Create a Post
cancel
Showing results for 
Search instead for 
Did you mean: 
Hugo_vd_Kooij
Advisor

Weird IPv6 issue in GAIA

I have a weird issue in GAIA. I can't even ping myself. This is a rather fresh R80.10 (Take 91) SmartCenter.

[Expert@fwmgmt:0]# ip addr show
1: lo: <LOOPBACK,UP,LOWER_UP> mtu 16436 qdisc noqueue
   link/loopback 00:00:00:00:00:00 brd ff:ff:ff:ff:ff:ff
   inet 127.0.0.1/8 brd 127.255.255.255 scope host lo
   inet6 ::1/128 scope host
      valid_lft forever preferred_lft forever
2: eth0: <BROADCAST,MULTICAST,UP,LOWER_UP> mtu 1500 qdisc pfifo_fast qlen 1000
   link/ether 00:0c:29:6e:6b:2a brd ff:ff:ff:ff:ff:ff
   inet 10.0.0.1/24 brd 10.0.0.255 scope global eth0
   inet6 2001:470:c9f0:100::1/64 scope global tentative
      valid_lft forever preferred_lft forever
   inet6 fe80::20c:29ff:fe6e:6b2a/64 scope link
      valid_lft forever preferred_lft forever
3: sit0: <NOARP> mtu 1480 qdisc noop
   link/sit 0.0.0.0 brd 0.0.0.0


[Expert@fwmgmt:0]# ip -f inet6 route show
2001:470:c9f0:100::/64 dev eth0 metric 256 expires 21259440sec mtu 1500 advmss 1440 hoplimit 4294967295
2000::/3 via 2001:470:c9f0:100::254 dev eth0 proto routed metric 1024 expires 21259446sec mtu 1500 advmss 1440 hoplimit 4294967295
fe80::/64 dev eth0 metric 256 expires 21259440sec mtu 1500 advmss 1440 hoplimit 4294967295
default via 2001:470:c9f0:100::254 dev eth0 proto routed metric 1024 expires 21259446sec mtu 1500 advmss 1440 hoplimit 4294967295
[Expert@fwmgmt:0]# ping6 2001:470:c9f0:100::1
PING 2001:470:c9f0:100::1(2001:470:c9f0:100::1) 56 data bytes
From fe80::20c:29ff:fe6e:6b2a icmp_seq=1 Destination unreachable: Address unreachable
From fe80::20c:29ff:fe6e:6b2a icmp_seq=2 Destination unreachable: Address unreachable
From fe80::20c:29ff:fe6e:6b2a icmp_seq=3 Destination unreachable: Address unreachable
From fe80::20c:29ff:fe6e:6b2a icmp_seq=5 Destination unreachable: Address unreachable
From fe80::20c:29ff:fe6e:6b2a icmp_seq=6 Destination unreachable: Address unreachable
From fe80::20c:29ff:fe6e:6b2a icmp_seq=7 Destination unreachable: Address unreachable
From fe80::20c:29ff:fe6e:6b2a icmp_seq=9 Destination unreachable: Address unreachable
From fe80::20c:29ff:fe6e:6b2a icmp_seq=10 Destination unreachable: Address unreachable
From fe80::20c:29ff:fe6e:6b2a icmp_seq=11 Destination unreachable: Address unreachable

--- 2001:470:c9f0:100::1 ping statistics ---
13 packets transmitted, 0 received, +9 errors, 100% packet loss, time 11999ms

Has anyone seen this before?

I have a silly feeling it might be known issue of the IPv6 stack in the Linux kernel.

Based on kernel/git/torvalds/linux.git - Linux kernel source tree 

<< We make miracles happen while you wait. The impossible jobs take just a wee bit longer. >>
0 Kudos
4 Replies
Hugo_vd_Kooij
Advisor

The issue is ..... sort of solved.

It turns out the IPv6 address is down because IPv6 has a duplicate IP detection algorithm.

The IP address does not actually exist but the machine is convinced that it is still there. My guess is that this is some sort of weird combination bug that will happen only every 13th blue moon.

I changed the IP address to another address and now the machine is happy and I can even connect to my Smart Center over IPv6 from other networks.

<< We make miracles happen while you wait. The impossible jobs take just a wee bit longer. >>
PhoneBoy
Admin
Admin

That's...bizarre. 

0 Kudos
Hugo_vd_Kooij
Advisor

The TAC engineer was learning as well as we did the troubleshooting.

The tentative keyword was very usefull in finding out more.

A usefull side effect of the work around: Nothing lives on the IP address in the post anymore 😉

<< We make miracles happen while you wait. The impossible jobs take just a wee bit longer. >>
0 Kudos
Hugo_vd_Kooij
Advisor

Normal behaviour is so .... dull. Why not have a bit of fun while you work 😉

<< We make miracles happen while you wait. The impossible jobs take just a wee bit longer. >>
0 Kudos

Leaderboard

Epsum factorial non deposit quid pro quo hic escorol.

Upcoming Events

    CheckMates Events