Несколько адресов IPv6 присвоены единственному интерфейсу
Не уверенный существует правильный ответ на этот tbh.
RAID не является резервным копированием, RAID является дублированием, таким образом, поможет Вам, если диск перестанет работать, но если Вы удаляете что-то, что Вы не хотели Вам, все еще нуждаются в резервном копировании, таким образом, необходимо рассмотреть, как Вы сделали бы это, если бы у Вас не было резервного диска в сервере.
Я не использовал DL320 G5 а именно, но предположение, что он похож на другие Умные Контроллеры массива, необходимо найти, что это - просто случай включения нового диска.
DHCPv6 does not need to be configured to issue IPv6 addresses for a network to function. Rather, each machine assigns itself a link local address (within [FE80::]/16), optionally based on its MAC, but possibly random. It then listens for router advertisements, and based on the advertised prefixes, it will assign itself one or more addresses using the same logic. Typically, a host will assign itself one or more random addresses which it will use to initiate connections, thus protecting its privacy (eg. not disclosing its MAC to all hosts on the internet, and not keeping a consistent network address); it will keep a consistent one as well, for other hosts to initiate connections to. Hosts, having discovered a prefix, will then query DHCPv6 to figure out stuff like their DNS servers.
That would be where the extra addresses are coming from (your use of disabling router discovery confirms this). For it to assign itself an address within [1024::]/64, some router on the network (running a router advertisement daemon) would have to be broadcasting that subnet; a network capture could tell you which. The packets will show up as ICMPv6 NDP router advertisements.
The extra addresses look to me like privacy addresses, which really should be disabled in any sort of business environment. Try disabling IPv6 privacy addresses:
netsh interface ipv6 set privacy state=disabled store=active
netsh interface ipv6 set privacy state=disabled store=persistent
netsh interface ipv6 set global randomizeidentifiers=disabled store=active
netsh interface ipv6 set global randomizeidentifiers=disabled store=persistent
This configuration change requires a restart.
Unfortunately, there are no Group Policy objects that I'm aware of to make this change, which is a serious oversight on Microsoft's part.
1000::/4 is unassigned. I would imagine that it is discovering that address via ND. Fire up wireshark, set a capture filter of icmp6 and see if a local host is transmitting prefix advertisements.