aboutsummaryrefslogtreecommitdiffstats
path: root/Documentation/networking/ip-sysctl.txt
diff options
context:
space:
mode:
Diffstat (limited to 'Documentation/networking/ip-sysctl.txt')
-rw-r--r--Documentation/networking/ip-sysctl.txt80
1 files changed, 69 insertions, 11 deletions
diff --git a/Documentation/networking/ip-sysctl.txt b/Documentation/networking/ip-sysctl.txt
index f350c69b2bb4..bfe924217f24 100644
--- a/Documentation/networking/ip-sysctl.txt
+++ b/Documentation/networking/ip-sysctl.txt
@@ -11,7 +11,9 @@ ip_forward - BOOLEAN
11 for routers) 11 for routers)
12 12
13ip_default_ttl - INTEGER 13ip_default_ttl - INTEGER
14 default 64 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)
15 17
16ip_no_pmtu_disc - BOOLEAN 18ip_no_pmtu_disc - BOOLEAN
17 Disable Path MTU Discovery. 19 Disable Path MTU Discovery.
@@ -20,6 +22,15 @@ ip_no_pmtu_disc - BOOLEAN
20min_pmtu - INTEGER 22min_pmtu - INTEGER
21 default 562 - minimum discovered Path MTU 23 default 562 - minimum discovered Path MTU
22 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
23mtu_expires - INTEGER 34mtu_expires - INTEGER
24 Time, in seconds, that cached PMTU information is kept. 35 Time, in seconds, that cached PMTU information is kept.
25 36
@@ -135,6 +146,7 @@ tcp_adv_win_scale - INTEGER
135 Count buffering overhead as bytes/2^tcp_adv_win_scale 146 Count buffering overhead as bytes/2^tcp_adv_win_scale
136 (if tcp_adv_win_scale > 0) or bytes-bytes/2^(-tcp_adv_win_scale), 147 (if tcp_adv_win_scale > 0) or bytes-bytes/2^(-tcp_adv_win_scale),
137 if it is <= 0. 148 if it is <= 0.
149 Possible values are [-31, 31], inclusive.
138 Default: 2 150 Default: 2
139 151
140tcp_allowed_congestion_control - STRING 152tcp_allowed_congestion_control - STRING
@@ -175,7 +187,7 @@ tcp_cookie_size - INTEGER
175tcp_dsack - BOOLEAN 187tcp_dsack - BOOLEAN
176 Allows TCP to send "duplicate" SACKs. 188 Allows TCP to send "duplicate" SACKs.
177 189
178tcp_ecn - BOOLEAN 190tcp_ecn - INTEGER
179 Enable Explicit Congestion Notification (ECN) in TCP. ECN is only 191 Enable Explicit Congestion Notification (ECN) in TCP. ECN is only
180 used when both ends of the TCP flow support it. It is useful to 192 used when both ends of the TCP flow support it. It is useful to
181 avoid losses due to congestion (when the bottleneck router supports 193 avoid losses due to congestion (when the bottleneck router supports
@@ -268,6 +280,17 @@ tcp_max_orphans - INTEGER
268 more aggressively. Let me to remind again: each orphan eats 280 more aggressively. Let me to remind again: each orphan eats
269 up to ~64K of unswappable memory. 281 up to ~64K of unswappable memory.
270 282
283tcp_max_ssthresh - INTEGER
284 Limited Slow-Start for TCP with large congestion windows (cwnd) defined in
285 RFC3742. Limited slow-start is a mechanism to limit growth of the cwnd
286 on the region where cwnd is larger than tcp_max_ssthresh. TCP increases cwnd
287 by at most tcp_max_ssthresh segments, and by at least tcp_max_ssthresh/2
288 segments per RTT when the cwnd is above tcp_max_ssthresh.
289 If TCP connection increased cwnd to thousands (or tens of thousands) segments,
290 and thousands of packets were being dropped during slow-start, you can set
291 tcp_max_ssthresh to improve performance for new TCP connection.
292 Default: 0 (off)
293
271tcp_max_syn_backlog - INTEGER 294tcp_max_syn_backlog - INTEGER
272 Maximal number of remembered connection requests, which are 295 Maximal number of remembered connection requests, which are
273 still did not receive an acknowledgment from connecting client. 296 still did not receive an acknowledgment from connecting client.
@@ -323,7 +346,7 @@ tcp_orphan_retries - INTEGER
323 when RTO retransmissions remain unacknowledged. 346 when RTO retransmissions remain unacknowledged.
324 See tcp_retries2 for more details. 347 See tcp_retries2 for more details.
325 348
326 The default value is 7. 349 The default value is 8.
327 If your machine is a loaded WEB server, 350 If your machine is a loaded WEB server,
328 you should think about lowering this value, such sockets 351 you should think about lowering this value, such sockets
329 may consume significant resources. Cf. tcp_max_orphans. 352 may consume significant resources. Cf. tcp_max_orphans.
@@ -698,10 +721,28 @@ igmp_max_memberships - INTEGER
698 Change the maximum number of multicast groups we can subscribe to. 721 Change the maximum number of multicast groups we can subscribe to.
699 Default: 20 722 Default: 20
700 723
701conf/interface/* changes special settings per interface (where "interface" is 724 Theoretical maximum value is bounded by having to send a membership
702 the name of your network interface) 725 report in a single datagram (i.e. the report can't span multiple
703conf/all/* is special, changes the settings for all interfaces 726 datagrams, or risk confusing the switch and leaving groups you don't
727 intend to).
728
729 The number of supported groups 'M' is bounded by the number of group
730 report entries you can fit into a single datagram of 65535 bytes.
731
732 M = 65536-sizeof (ip header)/(sizeof(Group record))
733
734 Group records are variable length, with a minimum of 12 bytes.
735 So net.ipv4.igmp_max_memberships should not be set higher than:
736
737 (65536-24) / 12 = 5459
738
739 The value 5459 assumes no IP header options, so in practice
740 this number may be lower.
704 741
742 conf/interface/* changes special settings per interface (where
743 "interface" is the name of your network interface)
744
745 conf/all/* is special, changes the settings for all interfaces
705 746
706log_martians - BOOLEAN 747log_martians - BOOLEAN
707 Log packets with impossible addresses to kernel log. 748 Log packets with impossible addresses to kernel log.
@@ -1014,6 +1055,12 @@ conf/interface/*:
1014accept_ra - BOOLEAN 1055accept_ra - BOOLEAN
1015 Accept Router Advertisements; autoconfigure using them. 1056 Accept Router Advertisements; autoconfigure using them.
1016 1057
1058 Possible values are:
1059 0 Do not accept Router Advertisements.
1060 1 Accept Router Advertisements if forwarding is disabled.
1061 2 Overrule forwarding behaviour. Accept Router Advertisements
1062 even if forwarding is enabled.
1063
1017 Functional default: enabled if local forwarding is disabled. 1064 Functional default: enabled if local forwarding is disabled.
1018 disabled if local forwarding is enabled. 1065 disabled if local forwarding is enabled.
1019 1066
@@ -1075,7 +1122,12 @@ forwarding - BOOLEAN
1075 Note: It is recommended to have the same setting on all 1122 Note: It is recommended to have the same setting on all
1076 interfaces; mixed router/host scenarios are rather uncommon. 1123 interfaces; mixed router/host scenarios are rather uncommon.
1077 1124
1078 FALSE: 1125 Possible values are:
1126 0 Forwarding disabled
1127 1 Forwarding enabled
1128 2 Forwarding enabled (Hybrid Mode)
1129
1130 FALSE (0):
1079 1131
1080 By default, Host behaviour is assumed. This means: 1132 By default, Host behaviour is assumed. This means:
1081 1133
@@ -1085,18 +1137,24 @@ forwarding - BOOLEAN
1085 Advertisements (and do autoconfiguration). 1137 Advertisements (and do autoconfiguration).
1086 4. If accept_redirects is TRUE (default), accept Redirects. 1138 4. If accept_redirects is TRUE (default), accept Redirects.
1087 1139
1088 TRUE: 1140 TRUE (1):
1089 1141
1090 If local forwarding is enabled, Router behaviour is assumed. 1142 If local forwarding is enabled, Router behaviour is assumed.
1091 This means exactly the reverse from the above: 1143 This means exactly the reverse from the above:
1092 1144
1093 1. IsRouter flag is set in Neighbour Advertisements. 1145 1. IsRouter flag is set in Neighbour Advertisements.
1094 2. Router Solicitations are not sent. 1146 2. Router Solicitations are not sent.
1095 3. Router Advertisements are ignored. 1147 3. Router Advertisements are ignored unless accept_ra is 2.
1096 4. Redirects are ignored. 1148 4. Redirects are ignored.
1097 1149
1098 Default: FALSE if global forwarding is disabled (default), 1150 TRUE (2):
1099 otherwise TRUE. 1151
1152 Hybrid mode. Same behaviour as TRUE, except for:
1153
1154 2. Router Solicitations are being sent when necessary.
1155
1156 Default: 0 (disabled) if global forwarding is disabled (default),
1157 otherwise 1 (enabled).
1100 1158
1101hop_limit - INTEGER 1159hop_limit - INTEGER
1102 Default Hop Limit to set. 1160 Default Hop Limit to set.