|
ping6 weird behaviour
![[it]](/s/countries/it.gif) Carmen Sandiego on Monday, 31 May 2010 17:49:32
Hi. Before reporting the weird behaviour, I'll say something about the current network configuration that I've managed to obtain. I have a PC which acts as router with CentOS: it is connected to the IPv4 network through eth0 physical interface. In it I have installed and configured DHCPv4 (with dhcpd), DHCPv6 (dhcp6s and RADV), DNSv4/v6 (with BIND 9). I've installed AICCU to set the tunnel up to SixXS. I have two clients, one with Windows 7 and one with Ubuntu 10.04, connected to the router PC; every host has it's own dedicated network interface, lets say eth1 and eth3. Now, here's the deal.
Hosts' IPv6 works flawlessly. I can ping6 ipv6.google.com, www.sixxs.net, www.kame.net without problems. I can surf those sites, again with no problems.
On the other hand, if I try to ping6 from inside the router, as follows
ping6 ipv6.google.com
it fails. It doesn't say anything, just stays there blinking. Same behaviour by specifying the address (2001:1450:8007::93). The only way to make it work is
ping6 -I 2001:xxxx:xxx:1::1 ipv6.google.com
where 2001:xxxx:xxx:1::1 is the v6 address of eth1. ping6 works fine, i tcpdumped the sixxs interface and I can see ICMPv6 requests and replys.
So, I'm wondering why I cannot see answers if I don't specify the source network interface. I hope I made myself clear. If not, I'll try to explain better!
Thanks in advance.
ping6 weird behaviour
Carmen Sandiego on Monday, 31 May 2010 17:57:38
UPDATE
I found another weird thing. If I try
ping6 -I 2001:xxxx:xxx:1::1 ipv6.google.com
while monitoring
tcpdump -i eth1 icmp6
I cannot see any reply! All the replys are correctly shown (by tcpdump) by the sixxs virtual interface. How is that possible? ping6 works but tcpdump shows no answers?
Thanks again.
ping6 weird behaviour
tcpdump -i eth1 icmp6
Why monitor eth1? Is that where those packets are going over?
Note of course that tunneled packets won't match up because of the 'icmp6'.
Also note that even though you might see packets on a tunneled interface, they might not be properly transmitted to the other side of the tunnel.
ping6 weird behaviour
What does your routing table look like? Check both "ip -6 ro sho" and "ip -6 addr sho" most likely you have some double addresses/routes.
ping6 weird behaviour
Carmen Sandiego on Tuesday, 01 June 2010 10:12:27
OK, I'll try to explain myself as best as I can.
The router PC has 3 network interfaces. eth0 is used to gather IPv4 connection. eth1 gets a static /64 IPv6, 2001:xxxx:1c6:1::/64. sixxs is the virtual tunnelling interface, through which I simulate native IPv6 connection.
{eth0} ---> IPv4: 192.168.1.2
{eth1} ---> IPv6: 2001:xxxx:1c6:1::/64
{sixxs} ---> tunnel ---> IPv6 connectivity: 2001:xxxx:100:350::2/64
A Win7 host connected to eth1 can successfully ping6 any hostname/address.
Now, I move to the router PC. I try to ping6 google as follows:
ping6 ipv6.google.com
and no answers are received. So, with ping6 still active, I try to see what happens on the network interfaces:
tcpdump -i sixxs icmp6
here I can see that requests and replys are correctly shown. Seems like the source address is missing, so the requests are being sent but the replys can't go any further than the sixxs interface.
Then I try to specify the source address with -I option:
ping6 -I eth1 ipv6.google.com
but still no luck. The solution is to specify the IPv6 address of eth1, as follows:
ping6 -I 2001:xxxx:1c6:1::1 ipv6.google.com
and I can see ping6 working this way (I can see answers, milliseconds and all that stuff). OK, ping6 works as expected. But what about eth1 traffic? While ping6 still active, I try:
tcpdump -i eth1 icmp6
but no replys are shown. So, ping6 works, but no answers are shown by tcpdump.
--- --- --- --- --- --- ---
Also, here is the output of
ip -6 ro sho
unreachable ::/96 dev lo metric 1024 expires 21269632sec error -101 mtu 16436 advmss 16376 hoplimit 4294967295
unreachable ::ffff:0.0.0.0/96 dev lo metric 1024 expires 21269632sec error -101 mtu 16436 advmss 16376 hoplimit 4294967295
2001:xxxx:100:350::/64 dev sixxs metric 256 expires 20660893sec mtu 1280 advmss 1420 hoplimit 4294967295
2001:xxxx:1c6:1::/64 dev eth1 metric 256 expires 21269551sec mtu 1500 advmss 1440 hoplimit 4294967295
unreachable 2002:a00::/24 dev lo metric 1024 expires 21269632sec error -101 mtu 16436 advmss 16376 hoplimit 4294967295
unreachable 2002:7f00::/24 dev lo metric 1024 expires 21269632sec error -101 mtu 16436 advmss 16376 hoplimit 4294967295
unreachable 2002:a9fe::/32 dev lo metric 1024 expires 21269632sec error -101 mtu 16436 advmss 16376 hoplimit 4294967295
unreachable 2002:ac10::/28 dev lo metric 1024 expires 21269632sec error -101 mtu 16436 advmss 16376 hoplimit 4294967295
unreachable 2002:c0a8::/32 dev lo metric 1024 expires 21269632sec error -101 mtu 16436 advmss 16376 hoplimit 4294967295
unreachable 2002:e000::/19 dev lo metric 1024 expires 21269632sec error -101 mtu 16436 advmss 16376 hoplimit 4294967295
unreachable 3ffe:ffff::/32 dev lo metric 1024 expires 21269632sec error -101 mtu 16436 advmss 16376 hoplimit 4294967295
fe80::/64 dev sixxs metric 256 expires 20660893sec mtu 1280 advmss 1420 hoplimit 4294967295
fe80::/64 dev eth0 metric 256 expires 21269544sec mtu 1500 advmss 1440 hoplimit 4294967295
fe80::/64 dev eth1 metric 256 expires 21269548sec mtu 1500 advmss 1440 hoplimit 4294967295
default via 2001:xxxx:100:350::1 dev sixxs metric 1024 expires 21269325sec mtu 1280 advmss 1220 hoplimit 4294967295
-- -- --
And here is the output for
ip -6 addr sho
1: lo: <LOOPBACK,UP,LOWER_UP> mtu 16436
inet6 ::1/128 scope host
valid_lft forever preferred_lft forever
2: eth0: <BROADCAST,MULTICAST,UP,LOWER_UP> mtu 1500 qlen 1000
inet6 fe80::202:b3ff:fe5e:479b/64 scope link
valid_lft forever preferred_lft forever
3: eth1: <BROADCAST,MULTICAST,UP,LOWER_UP> mtu 1500 qlen 1000
inet6 2001:xxxx:1c6:1::1/64 scope global
valid_lft forever preferred_lft forever
inet6 fe80::202:b3ff:fe5e:4bde/64 scope link
valid_lft forever preferred_lft forever
9: sixxs: <POINTOPOINT,MULTICAST,NOARP,UP,LOWER_UP> mtu 1280 qlen 500
inet6 2001:xxxx:100:350::2/64 scope global
valid_lft forever preferred_lft forever
inet6 fe80::xxxx:100:350:2/64 scope link
valid_lft forever preferred_lft forever
Thanks so much for helping, I really appreciate it.
ping6 weird behaviour
The router PC has 3 network interfaces. eth0 is used to gather IPv4 connection. eth1 gets a static /64 IPv6, 2001:xxxx:1c6:1::/64. sixxs is the virtual tunnelling interface, through which I simulate native IPv6 connection. {eth0} ---> IPv4: 192.168.1.2 {eth1} ---> IPv6: 2001:xxxx:1c6:1::/64 {sixxs} ---> tunnel ---> IPv6 connectivity: 2001:xxxx:100:350::2/64
Where does this prefix on eth1 come from? Please do note that the SixXS PoPs only route the address space that you have been assigned from us, which in this case is only the tunnel endpoint, nothing else.
Try using a properly configured setup.
ping6 weird behaviour
Carmen Sandiego on Tuesday, 01 June 2010 10:34:43
2001:xxxx:100:350::2 is just my tunnel endpoint. Anyway, SixXS has given me a /64 subnet, which is 2001:xxxx:1c6:1::/64. That is where the eth1 prefix comes from.
Is anything misconfigured?
ping6 weird behaviour
I am very sure that you do not have a subnet routed over your tunnel and that you have not been given any subnet yet.
ping6 weird behaviour
Carmen Sandiego on Tuesday, 01 June 2010 15:15:05
Subnet information: R12525
Subnet Prefix: 2001:1418:1c6::/48
Tunnel Endpoint: 2001:1418:100:350::2
Tunnel ID: T28929
Subnet Name: experimental subnet
State: Enabled
As stated here, it seems that I have been actually given a subnet, isn't it?
ping6 weird behaviour
Is your name "Livio Bertacco" ?
ping6 weird behaviour
Carmen Sandiego on Tuesday, 01 June 2010 15:20:53
That name refers to my "manager", who has set up the account first. I am doing an IPv6 stage in an italian company.
ping6 weird behaviour
Please elaborate to your "manager" that account sharing is not accepted.
|
![[ch]](/s/countries/ch.gif)
on Monday, 31 May 2010 18:22:04
Posting is only allowed when you are