IPv6 radvd Config (Telekom VDSL)

[shadesh]

Moin Jungs,

ich habe, nach Umzug und Umstellung auf MultiWAN (PPPOE Telekom (Tier1) und PPP Telefonica LTE (Tier2)) ein Problem mit IPV6. Es scheint so, als würde für radvd keine Config mehr erstellt werden. Die Config ist einfach leer.

Code: [Select]

root@gw:~ # cat /var/etc/radvd.conf
# Automatically generated, do not edit

Im syslog.log stehen teilweise die Infos:

Code: [Select]

Feb  6 19:57:59 gw dhcp6c: dhcp6c REQUEST on re0 - running newipv6
Feb  6 19:57:59 gw opnsense: /usr/local/etc/rc.newwanipv6: IP renewal is starting on 're0'
Feb  6 19:57:59 gw opnsense: /usr/local/etc/rc.newwanipv6: On (IP address: ) (interface: VDSL[opt4]) (real interface: re0).
Feb  6 19:57:59 gw opnsense: /usr/local/etc/rc.newwanipv6: Failed to detect IP for VDSL[opt4]
Feb  6 19:57:59 gw dhcp6c[79145]: Sending Solicit
Feb  6 19:58:00 gw dhcp6c[79145]: Sending Request
Feb  6 19:58:00 gw dhcp6c[79145]: Received REPLY for REQUEST
Feb  6 19:58:00 gw dhcp6c[79145]: failed to remove an address on pppoe0: Can't assign requested address
Feb  6 19:58:00 gw dhcp6c[79145]: failed to update an address ::
PPPOE Interface hat jedoch ein /64 zugewiesen bekommen.

Code: [Select]

pppoe0: flags=88d1<UP,POINTOPOINT,RUNNING,NOARP,SIMPLEX,MULTICAST> metric 0 mtu 1492
inet6 fe80::20d:b9ff:***:***%pppoe0 prefixlen 64 scopeid 0xb
inet6 fe80::20d:b9ff:***:***%pppoe0 prefixlen 64 scopeid 0xb
inet6 2003:cd:d7ff:3701:aaaa:aaaa:aaaa:aaaa prefixlen 64 autoconf
inet 91.40.*.* --> 62.155.*.* netmask 0xffffffff
nd6 options=23<PERFORMNUD,ACCEPT_RTADV,AUTO_LINKLOCAL>
Config ist eigentlich Standard, DHCPv6 auf dem VDSL Interface, auf allen anderen ein "Track Interface"
Wieso wird keine radvd config erstellt und dadurch der radvd Deamon nicht gestartet? Bin für jeden Hinweis dankbar... ich weiß leider nicht mehr weiter.

[franco]

Hmm, ist die Config auch leer nach:

# pluginctl -s radvd restart


Grüsse
Franco

[shadesh]

Leider ja

Code: [Select]

root@gw:~ # pluginctl -s radvd restart
Service `radvd' has been restarted.
root@gw:~ # cat /var/etc/radvd.conf
# Automatically generated, do not edit
root@gw:~ # ps aux | grep radvd
root   78866  0.0  0.1 1058012  2844  0  S+   07:11     0:00.01 grep radvd

[franco]

Versteh ich nicht. Gibts da noch ne Meldung im System Log dazu?

[shadesh]

Wenigstens bin ich mit "Verstehe ich nicht" nicht mehr alleine
Leider keine weiteren Logeinträge, ausser die oben genannten... ich bin kurz davor das Teil neu zu installieren... -.-

[franco]

Nun ja, es gibt schon einen Grund warum die Config nicht geschrieben wird, dazu müssten wir erstmal die WAN und LAN Einstellungen auseinandernehmen, speziell für IPv6.


Grüsse
Franco

[shadesh]

So die Mühle neu installiert (20.1) und keine Config zurückgesichert. Macht aber immer noch zicken.
Das Log ist immer noch voll mit:

Code: [Select]

Feb  9 14:50:54 gw opnsense: plugins_configure dhcp (,inet6)
Feb  9 14:50:54 gw opnsense: plugins_configure dhcp (execute task : dhcpd_dhcp_configure(,inet6))
Feb  9 14:50:55 gw opnsense: plugins_configure hosts ()
Feb  9 14:50:55 gw opnsense: plugins_configure hosts (execute task : dnsmasq_hosts_generate())
Feb  9 14:50:55 gw opnsense: plugins_configure hosts (execute task : unbound_hosts_generate())
Feb  9 14:50:55 gw dhcp6c[15155]: Sending Solicit
Feb  9 14:50:56 gw dhcp6c[15155]: Sending Request
Feb  9 14:50:56 gw dhcp6c[15155]: Received REPLY for REQUEST
Feb  9 14:50:56 gw dhcp6c[15155]: status code for NA-0: no addresses
Feb  9 14:50:56 gw dhcp6c: dhcp6c REQUEST on pppoe0 - running newipv6
Feb  9 14:50:56 gw opnsense: /usr/local/etc/rc.newwanipv6: IP renewal is starting on 'pppoe0'
Feb  9 14:50:57 gw opnsense: /usr/local/etc/rc.newwanipv6: On (IP address: 2003:cd:d7ff:3db5:20d:*0
) (interface: VDSL[opt1]) (real interface: pppoe0).
Feb  9 14:50:57 gw opnsense: plugins_configure dhcp (,inet6)
Feb  9 14:50:57 gw opnsense: plugins_configure dhcp (execute task : dhcpd_dhcp_configure(,inet6))
Feb  9 14:50:57 gw opnsense: plugins_configure hosts ()
Feb  9 14:50:57 gw opnsense: plugins_configure hosts (execute task : dnsmasq_hosts_generate())
Feb  9 14:50:57 gw opnsense: plugins_configure hosts (execute task : unbound_hosts_generate())
Feb  9 14:50:57 gw dhcp6c[15155]: Sending Solicit
Feb  9 14:50:58 gw dhcp6c[15155]: Sending Request
Feb  9 14:50:58 gw dhcp6c[15155]: Received REPLY for REQUEST
Feb  9 14:50:58 gw dhcp6c[15155]: status code for NA-0: no addresses
Feb  9 14:50:58 gw dhcp6c: dhcp6c REQUEST on pppoe0 - running newipv6
Feb  9 14:50:59 gw opnsense: /usr/local/etc/rc.newwanipv6: IP renewal is starting on 'pppoe0'
Feb  9 14:50:59 gw opnsense: /usr/local/etc/rc.newwanipv6: On (IP address: 2003:cd:d7ff:3db5:*) (interface: VDSL[opt1]) (real interface: pppoe0).
Feb  9 14:50:59 gw opnsense: plugins_configure dhcp (,inet6)
Feb  9 14:50:59 gw opnsense: plugins_configure dhcp (execute task : dhcpd_dhcp_configure(,inet6))
Feb  9 14:51:00 gw opnsense: plugins_configure hosts ()
Feb  9 14:51:00 gw opnsense: plugins_configure hosts (execute task : dnsmasq_hosts_generate())
Feb  9 14:51:00 gw opnsense: plugins_configure hosts (execute task : unbound_hosts_generate())
Feb  9 14:51:00 gw dhcp6c[15155]: Sending Solicit
Feb  9 14:51:01 gw dhcp6c[15155]: Sending Request
Feb  9 14:51:01 gw dhcp6c[15155]: Received REPLY for REQUEST
Feb  9 14:51:01 gw dhcp6c[15155]: status code for NA-0: no addresses
Feb  9 14:51:01 gw dhcp6c: dhcp6c REQUEST on pppoe0 - running newipv6
Feb  9 14:51:01 gw opnsense: /usr/local/etc/rc.newwanipv6: IP renewal is starting on 'pppoe0'
Feb  9 14:51:01 gw opnsense: /usr/local/etc/rc.newwanipv6: On (IP address: 2003:cd:d7ff:3db5:20d:*) (interface: VDSL[opt1]) (real interface: pppoe0).
Feb  9 14:51:01 gw opnsense: plugins_configure dhcp (,inet6)
Feb  9 14:51:01 gw opnsense: plugins_configure dhcp (execute task : dhcpd_dhcp_configure(,inet6))
Feb  9 14:51:02 gw opnsense: plugins_configure hosts ()
Feb  9 14:51:02 gw opnsense: plugins_configure hosts (execute task : dnsmasq_hosts_generate())
Feb  9 14:51:02 gw opnsense: plugins_configure hosts (execute task : unbound_hosts_generate())
Feb  9 14:51:02 gw dhcp6c[15155]: Sending Solicit
Feb  9 14:51:03 gw dhcp6c[15155]: Sending Request
Feb  9 14:51:03 gw dhcp6c[15155]: Received REPLY for REQUEST
Feb  9 14:51:03 gw dhcp6c[15155]: status code for NA-0: no addresses
Feb  9 14:51:03 gw dhcp6c: dhcp6c REQUEST on pppoe0 - running newipv6
Feb  9 14:51:04 gw opnsense: /usr/local/etc/rc.newwanipv6: IP renewal is starting on 'pppoe0'
Feb  9 14:51:04 gw opnsense: /usr/local/etc/rc.newwanipv6: On (IP address: 2003:cd:d7ff:3db5:*) (interface: VDSL[opt1]) (real interface: pppoe0).
Feb  9 14:51:04 gw opnsense: plugins_configure dhcp (,inet6)
Feb  9 14:51:04 gw opnsense: plugins_configure dhcp (execute task : dhcpd_dhcp_configure(,inet6))
Feb  9 14:51:04 gw opnsense: plugins_configure hosts ()
Feb  9 14:51:04 gw opnsense: plugins_configure hosts (execute task : dnsmasq_hosts_generate())
Feb  9 14:51:04 gw opnsense: plugins_configure hosts (execute task : unbound_hosts_generate())
Feb  9 14:51:04 gw dhcp6c[15155]: Sending Solicit
Feb  9 14:51:05 gw dhcp6c[15155]: Sending Request
Feb  9 14:51:05 gw dhcp6c[15155]: Received REPLY for REQUEST
Feb  9 14:51:05 gw dhcp6c[15155]: status code for NA-0: no addresses
Feb  9 14:51:05 gw dhcp6c: dhcp6c REQUEST on pppoe0 - running newipv6
Feb  9 14:51:06 gw opnsense: /usr/local/etc/rc.newwanipv6: IP renewal is starting on 'pppoe0'
Feb  9 14:51:06 gw opnsense: /usr/local/etc/rc.newwanipv6: On (IP address: 2003:cd:d7ff:3db5:*) (interface: VDSL[opt1]) (real interface: pppoe0).
Feb  9 14:51:06 gw opnsense: plugins_configure dhcp (,inet6)
Feb  9 14:51:06 gw opnsense: plugins_configure dhcp (execute task : dhcpd_dhcp_configure(,inet6)) 
am Client sieht das dann so aus:

Code: [Select]

16 bytes from 2a00:1450:*, icmp_seq=162 hlim=57 time=7.678 ms
ping6: sendmsg: No route to host
ping6: wrote google.de 16 chars, ret=-1
16 bytes from 2a00:1450:4016:806::2003, icmp_seq=164 hlim=57 time=8.154 ms
16 bytes from 2a00:1450:4016:806::2003, icmp_seq=165 hlim=57 time=9.517 ms
16 bytes from 2a00:1450:4016:806::2003, icmp_seq=166 hlim=57 time=8.536 ms
16 bytes from 2a00:1450:4016:806::2003, icmp_seq=167 hlim=57 time=8.438 ms
16 bytes from 2a00:1450:4016:806::2003, icmp_seq=168 hlim=57 time=9.375 ms
16 bytes from 2a00:1450:4016:806::2003, icmp_seq=169 hlim=57 time=7.157 ms
ping6: sendmsg: No route to host
ping6: wrote google.de 16 chars, ret=-1
16 bytes from 2a00:1450:4016:806::2003, icmp_seq=171 hlim=57 time=9.684 ms
16 bytes from 2a00:1450:4016:806::2003, icmp_seq=172 hlim=57 time=7.156 ms
16 bytes from 2a00:1450:4016:806::2003, icmp_seq=173 hlim=57 time=9.620 ms
16 bytes from 2a00:1450:4016:806::2003, icmp_seq=174 hlim=57 time=7.130 ms
16 bytes from 2a00:1450:4016:806::2003, icmp_seq=175 hlim=57 time=10.130 ms
16 bytes from 2a00:1450:4016:806::2003, icmp
Es scheint, als würde das Teil versuchen alle paar Sekunden "/usr/local/etc/rc.newwanipv6: IP renewal is starting on 'pppoe0' " ausführen zu wollen... warum ist mir schleierhaft.

[marjohn56]

Feb 9 14:50:56 gw dhcp6c [15155]: status code for NA-0: no addresses

Your ISP is not offering you a WAN address, select prefix only in the dhcp6c settings and see if that makes a difference

[shadesh]

Looks better now, but new message:

Code: [Select]

Feb  9 21:15:47 gw dhcp6c[65699]: Received REPLY for RENEW
Feb  9 21:15:59 gw dhcp6c[91061]: Sending Solicit
Feb  9 21:15:59 gw dhcp6c[65699]: XID mismatch
Feb  9 21:18:04 gw dhcp6c[91061]: Sending Solicit
Feb  9 21:18:04 gw dhcp6c[65699]: XID mismatch
Feb  9 21:20:02 gw dhcp6c[91061]: Sending Solicit
Feb  9 21:20:02 gw dhcp6c[65699]: XID mismatch
Feb  9 21:21:53 gw dhcp6c[91061]: Sending Solicit
Feb  9 21:21:53 gw dhcp6c[65699]: XID mismatch
Feb  9 21:23:47 gw dhcp6c[91061]: Sending Solicit
Feb  9 21:23:47 gw dhcp6c[65699]: XID mismatch
Feb  9 21:25:39 gw dhcp6c[91061]: Sending Solicit
Feb  9 21:25:39 gw dhcp6c[65699]: XID mismatch
Feb  9 21:27:29 gw dhcp6c[91061]: Sending Solicit
Feb  9 21:27:29 gw dhcp6c[65699]: XID mismatch
Feb  9 21:29:23 gw dhcp6c[91061]: Sending Solicit
Feb  9 21:29:23 gw dhcp6c[65699]: XID mismatch
Feb  9 21:30:47 gw dhcp6c[65699]: Sending Renew
Feb  9 21:30:47 gw dhcp6c[65699]: Received REPLY for RENEW
Feb  9 21:31:31 gw dhcp6c[91061]: Sending Solicit
Feb  9 21:31:31 gw dhcp6c[65699]: XID mismatch
Feb  9 21:33:24 gw dhcp6c[91061]: Sending Solicit
Feb  9 21:33:24 gw dhcp6c[65699]: XID mismatch
Feb  9 21:35:21 gw dhcp6c[91061]: Sending Solicit
Feb  9 21:35:21 gw dhcp6c[65699]: XID mismatch
Feb  9 21:37:13 gw dhcp6c[91061]: Sending Solicit
Feb  9 21:37:13 gw dhcp6c[65699]: XID mismatch
Feb  9 21:39:14 gw dhcp6c[91061]: Sending Solicit
Feb  9 21:39:14 gw dhcp6c[65699]: XID mismatch
Process output:

Code: [Select]

root   68674   0.0  0.1 1058012  2852  0  S+   21:41     0:00.01 grep dhcpv6
root@gw:/var/log # ps aux | grep dhcp
dhcpd  18419   0.0  0.4   14748  7884  -  Ss   15:30     0:03.82 /usr/local/sbin/dhcpd -6 -user dhcpd -group dhcpd -chroot /var/dhcpd -cf /etc/dhcpdv6.c
root   65699   0.0  0.1 1057800  2800  -  Is   15:30     0:00.15 /usr/local/sbin/dhcp6c -d -c /var/etc/dhcp6c_opt1.conf -p /var/run/dhcp6c_pppoe0.pid pp
root   88555   0.0  0.1 1057636  2404  -  Is   15:30     0:00.00 /usr/local/sbin/dhcpleases6 -c /usr/local/sbin/configctl dhcpd update prefixes -l /var/
root   91061   0.0  0.1 1057800  2832  -  Is   15:30     0:00.12 /usr/local/sbin/dhcp6c -d -c /var/etc/dhcp6c_opt1.conf -p /var/run/dhcp6c_pppoe0.pid pp
root   36895   0.0  0.1 1058012  2852  0  S+   21:41     0:00.01 grep dhcp


[marjohn56]

The xid generated by dhcp6c is 'out of sync' with the xid on the ISP dhcp6 server, this happens when a session is dropped by the client ( you ) but the ISP is still holding on. Best way to resolve this is to disconnect  the VDSL modem connection completely or turn off the modem and leave it disconnected for a minimum of 15 minutes, 30 minutes is better. This will force the ISPs systems to reset. After that you should be able to reconnect and all should be good.


We used to see exactly the same thing happening with Sky UK, their dhcp6 implementation was also rubbish, perhaps they sold their ideas to your ISP!

[shadesh]

Thanks for that info. Maybe the reason can also be that there is a second dhcp6c process for the same interface?

Code: [Select]

root@gw:/var/log # ps uxawww | grep dhcp6c
root   65699   0.0  0.1 1057800  2800  -  Is   15:30     0:00.16 /usr/local/sbin/dhcp6c -d -c /var/etc/dhcp6c_opt1.conf -p /var/run/dhcp6c_pppoe0.pid pppoe0
root   91061   0.0  0.1 1057800  2832  -  Is   15:30     0:00.12 /usr/local/sbin/dhcp6c -d -c /var/etc/dhcp6c_opt1.conf -p /var/run/dhcp6c_pppoe0.pid pppoe0

Saw the hint for that here: https://docs.netgate.com/pfsense/en/latest/dhcp/dhcpv6-client-xid-mismatch.html

[marjohn56]

Yes, that will be it. I thought that issue was long gone. Does this happen from a reboot?

[marjohn56]

Thanks for that info. Maybe the reason can also be that there is a second dhcp6c process for the same interface?

Code: [Select]

root@gw:/var/log # ps uxawww | grep dhcp6c
root   65699   0.0  0.1 1057800  2800  -  Is   15:30     0:00.16 /usr/local/sbin/dhcp6c -d -c /var/etc/dhcp6c_opt1.conf -p /var/run/dhcp6c_pppoe0.pid pppoe0
root   91061   0.0  0.1 1057800  2832  -  Is   15:30     0:00.12 /usr/local/sbin/dhcp6c -d -c /var/etc/dhcp6c_opt1.conf -p /var/run/dhcp6c_pppoe0.pid pppoe0

Saw the hint for that here: https://docs.netgate.com/pfsense/en/latest/dhcp/dhcpv6-client-xid-mismatch.html

This is something we worked on way back in 18.1*, it was a really common issue and something I worked on. It was my impression this had been fixed once and for all, this could be a regression... though i can't see how as no-one else is has reported this issue. Can you issue the following commands in the shell.


# killall -9 -c dhcp6c


Then scp or cd into the /var/run folder and delete any dhcp6c*.pid file(s) that are left there. Restart and see if it behaves then.

[shadesh]

Yes, that will be it. I thought that issue was long gone. Does this happen from a reboot?

Yep, this happened after a reboot. Will try your fix later today. Have to work now :p

[shadesh]

# killall -9 -c dhcp6c

Then scp or cd into the /var/run folder and delete any dhcp6c*.pid file(s) that are left there. Restart and see if it behaves then.

There was only one .pid file after killing both processes. Looks good after reboot:

Code: [Select]

root@gw:~ # ps uxawww | grep dhcp6c
root   61996   0.0  0.1 1057800  2840  -  Is   16:47   0:00.01 /usr/local/sbin/dhcp6c -d -c /var/etc/dhcp6c_opt1.conf -p /var/run/dhcp6c_pppoe0.pid pppoe0
root   12468   0.0  0.1 1058012  2844  1  S+   16:48   0:00.01 grep dhcp6c

Thanks!

PS: One more thing: Can you explain to me, why the Tier2 Gateway is always marked as "Active" when using Multi WAN with Tier config? From my understanding the Tier1 should be displayed as active if it's not failed. Btw, ignore the IPv6 Gateway in the screenshot - It's not part of the multiwan config.