IPv6 problem with /48 routed over link-local

T

Taos

Hello,

I am having a problem that I do not understand. I can reboot the machine and send approximately 8 pings before it will no longer ping an outside address.

The configuration and output is as follows:

rc.conf
Code: 
ipv6_network_interface="em0"
ipv6_defaultrouter="fe80::1%em0"

ipv6_router_enable="YES"
ipv6_route="/usr/sbin/route6d"
ipv6_ifconfig_em0="2607:f2f8:a7c4:: prefixlen 64"
ipv6_ifconfig_em0_alias0="2607:f2f8:a7c4::1 prefixlen 64"
ipv6_ifconfig_em0_alias1="2607:f2f8:a7c4::2 prefixlen 64"

ipv6_gateway_enable="YES"
rtadvd_enable="YES"
rtadvd_interfaces="em0"

ifconfig output:
Code: 
em0: flags=8843<UP,BROADCAST,RUNNING,SIMPLEX,MULTICAST> metric 0 mtu 1500
	options=209b<RXCSUM,TXCSUM,VLAN_MTU,VLAN_HWTAGGING,VLAN_HWCSUM,WOL_MAGIC>
	ether 52:54:00:27:28:36
	inet6 fe80::5054:ff:fe27:2836%em0 prefixlen 64 scopeid 0x1 
	inet6 2607:f2f8:a7c4:: prefixlen 64 
	inet6 2607:f2f8:a7c4::1 prefixlen 64 
	inet6 2607:f2f8:a7c4::2 prefixlen 64 
	nd6 options=3<PERFORMNUD,ACCEPT_RTADV>
	media: Ethernet autoselect (1000baseT <full-duplex>)
	status: active
lo0: flags=8049<UP,LOOPBACK,RUNNING,MULTICAST> metric 0 mtu 16384
	options=3<RXCSUM,TXCSUM>
	inet 127.0.0.1 netmask 0xff000000 
	inet6 ::1 prefixlen 128 
	inet6 fe80::1%lo0 prefixlen 64 scopeid 0x2 
	nd6 options=3<PERFORMNUD,ACCEPT_RTADV>

rtadvd.conf
Code: 
em0:\
:addrs#1:addr="2607:f2f8:a7c4::":prefixlen#64:tc=default:

ping output:
Code: 
ping6 ipv6.google.com
PING6(56=40+8+8 bytes) 2607:f2f8:a7c4:: --> 2607:f8b0:4007:800::1014
16 bytes from 2607:f8b0:4007:800::1014, icmp_seq=0 hlim=58 time=1.936 ms
16 bytes from 2607:f8b0:4007:800::1014, icmp_seq=1 hlim=58 time=1.694 ms
16 bytes from 2607:f8b0:4007:800::1014, icmp_seq=2 hlim=58 time=1.506 ms
16 bytes from 2607:f8b0:4007:800::1014, icmp_seq=3 hlim=58 time=1.632 ms
16 bytes from 2607:f8b0:4007:800::1014, icmp_seq=4 hlim=58 time=1.995 ms
16 bytes from 2607:f8b0:4007:800::1014, icmp_seq=5 hlim=58 time=1.392 ms
16 bytes from 2607:f8b0:4007:800::1014, icmp_seq=6 hlim=58 time=1.511 ms
16 bytes from 2607:f8b0:4007:800::1014, icmp_seq=7 hlim=58 time=2.350 ms
^C
--- ipv6.l.google.com ping6 statistics ---
10 packets transmitted, 8 packets received, 20.0% packet loss
round-trip min/avg/max/std-dev = 1.392/1.752/2.350/0.299 ms

tcpdump output:
Code: 
tcpdump -iem0 |grep IP6
tcpdump: verbose output suppressed, use -v or -vv for full protocol decode
listening on em0, link-type EN10MB (Ethernet), capture size 96 bytes
22:40:46.022380 IP6 fe80::5054:ff:fe27:2836 > ff02::1: ICMP6, router advertisement, length 56
22:40:47.280576 IP6 2607:f2f8:a7c4:: > ff02::1:ff00:1: ICMP6, neighbor solicitation, who has fe80::1, length 32
22:40:47.281079 IP6 2607:f2f8:a7c0::1 > 2607:f2f8:a7c4::: ICMP6, neighbor advertisement, tgt is fe80::1, length 32
22:40:47.281121 IP6 2607:f2f8:a7c4:: > lax04s08-in-x14.1e100.net: ICMP6, echo request, seq 0, length 16
22:40:47.282267 IP6 lax04s08-in-x14.1e100.net > 2607:f2f8:a7c4::: ICMP6, echo reply, seq 0, length 16
22:40:48.292474 IP6 2607:f2f8:a7c4:: > lax04s08-in-x14.1e100.net: ICMP6, echo request, seq 1, length 16
22:40:48.293873 IP6 lax04s08-in-x14.1e100.net > 2607:f2f8:a7c4::: ICMP6, echo reply, seq 1, length 16
22:40:49.282503 IP6 2607:f2f8:a7c4:: > lax04s08-in-x14.1e100.net: ICMP6, echo request, seq 2, length 16
22:40:49.283715 IP6 lax04s08-in-x14.1e100.net > 2607:f2f8:a7c4::: ICMP6, echo reply, seq 2, length 16
22:40:50.275046 IP6 2607:f2f8:a7c4:: > lax04s08-in-x14.1e100.net: ICMP6, echo request, seq 3, length 16
22:40:50.276376 IP6 lax04s08-in-x14.1e100.net > 2607:f2f8:a7c4::: ICMP6, echo reply, seq 3, length 16
22:40:51.275130 IP6 2607:f2f8:a7c4:: > lax04s08-in-x14.1e100.net: ICMP6, echo request, seq 4, length 16
22:40:51.276840 IP6 lax04s08-in-x14.1e100.net > 2607:f2f8:a7c4::: ICMP6, echo reply, seq 4, length 16
22:40:52.277561 IP6 2607:f2f8:a7c0::1 > 2607:f2f8:a7c4::: ICMP6, neighbor solicitation, who has 2607:f2f8:a7c4::, length 32
22:40:52.282525 IP6 2607:f2f8:a7c4:: > lax04s08-in-x14.1e100.net: ICMP6, echo request, seq 5, length 16
22:40:52.283704 IP6 lax04s08-in-x14.1e100.net > 2607:f2f8:a7c4::: ICMP6, echo reply, seq 5, length 16
22:40:53.277262 IP6 2607:f2f8:a7c0::1 > 2607:f2f8:a7c4::: ICMP6, neighbor solicitation, who has 2607:f2f8:a7c4::, length 32
22:40:53.285170 IP6 2607:f2f8:a7c4:: > lax04s08-in-x14.1e100.net: ICMP6, echo request, seq 6, length 16
22:40:53.286379 IP6 lax04s08-in-x14.1e100.net > 2607:f2f8:a7c4::: ICMP6, echo reply, seq 6, length 16
22:40:54.275226 IP6 2607:f2f8:a7c4:: > lax04s08-in-x14.1e100.net: ICMP6, echo request, seq 7, length 16
22:40:54.277233 IP6 2607:f2f8:a7c0::1 > 2607:f2f8:a7c4::: ICMP6, neighbor solicitation, who has 2607:f2f8:a7c4::, length 32
22:40:54.277269 IP6 lax04s08-in-x14.1e100.net > 2607:f2f8:a7c4::: ICMP6, echo reply, seq 7, length 16
22:40:55.280213 IP6 2607:f2f8:a7c4:: > lax04s08-in-x14.1e100.net: ICMP6, echo request, seq 8, length 16
22:40:55.281220 IP6 fe80::1 > ff02::1:ff00:2: ICMP6, neighbor solicitation, who has fe80::2, length 32
22:40:56.276676 IP6 fe80::1 > ff02::1:ff00:2: ICMP6, neighbor solicitation, who has fe80::2, length 32
22:40:56.280355 IP6 2607:f2f8:a7c4:: > lax04s08-in-x14.1e100.net: ICMP6, echo request, seq 9, length 16

netstat:
Code: 
netstat -nr -f inet6
Routing tables

Internet6:
Destination                       Gateway                       Flags      Netif Expire
::/96                             ::1                           UGRS        lo0 =>
default                           fe80::1%em0                   UGS         em0
::1                               ::1                           UH          lo0
::ffff:0.0.0.0/96                 ::1                           UGRS        lo0
2607:f2f8:a7c4::                  link#1                        UHS         lo0 =>
2607:f2f8:a7c4::/64               link#1                        U           em0
2607:f2f8:a7c4::1                 link#1                        UHS         lo0
2607:f2f8:a7c4::2                 link#1                        UHS         lo0
2607:f2f8:a7c4:0:de:face:666:666  link#1                        UHS         lo0
2607:f2f8:a7c4:0:de:face:666:1980 link#1                        UHS         lo0
fe80::/10                         ::1                           UGRS        lo0
fe80::%em0/64                     link#1                        U           em0
fe80::5054:ff:fe27:2836%em0       link#1                        UHS         lo0
fe80::%lo0/64                     link#2                        U           lo0
fe80::1%lo0                       link#2                        UHS         lo0
ff01:1::/32                       fe80::5054:ff:fe27:2836%em0   U           em0
ff01:2::/32                       ::1                           U           lo0
ff02::/16                         ::1                           UGRS        lo0
ff02::%em0/32                     fe80::5054:ff:fe27:2836%em0   U           em0
ff02::%lo0/32                     ::1                           U           lo0

At what point am I messing this up? Any help would be greatly appreciated.

Thanks,

Taos
 
Taos said:
Code: 
ipv6_gateway_enable="YES"
rtadvd_enable="YES"
rtadvd_interfaces="em0"
Click to expand...
Try turning these off, you don't seem to need them. You only have one interface anyway.
 
Hello SirDice,

Thanks for the reply! I actually added these after the fact, but have removed them and am still getting the same problem. :(

It seems the problem happens when it attempts to:

Code: 
22:40:55.281220 IP6 fe80::1 > ff02::1:ff00:2: ICMP6, neighbor solicitation, who has fe80::2, length 32

At this point it stops receiving the ECHO replies. There is no interface with the local-link fe80::2 so I'm not sure how it will reply to the who has?

Any idea as to why this might be happening, or how to resolve this at all?
 
Why don't you use the global address of the router for ipv6_defaultrouter?
 
This is what I was initially told from the support staff:

Code: 
We don't assign /48's to VLANs directly (where the first hop is
the first IP).  There is no point to doing so b/c it precludes you
from further subnetting the /48.  And, if you're not going to subnet
a /48, why do you need the /48 in the first place? :)  That's why
/64's are assigned to VLANs and the enclosing /48 reserved for you
should you ever do a subnet'd setup.

In order to subnet the /48, we route it over your VPS' link-local
address (fe80::/64).
 
What you had mentioned is how I have set this up in the past, however. So I am a little lost as to what is actually causing the problem and why I am able to communicate briefly over IPv6 before it stops...
 
Use the full fe80:: address. I'm guessing fe80::1 gets assigned somewhere else too.
 
I had tried this previously (using the link-local for the em0 interface (fe80::5054:ff:fe27:2836%em0) and it wouldn't send ICMP at all with this.
 
No, you need to use the fe80:: address of the router, not from your own interface.

But don't they supply Router Advertisements? If they do all you have to do is:
Code: 
ifconfig_em0_ipv6="inet6 accept_rtadv"
 
I don't appear to be receiving anything on the interface. Could this possibly be a misconfiguration on their end of some sort?

I have added this to to rc.conf, but I believe this was saying the interface was already accepting router advertisements?

nd6 options=3<PERFORMNUD,ACCEPT_RTADV>


The gateway I was given was fe80::1/64

I am not sure if there is miscommunication between myself and the support staff, or if I am just borking the setup...
 
I just checked my VPS and it too has fe80::1 as a default gateway. I do get router advertisements though so it's possible something isn't working correctly on their side.

The only IPv6 related things in my rc.conf are:
Code: 
ipv6_activate_all_interfaces="YES"
ifconfig_em0_ipv6="inet6 accept_rtadv"
 
I actually had to assign fe80::2 to the em0 interface.

I'll go get my dunce cap now. Thanks for all the help SirDice!
 
You must log in or register to reply here.
Back
Top