We still need to hook a openvpn reconfigure in the apinger reload command

Just backed out another vlan patch that was left lingering, the next snapshot should be better now.

We need to have our vlan patches sorted, and that includes 802.1p.

He's back from vacation next week at it's earliest.

It's on the roadmap for a commercial client. You would normally couple the VPN server name to a dynamic DNS entry.

Did you know that we also support gateway groups for a dynamic dns entry now?

My hunch would be that this is related to the vlan patches

b and yes

I hoped that some of the package maintainers would have a bit more active stance with regards to Ipv6.

I do see good work, but most of it is targeted to IPv4 features, not even basic interface support for HAproxy for example is there. Nor varnish, squid and others.

I sent a email to the developers list last year pleading for developers to get a tunnel and work from there. I did not get much response at the time. Now a year later there is still almost nothing there.

So unless they step up the packages will ship as they are.

Not sure who the original author was. Maybe we can let Darren have a look and make this into a theme of sorts.

I'm pretty sure that nobody has touched the snort package to work with IPv6 yet. So that probably means that if you use snort you kill your IPv6.

You would at a minimum require the local interfaces to be setup in snort to prevent it from blocking it's own interfaces.

there are no messages from dhcp6c which is the dhcp6 client in there. We'll have to try again.

It appears the dhcp6c client has exited for a reason, can you see those in the system logs? Should be there.

I didn't mention it before, but Router Advertisments _always_ send a link-local address of the gateway, sure it may have a global address, but that's for traceroute. Link-local is always for directly reachable nodes.

So yes, link-local addresses are normal. On that note, it could be that the dhcp6 client exited, and took the WAN address with it, which is weird, because it appears to have left the Prefix Delegation address on the LAN.

Check if the dhcp6c process is still there, the logging on it is a bit brief though. Maybe it exited on lease renewal.

I think the headend is not sending out RA, but the dhcp6 server/relay at the headend is already active.

But without RA that means delivering broken IPv6. Sort off, the missing default route means it fails fast, but still.

It's on the list, but I have not gotten round to it yet.

Have you checked "Allow DNS server list to be overridden by DHCP/PPP on WAN "

Otherwise it would never use those.