefix length of Route Information in RA. 2822 2823 Route Information w/ prefix larger than this variable shall 2824 be ignored. 2825 2826 Functional default: 2827 2828 * 0 if accept_ra_rtr_pref is enabled. 2829 * -1 if accept_ra_rtr_pref is disabled. 2830 2831accept_ra_rtr_pref - BOOLEAN 2832 Accept Router Preference in RA. 2833 2834 Functional default: 2835 2836 - enabled if accept_ra is enabled. 2837 - disabled if accept_ra is disabled. 2838 2839accept_ra_mtu - BOOLEAN 2840 Apply the MTU value specified in RA option 5 (RFC4861). If 2841 disabled, the MTU specified in the RA will be ignored. 2842 2843 Functional default: 2844 2845 - enabled if accept_ra is enabled. 2846 - disabled if accept_ra is disabled. 2847 2848accept_redirects - BOOLEAN 2849 Accept Redirects. 2850 2851 Functional default: 2852 2853 - enabled if local forwarding is disabled. 2854 - disabled if local forwarding is enabled. 2855 2856accept_source_route - INTEGER 2857 Accept source routing (routing extension header). 2858 2859 - >= 0: Accept only routing header type 2. 2860 - < 0: Do not accept routing header. 2861 2862 Default: 0 2863 2864autoconf - BOOLEAN 2865 Autoconfigure addresses using Prefix Information in Router 2866 Advertisements. 2867 2868 Functional default: 2869 2870 - enabled if accept_ra_pinfo is enabled. 2871 - disabled if accept_ra_pinfo is disabled. 2872 2873dad_transmits - INTEGER 2874 The amount of Duplicate Address Detection probes to send. 2875 2876 Default: 1 2877 2878forwarding - INTEGER 2879 Configure interface-specific Host/Router behaviour. 2880 2881 .. note:: 2882 2883 It is recommended to have the same setting on all 2884 interfaces; mixed router/host scenarios are rather uncommon. 2885 2886 Possible values are: 2887 2888 - 0 Forwarding disabled 2889 - 1 Forwarding enabled 2890 2891 **FALSE (0)**: 2892 2893 By default, Host behaviour is assumed. This means: 2894 2895 1. IsRouter flag is not set in Neighbour Advertisements. 2896 2. If accept_ra is TRUE (default), transmit Router 2897 Solicitations. 2898 3. If accept_ra is TRUE (default), accept Router 2899 Advertisements (and do autoconfiguration). 2900 4. If accept_redirects is TRUE (default), accept Redirects. 2901 2902 **TRUE (1)**: 2903 2904 If local forwarding is enabled, Router behaviour is assumed. 2905 This means exactly the reverse from the above: 2906 2907 1. IsRouter flag is set in Neighbour Advertisements. 2908 2. Router Solicitations are not sent unless accept_ra is 2. 2909 3. Router Advertisements are ignored unless accept_ra is 2. 2910 4. Redirects are ignored. 2911 2912 Default: 0 (disabled) if global forwarding is disabled (default), 2913 otherwise 1 (enabled). 2914 2915hop_limit - INTEGER 2916 Default Hop Limit to set. 2917 2918 Default: 64 2919 2920mtu - INTEGER 2921 Default Maximum Transfer Unit 2922 2923 Default: 1280 (IPv6 required minimum) 2924 2925ip_nonlocal_bind - BOOLEAN 2926 If enabled, allows processes to bind() to non-local IPv6 addresses, 2927 which can be quite useful - but may break some applications. 2928 2929 Possible values: 2930 2931 - 0 (disabled) 2932 - 1 (enabled) 2933 2934 Default: 0 (disabled) 2935 2936router_probe_interval - INTEGER 2937 Minimum interval (in seconds) between Router Probing described 2938 in RFC4191. 2939 2940 Default: 60 2941 2942router_solicitation_delay - INTEGER 2943 Number of seconds to wait after interface is brought up 2944 before sending Router Solicitations. 2945 2946 Default: 1 2947 2948router_solicitation_interval - INTEGER 2949 Number of seconds to wait between Router Solicitations. 2950 2951 Default: 4 2952 2953router_solicitations - INTEGER 2954 Number of Router Solicitations to send until assuming no 2955 routers are present. 2956 2957 Default: 3 2958 2959use_oif_addrs_only - BOOLEAN 2960 When enabled, the candidate source addresses for destinations 2961 routed via this interface are restricted to the set of addresses 2962 configured on this interface (vis. RFC 6724, section 4). 2963 2964 Possible values: 2965 2966 - 0 (disabled) 2967 - 1 (enabled) 2968 2969 Default: 0 (disabled) 2970 2971use_tempaddr - INTEGER 2972 Preference for Privacy Extensions (RFC3041). 2973 2974 * <= 0 : disable Privacy Extensions 2975 * == 1 : enable Privacy Extensions, but prefer public 2976 addresses over temporary addresses. 2977 * > 1 : enable Privacy Extensions and prefer temporary 2978 addresses over public addresses. 2979 2980 Default: 2981 2982 * 0 (for most devices) 2983 * -1 (for point-to-point devices and loopback devices) 2984 2985temp_valid_lft - INTEGER 2986 valid lifetime (in seconds) for temporary addresses. If less than the 2987 minimum required lifetime (typically 5-7 seconds), temporary addresses 2988 will not be created. 2989 2990 Default: 172800 (2 days) 2991 2992temp_prefered_lft - INTEGER 2993 Preferred lifetime (in seconds) for temporary addresses. If 2994 temp_prefered_lft is less than the minimum required lifetime (typically 2995 5-7 seconds), the preferred lifetime is the minimum required. If 2996 temp_prefered_lft is greater than temp_valid_lft, the preferred lifetime 2997 is temp_valid_lft. 2998 2999 Default: 86400 (1 day) 3000 3001keep_addr_on_down - INTEGER 3002 Keep all IPv6 addresses on an interface down event. If set static 3003 global addresses with no expiration time are not flushed. 3004 3005 * >0 : enabled 3006 * 0 : system default 3007 * <0 : disabled 3008 3009 Default: 0 (addresses are removed) 3010 3011max_desync_factor - INTEGER 3012 Maximum value for DESYNC_FACTOR, which is a random value 3013 that ensures that clients don't synchronize with each 3014 other and generate new addresses at exactly the same time. 3015 value is in seconds. 3016 3017 Default: 600 3018 3019regen_min_advance - INTEGER 3020 How far in advance (in seconds), at minimum, to create a new temporary 3021 address before the current one is deprecated. This value is added to 3022 the amount of time that may be required for duplicate address detection 3023 to determine when to create a new address. Linux permits setting this 3024 value to less than the default of 2 seconds, but a value less than 2 3025 does not conform to RFC 8981. 3026 3027 Default: 2 3028 3029regen_max_retry - INTEGER 3030 Number of attempts before give up attempting to generate 3031 valid temporary addresses. 3032 3033 Default: 5 3034 3035max_addresses - INTEGER 3036 Maximum number of autoconfigured addresses per interface. Setting 3037 to zero disables the limitation. It is not recommended to set this 3038 value too large (or to zero) because it would be an easy way to 3039 crash the kernel by allowing too many addresses to be created. 3040 3041 Default: 16 3042 3043disable_ipv6 - BOOLEAN 3044 Disable IPv6 operation. If accept_dad is set to 2, this value 3045 will be dynamically set to TRUE if DAD fails for the link-local 3046 address. 3047 3048 Default: FALSE (enable IPv6 operation) 3049 3050 When this value is changed from 1 to 0 (IPv6 is being enabled), 3051 it will dynamically create a link-local address on the given 3052 interface and start Duplicate Address Detection, if necessary. 3053 3054 When this value is changed from 0 to 1 (IPv6 is being disabled), 3055 it will dynamically delete all addresses and routes on the given 3056 interface. From now on it will not possible to add addresses/routes 3057 to the selected interface. 3058 3059accept_dad - INTEGER 3060 Whether to accept DAD (Duplicate Address Detection). 3061 3062 == ============================================================== 3063 0 Disable DAD 3064 1 Enable DAD (default) 3065 2 Enable DAD, and disable IPv6 operation if MAC-based duplicate 3066 link-local address has been found. 3067 == ============================================================== 3068 3069 DAD operation and mode on a given interface will be selected according 3070 to the maximum value of conf/{all,interface}/accept_dad. 3071 3072force_tllao - BOOLEAN 3073 Enable sending the target link-layer address option even when 3074 responding to a unicast neighbor solicitation. 3075 3076 Default: FALSE 3077 3078 Quoting from RFC 2461, section 4.4, Target link-layer address: 3079 3080 "The option MUST be included for multicast solicitations in order to 3081 avoid infinite Neighbor Solicitation "recursion" when the peer node 3082 does not have a cache entry to return a Neighbor Advertisements 3083 message. When responding to unicast solicitations, the option can be 3084 omitted since the sender of the solicitation has the correct link- 3085 layer address; otherwise it would not have be able to send the unicast 3086 solicitation in the first place. However, including the link-layer 3087 address in this case adds little overhead and eliminates a potential 3088 race condition where the sender deletes the cached link-layer address 3089 prior to receiving a response to a previous solicitation." 3090 3091ndisc_notify - BOOLEAN 3092 Define mode for notification of address and device changes. 3093 3094 Possible values: 3095 3096 - 0 (disabled) - do nothing 3097 - 1 (enabled) - Generate unsolicited neighbour advertisements when device is brought 3098 up or hardware address changes. 3099 3100 Default: 0 (disabled) 3101 3102ndisc_tclass - INTEGER 3103 The IPv6 Traffic Class to use by default when sending IPv6 Neighbor 3104 Discovery (Router Solicitation, Router Advertisement, Neighbor 3105 Solicitation, Neighbor Advertisement, Redirect) messages. 3106 These 8 bits can be interpreted as 6 high order bits holding the DSCP 3107 value and 2 low order bits representing ECN (which you probably want 3108 to leave cleared). 3109 3110 * 0 - (default) 3111 3112ndisc_evict_nocarrier - BOOLEAN 3113 Clears the neighbor discovery table on NOCARRIER events. This option is 3114 important for wireless devices where the neighbor discovery cache should 3115 not be cleared when roaming between access points on the same network. 3116 In most cases this should remain as the default (1). 3117 3118 Possible values: 3119 3120 - 0 (disabled) - Do not clear neighbor discovery cache on NOCARRIER events. 3121 - 1 (enabled) - Clear neighbor discover cache on NOCARRIER events. 3122 3123 Default: 1 (enabled) 3124 3125mldv1_unsolicited_report_interval - INTEGER 3126 The interval in milliseconds in which the next unsolicited 3127 MLDv1 report retransmit will take place. 3128 3129 Default: 10000 (10 seconds) 3130 3131mldv2_unsolicited_report_interval - INTEGER 3132 The interval in milliseconds in which the next unsolicited 3133 MLDv2 report retransmit will take place. 3134 3135 Default: 1000 (1 second) 3136 3137force_mld_version - INTEGER 3138 * 0 - (default) No enforcement of a MLD version, MLDv1 fallback allowed 3139 * 1 - Enforce to use MLD version 1 3140 * 2 - Enforce to use MLD version 2 3141 3142suppress_frag_ndisc - INTEGER 3143 Control RFC 6980 (Security Implications of IPv6 Fragmentation 3144 with IPv6 Neighbor Discovery) behavior: 3145 3146 * 1 - (default) discard fragmented neighbor discovery packets 3147 * 0 - allow fragmented neighbor discovery packets 3148 3149optimistic_dad - BOOLEAN 3150 Whether to perform Optimistic Duplicate Address Detection (RFC 4429). 3151 3152 Optimistic Duplicate Address Detection for the interface will be enabled 3153 if at least one of conf/{all,interface}/optimistic_dad is set to 1, 3154 it will be disabled otherwise. 3155 3156 Possible values: 3157 3158 - 0 (disabled) 3159 - 1 (enabled) 3160 3161 Default: 0 (disabled) 3162 3163 3164use_optimistic - BOOLEAN 3165 If enabled, do not classify optimistic addresses as deprecated during 3166 source address selection. Preferred addresses will still be chosen 3167 before optimistic addresses, subject to other ranking in the source 3168 address selection algorithm. 3169 3170 This will be enabled if at least one of 3171 conf/{all,interface}/use_optimistic is set to 1, disabled otherwise. 3172 3173 Possible values: 3174 3175 - 0 (disabled) 3176 - 1 (enabled) 3177 3178 Default: 0 (disabled) 3179 3180stable_secret - IPv6 address 3181 This IPv6 address will be used as a secret to generate IPv6 3182 addresses for link-local addresses and autoconfigured 3183 ones. All addresses generated after setting this secret will 3184 be stable privacy ones by default. This can be changed via the 3185 addrgenmode ip-link. conf/default/stable_secret is used as the 3186 secret for the namespace, the interface specific ones can 3187 overwrite that. Writes to conf/all/stable_secret are refused. 3188 3189 It is recommended to generate this secret during installation 3190 of a system and keep it stable after that. 3191 3192 By default the stable secret is unset. 3193 3194addr_gen_mode - INTEGER 3195 Defines how link-local and autoconf addresses are generated. 3196 3197 = ================================================================= 3198 0 generate address based on EUI64 (default) 3199 1 do no generate a link-local address, use EUI64 for addresses 3200 generated from autoconf 3201 2 generate stable privacy addresses, using the secret from 3202 stable_secret (RFC7217) 3203 3 generate stable privacy addresses, using a random secret if unset 3204 = ================================================================= 3205 3206drop_unicast_in_l2_multicast - BOOLEAN 3207 Drop any unicast IPv6 packets that are received in link-layer 3208 multicast (or broadcast) frames. 3209 3210 Possible values: 3211 3212 - 0 (disabled) 3213 - 1 (enabled) 3214 3215 Default: 0 (disabled) 3216 3217drop_unsolicited_na - BOOLEAN 3218 Drop all unsolicited neighbor advertisements, for example if there's 3219 a known good NA proxy on the network and such frames need not be used 3220 (or in the case of 802.11, must not be used to prevent attacks.) 3221 3222 Possible values: 3223 3224 - 0 (disabled) 3225 - 1 (enabled) 3226 3227 Default: 0 (disabled). 3228 3229accept_untracked_na - INTEGER 3230 Define behavior for accepting neighbor advertisements for IPv6 addresses 3231 that are absent from the neighbor cache or whose entries are in FAILED 3232 state: 3233 3234 - 0 - (default) Do not create new neighbor cache entries or update 3235 FAILED entries from neighbor advertisements. 3236 3237 - 1 - For routers, add a new neighbor cache entry or update an existing 3238 FAILED entry to STALE upon receiving a neighbor advertisement (either 3239 solicited or unsolicited) with the target link-layer address option 3240 specified. Without this knob, NAs received for untracked addresses 3241 (absent from the neighbor cache or in FAILED state) are silently 3242 ignored. 3243 3244 This is as per router-side behavior documented in RFC9131. 3245 3246 This has lower precedence than drop_unsolicited_na. 3247 3248 This will optimize the return path for the initial off-link 3249 communication that is initiated by a directly connected host, by 3250 ensuring that the first-hop router which turns on this setting doesn't 3251 have to buffer the initial return packets to do neighbor-solicitation. 3252 The prerequisite is that the host is configured to send unsolicited 3253 neighbor advertisements on interface bringup. This setting should be 3254 used in conjunction with the ndisc_notify setting on the host to 3255 satisfy this prerequisite. 3256 3257 - 2 - Extend option (1) to add a new neighbor cache entry or update a 3258 FAILED entry only if the source IP address is in the same subnet as 3259 an address configured on the interface that received the neighbor 3260 advertisement. 3261 3262enhanced_dad - BOOLEAN 3263 Include a nonce option in the IPv6 neighbor solicitation messages used for 3264 duplicate address detection per RFC7527. A received DAD NS will only signal 3265 a duplicate address if the nonce is different. This avoids any false 3266 detection of duplicates due to loopback of the NS messages that we send. 3267 The nonce option will be sent on an interface unless both of 3268 conf/{all,interface}/enhanced_dad are set to FALSE. 3269 3270 Possible values: 3271 3272 - 0 (disabled) 3273 - 1 (enabled) 3274 3275 Default: 1 (enabled) 3276 3277``icmp/*``: 3278=========== 3279 3280ratelimit - INTEGER 3281 Limit the maximal rates for sending ICMPv6 messages to a particular 3282 peer. 3283 3284 0 to disable any limiting, 3285 otherwise the space between responses in milliseconds. 3286 3287 Default: 100 3288 3289ratemask - list of comma separated ranges 3290 For ICMPv6 message types matching the ranges in the ratemask, limit 3291 the sending of the message according to ratelimit parameter. 3292 3293 The format used for both input and output is a comma separated 3294 list of ranges (e.g. "0-127,129" for ICMPv6 message type 0 to 127 and 3295 129). Writing to the file will clear all previous ranges of ICMPv6 3296 message types and update the current list with the input. 3297 3298 Refer to: https://www.iana.org/assignments/icmpv6-parameters/icmpv6-parameters.xhtml 3299 for numerical values of ICMPv6 message types, e.g. echo request is 128 3300 and echo reply is 129. 3301 3302 Default: 0-1,3-127 (rate limit ICMPv6 errors except Packet Too Big) 3303 3304echo_ignore_all - BOOLEAN 3305 If enabled, then the kernel will ignore all ICMP ECHO 3306 requests sent to it over the IPv6 protocol. 3307 3308 Possible values: 3309 3310 - 0 (disabled) 3311 - 1 (enabled) 3312 3313 Default: 0 (disabled) 3314 3315echo_ignore_multicast - BOOLEAN 3316 If enabled, then the kernel will ignore all ICMP ECHO 3317 requests sent to it over the IPv6 protocol via multicast. 3318 3319 Possible values: 3320 3321 - 0 (disabled) 3322 - 1 (enabled) 3323 3324 Default: 0 (disabled) 3325 3326echo_ignore_anycast - BOOLEAN 3327 If enabled, then the kernel will ignore all ICMP ECHO 3328 requests sent to it over the IPv6 protocol destined to anycast address. 3329 3330 Possible values: 3331 3332 - 0 (disabled) 3333 - 1 (enabled) 3334 3335 Default: 0 (disabled) 3336 3337error_anycast_as_unicast - BOOLEAN 3338 If enabled, then the kernel will respond with ICMP Errors 3339 resulting from requests sent to it over the IPv6 protocol destined 3340 to anycast address essentially treating anycast as unicast. 3341 3342 Possible values: 3343 3344 - 0 (disabled) 3345 - 1 (enabled) 3346 3347 Default: 0 (disabled) 3348 3349errors_extension_mask - UNSIGNED INTEGER 3350 Bitmask of ICMP extensions to append to ICMPv6 error messages 3351 ("Destination Unreachable" and "Time Exceeded"). The original datagram 3352 is trimmed / padded to 128 bytes in order to be compatible with 3353 applications that do not comply with RFC 4884. 3354 3355 Possible extensions are: 3356 3357 ==== ============================================================== 3358 0x01 Incoming IP interface information according to RFC 5837. 3359 Extension will include the index, IPv6 address (if present), 3360 name and MTU of the IP interface that received the datagram 3361 which elicited the ICMP error. 3362 ==== ============================================================== 3363 3364 Default: 0x00 (no extensions) 3365 3366xfrm6_gc_thresh - INTEGER 3367 (Obsolete since linux-4.14) 3368 The threshold at which we will start garbage collecting for IPv6 3369 destination cache entries. At twice this value the system will 3370 refuse new allocations. 3371 3372 3373IPv6 Update by: 3374Pekka Savola <pekkas@netcore.fi> 3375YOSHIFUJI Hideaki / USAGI Project <yoshfuji@linux-ipv6.org> 3376 3377 3378/proc/sys/net/bridge/* Variables: 3379================================= 3380 3381bridge-nf-call-arptables - BOOLEAN 3382 3383 Possible values: 3384 3385 - 0 (disabled) - disable this. 3386 - 1 (enabled) - pass bridged ARP traffic to arptables' FORWARD chain. 3387 3388 Default: 1 (enabled) 3389 3390bridge-nf-call-iptables - BOOLEAN 3391 3392 Possible values: 3393 3394 - 0 (disabled) - disable this. 3395 - 1 (enabled) - pass bridged IPv4 traffic to iptables' chains. 3396 3397 Default: 1 (enabled) 3398 3399bridge-nf-call-ip6tables - BOOLEAN 3400 3401 Possible values: 3402 3403 - 0 (disabled) - disable this. 3404 - 1 (enabled) - pass bridged IPv6 traffic to ip6tables' chains. 3405 3406 Default: 1 (enabled) 3407 3408bridge-nf-filter-vlan-tagged - BOOLEAN 3409 3410 Possible values: 3411 3412 - 0 (disabled) - disable this. 3413 - 1 (enabled) - pass bridged vlan-tagged ARP/IP/IPv6 traffic to {arp,ip,ip6}tables 3414 3415 Default: 0 (disabled) 3416 3417bridge-nf-filter-pppoe-tagged - BOOLEAN 3418 3419 Possible values: 3420 3421 - 0 (disabled) - disable this. 3422 - 1 (enabled) - pass bridged pppoe-tagged IP/IPv6 traffic to {ip,ip6}tables. 3423 3424 Default: 0 (disabled) 3425 3426bridge-nf-pass-vlan-input-dev - BOOLEAN 3427 - 1: if bridge-nf-filter-vlan-tagged is enabled, try to find a vlan 3428 interface on the bridge and set the netfilter input device to the 3429 vlan. This allows use of e.g. "iptables -i br0.1" and makes the 3430 REDIRECT target work with vlan-on-top-of-bridge interfaces. When no 3431 matching vlan interface is found, or this switch is off, the input 3432 device is set to the bridge interface. 3433 3434 - 0: disable bridge netfilter vlan interface lookup. 3435 3436 Default: 0 3437 3438``proc/sys/net/sctp/*`` Variables: 3439================================== 3440 3441addip_enable - BOOLEAN 3442 Enable or disable extension of Dynamic Address Reconfiguration 3443 (ADD-IP) functionality specified in RFC5061. This extension provides 3444 the ability to dynamically add and remove new addresses for the SCTP 3445 associations. 3446 3447 Possible values: 3448 3449 - 0 (disabled) - disable extension. 3450 - 1 (enabled) - enable extension 3451 3452 Default: 0 (disabled) 3453 3454pf_enable - INTEGER 3455 Enable or disable pf (pf is short for potentially failed) state. A value 3456 of pf_retrans > path_max_retrans also disables pf state. That is, one of 3457 both pf_enable and pf_retrans > path_max_retrans can disable pf state. 3458 Since pf_retrans and path_max_retrans can be changed by userspace 3459 application, sometimes user expects to disable pf state by the value of 3460 pf_retrans > path_max_retrans, but occasionally the value of pf_retrans 3461 or path_max_retrans is changed by the user application, this pf state is 3462 enabled. As such, it is necessary to add this to dynamically enable 3463 and disable pf state. See: 3464 https://datatracker.ietf.org/doc/draft-ietf-tsvwg-sctp-failover for 3465 details. 3466 3467 Possible values: 3468 3469 - 1: Enable pf. 3470 - 0: Disable pf. 3471 3472 Default: 1 3473 3474pf_expose - INTEGER 3475 Unset or enable/disable pf (pf is short for potentially failed) state 3476 exposure. Applications can control the exposure of the PF path state 3477 in the SCTP_PEER_ADDR_CHANGE event and access of SCTP_PF-state 3478 transport info via SCTP_GET_PEER_ADDR_INFO sockopt. 3479 3480 Possible values: 3481 3482 - 0: Unset pf state exposure (compatible with old applications). No 3483 event will be sent but the transport info can be queried. 3484 - 1: Disable pf state exposure. No event will be sent and trying to 3485 obtain transport info will return -EACCESS. 3486 - 2: Enable pf state exposure. The event will be sent for a transport 3487 becoming SCTP_PF state and transport info can be obtained. 3488 3489 Default: 0 3490 3491addip_noauth_enable - BOOLEAN 3492 Dynamic Address Reconfiguration (ADD-IP) requires the use of 3493 authentication to protect the operations of adding or removing new 3494 addresses. This requirement is mandated so that unauthorized hosts 3495 would not be able to hijack associations. However, older 3496 implementations may not have implemented this requirement while 3497 allowing the ADD-IP extension. For reasons of interoperability, 3498 we provide this variable to control the enforcement of the 3499 authentication requirement. 3500 3501 == =============================================================== 3502 1 Allow ADD-IP extension to be used without authentication. This 3503 should only be set in a closed environment for interoperability 3504 with older implementations. 3505 3506 0 Enforce the authentication requirement 3507 == =============================================================== 3508 3509 Default: 0 3510 3511auth_enable - BOOLEAN 3512 Enable or disable Authenticated Chunks extension. This extension 3513 provides the ability to send and receive authenticated chunks and is 3514 required for secure operation of Dynamic Address Reconfiguration 3515 (ADD-IP) extension. 3516 3517 Possible values: 3518 3519 - 0 (disabled) - disable extension. 3520 - 1 (enabled) - enable extension 3521 3522 Default: 0 (disabled) 3523 3524prsctp_enable - BOOLEAN 3525 Enable or disable the Partial Reliability extension (RFC3758) which 3526 is used to notify peers that a given DATA should no longer be expected. 3527 3528 Possible values: 3529 3530 - 0 (disabled) - disable extension. 3531 - 1 (enabled) - enable extension 3532 3533 Default: 1 (enabled) 3534 3535max_burst - INTEGER 3536 The limit of the number of new packets that can be initially sent. It 3537 controls how bursty the generated traffic can be. 3538 3539 Default: 4 3540 3541association_max_retrans - INTEGER 3542 Set the maximum number for retransmissions that an association can 3543 attempt deciding that the remote end is unreachable. If this value 3544 is exceeded, the association is terminated. 3545 3546 Default: 10 3547 3548max_init_retransmits - INTEGER 3549 The maximum number of retransmissions of INIT and COOKIE-ECHO chunks 3550 that an association will attempt before declaring the destination 3551 unreachable and terminating. 3552 3553 Default: 8 3554 3555path_max_retrans - INTEGER 3556 The maximum number of retransmissions that will be attempted on a given 3557 path. Once this threshold is exceeded, the path is considered 3558 unreachable, and new traffic will use a different path when the 3559 association is multihomed. 3560 3561 Default: 5 3562 3563pf_retrans - INTEGER 3564 The number of retransmissions that will be attempted on a given path 3565 before traffic is redirected to an alternate transport (should one 3566 exist). Note this is distinct from path_max_retrans, as a path that 3567 passes the pf_retrans threshold can still be used. Its only 3568 deprioritized when a transmission path is selected by the stack. This 3569 setting is primarily used to enable fast failover mechanisms without 3570 having to reduce path_max_retrans to a very low value. See: 3571 http://www.ietf.org/id/draft-nishida-tsvwg-sctp-failover-05.txt 3572 for details. Note also that a value of pf_retrans > path_max_retrans 3573 disables this feature. Since both pf_retrans and path_max_retrans can 3574 be changed by userspace application, a variable pf_enable is used to 3575 disable pf state. 3576 3577 Default: 0 3578 3579ps_retrans - INTEGER 3580 Primary.Switchover.Max.Retrans (PSMR), it's a tunable parameter coming 3581 from section-5 "Primary Path Switchover" in rfc7829. The primary path 3582 will be changed to another active path when the path error counter on 3583 the old primary path exceeds PSMR, so that "the SCTP sender is allowed 3584 to continue data transmission on a new working path even when the old 3585 primary destination address becomes active again". Note this feature 3586 is disabled by initializing 'ps_retrans' per netns as 0xffff by default, 3587 and its value can't be less than 'pf_retrans' when changing by sysctl. 3588 3589 Default: 0xffff 3590 3591rto_initial - INTEGER 3592 The initial round trip timeout value in milliseconds that will be used 3593 in calculating round trip times. This is the initial time interval 3594 for retransmissions. 3595 3596 Default: 3000 3597 3598rto_max - INTEGER 3599 The maximum value (in milliseconds) of the round trip timeout. This 3600 is the largest time interval that can elapse between retransmissions. 3601 3602 Default: 60000 3603 3604rto_min - INTEGER 3605 The minimum value (in milliseconds) of the round trip timeout. This 3606 is the smallest time interval the can elapse between retransmissions. 3607 3608 Default: 1000 3609 3610hb_interval - INTEGER 3611 The interval (in milliseconds) between HEARTBEAT chunks. These chunks 3612 are sent at the specified interval on idle paths to probe the state of 3613 a given path between 2 associations. 3614 3615 Default: 30000 3616 3617sack_timeout - INTEGER 3618 The amount of time (in milliseconds) that the implementation will wait 3619 to send a SACK. 3620 3621 Default: 200 3622 3623valid_cookie_life - INTEGER 3624 The default lifetime of the SCTP cookie (in milliseconds). The cookie 3625 is used during association establishment. 3626 3627 Default: 60000 3628 3629cookie_preserve_enable - BOOLEAN 3630 Enable or disable the ability to extend the lifetime of the SCTP cookie 3631 that is used during the establishment phase of SCTP association 3632 3633 Possible values: 3634 3635 - 0 (disabled) - disable. 3636 - 1 (enabled) - enable cookie lifetime extension. 3637 3638 Default: 1 (enabled) 3639 3640cookie_hmac_alg - STRING 3641 Select the hmac algorithm used when generating the cookie value sent by 3642 a listening sctp socket to a connecting client in the INIT-ACK chunk. 3643 Valid values are: 3644 3645 * sha256 3646 * none 3647 3648 Default: sha256 3649 3650rcvbuf_policy - INTEGER 3651 Determines if the receive buffer is attributed to the socket or to 3652 association. SCTP supports the capability to create multiple 3653 associations on a single socket. When using this capability, it is 3654 possible that a single stalled association that's buffering a lot 3655 of data may block other associations from delivering their data by 3656 consuming all of the receive buffer space. To work around this, 3657 the rcvbuf_policy could be set to attribute the receiver buffer space 3658 to each association instead of the socket. This prevents the described 3659 blocking. 3660 3661 - 1: rcvbuf space is per association 3662 - 0: rcvbuf space is per socket 3663 3664 Default: 0 3665 3666sndbuf_policy - INTEGER 3667 Similar to rcvbuf_policy above, this applies to send buffer space. 3668 3669 - 1: Send buffer is tracked per association 3670 - 0: Send buffer is tracked per socket. 3671 3672 Default: 0 3673 3674sctp_mem - vector of 3 INTEGERs: min, pressure, max 3675 Number of pages allowed for queueing by all SCTP sockets. 3676 3677 * min: Below this number of pages SCTP is not bothered about its 3678 memory usage. When amount of memory allocated by SCTP exceeds 3679 this number, SCTP starts to moderate memory usage. 3680 * pressure: This value was introduced to follow format of tcp_mem. 3681 * max: Maximum number of allowed pages. 3682 3683 Default is calculated at boot time from amount of available memory. 3684 3685sctp_rmem - vector of 3 INTEGERs: min, default, max 3686 Only the first value ("min") is used, "default" and "max" are 3687 ignored. 3688 3689 * min: Minimal size of receive buffer used by SCTP socket. 3690 It is guaranteed to each SCTP socket (but not association) even 3691 under moderate memory pressure. 3692 3693 Default: 4K 3694 3695sctp_wmem - vector of 3 INTEGERs: min, default, max 3696 Only the first value ("min") is used, "default" and "max" are 3697 ignored. 3698 3699 * min: Minimum size of send buffer that can be used by SCTP sockets. 3700 It is guaranteed to each SCTP socket (but not association) even 3701 under moderate memory pressure. 3702 3703 Default: 4K 3704 3705addr_scope_policy - INTEGER 3706 Control IPv4 address scoping (see 3707 https://datatracker.ietf.org/doc/draft-stewart-tsvwg-sctp-ipv4/00/ 3708 for details). 3709 3710 - 0 - Disable IPv4 address scoping 3711 - 1 - Enable IPv4 address scoping 3712 - 2 - Follow draft but allow IPv4 private addresses 3713 - 3 - Follow draft but allow IPv4 link local addresses 3714 3715 Default: 1 3716 3717udp_port - INTEGER 3718 The listening port for the local UDP tunneling sock. Normally it's 3719 using the IANA-assigned UDP port number 9899 (sctp-tunneling). 3720 3721 This UDP sock is used for processing the incoming UDP-encapsulated 3722 SCTP packets (from RFC6951), and shared by all applications in the 3723 same net namespace. This UDP sock will be closed when the value is 3724 set to 0. 3725 3726 The value will also be used to set the src port of the UDP header 3727 for the outgoing UDP-encapsulated SCTP packets. For the dest port, 3728 please refer to 'encap_port' below. 3729 3730 Default: 0 3731 3732encap_port - INTEGER 3733 The default remote UDP encapsulation port. 3734 3735 This value is used to set the dest port of the UDP header for the 3736 outgoing UDP-encapsulated SCTP packets by default. Users can also 3737 change the value for each sock/asoc/transport by using setsockopt. 3738 For further information, please refer to RFC6951. 3739 3740 Note that when connecting to a remote server, the client should set 3741 this to the port that the UDP tunneling sock on the peer server is 3742 listening to and the local UDP tunneling sock on the client also 3743 must be started. On the server, it would get the encap_port from 3744 the incoming packet's source port. 3745 3746 Default: 0 3747 3748plpmtud_probe_interval - INTEGER 3749 The time interval (in milliseconds) for the PLPMTUD probe timer, 3750 which is configured to expire after this period to receive an 3751 acknowledgment to a probe packet. This is also the time interval 3752 between the probes for the current pmtu when the probe search 3753 is done. 3754 3755 PLPMTUD will be disabled when 0 is set, and other values for it 3756 must be >= 5000. 3757 3758 Default: 0 3759 3760reconf_enable - BOOLEAN 3761 Enable or disable extension of Stream Reconfiguration functionality 3762 specified in RFC6525. This extension provides the ability to "reset" 3763 a stream, and it includes the Parameters of "Outgoing/Incoming SSN 3764 Reset", "SSN/TSN Reset" and "Add Outgoing/Incoming Streams". 3765 3766 Possible values: 3767 3768 - 0 (disabled) - Disable extension. 3769 - 1 (enabled) - Enable extension. 3770 3771 Default: 0 (disabled) 3772 3773intl_enable - BOOLEAN 3774 Enable or disable extension of User Message Interleaving functionality 3775 specified in RFC8260. This extension allows the interleaving of user 3776 messages sent on different streams. With this feature enabled, I-DATA 3777 chunk will replace DATA chunk to carry user messages if also supported 3778 by the peer. Note that to use this feature, one needs to set this option 3779 to 1 and also needs to set socket options SCTP_FRAGMENT_INTERLEAVE to 2 3780 and SCTP_INTERLEAVING_SUPPORTED to 1. 3781 3782 Possible values: 3783 3784 - 0 (disabled) - Disable extension. 3785 - 1 (enabled) - Enable extension. 3786 3787 Default: 0 (disabled) 3788 3789ecn_enable - BOOLEAN 3790 Control use of Explicit Congestion Notification (ECN) by SCTP. 3791 Like in TCP, ECN is used only when both ends of the SCTP connection 3792 indicate support for it. This feature is useful in avoiding losses 3793 due to congestion by allowing supporting routers to signal congestion 3794 before having to drop packets. 3795 3796 Possible values: 3797 3798 - 0 (disabled) - Disable ecn. 3799 - 1 (enabled) - Enable ecn. 3800 3801 Default: 1 (enabled) 3802 3803l3mdev_accept - BOOLEAN 3804 Enabling this option allows a "global" bound socket to work 3805 across L3 master domains (e.g., VRFs) with packets capable of 3806 being received regardless of the L3 domain in which they 3807 originated. Only valid when the kernel was compiled with 3808 CONFIG_NET_L3_MASTER_DEV. 3809 3810 Possible values: 3811 3812 - 0 (disabled) 3813 - 1 (enabled) 3814 3815 Default: 1 (enabled) 3816 3817 3818``/proc/sys/net/core/*`` 3819======================== 3820 3821 Please see: Documentation/admin-guide/sysctl/net.rst for descriptions of these entries. 3822 3823 3824``/proc/sys/net/unix/*`` 3825======================== 3826 3827max_dgram_qlen - INTEGER 3828 The maximum length of dgram socket receive queue 3829 3830 Possible values: 0 - INT_MAX 3831 3832 Default: 10 3833