| Current Path : /usr/src/share/examples/IPv6/ |
FreeBSD hs32.drive.ne.jp 9.1-RELEASE FreeBSD 9.1-RELEASE #1: Wed Jan 14 12:18:08 JST 2015 root@hs32.drive.ne.jp:/sys/amd64/compile/hs32 amd64 |
| Current File : //usr/src/share/examples/IPv6/USAGE |
USAGE
KAME Project
$KAME: USAGE,v 1.33 2000/11/22 10:22:57 itojun Exp $
$FreeBSD: release/9.1.0/share/examples/IPv6/USAGE 108106 2002-12-19 22:56:20Z trhodes $
This is an introduction of how to use the commands provided in the KAME
kit. For more information, please refer to each man page.
<<<ifconfig>>>
A link-local address is automatically assigned to each interface, when
the interface becomes up for the first time. Even if you find an interface
without a link-local address, do not panic. The link-local address will be
assigned when it becomes up (with "ifconfig IF up").
If you do not see a link-local address assigned to an interface on "ifconfig
up", the interface does not support IPv6 for some reasons - for example,
if the interface does not support link-layer multicast (IFF_MULTICAST is not
set), the interface cannot be used for IPv6.
Some network drivers allow an interface to become up even without a
hardware address (for example, PCMCIA network cards). In such cases, it is
possible that an interface has no link-local address even if the
interface is up. If you see such situation, please disable the
interface once and then re-enable it (i.e. do `ifconfig IF down;
ifconfig IF up').
Pseudo interfaces (like "gif" tunnel device) will borrow IPv6
interface identifier (lowermost 64bit of the address) from
EUI64/IEEE802 sources, like ethernet cards. Pseudo interfaces will be
able to get an IPv6 link-local address, if you have other "real"
interface configured beforehand. If you have no EUI64/IEEE802 sources
on the node, we have last-resort code in the kernel, which generates
interface identifier from MD5(hostname). MD5(hostname) may not be suitable
for your usage (for example, if you configure same hostname on both sides of
gif tunnel, you will be doomed), and if so, you may need to configure
link-local address manually.
See RFC2472 for more discussion on how to generate an interface ID for
pseudo interfaces.
If you have a router announcing Router Advertisement,
global addresses will be assigned automatically. So, neither
"ifconfig" nor "prefix" is necessary for your *host* (non-router node).
(Please refer to "sysctl" section for configuring a host to accept
Router Advertisement.)
If you want to set up a router, you need to assign global addresses
for two or more interfaces by "ifconfig" or "prefix" (prefix command
is described at next section).
If you want to assign a global address by "ifconfig", don't forget to
specify the "alias" argument to keep the link-local address.
# ifconfig de0 inet6 3ffe:501:808:1:200:f8ff:fe01:6317 prefixlen 64 alias
# ifconfig de0
de0: flags=8843<UP,BROADCAST,RUNNING,SIMPLEX,MULTICAST> mtu 1500
inet6 fe80::200:f8ff:fe01:6317%de0 prefixlen 64 scopeid 0x1
inet 163.221.202.12 netmask 0xffffff00 broadcast 163.221.202.255
inet6 3ffe:501:808:1:200:f8ff:fe01:6317 prefixlen 64
ether 00:00:f8:01:63:17
media: 100baseTX status: active
See also "/etc/rc.network6" for actual examples.
<<prefix>>
In the IPv6 architecture, an IPv6 address of an interface can be
generated from a prefix assigned to the interface, and a
link-dependent identifier for the interface. So assigning a full IPv6
address by ifconfig is not necessary anymore, because user can only
take care of prefix, by letting system take care of interface
identifier.
The newly added "prefix" command enables user to just assign prefixes
for interfaces, and let your system automatically generate IPv6
addresses. Prefixes added by the "prefix" command is maintained in
the kernel consistently with prefixes assigned by Router
Advertisement (in case of hosts) and with prefixes assigned by Router
Renumbering (in case of routers). Manual assignment of prefixes or
change of prefix properties take precedence over ones assigned by
Router Advertisement or Router Renumbering.
prefix command works only on routers.
If you want to assign a prefix (and consequently address) manually, do
as follows:
# ifconfig de0
de0: flags=8843<UP,BROADCAST,RUNNING,SIMPLEX,MULTICAST> mtu 1500
inet6 fe80::200:f8ff:fe01:6317%de0 prefixlen 64 scopeid 0x1
inet 163.221.202.12 netmask 0xffffff00 broadcast 163.221.202.255
ether 00:00:f8:01:63:17
media: 100baseTX status: active
# prefix de0 3ffe:501:808:1::
# ifconfig de0
de0: flags=8843<UP,BROADCAST,RUNNING,SIMPLEX,MULTICAST> mtu 1500
inet6 fe80::200:f8ff:fe01:6317%de0 prefixlen 64 scopeid 0x1
inet 163.221.202.12 netmask 0xffffff00 broadcast 163.221.202.255
inet6 3ffe:501:808:1:200:f8ff:fe01:6317 prefixlen 64
ether 00:00:f8:01:63:17
media: 100baseTX status: active
To check assigned prefix, use the "ndp" command (See description of
ndp command about its usage).
# ndp -p
3ffe:501:808:1::/64 if=de0
flags=LA, vltime=2592000, pltime=604800, expire=Never, origin=RR
No advertising router
The "prefix" command also has node internal prefix renumbering
ability.
If you have multiple prefixes which have 3ffe:501:808:/48 at the top,
and would like to renumber them to 3ffe:501:4819:/48, then use the
"prefix" command with the "matchpr" argument and the "usepr" argument.
Suppose that current state of before renumbering as follows:
# ifconfig de0
de0: flags=8843<UP,BROADCAST,RUNNING,SIMPLEX,MULTICAST> mtu 1500
inet6 fe80::200:f8ff:fe01:6317%de0 prefixlen 64 scopeid 0x1
inet 163.221.202.12 netmask 0xffffff00 broadcast 163.221.202.255
inet6 3ffe:501:808:1:200:f8ff:fe01:6317 prefixlen 64
ether 00:00:f8:01:63:17
media: 100baseTX status: active
# ifconfig de1
de1: flags=8843<UP,BROADCAST,RUNNING,SIMPLEX,MULTICAST> mtu 1500
inet6 fe80::200:f8ff:fe55:7011%de1 prefixlen 64 scopeid 0x2
inet 163.221.203.12 netmask 0xffffff00 broadcast 163.221.203.255
inet6 3ffe:501:808:2:200:f8ff:fe55:7011 prefixlen 64
ether 00:00:f8:55:70:11
media: 100baseTX status: active
# ndp -p
3ffe:501:808:1::/64 if=de0
flags=LA, vltime=2592000, pltime=604800, expire=Never, origin=RR
No advertising router
3ffe:501:808:2::/64 if=de1
flags=LA, vltime=2592000, pltime=604800, expire=Never, origin=RR
No advertising router
Then do as follows:
# prefix -a matchpr 3ffe:501:808:: mp_len 48 usepr 3ffe:501:4819:: up_uselen 48 change
If command is successful, prefixes and addresses will be renumbered as
follows.
# ifconfig de0
de0: flags=8843<UP,BROADCAST,RUNNING,SIMPLEX,MULTICAST> mtu 1500
inet6 fe80::200:f8ff:fe01:6317%de0 prefixlen 64 scopeid 0x1
inet 163.221.202.12 netmask 0xffffff00 broadcast 163.221.202.255
inet6 3ffe:501:4819:1:200:f8ff:fe01:6317 prefixlen 64
ether 00:00:f8:01:63:17
media: 100baseTX status: active
# ifconfig de1
de1: flags=8843<UP,BROADCAST,RUNNING,SIMPLEX,MULTICAST> mtu 1500
inet6 fe80::200:f8ff:fe55:7011%de0 prefixlen 64 scopeid 0x2
inet 163.221.203.12 netmask 0xffffff00 broadcast 163.221.203.255
inet6 3ffe:501:4819:2:200:f8ff:fe55:7011 prefixlen 64
ether 00:00:f8:55:70:11
media: 100baseTX status: active
# ndp -p
3ffe:501:4819:1::/64 if=de0
flags=LA, vltime=2592000, pltime=604800, expire=Never, origin=RR
No advertising router
3ffe:501:4819:2::/64 if=de1
flags=LA, vltime=2592000, pltime=604800, expire=Never, origin=RR
No advertising router
See also "/etc/rc.network6" for actual examples.
<<<route>>>
If there is a router announcing Router Advertisement on a subnet,
you need not to add a default route for your host by hand
(Please refer to "sysctl" section to accept Router Advertisement).
If you want to add a default route manually, do like:
# route add -inet6 default fe80::200:a2ff:fe0e:7543%ed0
"default" means ::/0. In other cases, if "prefixlen" is omitted, 64
is assumed for "prefixlen" to get along with the aggregatable address.
Note that, in IPv6, a link-local address should be used as gateway
("fe80::200:a2ff:fe0e:7543%ed0" in the above). If you use global addresses,
ICMPv6 redirect will not work properly. Also note that we use a special form
of link-local address as gateway. See Section 1.3 of IMPLEMENTATION for
more details.
For ease of configuration we recommend you to avoid static routes and run
a routing daemon (route6d for example) instead.
<<<ping6>>>
Reachability can be checked by "ping6". This "ping6" allows multicast
for its argument.
% ping6 -n -I ed0 ff02::1
PING6(56=40+8+8 bytes) fe80::5254:ff:feda:cb7d --> ff02::1%ed0
56 bytes from fe80::5254:ff:feda:cb7d%lo0, icmp_seq=0 hlim=64 time=0.25 ms
56 bytes from fe80::2a0:c9ff:fe84:ed6c%ed0, icmp_seq=0 hlim=64 time=1.333 ms(DUP!)
56 bytes from fe80::5254:ff:feda:d161%ed0, icmp_seq=0 hlim=64 time=1.459 ms(DUP!)
56 bytes from fe80::260:97ff:fec2:80bf%ed0, icmp_seq=0 hlim=64 time=1.538 ms(DUP!)
56 bytes from 3ffe:501:4819:2000:5054:ff:fedb:aa46, icmp_seq=0 hlim=255 time=1.615 ms(DUP!)
<<<ping6 -w>>>
Name resolution is possible by ICMPv6 node information query message.
This is very convenient for link-local addresses whose host name cannot be
resolved by DNS. Specify the "-w" option to "ping6".
% ping6 -n -I ed0 -w ff02::1
64 bytes from fe80::5254:ff:feda:cb7d%lo0: fto.kame.net
67 bytes from fe80::5254:ff:feda:d161%ed0: banana.kame.net
69 bytes from fe80::2a0:c9ff:fe84:ebd9%ed0: paradise.kame.net
66 bytes from fe80::260:8ff:fe8b:447f%ed0: taroh.kame.net
66 bytes from fe80::2a0:c9ff:fe84:ed6c%ed0: ayame.kame.net
<<<traceroute6>>>
The route for a target host can be checked by "traceroute6".
% traceroute6 tokyo.v6.wide.ad.jp
traceroute to tokyo.v6.wide.ad.jp (3ffe:501:0:401:200:e8ff:fed5:8923), 30 hops max, 12 byte packets
1 nr60.v6.kame.net 1.239 ms 0.924 ms 0.908 ms
2 otemachi.v6.wide.ad.jp 28.953 ms 31.451 ms 26.567 ms
3 tokyo.v6.wide.ad.jp 26.549 ms 26.58 ms 26.186 ms
If the -l option is specified, both address and name are shown in each line.
% traceroute6 -l tokyo.v6.wide.ad.jp
traceroute to tokyo.v6.wide.ad.jp (3ffe:501:0:401:200:e8ff:fed5:8923), 30 hops max, 12 byte packets
1 nr60.v6.kame.net (3ffe:501:4819:2000:260:97ff:fec2:80bf) 1.23 ms 0.952 ms 0.92 ms
2 otemachi.v6.wide.ad.jp (3ffe:501:0:1802:260:97ff:feb6:7ff0) 27.345 ms 26.706 ms 26.563 ms
3 tokyo.v6.wide.ad.jp (3ffe:501:0:401:200:e8ff:fed5:8923) 26.329 ms 26.36 ms 28.63 ms
<<<ndp>>>
To display the current Neighbor cache, use "ndp":
% ndp -a
Neighbor Linklayer Address Netif Expire St Flgs Prbs
nr60.v6.kame.net 0:60:97:c2:80:bf ed0 expired S R
3ffe:501:4819:2000:2c0:cff:fe 0:c0:c:10:3a:53 ed0 permanent R
paradise.v6.kame.net 52:54:0:dc:52:17 ed0 expired S R
fe80::200:eff:fe49:f929%ed0 0:0:e:49:f9:29 ed0 expired S R
fe80::200:86ff:fe05:80da%ed0 0:0:86:5:80:da ed0 expired S
fe80::200:86ff:fe05:c2d8%ed0 0:0:86:5:c2:d8 ed0 9s R
To flush all of the NDP cache entries, execute the following as root.
# ndp -c
To display the prefix list:
% ndp -p
3ffe:501:4819:2000::/64 if=ed0
flags=LA, vltime=2592000, pltime=604800, expire=29d23h59m58s, origin=RA
advertised by
fe80::5254:ff:fedc:5217%ed0 (reachable)
fe80::260:97ff:fec2:80bf%ed0 (reachable)
fe80::200:eff:fe49:f929%ed0 (no neighbor state)
To display the default router list:
% ndp -r
fe80::260:97ff:fec2:80bf if=ed0, flags=, expire=29m55s
fe80::5254:ff:fedc:5217 if=ed0, flags=, expire=29m7s
fe80::200:eff:fe49:f929 if=ed0, flags=, expire=28m47s
<<<rtsol>>>
To generate a Router Solicitation message right now to get global
addresses, use "rtsol".
# ifconfig ef0
ef0: flags=8863<UP,BROADCAST,NOTRAILERS,RUNNING,SIMPLEX,MULTICAST>
link type ether 0:a0:24:ab:83:9b mtu 1500 speed 10Mbps
media 10baseT status active
inet6 fe80::2a0:24ff:feab:839b%ef0 prefixlen 64 scopeid 0x2
# rtsol ef0
# ifconfig ef0
ef0: flags=8863<UP,BROADCAST,NOTRAILERS,RUNNING,SIMPLEX,MULTICAST>
link type ether 0:a0:24:ab:83:9b mtu 1500 speed 10Mbps
media 10baseT status active
inet6 fe80::2a0:24ff:feab:839b%ef0 prefixlen 64 scopeid 0x2
inet6 3ffe:501:4819:2000:2a0:24ff:feab:839b prefixlen 64
<<<rtsold>>>
rtsold is a daemon version of rtsol. If you run KAME IPv6 on a laptop
computer and frequently move with it, the daemon is useful since it watches
the interface and sends router solicitations when the status of the interface
changes. Note, however, that the feature is disabled by default. Please
add -m option when invocation of rtsold.
rtsold also supports multiple interfaces. For example, you can
invoke the daemon as follows:
# rtsold -m ep0 cnw0
<<<netstat>>>
To see routing table:
# netstat -nr
# netstat -nrl
long format with Ref and Use. Note that bsdi4 does not support the
-l option. You should use the -O option instead.
<<<sysctl>>>
If "net.inet6.ip6.accept_rtadv" is 1, Router Advertisement is
accepted. This means that global addresses and default route are
automatically set up. Otherwise, the announcement is rejected. The
default value is 0. To set "net.inet6.ip6.accept_rtadv" to 1, execute
as follows:
# sysctl net.inet6.ip6.accept_rtadv=1
<<<gifconfig>>>
"gif" interface enables you to perform IPv{4,6} over IPv{4,6}
protocol tunneling. To use this interface, you must specify the
outer IPv{4,6} address by using gifconfig, like:
# gifconfig gif0 163.221.198.61 163.221.11.21
"ifconfig gif0" will configure the address pair used for inner
IPv{4,6} header.
It is not required to configure inner IPv{4,6} address pair. If
you do not configure inner IPv{4,6} address pair, tunnel link is
considered as un-numbered link and the source address of inner
IPv{4,6} address pair will be borrowed from other interfaces.
The following example configures un-numbered IPv6-over-IPv4 tunnel:
# gifconfig gif0 10.0.0.1 10.0.0.1 netmask 255.255.255.0
The following example configures numbered IPv6-over-IPv4 tunnel:
# gifconfig gif0 10.0.0.1 10.0.0.1 netmask 255.255.255.0
# ifconfig gif0 inet6 3ffe:501:808:5::1 3ffe:501:808:5::2 prefixlen 64 alias
IPv6 spec allows you to use point-to-point link without global IPv6
address assigned to the interface. Routing protocol (such as RIPng)
uses link-local addresses only. If you are to configure IPv6-over-IPv4
tunnel, you need not to configure an address pair for inner IPv6
header. We suggest you to use the former example (un-numbered
IPv6-over-IPv4 tunnel) to connect to 6bone for simplicity.
Note that it is so easy to make an infinite routing loop using gif
interface, if you configure a tunnel using the same protocol family
for inner and outer header (i.e. IPv4-over-IPv4).
Refer to gifconfig(8) for more details.
<<<6to4>>>
WARNING: malicious party can abuse 6to4 relay routers/sites, read through
internet draft draft-itojun-ipv6-transition-abuse-xx.txt before configuring it.
"stf" interface enables you to perform 6to4 IPv6-over-IPv4 encapsulation,
as documented in draft-ietf-ngtrans-6to4-06.txt. See stf(4) for details.
<<<inetd>>>
Inetd supports AF_INET and AF_INET6 sockets, with IPsec policy
configuration support.
Refer to inetd(8) for more details.
<<<IPsec>>>
IPsec requires fairly complex configuration, so here we show transport
mode only. http://www.kame.net/newsletter/ has more comprehensive
examples.
Let us setup security association to deploy a secure channel between
HOST A (10.2.3.4) and HOST B (10.6.7.8). Here we show a little
complicated example. From HOST A to HOST B, only old AH is used.
From HOST B to HOST A, new AH and new ESP are combined.
Now we should choose algorithm to be used corresponding to "AH"/"new
AH"/"ESP"/"new ESP". Please refer to the "setkey" man page to know
algorithm names. Our choice is MD5 for AH, new-HMAC-SHA1 for new AH,
and new-DES-expIV with 8 byte IV for new ESP.
Key length highly depends on each algorithm. For example, key
length must be equal to 16 bytes for MD5, 20 for new-HMAC-SHA1,
and 8 for new-DES-expIV. Now we choose "MYSECRETMYSECRET",
"KAMEKAMEKAMEKAMEKAME", "PASSWORD", respectively.
OK, let us assign SPI (Security Parameter Index) for each protocol.
Please note that we need 3 SPIs for this secure channel since three
security headers are produced (one for from HOST A to HOST B, two for
from HOST B to HOST A). Please also note that SPI MUST be greater
than or equal to 256. We choose, 1000, 2000, and 3000, respectively.
(1)
HOST A ------> HOST B
(1)PROTO=AH
ALG=MD5(RFC1826)
KEY=MYSECRETMYSECRET
SPI=1000
(2.1)
HOST A <------ HOST B
<------
(2.2)
(2.1)
PROTO=AH
ALG=new-HMAC-SHA1(new AH)
KEY=KAMEKAMEKAMEKAMEKAME
SPI=2000
(2.2)
PROTO=ESP
ALG=new-DES-expIV(new ESP)
IV length = 8
KEY=PASSWORD
SPI=3000
Now, let us setup security association. Execute "setkey" on both HOST
A and B:
# setkey -c
add 10.2.3.4 10.6.7.8 ah 1000 -m transport -A keyed-md5 "MYSECRETMYSECRET" ;
add 10.6.7.8 10.2.3.4 ah 2000 -m transport -A hmac-sha1 "KAMEKAMEKAMEKAMEKAME" ;
add 10.6.7.8 10.2.3.4 esp 3000 -m transport -E des-cbc "PASSWORD" ;
^D
Actually, IPsec communication doesn't process until security policy
entries will be defined. In this case, you must setup each host.
At A:
# setkey -c
spdadd 10.2.3.4 10.6.7.8 any -P out ipsec
ah/transport/10.2.3.4-10.6.7.8/require ;
^D
At B:
spdadd 10.6.7.8 10.2.3.4 any -P out ipsec
esp/transport//require
ah/transport//require ;
^D
To utilize the security associations installed into the kernel, you
must set the socket security level by using setsockopt().
This is per-application (or per-socket) security. For example,
the "ping" command has the -P option with parameter to enable AH and/or ESP.
For example:
% ping -P "out ipsec \
ah/transport//use \
esp/tunnel/10.0.1.1-10.0.1.2/require" 10.0.2.2
If there are proper SAs, this policy specification causes ICMP packet
to be AH transport mode inner ESP tunnel mode like below.
HOST C -----------> GATEWAY D ----------> HOST E
10.0.1.1 10.0.1.2 10.0.2.1 10.0.2.2
| | | |
| ======= ESP ======= |
==================== AH ==================
<<<EDNS0>>>
EDNS0 is defined in RFC2671. With EDNS0, the resolver library can tell DNS
server of its receiving buffer size, and permit DNS server to transmit large
reply packet. EDNS0 is necessary to take advantage of larger minimum MTU
in IPv6. KAME libinet6 includes resolver side support for EDNS0.
Server side support for EDNS0 is included in ISC BIND9.
query packet with EDNS0
tells receive buffer size
KAME box -----------------------------> BIND9 DNS server
KAME box <----------------------------- BIND9 DNS server
can transmit jumbo reply, since DNS server
knows receive buffer size of KAME box
How to play with it:
- prepare KAME box and BIND9 DNS server (can be a same node)
- add the following into /etc/resolv.conf on KAME box:
options edns0 <--- enables EDNS0
nameserver <IPv4 or v6 address of BIND9 box>
- run applications compiled with libinet6 (like /usr/local/v6/bin/telnet),
see EDNS0 packet fly on the wire by tcpdump or some other method.
Caveats:
- BIND 4/8 DNS server will choke with EDNS0 packet, so you must not
turn the option on if you have BIND 4/8 DNS server. If you enable
"options edns0" against BIND 4/8 DNS server, you will never be able
to resolve names.
- If you use IPv6 UDP as DNS transport, path MTU discovery may
affect the traffic. KAME box tries to fragment packet to 1280
bytes, however, BIND9 may not.
- Some of our platforms do not use our extended resolver code in libinet6.
See COVERAGE for detail.
<<Further readings>>
http://www.netbsd.org/Documentation/network/ipv6/
Even if you are on non-netbsd operating system, the URL should be
useful.
http://www.kame.net/
<end of USAGE>