My Tumbleweed system does not get a public IPv6 address. The reason is unknown, but it must be a configuration problem, because all other dual boot OS, TW in rescue mode and even the KVM VMs get an IPv6 address on the same hardware. The router is OpenWrt based and seems to work fine. I already tried to copy the network configuration from the rescue system. I use a Wicked based network configuration, set up with YaST. For privacy reasons I replaced some numbers with "xx". # wicked ifstatus br0 br0 up link: #16, state up, mtu 1454 type: bridge, hwaddr xx:xx:xx:xx:xx:xx config: compat:suse:/etc/sysconfig/network/ifcfg-br0 leases: ipv4 dhcp granted leases: ipv6 dhcp requesting addr: ipv4 192.168.xxx.xxx/24 [dhcp] route: ipv4 default via 192.168.xxx.xxx proto dhcp # cat /etc/sysconfig/network/ifcfg-br0 BOOTPROTO='dhcp' STARTMODE='auto' ZONE='home' BRIDGE='yes' BRIDGE_PORTS='eno1' BRIDGE_STP='on' BRIDGE_FORWARDDELAY='2' # ip -6 addr show br0 16: br0: <BROADCAST,MULTICAST,UP,LOWER_UP> mtu 1454 qdisc noqueue state UP group default qlen 1000 inet6 fe80::406e:b2ff:fe1d:ae7/64 scope link valid_lft forever preferred_lft forever # systemctl status wickedd-dhcp6.service ● wickedd-dhcp6.service - wicked DHCPv6 supplicant service Loaded: loaded (/usr/lib/systemd/system/wickedd-dhcp6.service; enabled; vendor preset: disabled) Active: active (running) since Mon 2021-05-31 11:37:15 CEST; 53min ago Main PID: 26380 (wickedd-dhcp6) Tasks: 1 (limit: 4915) CPU: 48ms CGroup: /system.slice/wickedd-dhcp6.service └─26380 /usr/libexec/wicked/bin/wickedd-dhcp6 --systemd --foreground May 31 11:37:15 mybox systemd[1]: Starting wicked DHCPv6 supplicant service... May 31 11:37:15 mybox systemd[1]: Started wicked DHCPv6 supplicant service. May 31 11:38:46 mybox wickedd-dhcp6[26380]: br0: Request to acquire DHCPv6 lease with UUID xxxxxxxx-xxxx-xxxx-xxxx-00001a0> May 31 11:47:16 mybox wickedd-dhcp6[26380]: br0: Request to release DHCPv6 lease using UUID xxxxxxxx-xxxx-xxxx-xxxx-00001a> May 31 11:47:33 mybox wickedd-dhcp6[26380]: br0: Request to acquire DHCPv6 lease with UUID xxxxxxxx-xxxx-xxxx-xxxx-0000200> May 31 12:20:04 mybox wickedd-dhcp6[26380]: br0: Request to release DHCPv6 lease using UUID xxxxxxxx-xxxx-xxxx-xxxx-000020> May 31 12:24:43 mybox wickedd-dhcp6[26380]: br0: Request to acquire DHCPv6 lease with UUID xxxxxxxx-xxxx-xxxx-xxxx-0000340> # ping6 www.heise.de ping6: connect: Network is unreachable # sysctl net.ipv6.conf.all.disable_ipv6 net.ipv6.conf.all.disable_ipv6 = 0 # sysctl net.ipv6.conf.all.forwarding net.ipv6.conf.all.forwarding = 1 Any ideas? Greetings, Björn
On 31. 05. 21, 12:34, Bjoern Voigt wrote:
My Tumbleweed system does not get a public IPv6 address. The reason is unknown, but it must be a configuration problem, because all other dual boot OS, TW in rescue mode and even the KVM VMs get an IPv6 address on the same hardware. The router is OpenWrt based and seems to work fine. I already tried to copy the network configuration from the rescue system. I use a Wicked based network configuration, set up with YaST.
Are you sure the other systems use ipv6+dhcp? What does output of ip -6 addr on the other systems look like?
# sysctl net.ipv6.conf.all.disable_ipv6 net.ipv6.conf.all.disable_ipv6 = 0 # sysctl net.ipv6.conf.all.forwarding net.ipv6.conf.all.forwarding = 1
What about net.ipv6.conf.all.autoconf and net.ipv6.conf.all.accept_ra here and on other systems? regards, -- js suse labs
Jiri Slaby wrote:
On 31. 05. 21, 12:34, Bjoern Voigt wrote:
My Tumbleweed system does not get a public IPv6 address. The reason is unknown, but it must be a configuration problem, because all other dual boot OS, TW in rescue mode and even the KVM VMs get an IPv6 address on the same hardware. The router is OpenWrt based and seems to work fine. I already tried to copy the network configuration from the rescue system. I use a Wicked based network configuration, set up with YaST.
Are you sure the other systems use ipv6+dhcp? What does output of ip -6 addr on the other systems look like? My Linux Minut 19 KVM VM uses Network Manager with the following DHCPv6 configuration:
/sbin/dhclient -d -q -sf /usr/lib/NetworkManager/nm-dhcp-helper -pf /run/dhclient-ens3.pid -lf /var/lib/NetworkManager/dhclient-xxxxxxxx-xxxx-xxxx-xxxx-xxxxxxxxxxxx-ens3.lease -cf /var/lib/NetworkManager/dhclient-ens3.conf ens3 /var/lib/NetworkManager/dhclient-ens3.conf: # Created by NetworkManager send fqdn.fqdn "mybox-mint"; # added by NetworkManager send fqdn.server-update on; also request dhcp6.name-servers; also request dhcp6.domain-search; also request dhcp6.client-id; # ip -6 addr 1: lo: <LOOPBACK,UP,LOWER_UP> mtu 65536 state UNKNOWN qlen 1000 inet6 ::1/128 scope host valid_lft forever preferred_lft forever 2: ens3: <BROADCAST,MULTICAST,UP,LOWER_UP> mtu 1500 state UP qlen 1000 inet6 fdc6:35fa:xxxx::xxx/128 scope global noprefixroute valid_lft forever preferred_lft forever inet6 2a01:5241:xxx:xxxx::xxx/128 scope global dynamic noprefixroute valid_lft 76615sec preferred_lft 76615sec inet6 fdc6:35fa:xxxx:0:xxxx:xxxx:xxxx:xxxx/64 scope global temporary dynamic valid_lft 604575sec preferred_lft 85590sec inet6 fdc6:35fa:xxxx:0:xxxx:xxxx:xxxx:xxxx/64 scope global mngtmpaddr noprefixroute valid_lft forever preferred_lft forever inet6 2a01:5241:xxx:xxxx:xxxx:xxxx:xxxx:xxxx/64 scope global temporary dynamic valid_lft 76614sec preferred_lft 76614sec inet6 2a01:5241:xxx:xxxx:xxxx:xxxx:xxxx:xxxx/64 scope global dynamic mngtmpaddr noprefixroute valid_lft 76614sec preferred_lft 76614sec inet6 fe80::f4b0:xxxx:xxxx:xxxx/64 scope link noprefixroute valid_lft forever preferred_lft forever net.ipv6.conf.all.accept_ra = 1 net.ipv6.conf.ens3.forwarding = 0
# sysctl net.ipv6.conf.all.disable_ipv6 net.ipv6.conf.all.disable_ipv6 = 0 # sysctl net.ipv6.conf.all.forwarding net.ipv6.conf.all.forwarding = 1
What about net.ipv6.conf.all.autoconf and net.ipv6.conf.all.accept_ra here and on other systems? net.ipv6.conf.all.autoconf = 1 net.ipv6.conf.all.forwarding = 1
I found here https://bugzilla.opensuse.org/show_bug.cgi?id=1166358 two tips which together solve the problem: sysctl -w net.ipv6.conf.all.accept_ra=2 sysctl -w net.ipv6.conf.br0.forwarding=0 Unfortunately the bug report is closed with RESOLVED INVALID. I do not agree that the report has been closed. There bug report contains only temporary solutions (with sysctl or echo). Also YaST oder Wicked could be patched, so that they can detect an inconsistent configuration (DHCPv6 is activated, but net.ipv6.conf is wrong) and report this to the user (YaST) or log this (Wicked). Jiri, do you agree and could you reopen the bug report? Greetings, Björn* *
On 01. 06. 21, 11:48, Bjoern Voigt wrote:
I found here https://bugzilla.opensuse.org/show_bug.cgi?id=1166358 two tips which together solve the problem:
sysctl -w net.ipv6.conf.all.accept_ra=2 sysctl -w net.ipv6.conf.br0.forwarding=0
Unfortunately the bug report is closed with RESOLVED INVALID. I do not agree that the report has been closed.
There bug report contains only temporary solutions (with sysctl or echo). Also YaST oder Wicked could be patched, so that they can detect an inconsistent configuration (DHCPv6 is activated, but net.ipv6.conf is wrong) and report this to the user (YaST) or log this (Wicked).
Jiri, do you agree and could you reopen the bug report?
I suggest you to "Clone This Bug", instead. regards, -- js suse labs
participants (2)
-
Bjoern Voigt -
Jiri Slaby