Forum - Slow behavior :: SixXS - IPv6 Deployment & Tunnel Broker

Slow behavior

Carmen Sandiego on Monday, 20 June 2005 16:47:25

hi there, ive set up a gentoo router using a subnet i got here to give my clients ipv6 access. for now only one client is using it. also a gentoo linux box. I have set up radvd and a dhcpv6 server called dibbler. They work great only i have some questions that aren't clear to me at all. - First: whats the use of radvd when u have a dhcpv6 server? i get 2 adresses. One that i want from dhcpv6 and one from radvd that i dont want. By default the ip i got from radvd is used what i don't want. how do i solve this. - Second: The router box has an ip 2001:960:62f::11. The clientbox has an ip 2001:960:62f::3. I can ping in both ways correctly. Also a route is added to my clientbox:

Kernel IPv6 routing table Destination Next Hop Flags Metric Ref Use Iface ::1/128 :: U 0 3 0 lo 2001:960:62f::3/128 :: U 0 481 0 lo 2001:960:62f:0:250:daff:fede:adc9/128 :: U 0 134 0 lo 2001:960:62f::/64 :: UA 256 6 0 eth0 fe80::250:daff:fede:adc9/128 :: U 0 86 0 lo fe80::/64 :: U 256 0 0 eth0 ff02::1:2/128 ff02::1:2 UC 0 1 1 eth0 ff00::/8 :: U 256 0 0 eth0 ::/0 2001:960:62f::11 UG 1 660 0 eth0 ::/0 fe80::220:18ff:fea0:43b4 UGDA 1024 10 0 eth0

As you can see the last one is the one to use. I had to add that one myself cuz it wasnt there. Without the rule i couldn't ping to any outside ipv6 address (logical). Now when i try to surf to sixxs.net for example. Firefox resolves the ip address and starts connecting. But it doensn't work unless i keep pinging 2001:960:62f::11 infinitely. After a while pinging that doesn't work either anymore until i restart the ping. How can i solve this wacko problem? Here you see the ifconfig from my client:

eth0 Link encap:Ethernet HWaddr 00:50:DA:DE:AD:C9 inet addr:192.168.0.7 Bcast:192.168.0.255 Mask:255.255.255.0 inet6 addr: fe80::250:daff:fede:adc9/64 Scope:Link inet6 addr: 2001:960:62f:0:250:daff:fede:adc9/64 Scope:Global inet6 addr: 2001:960:62f::3/128 Scope:Global UP BROADCAST RUNNING MULTICAST MTU:1500 Metric:1 RX packets:15959 errors:0 dropped:0 overruns:66 frame:0 TX packets:8770 errors:0 dropped:0 overruns:0 carrier:0 collisions:1 txqueuelen:1000 RX bytes:2912691 (2.7 Mb) TX bytes:859897 (839.7 Kb) Interrupt:11 Base address:0xd800

Here it is from my router:

eth0 Link encap:Ethernet HWaddr 00:20:18:A0:43:B4 inet addr:192.168.0.8 Bcast:192.168.0.255 Mask:255.255.255.0 inet6 addr: fe80::220:18ff:fea0:43b4/64 Scope:Link inet6 addr: 2001:960:62f::11/0 Scope:Global UP BROADCAST RUNNING MULTICAST MTU:1500 Metric:1 RX packets:4057846 errors:0 dropped:64 overruns:0 frame:0 TX packets:785130 errors:0 dropped:0 overruns:0 carrier:0 collisions:0 txqueuelen:1000 RX bytes:462851519 (441.4 Mb) TX bytes:257315128 (245.3 Mb) Interrupt:11 lo Link encap:Local Loopback inet addr:127.0.0.1 Mask:255.0.0.0 inet6 addr: ::1/128 Scope:Host UP LOOPBACK RUNNING MTU:16436 Metric:1 RX packets:1087682 errors:0 dropped:0 overruns:0 frame:0 TX packets:1087682 errors:0 dropped:0 overruns:0 carrier:0 collisions:0 txqueuelen:0 RX bytes:79280508 (75.6 Mb) TX bytes:79280508 (75.6 Mb) sit0 Link encap:IPv6-in-IPv4 inet6 addr: ::192.168.0.8/96 Scope:Compat inet6 addr: ::127.0.0.1/96 Scope:Unknown UP RUNNING NOARP MTU:1280 Metric:1 RX packets:0 errors:0 dropped:0 overruns:0 frame:0 TX packets:0 errors:42 dropped:0 overruns:0 carrier:0 collisions:0 txqueuelen:0 RX bytes:0 (0.0 b) TX bytes:0 (0.0 b) sit1 Link encap:IPv6-in-IPv4 inet6 addr: fe80::c0a8:8/64 Scope:Link inet6 addr: 2001:960:62f::1/64 Scope:Global inet6 addr: 2001:960:62f::2/64 Scope:Global inet6 addr: 2001:960:2:2cc::2/64 Scope:Global UP POINTOPOINT RUNNING NOARP MTU:1280 Metric:1 RX packets:204636 errors:0 dropped:0 overruns:0 frame:0 TX packets:175089 errors:0 dropped:0 overruns:0 carrier:0 collisions:0 txqueuelen:0 RX bytes:65508297 (62.4 Mb) TX bytes:18832949 (17.9 Mb)

Slow behavior

Jeroen Massar SixXS Staff

on Tuesday, 21 June 2005 11:15:18

- First: whats the use of radvd when u have a dhcpv6 server? i get 2 adresses.

One that i want from dhcpv6 and one from radvd that i dont want.

By default the ip i got from radvd is used what i don't want. how do i

solve this.

By specifying the "AdvManagedFlag on" and "AdvAutonomous off" flags in the radvd.conf, the first causes the client to know that it is supposed to use the DHCP address, the second mentions it to not use this prefix to configure itself. Having multiple addresses on a link is btw one of the good properties of IPv6. That your OS most likely does not support source address selection is your OS's problem...

- Second: The router box has an ip 2001:960:62f::11. The clientbox has an ip

2001:960:62f::3. I can ping in both ways correctly. Also a route is added to

my clientbox:

You have 2 default routes, though only the 2001:960:62f::11 will be used because of the metric

As you can see the last one is the one to use. I had to add that one myself

cuz it wasnt there. Without the rule i couldn't ping to any outside ipv6

address (logical). Now when i try to surf to sixxs.net for example.

Firefox resolves the ip address and starts connecting. But it doensn't

work unless i keep pinging 2001:960:62f::11 infinitely. After a while

pinging that doesn't work either anymore until i restart the ping.

How can i solve this wacko problem?

You will first want to fix this one: "inet6 addr: 2001:960:62f::11/0 Scope:Global" as that means that the whole internet is attached to your eth0. You can throw out sit0, as you are not using it anyway. Note that since quite a number of years the sit0/sit1 principle is not used anymore. Then you can also resort to using tcpdump and checking if Neighbour Discovery actually works on your network. Also check the Contact page for the "Reporting Problems" section which lists a good checklist. Also you seem to have 2001:960::62f::/64 assigned on both sit1 and on eth0, this will *not* work. You got a /48, that means you have 65536 /64's, you might want to use on of those 65536 networks on eth0. Having additional addresses on your tunnel interface doesn't really make sense. IP's are supposed to point to a host, to enable that host to talk to another host.

Slow behavior

Carmen Sandiego on Tuesday, 21 June 2005 19:14:35

thx, this helped me alot, before i didn't know sure what i was doing :p

Please note Posting is only allowed when you are logged in.