diff options
| author | Michael Kerrisk <mtk.manpages@gmail.com> | 2007-06-23 07:56:56 +0000 |
|---|---|---|
| committer | Michael Kerrisk <mtk.manpages@gmail.com> | 2007-06-23 07:56:56 +0000 |
| commit | f19a0f03eefac77a52687989582a5bcba2e109d3 (patch) | |
| tree | 3fed6b3b2dfd6230eda1c873b043bb68967a283b /man7/tcp.7 | |
| parent | 66ee0c7e89fe01348878638515edb8b2a51b9a2e (diff) | |
| download | man-pages-f19a0f03eefac77a52687989582a5bcba2e109d3.tar.gz | |
ffix
Diffstat (limited to 'man7/tcp.7')
| -rw-r--r-- | man7/tcp.7 | 40 |
1 files changed, 20 insertions, 20 deletions
diff --git a/man7/tcp.7 b/man7/tcp.7 index d54c61d415..61c2787151 100644 --- a/man7/tcp.7 +++ b/man7/tcp.7 @@ -71,9 +71,9 @@ latency or bandwidth. To make use of them, the send and receive buffer sizes must be increased. They can be set globally with the -.B net.ipv4.tcp_wmem +.I net.ipv4.tcp_wmem and -.B net.ipv4.tcp_rmem +.I net.ipv4.tcp_rmem sysctl variables, or on individual sockets by using the .B SO_SNDBUF and @@ -87,9 +87,9 @@ The maximum sizes for socket buffers declared via the and .B SO_RCVBUF mechanisms are limited by the global -.B net.core.rmem_max +.I net.core.rmem_max and -.B net.core.wmem_max +.I net.core.wmem_max sysctls. Note that TCP actually allocates twice the size of the buffer requested in the @@ -210,7 +210,7 @@ The rest of the space is used as the "application" buffer, used to isolate the network from scheduling and application latencies. The -.BR tcp_adv_win_scale +.IR tcp_adv_win_scale default value of 2 implies that the space used for the application buffer is one fourth that of the total. @@ -357,7 +357,7 @@ depending on the memory in the system. If this number is exceeded, the socket is closed and a warning is printed. .TP -.BR tcp_mem +.IR tcp_mem This is a vector of 3 integers: [low, pressure, high]. These bounds are used by TCP to track its memory usage. The @@ -430,7 +430,7 @@ if a RST is received in TIME_WAIT state, we close the socket immediately without waiting for the end of the TIME_WAIT period. .TP -.BR tcp_rmem +.IR tcp_rmem This is a vector of 3 integers: [min, default, max]. These parameters are used by TCP to regulate receive @@ -457,7 +457,7 @@ on a socket. - the default size of the receive buffer for a TCP socket. This value overwrites the initial default buffer size from the generic global -.B net.core.rmem_default +.I net.core.rmem_default defined for all protocols. The default value is 87380 bytes, and is lowered to 43689 in low-memory systems. @@ -465,14 +465,14 @@ If larger receive buffer sizes are desired, this value should be increased (to affect all sockets). To employ large TCP windows, the -.B net.ipv4.tcp_window_scaling +.I net.ipv4.tcp_window_scaling must be enabled (default). .I max - the maximum size of the receive buffer used by each TCP socket. This value does not override the global -.BR net.core.rmem_max . +.IR net.core.rmem_max . This is not used to limit the size of the receive buffer declared using .B SO_RCVBUF @@ -519,10 +519,10 @@ It is not recommended as a tuning mechanism for heavily loaded servers to help with overloaded or misconfigured conditions. For recommended alternatives see -.BR tcp_max_syn_backlog , -.BR tcp_synack_retries , +.IR tcp_max_syn_backlog , +.IR tcp_synack_retries , and -.BR tcp_abort_on_overflow . +.IR tcp_abort_on_overflow . .TP .BR tcp_syn_retries " (integer; default: 5)" The maximum number of times initial SYNs for an active TCP @@ -560,7 +560,7 @@ windows are desired, applications can increase the size of their socket buffers and the window scaling option will be employed. If -.BR tcp_window_scaling +.IR tcp_window_scaling is disabled, TCP will not negotiate the use of window scaling with the other end during connection setup. .\" @@ -592,7 +592,7 @@ account the bandwidth used at the time congestion is experienced. TCP Westwood+ significantly increases fairness with respect to TCP Reno in wired networks and throughput over wireless links. .TP -.BR tcp_wmem +.IR tcp_wmem This is a vector of 3 integers: [min, default, max]. These parameters are used by TCP to regulate send buffer sizes. TCP dynamically adjusts the size of the send buffer from the @@ -614,20 +614,20 @@ on a socket. - the default size of the send buffer for a TCP socket. This value overwrites the initial default buffer size from the generic global -.B net.core.wmem_default +.I net.core.wmem_default defined for all protocols. The default value is 16K bytes. If larger send buffer sizes are desired, this value should be increased (to affect all sockets). To employ large TCP windows, the sysctl variable -.B net.ipv4.tcp_window_scaling +.I net.ipv4.tcp_window_scaling must be enabled (default). .I max - the maximum size of the send buffer used by each TCP socket. This value does not override the global -.BR net.core.wmem_max . +.IR net.core.wmem_max . This is not used to limit the size of the send buffer declared using .B SO_SNDBUF @@ -706,7 +706,7 @@ portable. .B TCP_LINGER2 The lifetime of orphaned FIN_WAIT2 state sockets. This option can be used to override the system wide sysctl -.B tcp_fin_timeout +.I tcp_fin_timeout on this socket. This is not to be confused with the .BR socket (7) @@ -891,7 +891,7 @@ pointer field by default. This violates RFC\ 1122, but is required for interoperability with other stacks. It can be changed by the -.B tcp_stdurg +.I tcp_stdurg sysctl. .SH BUGS Not all errors are documented. |