| 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 |  | 
|  | 35 | rmem_default | 
|  | 36 | ------------ | 
|  | 37 |  | 
|  | 38 | The default setting of the socket receive buffer in bytes. | 
|  | 39 |  | 
|  | 40 | rmem_max | 
|  | 41 | -------- | 
|  | 42 |  | 
|  | 43 | The maximum receive socket buffer size in bytes. | 
|  | 44 |  | 
|  | 45 | wmem_default | 
|  | 46 | ------------ | 
|  | 47 |  | 
|  | 48 | The default setting (in bytes) of the socket send buffer. | 
|  | 49 |  | 
|  | 50 | wmem_max | 
|  | 51 | -------- | 
|  | 52 |  | 
|  | 53 | The maximum send socket buffer size in bytes. | 
|  | 54 |  | 
|  | 55 | message_burst and message_cost | 
|  | 56 | ------------------------------ | 
|  | 57 |  | 
|  | 58 | These parameters  are used to limit the warning messages written to the kernel | 
|  | 59 | log from  the  networking  code.  They  enforce  a  rate  limit  to  make  a | 
|  | 60 | denial-of-service attack  impossible. A higher message_cost factor, results in | 
|  | 61 | fewer messages that will be written. Message_burst controls when messages will | 
|  | 62 | be dropped.  The  default  settings  limit  warning messages to one every five | 
|  | 63 | seconds. | 
|  | 64 |  | 
|  | 65 | warnings | 
|  | 66 | -------- | 
|  | 67 |  | 
|  | 68 | This controls console messages from the networking stack that can occur because | 
|  | 69 | of problems on the network like duplicate address or bad checksums. Normally, | 
|  | 70 | this should be enabled, but if the problem persists the messages can be | 
|  | 71 | disabled. | 
|  | 72 |  | 
|  | 73 | netdev_budget | 
|  | 74 | ------------- | 
|  | 75 |  | 
|  | 76 | Maximum number of packets taken from all interfaces in one polling cycle (NAPI | 
|  | 77 | poll). In one polling cycle interfaces which are registered to polling are | 
|  | 78 | probed in a round-robin manner. The limit of packets in one such probe can be | 
|  | 79 | set per-device via sysfs class/net/<device>/weight . | 
|  | 80 |  | 
|  | 81 | netdev_max_backlog | 
|  | 82 | ------------------ | 
|  | 83 |  | 
|  | 84 | Maximum number  of  packets,  queued  on  the  INPUT  side, when the interface | 
|  | 85 | receives packets faster than kernel can process them. | 
|  | 86 |  | 
| Eric Dumazet | 3b098e2 | 2010-05-15 23:57:10 -0700 | [diff] [blame] | 87 | netdev_tstamp_prequeue | 
|  | 88 | ---------------------- | 
|  | 89 |  | 
|  | 90 | If set to 0, RX packet timestamps can be sampled after RPS processing, when | 
|  | 91 | the target CPU processes packets. It might give some delay on timestamps, but | 
|  | 92 | permit to distribute the load on several cpus. | 
|  | 93 |  | 
|  | 94 | If set to 1 (default), timestamps are sampled as soon as possible, before | 
|  | 95 | queueing. | 
|  | 96 |  | 
| Shen Feng | 760df93 | 2009-04-02 16:57:20 -0700 | [diff] [blame] | 97 | optmem_max | 
|  | 98 | ---------- | 
|  | 99 |  | 
|  | 100 | Maximum ancillary buffer size allowed per socket. Ancillary data is a sequence | 
|  | 101 | of struct cmsghdr structures with appended data. | 
|  | 102 |  | 
|  | 103 | 2. /proc/sys/net/unix - Parameters for Unix domain sockets | 
|  | 104 | ------------------------------------------------------- | 
|  | 105 |  | 
| Li Xiaodong | 45dad7b | 2009-04-02 16:57:21 -0700 | [diff] [blame] | 106 | There is only one file in this directory. | 
|  | 107 | 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] | 108 | 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] | 109 |  | 
|  | 110 |  | 
|  | 111 | 3. /proc/sys/net/ipv4 - IPV4 settings | 
|  | 112 | ------------------------------------------------------- | 
|  | 113 | Please see: Documentation/networking/ip-sysctl.txt and ipvs-sysctl.txt for | 
|  | 114 | descriptions of these entries. | 
|  | 115 |  | 
|  | 116 |  | 
|  | 117 | 4. Appletalk | 
|  | 118 | ------------------------------------------------------- | 
|  | 119 |  | 
|  | 120 | The /proc/sys/net/appletalk  directory  holds the Appletalk configuration data | 
|  | 121 | when Appletalk is loaded. The configurable parameters are: | 
|  | 122 |  | 
|  | 123 | aarp-expiry-time | 
|  | 124 | ---------------- | 
|  | 125 |  | 
|  | 126 | The amount  of  time  we keep an ARP entry before expiring it. Used to age out | 
|  | 127 | old hosts. | 
|  | 128 |  | 
|  | 129 | aarp-resolve-time | 
|  | 130 | ----------------- | 
|  | 131 |  | 
|  | 132 | The amount of time we will spend trying to resolve an Appletalk address. | 
|  | 133 |  | 
|  | 134 | aarp-retransmit-limit | 
|  | 135 | --------------------- | 
|  | 136 |  | 
|  | 137 | The number of times we will retransmit a query before giving up. | 
|  | 138 |  | 
|  | 139 | aarp-tick-time | 
|  | 140 | -------------- | 
|  | 141 |  | 
|  | 142 | Controls the rate at which expires are checked. | 
|  | 143 |  | 
|  | 144 | The directory  /proc/net/appletalk  holds the list of active Appletalk sockets | 
|  | 145 | on a machine. | 
|  | 146 |  | 
|  | 147 | The fields  indicate  the DDP type, the local address (in network:node format) | 
|  | 148 | the remote  address,  the  size of the transmit pending queue, the size of the | 
|  | 149 | received queue  (bytes waiting for applications to read) the state and the uid | 
|  | 150 | owning the socket. | 
|  | 151 |  | 
|  | 152 | /proc/net/atalk_iface lists  all  the  interfaces  configured for appletalk.It | 
|  | 153 | shows the  name  of the interface, its Appletalk address, the network range on | 
|  | 154 | that address  (or  network number for phase 1 networks), and the status of the | 
|  | 155 | interface. | 
|  | 156 |  | 
|  | 157 | /proc/net/atalk_route lists  each  known  network  route.  It lists the target | 
|  | 158 | (network) that the route leads to, the router (may be directly connected), the | 
|  | 159 | route flags, and the device the route is using. | 
|  | 160 |  | 
|  | 161 |  | 
|  | 162 | 5. IPX | 
|  | 163 | ------------------------------------------------------- | 
|  | 164 |  | 
|  | 165 | The IPX protocol has no tunable values in proc/sys/net. | 
|  | 166 |  | 
|  | 167 | The IPX  protocol  does,  however,  provide  proc/net/ipx. This lists each IPX | 
|  | 168 | socket giving  the  local  and  remote  addresses  in  Novell  format (that is | 
|  | 169 | network:node:port). In  accordance  with  the  strange  Novell  tradition, | 
|  | 170 | everything but the port is in hex. Not_Connected is displayed for sockets that | 
|  | 171 | are not  tied to a specific remote address. The Tx and Rx queue sizes indicate | 
|  | 172 | the number  of  bytes  pending  for  transmission  and  reception.  The  state | 
|  | 173 | indicates the  state  the  socket  is  in and the uid is the owning uid of the | 
|  | 174 | socket. | 
|  | 175 |  | 
|  | 176 | The /proc/net/ipx_interface  file lists all IPX interfaces. For each interface | 
|  | 177 | it gives  the network number, the node number, and indicates if the network is | 
|  | 178 | the primary  network.  It  also  indicates  which  device  it  is bound to (or | 
|  | 179 | Internal for  internal  networks)  and  the  Frame  Type if appropriate. Linux | 
|  | 180 | supports 802.3,  802.2,  802.2  SNAP  and DIX (Blue Book) ethernet framing for | 
|  | 181 | IPX. | 
|  | 182 |  | 
|  | 183 | The /proc/net/ipx_route  table  holds  a list of IPX routes. For each route it | 
|  | 184 | gives the  destination  network, the router node (or Directly) and the network | 
|  | 185 | address of the router (or Connected) for internal networks. |