BertHubert Netherlabs BV

bert.hubert@netherlabs.nl

Thomas Graf (Section Author)

tgraf%suug.ch

Gregory Maxwell (Section Author) Remco van Mook (Section Author)

remco@virtu.nl

Martijn van Oosterhout (Section Author)

kleptog@cupid.suninternet.com

Paul B Schroeder (Section Author)

paulsch@us.ibm.com

Jasper Spaans (Section Author)

jasper@spaans.ds9a.nl

Pedro Larroy (Section Author)

piotr%member.fsf.org

$Revision: 1.43 $ $Date: 2003/10/29 12:33:38 $ DocBook Edition A very hands-on approach to iproute2, traffic shaping and a bit of netfilter. This document is dedicated to lots of people, and is my attempt to do something back. To list but a few: Rusty Russell Alexey N. Kuznetsov The good folks from Google The staff of Casema Internet Welcome, gentle reader. This document hopes to enlighten you on how to do more with Linux 2.2/2.4 routing. Unbeknownst to most users, you already run tools which allow you to do spectacular things. Commands like route and ifconfig are actually very thin wrappers for the very powerful iproute2 infrastructure. I hope that this HOWTO will become as readable as the ones by Rusty Russell of (amongst other things) netfilter fame. You can always reach us by writing to the HOWTO team. However, please consider posting to the mailing list (see the relevant section) if you have questions which are not directly related to this HOWTO. We are no free helpdesk, but we often will answer questions asked on the list. Before losing your way in this HOWTO, if all you want to do is simple traffic shaping, skip everything and head to the chapter, and read about CBQ.init. This document is distributed in the hope that it will be useful, but WITHOUT ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. In short, if your STM-64 backbone breaks down and distributes pornography to your most esteemed customers - it's never our fault. Sorry. Copyright (c) 2002 by bert hubert, Gregory Maxwell, Martijn van Oosterhout, Remco van Mook, Paul B. Schroeder and others. This material may be distributed only subject to the terms and conditions set forth in the Open Publication License, v1.0 or later (the latest version is presently available at http://www.opencontent.org/openpub/). Please freely copy and distribute (sell or give away) this document in any format. It's requested that corrections and/or comments be forwarded to the document maintainer. It is also requested that if you publish this HOWTO in hardcopy that you send the authors some samples for review purposes :-) As the title implies, this is the Advanced HOWTO. While by no means rocket science, some prior knowledge is assumed. Here are some other references which might help teach you more: Rusty Russell's networking-concepts-HOWTO Very nice introduction, explaining what a network is, and how it is connected to other networks. Linux Networking-HOWTO (Previously the Net-3 HOWTO) Great stuff, although very verbose. It teaches you a lot of stuff that's already configured if you are able to connect to the Internet. Should be located in /usr/doc/HOWTO/NET3-4-HOWTO.txt but can be also be found online. A small list of things that are possible: Throttle bandwidth for certain computers Throttle bandwidth TO certain computers Help you to fairly share your bandwidth Protect your network from DoS attacks Protect the Internet from your customers Multiplex several servers as one, for load balancing or enhanced availability Restrict access to your computers Limit access of your users to other hosts Do routing based on user id (yes!), MAC address, source IP address, port, type of service, time of day or content Currently, not many people are using these advanced features. This is for several reasons. While the provided documentation is verbose, it is not very hands-on. Traffic control is almost undocumented. There are several things which should be noted about this document. While I wrote most of it, I really don't want it to stay that way. I am a strong believer in Open Source, so I encourage you to send feedback, updates, patches etcetera. Do not hesitate to inform me of typos or plain old errors. If my English sounds somewhat wooden, please realize that I'm not a native speaker. Feel free to send suggestions. If you feel you are better qualified to maintain a section, or think that you can author and maintain new sections, you are welcome to do so. The SGML of this HOWTO is available via CVS, I very much envision more people working on it. In aid of this, you will find lots of FIXME notices. Patches are always welcome! Wherever you find a FIXME, you should know that you are treading in unknown territory. This is not to say that there are no errors elsewhere, but be extra careful. If you have validated something, please let us know so we can remove the FIXME notice. About this HOWTO, I will take some liberties along the road. For example, I postulate a 10Mbit Internet connection, while I know full well that those are not very common. The canonical location for the HOWTO is here. We now have anonymous CVS access available to the world at large. This is good in a number of ways. You can easily upgrade to newer versions of this HOWTO and submitting patches is no work at all. Furthermore, it allows the authors to work on the source independently, which is good too. $ export CVSROOT=:pserver:anon@outpost.ds9a.nl:/var/cvsroot $ cvs login CVS password: [enter 'cvs' (without 's)] $ cvs co 2.4routing cvs server: Updating 2.4routing U 2.4routing/lartc.db If you made changes and want to contribute them, run cvs -z3 diff -uBb, and mail the output to howto@ds9a.nl, we can then integrate it easily. Thanks! Please make sure that you edit the .db file, by the way, the other files are generated from that one. A Makefile is supplied which should help you create postscript, dvi, pdf, html and plain text. You may need to install docbook, docbook-utils, ghostscript and tetex to get all formats. Be careful not to edit 2.4routing.sgml! It contains an older version of the HOWTO. The right file is lartc.db. The authors receive an increasing amount of mail about this HOWTO. Because of the clear interest of the community, it has been decided to start a mailinglist where people can talk to each other about Advanced Routing and Traffic Control. You can subscribe to the list here. It should be pointed out that the authors are very hesitant of answering questions not asked on the list. We would like the archive of the list to become some kind of knowledge base. If you have a question, please search the archive, and then post to the mailinglist. We will be doing interesting stuff almost immediately, which also means that there will initially be parts that are explained incompletely or are not perfect. Please gloss over these parts and assume that all will become clear. Routing and filtering are two distinct things. Filtering is documented very well by Rusty's HOWTOs, available here: Rusty's Remarkably Unreliable Guides We will be focusing mostly on what is possible by combining netfilter and iproute2. Most Linux distributions, and most UNIX's, currently use the venerable arp, ifconfig and route commands. While these tools work, they show some unexpected behaviour under Linux 2.2 and up. For example, GRE tunnels are an integral part of routing these days, but require completely different tools. With iproute2, tunnels are an integral part of the tool set. The 2.2 and above Linux kernels include a completely redesigned network subsystem. This new networking code brings Linux performance and a feature set with little competition in the general OS arena. In fact, the new routing, filtering, and classifying code is more featureful than the one provided by many dedicated routers and firewalls and traffic shaping products. As new networking concepts have been invented, people have found ways to plaster them on top of the existing framework in existing OSes. This constant layering of cruft has lead to networking code that is filled with strange behaviour, much like most human languages. In the past, Linux emulated SunOS's handling of many of these things, which was not ideal. This new framework makes it possible to clearly express features previously beyond Linux's reach. Linux has a sophisticated system for bandwidth provisioning called Traffic Control. This system supports various method for classifying, prioritizing, sharing, and limiting both inbound and outbound traffic. We'll start off with a tiny tour of iproute2 possibilities. You should make sure that you have the userland tools installed. This package is called 'iproute' on both RedHat and Debian, and may otherwise be found at ftp://ftp.inr.ac.ru/ip-routing/iproute2-2.2.4-now-ss??????.tar.gz". You can also try here for the latest version. Some parts of iproute require you to have certain kernel options enabled. It should also be noted that all releases of RedHat up to and including 6.2 come without most of the traffic control features in the default kernel. RedHat 7.2 has everything in by default. Also make sure that you have netlink support, should you choose to roll your own kernel. Iproute2 needs it. This may come as a surprise, but iproute2 is already configured! The current commands ifconfig and route are already using the advanced syscalls, but mostly with very default (ie. boring) settings. The ip tool is central, and we'll ask it to display our interfaces for us. [ahu@home ahu]$ ip link list 1: lo: <LOOPBACK,UP> mtu 3924 qdisc noqueue link/loopback 00:00:00:00:00:00 brd 00:00:00:00:00:00 2: dummy: <BROADCAST,NOARP> mtu 1500 qdisc noop link/ether 00:00:00:00:00:00 brd ff:ff:ff:ff:ff:ff 3: eth0: <BROADCAST,MULTICAST,PROMISC,UP> mtu 1400 qdisc pfifo_fast qlen 100 link/ether 48:54:e8:2a:47:16 brd ff:ff:ff:ff:ff:ff 4: eth1: <BROADCAST,MULTICAST,PROMISC,UP> mtu 1500 qdisc pfifo_fast qlen 100 link/ether 00:e0:4c:39:24:78 brd ff:ff:ff:ff:ff:ff 3764: ppp0: <POINTOPOINT,MULTICAST,NOARP,UP> mtu 1492 qdisc pfifo_fast qlen 10 link/ppp Your mileage may vary, but this is what it shows on my NAT router at home. I'll only explain part of the output as not everything is directly relevant. We first see the loopback interface. While your computer may function somewhat without one, I'd advise against it. The MTU size (Maximum Transfer Unit) is 3924 octets, and it is not supposed to queue. Which makes sense because the loopback interface is a figment of your kernel's imagination. I'll skip the dummy interface for now, and it may not be present on your computer. Then there are my two physical network interfaces, one at the side of my cable modem, the other one serves my home ethernet segment. Furthermore, we see a ppp0 interface. Note the absence of IP addresses. iproute disconnects the concept of 'links' and 'IP addresses'. With IP aliasing, the concept of 'the' IP address had become quite irrelevant anyhow. It does show us the MAC addresses though, the hardware identifier of our ethernet interfaces. [ahu@home ahu]$ ip address show 1: lo: <LOOPBACK,UP> mtu 3924 qdisc noqueue link/loopback 00:00:00:00:00:00 brd 00:00:00:00:00:00 inet 127.0.0.1/8 brd 127.255.255.255 scope host lo 2: dummy: <BROADCAST,NOARP> mtu 1500 qdisc noop link/ether 00:00:00:00:00:00 brd ff:ff:ff:ff:ff:ff 3: eth0: <BROADCAST,MULTICAST,PROMISC,UP> mtu 1400 qdisc pfifo_fast qlen 100 link/ether 48:54:e8:2a:47:16 brd ff:ff:ff:ff:ff:ff inet 10.0.0.1/8 brd 10.255.255.255 scope global eth0 4: eth1: <BROADCAST,MULTICAST,PROMISC,UP> mtu 1500 qdisc pfifo_fast qlen 100 link/ether 00:e0:4c:39:24:78 brd ff:ff:ff:ff:ff:ff 3764: ppp0: <POINTOPOINT,MULTICAST,NOARP,UP> mtu 1492 qdisc pfifo_fast qlen 10 link/ppp inet 212.64.94.251 peer 212.64.94.1/32 scope global ppp0 This contains more information. It shows all our addresses, and to which cards they belong. 'inet' stands for Internet (IPv4). There are lots of other address families, but these don't concern us right now. Let's examine eth0 somewhat closer. It says that it is related to the inet address '10.0.0.1/8'. What does this mean? The /8 stands for the number of bits that are in the Network Address. There are 32 bits, so we have 24 bits left that are part of our network. The first 8 bits of 10.0.0.1 correspond to 10.0.0.0, our Network Address, and our netmask is 255.0.0.0. The other bits are connected to this interface, so 10.250.3.13 is directly available on eth0, as is 10.0.0.1 for example. With ppp0, the same concept goes, though the numbers are different. Its address is 212.64.94.251, without a subnet mask. This means that we have a point-to-point connection and that every address, with the exception of 212.64.94.251, is remote. There is more information, however. It tells us that on the other side of the link there is, yet again, only one address, 212.64.94.1. The /32 tells us that there are no 'network bits'. It is absolutely vital that you grasp these concepts. Refer to the documentation mentioned at the beginning of this HOWTO if you have trouble. You may also note 'qdisc', which stands for Queueing Discipline. This will become vital later on. Well, we now know how to find 10.x.y.z addresses, and we are able to reach 212.64.94.1. This is not enough however, so we need instructions on how to reach the world. The Internet is available via our ppp connection, and it appears that 212.64.94.1 is willing to spread our packets around the world, and deliver results back to us. [ahu@home ahu]$ ip route show 212.64.94.1 dev ppp0 proto kernel scope link src 212.64.94.251 10.0.0.0/8 dev eth0 proto kernel scope link src 10.0.0.1 127.0.0.0/8 dev lo scope link default via 212.64.94.1 dev ppp0 This is pretty much self explanatory. The first 3 lines of output explicitly state what was already implied by ip address show, the last line tells us that the rest of the world can be found via 212.64.94.1, our default gateway. We can see that it is a gateway because of the word via, which tells us that we need to send packets to 212.64.94.1, and that it will take care of things. For reference, this is what the old route utility shows us: [ahu@home ahu]$ route -n Kernel IP routing table Destination Gateway Genmask Flags Metric Ref Use Iface 212.64.94.1 0.0.0.0 255.255.255.255 UH 0 0 0 ppp0 10.0.0.0 0.0.0.0 255.0.0.0 U 0 0 0 eth0 127.0.0.0 0.0.0.0 255.0.0.0 U 0 0 0 lo 0.0.0.0 212.64.94.1 0.0.0.0 UG 0 0 0 ppp0 ARP is the Address Resolution Protocol as described in RFC 826. ARP is used by a networked machine to resolve the hardware location/address of another machine on the same local network. Machines on the Internet are generally known by their names which resolve to IP addresses. This is how a machine on the foo.com network is able to communicate with another machine which is on the bar.net network. An IP address, though, cannot tell you the physical location of a machine. This is where ARP comes into the picture. Let's take a very simple example. Suppose I have a network composed of several machines. Two of the machines which are currently on my network are foo with an IP address of 10.0.0.1 and bar with an IP address of 10.0.0.2. Now foo wants to ping bar to see that he is alive, but alas, foo has no idea where bar is. So when foo decides to ping bar he will need to send out an ARP request. This ARP request is akin to foo shouting out on the network "Bar (10.0.0.2)! Where are you?" As a result of this every machine on the network will hear foo shouting, but only bar (10.0.0.2) will respond. Bar will then send an ARP reply directly back to foo which is akin bar saying, "Foo (10.0.0.1) I am here at 00:60:94:E9:08:12." After this simple transaction that's used to locate his friend on the network, foo is able to communicate with bar until he (his arp cache) forgets where bar is (typically after 15 minutes on Unix). Now let's see how this works. You can view your machines current arp/neighbor cache/table like so: [root@espa041 /home/src/iputils]# ip neigh show 9.3.76.42 dev eth0 lladdr 00:60:08:3f:e9:f9 nud reachable 9.3.76.1 dev eth0 lladdr 00:06:29:21:73:c8 nud reachable As you can see my machine espa041 (9.3.76.41) knows where to find espa042 (9.3.76.42) and espagate (9.3.76.1). Now let's add another machine to the arp cache. [root@espa041 /home/paulsch/.gnome-desktop]# ping -c 1 espa043 PING espa043.austin.ibm.com (9.3.76.43) from 9.3.76.41 : 56(84) bytes of data. 64 bytes from 9.3.76.43: icmp_seq=0 ttl=255 time=0.9 ms --- espa043.austin.ibm.com ping statistics --- 1 packets transmitted, 1 packets received, 0% packet loss round-trip min/avg/max = 0.9/0.9/0.9 ms [root@espa041 /home/src/iputils]# ip neigh show 9.3.76.43 dev eth0 lladdr 00:06:29:21:80:20 nud reachable 9.3.76.42 dev eth0 lladdr 00:60:08:3f:e9:f9 nud reachable 9.3.76.1 dev eth0 lladdr 00:06:29:21:73:c8 nud reachable As a result of espa041 trying to contact espa043, espa043's hardware address/location has now been added to the arp/neighbor cache. So until the entry for espa043 times out (as a result of no communication between the two) espa041 knows where to find espa043 and has no need to send an ARP request. Now let's delete espa043 from our arp cache: [root@espa041 /home/src/iputils]# ip neigh delete 9.3.76.43 dev eth0 [root@espa041 /home/src/iputils]# ip neigh show 9.3.76.43 dev eth0 nud failed 9.3.76.42 dev eth0 lladdr 00:60:08:3f:e9:f9 nud reachable 9.3.76.1 dev eth0 lladdr 00:06:29:21:73:c8 nud stale Now espa041 has again forgotten where to find espa043 and will need to send another ARP request the next time he needs to communicate with espa043. You can also see from the above output that espagate (9.3.76.1) has been changed to the "stale" state. This means that the location shown is still valid, but it will have to be confirmed at the first transaction to that machine. If you have a large router, you may well cater for the needs of different people, who should be served differently. The routing policy database allows you to do this by having multiple sets of routing tables. If you want to use this feature, make sure that your kernel is compiled with the "IP: advanced router" and "IP: policy routing" features. When the kernel needs to make a routing decision, it finds out which table needs to be consulted. By default, there are three tables. The old 'route' tool modifies the main and local tables, as does the ip tool (by default). The default rules: [ahu@home ahu]$ ip rule list 0: from all lookup local 32766: from all lookup main 32767: from all lookup default This lists the priority of all rules. We see that all rules apply to all packets ('from all'). We've seen the 'main' table before, it is output by ip route ls, but the 'local' and 'default' table are new. If we want to do fancy things, we generate rules which point to different tables which allow us to override system wide routing rules. For the exact semantics on what the kernel does when there are more matching rules, see Alexey's ip-cref documentation. Let's take a real example once again, I have 2 (actually 3, about time I returned them) cable modems, connected to a Linux NAT ('masquerading') router. People living here pay me to use the Internet. Suppose one of my house mates only visits hotmail and wants to pay less. This is fine with me, but they'll end up using the low-end cable modem. The 'fast' cable modem is known as 212.64.94.251 and is a PPP link to 212.64.94.1. The 'slow' cable modem is known by various ip addresses, 212.64.78.148 in this example and is a link to 195.96.98.253. The local table: [ahu@home ahu]$ ip route list table local broadcast 127.255.255.255 dev lo proto kernel scope link src 127.0.0.1 local 10.0.0.1 dev eth0 proto kernel scope host src 10.0.0.1 broadcast 10.0.0.0 dev eth0 proto kernel scope link src 10.0.0.1 local 212.64.94.251 dev ppp0 proto kernel scope host src 212.64.94.251 broadcast 10.255.255.255 dev eth0 proto kernel scope link src 10.0.0.1 broadcast 127.0.0.0 dev lo proto kernel scope link src 127.0.0.1 local 212.64.78.148 dev ppp2 proto kernel scope host src 212.64.78.148 local 127.0.0.1 dev lo proto kernel scope host src 127.0.0.1 local 127.0.0.0/8 dev lo proto kernel scope host src 127.0.0.1 Lots of obvious things, but things that need to be specified somewhere. Well, here they are. The default table is empty. Let's view the 'main' table: [ahu@home ahu]$ ip route list table main 195.96.98.253 dev ppp2 proto kernel scope link src 212.64.78.148 212.64.94.1 dev ppp0 proto kernel scope link src 212.64.94.251 10.0.0.0/8 dev eth0 proto kernel scope link src 10.0.0.1 127.0.0.0/8 dev lo scope link default via 212.64.94.1 dev ppp0 We now generate a new rule which we call 'John', for our hypothetical house mate. Although we can work with pure numbers, it's far easier if we add our tables to /etc/iproute2/rt_tables. # echo 200 John >> /etc/iproute2/rt_tables # ip rule add from 10.0.0.10 table John # ip rule ls 0: from all lookup local 32765: from 10.0.0.10 lookup John 32766: from all lookup main 32767: from all lookup default Now all that is left is to generate John's table, and flush the route cache: # ip route add default via 195.96.98.253 dev ppp2 table John # ip route flush cache And we are done. It is left as an exercise for the reader to implement this in ip-up. A common configuration is the following, in which there are two providers that connect a local network (or even a single machine) to the big Internet. ________ +------------+ / | | | +-------------+ Provider 1 +------- __ | | | / ___/ \_ +------+-------+ +------------+ | _/ \__ | if1 | / / \ | | | | Local network -----+ Linux router | | Internet \_ __/ | | | \__ __/ | if2 | \ \___/ +------+-------+ +------------+ | | | | \ +-------------+ Provider 2 +------- | | | +------------+ \________ There are usually two questions given this setup. The first is how to route answers to packets coming in over a particular provider, say Provider 1, back out again over that same provider. Let us first set some symbolical names. Let $IF1 be the name of the first interface (if1 in the picture above) and $IF2 the name of the second interface. Then let $IP1 be the IP address associated with $IF1 and $IP2 the IP address associated with $IF2. Next, let $P1 be the IP address of the gateway at Provider 1, and $P2 the IP address of the gateway at provider 2. Finally, let $P1_NET be the IP network $P1 is in, and $P2_NET the IP network $P2 is in. One creates two additional routing tables, say T1 and T2. These are added in /etc/iproute2/rt_tables. Then you set up routing in these tables as follows: ip route add $P1_NET dev $IF1 src $IP1 table T1 ip route add default via $P1 table T1 ip route add $P2_NET dev $IF2 src $IP2 table T2 ip route add default via $P2 table T2 Nothing spectacular, just build a route to the gateway and build a default route via that gateway, as you would do in the case of a single upstream provider, but put the routes in a separate table per provider. Note that the network route suffices, as it tells you how to find any host in that network, which includes the gateway, as specified above. Next you set up the main routing table. It is a good idea to route things to the direct neighbour through the interface connected to that neighbour. Note the `src' arguments, they make sure the right outgoing IP address is chosen. ip route add $P1_NET dev $IF1 src $IP1 ip route add $P2_NET dev $IF2 src $IP2 Then, your preference for default route: ip route add default via $P1 Next, you set up the routing rules. These actually choose what routing table to route with. You want to make sure that you route out a given interface if you already have the corresponding source address: ip rule add from $IP1 table T1 ip rule add from $IP2 table T2 This set of commands makes sure all answers to traffic coming in on a particular interface get answered from that interface. Reader Rod Roark notes: 'If $P0_NET is the local network and $IF0 is its interface, the following additional entries are desirable: ip route add $P0_NET dev $IF0 table T1 ip route add $P2_NET dev $IF2 table T1 ip route add 127.0.0.0/8 dev lo table T1 ip route add $P0_NET dev $IF0 table T2 ip route add $P1_NET dev $IF1 table T2 ip route add 127.0.0.0/8 dev lo table T2 ' Now, this is just the very basic setup. It will work for all processes running on the router itself, and for the local network, if it is masqueraded. If it is not, then you either have IP space from both providers 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. The second question is how to balance traffic going out over the two providers. This is actually not hard if you already have set up split access as above. Instead of choosing one of the two providers as your default route, you now set up the default route to be a multipath route. In the default kernel this will balance routes over the two providers. It is done as follows (once more building on the example in the section on split-access): ip route add default scope global nexthop via $P1 dev $IF1 weight 1 \ nexthop via $P2 dev $IF2 weight 1 This will balance the routes over both providers. The weight parameters can be tweaked to favor one provider over the other. Note that balancing will not be perfect, as it is route based, and routes are cached. This means that routes to often-used sites will always be over the same provider. Furthermore, if you really want to do this, you probably also want to look at Julian Anastasov's patches at http://www.ssi.bg/~ja/#routes , Julian's route patch page. They will make things nicer to work with. There are 3 kinds of tunnels in Linux. There's IP in IP tunneling, GRE tunneling and tunnels that live outside the kernel (like, for example PPTP). Tunnels can be used to do some very unusual and very cool stuff. They can also make things go horribly wrong when you don't configure them right. Don't point your default route to a tunnel device unless you know EXACTLY what you are doing :-). Furthermore, tunneling increases overhead, because it needs an extra set of IP headers. Typically this is 20 bytes per packet, so if the normal packet size (MTU) on a network is 1500 bytes, a packet that is sent through a tunnel can only be 1480 bytes big. This is not necessarily a problem, but be sure to read up on IP packet fragmentation/reassembly when you plan to connect large networks with tunnels. Oh, and of course, the fastest way to dig a tunnel is to dig at both sides. This kind of tunneling has been available in Linux for a long time. It requires 2 kernel modules, ipip.o and new_tunnel.o. Let's say you have 3 networks: Internal networks A and B, and intermediate network C (or let's say, Internet). So we have network A: network 10.0.1.0 netmask 255.255.255.0 router 10.0.1.1 The router has address 172.16.17.18 on network C. and network B: network 10.0.2.0 netmask 255.255.255.0 router 10.0.2.1 The router has address 172.19.20.21 on network C. As far as network C is concerned, we assume that it will pass any packet sent from A to B and vice versa. You might even use the Internet for this. Here's what you do: First, make sure the modules are installed: insmod ipip.o insmod new_tunnel.o Then, on the router of network A, you do the following: ifconfig tunl0 10.0.1.1 pointopoint 172.19.20.21 route add -net 10.0.2.0 netmask 255.255.255.0 dev tunl0 And on the router of network B: ifconfig tunl0 10.0.2.1 pointopoint 172.16.17.18 route add -net 10.0.1.0 netmask 255.255.255.0 dev tunl0 And if you're finished with your tunnel: ifconfig tunl0 down Presto, you're done. You can't forward broadcast or IPv6 traffic through an IP-in-IP tunnel, though. You just connect 2 IPv4 networks that normally wouldn't be able to talk to each other, that's all. As far as compatibility goes, this code has been around a long time, so it's compatible all the way back to 1.3 kernels. Linux IP-in-IP tunneling doesn't work with other Operating Systems or routers, as far as I know. It's simple, it works. Use it if you have to, otherwise use GRE. GRE is a tunneling protocol that was originally developed by Cisco, and it can do a few more things than IP-in-IP tunneling. For example, you can also transport multicast traffic and IPv6 through a GRE tunnel. In Linux, you'll need the ip_gre.o module. Let's do IPv4 tunneling first: Let's say you have 3 networks: Internal networks A and B, and intermediate network C (or let's say, Internet). So we have network A: network 10.0.1.0 netmask 255.255.255.0 router 10.0.1.1 The router has address 172.16.17.18 on network C. Let's call this network neta (ok, hardly original) and network B: network 10.0.2.0 netmask 255.255.255.0 router 10.0.2.1 The router has address 172.19.20.21 on network C. Let's call this network netb (still not original) As far as network C is concerned, we assume that it will pass any packet sent from A to B and vice versa. How and why, we do not care. On the router of network A, you do the following: ip tunnel add netb mode gre remote 172.19.20.21 local 172.16.17.18 ttl 255 ip link set netb up ip addr add 10.0.1.1 dev netb ip route add 10.0.2.0/24 dev netb Let's discuss this for a bit. In line 1, we added a tunnel device, and called it netb (which is kind of obvious because that's where we want it to go). Furthermore we told it to use the GRE protocol (mode gre), that the remote address is 172.19.20.21 (the router at the other end), that our tunneling packets should originate from 172.16.17.18 (which allows your router to have several IP addresses on network C and let you decide which one to use for tunneling) and that the TTL field of the packet should be set to 255 (ttl 255). The second line enables the device. In the third line we gave the newly born interface netb the address 10.0.1.1. This is OK for smaller networks, but when you're starting up a mining expedition (LOTS of tunnels), you might want to consider using another IP range for tunneling interfaces (in this example, you could use 10.0.3.0). In the fourth line we set the route for network B. Note the different notation for the netmask. If you're not familiar with this notation, here's how it works: you write out the netmask in binary form, and you count all the ones. If you don't know how to do that, just remember that 255.0.0.0 is /8, 255.255.0.0 is /16 and 255.255.255.0 is /24. Oh, and 255.255.254.0 is /23, in case you were wondering. But enough about this, let's go on with the router of network B. ip tunnel add neta mode gre remote 172.16.17.18 local 172.19.20.21 ttl 255 ip link set neta up ip addr add 10.0.2.1 dev neta ip route add 10.0.1.0/24 dev neta And when you want to remove the tunnel on router A: ip link set netb down ip tunnel del netb Of course, you can replace netb with neta for router B. See Section 6 for a short bit about IPv6 Addresses. On with the tunnels. Let's assume that you have the following IPv6 network, and you want to connect it to 6bone, or a friend. Network 3ffe:406:5:1:5:a:2:1/96 Your IPv4 address is 172.16.17.18, and the 6bone router has IPv4 address 172.22.23.24. ip tunnel add sixbone mode sit remote 172.22.23.24 local 172.16.17.18 ttl 255 ip link set sixbone up ip addr add 3ffe:406:5:1:5:a:2:1/96 dev sixbone ip route add 3ffe::/15 dev sixbone Let's discuss this. In the first line, we created a tunnel device called sixbone. We gave it mode sit (which is IPv6 in IPv4 tunneling) and told it where to go to (remote) and where to come from (local). TTL is set to maximum, 255. Next, we made the device active (up). After that, we added our own network address, and set a route for 3ffe::/15 (which is currently all of 6bone) through the tunnel. GRE tunnels are currently the preferred type of tunneling. It's a standard that is also widely adopted outside the Linux community and therefore a Good Thing. There are literally dozens of implementations of tunneling outside the kernel. Best known are of course PPP and PPTP, but there are lots more (some proprietary, some secure, some that don't even use IP) and that is really beyond the scope of this HOWTO. By Marco Davids <marco@sara.nl> NOTE to maintainer: As far as I am concerned, this IPv6-IPv4 tunneling is not per definition GRE tunneling. You could tunnel IPv6 over IPv4 by means of GRE tunnel devices (GRE tunnels ANY to IPv4), but the device used here ("sit") only tunnels IPv6 over IPv4 and is therefore something different. This is another application of the tunneling capabilities of Linux. It is popular among the IPv6 early adopters, or pioneers if you like. The 'hands-on' example described below is certainly not the only way to do IPv6 tunneling. However, it is the method that is often used to tunnel between Linux and a Cisco IPv6 capable router and experience tells us that this is just the thing many people are after. Ten to one this applies to you too ;-) A short bit about IPv6 addresses: IPv6 addresses are, compared to IPv4 addresses, really big: 128 bits against 32 bits. And this provides us just with the thing we need: many, many IP-addresses: 340,282,266,920,938,463,463,374,607,431,768,211,465 to be precise. Apart from this, IPv6 (or IPng, for IP Next Generation) is supposed to provide for smaller routing tables on the Internet's backbone routers, simpler configuration of equipment, better security at the IP level and better support for QoS. An example: 2002:836b:9820:0000:0000:0000:836b:9886 Writing down IPv6 addresses can be quite a burden. Therefore, to make life easier there are some rules: Don't use leading zeroes. Same as in IPv4. Use colons to separate every 16 bits or two bytes. When you have lots of consecutive zeroes, you can write this down as ::. You can only do this once in an address and only for quantities of 16 bits, though. The address 2002:836b:9820:0000:0000:0000:836b:9886 can be written down as 2002:836b:9820::836b:9886, which is somewhat friendlier. Another example, the address 3ffe:0000:0000:0000:0000:0020:34A1:F32C can be written down as 3ffe::20:34A1:F32C, which is a lot shorter. IPv6 is intended to be the successor of the current IPv4. Because it is relatively new technology, there is no worldwide native IPv6 network yet. To be able to move forward swiftly, the 6bone was introduced. Native IPv6 networks are connected to each other by encapsulating the IPv6 protocol in IPv4 packets and sending them over the existing IPv4 infrastructure from one IPv6 site to another. That is precisely where the tunnel steps in. To be able to use IPv6, we should have a kernel that supports it. There are many good documents on how to achieve this. But it all comes down to a few steps: Get yourself a recent Linux distribution, with suitable glibc. Then get yourself an up-to-date kernel source. If you are all set, then you can go ahead and compile an IPv6 capable kernel: Go to /usr/src/linux and type: make menuconfig Choose "Networking Options" Select "The IPv6 protocol", "IPv6: enable EUI-64 token format", "IPv6: disable provider based addresses" HINT: Don't go for the 'module' option. Often this won't work well. In other words, compile IPv6 as 'built-in' in your kernel. You can then save your config like usual and go ahead with compiling the kernel. HINT: Before doing so, consider editing the Makefile: EXTRAVERSION = -x ; --> ; EXTRAVERSION = -x-IPv6 There is a lot of good documentation about compiling and installing a kernel, however this document is about something else. If you run into problems at this stage, go and look for documentation about compiling a Linux kernel according to your own specifications. The file /usr/src/linux/README might be a good start. After you accomplished all this, and rebooted with your brand new kernel, you might want to issue an '/sbin/ifconfig -a' and notice the brand new 'sit0-device'. SIT stands for Simple Internet Transition. You may give yourself a compliment; you are now one major step closer to IP, the Next Generation ;-) Now on to the next step. You want to connect your host, or maybe even your entire LAN to another IPv6 capable network. This might be the "6bone" that is setup especially for this particular purpose. Let's assume that you have the following IPv6 network: 3ffe:604:6:8::/64 and you want to connect it to 6bone, or a friend. Please note that the /64 subnet notation works just like with regular IP addresses. Your IPv4 address is 145.100.24.181 and the 6bone router has IPv4 address 145.100.1.5 # ip tunnel add sixbone mode sit remote 145.100.1.5 [local 145.100.24.181 ttl 255] # ip link set sixbone up # ip addr add 3FFE:604:6:7::2/126 dev sixbone # ip route add 3ffe::0/16 dev sixbone Let's discuss this. In the first line, we created a tunnel device called sixbone. We gave it mode sit (which is IPv6 in IPv4 tunneling) and told it where to go to (remote) and where to come from (local). TTL is set to maximum, 255. Next, we made the device active (up). After that, we added our own network address, and set a route for 3ffe::/15 (which is currently all of 6bone) through the tunnel. If the particular machine you run this on is your IPv6 gateway, then consider adding the following lines: # echo 1 >/proc/sys/net/ipv6/conf/all/forwarding # /usr/local/sbin/radvd The latter, radvd is -like zebra- a router advertisement daemon, to support IPv6's autoconfiguration features. Search for it with your favourite search-engine if you like. You can check things like this: # /sbin/ip -f inet6 addr If you happen to have radvd running on your IPv6 gateway and boot your IPv6 capable Linux on a machine on your local LAN, you would be able to enjoy the benefits of IPv6 autoconfiguration: # /sbin/ip -f inet6 addr 1: lo: <LOOPBACK,UP> mtu 3924 qdisc noqueue inet6 ::1/128 scope host 3: eth0: <BROADCAST,MULTICAST,UP> mtu 1500 qdisc pfifo_fast qlen 100 inet6 3ffe:604:6:8:5054:4cff:fe01:e3d6/64 scope global dynamic valid_lft forever preferred_lft 604646sec inet6 fe80::5054:4cff:fe01:e3d6/10 scope link You could go ahead and configure your bind for IPv6 addresses. The A type has an equivalent for IPv6: AAAA. The in-addr.arpa's equivalent is: ip6.int. There's a lot of information available on this topic. There is an increasing number of IPv6-aware applications available, including secure shell, telnet, inetd, Mozilla the browser, Apache the webserver and a lot of others. But this is all outside the scope of this Routing document ;-) On the Cisco side the configuration would be something like this: ! interface Tunnel1 description IPv6 tunnel no ip address no ip directed-broadcast ipv6 address 3FFE:604:6:7::1/126 tunnel source Serial0 tunnel destination 145.100.24.181 tunnel mode ipv6ip ! ipv6 route 3FFE:604:6:8::/64 Tunnel1 But if you don't have a Cisco at your disposal, try one of the many IPv6 tunnel brokers available on the Internet. They are willing to configure their Cisco with an extra tunnel for you. Mostly by means of a friendly web interface. Search for "ipv6 tunnel broker" on your favourite search engine. There are two kinds of IPSEC available for Linux these days. For 2.2 and 2.4, there is FreeS/WAN, which was the first major implementation. They have an official site and an unofficial one that is actually maintained. FreeS/WAN has traditionally not been merged with the mainline kernel for a number of reasons. Most often mentioned are 'political' issues with Americans working on crypto tainting its exportability. Furthermore, it does not integrate too well with the Linux kernel, leading it to be a bad candidate for actual merging. Additionally, many parties have voiced worries about the quality of the code. To setup FreeS/WAN, a lot of documentation is available. As of Linux 2.5.47, there is a native IPSEC implementation in the kernel. It was written by Alexey Kuznetsov and Dave Miller, inspired by the work of the USAGI IPv6 group. With its merge, James Morris' CrypoAPI also became part of the kernel - it does the actual crypting. This HOWTO will only document the 2.5+ version of IPSEC. FreeS/WAN is recommended for Linux 2.4 users for now, but be aware that its configuration will differ from the native IPSEC. In related news, there are now patches to make the FreeS/WAN userspace code work with the native Linux IPSEC. As of 2.5.49, IPSEC works without further patches. Userspace tools appear to be available here. There are multiple programs available, the one linked here is based on Racoon. When compiling your kernel, be sure to turn on 'PF_KEY', 'AH', 'ESP' and everything in the CryptoAPI! The author of this chapter is a complete IPSEC nitwit! If you find the inevitable mistakes, please email bert hubert ahu@ds9a.nl. First, we'll show how to manually setup secure communication between two hosts. A large part of this process can also be automated, but here we'll do it by hand so as to acquaint ourselves with what is going on 'under the hood'. Feel free to skip the following section if you are only interested in automatic keying but be aware that some understanding of manual keying is useful. IPSEC is a complicated subject. A lot of information is available online, this HOWTO will concentrate on getting you up and running and explaining the basic principles. All examples are based on Racoon as found on the link above. Many iptables configurations drop IPSEC packets! To pass IPSEC, use: 'iptables -A xxx -p 50 -j ACCEPT' and 'iptables -A xxx -p 51 -j ACCEPT' IPSEC offers a secure version of the Internet Protocol. Security in this context means two different things: encryption and authentication. A naive vision of security offers only encryption but it can easily be shown that is insufficient - you may be communicating encyphered, but no guarantee is offered that the remote party is the one you expect it to be. IPSEC supports 'Encapsulated Security Payload' (ESP) for encryption and 'Authentication Header' (AH) for authenticating the remote partner. You can configure both of them, or decided to do only either. Both ESP and AH rely on security associations. A security association (SA) consists of a source, a destination and an instruction. A sample authentication SA may look like this: add 10.0.0.11 10.0.0.216 ah 15700 -A hmac-md5 "1234567890123456"; This says 'traffic going from 10.0.0.11 to 10.0.0.216 that needs an AH can be signed using HMAC-MD5 using secret 1234567890123456'. This instruction is labelled with SPI ('Security Parameter Index') id '15700', more about that later. The interesting bit about SAs is that they are symmetrical. Both sides of a conversation share exactly the same SA, it is not mirrored on the other side. Do note however that there is no 'autoreverse' rule - this SA only describes a possible authentication from 10.0.0.11 to 10.0.0.216. For two-way traffic, two SAs are needed. A sample ESP SA: add 10.0.0.11 10.0.0.216 esp 15701 -E 3des-cbc "123456789012123456789012"; This says 'traffic going from 10.0.0.11 to 10.0.0.216 that needs encryption can be encyphered using 3des-cbc with key 123456789012123456789012'. The SPI id is '15701'. So far, we've seen that SAs describe possible instructions, but do not in fact describe policy as to when these need to be used. In fact, there could be an arbitrary number of nearly identical SAs with only differing SPI ids. Incidentally, SPI stands for Security Parameter Index. To do actual crypto, we need to describe a policy. This policy can include things as 'use ipsec if available' or 'drop traffic unless we have ispec'. A typical simple Security Policy (SP) looks like this: spdadd 10.0.0.216 10.0.0.11 any -P out ipsec esp/transport//require ah/transport//require; If entered on host 10.0.0.216, this means that all traffic going out to 10.0.0.11 must be encrypted and be wrapped in an AH authenticating header. Note that this does not describe which SA is to be used, that is left as an exercise for the kernel to determine. In other words, a Security Policy specifies WHAT we want; a Security Association describes HOW we want it. Outgoing packets are labelled with the SA SPI ('the how') which the kernel used for encryption and authentication so the remote can lookup the corresponding verification and decryption instruction. What follows is a very simple configuration for talking from host 10.0.0.216 to 10.0.0.11 using encryption and authentication. Note that the reverse path is plaintext in this first version and that this configuration should not be deployed. On host 10.0.0.216: #!/sbin/setkey -f add 10.0.0.216 10.0.0.11 ah 24500 -A hmac-md5 "1234567890123456"; add 10.0.0.216 10.0.0.11 esp 24501 -E 3des-cbc "123456789012123456789012"; spdadd 10.0.0.216 10.0.0.11 any -P out ipsec esp/transport//require ah/transport//require; On host 10.0.0.11, the same Security Associations, no Security Policy: #!/sbin/setkey -f add 10.0.0.216 10.0.0.11 ah 24500 -A hmac-md5 "1234567890123456"; add 10.0.0.216 10.0.0.11 esp 24501 -E 3des-cbc "123456789012123456789012"; With the above configuration in place (these files can be executed if 'setkey' is installed in /sbin), 'ping 10.0.0.11' from 10.0.0.216 looks like this using tcpdump: 22:37:52 10.0.0.216 > 10.0.0.11: AH(spi=0x00005fb4,seq=0xa): ESP(spi=0x00005fb5,seq=0xa) (DF) 22:37:52 10.0.0.11 > 10.0.0.216: icmp: echo reply Note how the ping back from 10.0.0.11 is indeed plainly visible. The forward ping cannot be read by tcpdump of course, but it does show the Security Parameter Index of AH and ESP, which tells 10.0.0.11 how to verify the authenticity of our packet and how to decrypt it. A few things must be mentioned however. The configuration above is shown in a lot of IPSEC examples and it is very dangerous. The problem is that the above contains policy on how 10.0.0.216 should treat packets going to 10.0.0.11, and that it explains how 10.0.0.11 should treat those packets but it does NOT instruct 10.0.0.11 to discard unauthenticated or unencrypted traffic! Anybody can now insert spoofed and completely unencrypted data and 10.0.0.11 will accept it. To remedy the above, we need an incoming Security Policy on 10.0.0.11, as follows: #!/sbin/setkey -f spdadd 10.0.0.216 10.0.0.11 any -P IN ipsec esp/transport//require ah/transport//require; This instructs 10.0.0.11 that any traffic coming to it from 10.0.0.216 is required to have valid ESP and AH. Now, to complete this configuration, we need return traffic to be encrypted and authenticated as well of course. The full configuration on 10.0.0.216: #!/sbin/setkey -f flush; spdflush; # AH add 10.0.0.11 10.0.0.216 ah 15700 -A hmac-md5 "1234567890123456"; add 10.0.0.216 10.0.0.11 ah 24500 -A hmac-md5 "1234567890123456"; # ESP add 10.0.0.11 10.0.0.216 esp 15701 -E 3des-cbc "123456789012123456789012"; add 10.0.0.216 10.0.0.11 esp 24501 -E 3des-cbc "123456789012123456789012"; spdadd 10.0.0.216 10.0.0.11 any -P out ipsec esp/transport//require ah/transport//require; spdadd 10.0.0.11 10.0.0.216 any -P in ipsec esp/transport//require ah/transport//require; And on 10.0.0.11: #!/sbin/setkey -f flush; spdflush; # AH add 10.0.0.11 10.0.0.216 ah 15700 -A hmac-md5 "1234567890123456"; add 10.0.0.216 10.0.0.11 ah 24500 -A hmac-md5 "1234567890123456"; # ESP add 10.0.0.11 10.0.0.216 esp 15701 -E 3des-cbc "123456789012123456789012"; add 10.0.0.216 10.0.0.11 esp 24501 -E 3des-cbc "123456789012123456789012"; spdadd 10.0.0.11 10.0.0.216 any -P out ipsec esp/transport//require ah/transport//require; spdadd 10.0.0.216 10.0.0.11 any -P in ipsec esp/transport//require ah/transport//require; Note that in this example we used identical keys for both directions of traffic. This is not in any way required however. To examine the configuration we just created, execute setkey -D, which shows the Security Associations or setkey -DP which shows the configured policies. In the previous section, encryption was configured using simple shared secrets. In other words, to remain secure, we need to transfer our encryption configuration over a trusted channel. If we were to configure the remote host over telnet, any third party would know our shared secret and the setup would not be secure. Furthermore, because the secret is shared, it is not a secret. The remote can't do a lot with our secret, but we do need to make sure that we use a different secret for communicating with all our partners. This requires a large number of keys, if there are 10 parties, this needs at least 50 different secrets. Besides the symmetric key problem, there is also the need for key rollover. If a third party manages to sniff enough traffic, it may be in a position to reverse engineer the key. This is prevented by moving to a new key every once in a while but that is a process that needs to be automated. Another problem is that with manual keying as described above we exactly define the algorithms and key lengths used, something that requires a lot of coordination with the remote party. It is desirable to be able to have the ability to describe a broader key policy such as 'We can do 3DES and Blowfish with at least the following key lengths'. To solve these isses, IPSEC provides Internet Key Exchange to automatically exchange randomly generated keys which are transmitted using asymmetric encryption technology, according to negotiated algorithm details. The Linux 2.5 IPSEC implementation works with the KAME 'racoon' IKE daemon. As of 9 November, the racoon version in Alexey's iptools distribution can be compiled, although you may need to remove #include <net/route.h> in two files. Alternatively, I've supplied a precompiled version. IKE needs access to UDP port 500, be sure that iptables does not block it. As explained before, automatic keying does a lot of the work for us. Specifically, it creates Security Associations on the fly. It does not however set policy for us, which is as it should be. So, to benefit from IKE, setup a policy, but do not supply any SAs. If the kernel discovers that there is an IPSEC policy, but no Security Association, it will notify the IKE daemon, which then goes to work on trying to negotiate one. Reiterating, a Security Policy specifies WHAT we want; a Security Association describes HOW we want it. Using automatic keying lets us get away with only specifying what we want. Kame racoon comes with a grand host of options, most of which have very fine default values, so we don't need to touch them. As described above, the operator needs to define a Security Policy, but no Security Associations. We leave their negotiation to the IKE daemon. In this example, 10.0.0.11 and 10.0.0.216 are once again going to setup secure communications, but this time with help from racoon. For simplicity this configuration will be using pre-shared keys, the dreaded 'shared secrets'. X.509 certificates are discussed in a separate section, see . We're going to stick to almost the default configuration, identical on both hosts: path pre_shared_key "/usr/local/etc/racoon/psk.txt"; remote anonymous { exchange_mode aggressive,main; doi ipsec_doi; situation identity_only; my_identifier address; lifetime time 2 min; # sec,min,hour initial_contact on; proposal_check obey; # obey, strict or claim proposal { encryption_algorithm 3des; hash_algorithm sha1; authentication_method pre_shared_key; dh_group 2 ; } } sainfo anonymous { pfs_group 1; lifetime time 2 min; encryption_algorithm 3des ; authentication_algorithm hmac_sha1; compression_algorithm deflate ; } Lots of settings - I think yet more can be removed to get closer to the default configuration. A few noteworthy things. We've configured two anonymous settings which hold for all remotes, making further configuration easy. There is no need for per-host stanzas here, unless we really want them. Furthermore, we've set it up such that we identify ourselves based on our IP address ('my_identifier address'), and declare that we can do 3des, sha1, and that we will be using a pre-shared key, located in psk.txt. In psk.txt, we now setup two entries, which do differ on both hosts. On 10.0.0.11: 10.0.0.216 password2 On 10.0.0.216: 10.0.0.11 password2 Make sure these files are owned by root, and set to mode 0600, racoon will not trust their contents otherwise. Note that these files are mirrors from eachother. Now we are ready to setup our desired policy, which is simple enough. On host 10.0.0.216: #!/sbin/setkey -f flush; spdflush; spdadd 10.0.0.216 10.0.0.11 any -P out ipsec esp/transport//require; spdadd 10.0.0.11 10.0.0.216 any -P in ipsec esp/transport//require; And on 10.0.0.11: #!/sbin/setkey -f flush; spdflush; spdadd 10.0.0.11 10.0.0.216 any -P out ipsec esp/transport//require; spdadd 10.0.0.216 10.0.0.11 any -P in ipsec esp/transport//require; Note how again these policies are mirrored. We are now ready to launch racoon! Once launched, the moment we try to telnet from 10.0.0.11 to 10.0.0.216, or the other way around, racoon will start negotiating: 12:18:44: INFO: isakmp.c:1689:isakmp_post_acquire(): IPsec-SA request for 10.0.0.11 queued due to no phase1 found. 12:18:44: INFO: isakmp.c:794:isakmp_ph1begin_i(): initiate new phase 1 negotiation: 10.0.0.216[500]<=>10.0.0.11[500] 12:18:44: INFO: isakmp.c:799:isakmp_ph1begin_i(): begin Aggressive mode. 12:18:44: INFO: vendorid.c:128:check_vendorid(): received Vendor ID: KAME/racoon 12:18:44: NOTIFY: oakley.c:2037:oakley_skeyid(): couldn't find the proper pskey, try to get one by the peer's address. 12:18:44: INFO: isakmp.c:2417:log_ph1established(): ISAKMP-SA established 10.0.0.216[500]-10.0.0.11[500] spi:044d25dede78a4d1:ff01e5b4804f0680 12:18:45: INFO: isakmp.c:938:isakmp_ph2begin_i(): initiate new phase 2 negotiation: 10.0.0.216[0]<=>10.0.0.11[0] 12:18:45: INFO: pfkey.c:1106:pk_recvupdate(): IPsec-SA established: ESP/Transport 10.0.0.11->10.0.0.216 spi=44556347(0x2a7e03b) 12:18:45: INFO: pfkey.c:1318:pk_recvadd(): IPsec-SA established: ESP/Transport 10.0.0.216->10.0.0.11 spi=15863890(0xf21052) If we now run setkey -D, which shows the Security Associations, they are indeed there: 10.0.0.216 10.0.0.11 esp mode=transport spi=224162611(0x0d5c7333) reqid=0(0x00000000) E: 3des-cbc 5d421c1b d33b2a9f 4e9055e3 857db9fc 211d9c95 ebaead04 A: hmac-sha1 c5537d66 f3c5d869 bd736ae2 08d22133 27f7aa99 seq=0x00000000 replay=4 flags=0x00000000 state=mature created: Nov 11 12:28:45 2002 current: Nov 11 12:29:16 2002 diff: 31(s) hard: 600(s) soft: 480(s) last: Nov 11 12:29:12 2002 hard: 0(s) soft: 0(s) current: 304(bytes) hard: 0(bytes) soft: 0(bytes) allocated: 3 hard: 0 soft: 0 sadb_seq=1 pid=17112 refcnt=0 10.0.0.11 10.0.0.216 esp mode=transport spi=165123736(0x09d79698) reqid=0(0x00000000) E: 3des-cbc d7af8466 acd4f14c 872c5443 ec45a719 d4b3fde1 8d239d6a A: hmac-sha1 41ccc388 4568ac49 19e4e024 628e240c 141ffe2f seq=0x00000000 replay=4 flags=0x00000000 state=mature created: Nov 11 12:28:45 2002 current: Nov 11 12:29:16 2002 diff: 31(s) hard: 600(s) soft: 480(s) last: hard: 0(s) soft: 0(s) current: 231(bytes) hard: 0(bytes) soft: 0(bytes) allocated: 2 hard: 0 soft: 0 sadb_seq=0 pid=17112 refcnt=0 As are the Security Policies we configured ourselves: 10.0.0.11[any] 10.0.0.216[any] tcp in ipsec esp/transport//require created:Nov 11 12:28:28 2002 lastused:Nov 11 12:29:12 2002 lifetime:0(s) validtime:0(s) spid=3616 seq=5 pid=17134 refcnt=3 10.0.0.216[any] 10.0.0.11[any] tcp out ipsec esp/transport//require created:Nov 11 12:28:28 2002 lastused:Nov 11 12:28:44 2002 lifetime:0(s) validtime:0(s) spid=3609 seq=4 pid=17134 refcnt=3 If this does not work, check that all configuration files are owned by root, and can only be read by root. To start racoon on the foreground, use '-F'. To force it to read a certain configuration file, instead of at the compiled location, use '-f'. For staggering amounts of detail, add a 'log debug;' statement to racoon.conf. As mentioned before, the use of shared secrets is hard because they aren't easily shared and once shared, are no longer secret. Luckily, there is asymmetric encryption technology to help resolve this. If each IPSEC participant makes a public and a private key, secure communications can be setup by both parties publishing their public key, and configuring policy. Building a key is relatively easy, although it requires some work. The following is based on the 'openssl' tool. OpenSSL has a lot of infrastructure for keys that may or may not be signed by certificate authorities. Right now, we need to circumvent all that infrastructure and practice some good old Snake Oil security, and do without a certificate authority. First we issue a 'certificate request' for our host, called 'laptop': $ openssl req -new -nodes -newkey rsa:1024 -sha1 -keyform PEM -keyout \ laptop.private -outform PEM -out request.pem This asks us some questions: Country Name (2 letter code) [AU]:NL State or Province Name (full name) [Some-State]:. Locality Name (eg, city) []:Delft Organization Name (eg, company) [Internet Widgits Pty Ltd]:Linux Advanced Routing & Traffic Control Organizational Unit Name (eg, section) []:laptop Common Name (eg, YOUR name) []:bert hubert Email Address []:ahu@ds9a.nl Please enter the following 'extra' attributes to be sent with your certificate request A challenge password []: An optional company name []: It is left to your own discretion how completely you want to fill this out. You may or may not want to put your hostname in there, depending on your security needs. In this example, we have. We'll now 'self sign' this request: $ openssl x509 -req -in request.pem -signkey laptop.private -out \ laptop.public Signature ok subject=/C=NL/L=Delft/O=Linux Advanced Routing & Traffic \ Control/OU=laptop/CN=bert hubert/Email=ahu@ds9a.nl Getting Private key The 'request.pem' file can now be discarded. Repeat this procedure for all hosts you need a key for. You can distribute the '.public' file with impunity, but keep the '.private' one private! Once we have a public and a private key for our hosts we can tell racoon to use them. We return to our previous configuration and the two hosts, 10.0.0.11 ('upstairs') and 10.0.0.216 ('laptop'). To the racoon.conf file on 10.0.0.11, we add: path certificate "/usr/local/etc/racoon/certs"; remote 10.0.0.216 { exchange_mode aggressive,main; my_identifier asn1dn; peers_identifier asn1dn; certificate_type x509 "upstairs.public" "upstairs.private"; peers_certfile "laptop.public"; proposal { encryption_algorithm 3des; hash_algorithm sha1; authentication_method rsasig; dh_group 2 ; } } This tells racoon that certificates are to be found in /usr/local/etc/racoon/certs/. Furthermore, it contains configuration items specific for remote 10.0.0.216. The 'asn1dn' lines tell racoon that the identifier for both the local and remote ends are to be extracted from the public keys. This is the 'subject=/C=NL/L=Delft/O=Linux Advanced Routing & Traffic Control/OU=laptop/CN=bert hubert/Email=ahu@ds9a.nl' output from above. The certificate_type line configures the local public and private key. The peers_certfile statement configures racoon to read the public key of the remote peer from the file laptop.public. The proposal stanza is unchanged from what we've seen earlier, with the exception that the authentication_method is now rsasig, indicating the use of RSA public/private keys for authentication. The addition to the configuration of 10.0.0.216 is nearly identical, except for the usual mirroring: path certificate "/usr/local/etc/racoon/certs"; remote 10.0.0.11 { exchange_mode aggressive,main; my_identifier asn1dn; peers_identifier asn1dn; certificate_type x509 "laptop.public" "laptop.private"; peers_certfile "upstairs.public"; proposal { encryption_algorithm 3des; hash_algorithm sha1; authentication_method rsasig; dh_group 2 ; } } Now that we've added these statements to both hosts, we only need to move the key files in place. The 'upstairs' machine needs upstairs.private, upstairs.public, and laptop.public in /usr/local/etc/racoon/certs. Make sure that this directory is owned by root and has mode 0700 or racoon may refuse to read it! The 'laptop' machine needs laptop.private, laptop.public, and upstairs.public in /usr/local/etc/racoon/certs. In other words, each host needs its own public and private key and additionally, the public key of the remote. Verify that a Security Policy is in place (execute the 'spdadd' lines in ). Then launch racoon and everything should work. To setup secure communications with a remote party, we must exchange public keys. While the public key does not need to be kept a secret, on the contrary, it is very important to be sure that it is in fact the unaltered key. In other words, you need to be certain there is no 'man in the middle'. To make this easy, OpenSSL provides the 'digest' command: $ openssl dgst upstairs.public MD5(upstairs.public)= 78a3bddafb4d681c1ca8ed4d23da4ff1 Now all we need to do is verify if our remote partner sees the same digest. This might be done by meeting in real life or perhaps over the phone, making sure the number of the remote party was not in fact sent over the same email containing the key! Another way of doing this is the use of a Trusted Third Party which runs a Certificate Authority. This CA would then sign your key, which we've done ourselves above. So far, we've only seen IPSEC in so called 'transport' mode where both endpoints understand IPSEC directly. As this is often not the case, it may be necessary to have only routers understand IPSEC, and have them do the work for the hosts behind them. This is called 'tunnel mode'. Setting this up is a breeze. To tunnel all traffic to 130.161.0.0/16 from 10.0.0.216 via 10.0.0.11, we issue the following on 10.0.0.216: #!/sbin/setkey -f flush; spdflush; add 10.0.0.216 10.0.0.11 esp 34501 -m tunnel -E 3des-cbc "123456789012123456789012"; spdadd 10.0.0.0/24 130.161.0.0/16 any -P out ipsec esp/tunnel/10.0.0.216-10.0.0.11/require; Note the '-m tunnel', it is vitally important! This first configures an ESP encryption SA between our tunnel endpoints, 10.0.0.216 and 10.0.0.11. Next the actual tunnel is configured. It instructs the kernel to encrypt all traffic it has to route from 10.0.0.0/24 to 130.161.0.0. Furthermore, this traffic then has to be shipped to 10.0.0.11. 10.0.0.11 also needs some configuration: #!/sbin/setkey -f flush; spdflush; add 10.0.0.216 10.0.0.11 esp 34501 -m tunnel -E 3des-cbc "123456789012123456789012"; spdadd 10.0.0.0/24 130.161.0.0/16 any -P in ipsec esp/tunnel/10.0.0.216-10.0.0.11/require; Note that this is exactly identical, except for the change from '-P out' to '-P in'. As with earlier examples, we've now only configured traffic going one way. Completing the other half of the tunnel is left as an exercise for the reader. Another name for this setup is 'proxy ESP', which is somewhat clearer. The IPSEC tunnel needs to have IP Forwarding enabled in the kernel! Thomas Walpuski reports that he wrote a patch to make OpenBSD isakpmd work with Linux 2.5 IPSEC. Furthermore, the main isakpmd CVS repository now contains this code! Some notes are on his page. isakpmd is quite different from racoon mentioned above but many people like it. It can be found here. Read more about OpenBSD CVS here. Thomas also made a tarball available for those uncomfortable with CVS or patch. Furthermore, there are patches to make the FreeS/WAN userspace tools work with the native Linux 2.5 IPSEC, you can find them here. FIXME: Write this Andreas Jellinghaus <aj@dungeon.inka.de> reports: "win2k: it works. pre_shared key with ip address for authentication (I don't think windows supports fqdn or userfqdn strings). Certificates should also work, didn't try.". Peter Bieringer reports: Here are some results (tunnel mode only tested, auth=SHA1): DES: ok 3DES: ok AES-128: ok AES-192: not supported by CP VPN-1 AES-256: ok CAST* : not supported by used Linux kernel Tested version: FP4 aka R54 aka w/AI More information here. FIXME: Editor Vacancy! The Multicast-HOWTO is ancient (relatively-speaking) and may be inaccurate or misleading in places, for that reason. Before you can do any multicast routing, you need to configure the Linux kernel to support the type of multicast routing you want to do. This, in turn, requires you to decide what type of multicast routing you expect to be using. There are essentially four "common" types - DVMRP (the Multicast version of the RIP unicast protocol), MOSPF (the same, but for OSPF), PIM-SM ("Protocol Independent Multicasting - Sparse Mode", which assumes that users of any multicast group are spread out, rather than clumped) and PIM-DM (the same, but "Dense Mode", which assumes that there will be significant clumps of users of the same multicast group). In the Linux kernel, you will notice that these options don't appear. This is because the protocol itself is handled by a routing application, such as Zebra, mrouted, or pimd. However, you still have to have a good idea of which you're going to use, to select the right options in the kernel. For all multicast routing, you will definitely need to enable "multicasting" and "multicast routing". For DVMRP and MOSPF, this is sufficient. If you are going to use PIM, you must also enable PIMv1 or PIMv2, depending on whether the network you are connecting to uses version 1 or 2 of the PIM protocol. Once you have all that sorted out, and your new Linux kernel compiled, you will see that the IP protocols listed, at boot time, now include IGMP. This is a protocol for managing multicast groups. At the time of writing, Linux supports IGMP versions 1 and 2 only, although version 3 does exist and has been documented. This doesn't really affect us that much, as IGMPv3 is still new enough that the extra capabilities of IGMPv3 aren't going to be that much use. Because IGMP deals with groups, only the features present in the simplest version of IGMP over the entire group are going to be used. For the most part, that will be IGMPv2, although IGMPv1 is sill going to be encountered. So far, so good. We've enabled multicasting. Now, we have to tell the Linux kernel to actually do something with it, so we can start routing. This means adding the Multicast virtual network to the router table: ip route add 224.0.0.0/4 dev eth0 (Assuming, of course, that you're multicasting over eth0! Substitute the device of your choice, for this.) Now, tell Linux to forward packets... echo 1 > /proc/sys/net/ipv4/ip_forward At this point, you may be wondering if this is ever going to do anything. So, to test our connection, we ping the default group, 224.0.0.1, to see if anyone is alive. All machines on your LAN with multicasting enabled should respond, but nothing else. You'll notice that none of the machines that respond have an IP address of 224.0.0.1. What a surprise! :) This is a group address (a "broadcast" to subscribers), and all members of the group will respond with their own address, not the group address. ping -c 2 224.0.0.1 At this point, you're ready to do actual multicast routing. Well, assuming that you have two networks to route between. (To Be Continued!) Now, when I discovered this, it really blew me away. Linux 2.2/2.4 comes with everything to manage bandwidth in ways comparable to high-end dedicated bandwidth management systems. Linux even goes far beyond what Frame and ATM provide. Just to prevent confusion, tc uses the following rules for bandwith specification: mbps = 1024 kbps = 1024 * 1024 bps => byte/s mbit = 1024 kbit => kilo bit/s. mb = 1024 kb = 1024 * 1024 b => byte mbit = 1024 kbit => kilo bit. Internally, the number is stored in bps and b. But when tc prints the rate, it uses following : 1Mbit = 1024 Kbit = 1024 * 1024 bps => byte/s With queueing we determine the way in which data is SENT. It is important to realise that we can only shape data that we transmit. With the way the Internet works, we have no direct control of what people send us. It's a bit like your (physical!) mailbox at home. There is no way you can influence the world to modify the amount of mail they send you, short of contacting everybody. However, the Internet is mostly based on TCP/IP which has a few features that help us. TCP/IP has no way of knowing the capacity of the network between two hosts, so it just starts sending data faster and faster ('slow start') and when packets start getting lost, because there is no room to send them, it will slow down. In fact it is a bit smarter than this, but more about that later. This is the equivalent of not reading half of your mail, and hoping that people will stop sending it to you. With the difference that it works for the Internet :-) If you have a router and wish to prevent certain hosts within your network from downloading too fast, you need to do your shaping on the *inner* interface of your router, the one that sends data to your own computers. You also have to be sure you are controlling the bottleneck of the link. If you have a 100Mbit NIC and you have a router that has a 256kbit link, you have to make sure you are not sending more data than your router can handle. Otherwise, it will be the router who is controlling the link and shaping the available bandwith. We need to 'own the queue' so to speak, and be the slowest link in the chain. Luckily this is easily possible. As said, with queueing disciplines, we change the way data is sent. Classless queueing disciplines are those that, by and large accept data and only reschedule, delay or drop it. These can be used to shape traffic for an entire interface, without any subdivisions. It is vital that you understand this part of queueing before we go on the classful qdisc-containing-qdiscs! By far the most widely used discipline is the pfifo_fast qdisc - this is the default. This also explains why these advanced features are so robust. They are nothing more than 'just another queue'. Each of these queues has specific strengths and weaknesses. Not all of them may be as well tested. This queue is, as the name says, First In, First Out, which means that no packet receives special treatment. At least, not quite. This queue has 3 so called 'bands'. Within each band, FIFO rules apply. However, as long as there are packets waiting in band 0, band 1 won't be processed. Same goes for band 1 and band 2. The kernel honors the so called Type of Service flag of packets, and takes care to insert 'minimum delay' packets in band 0. Do not confuse this classless simple qdisc with the classful PRIO one! Although they behave similarly, pfifo_fast is classless and you cannot add other qdiscs to it with the tc command. You can't configure the pfifo_fast qdisc as it is the hardwired default. This is how it is configured by default: priomap Determines how packet priorities, as assigned by the kernel, map to bands. Mapping occurs based on the TOS octet of the packet, which looks like this: 0 1 2 3 4 5 6 7 +-----+-----+-----+-----+-----+-----+-----+-----+ | | | | | PRECEDENCE | TOS | MBZ | | | | | +-----+-----+-----+-----+-----+-----+-----+-----+ The four TOS bits (the 'TOS field') are defined as: Binary Decimcal Meaning ----------------------------------------- 1000 8 Minimize delay (md) 0100 4 Maximize throughput (mt) 0010 2 Maximize reliability (mr) 0001 1 Minimize monetary cost (mmc) 0000 0 Normal Service As there is 1 bit to the right of these four bits, the actual value of the TOS field is double the value of the TOS bits. Tcpdump -v -v shows you the value of the entire TOS field, not just the four bits. It is the value you see in the first column of this table: TOS Bits Means Linux Priority Band ------------------------------------------------------------ 0x0 0 Normal Service 0 Best Effort 1 0x2 1 Minimize Monetary Cost 1 Filler 2 0x4 2 Maximize Reliability 0 Best Effort 1 0x6 3 mmc+mr 0 Best Effort 1 0x8 4 Maximize Throughput 2 Bulk 2 0xa 5 mmc+mt 2 Bulk 2 0xc 6 mr+mt 2 Bulk 2 0xe 7 mmc+mr+mt 2 Bulk 2 0x10 8 Minimize Delay 6 Interactive 0 0x12 9 mmc+md 6 Interactive 0 0x14 10 mr+md 6 Interactive 0 0x16 11 mmc+mr+md 6 Interactive 0 0x18 12 mt+md 4 Int. Bulk 1 0x1a 13 mmc+mt+md 4 Int. Bulk 1 0x1c 14 mr+mt+md 4 Int. Bulk 1 0x1e 15 mmc+mr+mt+md 4 Int. Bulk 1 Lots of numbers. The second column contains the value of the relevant four TOS bits, followed by their translated meaning. For example, 15 stands for a packet wanting Minimal Monetary Cost, Maximum Reliability, Maximum Throughput AND Minimum Delay. I would call this a 'Dutch Packet'. The fourth column lists the way the Linux kernel interprets the TOS bits, by showing to which Priority they are mapped. The last column shows the result of the default priomap. On the command line, the default priomap looks like this: 1, 2, 2, 2, 1, 2, 0, 0 , 1, 1, 1, 1, 1, 1, 1, 1 This means that priority 4, for example, gets mapped to band number 1. The priomap also allows you to list higher priorities (> 7) which do not correspond to TOS mappings, but which are set by other means. This table from RFC 1349 (read it for more details) tells you how applications might very well set their TOS bits: TELNET 1000 (minimize delay) FTP Control 1000 (minimize delay) Data 0100 (maximize throughput) TFTP 1000 (minimize delay) SMTP Command phase 1000 (minimize delay) DATA phase 0100 (maximize throughput) Domain Name Service UDP Query 1000 (minimize delay) TCP Query 0000 Zone Transfer 0100 (maximize throughput) NNTP 0001 (minimize monetary cost) ICMP Errors 0000 Requests 0000 (mostly) Responses <same as request> (mostly) txqueuelen The length of this queue is gleaned from the interface configuration, which you can see and set with ifconfig and ip. To set the queue length to 10, execute: ifconfig eth0 txqueuelen 10 You can't set this parameter with tc! The Token Bucket Filter (TBF) is a simple qdisc that only passes packets arriving at a rate which is not exceeding some administratively set rate, but with the possibility to allow short bursts in excess of this rate. TBF is very precise, network- and processor friendly. It should be your first choice if you simply want to slow an interface down! The TBF implementation consists of a buffer (bucket), constantly filled by some virtual pieces of information called tokens, at a specific rate (token rate). The most important parameter of the bucket is its size, that is the number of tokens it can store. Each arriving token collects one incoming data packet from the data queue and is then deleted from the bucket. Associating this algorithm with the two flows -- token and data, gives us three possible scenarios: The data arrives in TBF at a rate that's equal to the rate of incoming tokens. In this case each incoming packet has its matching token and passes the queue without delay. The data arrives in TBF at a rate that's smaller than the token rate. Only a part of the tokens are deleted at output of each data packet that's sent out the queue, so the tokens accumulate, up to the bucket size. The unused tokens can then be used to send data at a speed that's exceeding the standard token rate, in case short data bursts occur. The data arrives in TBF at a rate bigger than the token rate. This means that the bucket will soon be devoid of tokens, which causes the TBF to throttle itself for a while. This is called an 'overlimit situation'. If packets keep coming in, packets will start to get dropped. The last scenario is very important, because it allows to administratively shape the bandwidth available to data that's passing the filter. The accumulation of tokens allows a short burst of overlimit data to be still passed without loss, but any lasting overload will cause packets to be constantly delayed, and then dropped. Please note that in the actual implementation, tokens correspond to bytes, not packets. Even though you will probably not need to change them, tbf has some knobs available. First the parameters that are always available: limit or latency Limit is the number of bytes that can be queued waiting for tokens to become available. You can also specify this the other way around by setting the 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 Size of the bucket, in bytes. This is the maximum amount of bytes that tokens can be available for instantaneously. In general, larger shaping rates require a larger buffer. For 10mbit/s on Intel, you need at least 10kbyte buffer if you want to reach your configured rate! If your buffer is too small, packets may be dropped because more tokens arrive per timer tick than fit in your bucket. mpu 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 The speedknob. See remarks above about limits! If the bucket contains tokens and is allowed to empty, by default it does so at infinite speed. If this is unacceptable, use the following parameters: peakrate If tokens are available, and packets arrive, they are sent out immediately by default, at 'lightspeed' so to speak. That may not be what you want, especially if you have a large bucket. The peakrate can be used to specify how quickly the bucket is allowed to be depleted. If doing everything by the book, this is achieved by releasing a packet, and then wait just long enough, and release the next. We calculated our waits so we send just at peakrate. However, due to the default 10ms timer resolution of Unix, with 10.000 bits average packets, we are limited to 1mbit/s of peakrate! mtu/minburst The 1mbit/s peakrate is not very useful if your regular rate is more than that. A higher peakrate is possible by sending out more packets per timertick, which effectively means that we create a second bucket! This second bucket defaults to a single packet, which is not a bucket at 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). A simple but *very* useful configuration is this: # tc qdisc add dev ppp0 root tbf rate 220kbit latency 50ms burst 1540 Ok, why is this useful? If you have a networking device with a large queue, like a DSL modem or a cable modem, and you talk to it over a fast device, like over an ethernet interface, you will find that uploading absolutely destroys interactivity. This is because uploading will fill the queue in the modem, which is probably *huge* because this helps actually achieving good data throughput uploading. But this is not what you want, you want to have the queue not too big so interactivity remains and you can still do other stuff while sending data. The line above slows down sending to a rate that does not lead to a queue in the modem - the queue will be in Linux, where we can control it to a limited size. Change 220kbit to your uplink's *actual* speed, minus a few percent. If you have a really fast modem, raise 'burst' a bit. Stochastic Fairness Queueing (SFQ) is a simple implementation of the fair queueing algorithms family. It's less accurate than others, but it also requires less calculations while being almost perfectly fair. The key word in SFQ is conversation (or flow), which mostly corresponds to a TCP session or a UDP stream. Traffic is divided into a pretty large number of FIFO queues, one for each conversation. Traffic is then sent in a round robin fashion, giving each session the chance to send data in turn. This leads to very fair behaviour and disallows any single conversation from drowning out the rest. SFQ is called 'Stochastic' because it doesn't really allocate a queue for each session, it has an algorithm which divides traffic over a limited number of queues using a hashing algorithm. Because of the hash, multiple sessions might end up in the same bucket, which would halve each session's chance of sending a packet, thus halving the effective speed available. To prevent this situation from becoming noticeable, SFQ changes its hashing algorithm quite often so that any two colliding sessions will only do so for a small number of seconds. It is important to note that SFQ is only useful in case your actual outgoing interface is really full! If it isn't then there will be no queue on your linux machine and hence no effect. Later on we will describe how to combine SFQ with other qdiscs to get a best-of-both worlds situation. Specifically, setting SFQ on the ethernet interface heading to your cable modem or DSL router is pointless without further shaping! The SFQ is pretty much self tuning: perturb Reconfigure hashing once this many seconds. If unset, hash will never be reconfigured. Not recommended. 10 seconds is probably a good value. quantum 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! limit The total number of packets that will be queued by this SFQ (after that it starts dropping them). If you have a device which has identical link speed and actual available rate, like a phone modem, this configuration will help promote fairness: # tc qdisc add dev ppp0 root sfq perturb 10 # tc -s -d qdisc ls qdisc sfq 800c: dev ppp0 quantum 1514b limit 128p flows 128/1024 perturb 10sec Sent 4812 bytes 62 pkts (dropped 0, overlimits 0) The number 800c: is the automatically assigned handle number, limit means that 128 packets can wait in this queue. There are 1024 hashbuckets available for accounting, of which 128 can be active at a time (no more packets fit in the queue!) Once every 10 seconds, the hashes are reconfigured. Summarizing, these are the simple queues that actually manage traffic by reordering, slowing or dropping packets. The following tips may help in choosing which queue to use. It mentions some qdiscs described in the chapter. To purely slow down outgoing traffic, use the Token Bucket Filter. Works up to huge bandwidths, if you scale the bucket. If your link is truly full and you want to make sure that no single session can dominate your outgoing bandwidth, use Stochastical Fairness Queueing. If you have a big backbone and know what you are doing, consider Random Early Drop (see Advanced chapter). To 'shape' incoming traffic which you are not forwarding, use the Ingress Policer. Incoming shaping is called 'policing', by the way, not 'shaping'. If you *are* forwarding it, use a TBF on the interface you are forwarding the data to. Unless you want to shape traffic that may go out over several interfaces, in which case the only common factor is the incoming interface. In that case use the Ingress Policer. If you don't want to shape, but only want to see if your interface is so loaded that it has to queue, use the pfifo queue (not pfifo_fast). It lacks internal bands but does account the size of its backlog. Finally - you can also do social shaping. You may not always be able to use technology to achieve what you want. Users experience technical constraints as hostile. A kind word may also help with getting your bandwidth to be divided right! To properly understand more complicated configurations it is necessary to explain a few concepts first. Because of the complexity and the relative youth of the subject, a lot of different words are used when people in fact mean the same thing. The following is loosely based on draft-ietf-diffserv-model-06.txt, An Informal Management Model for Diffserv Routers. It can currently be found at http://www.ietf.org/internet-drafts/draft-ietf-diffserv-model-06.txt . Read it for the strict definitions of the terms used. Queueing Discipline (qdisc) An algorithm that manages the queue of a device, either incoming (ingress) or outgoing (egress). root qdisc The root qdisc is the qdisc attached to the device. Classless qdisc A qdisc with no configurable internal subdivisions. Classful qdisc A classful qdisc contains multiple classes. Some of these classes contains a further qdisc, which may again be classful, but need not be. According to the strict definition, pfifo_fast *is* classful, because it contains three 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 A classful qdisc may have many classes, each of which is internal to the qdisc. A class, in turn, may have several classes added to it. So a class can have a qdisc as parent or an other class. A leaf class is a class with no child classes. This class has 1 qdisc attached to it. This qdisc is responsible to send the data from that class. When you create a class, a fifo qdisc is attached to it. When you add a child class, this qdisc is removed. For a leaf class, this fifo qdisc can be replaced with an other more suitable qdisc. You can even replace this fifo qdisc with a classful qdisc so you can add extra classes. Classifier Each classful qdisc needs to determine to which class it needs to send a packet. This is done using the classifier. Filter Classification can be performed using filters. A filter contains a number of conditions which if matched, make the filter match. Scheduling A qdisc may, with the help of a classifier, decide that some packets need to 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 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 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 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 Some queues, like for example the Token Bucket Filter, may need to hold on to a packet for a certain time in order to limit the bandwidth. This means that they sometimes refuse to pass a packet, even though they have one available. Now that we have our terminology straight, let's see where all these things are. Userspace programs ^ | +---------------+-----------------------------------------+ | Y | | -------> IP Stack | | | | | | | Y | | | Y | | ^ | | | | / ----------> Forwarding -> | | ^ / | | | |/ Y | | | | | | ^ Y /-qdisc1-\ | | | Egress /--qdisc2--\ | --->->Ingress Classifier ---qdisc3---- | -> | Qdisc \__qdisc4__/ | | \-qdiscN_/ | | | +----------------------------------------------------------+ Thanks to Jamal Hadi Salim for this ASCII representation. The big block represents the kernel. The leftmost arrow represents traffic entering your machine from the network. It is then fed to the Ingress Qdisc which may apply Filters to a packet, and decide to drop it. This is called 'Policing'. This happens at a very early stage, before it has seen a lot of the kernel. It is therefore a very good place to drop traffic very early, without consuming a lot of CPU power. If the packet is allowed to continue, it may be destined for a local application, in which case it enters the IP stack in order to be processed, and handed over to a userspace program. The packet may also be forwarded without entering an application, in which case it is destined for egress. Userspace programs may also deliver data, which is then examined and forwarded to the Egress Classifier. There it is investigated and enqueued to any of a number of qdiscs. In the unconfigured default case, there is only one egress qdisc installed, the pfifo_fast, which always receives the packet. This is called 'enqueueing'. The packet now sits in the qdisc, waiting for the kernel to ask for it for transmission over the network interface. This is called 'dequeueing'. This picture also holds in case there is only one network adaptor - the arrows entering and leaving the kernel should not be taken too literally. Each network adaptor has both ingress and egress hooks. Classful qdiscs are very useful if you have different kinds of traffic which should have differing treatment. One of the classful qdiscs is called 'CBQ', 'Class Based Queueing' and it is so widely mentioned that people identify queueing with classes solely with CBQ, but this is not the case. CBQ is merely the oldest kid on the block - and also the most complex one. It may not always do what you want. This may come as something of a shock to many who fell for the 'sendmail effect', which teaches us that any complex technology which doesn't come with documentation must be the best available. More about CBQ and its alternatives shortly. When traffic enters a classful qdisc, it needs to be sent to any of the classes within - it needs to be 'classified'. To determine what to do with a packet, the so called 'filters' are consulted. It is important to know that the filters are called from within a qdisc, and not the other way around! The filters attached to that qdisc then return with a decision, and the qdisc uses this to enqueue the packet into one of the classes. Each subclass may try other filters to see if further instructions apply. If not, the class enqueues the packet to the qdisc it contains. Besides containing other qdiscs, most classful qdiscs also perform shaping. This is useful to perform both packet scheduling (with SFQ, for example) and rate control. You need this in cases where you have a high speed interface (for example, ethernet) to a slower device (a cable modem). If you were only to run SFQ, nothing would happen, as packets enter & leave your router without delay: the output interface is far faster than your actual link speed. There is no queue to schedule then. Each interface has one egress 'root qdisc'. By default, it is the earlier mentioned classless pfifo_fast queueing discipline. Each qdisc and class is assigned a handle, which can be used by later configuration statements to refer to that qdisc. Besides an egress qdisc, an interface may also have an ingress qdisc , which polices traffic coming in. The handles of these qdiscs consist of two parts, a major number and a minor number : <major>:<minor>. It is customary to name the root qdisc '1:', which is equal to '1:0'. The minor number of a qdisc is always 0. Classes need to have the same major number as their parent. This major number must be unique within a egress or ingress setup. The minor number must be unique within a qdisc and his classes. Recapping, a typical hierarchy might look like this: 1: root qdisc | 1:1 child class / | \ / | \ / | \ / | \ 1:10 1:11 1:12 child classes | | | | 11: | leaf class | | 10: 12: qdisc / \ / \ 10:1 10:2 12:1 12:2 leaf classes But don't let this tree fool you! You should *not* imagine the kernel to be at the apex of the tree and the network below, that is just not the case. Packets get enqueued and dequeued at the root qdisc, which is the only thing the kernel talks to. A packet might get classified in a chain like this: 1: -> 1:1 -> 1:12 -> 12: -> 12:2 The packet now resides in a queue in a qdisc attached to class 12:2. In this example, a filter was attached to each 'node' in the tree, each choosing a branch to take next. This can make sense. However, this is also possible: 1: -> 12:2 In this case, a filter attached to the root decided to send the packet directly to 12:2. When the kernel decides that it needs to extract packets to send to the interface, the root qdisc 1: gets a dequeue request, which is passed to 1:1, which is in turn passed to 10:, 11: and 12:, each of which queries its siblings, and tries to dequeue() from them. In this case, the kernel needs to walk the entire tree, because only 12:2 contains a packet. In short, nested classes ONLY talk to their parent qdiscs, never to an interface. Only the root qdisc gets dequeued by the kernel! The upshot of this is that classes never get dequeued faster than their parents allow. And this is exactly what we want: this way we can have SFQ in an inner class, which doesn't do any shaping, only scheduling, and have a shaping outer qdisc, which does the shaping. The PRIO qdisc doesn't actually shape, it only subdivides traffic based on how you configured your filters. You can consider the PRIO qdisc a kind of pfifo_fast on steroids, whereby each band is a separate class instead of a simple FIFO. When a packet is enqueued to the PRIO qdisc, a class is chosen based on the filter commands you gave. By default, three classes are created. These classes by default contain pure FIFO qdiscs with no internal structure, but you can replace these by any qdisc you have available. Whenever a packet needs to be dequeued, class :1 is tried first. Higher classes are only used if lower bands all did not give up a packet. This qdisc is very useful in case you want to prioritize certain kinds of traffic without using only TOS-flags but using all the power of the tc filters. You can also add an other qdisc to the 3 predefined classes, whereas pfifo_fast is limited to simple fifo qdiscs. Because it doesn't actually shape, the same warning as for SFQ holds: either use it only if your physical link is really full or wrap it inside a classful qdisc that does shape. The latter holds for almost all cable modems and DSL devices. In formal words, the PRIO qdisc is a Work-Conserving scheduler. The following parameters are recognized by tc: bands Number of bands to create. Each band is in fact a class. If you change this number, you must also change: priomap If you do not provide tc filters to classify traffic, the PRIO qdisc looks 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 your PRIO qdisc is called 12:, tc filter traffic to 12:1 to grant it more priority. Reiterating, band 0 goes to minor number 1! Band 1 to minor number 2, etc. We will create this tree: 1: root qdisc / | \ / | \ / | \ 1:1 1:2 1:3 classes | | | 10: 20: 30: qdiscs qdiscs sfq tbf sfq band 0 1 2 Bulk traffic will go to 30:, interactive traffic to 20: or 10:. Command lines: # tc qdisc add dev eth0 root handle 1: prio ## This *instantly* creates classes 1:1, 1:2, 1:3 # tc qdisc add dev eth0 parent 1:1 handle 10: sfq # tc qdisc add dev eth0 parent 1:2 handle 20: tbf rate 20kbit buffer 1600 limit 3000 # tc qdisc add dev eth0 parent 1:3 handle 30: sfq Now let's see what we created: # tc -s qdisc ls dev eth0 qdisc sfq 30: quantum 1514b Sent 0 bytes 0 pkts (dropped 0, overlimits 0) qdisc tbf 20: rate 20Kbit burst 1599b lat 667.6ms Sent 0 bytes 0 pkts (dropped 0, overlimits 0) qdisc sfq 10: quantum 1514b Sent 132 bytes 2 pkts (dropped 0, overlimits 0) qdisc prio 1: bands 3 priomap 1 2 2 2 1 2 0 0 1 1 1 1 1 1 1 1 Sent 174 bytes 3 pkts (dropped 0, overlimits 0) As you can see, band 0 has already had some traffic, and one packet was sent while running this command! We now do some bulk data transfer with a tool that properly sets TOS flags, and take another look: # scp tc ahu@10.0.0.11:./ ahu@10.0.0.11's password: tc 100% |*****************************| 353 KB 00:00 # tc -s qdisc ls dev eth0 qdisc sfq 30: quantum 1514b Sent 384228 bytes 274 pkts (dropped 0, overlimits 0) qdisc tbf 20: rate 20Kbit burst 1599b lat 667.6ms Sent 2640 bytes 20 pkts (dropped 0, overlimits 0) qdisc sfq 10: quantum 1514b Sent 2230 bytes 31 pkts (dropped 0, overlimits 0) qdisc prio 1: bands 3 priomap 1 2 2 2 1 2 0 0 1 1 1 1 1 1 1 1 Sent 389140 bytes 326 pkts (dropped 0, overlimits 0) As you can see, all traffic went to handle 30:, which is the lowest priority band, just as intended. Now to verify that interactive traffic goes to higher bands, we create some interactive traffic: # tc -s qdisc ls dev eth0 qdisc sfq 30: quantum 1514b Sent 384228 bytes 274 pkts (dropped 0, overlimits 0) qdisc tbf 20: rate 20Kbit burst 1599b lat 667.6ms Sent 2640 bytes 20 pkts (dropped 0, overlimits 0) qdisc sfq 10: quantum 1514b Sent 14926 bytes 193 pkts (dropped 0, overlimits 0) qdisc prio 1: bands 3 priomap 1 2 2 2 1 2 0 0 1 1 1 1 1 1 1 1 Sent 401836 bytes 488 pkts (dropped 0, overlimits 0) It worked - all additional traffic has gone to 10:, which is our highest priority qdisc. No traffic was sent to the lowest priority, which previously received our entire scp. As said before, CBQ is the most complex qdisc available, the most hyped, the least understood, and probably the trickiest one to get right. This is not because the authors are evil or incompetent, far from it, it's just that the CBQ algorithm isn't all that precise and doesn't really match the way Linux works. Besides being classful, CBQ is also a shaper and it is in that aspect that it really doesn't work very well. It should work like this. If you try to shape a 10mbit/s connection to 1mbit/s, the link should be idle 90% of the time. If it isn't, we need to throttle so that it IS idle 90% of the time. This is pretty hard to measure, so CBQ instead derives the idle time from the number of microseconds that elapse between requests from the hardware layer for more data. Combined, this can be used to approximate how full or empty the link is. This is rather tortuous and doesn't always arrive at proper results. For example, what if the actual link speed of an interface that is not really able to transmit the full 100mbit/s of data, perhaps because of a badly implemented driver? A PCMCIA network card will also never achieve 100mbit/s because of the way the bus is designed - again, how do we calculate the idle time? It gets even worse if we consider not-quite-real network devices like PPP over Ethernet or PPTP over TCP/IP. The effective bandwidth in that case is probably determined by the efficiency of pipes to userspace - which is huge. People who have done measurements discover that CBQ is not always very accurate and sometimes completely misses the mark. In many circumstances however it works well. With the documentation provided here, you should be able to configure it to work well in most cases. As said before, CBQ works by making sure that the link is idle just long enough to bring down the real bandwidth to the configured rate. To do so, it calculates the time that should pass between average packets. During operations, the effective idletime is measured using an exponential weighted moving average (EWMA), which considers recent packets to be exponentially more important than past ones. The UNIX loadaverage is calculated in the same way. The calculated idle time is subtracted from the EWMA measured one, the resulting number is called 'avgidle'. A perfectly loaded link has an avgidle of zero: packets arrive exactly once every calculated interval. An overloaded link has a negative avgidle and if it gets too negative, CBQ shuts down for a while and is then 'overlimit'. Conversely, an idle link might amass a huge avgidle, which would then allow infinite bandwidths after a few hours of silence. To prevent this, avgidle is capped at maxidle. If overlimit, in theory, the CBQ could throttle itself for exactly the amount of time that was calculated to pass between packets, and then pass one packet, and throttle again. But see the 'minburst' parameter below. These are parameters you can specify in order to configure shaping: avpkt Average size of a packet, measured in bytes. Needed for calculating maxidle, which is derived from maxburst, which is specified in packets. bandwidth The physical bandwidth of your device, needed for idle time calculations. cell The time a packet takes to be transmitted over a device may grow in steps, 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 This number of packets is used to calculate maxidle so that when avgidle is 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 As mentioned before, CBQ needs to throttle in case of overlimit. The ideal solution is to do so for exactly the calculated idle time, and pass 1 packet. For Unix kernels, however, it is generally hard to schedule events shorter than 10ms, so it is better to throttle for a longer period, and then pass minburst packets in one go, and then sleep minburst times longer. 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 If avgidle is below 0, we are overlimits and need to wait until avgidle will be big enough to send one packet. To prevent a sudden burst from shutting down the link for a prolonged period of time, avgidle is reset to minidle if it gets too low. Minidle is specified in negative microseconds, so 10 means that avgidle is capped at -10us. mpu 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 Desired rate of traffic leaving this qdisc - this is the 'speed knob'! Internally, CBQ has a lot of fine tuning. For example, classes which are known not to have data enqueued to them aren't queried. Overlimit classes are penalized by lowering their effective priority. All very smart & complicated. Besides shaping, using the aforementioned idletime approximations, CBQ also acts like the PRIO queue in the sense that classes can have differing priorities and that lower priority numbers will be polled before the higher priority ones. Each time a packet is requested by the hardware layer to be sent out to the network, a weighted round robin process ('WRR') starts, beginning with