我在树莓派 3 上运行着 Ubuntu 22.04 LTS（GNU/Linux 5.15.0-1008-raspi armv7l）。
由于某种原因，在正常运行约 2 天后，它开始随机删除其全局 IPv6 地址。一旦计算机收到 ICMP RA 数据包，IPv6 地址就会恢复。
重新启动系统可确保接下来两天都不会发生任何事故。
这在新安装的系统上也发生过。
我们的路由器宣传路由器寿命为 1800 秒，前缀寿命为 3600 秒。Tcpdump 表示我们的路由器每 600 秒发送一次 ICMP RA。即使使用 watch 来观察 的输出ip addr，当地址被删除时，倒计时上仍然剩下 4 位数字。（恰巧 IPv6 地址在收到 RA 后 2 秒内丢失）
当系统工作时，ip addr看起来像
1: lo: <LOOPBACK,UP,LOWER_UP> mtu 65536 qdisc noqueue state UNKNOWN group default qlen 1000
link/loopback 00:00:00:00:00:00 brd 00:00:00:00:00:00
inet 127.0.0.1/8 scope host lo
valid_lft forever preferred_lft forever
inet6 ::1/128 scope host
valid_lft forever preferred_lft forever
2: eth0: <BROADCAST,MULTICAST,UP,LOWER_UP> mtu 1500 qdisc fq_codel state UP group default qlen 1000
link/ether zz:zz:zz:zz:zz:zz brd ff:ff:ff:ff:ff:ff
inet 192.168.178.53/24 metric 100 brd 192.168.178.255 scope global dynamic eth0
valid_lft 624462sec preferred_lft 624462sec
inet6 yyyy:yyyy:yyyy:yyyy:xxxx:xxxx:xxxx:xxxx/64 scope global dynamic mngtmpaddr noprefixroute
valid_lft 7197sec preferred_lft 3597sec
inet6 fe80::xxxx:xxxx:xxxx:xxxx/64 scope link
valid_lft forever preferred_lft forever
3: wlan0: <BROADCAST,MULTICAST> mtu 1500 qdisc noop state DOWN group default qlen 1000
link/ether b8:27:eb:a3:c0:93 brd ff:ff:ff:ff:ff:ff
4: teredo: <POINTOPOINT,MULTICAST,NOARP,UP,LOWER_UP> mtu 1280 qdisc fq_codel state UNKNOWN group default qlen 500
link/none
inet6 2001:0:zzzz:zzzz:zzzz:zzzz/32 scope global
valid_lft forever preferred_lft forever
inet6 fe80::ffff:ffff:ffff/64 scope link
valid_lft forever preferred_lft forever
inet6 fe80::9a37:a9a:298d:ac5c/64 scope link stable-privacy
valid_lft forever preferred_lft forever
当它坏了的时候，系统看起来就像
1: lo: <LOOPBACK,UP,LOWER_UP> mtu 65536 qdisc noqueue state UNKNOWN group default qlen 1000
link/loopback 00:00:00:00:00:00 brd 00:00:00:00:00:00
inet 127.0.0.1/8 scope host lo
valid_lft forever preferred_lft forever
inet6 ::1/128 scope host
valid_lft forever preferred_lft forever
2: eth0: <BROADCAST,MULTICAST,UP,LOWER_UP> mtu 1500 qdisc fq_codel state UP group default qlen 1000
link/ether zz:zz:zz:zz:zz:zz brd ff:ff:ff:ff:ff:ff
inet 192.168.178.53/24 metric 100 brd 192.168.178.255 scope global dynamic eth0
valid_lft 622714sec preferred_lft 622714sec
inet6 fe80::xxxx:xxxx:xxxx:xxxx/64 scope link
valid_lft forever preferred_lft forever
3: wlan0: <BROADCAST,MULTICAST> mtu 1500 qdisc noop state DOWN group default qlen 1000
link/ether b8:27:eb:a3:c0:93 brd ff:ff:ff:ff:ff:ff
4: teredo: <POINTOPOINT,MULTICAST,NOARP,UP,LOWER_UP> mtu 1280 qdisc fq_codel state UNKNOWN group default qlen 500
link/none
inet6 2001:0:zzzz:zzzz:zzzz:zzzz/32 scope global
valid_lft forever preferred_lft forever
inet6 fe80::ffff:ffff:ffff/64 scope link
valid_lft forever preferred_lft forever
inet6 fe80::9a37:a9a:298d:ac5c/64 scope link stable-privacy
valid_lft forever preferred_lft forever
系统日志没有给出关于此问题的任何线索（仅有症状）：[完整文件]
Jun 7 17:17:57 rp02 systemd[1]: Starting Daily apt download activities...
Jun 7 17:18:46 rp02 systemd[1]: apt-daily.service: Deactivated successfully.
Jun 7 17:18:46 rp02 systemd[1]: Finished Daily apt download activities.
Jun 7 17:18:46 rp02 systemd[1]: apt-daily.service: Consumed 47.031s CPU time.
Jun 7 18:15:28 rp02 systemd[1]: Created slice User Slice of UID 1000.
Jun 7 18:15:28 rp02 systemd[1]: Starting User Runtime Directory /run/user/1000...
Jun 7 18:15:28 rp02 systemd[1]: Finished User Runtime Directory /run/user/1000.
Jun 7 18:15:28 rp02 systemd[1]: Starting User Manager for UID 1000...
Jun 7 18:15:30 rp02 systemd[5522]: Queued start job for default target Main User Target.
Jun 7 18:15:30 rp02 systemd[5522]: Created slice User Application Slice.
Jun 7 18:15:30 rp02 systemd[5522]: Reached target Paths.
Jun 7 18:15:30 rp02 systemd[5522]: Reached target Timers.
Jun 7 18:15:30 rp02 systemd[5522]: Starting D-Bus User Message Bus Socket...
Jun 7 18:15:30 rp02 systemd[5522]: Listening on GnuPG network certificate management daemon.
Jun 7 18:15:30 rp02 systemd[5522]: Listening on GnuPG cryptographic agent and passphrase cache (access for web browsers).
Jun 7 18:15:30 rp02 systemd[5522]: Listening on GnuPG cryptographic agent and passphrase cache (restricted).
Jun 7 18:15:30 rp02 systemd[5522]: Listening on GnuPG cryptographic agent (ssh-agent emulation).
Jun 7 18:15:30 rp02 systemd[5522]: Listening on GnuPG cryptographic agent and passphrase cache.
Jun 7 18:15:30 rp02 systemd[5522]: Listening on debconf communication socket.
Jun 7 18:15:30 rp02 systemd[5522]: Listening on REST API socket for snapd user session agent.
Jun 7 18:15:30 rp02 systemd[5522]: Listening on D-Bus User Message Bus Socket.
Jun 7 18:15:30 rp02 systemd[5522]: Reached target Sockets.
Jun 7 18:15:30 rp02 systemd[5522]: Reached target Basic System.
Jun 7 18:15:30 rp02 systemd[5522]: Reached target Main User Target.
Jun 7 18:15:30 rp02 systemd[5522]: Startup finished in 2.101s.
Jun 7 18:15:30 rp02 systemd[1]: Started User Manager for UID 1000.
Jun 7 18:15:30 rp02 systemd[1]: Started Session 59 of User fernando.
Jun 7 18:17:01 rp02 CRON[5619]: (root) CMD ( cd / && run-parts --report /etc/cron.hourly)
Jun 7 18:23:42 rp02 avahi-daemon[699]: Withdrawing address record for yyyy:yyyy:yyyy:yyyy:xxxx:xxxx:xxxx:xxxx on eth0.
Jun 7 18:23:42 rp02 avahi-daemon[699]: Leaving mDNS multicast group on interface eth0.IPv6 with address yyyy:yyyy:yyyy:yyyy:xxxx:xxxx:xxxx:xxxx.
Jun 7 18:23:42 rp02 avahi-daemon[699]: Joining mDNS multicast group on interface eth0.IPv6 with address fe80::xxxx:xxxx:xxxx:xxxx.
Jun 7 18:23:42 rp02 avahi-daemon[699]: Registering new address record for fe80::xxxx:xxxx:xxxx:xxxx on eth0.*.
Jun 7 18:29:48 rp02 systemd[1]: Starting Download data for packages that failed at package install time...
Jun 7 18:29:48 rp02 systemd[1]: update-notifier-download.service: Deactivated successfully.
Jun 7 18:29:48 rp02 systemd[1]: Finished Download data for packages that failed at package install time.
Jun 7 18:32:43 rp02 avahi-daemon[699]: Leaving mDNS multicast group on interface eth0.IPv6 with address fe80::xxxx:xxxx:xxxx:xxxx.
Jun 7 18:32:43 rp02 avahi-daemon[699]: Joining mDNS multicast group on interface eth0.IPv6 with address yyyy:yyyy:yyyy:yyyy:xxxx:xxxx:xxxx:xxxx.
Jun 7 18:32:43 rp02 avahi-daemon[699]: Registering new address record for yyyy:yyyy:yyyy:yyyy:xxxx:xxxx:xxxx:xxxx on eth0.*.
Jun 7 18:32:43 rp02 avahi-daemon[699]: Withdrawing address record for fe80::xxxx:xxxx:xxxx:xxxx on eth0.
Jun 7 18:35:29 rp02 snapd[721]: storehelpers.go:722: cannot refresh: snap has no updates available: "certbot", "core20", "lxd", "snapd"
Jun 7 18:35:29 rp02 snapd[721]: autorefresh.go:539: auto-refresh: all snaps are up-to-date
Jun 7 18:39:57 rp02 systemd[1]: Starting Cleanup of Temporary Directories...
Jun 7 18:39:57 rp02 systemd[1]: systemd-tmpfiles-clean.service: Deactivated successfully.
Jun 7 18:39:57 rp02 systemd[1]: Finished Cleanup of Temporary Directories.
Jun 7 18:42:07 rp02 avahi-daemon[699]: Withdrawing address record for yyyy:yyyy:yyyy:yyyy:xxxx:xxxx:xxxx:xxxx on eth0.
Jun 7 18:42:07 rp02 avahi-daemon[699]: Leaving mDNS multicast group on interface eth0.IPv6 with address yyyy:yyyy:yyyy:yyyy:xxxx:xxxx:xxxx:xxxx.
Jun 7 18:42:07 rp02 avahi-daemon[699]: Joining mDNS multicast group on interface eth0.IPv6 with address fe80::xxxx:xxxx:xxxx:xxxx.
Jun 7 18:42:07 rp02 avahi-daemon[699]: Registering new address record for fe80::xxxx:xxxx:xxxx:xxxx on eth0.*.
Jun 7 18:43:54 rp02 avahi-daemon[699]: Leaving mDNS multicast group on interface eth0.IPv6 with address fe80::xxxx:xxxx:xxxx:xxxx.
Jun 7 18:43:54 rp02 avahi-daemon[699]: Joining mDNS multicast group on interface eth0.IPv6 with address yyyy:yyyy:yyyy:yyyy:xxxx:xxxx:xxxx:xxxx.
Jun 7 18:43:54 rp02 avahi-daemon[699]: Registering new address record for yyyy:yyyy:yyyy:yyyy:xxxx:xxxx:xxxx:xxxx on eth0.*.
Jun 7 18:43:54 rp02 avahi-daemon[699]: Withdrawing address record for fe80::xxxx:xxxx:xxxx:xxxx on eth0.
Jun 7 18:53:03 rp02 avahi-daemon[699]: Withdrawing address record for yyyy:yyyy:yyyy:yyyy:xxxx:xxxx:xxxx:xxxx on eth0.
Jun 7 18:53:03 rp02 avahi-daemon[699]: Leaving mDNS multicast group on interface eth0.IPv6 with address yyyy:yyyy:yyyy:yyyy:xxxx:xxxx:xxxx:xxxx.
Jun 7 18:53:03 rp02 avahi-daemon[699]: Joining mDNS multicast group on interface eth0.IPv6 with address fe80::xxxx:xxxx:xxxx:xxxx.
Jun 7 18:53:03 rp02 avahi-daemon[699]: Registering new address record for fe80::xxxx:xxxx:xxxx:xxxx on eth0.*.
Jun 7 19:02:33 rp02 avahi-daemon[699]: Leaving mDNS multicast group on interface eth0.IPv6 with address fe80::xxxx:xxxx:xxxx:xxxx.
Jun 7 19:02:33 rp02 avahi-daemon[699]: Joining mDNS multicast group on interface eth0.IPv6 with address yyyy:yyyy:yyyy:yyyy:xxxx:xxxx:xxxx:xxxx.
Jun 7 19:02:33 rp02 avahi-daemon[699]: Registering new address record for yyyy:yyyy:yyyy:yyyy:xxxx:xxxx:xxxx:xxxx on eth0.*.
Jun 7 19:02:33 rp02 avahi-daemon[699]: Withdrawing address record for fe80::xxxx:xxxx:xxxx:xxxx on eth0.
Jun 7 19:03:10 rp02 avahi-daemon[699]: Withdrawing address record for yyyy:yyyy:yyyy:yyyy:xxxx:xxxx:xxxx:xxxx on eth0.
Jun 7 19:03:10 rp02 avahi-daemon[699]: Leaving mDNS multicast group on interface eth0.IPv6 with address yyyy:yyyy:yyyy:yyyy:xxxx:xxxx:xxxx:xxxx.
Jun 7 19:03:10 rp02 avahi-daemon[699]: Joining mDNS multicast group on interface eth0.IPv6 with address fe80::xxxx:xxxx:xxxx:xxxx.
Jun 7 19:03:10 rp02 avahi-daemon[699]: Registering new address record for fe80::xxxx:xxxx:xxxx:xxxx on eth0.*.
Jun 7 19:10:43 rp02 avahi-daemon[699]: Leaving mDNS multicast group on interface eth0.IPv6 with address fe80::xxxx:xxxx:xxxx:xxxx.
Jun 7 19:10:43 rp02 avahi-daemon[699]: Joining mDNS multicast group on interface eth0.IPv6 with address yyyy:yyyy:yyyy:yyyy:xxxx:xxxx:xxxx:xxxx.
Jun 7 19:10:43 rp02 avahi-daemon[699]: Registering new address record for yyyy:yyyy:yyyy:yyyy:xxxx:xxxx:xxxx:xxxx on eth0.*.
Jun 7 19:10:43 rp02 avahi-daemon[699]: Withdrawing address record for fe80::xxxx:xxxx:xxxx:xxxx on eth0.
Jun 7 19:13:35 rp02 systemd[1]: Starting Service for snap application certbot.renew...
Jun 7 19:13:42 rp02 systemd[1]: snap.certbot.renew.service: Deactivated successfully.
Jun 7 19:13:42 rp02 systemd[1]: Finished Service for snap application certbot.renew.
Jun 7 19:13:42 rp02 systemd[1]: snap.certbot.renew.service: Consumed 6.641s CPU time.
Jun 7 19:17:01 rp02 CRON[5717]: (root) CMD ( cd / && run-parts --report /etc/cron.hourly)
Jun 7 19:19:41 rp02 avahi-daemon[699]: Withdrawing address record for yyyy:yyyy:yyyy:yyyy:xxxx:xxxx:xxxx:xxxx on eth0.
Jun 7 19:19:41 rp02 avahi-daemon[699]: Leaving mDNS multicast group on interface eth0.IPv6 with address yyyy:yyyy:yyyy:yyyy:xxxx:xxxx:xxxx:xxxx.
Jun 7 19:19:41 rp02 avahi-daemon[699]: Joining mDNS multicast group on interface eth0.IPv6 with address fe80::xxxx:xxxx:xxxx:xxxx.
Jun 7 19:19:41 rp02 avahi-daemon[699]: Registering new address record for fe80::xxxx:xxxx:xxxx:xxxx on eth0.*.
Jun 7 19:20:55 rp02 avahi-daemon[699]: Leaving mDNS multicast group on interface eth0.IPv6 with address fe80::xxxx:xxxx:xxxx:xxxx.
Jun 7 19:20:55 rp02 avahi-daemon[699]: Joining mDNS multicast group on interface eth0.IPv6 with address yyyy:yyyy:yyyy:yyyy:xxxx:xxxx:xxxx:xxxx.
重启两天就能解决问题，因此很难尝试任何方法
进一步调查并将 systemd 转为调试模式表明我的问题基本上https://archlinuxarm.org/forum/viewtopic.php?f=60&t=15566&sid=cf1d054d50c16df15c74079e2a6871dd&start=10。但这对我没有帮助，因为 systemd 没有可以降级到的版本 248