linux/Documentation/networking/ip-sysctl.txt
<<
>>
Prefs
   1/proc/sys/net/ipv4/* Variables:
   2
   3ip_forward - BOOLEAN
   4        0 - disabled (default)
   5        not 0 - enabled
   6
   7        Forward Packets between interfaces.
   8
   9        This variable is special, its change resets all configuration
  10        parameters to their default state (RFC1122 for hosts, RFC1812
  11        for routers)
  12
  13ip_default_ttl - INTEGER
  14        Default value of TTL field (Time To Live) for outgoing (but not
  15        forwarded) IP packets. Should be between 1 and 255 inclusive.
  16        Default: 64 (as recommended by RFC1700)
  17
  18ip_no_pmtu_disc - BOOLEAN
  19        Disable Path MTU Discovery.
  20        default FALSE
  21
  22min_pmtu - INTEGER
  23        default 552 - minimum discovered Path MTU
  24
  25route/max_size - INTEGER
  26        Maximum number of routes allowed in the kernel.  Increase
  27        this when using large numbers of interfaces and/or routes.
  28
  29neigh/default/gc_thresh3 - INTEGER
  30        Maximum number of neighbor entries allowed.  Increase this
  31        when using large numbers of interfaces and when communicating
  32        with large numbers of directly-connected peers.
  33
  34neigh/default/unres_qlen_bytes - INTEGER
  35        The maximum number of bytes which may be used by packets
  36        queued for each unresolved address by other network layers.
  37        (added in linux 3.3)
  38
  39neigh/default/unres_qlen - INTEGER
  40        The maximum number of packets which may be queued for each
  41        unresolved address by other network layers.
  42        (deprecated in linux 3.3) : use unres_qlen_bytes instead.
  43
  44mtu_expires - INTEGER
  45        Time, in seconds, that cached PMTU information is kept.
  46
  47min_adv_mss - INTEGER
  48        The advertised MSS depends on the first hop route MTU, but will
  49        never be lower than this setting.
  50
  51IP Fragmentation:
  52
  53ipfrag_high_thresh - INTEGER
  54        Maximum memory used to reassemble IP fragments. When
  55        ipfrag_high_thresh bytes of memory is allocated for this purpose,
  56        the fragment handler will toss packets until ipfrag_low_thresh
  57        is reached.
  58
  59ipfrag_low_thresh - INTEGER
  60        See ipfrag_high_thresh
  61
  62ipfrag_time - INTEGER
  63        Time in seconds to keep an IP fragment in memory.
  64
  65ipfrag_secret_interval - INTEGER
  66        Regeneration interval (in seconds) of the hash secret (or lifetime
  67        for the hash secret) for IP fragments.
  68        Default: 600
  69
  70ipfrag_max_dist - INTEGER
  71        ipfrag_max_dist is a non-negative integer value which defines the
  72        maximum "disorder" which is allowed among fragments which share a
  73        common IP source address. Note that reordering of packets is
  74        not unusual, but if a large number of fragments arrive from a source
  75        IP address while a particular fragment queue remains incomplete, it
  76        probably indicates that one or more fragments belonging to that queue
  77        have been lost. When ipfrag_max_dist is positive, an additional check
  78        is done on fragments before they are added to a reassembly queue - if
  79        ipfrag_max_dist (or more) fragments have arrived from a particular IP
  80        address between additions to any IP fragment queue using that source
  81        address, it's presumed that one or more fragments in the queue are
  82        lost. The existing fragment queue will be dropped, and a new one
  83        started. An ipfrag_max_dist value of zero disables this check.
  84
  85        Using a very small value, e.g. 1 or 2, for ipfrag_max_dist can
  86        result in unnecessarily dropping fragment queues when normal
  87        reordering of packets occurs, which could lead to poor application
  88        performance. Using a very large value, e.g. 50000, increases the
  89        likelihood of incorrectly reassembling IP fragments that originate
  90        from different IP datagrams, which could result in data corruption.
  91        Default: 64
  92
  93INET peer storage:
  94
  95inet_peer_threshold - INTEGER
  96        The approximate size of the storage.  Starting from this threshold
  97        entries will be thrown aggressively.  This threshold also determines
  98        entries' time-to-live and time intervals between garbage collection
  99        passes.  More entries, less time-to-live, less GC interval.
 100
 101inet_peer_minttl - INTEGER
 102        Minimum time-to-live of entries.  Should be enough to cover fragment
 103        time-to-live on the reassembling side.  This minimum time-to-live  is
 104        guaranteed if the pool size is less than inet_peer_threshold.
 105        Measured in seconds.
 106
 107inet_peer_maxttl - INTEGER
 108        Maximum time-to-live of entries.  Unused entries will expire after
 109        this period of time if there is no memory pressure on the pool (i.e.
 110        when the number of entries in the pool is very small).
 111        Measured in seconds.
 112
 113TCP variables:
 114
 115somaxconn - INTEGER
 116        Limit of socket listen() backlog, known in userspace as SOMAXCONN.
 117        Defaults to 128.  See also tcp_max_syn_backlog for additional tuning
 118        for TCP sockets.
 119
 120tcp_abc - INTEGER
 121        Controls Appropriate Byte Count (ABC) defined in RFC3465.
 122        ABC is a way of increasing congestion window (cwnd) more slowly
 123        in response to partial acknowledgments.
 124        Possible values are:
 125                0 increase cwnd once per acknowledgment (no ABC)
 126                1 increase cwnd once per acknowledgment of full sized segment
 127                2 allow increase cwnd by two if acknowledgment is
 128                  of two segments to compensate for delayed acknowledgments.
 129        Default: 0 (off)
 130
 131tcp_abort_on_overflow - BOOLEAN
 132        If listening service is too slow to accept new connections,
 133        reset them. Default state is FALSE. It means that if overflow
 134        occurred due to a burst, connection will recover. Enable this
 135        option _only_ if you are really sure that listening daemon
 136        cannot be tuned to accept connections faster. Enabling this
 137        option can harm clients of your server.
 138
 139tcp_adv_win_scale - INTEGER
 140        Count buffering overhead as bytes/2^tcp_adv_win_scale
 141        (if tcp_adv_win_scale > 0) or bytes-bytes/2^(-tcp_adv_win_scale),
 142        if it is <= 0.
 143        Possible values are [-31, 31], inclusive.
 144        Default: 1
 145
 146tcp_allowed_congestion_control - STRING
 147        Show/set the congestion control choices available to non-privileged
 148        processes. The list is a subset of those listed in
 149        tcp_available_congestion_control.
 150        Default is "reno" and the default setting (tcp_congestion_control).
 151
 152tcp_app_win - INTEGER
 153        Reserve max(window/2^tcp_app_win, mss) of window for application
 154        buffer. Value 0 is special, it means that nothing is reserved.
 155        Default: 31
 156
 157tcp_available_congestion_control - STRING
 158        Shows the available congestion control choices that are registered.
 159        More congestion control algorithms may be available as modules,
 160        but not loaded.
 161
 162tcp_base_mss - INTEGER
 163        The initial value of search_low to be used by the packetization layer
 164        Path MTU discovery (MTU probing).  If MTU probing is enabled,
 165        this is the initial MSS used by the connection.
 166
 167tcp_congestion_control - STRING
 168        Set the congestion control algorithm to be used for new
 169        connections. The algorithm "reno" is always available, but
 170        additional choices may be available based on kernel configuration.
 171        Default is set as part of kernel configuration.
 172        For passive connections, the listener congestion control choice
 173        is inherited.
 174        [see setsockopt(listenfd, SOL_TCP, TCP_CONGESTION, "name" ...) ]
 175
 176tcp_cookie_size - INTEGER
 177        Default size of TCP Cookie Transactions (TCPCT) option, that may be
 178        overridden on a per socket basis by the TCPCT socket option.
 179        Values greater than the maximum (16) are interpreted as the maximum.
 180        Values greater than zero and less than the minimum (8) are interpreted
 181        as the minimum.  Odd values are interpreted as the next even value.
 182        Default: 0 (off).
 183
 184tcp_dsack - BOOLEAN
 185        Allows TCP to send "duplicate" SACKs.
 186
 187tcp_early_retrans - INTEGER
 188        Enable Early Retransmit (ER), per RFC 5827. ER lowers the threshold
 189        for triggering fast retransmit when the amount of outstanding data is
 190        small and when no previously unsent data can be transmitted (such
 191        that limited transmit could be used).
 192        Possible values:
 193                0 disables ER
 194                1 enables ER
 195                2 enables ER but delays fast recovery and fast retransmit
 196                  by a fourth of RTT. This mitigates connection falsely
 197                  recovers when network has a small degree of reordering
 198                  (less than 3 packets).
 199        Default: 2
 200
 201tcp_ecn - INTEGER
 202        Enable Explicit Congestion Notification (ECN) in TCP. ECN is only
 203        used when both ends of the TCP flow support it. It is useful to
 204        avoid losses due to congestion (when the bottleneck router supports
 205        ECN).
 206        Possible values are:
 207                0 disable ECN
 208                1 ECN enabled
 209                2 Only server-side ECN enabled. If the other end does
 210                  not support ECN, behavior is like with ECN disabled.
 211        Default: 2
 212
 213tcp_fack - BOOLEAN
 214        Enable FACK congestion avoidance and fast retransmission.
 215        The value is not used, if tcp_sack is not enabled.
 216
 217tcp_fin_timeout - INTEGER
 218        Time to hold socket in state FIN-WAIT-2, if it was closed
 219        by our side. Peer can be broken and never close its side,
 220        or even died unexpectedly. Default value is 60sec.
 221        Usual value used in 2.2 was 180 seconds, you may restore
 222        it, but remember that if your machine is even underloaded WEB server,
 223        you risk to overflow memory with kilotons of dead sockets,
 224        FIN-WAIT-2 sockets are less dangerous than FIN-WAIT-1,
 225        because they eat maximum 1.5K of memory, but they tend
 226        to live longer. Cf. tcp_max_orphans.
 227
 228tcp_frto - INTEGER
 229        Enables Forward RTO-Recovery (F-RTO) defined in RFC4138.
 230        F-RTO is an enhanced recovery algorithm for TCP retransmission
 231        timeouts.  It is particularly beneficial in wireless environments
 232        where packet loss is typically due to random radio interference
 233        rather than intermediate router congestion.  F-RTO is sender-side
 234        only modification. Therefore it does not require any support from
 235        the peer.
 236
 237        If set to 1, basic version is enabled.  2 enables SACK enhanced
 238        F-RTO if flow uses SACK.  The basic version can be used also when
 239        SACK is in use though scenario(s) with it exists where F-RTO
 240        interacts badly with the packet counting of the SACK enabled TCP
 241        flow.
 242
 243tcp_frto_response - INTEGER
 244        When F-RTO has detected that a TCP retransmission timeout was
 245        spurious (i.e, the timeout would have been avoided had TCP set a
 246        longer retransmission timeout), TCP has several options what to do
 247        next. Possible values are:
 248                0 Rate halving based; a smooth and conservative response,
 249                  results in halved cwnd and ssthresh after one RTT
 250                1 Very conservative response; not recommended because even
 251                  though being valid, it interacts poorly with the rest of
 252                  Linux TCP, halves cwnd and ssthresh immediately
 253                2 Aggressive response; undoes congestion control measures
 254                  that are now known to be unnecessary (ignoring the
 255                  possibility of a lost retransmission that would require
 256                  TCP to be more cautious), cwnd and ssthresh are restored
 257                  to the values prior timeout
 258        Default: 0 (rate halving based)
 259
 260tcp_keepalive_time - INTEGER
 261        How often TCP sends out keepalive messages when keepalive is enabled.
 262        Default: 2hours.
 263
 264tcp_keepalive_probes - INTEGER
 265        How many keepalive probes TCP sends out, until it decides that the
 266        connection is broken. Default value: 9.
 267
 268tcp_keepalive_intvl - INTEGER
 269        How frequently the probes are send out. Multiplied by
 270        tcp_keepalive_probes it is time to kill not responding connection,
 271        after probes started. Default value: 75sec i.e. connection
 272        will be aborted after ~11 minutes of retries.
 273
 274tcp_low_latency - BOOLEAN
 275        If set, the TCP stack makes decisions that prefer lower
 276        latency as opposed to higher throughput.  By default, this
 277        option is not set meaning that higher throughput is preferred.
 278        An example of an application where this default should be
 279        changed would be a Beowulf compute cluster.
 280        Default: 0
 281
 282tcp_max_orphans - INTEGER
 283        Maximal number of TCP sockets not attached to any user file handle,
 284        held by system. If this number is exceeded orphaned connections are
 285        reset immediately and warning is printed. This limit exists
 286        only to prevent simple DoS attacks, you _must_ not rely on this
 287        or lower the limit artificially, but rather increase it
 288        (probably, after increasing installed memory),
 289        if network conditions require more than default value,
 290        and tune network services to linger and kill such states
 291        more aggressively. Let me to remind again: each orphan eats
 292        up to ~64K of unswappable memory.
 293
 294tcp_max_ssthresh - INTEGER
 295        Limited Slow-Start for TCP with large congestion windows (cwnd) defined in
 296        RFC3742. Limited slow-start is a mechanism to limit growth of the cwnd
 297        on the region where cwnd is larger than tcp_max_ssthresh. TCP increases cwnd
 298        by at most tcp_max_ssthresh segments, and by at least tcp_max_ssthresh/2
 299        segments per RTT when the cwnd is above tcp_max_ssthresh.
 300        If TCP connection increased cwnd to thousands (or tens of thousands) segments,
 301        and thousands of packets were being dropped during slow-start, you can set
 302        tcp_max_ssthresh to improve performance for new TCP connection.
 303        Default: 0 (off)
 304
 305tcp_max_syn_backlog - INTEGER
 306        Maximal number of remembered connection requests, which have not
 307        received an acknowledgment from connecting client.
 308        The minimal value is 128 for low memory machines, and it will
 309        increase in proportion to the memory of machine.
 310        If server suffers from overload, try increasing this number.
 311
 312tcp_max_tw_buckets - INTEGER
 313        Maximal number of timewait sockets held by system simultaneously.
 314        If this number is exceeded time-wait socket is immediately destroyed
 315        and warning is printed. This limit exists only to prevent
 316        simple DoS attacks, you _must_ not lower the limit artificially,
 317        but rather increase it (probably, after increasing installed memory),
 318        if network conditions require more than default value.
 319
 320tcp_mem - vector of 3 INTEGERs: min, pressure, max
 321        min: below this number of pages TCP is not bothered about its
 322        memory appetite.
 323
 324        pressure: when amount of memory allocated by TCP exceeds this number
 325        of pages, TCP moderates its memory consumption and enters memory
 326        pressure mode, which is exited when memory consumption falls
 327        under "min".
 328
 329        max: number of pages allowed for queueing by all TCP sockets.
 330
 331        Defaults are calculated at boot time from amount of available
 332        memory.
 333
 334tcp_moderate_rcvbuf - BOOLEAN
 335        If set, TCP performs receive buffer auto-tuning, attempting to
 336        automatically size the buffer (no greater than tcp_rmem[2]) to
 337        match the size required by the path for full throughput.  Enabled by
 338        default.
 339
 340tcp_mtu_probing - INTEGER
 341        Controls TCP Packetization-Layer Path MTU Discovery.  Takes three
 342        values:
 343          0 - Disabled
 344          1 - Disabled by default, enabled when an ICMP black hole detected
 345          2 - Always enabled, use initial MSS of tcp_base_mss.
 346
 347tcp_no_metrics_save - BOOLEAN
 348        By default, TCP saves various connection metrics in the route cache
 349        when the connection closes, so that connections established in the
 350        near future can use these to set initial conditions.  Usually, this
 351        increases overall performance, but may sometimes cause performance
 352        degradation.  If set, TCP will not cache metrics on closing
 353        connections.
 354
 355tcp_orphan_retries - INTEGER
 356        This value influences the timeout of a locally closed TCP connection,
 357        when RTO retransmissions remain unacknowledged.
 358        See tcp_retries2 for more details.
 359
 360        The default value is 8.
 361        If your machine is a loaded WEB server,
 362        you should think about lowering this value, such sockets
 363        may consume significant resources. Cf. tcp_max_orphans.
 364
 365tcp_reordering - INTEGER
 366        Maximal reordering of packets in a TCP stream.
 367        Default: 3
 368
 369tcp_retrans_collapse - BOOLEAN
 370        Bug-to-bug compatibility with some broken printers.
 371        On retransmit try to send bigger packets to work around bugs in
 372        certain TCP stacks.
 373
 374tcp_retries1 - INTEGER
 375        This value influences the time, after which TCP decides, that
 376        something is wrong due to unacknowledged RTO retransmissions,
 377        and reports this suspicion to the network layer.
 378        See tcp_retries2 for more details.
 379
 380        RFC 1122 recommends at least 3 retransmissions, which is the
 381        default.
 382
 383tcp_retries2 - INTEGER
 384        This value influences the timeout of an alive TCP connection,
 385        when RTO retransmissions remain unacknowledged.
 386        Given a value of N, a hypothetical TCP connection following
 387        exponential backoff with an initial RTO of TCP_RTO_MIN would
 388        retransmit N times before killing the connection at the (N+1)th RTO.
 389
 390        The default value of 15 yields a hypothetical timeout of 924.6
 391        seconds and is a lower bound for the effective timeout.
 392        TCP will effectively time out at the first RTO which exceeds the
 393        hypothetical timeout.
 394
 395        RFC 1122 recommends at least 100 seconds for the timeout,
 396        which corresponds to a value of at least 8.
 397
 398tcp_rfc1337 - BOOLEAN
 399        If set, the TCP stack behaves conforming to RFC1337. If unset,
 400        we are not conforming to RFC, but prevent TCP TIME_WAIT
 401        assassination.
 402        Default: 0
 403
 404tcp_rmem - vector of 3 INTEGERs: min, default, max
 405        min: Minimal size of receive buffer used by TCP sockets.
 406        It is guaranteed to each TCP socket, even under moderate memory
 407        pressure.
 408        Default: 1 page
 409
 410        default: initial size of receive buffer used by TCP sockets.
 411        This value overrides net.core.rmem_default used by other protocols.
 412        Default: 87380 bytes. This value results in window of 65535 with
 413        default setting of tcp_adv_win_scale and tcp_app_win:0 and a bit
 414        less for default tcp_app_win. See below about these variables.
 415
 416        max: maximal size of receive buffer allowed for automatically
 417        selected receiver buffers for TCP socket. This value does not override
 418        net.core.rmem_max.  Calling setsockopt() with SO_RCVBUF disables
 419        automatic tuning of that socket's receive buffer size, in which
 420        case this value is ignored.
 421        Default: between 87380B and 6MB, depending on RAM size.
 422
 423tcp_sack - BOOLEAN
 424        Enable select acknowledgments (SACKS).
 425
 426tcp_slow_start_after_idle - BOOLEAN
 427        If set, provide RFC2861 behavior and time out the congestion
 428        window after an idle period.  An idle period is defined at
 429        the current RTO.  If unset, the congestion window will not
 430        be timed out after an idle period.
 431        Default: 1
 432
 433tcp_stdurg - BOOLEAN
 434        Use the Host requirements interpretation of the TCP urgent pointer field.
 435        Most hosts use the older BSD interpretation, so if you turn this on
 436        Linux might not communicate correctly with them.
 437        Default: FALSE
 438
 439tcp_synack_retries - INTEGER
 440        Number of times SYNACKs for a passive TCP connection attempt will
 441        be retransmitted. Should not be higher than 255. Default value
 442        is 5, which corresponds to ~180seconds.
 443
 444tcp_syncookies - BOOLEAN
 445        Only valid when the kernel was compiled with CONFIG_SYNCOOKIES
 446        Send out syncookies when the syn backlog queue of a socket
 447        overflows. This is to prevent against the common 'SYN flood attack'
 448        Default: FALSE
 449
 450        Note, that syncookies is fallback facility.
 451        It MUST NOT be used to help highly loaded servers to stand
 452        against legal connection rate. If you see SYN flood warnings
 453        in your logs, but investigation shows that they occur
 454        because of overload with legal connections, you should tune
 455        another parameters until this warning disappear.
 456        See: tcp_max_syn_backlog, tcp_synack_retries, tcp_abort_on_overflow.
 457
 458        syncookies seriously violate TCP protocol, do not allow
 459        to use TCP extensions, can result in serious degradation
 460        of some services (f.e. SMTP relaying), visible not by you,
 461        but your clients and relays, contacting you. While you see
 462        SYN flood warnings in logs not being really flooded, your server
 463        is seriously misconfigured.
 464
 465tcp_fastopen - INTEGER
 466        Enable TCP Fast Open feature (draft-ietf-tcpm-fastopen) to send data
 467        in the opening SYN packet. To use this feature, the client application
 468        must not use connect(). Instead, it should use sendmsg() or sendto()
 469        with MSG_FASTOPEN flag which performs a TCP handshake automatically.
 470
 471        The values (bitmap) are:
 472        1: Enables sending data in the opening SYN on the client
 473        5: Enables sending data in the opening SYN on the client regardless
 474           of cookie availability.
 475
 476        Default: 0
 477
 478tcp_syn_retries - INTEGER
 479        Number of times initial SYNs for an active TCP connection attempt
 480        will be retransmitted. Should not be higher than 255. Default value
 481        is 5, which corresponds to ~180seconds.
 482
 483tcp_timestamps - BOOLEAN
 484        Enable timestamps as defined in RFC1323.
 485
 486tcp_tso_win_divisor - INTEGER
 487        This allows control over what percentage of the congestion window
 488        can be consumed by a single TSO frame.
 489        The setting of this parameter is a choice between burstiness and
 490        building larger TSO frames.
 491        Default: 3
 492
 493tcp_tw_recycle - BOOLEAN
 494        Enable fast recycling TIME-WAIT sockets. Default value is 0.
 495        It should not be changed without advice/request of technical
 496        experts.
 497
 498tcp_tw_reuse - BOOLEAN
 499        Allow to reuse TIME-WAIT sockets for new connections when it is
 500        safe from protocol viewpoint. Default value is 0.
 501        It should not be changed without advice/request of technical
 502        experts.
 503
 504tcp_window_scaling - BOOLEAN
 505        Enable window scaling as defined in RFC1323.
 506
 507tcp_wmem - vector of 3 INTEGERs: min, default, max
 508        min: Amount of memory reserved for send buffers for TCP sockets.
 509        Each TCP socket has rights to use it due to fact of its birth.
 510        Default: 1 page
 511
 512        default: initial size of send buffer used by TCP sockets.  This
 513        value overrides net.core.wmem_default used by other protocols.
 514        It is usually lower than net.core.wmem_default.
 515        Default: 16K
 516
 517        max: Maximal amount of memory allowed for automatically tuned
 518        send buffers for TCP sockets. This value does not override
 519        net.core.wmem_max.  Calling setsockopt() with SO_SNDBUF disables
 520        automatic tuning of that socket's send buffer size, in which case
 521        this value is ignored.
 522        Default: between 64K and 4MB, depending on RAM size.
 523
 524tcp_workaround_signed_windows - BOOLEAN
 525        If set, assume no receipt of a window scaling option means the
 526        remote TCP is broken and treats the window as a signed quantity.
 527        If unset, assume the remote TCP is not broken even if we do
 528        not receive a window scaling option from them.
 529        Default: 0
 530
 531tcp_dma_copybreak - INTEGER
 532        Lower limit, in bytes, of the size of socket reads that will be
 533        offloaded to a DMA copy engine, if one is present in the system
 534        and CONFIG_NET_DMA is enabled.
 535        Default: 4096
 536
 537tcp_thin_linear_timeouts - BOOLEAN
 538        Enable dynamic triggering of linear timeouts for thin streams.
 539        If set, a check is performed upon retransmission by timeout to
 540        determine if the stream is thin (less than 4 packets in flight).
 541        As long as the stream is found to be thin, up to 6 linear
 542        timeouts may be performed before exponential backoff mode is
 543        initiated. This improves retransmission latency for
 544        non-aggressive thin streams, often found to be time-dependent.
 545        For more information on thin streams, see
 546        Documentation/networking/tcp-thin.txt
 547        Default: 0
 548
 549tcp_thin_dupack - BOOLEAN
 550        Enable dynamic triggering of retransmissions after one dupACK
 551        for thin streams. If set, a check is performed upon reception
 552        of a dupACK to determine if the stream is thin (less than 4
 553        packets in flight). As long as the stream is found to be thin,
 554        data is retransmitted on the first received dupACK. This
 555        improves retransmission latency for non-aggressive thin
 556        streams, often found to be time-dependent.
 557        For more information on thin streams, see
 558        Documentation/networking/tcp-thin.txt
 559        Default: 0
 560
 561tcp_limit_output_bytes - INTEGER
 562        Controls TCP Small Queue limit per tcp socket.
 563        TCP bulk sender tends to increase packets in flight until it
 564        gets losses notifications. With SNDBUF autotuning, this can
 565        result in a large amount of packets queued in qdisc/device
 566        on the local machine, hurting latency of other flows, for
 567        typical pfifo_fast qdiscs.
 568        tcp_limit_output_bytes limits the number of bytes on qdisc
 569        or device to reduce artificial RTT/cwnd and reduce bufferbloat.
 570        Note: For GSO/TSO enabled flows, we try to have at least two
 571        packets in flight. Reducing tcp_limit_output_bytes might also
 572        reduce the size of individual GSO packet (64KB being the max)
 573        Default: 131072
 574
 575tcp_challenge_ack_limit - INTEGER
 576        Limits number of Challenge ACK sent per second, as recommended
 577        in RFC 5961 (Improving TCP's Robustness to Blind In-Window Attacks)
 578        Default: 100
 579
 580UDP variables:
 581
 582udp_mem - vector of 3 INTEGERs: min, pressure, max
 583        Number of pages allowed for queueing by all UDP sockets.
 584
 585        min: Below this number of pages UDP is not bothered about its
 586        memory appetite. When amount of memory allocated by UDP exceeds
 587        this number, UDP starts to moderate memory usage.
 588
 589        pressure: This value was introduced to follow format of tcp_mem.
 590
 591        max: Number of pages allowed for queueing by all UDP sockets.
 592
 593        Default is calculated at boot time from amount of available memory.
 594
 595udp_rmem_min - INTEGER
 596        Minimal size of receive buffer used by UDP sockets in moderation.
 597        Each UDP socket is able to use the size for receiving data, even if
 598        total pages of UDP sockets exceed udp_mem pressure. The unit is byte.
 599        Default: 1 page
 600
 601udp_wmem_min - INTEGER
 602        Minimal size of send buffer used by UDP sockets in moderation.
 603        Each UDP socket is able to use the size for sending data, even if
 604        total pages of UDP sockets exceed udp_mem pressure. The unit is byte.
 605        Default: 1 page
 606
 607CIPSOv4 Variables:
 608
 609cipso_cache_enable - BOOLEAN
 610        If set, enable additions to and lookups from the CIPSO label mapping
 611        cache.  If unset, additions are ignored and lookups always result in a
 612        miss.  However, regardless of the setting the cache is still
 613        invalidated when required when means you can safely toggle this on and
 614        off and the cache will always be "safe".
 615        Default: 1
 616
 617cipso_cache_bucket_size - INTEGER
 618        The CIPSO label cache consists of a fixed size hash table with each
 619        hash bucket containing a number of cache entries.  This variable limits
 620        the number of entries in each hash bucket; the larger the value the
 621        more CIPSO label mappings that can be cached.  When the number of
 622        entries in a given hash bucket reaches this limit adding new entries
 623        causes the oldest entry in the bucket to be removed to make room.
 624        Default: 10
 625
 626cipso_rbm_optfmt - BOOLEAN
 627        Enable the "Optimized Tag 1 Format" as defined in section 3.4.2.6 of
 628        the CIPSO draft specification (see Documentation/netlabel for details).
 629        This means that when set the CIPSO tag will be padded with empty
 630        categories in order to make the packet data 32-bit aligned.
 631        Default: 0
 632
 633cipso_rbm_structvalid - BOOLEAN
 634        If set, do a very strict check of the CIPSO option when
 635        ip_options_compile() is called.  If unset, relax the checks done during
 636        ip_options_compile().  Either way is "safe" as errors are caught else
 637        where in the CIPSO processing code but setting this to 0 (False) should
 638        result in less work (i.e. it should be faster) but could cause problems
 639        with other implementations that require strict checking.
 640        Default: 0
 641
 642IP Variables:
 643
 644ip_local_port_range - 2 INTEGERS
 645        Defines the local port range that is used by TCP and UDP to
 646        choose the local port. The first number is the first, the
 647        second the last local port number. The default values are
 648        32768 and 61000 respectively.
 649
 650ip_local_reserved_ports - list of comma separated ranges
 651        Specify the ports which are reserved for known third-party
 652        applications. These ports will not be used by automatic port
 653        assignments (e.g. when calling connect() or bind() with port
 654        number 0). Explicit port allocation behavior is unchanged.
 655
 656        The format used for both input and output is a comma separated
 657        list of ranges (e.g. "1,2-4,10-10" for ports 1, 2, 3, 4 and
 658        10). Writing to the file will clear all previously reserved
 659        ports and update the current list with the one given in the
 660        input.
 661
 662        Note that ip_local_port_range and ip_local_reserved_ports
 663        settings are independent and both are considered by the kernel
 664        when determining which ports are available for automatic port
 665        assignments.
 666
 667        You can reserve ports which are not in the current
 668        ip_local_port_range, e.g.:
 669
 670        $ cat /proc/sys/net/ipv4/ip_local_port_range
 671        32000   61000
 672        $ cat /proc/sys/net/ipv4/ip_local_reserved_ports
 673        8080,9148
 674
 675        although this is redundant. However such a setting is useful
 676        if later the port range is changed to a value that will
 677        include the reserved ports.
 678
 679        Default: Empty
 680
 681ip_nonlocal_bind - BOOLEAN
 682        If set, allows processes to bind() to non-local IP addresses,
 683        which can be quite useful - but may break some applications.
 684        Default: 0
 685
 686ip_dynaddr - BOOLEAN
 687        If set non-zero, enables support for dynamic addresses.
 688        If set to a non-zero value larger than 1, a kernel log
 689        message will be printed when dynamic address rewriting
 690        occurs.
 691        Default: 0
 692
 693icmp_echo_ignore_all - BOOLEAN
 694        If set non-zero, then the kernel will ignore all ICMP ECHO
 695        requests sent to it.
 696        Default: 0
 697
 698icmp_echo_ignore_broadcasts - BOOLEAN
 699        If set non-zero, then the kernel will ignore all ICMP ECHO and
 700        TIMESTAMP requests sent to it via broadcast/multicast.
 701        Default: 1
 702
 703icmp_ratelimit - INTEGER
 704        Limit the maximal rates for sending ICMP packets whose type matches
 705        icmp_ratemask (see below) to specific targets.
 706        0 to disable any limiting,
 707        otherwise the minimal space between responses in milliseconds.
 708        Default: 1000
 709
 710icmp_ratemask - INTEGER
 711        Mask made of ICMP types for which rates are being limited.
 712        Significant bits: IHGFEDCBA9876543210
 713        Default mask:     0000001100000011000 (6168)
 714
 715        Bit definitions (see include/linux/icmp.h):
 716                0 Echo Reply
 717                3 Destination Unreachable *
 718                4 Source Quench *
 719                5 Redirect
 720                8 Echo Request
 721                B Time Exceeded *
 722                C Parameter Problem *
 723                D Timestamp Request
 724                E Timestamp Reply
 725                F Info Request
 726                G Info Reply
 727                H Address Mask Request
 728                I Address Mask Reply
 729
 730        * These are rate limited by default (see default mask above)
 731
 732icmp_ignore_bogus_error_responses - BOOLEAN
 733        Some routers violate RFC1122 by sending bogus responses to broadcast
 734        frames.  Such violations are normally logged via a kernel warning.
 735        If this is set to TRUE, the kernel will not give such warnings, which
 736        will avoid log file clutter.
 737        Default: FALSE
 738
 739icmp_errors_use_inbound_ifaddr - BOOLEAN
 740
 741        If zero, icmp error messages are sent with the primary address of
 742        the exiting interface.
 743
 744        If non-zero, the message will be sent with the primary address of
 745        the interface that received the packet that caused the icmp error.
 746        This is the behaviour network many administrators will expect from
 747        a router. And it can make debugging complicated network layouts
 748        much easier.
 749
 750        Note that if no primary address exists for the interface selected,
 751        then the primary address of the first non-loopback interface that
 752        has one will be used regardless of this setting.
 753
 754        Default: 0
 755
 756igmp_max_memberships - INTEGER
 757        Change the maximum number of multicast groups we can subscribe to.
 758        Default: 20
 759
 760        Theoretical maximum value is bounded by having to send a membership
 761        report in a single datagram (i.e. the report can't span multiple
 762        datagrams, or risk confusing the switch and leaving groups you don't
 763        intend to).
 764
 765        The number of supported groups 'M' is bounded by the number of group
 766        report entries you can fit into a single datagram of 65535 bytes.
 767
 768        M = 65536-sizeof (ip header)/(sizeof(Group record))
 769
 770        Group records are variable length, with a minimum of 12 bytes.
 771        So net.ipv4.igmp_max_memberships should not be set higher than:
 772
 773        (65536-24) / 12 = 5459
 774
 775        The value 5459 assumes no IP header options, so in practice
 776        this number may be lower.
 777
 778        conf/interface/*  changes special settings per interface (where
 779        "interface" is the name of your network interface)
 780
 781        conf/all/*        is special, changes the settings for all interfaces
 782
 783log_martians - BOOLEAN
 784        Log packets with impossible addresses to kernel log.
 785        log_martians for the interface will be enabled if at least one of
 786        conf/{all,interface}/log_martians is set to TRUE,
 787        it will be disabled otherwise
 788
 789accept_redirects - BOOLEAN
 790        Accept ICMP redirect messages.
 791        accept_redirects for the interface will be enabled if:
 792        - both conf/{all,interface}/accept_redirects are TRUE in the case
 793          forwarding for the interface is enabled
 794        or
 795        - at least one of conf/{all,interface}/accept_redirects is TRUE in the
 796          case forwarding for the interface is disabled
 797        accept_redirects for the interface will be disabled otherwise
 798        default TRUE (host)
 799                FALSE (router)
 800
 801forwarding - BOOLEAN
 802        Enable IP forwarding on this interface.
 803
 804mc_forwarding - BOOLEAN
 805        Do multicast routing. The kernel needs to be compiled with CONFIG_MROUTE
 806        and a multicast routing daemon is required.
 807        conf/all/mc_forwarding must also be set to TRUE to enable multicast
 808        routing for the interface
 809
 810medium_id - INTEGER
 811        Integer value used to differentiate the devices by the medium they
 812        are attached to. Two devices can have different id values when
 813        the broadcast packets are received only on one of them.
 814        The default value 0 means that the device is the only interface
 815        to its medium, value of -1 means that medium is not known.
 816
 817        Currently, it is used to change the proxy_arp behavior:
 818        the proxy_arp feature is enabled for packets forwarded between
 819        two devices attached to different media.
 820
 821proxy_arp - BOOLEAN
 822        Do proxy arp.
 823        proxy_arp for the interface will be enabled if at least one of
 824        conf/{all,interface}/proxy_arp is set to TRUE,
 825        it will be disabled otherwise
 826
 827proxy_arp_pvlan - BOOLEAN
 828        Private VLAN proxy arp.
 829        Basically allow proxy arp replies back to the same interface
 830        (from which the ARP request/solicitation was received).
 831
 832        This is done to support (ethernet) switch features, like RFC
 833        3069, where the individual ports are NOT allowed to
 834        communicate with each other, but they are allowed to talk to
 835        the upstream router.  As described in RFC 3069, it is possible
 836        to allow these hosts to communicate through the upstream
 837        router by proxy_arp'ing. Don't need to be used together with
 838        proxy_arp.
 839
 840        This technology is known by different names:
 841          In RFC 3069 it is called VLAN Aggregation.
 842          Cisco and Allied Telesyn call it Private VLAN.
 843          Hewlett-Packard call it Source-Port filtering or port-isolation.
 844          Ericsson call it MAC-Forced Forwarding (RFC Draft).
 845
 846shared_media - BOOLEAN
 847        Send(router) or accept(host) RFC1620 shared media redirects.
 848        Overrides ip_secure_redirects.
 849        shared_media for the interface will be enabled if at least one of
 850        conf/{all,interface}/shared_media is set to TRUE,
 851        it will be disabled otherwise
 852        default TRUE
 853
 854secure_redirects - BOOLEAN
 855        Accept ICMP redirect messages only for gateways,
 856        listed in default gateway list.
 857        secure_redirects for the interface will be enabled if at least one of
 858        conf/{all,interface}/secure_redirects is set to TRUE,
 859        it will be disabled otherwise
 860        default TRUE
 861
 862send_redirects - BOOLEAN
 863        Send redirects, if router.
 864        send_redirects for the interface will be enabled if at least one of
 865        conf/{all,interface}/send_redirects is set to TRUE,
 866        it will be disabled otherwise
 867        Default: TRUE
 868
 869bootp_relay - BOOLEAN
 870        Accept packets with source address 0.b.c.d destined
 871        not to this host as local ones. It is supposed, that
 872        BOOTP relay daemon will catch and forward such packets.
 873        conf/all/bootp_relay must also be set to TRUE to enable BOOTP relay
 874        for the interface
 875        default FALSE
 876        Not Implemented Yet.
 877
 878accept_source_route - BOOLEAN
 879        Accept packets with SRR option.
 880        conf/all/accept_source_route must also be set to TRUE to accept packets
 881        with SRR option on the interface
 882        default TRUE (router)
 883                FALSE (host)
 884
 885accept_local - BOOLEAN
 886        Accept packets with local source addresses. In combination
 887        with suitable routing, this can be used to direct packets
 888        between two local interfaces over the wire and have them
 889        accepted properly.
 890
 891        rp_filter must be set to a non-zero value in order for
 892        accept_local to have an effect.
 893
 894        default FALSE
 895
 896route_localnet - BOOLEAN
 897        Do not consider loopback addresses as martian source or destination
 898        while routing. This enables the use of 127/8 for local routing purposes.
 899        default FALSE
 900
 901rp_filter - INTEGER
 902        0 - No source validation.
 903        1 - Strict mode as defined in RFC3704 Strict Reverse Path
 904            Each incoming packet is tested against the FIB and if the interface
 905            is not the best reverse path the packet check will fail.
 906            By default failed packets are discarded.
 907        2 - Loose mode as defined in RFC3704 Loose Reverse Path
 908            Each incoming packet's source address is also tested against the FIB
 909            and if the source address is not reachable via any interface
 910            the packet check will fail.
 911
 912        Current recommended practice in RFC3704 is to enable strict mode
 913        to prevent IP spoofing from DDos attacks. If using asymmetric routing
 914        or other complicated routing, then loose mode is recommended.
 915
 916        The max value from conf/{all,interface}/rp_filter is used
 917        when doing source validation on the {interface}.
 918
 919        Default value is 0. Note that some distributions enable it
 920        in startup scripts.
 921
 922arp_filter - BOOLEAN
 923        1 - Allows you to have multiple network interfaces on the same
 924        subnet, and have the ARPs for each interface be answered
 925        based on whether or not the kernel would route a packet from
 926        the ARP'd IP out that interface (therefore you must use source
 927        based routing for this to work). In other words it allows control
 928        of which cards (usually 1) will respond to an arp request.
 929
 930        0 - (default) The kernel can respond to arp requests with addresses
 931        from other interfaces. This may seem wrong but it usually makes
 932        sense, because it increases the chance of successful communication.
 933        IP addresses are owned by the complete host on Linux, not by
 934        particular interfaces. Only for more complex setups like load-
 935        balancing, does this behaviour cause problems.
 936
 937        arp_filter for the interface will be enabled if at least one of
 938        conf/{all,interface}/arp_filter is set to TRUE,
 939        it will be disabled otherwise
 940
 941arp_announce - INTEGER
 942        Define different restriction levels for announcing the local
 943        source IP address from IP packets in ARP requests sent on
 944        interface:
 945        0 - (default) Use any local address, configured on any interface
 946        1 - Try to avoid local addresses that are not in the target's
 947        subnet for this interface. This mode is useful when target
 948        hosts reachable via this interface require the source IP
 949        address in ARP requests to be part of their logical network
 950        configured on the receiving interface. When we generate the
 951        request we will check all our subnets that include the
 952        target IP and will preserve the source address if it is from
 953        such subnet. If there is no such subnet we select source
 954        address according to the rules for level 2.
 955        2 - Always use the best local address for this target.
 956        In this mode we ignore the source address in the IP packet
 957        and try to select local address that we prefer for talks with
 958        the target host. Such local address is selected by looking
 959        for primary IP addresses on all our subnets on the outgoing
 960        interface that include the target IP address. If no suitable
 961        local address is found we select the first local address
 962        we have on the outgoing interface or on all other interfaces,
 963        with the hope we will receive reply for our request and
 964        even sometimes no matter the source IP address we announce.
 965
 966        The max value from conf/{all,interface}/arp_announce is used.
 967
 968        Increasing the restriction level gives more chance for
 969        receiving answer from the resolved target while decreasing
 970        the level announces more valid sender's information.
 971
 972arp_ignore - INTEGER
 973        Define different modes for sending replies in response to
 974        received ARP requests that resolve local target IP addresses:
 975        0 - (default): reply for any local target IP address, configured
 976        on any interface
 977        1 - reply only if the target IP address is local address
 978        configured on the incoming interface
 979        2 - reply only if the target IP address is local address
 980        configured on the incoming interface and both with the
 981        sender's IP address are part from same subnet on this interface
 982        3 - do not reply for local addresses configured with scope host,
 983        only resolutions for global and link addresses are replied
 984        4-7 - reserved
 985        8 - do not reply for all local addresses
 986
 987        The max value from conf/{all,interface}/arp_ignore is used
 988        when ARP request is received on the {interface}
 989
 990arp_notify - BOOLEAN
 991        Define mode for notification of address and device changes.
 992        0 - (default): do nothing
 993        1 - Generate gratuitous arp requests when device is brought up
 994            or hardware address changes.
 995
 996arp_accept - BOOLEAN
 997        Define behavior for gratuitous ARP frames who's IP is not
 998        already present in the ARP table:
 999        0 - don't create new entries in the ARP table
1000        1 - create new entries in the ARP table
1001
1002        Both replies and requests type gratuitous arp will trigger the
1003        ARP table to be updated, if this setting is on.
1004
1005        If the ARP table already contains the IP address of the
1006        gratuitous arp frame, the arp table will be updated regardless
1007        if this setting is on or off.
1008
1009
1010app_solicit - INTEGER
1011        The maximum number of probes to send to the user space ARP daemon
1012        via netlink before dropping back to multicast probes (see
1013        mcast_solicit).  Defaults to 0.
1014
1015disable_policy - BOOLEAN
1016        Disable IPSEC policy (SPD) for this interface
1017
1018disable_xfrm - BOOLEAN
1019        Disable IPSEC encryption on this interface, whatever the policy
1020
1021
1022
1023tag - INTEGER
1024        Allows you to write a number, which can be used as required.
1025        Default value is 0.
1026
1027Alexey Kuznetsov.
1028kuznet@ms2.inr.ac.ru
1029
1030Updated by:
1031Andi Kleen
1032ak@muc.de
1033Nicolas Delon
1034delon.nicolas@wanadoo.fr
1035
1036
1037
1038
1039/proc/sys/net/ipv6/* Variables:
1040
1041IPv6 has no global variables such as tcp_*.  tcp_* settings under ipv4/ also
1042apply to IPv6 [XXX?].
1043
1044bindv6only - BOOLEAN
1045        Default value for IPV6_V6ONLY socket option,
1046        which restricts use of the IPv6 socket to IPv6 communication
1047        only.
1048                TRUE: disable IPv4-mapped address feature
1049                FALSE: enable IPv4-mapped address feature
1050
1051        Default: FALSE (as specified in RFC3493)
1052
1053IPv6 Fragmentation:
1054
1055ip6frag_high_thresh - INTEGER
1056        Maximum memory used to reassemble IPv6 fragments. When
1057        ip6frag_high_thresh bytes of memory is allocated for this purpose,
1058        the fragment handler will toss packets until ip6frag_low_thresh
1059        is reached.
1060
1061ip6frag_low_thresh - INTEGER
1062        See ip6frag_high_thresh
1063
1064ip6frag_time - INTEGER
1065        Time in seconds to keep an IPv6 fragment in memory.
1066
1067ip6frag_secret_interval - INTEGER
1068        Regeneration interval (in seconds) of the hash secret (or lifetime
1069        for the hash secret) for IPv6 fragments.
1070        Default: 600
1071
1072conf/default/*:
1073        Change the interface-specific default settings.
1074
1075
1076conf/all/*:
1077        Change all the interface-specific settings.
1078
1079        [XXX:  Other special features than forwarding?]
1080
1081conf/all/forwarding - BOOLEAN
1082        Enable global IPv6 forwarding between all interfaces.
1083
1084        IPv4 and IPv6 work differently here; e.g. netfilter must be used
1085        to control which interfaces may forward packets and which not.
1086
1087        This also sets all interfaces' Host/Router setting
1088        'forwarding' to the specified value.  See below for details.
1089
1090        This referred to as global forwarding.
1091
1092proxy_ndp - BOOLEAN
1093        Do proxy ndp.
1094
1095conf/interface/*:
1096        Change special settings per interface.
1097
1098        The functional behaviour for certain settings is different
1099        depending on whether local forwarding is enabled or not.
1100
1101accept_ra - INTEGER
1102        Accept Router Advertisements; autoconfigure using them.
1103
1104        It also determines whether or not to transmit Router
1105        Solicitations. If and only if the functional setting is to
1106        accept Router Advertisements, Router Solicitations will be
1107        transmitted.
1108
1109        Possible values are:
1110                0 Do not accept Router Advertisements.
1111                1 Accept Router Advertisements if forwarding is disabled.
1112                2 Overrule forwarding behaviour. Accept Router Advertisements
1113                  even if forwarding is enabled.
1114
1115        Functional default: enabled if local forwarding is disabled.
1116                            disabled if local forwarding is enabled.
1117
1118accept_ra_defrtr - BOOLEAN
1119        Learn default router in Router Advertisement.
1120
1121        Functional default: enabled if accept_ra is enabled.
1122                            disabled if accept_ra is disabled.
1123
1124accept_ra_pinfo - BOOLEAN
1125        Learn Prefix Information in Router Advertisement.
1126
1127        Functional default: enabled if accept_ra is enabled.
1128                            disabled if accept_ra is disabled.
1129
1130accept_ra_rt_info_max_plen - INTEGER
1131        Maximum prefix length of Route Information in RA.
1132
1133        Route Information w/ prefix larger than or equal to this
1134        variable shall be ignored.
1135
1136        Functional default: 0 if accept_ra_rtr_pref is enabled.
1137                            -1 if accept_ra_rtr_pref is disabled.
1138
1139accept_ra_rtr_pref - BOOLEAN
1140        Accept Router Preference in RA.
1141
1142        Functional default: enabled if accept_ra is enabled.
1143                            disabled if accept_ra is disabled.
1144
1145accept_redirects - BOOLEAN
1146        Accept Redirects.
1147
1148        Functional default: enabled if local forwarding is disabled.
1149                            disabled if local forwarding is enabled.
1150
1151accept_source_route - INTEGER
1152        Accept source routing (routing extension header).
1153
1154        >= 0: Accept only routing header type 2.
1155        < 0: Do not accept routing header.
1156
1157        Default: 0
1158
1159autoconf - BOOLEAN
1160        Autoconfigure addresses using Prefix Information in Router
1161        Advertisements.
1162
1163        Functional default: enabled if accept_ra_pinfo is enabled.
1164                            disabled if accept_ra_pinfo is disabled.
1165
1166dad_transmits - INTEGER
1167        The amount of Duplicate Address Detection probes to send.
1168        Default: 1
1169
1170forwarding - INTEGER
1171        Configure interface-specific Host/Router behaviour.
1172
1173        Note: It is recommended to have the same setting on all
1174        interfaces; mixed router/host scenarios are rather uncommon.
1175
1176        Possible values are:
1177                0 Forwarding disabled
1178                1 Forwarding enabled
1179
1180        FALSE (0):
1181
1182        By default, Host behaviour is assumed.  This means:
1183
1184        1. IsRouter flag is not set in Neighbour Advertisements.
1185        2. If accept_ra is TRUE (default), transmit Router
1186           Solicitations.
1187        3. If accept_ra is TRUE (default), accept Router
1188           Advertisements (and do autoconfiguration).
1189        4. If accept_redirects is TRUE (default), accept Redirects.
1190
1191        TRUE (1):
1192
1193        If local forwarding is enabled, Router behaviour is assumed.
1194        This means exactly the reverse from the above:
1195
1196        1. IsRouter flag is set in Neighbour Advertisements.
1197        2. Router Solicitations are not sent unless accept_ra is 2.
1198        3. Router Advertisements are ignored unless accept_ra is 2.
1199        4. Redirects are ignored.
1200
1201        Default: 0 (disabled) if global forwarding is disabled (default),
1202                 otherwise 1 (enabled).
1203
1204hop_limit - INTEGER
1205        Default Hop Limit to set.
1206        Default: 64
1207
1208mtu - INTEGER
1209        Default Maximum Transfer Unit
1210        Default: 1280 (IPv6 required minimum)
1211
1212router_probe_interval - INTEGER
1213        Minimum interval (in seconds) between Router Probing described
1214        in RFC4191.
1215
1216        Default: 60
1217
1218router_solicitation_delay - INTEGER
1219        Number of seconds to wait after interface is brought up
1220        before sending Router Solicitations.
1221        Default: 1
1222
1223router_solicitation_interval - INTEGER
1224        Number of seconds to wait between Router Solicitations.
1225        Default: 4
1226
1227router_solicitations - INTEGER
1228        Number of Router Solicitations to send until assuming no
1229        routers are present.
1230        Default: 3
1231
1232use_tempaddr - INTEGER
1233        Preference for Privacy Extensions (RFC3041).
1234          <= 0 : disable Privacy Extensions
1235          == 1 : enable Privacy Extensions, but prefer public
1236                 addresses over temporary addresses.
1237          >  1 : enable Privacy Extensions and prefer temporary
1238                 addresses over public addresses.
1239        Default:  0 (for most devices)
1240                 -1 (for point-to-point devices and loopback devices)
1241
1242temp_valid_lft - INTEGER
1243        valid lifetime (in seconds) for temporary addresses.
1244        Default: 604800 (7 days)
1245
1246temp_prefered_lft - INTEGER
1247        Preferred lifetime (in seconds) for temporary addresses.
1248        Default: 86400 (1 day)
1249
1250max_desync_factor - INTEGER
1251        Maximum value for DESYNC_FACTOR, which is a random value
1252        that ensures that clients don't synchronize with each
1253        other and generate new addresses at exactly the same time.
1254        value is in seconds.
1255        Default: 600
1256
1257regen_max_retry - INTEGER
1258        Number of attempts before give up attempting to generate
1259        valid temporary addresses.
1260        Default: 5
1261
1262max_addresses - INTEGER
1263        Maximum number of autoconfigured addresses per interface.  Setting
1264        to zero disables the limitation.  It is not recommended to set this
1265        value too large (or to zero) because it would be an easy way to
1266        crash the kernel by allowing too many addresses to be created.
1267        Default: 16
1268
1269disable_ipv6 - BOOLEAN
1270        Disable IPv6 operation.  If accept_dad is set to 2, this value
1271        will be dynamically set to TRUE if DAD fails for the link-local
1272        address.
1273        Default: FALSE (enable IPv6 operation)
1274
1275        When this value is changed from 1 to 0 (IPv6 is being enabled),
1276        it will dynamically create a link-local address on the given
1277        interface and start Duplicate Address Detection, if necessary.
1278
1279        When this value is changed from 0 to 1 (IPv6 is being disabled),
1280        it will dynamically delete all address on the given interface.
1281
1282accept_dad - INTEGER
1283        Whether to accept DAD (Duplicate Address Detection).
1284        0: Disable DAD
1285        1: Enable DAD (default)
1286        2: Enable DAD, and disable IPv6 operation if MAC-based duplicate
1287           link-local address has been found.
1288
1289force_tllao - BOOLEAN
1290        Enable sending the target link-layer address option even when
1291        responding to a unicast neighbor solicitation.
1292        Default: FALSE
1293
1294        Quoting from RFC 2461, section 4.4, Target link-layer address:
1295
1296        "The option MUST be included for multicast solicitations in order to
1297        avoid infinite Neighbor Solicitation "recursion" when the peer node
1298        does not have a cache entry to return a Neighbor Advertisements
1299        message.  When responding to unicast solicitations, the option can be
1300        omitted since the sender of the solicitation has the correct link-
1301        layer address; otherwise it would not have be able to send the unicast
1302        solicitation in the first place. However, including the link-layer
1303        address in this case adds little overhead and eliminates a potential
1304        race condition where the sender deletes the cached link-layer address
1305        prior to receiving a response to a previous solicitation."
1306
1307icmp/*:
1308ratelimit - INTEGER
1309        Limit the maximal rates for sending ICMPv6 packets.
1310        0 to disable any limiting,
1311        otherwise the minimal space between responses in milliseconds.
1312        Default: 1000
1313
1314
1315IPv6 Update by:
1316Pekka Savola <pekkas@netcore.fi>
1317YOSHIFUJI Hideaki / USAGI Project <yoshfuji@linux-ipv6.org>
1318
1319
1320/proc/sys/net/bridge/* Variables:
1321
1322bridge-nf-call-arptables - BOOLEAN
1323        1 : pass bridged ARP traffic to arptables' FORWARD chain.
1324        0 : disable this.
1325        Default: 1
1326
1327bridge-nf-call-iptables - BOOLEAN
1328        1 : pass bridged IPv4 traffic to iptables' chains.
1329        0 : disable this.
1330        Default: 1
1331
1332bridge-nf-call-ip6tables - BOOLEAN
1333        1 : pass bridged IPv6 traffic to ip6tables' chains.
1334        0 : disable this.
1335        Default: 1
1336
1337bridge-nf-filter-vlan-tagged - BOOLEAN
1338        1 : pass bridged vlan-tagged ARP/IP/IPv6 traffic to {arp,ip,ip6}tables.
1339        0 : disable this.
1340        Default: 0
1341
1342bridge-nf-filter-pppoe-tagged - BOOLEAN
1343        1 : pass bridged pppoe-tagged IP/IPv6 traffic to {ip,ip6}tables.
1344        0 : disable this.
1345        Default: 0
1346
1347bridge-nf-pass-vlan-input-dev - BOOLEAN
1348        1: if bridge-nf-filter-vlan-tagged is enabled, try to find a vlan
1349        interface on the bridge and set the netfilter input device to the vlan.
1350        This allows use of e.g. "iptables -i br0.1" and makes the REDIRECT
1351        target work with vlan-on-top-of-bridge interfaces.  When no matching
1352        vlan interface is found, or this switch is off, the input device is
1353        set to the bridge interface.
1354        0: disable bridge netfilter vlan interface lookup.
1355        Default: 0
1356
1357proc/sys/net/sctp/* Variables:
1358
1359addip_enable - BOOLEAN
1360        Enable or disable extension of  Dynamic Address Reconfiguration
1361        (ADD-IP) functionality specified in RFC5061.  This extension provides
1362        the ability to dynamically add and remove new addresses for the SCTP
1363        associations.
1364
1365        1: Enable extension.
1366
1367        0: Disable extension.
1368
1369        Default: 0
1370
1371addip_noauth_enable - BOOLEAN
1372        Dynamic Address Reconfiguration (ADD-IP) requires the use of
1373        authentication to protect the operations of adding or removing new
1374        addresses.  This requirement is mandated so that unauthorized hosts
1375        would not be able to hijack associations.  However, older
1376        implementations may not have implemented this requirement while
1377        allowing the ADD-IP extension.  For reasons of interoperability,
1378        we provide this variable to control the enforcement of the
1379        authentication requirement.
1380
1381        1: Allow ADD-IP extension to be used without authentication.  This
1382           should only be set in a closed environment for interoperability
1383           with older implementations.
1384
1385        0: Enforce the authentication requirement
1386
1387        Default: 0
1388
1389auth_enable - BOOLEAN
1390        Enable or disable Authenticated Chunks extension.  This extension
1391        provides the ability to send and receive authenticated chunks and is
1392        required for secure operation of Dynamic Address Reconfiguration
1393        (ADD-IP) extension.
1394
1395        1: Enable this extension.
1396        0: Disable this extension.
1397
1398        Default: 0
1399
1400prsctp_enable - BOOLEAN
1401        Enable or disable the Partial Reliability extension (RFC3758) which
1402        is used to notify peers that a given DATA should no longer be expected.
1403
1404        1: Enable extension
1405        0: Disable
1406
1407        Default: 1
1408
1409max_burst - INTEGER
1410        The limit of the number of new packets that can be initially sent.  It
1411        controls how bursty the generated traffic can be.
1412
1413        Default: 4
1414
1415association_max_retrans - INTEGER
1416        Set the maximum number for retransmissions that an association can
1417        attempt deciding that the remote end is unreachable.  If this value
1418        is exceeded, the association is terminated.
1419
1420        Default: 10
1421
1422max_init_retransmits - INTEGER
1423        The maximum number of retransmissions of INIT and COOKIE-ECHO chunks
1424        that an association will attempt before declaring the destination
1425        unreachable and terminating.
1426
1427        Default: 8
1428
1429path_max_retrans - INTEGER
1430        The maximum number of retransmissions that will be attempted on a given
1431        path.  Once this threshold is exceeded, the path is considered
1432        unreachable, and new traffic will use a different path when the
1433        association is multihomed.
1434
1435        Default: 5
1436
1437pf_retrans - INTEGER
1438        The number of retransmissions that will be attempted on a given path
1439        before traffic is redirected to an alternate transport (should one
1440        exist).  Note this is distinct from path_max_retrans, as a path that
1441        passes the pf_retrans threshold can still be used.  Its only
1442        deprioritized when a transmission path is selected by the stack.  This
1443        setting is primarily used to enable fast failover mechanisms without
1444        having to reduce path_max_retrans to a very low value.  See:
1445        http://www.ietf.org/id/draft-nishida-tsvwg-sctp-failover-05.txt
1446        for details.  Note also that a value of pf_retrans > path_max_retrans
1447        disables this feature
1448
1449        Default: 0
1450
1451rto_initial - INTEGER
1452        The initial round trip timeout value in milliseconds that will be used
1453        in calculating round trip times.  This is the initial time interval
1454        for retransmissions.
1455
1456        Default: 3000
1457
1458rto_max - INTEGER
1459        The maximum value (in milliseconds) of the round trip timeout.  This
1460        is the largest time interval that can elapse between retransmissions.
1461
1462        Default: 60000
1463
1464rto_min - INTEGER
1465        The minimum value (in milliseconds) of the round trip timeout.  This
1466        is the smallest time interval the can elapse between retransmissions.
1467
1468        Default: 1000
1469
1470hb_interval - INTEGER
1471        The interval (in milliseconds) between HEARTBEAT chunks.  These chunks
1472        are sent at the specified interval on idle paths to probe the state of
1473        a given path between 2 associations.
1474
1475        Default: 30000
1476
1477sack_timeout - INTEGER
1478        The amount of time (in milliseconds) that the implementation will wait
1479        to send a SACK.
1480
1481        Default: 200
1482
1483valid_cookie_life - INTEGER
1484        The default lifetime of the SCTP cookie (in milliseconds).  The cookie
1485        is used during association establishment.
1486
1487        Default: 60000
1488
1489cookie_preserve_enable - BOOLEAN
1490        Enable or disable the ability to extend the lifetime of the SCTP cookie
1491        that is used during the establishment phase of SCTP association
1492
1493        1: Enable cookie lifetime extension.
1494        0: Disable
1495
1496        Default: 1
1497
1498rcvbuf_policy - INTEGER
1499        Determines if the receive buffer is attributed to the socket or to
1500        association.   SCTP supports the capability to create multiple
1501        associations on a single socket.  When using this capability, it is
1502        possible that a single stalled association that's buffering a lot
1503        of data may block other associations from delivering their data by
1504        consuming all of the receive buffer space.  To work around this,
1505        the rcvbuf_policy could be set to attribute the receiver buffer space
1506        to each association instead of the socket.  This prevents the described
1507        blocking.
1508
1509        1: rcvbuf space is per association
1510        0: recbuf space is per socket
1511
1512        Default: 0
1513
1514sndbuf_policy - INTEGER
1515        Similar to rcvbuf_policy above, this applies to send buffer space.
1516
1517        1: Send buffer is tracked per association
1518        0: Send buffer is tracked per socket.
1519
1520        Default: 0
1521
1522sctp_mem - vector of 3 INTEGERs: min, pressure, max
1523        Number of pages allowed for queueing by all SCTP sockets.
1524
1525        min: Below this number of pages SCTP is not bothered about its
1526        memory appetite. When amount of memory allocated by SCTP exceeds
1527        this number, SCTP starts to moderate memory usage.
1528
1529        pressure: This value was introduced to follow format of tcp_mem.
1530
1531        max: Number of pages allowed for queueing by all SCTP sockets.
1532
1533        Default is calculated at boot time from amount of available memory.
1534
1535sctp_rmem - vector of 3 INTEGERs: min, default, max
1536        Only the first value ("min") is used, "default" and "max" are
1537        ignored.
1538
1539        min: Minimal size of receive buffer used by SCTP socket.
1540        It is guaranteed to each SCTP socket (but not association) even
1541        under moderate memory pressure.
1542
1543        Default: 1 page
1544
1545sctp_wmem  - vector of 3 INTEGERs: min, default, max
1546        Currently this tunable has no effect.
1547
1548addr_scope_policy - INTEGER
1549        Control IPv4 address scoping - draft-stewart-tsvwg-sctp-ipv4-00
1550
1551        0   - Disable IPv4 address scoping
1552        1   - Enable IPv4 address scoping
1553        2   - Follow draft but allow IPv4 private addresses
1554        3   - Follow draft but allow IPv4 link local addresses
1555
1556        Default: 1
1557
1558
1559/proc/sys/net/core/*
1560        Please see: Documentation/sysctl/net.txt for descriptions of these entries.
1561
1562
1563/proc/sys/net/unix/*
1564max_dgram_qlen - INTEGER
1565        The maximum length of dgram socket receive queue
1566
1567        Default: 10
1568
1569
1570UNDOCUMENTED:
1571
1572/proc/sys/net/irda/*
1573        fast_poll_increase FIXME
1574        warn_noreply_time FIXME
1575        discovery_slots FIXME
1576        slot_timeout FIXME
1577        max_baud_rate FIXME
1578        discovery_timeout FIXME
1579        lap_keepalive_time FIXME
1580        max_noreply_time FIXME
1581        max_tx_data_size FIXME
1582        max_tx_window FIXME
1583        min_tx_turn_time FIXME
1584
lxr.linux.no kindly hosted by Redpill Linpro AS, provider of Linux consulting and operations services since 1995.