diff --git a/LDP/howto/docbook/Adv-Routing-HOWTO.sgml b/LDP/howto/docbook/Adv-Routing-HOWTO.sgml index 568946bd..d07b263a 100644 --- a/LDP/howto/docbook/Adv-Routing-HOWTO.sgml +++ b/LDP/howto/docbook/Adv-Routing-HOWTO.sgml @@ -915,7 +915,7 @@ that connect a local network (or even a single machine) to the big Internet. There are usually two questions given this setup. - + Split access The first is how to route answers to packets coming in over a @@ -986,6 +986,7 @@ There are usually two questions given this setup. or you are going to want to masquerade to one of the two providers. In both cases you will want to add rules selecting which provider to route out from based on the IP address of the machine in the local network. + Load balancing @@ -1980,6 +1981,7 @@ ICMP + txqueuelen @@ -1993,6 +1995,7 @@ execute: ifconfig eth0 txqueuelen 10 You can't set this parameter with tc! + @@ -2100,6 +2103,7 @@ latency parameter, which specifies the maximum amount of time a packet can sit in the TBF. The latter calculation takes into account the size of the bucket, the rate and possibly the peakrate (if set). + burst/buffer/maxburst @@ -2115,6 +2119,7 @@ rates require a larger buffer. For 10mbit/s on Intel, you need at least If your buffer is too small, packets may be dropped because more tokens arrive per timer tick than fit in your bucket. + mpu @@ -2124,6 +2129,7 @@ A zero-sized packet does not use zero bandwidth. For ethernet, no packet uses less than 64 bytes. The Minimum Packet Unit determines the minimal token usage for a packet. + rate @@ -2131,6 +2137,7 @@ token usage for a packet. The speedknob. See remarks above about limits! + @@ -2163,6 +2170,7 @@ our waits so we send just at peakrate. However, due to de default 10ms timer resolution of Unix, with 10.000 bits average packets, we are limited to 1mbit/s of peakrate! + mtu/minburst @@ -2182,6 +2190,7 @@ all. To calculate the maximum possible peakrate, multiply the configured mtu by 100 (or more correctly, HZ, which is 100 on Intel, 1024 on Alpha). + @@ -2287,6 +2296,7 @@ The SFQ is pretty much self tuning: Reconfigure hashing once this many seconds. If unset, hash will never be reconfigured. Not recommended. 10 seconds is probably a good value. + quantum @@ -2296,6 +2306,7 @@ Amount of bytes a stream is allowed to dequeue before the next queue gets a turn. Defaults to 1 maximum sized packet (MTU-sized). Do not set below the MTU! + @@ -2434,6 +2445,7 @@ Read it for the strict definitions of the terms used. An algorithm that manages the queue of a device, either incoming (ingress) or outgoing (egress). + Classless qdisc @@ -2441,6 +2453,7 @@ or outgoing (egress). A qdisc with no configurable internal subdivisions. + Classful qdisc @@ -2453,6 +2466,7 @@ bands which are, in fact, classes. However, from the user's configuration perspective, it is classless as the classes can't be touched with the tc tool. + Classes @@ -2461,6 +2475,7 @@ tool. A classful qdisc may have many classes, which each are internal to the qdisc. Each of these classes may contain a real qdisc. + Classifier @@ -2469,6 +2484,7 @@ qdisc. Each of these classes may contain a real qdisc. Each classful qdisc needs to determine to which class it needs to send a packet. This is done using the classifier. + Filter @@ -2477,6 +2493,7 @@ packet. This is done using the classifier. Classification can be performed using filters. A filter contains a number of conditions which if matched, make the filter match. + Scheduling @@ -2487,6 +2504,7 @@ go out earlier than others. This process is called Scheduling, and is performed for example by the pfifo_fast qdisc mentioned earlier. Scheduling is also called 'reordering', but this is confusing. + Shaping @@ -2496,6 +2514,7 @@ The process of delaying packets before they go out to make traffic confirm to a configured maximum rate. Shaping is performed on egress. Colloquially, dropping packets to slow traffic down is also often called Shaping. + Policing @@ -2505,6 +2524,7 @@ Delaying or dropping packets in order to make traffic stay below a configured bandwidth. In Linux, policing can only drop a packet and not delay it - there is no 'ingress queue'. + Work-Conserving @@ -2514,6 +2534,7 @@ A work-conserving qdisc always delivers a packet if one is available. In other words, it never delays a packet if the network adaptor is ready to send one (in the case of an egress qdisc). + non-Work-Conserving @@ -2524,6 +2545,7 @@ to a packet for a certain time in order to limit the bandwidth. This means that they sometimes refuse to give up a packet, even though they have one available. + @@ -2811,6 +2833,7 @@ The following parameters are recognized by tc: Number of bands to create. Each band is in fact a class. If you change this number, you must also change: + priomap @@ -2824,6 +2847,7 @@ at the TC_PRIO priority to decide how to enqueue traffic. This works just like with the pfifo_fast qdisc mentioned earlier, see there for lots of detail. + The bands are classes, and are called major:1 to major:3 by default, so if @@ -3051,6 +3075,7 @@ These are parameters you can specify in order to configure shaping: Average size of a packet, measured in bytes. Needed for calculating maxidle, which is derived from maxburst, which is specified in packets. + bandwidth @@ -3059,6 +3084,7 @@ which is derived from maxburst, which is specified in packets. The physical bandwidth of your device, needed for idle time calculations. + cell @@ -3069,6 +3095,7 @@ based on the packet size. An 800 and an 806 size packet may take just as long to send, for example - this sets the granularity. Most often set to '8'. Must be an integral power of two. + maxburst @@ -3079,6 +3106,7 @@ at maxidle, this number of average packets can be burst before avgidle drops to 0. Set it higher to be more tolerant of bursts. You can't set maxidle directly, only via this parameter. + minburst @@ -3096,6 +3124,7 @@ The time to wait is called the offtime. Higher values of minburst lead to more accurate shaping in the long term, but to bigger bursts at millisecond timescales. + minidle @@ -3111,6 +3140,7 @@ it gets too low. Minidle is specified in negative microseconds, so 10 means that avgidle is capped at -10us. + mpu @@ -3120,6 +3150,7 @@ Minimum packet size - needed because even a zero size packet is padded to 64 bytes on ethernet, and so takes a certain time to transmit. CBQ needs to know this to accurately calculate the idle time. + rate @@ -3127,6 +3158,7 @@ to know this to accurately calculate the idle time. Desired rate of traffic leaving this qdisc - this is the 'speed knob'! + @@ -3176,6 +3208,7 @@ the 'priority' parameter. Each time a class gets its turn, it can only send out a limited amount of data. 'Allot' is the base unit of this amount. See the 'weight' parameter for more information. + prio @@ -3185,6 +3218,7 @@ The CBQ can also act like the PRIO device. Inner classes with lower priority are tried first and as long as they have traffic, other classes are not polled for traffic. + weight @@ -3203,6 +3237,7 @@ using 'rate/10' as a rule of thumb and it appears to work well. The renormalized weight is multiplied by the 'allot' parameter to determine how much data can be sent in one round. + @@ -3239,6 +3274,7 @@ agencies on your link who do want to give each other freebies. The control program tc also knows about 'sharing', which is the reverse of 'isolated'. + bounded/borrow @@ -3248,6 +3284,7 @@ A class can also be 'bounded', which means that it will not try to borrow bandwidth from sibling classes. tc also knows about 'borrow', which is the reverse of 'bounded'. + A typical situation might be where you have two agencies on your link which @@ -3750,6 +3787,7 @@ packet. Source mask 'match ip src 1.2.3.0/24', destination mask 'match ip dst 4.3.2.0/24'. To match a single host, use /32, or omit the mask. + On source/destination port, all IP protocols @@ -3757,6 +3795,7 @@ Source mask 'match ip src 1.2.3.0/24', destination mask 'match ip dst Source: 'match ip sport 80 0xffff', 'match ip dport 0xffff' + On ip protocol (tcp, udp, icmp, gre, ipsec) @@ -3765,6 +3804,7 @@ Source: 'match ip sport 80 0xffff', 'match ip dport 0xffff' Use the numbers from /etc/protocols, for example, icmp is 1: 'match ip protocol 1 0xff'. + On fwmark @@ -3791,6 +3831,7 @@ The number 6 is arbitrary. If you don't want to understand the full tc filter syntax, just use iptables, and only learn to select on fwmark. + On the TOS field @@ -3806,6 +3847,7 @@ To select interactive, minimum delay traffic: Use 0x08 0xff for bulk traffic. + @@ -4240,6 +4282,7 @@ Bases the decision on how the firewall has marked the packet. This can be the easy way out if you don't want to learn tc filter syntax. See the Queueing chapter for details. + u32 @@ -4247,6 +4290,7 @@ Queueing chapter for details. Bases the decision on fields within the packet (i.e. source IP address, etc) + route @@ -4254,6 +4298,7 @@ Bases the decision on fields within the packet (i.e. source IP address, etc) Bases the decision on which route the packet will be routed by + rsvp, rsvp6 @@ -4265,6 +4310,7 @@ URL="http://www.isi.edu/div7/rsvp/overview.html" >. Only useful on networks you control - the Internet does not respect RSVP. + tcindex @@ -4272,6 +4318,7 @@ on networks you control - the Internet does not respect RSVP. Used in the DSMARK qdisc, see the relevant section. + @@ -4297,6 +4344,7 @@ here for convenience: The protocol this classifier will accept. Generally you will only be accepting only IP traffic. Required. + parent @@ -4305,6 +4353,7 @@ accepting only IP traffic. Required. The handle this classifier is to be attached to. This handle must be an already existing class. Required. + prio @@ -4312,6 +4361,7 @@ an already existing class. Required. The priority of this classifier. Lower numbers get tested first. + handle @@ -4319,6 +4369,7 @@ The priority of this classifier. Lower numbers get tested first. This handle means different things to different filters. + @@ -4828,6 +4879,7 @@ Currently, three actions are available: Causes this filter not to match, but perhaps other filters will. + drop @@ -4839,6 +4891,7 @@ For example, you may have a name server that falls over if offered more than 5mbit/s of packets, in which case an ingress filter could be used to make sure no more is ever offered. + Pass/OK @@ -4847,6 +4900,7 @@ sure no more is ever offered. Pass on traffic ok. Might be used to disable a complicated filter, but leave it in place. + reclassify @@ -4855,6 +4909,7 @@ it in place. Most often comes down to reclassification to Best Effort. This is the default action. + @@ -5158,6 +5213,7 @@ Kuznetsov <kuznet@ms2.inr.ac.ru> and Andi Kleen <ak@muc.de> If the kernel decides that it can't deliver a packet, it will drop it, and send the source of the packet an ICMP notice to this effect. + /proc/sys/net/ipv4/icmp_echo_ignore_all @@ -5166,6 +5222,7 @@ send the source of the packet an ICMP notice to this effect. Don't act on echo packets at all. Please don't set this by default, but if you are used as a relay in a DoS attack, it may be useful. + /proc/sys/net/ipv4/icmp_echo_ignore_broadcasts [Useful] @@ -5175,6 +5232,7 @@ If you ping the broadcast address of a network, all hosts are supposed to respond. This makes for a dandy denial-of-service tool. Set this to 1 to ignore these broadcast messages. + /proc/sys/net/ipv4/icmp_echoreply_rate @@ -5182,6 +5240,7 @@ ignore these broadcast messages. The rate at which echo replies are sent to any one destination. + /proc/sys/net/ipv4/icmp_ignore_bogus_error_responses @@ -5190,6 +5249,7 @@ The rate at which echo replies are sent to any one destination. Set this to ignore ICMP errors caused by hosts in the network reacting badly to frames sent to what they perceive to be the broadcast address. + /proc/sys/net/ipv4/icmp_paramprob_rate @@ -5199,6 +5259,7 @@ A relatively unknown ICMP message, which is sent in response to incorrect packets with broken IP or TCP headers. With this file you can control the rate at which it is sent. + /proc/sys/net/ipv4/icmp_timeexceed_rate @@ -5207,6 +5268,7 @@ rate at which it is sent. This the famous cause of the 'Solaris middle star' in traceroutes. Limits number of ICMP Time Exceeded messages sent. + /proc/sys/net/ipv4/igmp_max_memberships @@ -5215,6 +5277,7 @@ number of ICMP Time Exceeded messages sent. Maximum number of listening igmp (multicast) sockets on the host. FIXME: Is this true? + /proc/sys/net/ipv4/inet_peer_gc_maxtime @@ -5226,6 +5289,7 @@ Minimum interval between garbage collection passes. This interval is in effect under low (or absent) memory pressure on the pool. Measured in jiffies. + /proc/sys/net/ipv4/inet_peer_gc_mintime @@ -5234,6 +5298,7 @@ jiffies. Minimum interval between garbage collection passes. This interval is in effect under high memory pressure on the pool. Measured in jiffies. + /proc/sys/net/ipv4/inet_peer_maxttl @@ -5243,6 +5308,7 @@ Maximum time-to-live of entries. Unused entries will expire after this period of time if there is no memory pressure on the pool (i.e. when the number of entries in the pool is very small). Measured in jiffies. + /proc/sys/net/ipv4/inet_peer_minttl @@ -5253,6 +5319,7 @@ time-to-live on the reassembling side. This minimum time-to-live is guaranteed if the pool size is less than inet_peer_threshold. Measured in jiffies. + /proc/sys/net/ipv4/inet_peer_threshold @@ -5263,6 +5330,7 @@ entries will be thrown aggressively. This threshold also determines entries' time-to-live and time intervals between garbage collection passes. More entries, less time-to-live, less GC interval. + /proc/sys/net/ipv4/ip_autoconfig @@ -5271,6 +5339,7 @@ More entries, less time-to-live, less GC interval. This file contains the number one if the host received its IP configuration by RARP, BOOTP, DHCP or a similar mechanism. Otherwise it is zero. + /proc/sys/net/ipv4/ip_default_ttl @@ -5280,6 +5349,7 @@ Time To Live of packets. Set to a safe 64. Raise it if you have a huge network. Don't do so for fun - routing loops cause much more damage that way. You might even consider lowering it in some circumstances. + /proc/sys/net/ipv4/ip_dynaddr @@ -5290,6 +5360,7 @@ address. Once your demand interface comes up, any local TCP sockets which haven' connection that brings up your interface itself does not work, but the second try does. + /proc/sys/net/ipv4/ip_forward @@ -5297,6 +5368,7 @@ second try does. If the kernel should attempt to forward packets. Off by default. + /proc/sys/net/ipv4/ip_local_port_range @@ -5305,6 +5377,7 @@ If the kernel should attempt to forward packets. Off by default. Range of local ports for outgoing connections. Actually quite small by default, 1024 to 4999. + /proc/sys/net/ipv4/ip_no_pmtu_disc @@ -5315,6 +5388,7 @@ determine the largest Maximum Transfer Unit possible on your path. See also the section on Path MTU discovery in the chapter. + /proc/sys/net/ipv4/ipfrag_high_thresh @@ -5325,6 +5399,7 @@ ipfrag_high_thresh bytes of memory is allocated for this purpose, the fragment handler will toss packets until ipfrag_low_thresh is reached. + /proc/sys/net/ipv4/ip_nonlocal_bind @@ -5335,6 +5410,7 @@ which doesn't belong to a device on your system. This can be useful when your machine is on a non-permanent (or even dynamic) link, so your services are able to start up and bind to a specific address when your link is down. + /proc/sys/net/ipv4/ipfrag_low_thresh @@ -5342,6 +5418,7 @@ are able to start up and bind to a specific address when your link is down. Minimum memory used to reassemble IP fragments. + /proc/sys/net/ipv4/ipfrag_time @@ -5349,6 +5426,7 @@ Minimum memory used to reassemble IP fragments. Time in seconds to keep an IP fragment in memory. + /proc/sys/net/ipv4/tcp_abort_on_overflow @@ -5358,6 +5436,7 @@ A boolean flag controlling the behaviour under lots of incoming connections. When enabled, this causes the kernel to actively send RST packets when a service is overloaded. + /proc/sys/net/ipv4/tcp_fin_timeout @@ -5371,6 +5450,7 @@ to overflow memory with kilotons of dead sockets, FIN-WAIT-2 sockets are less dangerous than FIN-WAIT-1, because they eat maximum 1.5K of memory, but they tend to live longer. Cf. tcp_max_orphans. + /proc/sys/net/ipv4/tcp_keepalive_time @@ -5380,6 +5460,7 @@ How often TCP sends out keepalive messages when keepalive is enabled. Default: 2hours. + /proc/sys/net/ipv4/tcp_keepalive_intvl @@ -5389,6 +5470,7 @@ How frequent probes are retransmitted, when a probe isn't acknowledged. Default: 75 seconds. + /proc/sys/net/ipv4/tcp_keepalive_probes @@ -5402,6 +5484,7 @@ Default value: 9. Multiplied with tcp_keepalive_intvl, this gives the time a link can be non-responsive after a keepalive has been sent. + /proc/sys/net/ipv4/tcp_max_orphans @@ -5416,6 +5499,7 @@ network conditions require more than default value, and tune network services to linger and kill such states more aggressively. Let me remind you again: each orphan eats up to  64K of unswappable memory. + /proc/sys/net/ipv4/tcp_orphan_retries @@ -5426,6 +5510,7 @@ Default value 7 corresponds to  50sec-16min depending on RTO. If your machi is a loaded WEB server, you should think about lowering this value, such sockets may consume significant resources. Cf. tcp_max_orphans. + /proc/sys/net/ipv4/tcp_max_syn_backlog @@ -5439,6 +5524,7 @@ make it greater than 1024, it would be better to change TCP_SYNQ_HSIZE in include/net/tcp.h to keep TCP_SYNQ_HSIZE*16<=tcp_max_syn_backlog and to recompile kernel. + /proc/sys/net/ipv4/tcp_max_tw_buckets @@ -5451,6 +5537,7 @@ not lower the limit artificially, but rather increase it (probably, after increasing installed memory), if network conditions require more than default value. + /proc/sys/net/ipv4/tcp_retrans_collapse @@ -5460,6 +5547,7 @@ Bug-to-bug compatibility with some broken printers. On retransmit try to send bigger packets to work around bugs in certain TCP stacks. + /proc/sys/net/ipv4/tcp_retries1 @@ -5470,6 +5558,7 @@ and it is necessary to report this suspicion to network layer. Minimal RFC value is 3, it is default, which corresponds to  3sec-8min depending on RTO. + /proc/sys/net/ipv4/tcp_retries2 @@ -5484,6 +5573,7 @@ says that the limit should be longer than 100 sec. It is too small number. Default value 15 corresponds to  13-30min depending on RTO. + /proc/sys/net/ipv4/tcp_rfc1337 @@ -5495,6 +5585,7 @@ sockets in the time-wait state. Default: 0 + /proc/sys/net/ipv4/tcp_sack @@ -5503,6 +5594,7 @@ Default: 0 Use Selective ACK which can be used to signify that specific packets are missing - therefore helping fast recovery. + /proc/sys/net/ipv4/tcp_stdurg @@ -5516,6 +5608,7 @@ Linux might not communicate correctly with them. Default: FALSE + /proc/sys/net/ipv4/tcp_syn_retries @@ -5524,6 +5617,7 @@ Default: FALSE Number of SYN packets the kernel will send before giving up on the new connection. + /proc/sys/net/ipv4/tcp_synack_retries @@ -5534,6 +5628,7 @@ piggybacked ACK on it, to acknowledge the earlier received SYN. This is part 2 of the threeway handshake. This setting determines the number of SYN+ACK packets sent before the kernel gives up on the connection. + /proc/sys/net/ipv4/tcp_timestamps @@ -5544,6 +5639,7 @@ sequence numbers. A 1 gigabit link might conceivably re-encounter a previous sequence number with an out-of-line value, because it was of a previous generation. The timestamp will let it recognize this 'ancient packet'. + /proc/sys/net/ipv4/tcp_tw_recycle @@ -5552,6 +5648,7 @@ generation. The timestamp will let it recognize this 'ancient packet'. Enable fast recycling TIME-WAIT sockets. Default value is 1. It should not be changed without advice/request of technical experts. + /proc/sys/net/ipv4/tcp_window_scaling @@ -5561,6 +5658,7 @@ TCP/IP normally allows windows up to 65535 bytes big. For really fast networks, this may not be enough. The window scaling options allows for almost gigabyte windows, which is good for high bandwidth*delay products. + @@ -5584,6 +5682,7 @@ to resend your packet on the same interface), it will send us a ICMP Redirect. This is a slight security risk however, so you may want to turn it off, or use secure redirects. + /proc/sys/net/ipv4/conf/DEV/accept_source_route @@ -5593,6 +5692,7 @@ Not used very much anymore. You used to be able to give a packet a list of IP addresses it should visit on its way. Linux can be made to honor this IP option. + /proc/sys/net/ipv4/conf/DEV/bootp_relay @@ -5607,6 +5707,7 @@ such packets. The default is 0, since this feature is not implemented yet (kernel version 2.2.12). + /proc/sys/net/ipv4/conf/DEV/forwarding @@ -5614,6 +5715,7 @@ The default is 0, since this feature is not implemented yet (kernel version Enable or disable IP forwarding on this interface. + /proc/sys/net/ipv4/conf/DEV/log_martians @@ -5622,6 +5724,7 @@ Enable or disable IP forwarding on this interface. See the section on . + /proc/sys/net/ipv4/conf/DEV/mc_forwarding @@ -5629,6 +5732,7 @@ See the section on If we do multicast forwarding on this interface + /proc/sys/net/ipv4/conf/DEV/proxy_arp @@ -5640,6 +5744,7 @@ pseudo bridges'. Do take care that your netmasks are very correct before enabling this! Also be aware that the rp_filter, mentioned elsewhere, also operates on ARP queries! + /proc/sys/net/ipv4/conf/DEV/rp_filter @@ -5648,6 +5753,7 @@ operates on ARP queries! See the section on . + /proc/sys/net/ipv4/conf/DEV/secure_redirects @@ -5656,6 +5762,7 @@ See the section on Accept ICMP redirect messages only for gateways, listed in default gateway list. Enabled by default. + /proc/sys/net/ipv4/conf/DEV/send_redirects @@ -5663,6 +5770,7 @@ list. Enabled by default. If we send the above mentioned redirects. + /proc/sys/net/ipv4/conf/DEV/shared_media @@ -5671,6 +5779,7 @@ If we send the above mentioned redirects. If it is not set the kernel does not assume that different subnets on this device can communicate directly. Default setting is 'yes'. + /proc/sys/net/ipv4/conf/DEV/tag @@ -5678,6 +5787,7 @@ device can communicate directly. Default setting is 'yes'. FIXME: fill this in + @@ -5700,6 +5810,7 @@ Maximum for random delay of answers to neighbor solicitation messages in jiffies (1/100 sec). Not yet implemented (Linux does not have anycast support yet). + /proc/sys/net/ipv4/neigh/DEV/app_solicit @@ -5708,6 +5819,7 @@ yet). Determines the number of requests to send to the user level ARP daemon. Use 0 to turn off. + /proc/sys/net/ipv4/neigh/DEV/base_reachable_time @@ -5716,6 +5828,7 @@ to turn off. A base value used for computing the random reachable time value as specified in RFC2461. + /proc/sys/net/ipv4/neigh/DEV/delay_first_probe_time @@ -5724,6 +5837,7 @@ in RFC2461. Delay for the first time probe if the neighbor is reachable. (see gc_stale_time) + /proc/sys/net/ipv4/neigh/DEV/gc_stale_time @@ -5735,6 +5849,7 @@ to another machine). When ucast_solicit is greater than 0 it first tries to send an ARP packet directly to the known host When that fails and mcast_solicit is greater than 0, an ARP request is broadcast. + /proc/sys/net/ipv4/neigh/DEV/locktime @@ -5743,6 +5858,7 @@ mcast_solicit is greater than 0, an ARP request is broadcast. An ARP/neighbor entry is only replaced with a new one if the old is at least locktime old. This prevents ARP cache thrashing. + /proc/sys/net/ipv4/neigh/DEV/mcast_solicit @@ -5750,6 +5866,7 @@ locktime old. This prevents ARP cache thrashing. Maximum number of retries for multicast solicitation. + /proc/sys/net/ipv4/neigh/DEV/proxy_delay @@ -5759,6 +5876,7 @@ Maximum time (real time is random [0..proxytime]) before answering to an request for which we have an proxy ARP entry. In some cases, this is used to prevent network flooding. + /proc/sys/net/ipv4/neigh/DEV/proxy_qlen @@ -5766,6 +5884,7 @@ prevent network flooding. Maximum queue length of the delayed proxy arp timer. (see proxy_delay). + /proc/sys/net/ipv4/neigh/DEV/retrans_time @@ -5775,6 +5894,7 @@ The time, expressed in jiffies (1/100 sec), between retransmitted Neighbor Solicitation messages. Used for address resolution and to determine if a neighbor is unreachable. + /proc/sys/net/ipv4/neigh/DEV/ucast_solicit @@ -5782,6 +5902,7 @@ neighbor is unreachable. Maximum number of retries for unicast solicitation. + /proc/sys/net/ipv4/neigh/DEV/unres_qlen @@ -5790,6 +5911,7 @@ Maximum number of retries for unicast solicitation. Maximum queue length for a pending arp request - the number of packets which are accepted from other layers while the ARP address is still resolved. + Internet QoS: Architectures and Mechanisms for Quality of Service, @@ -5799,6 +5921,7 @@ Zheng Wang, ISBN 1-55860-608-4 Hardcover textbook covering topics related to Quality of Service. Good for understanding basic concepts. + @@ -5820,6 +5943,7 @@ log from the routing code. The higher the error_cost factor is, the fewer messages will be written. Error_burst controls when messages will be dropped. The default settings limit warning messages to one every five seconds. + /proc/sys/net/ipv4/route/error_cost @@ -5830,6 +5954,7 @@ log from the routing code. The higher the error_cost factor is, the fewer messages will be written. Error_burst controls when messages will be dropped. The default settings limit warning messages to one every five seconds. + /proc/sys/net/ipv4/route/flush @@ -5837,6 +5962,7 @@ The default settings limit warning messages to one every five seconds. Writing to this file results in a flush of the routing cache. + /proc/sys/net/ipv4/route/gc_elasticity @@ -5855,6 +5981,7 @@ URL="http://mailman.ds9a.nl/pipermail/lartc/2002q1/002667.html" >this post by Ard van Breemen. + /proc/sys/net/ipv4/route/gc_interval @@ -5862,6 +5989,7 @@ URL="http://mailman.ds9a.nl/pipermail/lartc/2002q1/002667.html" See /proc/sys/net/ipv4/route/gc_elasticity. + /proc/sys/net/ipv4/route/gc_min_interval @@ -5869,6 +5997,7 @@ See /proc/sys/net/ipv4/route/gc_elasticity. See /proc/sys/net/ipv4/route/gc_elasticity. + /proc/sys/net/ipv4/route/gc_thresh @@ -5876,6 +6005,7 @@ See /proc/sys/net/ipv4/route/gc_elasticity. See /proc/sys/net/ipv4/route/gc_elasticity. + /proc/sys/net/ipv4/route/gc_timeout @@ -5883,6 +6013,7 @@ See /proc/sys/net/ipv4/route/gc_elasticity. See /proc/sys/net/ipv4/route/gc_elasticity. + /proc/sys/net/ipv4/route/max_delay @@ -5890,6 +6021,7 @@ See /proc/sys/net/ipv4/route/gc_elasticity. Delays for flushing the routing cache. + /proc/sys/net/ipv4/route/max_size @@ -5898,6 +6030,7 @@ Delays for flushing the routing cache. Maximum size of the routing cache. Old entries will be purged once the cache reached has this size. + /proc/sys/net/ipv4/route/min_adv_mss @@ -5905,6 +6038,7 @@ reached has this size. FIXME: fill this in + /proc/sys/net/ipv4/route/min_delay @@ -5912,6 +6046,7 @@ FIXME: fill this in Delays for flushing the routing cache. + /proc/sys/net/ipv4/route/min_pmtu @@ -5919,6 +6054,7 @@ Delays for flushing the routing cache. FIXME: fill this in + /proc/sys/net/ipv4/route/mtu_expires @@ -5926,6 +6062,7 @@ FIXME: fill this in FIXME: fill this in + /proc/sys/net/ipv4/route/redirect_load @@ -5935,6 +6072,7 @@ Factors which determine if more ICMP redirects should be sent to a specific host. No redirects will be sent once the load limit or the maximum number of redirects has been reached. + /proc/sys/net/ipv4/route/redirect_number @@ -5942,6 +6080,7 @@ redirects has been reached. See /proc/sys/net/ipv4/route/redirect_load. + /proc/sys/net/ipv4/route/redirect_silence @@ -5950,6 +6089,7 @@ See /proc/sys/net/ipv4/route/redirect_load. Timeout for redirects. After this period redirects will be sent again, even if this has been stopped, because the load or number limit has been reached. + @@ -5997,6 +6137,7 @@ Specifies the length of the queue. Measured in bytes for bfifo, in packets for pfifo. Defaults to the interface txqueuelen (see pfifo_fast chapter) packets long or txqueuelen*mtu bytes for bfifo. + @@ -6650,8 +6791,7 @@ URL="http://linux-atm.sourceforge.net/" This qdisc is not included in the standard kernels but can be downloaded from �. +/>. Currently the qdisc is only tested with Linux 2.2 kernels but it will probably work with 2.4/2.5 kernels too. @@ -7070,6 +7210,7 @@ BUT This would put additional load on the router and some commercial routers might not even support this. + Using a Layer 4 switch. @@ -7087,6 +7228,7 @@ The cost for this equipment is usually very high. Typical layer 4 switch would normally cost more than a typical router+good linux server. + Using cache server as network's gateway. @@ -7107,6 +7249,7 @@ or the server itself might crash and no one on the network will be able to access the Internet if that occurs. + Linux+NetFilter router. @@ -7117,6 +7260,7 @@ which is using NetFilter for "mark"ing the packets with destination port 80 and using iproute2 to route the "mark"ed packets to the Squid server. + @@ -7255,6 +7399,7 @@ naret# ip route add default via 10.0.0.2 dev eth0 table www.out naret# ip route flush cache + @@ -7581,6 +7726,7 @@ This means that downloading or uploading files should not disturb SSH or even telnet. These are the most important things, even 200ms latency is sluggish to work over. + Allow 'surfing' at reasonable speeds while up or downloading @@ -7589,6 +7735,7 @@ sluggish to work over. Even though http is 'bulk' traffic, other traffic should not drown it out too much. + Make sure uploads don't harm downloads, and the other way around @@ -7597,6 +7744,7 @@ too much. This is a much observed phenomenon where upstream traffic simply destroys download speed. + It turns out that all this is possible, at the cost of a tiny bit of @@ -7652,6 +7800,7 @@ eliminated, and moved to your Linux router. Luckily this is possible. By limiting our upload speed to slightly less than the truly available rate, no queues are built up in our modem. The queue is now moved to Linux. + Limit download speed @@ -7663,6 +7812,7 @@ which causes TCP/IP to slow down to just the rate we want. Because we don't want to drop traffic unnecessarily, we configure a 'burst' size we allow at higher speed. + @@ -8348,6 +8498,7 @@ URL="http://scry.wanfear.com/~greear/vlan/cisco_howto.html" Update: has been included in the kernel as of 2.4.14 (perhaps 13). + Alternate 802.1Q VLAN Implementation for Linux + Linux Virtual Server + CBQ.init + Chronox easy shaping scripts + Virtual Router @@ -8523,6 +8678,7 @@ Not *one* ping packet was lost! Just after packet 4, I disconnected my P200 from the network, and my 486 took over, which you can see from the higher latency. + @@ -8544,6 +8700,7 @@ URL="http://snafu.freedom.org/linux2.2/iproute-notes.html" Contains lots of technical information, comments from the kernel + Slides by Jamal Hadi Salim, one of the authors of Linux traffic control + + + Differentiated Services on Linux @@ -8590,6 +8750,7 @@ URL="ftp://icaftp.epfl.ch/pub/linux/diffserv/misc/dsid-01.txt.gz" Kuznetsov describes DiffServ facilities in the Linux kernel, amongst which are TBF, GRED, the DSMARK qdisc and the tcindex classifier. + + + Docum experimental site + TCP/IP Illustrated, volume 1, W. Richard Stevens, ISBN 0-201-63346-9 @@ -8636,6 +8800,7 @@ of practical information, examples, tests and also points out some CBQ/tc bugs. Required reading if you truly want to understand TCP/IP. Entertaining as well. +