大网数据广告活动
服务器运维问答

IPv6 broken and cannot even ping the own address

· 52CE 我爱测

I have broken IPv6 in some LXC containers, where the strange thing is, that i cannot even ping the address on the interface. $ ip -6 r prefix::/112 dev eth0 proto kernel metric 256 $ ip -6 a show dev eth0 69: eth0@if70: <BROADCAS…

问题原文

原帖链接 · allo · CC BY-SA 3.0

I have broken IPv6 in some LXC containers, where the strange thing is, that i cannot even ping the address on the interface.

$ ip -6 r
prefix::/112 dev eth0  proto kernel  metric 256
$ ip -6 a show dev eth0
69: eth0@if70: <BROADCAST,MULTICAST,UP,LOWER_UP> mtu 1500 state UP qlen 1000
inet6 prefix::3/112 scope global tentative dadfailed 
   valid_lft forever preferred_lft forever

Now a ping6 prefix::3 fails with

64 bytes from localhost (::1): Destination unreachable: Address unreachable

I tried disabling the firewall (which shouldn't have any effect, as the local packets should not hit the INPUT chain), but this didn't change anything.

Another thing which is strange: The interface did not get a link-local address.

The OS is a rather minimal installation of debian jessie. Other containers with debian jessie installations work fine with IPv6.

已采纳解答

原帖链接 · Torsten-Sven Urbas · CC BY-SA 3.0

If you look at the output you posted, you will notice tentative dadfailed.

This means that the IPv6 duplicate address detection has kicked in for this IP address. Are you absolutely sure that the IP is not in use anywhere else on the same network segment?

You can run lxc-ls -f on your LXC hostnodes to figure out if some other container might be using that IP.

摘自公开归档,保留原文语言,版式经整理,外部媒体与邮箱地址已移除。

资料来源与适用范围

资料来源
Server Fault
适用范围
以原帖系统、软件版本与发布日期为准(2017-01-31)
内容语言
中文;命令、路径和错误标识保留技术原文
本地收录日期
2026-11-15
授权说明
CC BY-SA 3.0

正文与所需资源由本站本地存储。查看来源及更新记录 ↗

请按适用版本和实际日志确认原因;本站没有对每一种系统或设备逐项执行文中的操作。