IPv6 stopped working, reboot fixed it - ideas?

[mope]

I've been using IPv6 for a few months now with no problem. However, today it stopped working for some reason. I got 0/10 on the test-ipv6.com site. I rebooted the OPNsense router and it's all fine now. There were no stopped services in OPNsense, radvd and the WAN interfaces were fine, as was IPv4.

Any ideas on what I could check to find a cause for this?

[tiermutter]

You can check logs for the time where the problem occured, but did you also check v6 connectivity on other devices? Maybe it was a client sided problem or ISP...

[tiermutter]

BTW:
Last days I have similar issues on Win10 laptop with multiple LAN interfaces, where the OS drops v6 addreses. But there is no OPNsense on this site...

[mope]

You can check logs for the time where the problem occured, but did you also check v6 connectivity on other devices?

Yes I checked other devices and they showed the same issue. All devices worked fine after router was rebooted... I get a bit lost in the diagnostics section of OPNsense as I'm a long time pfSense user. Could you suggest a good place to look for this problem?

[tiermutter]

Since the problem is already (for now) fixed, it's hard to find out anything. Only the logs remain, no idea what to look for, as the problem can be manifold...
Did you check if clients still had (public) v6 addresses? If it occurs again I would first check this, then looking at the sense´s WAN and LAN interface: is there a public v6 assigned? If so, start a traceroute from sense (or client if it still has a v6 address) to somewhere in WAN (eg google.com). Also try to ping or traceroute v6 IPs on WAN (and sense itself), to see if it is just a DNS problem.

[mope]

Since the problem is already (for now) fixed, it's hard to find out anything. Only the logs remain, no idea what to look for, as the problem can be manifold...
Did you check if clients still had (public) v6 addresses? If it occurs again I would first check this, then looking at the sense´s WAN and LAN interface: is there a public v6 assigned? If so, start a traceroute from sense (or client if it still has a v6 address) to somewhere in WAN (eg google.com). Also try to ping or traceroute v6 IPs on WAN (and sense itself), to see if it is just a DNS problem.

I didn't think to check if the clients had an IPv6 address, the WAN interface did have one though. I will get more logs next time. Thanks for replying.