In Yast LAN configuration, it's possible to create a VLAN interface with different IP addresses than the native LAN. However, one thing I've noticed is the VLANs have the same IPv6 link local address as the native LAN. Is it possible to change the link local address? Adding another doesn't work, as a valid subnet mask is required, but link local addresses don't use them. Having the same link local address, for two interfaces, can cause problems when used as a router, as the link local address is often used as the router address. tnx jk -- To unsubscribe, e-mail: opensuse+unsubscribe@opensuse.org To contact the owner, e-mail: opensuse+owner@opensuse.org
On 12/23/2014 03:22 PM, James Knott wrote:
In Yast LAN configuration, it's possible to create a VLAN interface with different IP addresses than the native LAN. However, one thing I've noticed is the VLANs have the same IPv6 link local address as the native LAN. Is it possible to change the link local address? Adding another doesn't work, as a valid subnet mask is required, but link local addresses don't use them. Having the same link local address, for two interfaces, can cause problems when used as a router, as the link local address is often used as the router address.
tnx jk
I just tried creating another VLAN and it does not get a link local address and when I try to add an IPv6 route to it, I get an error "RTNETLINK answers: No such device", even though I am able to add an IPv4 address to it and ping it. -- To unsubscribe, e-mail: opensuse+unsubscribe@opensuse.org To contact the owner, e-mail: opensuse+owner@opensuse.org
В Tue, 23 Dec 2014 15:22:37 -0500
James Knott
In Yast LAN configuration, it's possible to create a VLAN interface with different IP addresses than the native LAN. However, one thing I've noticed is the VLANs have the same IPv6 link local address as the native LAN. Is it possible to change the link local address?
No, link-local is by definition based on MAC address. You can manually add other addresses with link scope.
Adding another doesn't work, as a valid subnet mask is required, but link local addresses don't use them. Having the same link local address, for two interfaces, can cause problems when used as a router, as the link local address is often used as the router address.
Can you give example where it is a problem? -- To unsubscribe, e-mail: opensuse+unsubscribe@opensuse.org To contact the owner, e-mail: opensuse+owner@opensuse.org
On 12/25/2014 02:12 PM, Andrei Borzenkov wrote:
В Tue, 23 Dec 2014 15:22:37 -0500 James Knott
пишет: In Yast LAN configuration, it's possible to create a VLAN interface with different IP addresses than the native LAN. However, one thing I've noticed is the VLANs have the same IPv6 link local address as the native LAN. Is it possible to change the link local address? No, link-local is by definition based on MAC address. You can manually add other addresses with link scope.
I wasn't familiar with that link scope. I don't see any mention of scope in the IP man page. However, it still appears it can't be done in Yast LAN.
Adding another doesn't work, as a valid subnet mask is required, but link local addresses don't use them. Having the same link local address, for two interfaces, can cause problems when used as a router, as the link local address is often used as the router address.
Can you give example where it is a problem?
I was thinking about setting up routes, but after posting my message, I remembered that it's necessary to use the interface ID in addition to the link local address, so no problem. -- To unsubscribe, e-mail: opensuse+unsubscribe@opensuse.org To contact the owner, e-mail: opensuse+owner@opensuse.org
В Fri, 26 Dec 2014 10:47:51 -0500
James Knott
On 12/25/2014 02:12 PM, Andrei Borzenkov wrote:
В Tue, 23 Dec 2014 15:22:37 -0500 James Knott
пишет: In Yast LAN configuration, it's possible to create a VLAN interface with different IP addresses than the native LAN. However, one thing I've noticed is the VLANs have the same IPv6 link local address as the native LAN. Is it possible to change the link local address? No, link-local is by definition based on MAC address. You can manually add other addresses with link scope.
I wasn't familiar with that link scope. I don't see any mention of scope in the IP man page. However, it still appears it can't be done in Yast LAN.
IPv6 address scope is defined by format prefix (initial bits). Any address that starts with FE80 is link scoped. It should never be forwarded or routed by any gateway.
Adding another doesn't work, as a valid subnet mask is required, but link local addresses don't use them. Having the same link local address, for two interfaces, can cause problems when used as a router, as the link local address is often used as the router address.
Can you give example where it is a problem?
I was thinking about setting up routes, but after posting my message, I remembered that it's necessary to use the interface ID in addition to the link local address, so no problem.
Yes, standard explicitly permits identical link local addresses on multiple interfaces. IPv6 defines notion of "zone" to which interface is connected and disambiguates addresses by qualifying them with zone identifier. -- To unsubscribe, e-mail: opensuse+unsubscribe@opensuse.org To contact the owner, e-mail: opensuse+owner@opensuse.org
On 12/26/2014 03:50 PM, Andrei Borzenkov wrote:
В Fri, 26 Dec 2014 10:47:51 -0500 James Knott
пишет: On 12/25/2014 02:12 PM, Andrei Borzenkov wrote:
В Tue, 23 Dec 2014 15:22:37 -0500 James Knott
пишет: In Yast LAN configuration, it's possible to create a VLAN interface with different IP addresses than the native LAN. However, one thing I've noticed is the VLANs have the same IPv6 link local address as the native LAN. Is it possible to change the link local address? No, link-local is by definition based on MAC address. You can manually add other addresses with link scope. I wasn't familiar with that link scope. I don't see any mention of scope in the IP man page. However, it still appears it can't be done in Yast LAN.
IPv6 address scope is defined by format prefix (initial bits). Any address that starts with FE80 is link scoped. It should never be forwarded or routed by any gateway.
That I know.
Adding another doesn't work, as a valid subnet mask is required, but link local addresses don't use them. Having the same link local address, for two interfaces, can cause problems when used as a router, as the link local address is often used as the router address.
Can you give example where it is a problem? I was thinking about setting up routes, but after posting my message, I remembered that it's necessary to use the interface ID in addition to the link local address, so no problem.
Yes, standard explicitly permits identical link local addresses on multiple interfaces. IPv6 defines notion of "zone" to which interface is connected and disambiguates addresses by qualifying them with zone identifier.
What seems to be more of an issue at the moment is the 2nd VLAN I created doesn't get a link local address and when I try to forward IPv6 via the Interface, I get an error "RTNETLINK answers: No such device". This, of course, makes it impossible to run IPv6 over that VLAN. I already have one VLAN where IPv6 works fine. There is also another problem, which I have file a bugzilla report about is that when an interface is set to run IPv4 DHCP, it no longer gets a link local address and is again useless for IPv6. This is with the ifup method. The problem doesn't happen with network manager. -- To unsubscribe, e-mail: opensuse+unsubscribe@opensuse.org To contact the owner, e-mail: opensuse+owner@opensuse.org
В Fri, 26 Dec 2014 16:44:01 -0500
James Knott
What seems to be more of an issue at the moment is the 2nd VLAN I created doesn't get a link local address and when I try to forward IPv6 via the Interface, I get an error "RTNETLINK answers: No such device". This, of course, makes it impossible to run IPv6 over that VLAN. I already have one VLAN where IPv6 works fine. There is also another problem, which I have file a bugzilla report about is that when an interface is set to run IPv4 DHCP, it no longer gets a link local address and is again useless for IPv6. This is with the ifup method.
You did not mention openSUSE version. I cannot reproduce it in 13.2. I created VLAN, set it to DHCP and "Use DHCPv4 only" and it gets IPv6 link local address on startup. -- To unsubscribe, e-mail: opensuse+unsubscribe@opensuse.org To contact the owner, e-mail: opensuse+owner@opensuse.org
On 12/27/2014 03:04 AM, Andrei Borzenkov wrote:
В Fri, 26 Dec 2014 16:44:01 -0500 James Knott
пишет: What seems to be more of an issue at the moment is the 2nd VLAN I created doesn't get a link local address and when I try to forward IPv6 via the Interface, I get an error "RTNETLINK answers: No such device". This, of course, makes it impossible to run IPv6 over that VLAN. I already have one VLAN where IPv6 works fine. There is also another problem, which I have file a bugzilla report about is that when an interface is set to run IPv4 DHCP, it no longer gets a link local address and is again useless for IPv6. This is with the ifup method. You did not mention openSUSE version. I cannot reproduce it in 13.2. I created VLAN, set it to DHCP and "Use DHCPv4 only" and it gets IPv6 link local address on startup.
Sorry, I'm currently running 13.1 on my firewall/router. I prefer to run the Evergreen versions for my firewall. -- To unsubscribe, e-mail: opensuse+unsubscribe@opensuse.org To contact the owner, e-mail: opensuse+owner@opensuse.org
В Sat, 27 Dec 2014 06:53:46 -0500
James Knott
On 12/27/2014 03:04 AM, Andrei Borzenkov wrote:
В Fri, 26 Dec 2014 16:44:01 -0500 James Knott
пишет: What seems to be more of an issue at the moment is the 2nd VLAN I created doesn't get a link local address and when I try to forward IPv6 via the Interface, I get an error "RTNETLINK answers: No such device". This, of course, makes it impossible to run IPv6 over that VLAN. I already have one VLAN where IPv6 works fine. There is also another problem, which I have file a bugzilla report about is that when an interface is set to run IPv4 DHCP, it no longer gets a link local address and is again useless for IPv6. This is with the ifup method. You did not mention openSUSE version. I cannot reproduce it in 13.2. I created VLAN, set it to DHCP and "Use DHCPv4 only" and it gets IPv6 link local address on startup.
Sorry, I'm currently running 13.1 on my firewall/router. I prefer to run the Evergreen versions for my firewall.
Same on 13.1 - IPv6 with autoconfiguration is active on VLAN. What does the following print? find /proc/sys/net -name disable_ipv6 -exec grep -H . '{}' + -- To unsubscribe, e-mail: opensuse+unsubscribe@opensuse.org To contact the owner, e-mail: opensuse+owner@opensuse.org
On 12/27/2014 08:58 AM, Andrei Borzenkov wrote:
find /proc/sys/net -name disable_ipv6 -exec grep -H . '{}' +
# find /proc/sys/net -name disable_ipv6 -exec grep -H . '{}' + /proc/sys/net/ipv6/conf/all/disable_ipv6:0 /proc/sys/net/ipv6/conf/default/disable_ipv6:0 /proc/sys/net/ipv6/conf/eth1/disable_ipv6:0 /proc/sys/net/ipv6/conf/lo/disable_ipv6:0 /proc/sys/net/ipv6/conf/sit0/disable_ipv6:0 /proc/sys/net/ipv6/conf/sit1/disable_ipv6:0 /proc/sys/net/ipv6/conf/vlan5/disable_ipv6:0 One thing I see is that neither eth0 nor vlan2 are listed. Eth0 connects to the cable modem. -- To unsubscribe, e-mail: opensuse+unsubscribe@opensuse.org To contact the owner, e-mail: opensuse+owner@opensuse.org
В Sat, 27 Dec 2014 16:48:38 -0500
James Knott
On 12/27/2014 08:58 AM, Andrei Borzenkov wrote:
find /proc/sys/net -name disable_ipv6 -exec grep -H . '{}' +
# find /proc/sys/net -name disable_ipv6 -exec grep -H . '{}' + /proc/sys/net/ipv6/conf/all/disable_ipv6:0 /proc/sys/net/ipv6/conf/default/disable_ipv6:0 /proc/sys/net/ipv6/conf/eth1/disable_ipv6:0 /proc/sys/net/ipv6/conf/lo/disable_ipv6:0 /proc/sys/net/ipv6/conf/sit0/disable_ipv6:0 /proc/sys/net/ipv6/conf/sit1/disable_ipv6:0 /proc/sys/net/ipv6/conf/vlan5/disable_ipv6:0
One thing I see is that neither eth0 nor vlan2 are listed. Eth0 connects to the cable modem.
Could you show "ip link" output? -- To unsubscribe, e-mail: opensuse+unsubscribe@opensuse.org To contact the owner, e-mail: opensuse+owner@opensuse.org
On 12/28/2014 01:04 AM, Andrei Borzenkov wrote:
One thing I see is that neither eth0 nor vlan2 are listed. Eth0
connects to the cable modem. Could you show "ip link" output?
# ip link
1: lo:
On 12/28/2014 08:02 AM, James Knott wrote:
On 12/28/2014 01:04 AM, Andrei Borzenkov wrote:
One thing I see is that neither eth0 nor vlan2 are listed. Eth0
connects to the cable modem. Could you show "ip link" output?
ip link
1: lo:
mtu 65536 qdisc noqueue state UNKNOWN mode DEFAULT link/loopback 00:00:00:00:00:00 brd 00:00:00:00:00:00 2: eth1: mtu 1500 qdisc mq state UP mode DEFAULT qlen 1000 link/ether 00:16:17:a7:f2:d3 brd ff:ff:ff:ff:ff:ff 3: eth0: mtu 576 qdisc pfifo_fast state UP mode DEFAULT qlen 1000 link/ether 00:14:d1:2b:ed:ea brd ff:ff:ff:ff:ff:ff 4: vlan2@eth0: mtu 576 qdisc noqueue state UP mode DEFAULT link/ether 00:14:d1:2b:ed:ea brd ff:ff:ff:ff:ff:ff 5: vlan5@eth1: mtu 1500 qdisc noqueue state UP mode DEFAULT link/ether 00:16:17:a7:f2:d3 brd ff:ff:ff:ff:ff:ff 6: sit0: <NOARP> mtu 1480 qdisc noop state DOWN mode DEFAULT link/sit 0.0.0.0 brd 0.0.0.0 7: sit1: mtu 1280 qdisc noqueue state UNKNOWN mode DEFAULT link/sit 0.0.0.0 peer 64.86.88.116
I notiiced the 576 MTU on eth0 and vlan2. I thought this might be the
problem as IPv6 has a minimum MTU of 1280. I set them to 1500, but
vlan2 stll shows 576 with the ip link and ifconfig commands.
Regardless, both eth0 and vlan2 still don't get a link local address. I
have no idea why they were 576, as I didn't set any MTU prior to this.
# ip link
1: lo:
В Sun, 28 Dec 2014 08:28:10 -0500
James Knott
On 12/28/2014 08:02 AM, James Knott wrote:
On 12/28/2014 01:04 AM, Andrei Borzenkov wrote:
One thing I see is that neither eth0 nor vlan2 are listed. Eth0
connects to the cable modem. Could you show "ip link" output?
ip link
1: lo:
mtu 65536 qdisc noqueue state UNKNOWN mode DEFAULT link/loopback 00:00:00:00:00:00 brd 00:00:00:00:00:00 2: eth1: mtu 1500 qdisc mq state UP mode DEFAULT qlen 1000 link/ether 00:16:17:a7:f2:d3 brd ff:ff:ff:ff:ff:ff 3: eth0: mtu 576 qdisc pfifo_fast state UP mode DEFAULT qlen 1000 link/ether 00:14:d1:2b:ed:ea brd ff:ff:ff:ff:ff:ff 4: vlan2@eth0: mtu 576 qdisc noqueue state UP mode DEFAULT link/ether 00:14:d1:2b:ed:ea brd ff:ff:ff:ff:ff:ff 5: vlan5@eth1: mtu 1500 qdisc noqueue state UP mode DEFAULT link/ether 00:16:17:a7:f2:d3 brd ff:ff:ff:ff:ff:ff 6: sit0: <NOARP> mtu 1480 qdisc noop state DOWN mode DEFAULT link/sit 0.0.0.0 brd 0.0.0.0 7: sit1: mtu 1280 qdisc noqueue state UNKNOWN mode DEFAULT link/sit 0.0.0.0 peer 64.86.88.116 I notiiced the 576 MTU on eth0 and vlan2. I thought this might be the problem as IPv6 has a minimum MTU of 1280.
Yes. Linux kernel won't activate IPv6 on interface if MTU is set smaller than 1280.
I set them to 1500, but vlan2 stll shows 576 with the ip link and ifconfig commands.
How exactly did you set them?
Regardless, both eth0 and vlan2 still don't get a link local address.
If they still have MTU 576 this is expected.
I have no idea why they were 576, as I didn't set any MTU prior to this.
Most likely from your DHCP server. Both dhcpcd and dhclient query MTU by default.
3: eth0:
mtu 1500 qdisc pfifo_fast state UP mode DEFAULT qlen 1000 link/ether 00:14:d1:2b:ed:ea brd ff:ff:ff:ff:ff:ff
Does /proc/sys/net/ipv6/conf/eth0 exist? -- To unsubscribe, e-mail: opensuse+unsubscribe@opensuse.org To contact the owner, e-mail: opensuse+owner@opensuse.org
On 12/28/2014 01:01 PM, Andrei Borzenkov wrote:
I set them to 1500, but
vlan2 stll shows 576 with the ip link and ifconfig commands.
How exactly did you set them?
Initially in Yast. However, I just tried with "link set vlan2 mtu 1500" and then shutting down and restarting vlan2. It now has a link local address. Why would that require specifically setting MTU with link set, when it's set to 1500 in Yast? Why would vlan2 get 576 when vlan5 and eth1 get 1500 MTU, when I previously hadn't set any of them? I just used Wireshark to verify the DHCP MTU. It is indeed 576 bytes. Why on earth would they do that? I'm on a hybrid fibre/coax connection. I haven't used dial up in years. Incidentally, Wireshark also shows IPv6 traffic, such as Neighbour Discovery, coming from the same router as my default route. So, my ISP appears to be ready for IPv6 or close to it. I also noticed the DHCP server is doing gratuitous ACK, Offer and NAK broadcasts. The ACKs do not appear to have the MTU size option. -- To unsubscribe, e-mail: opensuse+unsubscribe@opensuse.org To contact the owner, e-mail: opensuse+owner@opensuse.org
On 12/28/2014 01:33 PM, James Knott wrote:
I just used Wireshark to verify the DHCP MTU. It is indeed 576 bytes.
Is it possible to override the DHCP assigned MTU? the file /etc/sysconfig/network/ifcfg-eth0 also shows 1500. -- To unsubscribe, e-mail: opensuse+unsubscribe@opensuse.org To contact the owner, e-mail: opensuse+owner@opensuse.org
On Sun, 28 Dec 2014 19:52, James Knott
On 12/28/2014 01:33 PM, James Knott wrote:
I just used Wireshark to verify the DHCP MTU. It is indeed 576 bytes.
Is it possible to override the DHCP assigned MTU? the file /etc/sysconfig/network/ifcfg-eth0 also shows 1500.
Yes, there is an option for "nomtu", see "man 8 dhcpcd" IIRC, "-M" or "--nomtu" That means to modify the calling instance of dhcpcd, let's look, the file "/etc/sysconfig/network/dhcp" contains the possibility to specify additional options, I'd try that first. - Yamaban. -- To unsubscribe, e-mail: opensuse+unsubscribe@opensuse.org To contact the owner, e-mail: opensuse+owner@opensuse.org
On Sun, 28 Dec 2014 19:52, James Knott
wrote: On 12/28/2014 01:33 PM, James Knott wrote:
I just used Wireshark to verify the DHCP MTU. It is indeed 576 bytes.
Is it possible to override the DHCP assigned MTU? the file /etc/sysconfig/network/ifcfg-eth0 also shows 1500.
Yes, there is an option for "nomtu", see "man 8 dhcpcd" IIRC, "-M" or "--nomtu"
That means to modify the calling instance of dhcpcd, let's look, the file "/etc/sysconfig/network/dhcp" contains the possibility to specify additional options, I'd try that first. I don't see anything related to MTU in the dhcp file. I tried disabling MTU in DHCP and setting it to 1500. With that, I can ping with 1500 byte packets, but I still don't get the link local address. When I
On 12/28/2014 02:58 PM, Yamaban wrote: tried static config, I was able to get a link local address. I suppose I could try editing the dhcp file to add "dhcp -M" to see what happens. -- To unsubscribe, e-mail: opensuse+unsubscribe@opensuse.org To contact the owner, e-mail: opensuse+owner@opensuse.org
On 12/28/2014 03:47 PM, James Knott wrote:
I suppose I could try editing the dhcp file to add "dhcp -M" to see what happens.
No, didn't work. -- To unsubscribe, e-mail: opensuse+unsubscribe@opensuse.org To contact the owner, e-mail: opensuse+owner@opensuse.org
On Sun, 28 Dec 2014 21:47, James Knott
On Sun, 28 Dec 2014 19:52, James Knott
wrote: On 12/28/2014 01:33 PM, James Knott wrote:
I just used Wireshark to verify the DHCP MTU. It is indeed 576 bytes.
Is it possible to override the DHCP assigned MTU? the file /etc/sysconfig/network/ifcfg-eth0 also shows 1500.
Yes, there is an option for "nomtu", see "man 8 dhcpcd" IIRC, "-M" or "--nomtu"
That means to modify the calling instance of dhcpcd, let's look, the file "/etc/sysconfig/network/dhcp" contains the possibility to specify additional options, I'd try that first. I don't see anything related to MTU in the dhcp file. I tried disabling MTU in DHCP and setting it to 1500. With that, I can ping with 1500 byte packets, but I still don't get the link local address. When I
On 12/28/2014 02:58 PM, Yamaban wrote: tried static config, I was able to get a link local address.
I suppose I could try editing the dhcp file to add "dhcp -M" to see what happens.
Urgs, no, you misunderstood. In the file "/etc/sysconfig/network/dhcp" You can define "USER_OPTIONS". E.g. DHCPCD_USER_OPTIONS="-M" and the similar option for the ISC dhclient in DHCLIENT_USER_OPTIONS="[disable mtu option]" and for the dhcp6c in DHCP6C_USER_OPTIONS="[disable mtu option]" . - Yamaban -- -- To unsubscribe, e-mail: opensuse+unsubscribe@opensuse.org To contact the owner, e-mail: opensuse+owner@opensuse.org
On 12/28/2014 05:50 PM, Yamaban wrote:
I suppose I could try editing the dhcp file to add "dhcp -M" to see what happens.
Urgs, no, you misunderstood.
In the file "/etc/sysconfig/network/dhcp" You can define "USER_OPTIONS".
E.g. DHCPCD_USER_OPTIONS="-M" and the similar option for the ISC dhclient in DHCLIENT_USER_OPTIONS="[disable mtu option]" and for the dhcp6c in DHCP6C_USER_OPTIONS="[disable mtu option]" .
That did it. I now have a link local address on that interface and can ping with a 1500 byte packet. tnx Now to explain to my ISP why their DHCP server config is wrong. ;-) -- To unsubscribe, e-mail: opensuse+unsubscribe@opensuse.org To contact the owner, e-mail: opensuse+owner@opensuse.org
On 12/28/2014 01:33 PM, James Knott wrote:
I just used Wireshark to verify the DHCP MTU. It is indeed 576 bytes. Why on earth would they do that? I'm on a hybrid fibre/coax connection. I haven't used dial up in years.
Their IPv6 router advertisements are showing 1500 MTU. So, they expect the proper size there, but make it impossible for any device to run IPv6. Brilliant!!! -- To unsubscribe, e-mail: opensuse+unsubscribe@opensuse.org To contact the owner, e-mail: opensuse+owner@opensuse.org
On 12/28/2014 01:01 PM, Andrei Borzenkov wrote:
3: eth0:
mtu 1500 qdisc pfifo_fast state UP mode DEFAULT qlen 1000 link/ether 00:14:d1:2b:ed:ea brd ff:ff:ff:ff:ff:ff Does /proc/sys/net/ipv6/conf/eth0 exist?
ls /proc/sys/net/ipv6/conf/eth0 accept_dad accept_source_route hop_limit router_probe_interval accept_ra autoconf max_addresses router_solicitation_delay accept_ra_defrtr dad_transmits max_desync_factor router_solicitation_interval accept_ra_pinfo disable_ipv6 mtu router_solicitations accept_ra_rt_info_max_plen force_mld_version ndisc_notify temp_prefered_lft accept_ra_rtr_pref force_tllao proxy_ndp temp_valid_lft accept_redirects forwarding regen_max_retry use_tempaddr And disable_ipv6 = 0, but mtu=1500. -- To unsubscribe, e-mail: opensuse+unsubscribe@opensuse.org To contact the owner, e-mail: opensuse+owner@opensuse.org
On 12/27/2014 03:04 AM, Andrei Borzenkov wrote:
В Fri, 26 Dec 2014 16:44:01 -0500 James Knott
пишет: What seems to be more of an issue at the moment is the 2nd VLAN I created doesn't get a link local address and when I try to forward IPv6 via the Interface, I get an error "RTNETLINK answers: No such device". This, of course, makes it impossible to run IPv6 over that VLAN. I already have one VLAN where IPv6 works fine. There is also another problem, which I have file a bugzilla report about is that when an interface is set to run IPv4 DHCP, it no longer gets a link local address and is again useless for IPv6. This is with the ifup method. You did not mention openSUSE version. I cannot reproduce it in 13.2. I created VLAN, set it to DHCP and "Use DHCPv4 only" and it gets IPv6 link local address on startup.
Perhaps I should describe my configuration. I'm running openSUSE 13.1 as my firewall, but I have run other versions for a few years, so this isn't the first version that has had the problem. Someone else has also experienced it and I have filed a bugzilla report about it. My network consists of that firewall computer, connected to a cable modem. That interface runs DHCP and does not have an IPv6 link local address. I also use a 6in4 tunnel to get a /56 IPv6 subnet, which consists of 256 /64 subnets. I used the first /64 on my main LAN and the 2nd on VLAN 5, which is my guest WiFi connection. When I tried to set up a 2nd VLAN, it also wouldn't get an IPv6 link local address, though I can configure it for IPv4. I have been running IPv6 for over 4.5 years. -- To unsubscribe, e-mail: opensuse+unsubscribe@opensuse.org To contact the owner, e-mail: opensuse+owner@opensuse.org
participants (3)
-
Andrei Borzenkov
-
James Knott
-
Yamaban