Upgrade to 22.1.4_1 no IPv6 connectivity

Started by staticznld, March 25, 2022, 08:41:46 PM

Previous topic - Next topic
Well with the patch applied my clients AND the OPNsense box are able to connect over IPv6.


***GOT REQUEST TO AUDIT CONNECTIVITY***
Currently running OPNsense 22.1.4_1 (amd64/OpenSSL) at Tue Mar 29 09:52:34 CEST 2022
Checking connectivity for host: pkg.opnsense.org -> 89.149.211.205
PING 89.149.211.205 (89.149.211.205): 1500 data bytes
1508 bytes from 89.149.211.205: icmp_seq=0 ttl=59 time=5.230 ms
1508 bytes from 89.149.211.205: icmp_seq=1 ttl=59 time=5.271 ms
1508 bytes from 89.149.211.205: icmp_seq=2 ttl=59 time=5.811 ms
1508 bytes from 89.149.211.205: icmp_seq=3 ttl=59 time=5.129 ms

--- 89.149.211.205 ping statistics ---
4 packets transmitted, 4 packets received, 0.0% packet loss
round-trip min/avg/max/stddev = 5.129/5.360/5.811/0.265 ms
Checking connectivity for repository (IPv4): https://pkg.opnsense.org/FreeBSD:13:amd64/22.1
Updating OPNsense repository catalogue...
Fetching meta.conf: . done
Fetching packagesite.txz: .......... done
Processing entries: .......... done
OPNsense repository update completed. 783 packages processed.
All repositories are up to date.
Checking connectivity for host: pkg.opnsense.org -> 2001:1af8:4f00:a005:5::
PING6(1548=40+8+1500 bytes) 2a02:a450:839:dddd:: --> 2001:1af8:4f00:a005:5::
1508 bytes from 2001:1af8:4f00:a005:5::, icmp_seq=0 hlim=59 time=3.404 ms
1508 bytes from 2001:1af8:4f00:a005:5::, icmp_seq=1 hlim=59 time=3.280 ms
1508 bytes from 2001:1af8:4f00:a005:5::, icmp_seq=2 hlim=59 time=3.301 ms
1508 bytes from 2001:1af8:4f00:a005:5::, icmp_seq=3 hlim=59 time=3.398 ms

--- 2001:1af8:4f00:a005:5:: ping6 statistics ---
4 packets transmitted, 4 packets received, 0.0% packet loss
round-trip min/avg/max/std-dev = 3.280/3.346/3.404/0.056 ms
Checking connectivity for repository (IPv6): https://pkg.opnsense.org/FreeBSD:13:amd64/22.1
Updating OPNsense repository catalogue...
Fetching meta.conf: . done
Fetching packagesite.txz: .......... done
Processing entries: .......... done
OPNsense repository update completed. 783 packages processed.
All repositories are up to date.
***DONE***

The patch really doesn't do much to be honest and is semantically the same except that it tries to clear the router file when rtsold is invoked. What I need to know:

What is the state of /tmp/pppoe0_routerv6 with and without the patch... and if the file exists what is the contents of it? (the name might differ, look for "/tmp/*_routerv6" files on your system.


Cheers,
Franco

At the moment i am unable to check without the patch installed.
But with the patch installed i see the default link local gateway.

root@router:~ # cat /tmp/pppoe0_routerv6
fe80::1251:72ff:fe23:3288
root@router:~ #


Before i installed the update i didn't see the default gateway in the gui.

I will try to lookup the file tonight.

Hmm, in any case here is a patch for 22.1.4 that emulates the previous behaviour.

https://github.com/opnsense/core/commit/4b6fa9e556690

However, it suggest that doing "DHCPv6 on a PPPoE that provides IPv6 connectivity" is one of the silliest things the code historically did... does dhcp6c even ever get an address? A prefix maybe at best?


Cheers,
Franco

Without the patch installed i did see the "IPv6 delegated prefix" xxxx:xxxx:xxxx::/48 in the overview.

prefix info is irrelevant. It moved to a different file but it's a read-only value. (It's mentioned in the patch BTW.)


Cheers,
Franco

Patch removed en reloaded the PPPoE device.

root@router:~ # cat /tmp/pppoe0_routerv6
fe80::1251:72ff:fe23:3288


Also the default IPv6 route is gone in the GUI.

Then applied https://github.com/opnsense/core/commit/4b6fa9e556690 and IPv6 is working again!
Also the default route is back in the GUI.

Hi,

I decided to investigate my IPv6-problem and it turns out Franco was right  :-[ . The problems I experienced after running patch 1a5dfc932f8 had to do with IPv6 fragments (observed while looking at a packet capture) and had nothing to do with the patch.

After checking (and re-setting) the MTU-size on the WAN-interface (MTU:1512) and the vlan on the WAN-interface (MTU:1508) I got an MTU of 1500 on my PPPoE interface after which I was able to ping using IPv6 again.

Checking connectivity for host: pkg.opnsense.org -> 2001:1af8:4f00:a005:5::
PING6(1548=40+8+1500 bytes) [opnsense-box_ipv6] --> 2001:1af8:4f00:a005:5::
1508 bytes from 2001:1af8:4f00:a005:5::, icmp_seq=0 hlim=59 time=3.960 ms
1508 bytes from 2001:1af8:4f00:a005:5::, icmp_seq=1 hlim=59 time=3.660 ms
1508 bytes from 2001:1af8:4f00:a005:5::, icmp_seq=2 hlim=59 time=3.657 ms
1508 bytes from 2001:1af8:4f00:a005:5::, icmp_seq=3 hlim=59 time=3.629 ms


Problem solved!Thanks for being so resolute. That made me look into it again and solve my problem  ;) .
Best regards.

Also investigating your MTU solution, but that isn't working for me.
Some websites aren't available at all!
I have a PPPoE tunnel over VLAN 6.
I have created an unused (only for setting MTU) interface IGB0 with an MTU of 1508, on top of that VLAN6, and on top of that PPPoE.

When looking at https://www.speedguide.net/analyzer.php i can see that I am using an MTU of 1500.

opnsense-patch 4b6fa9e
Did solve the problem for me.

Is there ever a nameserver file?

# cat /tmp/pppoe0_nameserverv6

You said there is a prefix assigned. Do you also request and address for WAN IPv6 or just the prefix?

Nameserver is there

root@router:~ # cat /tmp/pppoe0_nameserverv6
2a02:a47f:e000::53
2a02:a47f:e000::54


I only request an prefix.

Thanks, I guess this is fixed then (and only relevant for PPPoE before someone else jumps in again).


Cheers,
Franco

Hi Franco,

Just to confirm ... this patch will be included in the next update of OPNsense?

Best regards.

Yes, but happy about all additional feedback in the meantime.


Cheers,
Franco

Thanks for the patch!

If there is anything else you would like know, let me know and i take a look.