Shen Feng | 760df93 | 2009-04-02 16:57:20 -0700 | [diff] [blame] | 1 | Documentation for /proc/sys/net/* kernel version 2.4.0-test11-pre4 |
| 2 | (c) 1999 Terrehon Bowden <terrehon@pacbell.net> |
| 3 | Bodo Bauer <bb@ricochet.net> |
| 4 | (c) 2000 Jorge Nerin <comandante@zaralinux.com> |
| 5 | (c) 2009 Shen Feng <shen@cn.fujitsu.com> |
| 6 | |
| 7 | For general info and legal blurb, please look in README. |
| 8 | |
| 9 | ============================================================== |
| 10 | |
| 11 | This file contains the documentation for the sysctl files in |
| 12 | /proc/sys/net and is valid for Linux kernel version 2.4.0-test11-pre4. |
| 13 | |
| 14 | The interface to the networking parts of the kernel is located in |
| 15 | /proc/sys/net. The following table shows all possible subdirectories.You may |
| 16 | see only some of them, depending on your kernel's configuration. |
| 17 | |
| 18 | |
| 19 | Table : Subdirectories in /proc/sys/net |
| 20 | .............................................................................. |
| 21 | Directory Content Directory Content |
| 22 | core General parameter appletalk Appletalk protocol |
| 23 | unix Unix domain sockets netrom NET/ROM |
| 24 | 802 E802 protocol ax25 AX25 |
| 25 | ethernet Ethernet protocol rose X.25 PLP layer |
| 26 | ipv4 IP version 4 x25 X.25 protocol |
| 27 | ipx IPX token-ring IBM token ring |
| 28 | bridge Bridging decnet DEC net |
| 29 | ipv6 IP version 6 |
| 30 | .............................................................................. |
| 31 | |
| 32 | 1. /proc/sys/net/core - Network core options |
| 33 | ------------------------------------------------------- |
| 34 | |
Eric Dumazet | 0a14842 | 2011-04-20 09:27:32 +0000 | [diff] [blame] | 35 | bpf_jit_enable |
| 36 | -------------- |
| 37 | |
| 38 | This enables Berkeley Packet Filter Just in Time compiler. |
| 39 | Currently supported on x86_64 architecture, bpf_jit provides a framework |
| 40 | to speed packet filtering, the one used by tcpdump/libpcap for example. |
| 41 | Values : |
| 42 | 0 - disable the JIT (default value) |
| 43 | 1 - enable the JIT |
| 44 | 2 - enable the JIT and ask the compiler to emit traces on kernel log. |
| 45 | |
Shan Wei | c60f6aa | 2012-04-26 16:52:52 +0000 | [diff] [blame] | 46 | dev_weight |
| 47 | -------------- |
| 48 | |
| 49 | The maximum number of packets that kernel can handle on a NAPI interrupt, |
| 50 | it's a Per-CPU variable. |
| 51 | Default: 64 |
| 52 | |
Shen Feng | 760df93 | 2009-04-02 16:57:20 -0700 | [diff] [blame] | 53 | rmem_default |
| 54 | ------------ |
| 55 | |
| 56 | The default setting of the socket receive buffer in bytes. |
| 57 | |
| 58 | rmem_max |
| 59 | -------- |
| 60 | |
| 61 | The maximum receive socket buffer size in bytes. |
| 62 | |
| 63 | wmem_default |
| 64 | ------------ |
| 65 | |
| 66 | The default setting (in bytes) of the socket send buffer. |
| 67 | |
| 68 | wmem_max |
| 69 | -------- |
| 70 | |
| 71 | The maximum send socket buffer size in bytes. |
| 72 | |
| 73 | message_burst and message_cost |
| 74 | ------------------------------ |
| 75 | |
| 76 | These parameters are used to limit the warning messages written to the kernel |
| 77 | log from the networking code. They enforce a rate limit to make a |
| 78 | denial-of-service attack impossible. A higher message_cost factor, results in |
| 79 | fewer messages that will be written. Message_burst controls when messages will |
| 80 | be dropped. The default settings limit warning messages to one every five |
| 81 | seconds. |
| 82 | |
| 83 | warnings |
| 84 | -------- |
| 85 | |
| 86 | This controls console messages from the networking stack that can occur because |
| 87 | of problems on the network like duplicate address or bad checksums. Normally, |
| 88 | this should be enabled, but if the problem persists the messages can be |
| 89 | disabled. |
| 90 | |
| 91 | netdev_budget |
| 92 | ------------- |
| 93 | |
| 94 | Maximum number of packets taken from all interfaces in one polling cycle (NAPI |
| 95 | poll). In one polling cycle interfaces which are registered to polling are |
| 96 | probed in a round-robin manner. The limit of packets in one such probe can be |
| 97 | set per-device via sysfs class/net/<device>/weight . |
| 98 | |
| 99 | netdev_max_backlog |
| 100 | ------------------ |
| 101 | |
| 102 | Maximum number of packets, queued on the INPUT side, when the interface |
| 103 | receives packets faster than kernel can process them. |
| 104 | |
Eric Dumazet | 3b098e2 | 2010-05-15 23:57:10 -0700 | [diff] [blame] | 105 | netdev_tstamp_prequeue |
| 106 | ---------------------- |
| 107 | |
| 108 | If set to 0, RX packet timestamps can be sampled after RPS processing, when |
| 109 | the target CPU processes packets. It might give some delay on timestamps, but |
| 110 | permit to distribute the load on several cpus. |
| 111 | |
| 112 | If set to 1 (default), timestamps are sampled as soon as possible, before |
| 113 | queueing. |
| 114 | |
Shen Feng | 760df93 | 2009-04-02 16:57:20 -0700 | [diff] [blame] | 115 | optmem_max |
| 116 | ---------- |
| 117 | |
| 118 | Maximum ancillary buffer size allowed per socket. Ancillary data is a sequence |
| 119 | of struct cmsghdr structures with appended data. |
| 120 | |
| 121 | 2. /proc/sys/net/unix - Parameters for Unix domain sockets |
| 122 | ------------------------------------------------------- |
| 123 | |
Li Xiaodong | 45dad7b | 2009-04-02 16:57:21 -0700 | [diff] [blame] | 124 | There is only one file in this directory. |
| 125 | unix_dgram_qlen limits the max number of datagrams queued in Unix domain |
Li Zefan | ca8b995 | 2009-04-13 14:39:36 -0700 | [diff] [blame] | 126 | socket's buffer. It will not take effect unless PF_UNIX flag is specified. |
Shen Feng | 760df93 | 2009-04-02 16:57:20 -0700 | [diff] [blame] | 127 | |
| 128 | |
| 129 | 3. /proc/sys/net/ipv4 - IPV4 settings |
| 130 | ------------------------------------------------------- |
| 131 | Please see: Documentation/networking/ip-sysctl.txt and ipvs-sysctl.txt for |
| 132 | descriptions of these entries. |
| 133 | |
| 134 | |
| 135 | 4. Appletalk |
| 136 | ------------------------------------------------------- |
| 137 | |
| 138 | The /proc/sys/net/appletalk directory holds the Appletalk configuration data |
| 139 | when Appletalk is loaded. The configurable parameters are: |
| 140 | |
| 141 | aarp-expiry-time |
| 142 | ---------------- |
| 143 | |
| 144 | The amount of time we keep an ARP entry before expiring it. Used to age out |
| 145 | old hosts. |
| 146 | |
| 147 | aarp-resolve-time |
| 148 | ----------------- |
| 149 | |
| 150 | The amount of time we will spend trying to resolve an Appletalk address. |
| 151 | |
| 152 | aarp-retransmit-limit |
| 153 | --------------------- |
| 154 | |
| 155 | The number of times we will retransmit a query before giving up. |
| 156 | |
| 157 | aarp-tick-time |
| 158 | -------------- |
| 159 | |
| 160 | Controls the rate at which expires are checked. |
| 161 | |
| 162 | The directory /proc/net/appletalk holds the list of active Appletalk sockets |
| 163 | on a machine. |
| 164 | |
| 165 | The fields indicate the DDP type, the local address (in network:node format) |
| 166 | the remote address, the size of the transmit pending queue, the size of the |
| 167 | received queue (bytes waiting for applications to read) the state and the uid |
| 168 | owning the socket. |
| 169 | |
| 170 | /proc/net/atalk_iface lists all the interfaces configured for appletalk.It |
| 171 | shows the name of the interface, its Appletalk address, the network range on |
| 172 | that address (or network number for phase 1 networks), and the status of the |
| 173 | interface. |
| 174 | |
| 175 | /proc/net/atalk_route lists each known network route. It lists the target |
| 176 | (network) that the route leads to, the router (may be directly connected), the |
| 177 | route flags, and the device the route is using. |
| 178 | |
| 179 | |
| 180 | 5. IPX |
| 181 | ------------------------------------------------------- |
| 182 | |
| 183 | The IPX protocol has no tunable values in proc/sys/net. |
| 184 | |
| 185 | The IPX protocol does, however, provide proc/net/ipx. This lists each IPX |
| 186 | socket giving the local and remote addresses in Novell format (that is |
| 187 | network:node:port). In accordance with the strange Novell tradition, |
| 188 | everything but the port is in hex. Not_Connected is displayed for sockets that |
| 189 | are not tied to a specific remote address. The Tx and Rx queue sizes indicate |
| 190 | the number of bytes pending for transmission and reception. The state |
| 191 | indicates the state the socket is in and the uid is the owning uid of the |
| 192 | socket. |
| 193 | |
| 194 | The /proc/net/ipx_interface file lists all IPX interfaces. For each interface |
| 195 | it gives the network number, the node number, and indicates if the network is |
| 196 | the primary network. It also indicates which device it is bound to (or |
| 197 | Internal for internal networks) and the Frame Type if appropriate. Linux |
| 198 | supports 802.3, 802.2, 802.2 SNAP and DIX (Blue Book) ethernet framing for |
| 199 | IPX. |
| 200 | |
| 201 | The /proc/net/ipx_route table holds a list of IPX routes. For each route it |
| 202 | gives the destination network, the router node (or Directly) and the network |
| 203 | address of the router (or Connected) for internal networks. |