8800NL IPv6 on BT FTTC Keep alive needed?

[elwisty]

I have an 8800NL which I use on BT's FTTC service.

I have configured it correctly (I think!) for IPv6, and if I restart the router, or disconnect and reconnect the PPPoE link, I get an IPv6 connection - as far as the IPv6 testers are concerned - the WAN status shows an IPv6 address of the form 2a00::::::: and all my networked machines get IPv6 addresses. There is no Default IPv6 Gateway / IPv6 Address shown in the routers status page. I have incoming WAN ping's blocked for both IPv4 and IPv6.

However I don't knowingly use many(any?) IPv6 sites or services. I noticed that after a while (few hours?) if I reran the IPv6 tests it said I had no connectivity. There was no evidence of the router disconnecting (DSL stats show no drops, and the router connection time is of the order of days). The Router shows an IPV6 address of fe80::::::. A disconnect and reconnect of the PPPoE link brings the 2a00::::::: address back again and resets the connection time. Until the next time.

As an experiment I wrote a batch job that pings Google's ipv6 and ran it using the scheduler every five minutes. That kept the IPv6 address alive for a full day. Stopping the job resulted in the ipv6 connectivity going away in a couple of hours.

Has anyone else on BT seen similar symptoms? Do you think this is a BT issue, or a configuration issue on my 8800NL?.

Do I need ICMP enabled for IPv6?

[billion_fan]

I have an 8800NL which I use on BT's FTTC service.

I have configured it correctly (I think!) for IPv6, and if I restart the router, or disconnect and reconnect the PPPoE link, I get an IPv6 connection - as far as the IPv6 testers are concerned - the WAN status shows an IPv6 address of the form 2a00::::::: and all my networked machines get IPv6 addresses. There is no Default IPv6 Gateway / IPv6 Address shown in the routers status page. I have incoming WAN ping's blocked for both IPv4 and IPv6.

However I don't knowingly use many(any?) IPv6 sites or services. I noticed that after a while (few hours?) if I reran the IPv6 tests it said I had no connectivity. There was no evidence of the router disconnecting (DSL stats show no drops, and the router connection time is of the order of days). The Router shows an IPV6 address of fe80::::::. A disconnect and reconnect of the PPPoE link brings the 2a00::::::: address back again and resets the connection time. Until the next time.

As an experiment I wrote a batch job that pings Google's ipv6 and ran it using the scheduler every five minutes. That kept the IPv6 address alive for a full day. Stopping the job resulted in the ipv6 connectivity going away in a couple of hours.

Has anyone else on BT seen similar symptoms? Do you think this is a BT issue, or a configuration issue on my 8800NL?.

Do I need ICMP enabled for IPv6?

I haven't had any reports of a similar issue with Ipv6 on our devices.

It might be a BT issue, have a look at the following thread

https://community.bt.com/t5/Connected-D ... ue#M137174

[elwisty]

Thanks, I have posted a similar question to the BT forums.

As an aside, do the WAN addresses (Ipv4 and Ipv6) have lease times like the internal ones handed out by DHCP?

Is it the role of the router to re-request these before they expire? Or is it all handled differently?

[billion_fan]

Thanks, I have posted a similar question to the BT forums.

As an aside, do the WAN addresses (Ipv4 and Ipv6) have lease times like the internal ones handed out by DHCP?

Is it the role of the router to re-request these before they expire? Or is it all handled differently?

It depends in the ISP, and how they have configured their equipment, normally ISP will assign the IPv4/IPv6 if enabled on the router

[elwisty]

I have been experimenting with 1 hour, 30 minute and 15 minute "keep alive". The one hour keep alive dropped out after about 20 hours. A retry with 30 minutes lasted about 30 hours. The 15 minute schedule keep alive has now been OK for three days.

It looks as if something at BT's end (at least on my line) needs a bit of traffic at regular intervals to keep the IPv6 "lease" active. Is there any way to examine the lease on the Billion?

[billion_fan]

I have been experimenting with 1 hour, 30 minute and 15 minute "keep alive". The one hour keep alive dropped out after about 20 hours. A retry with 30 minutes lasted about 30 hours. The 15 minute schedule keep alive has now been OK for three days.

It looks as if something at BT's end (at least on my line) needs a bit of traffic at regular intervals to keep the IPv6 "lease" active. Is there any way to examine the lease on the Billion?

I don't think there is, as the lease is set by BT

[elwisty]

Spoke too soon. The 15 minute keep alive timed out after just over 3 days. the 19:15:00 ping succeeded but by 19:22 the router was showing was showing the fe80:::::: address.

The Router log showed
Jan 24 19:15:34 daemon warn radvd[11151]: poll error: Interrupted system call
Jan 24 19:15:34 daemon warn radvd[11151]: Exiting, sigterm received.
Jan 24 19:15:34 daemon info radvd[8382]: version 1.8 started
Jan 24 19:15:34 daemon warn radvd[8382]: poll error: Interrupted system call
Jan 24 19:15:34 daemon warn radvd[8382]: Exiting, sigterm received.
Jan 24 19:15:34 daemon info radvd[8390]: version 1.8 started
Jan 24 19:15:34 daemon warn radvd[8390]: poll error: Interrupted system call
Jan 24 19:15:34 daemon warn radvd[8390]: Exiting, sigterm received.
Jan 24 19:15:34 daemon info radvd[8394]: version 1.8 started
Jan 24 19:15:34 daemon warn radvd[8394]: poll error: Interrupted system call
Jan 24 19:15:34 daemon warn radvd[8394]: Exiting, sigterm received.
Jan 24 19:15:34 daemon info radvd[8396]: version 1.8 started
Jan 24 19:15:35 daemon warn radvd[8396]: poll error: Interrupted system call
Jan 24 19:15:35 daemon warn radvd[8396]: Exiting, sigterm received.
Jan 24 19:15:35 daemon info radvd[8459]: version 1.8 started

A disconnect reconnect cycle brought back the 2a00::::::: address

The Router log covering the reconnect was very similar except for a dhcp6c restart.

Jan 24 19:23:06 daemon notice syslog: pppd:Terminating connection due to link down.
Jan 24 19:23:06 daemon crit syslog: Clear IP addresses. PPP connection DOWN.
Jan 24 19:23:06 daemon crit syslog: Clear IP addresses. Connection DOWN.
Jan 24 19:23:06 daemon info radvd[8459]: attempting to reread config file
Jan 24 19:23:06 daemon notice syslog: pppd:Couldn't increase MTU to 1500.
Jan 24 19:23:06 daemon err syslog: pppd:Couldn't increase MRU to 1500
Jan 24 19:23:06 daemon info radvd[8459]: resuming normal operation
Jan 24 19:23:06 daemon notice syslog: pppd:Connection terminated.
Jan 24 19:23:06 daemon notice syslog: pppd:Connect time 2151.5 minutes.
Jan 24 19:23:06 daemon notice syslog: pppd:Sent 589002663 bytes, received 1668066257 bytes.
Jan 24 19:23:06 daemon info radvd[8459]: attempting to reread config file
Jan 24 19:23:06 daemon info radvd[8459]: resuming normal operation
Jan 24 19:23:06 daemon notice syslog: pppd:Doing disconnect
Jan 24 19:23:12 daemon warn radvd[8459]: poll error: Interrupted system call
Jan 24 19:23:12 daemon warn radvd[8459]: Exiting, sigterm received.
Jan 24 19:23:12 daemon info radvd[11117]: version 1.8 started
Jan 24 19:23:12 daemon warn radvd[11117]: poll error: Interrupted system call
Jan 24 19:23:12 daemon warn radvd[11117]: Exiting, sigterm received.
Jan 24 19:23:12 daemon info radvd[11124]: version 1.8 started
Jan 24 19:23:12 daemon err dhcp6c[11023]: client6_send: transmit failed: Network is unreachable
Jan 24 19:23:15 daemon notice syslog: PPP: Start to connect ...
Jan 24 19:23:20 daemon crit syslog: PPP server detected.
Jan 24 19:23:20 daemon crit syslog: PPP session established.
Jan 24 19:23:20 daemon info radvd[11124]: attempting to reread config file
Jan 24 19:23:20 daemon info radvd[11124]: resuming normal operation
Jan 24 19:23:20 daemon warn kernel: netdev path : ppp1.1 -> ptm0.1 -> ptm0
Jan 24 19:23:20 daemon notice syslog: pppd:Using interface ppp1.1
Jan 24 19:23:20 daemon notice syslog: pppd:Connect: ppp1.1 <--> ptm0.1
Jan 24 19:23:20 daemon notice syslog: pppd:Couldn't increase MTU to 1500.
Jan 24 19:23:20 daemon err syslog: pppd:Couldn't increase MRU to 1500
Jan 24 19:23:20 daemon err syslog: pppd:Couldn't increase MRU to 1500
Jan 24 19:23:20 daemon crit syslog: PPP LCP UP.
Jan 24 19:23:20 daemon notice syslog: pppd:Remote message: CHAP authentication success
Jan 24 19:23:20 daemon info radvd[11124]: attempting to reread config file
Jan 24 19:23:20 daemon notice syslog: pppd:local LL address fe80::11cd:8b88:e784:5d40
Jan 24 19:23:20 daemon notice syslog: pppd:remote LL address fe80::0221:05ff:feae:0424
Jan 24 19:23:20 daemon notice syslog: pppd:local IP address 109.153.139.232
Jan 24 19:23:20 daemon notice syslog: pppd:remote IP address 172.16.14.2
Jan 24 19:23:20 daemon notice syslog: pppd:primary DNS address 81.139.57.100
Jan 24 19:23:20 daemon notice syslog: pppd:secondary DNS address 81.139.56.100
Jan 24 19:23:20 daemon crit syslog: Received valid IP address from server. Connection UP.
Jan 24 19:23:20 daemon info radvd[11124]: resuming normal operation
Jan 24 19:23:21 daemon info syslog: Hop limit : 64
Jan 24 19:23:21 daemon info syslog: Stateful address conf. : Yes
Jan 24 19:23:21 daemon info syslog: Stateful other conf. : No
Jan 24 19:23:21 daemon info syslog: Router preference : medium
Jan 24 19:23:21 daemon info syslog: Router lifetime : 4500 seconds
Jan 24 19:23:21 daemon info syslog: Reachable time : 0 milliseconds
Jan 24 19:23:21 daemon info syslog: Retransmit time : 0 milliseconds
Jan 24 19:23:24 daemon info radvd[11124]: attempting to reread config file
Jan 24 19:23:24 daemon warn kernel: ^[[0;36;44mBroadcom Packet Flow Cache flushing the flows^[[0m
Jan 24 19:23:24 daemon info radvd[11124]: resuming normal operation
Jan 24 19:23:24 daemon info syslog: dhcp6c restart..
Jan 24 19:23:36 daemon warn radvd[11124]: poll error: Interrupted system call
Jan 24 19:23:36 daemon warn radvd[11124]: Exiting, sigterm received.
Jan 24 19:23:37 daemon info radvd[11636]: version 1.8 started
Jan 24 19:23:37 daemon warn radvd[11636]: poll error: Interrupted system call
Jan 24 19:23:37 daemon warn radvd[11636]: Exiting, sigterm received.
Jan 24 19:23:37 daemon info radvd[11648]: version 1.8 started
Jan 24 19:23:37 daemon warn radvd[11648]: poll error: Interrupted system call
Jan 24 19:23:37 daemon warn radvd[11648]: Exiting, sigterm received.
Jan 24 19:23:37 daemon info radvd[11668]: version 1.8 started
Jan 24 19:23:37 daemon warn radvd[11668]: poll error: Interrupted system call
Jan 24 19:23:37 daemon warn radvd[11668]: Exiting, sigterm received.
Jan 24 19:23:37 daemon info radvd[11690]: version 1.8 started

Does this give any clues as to what is going on?

The rest of the log seems to be just Ipv4 DHCP OFFERs and ACKs

[billion_fan]

Spoke too soon. The 15 minute keep alive timed out after just over 3 days. the 19:15:00 ping succeeded but by 19:22 the router was showing was showing the fe80:::::: address.

The Router log showed
Jan 24 19:15:34 daemon warn radvd[11151]: poll error: Interrupted system call
Jan 24 19:15:34 daemon warn radvd[11151]: Exiting, sigterm received.
Jan 24 19:15:34 daemon info radvd[8382]: version 1.8 started
Jan 24 19:15:34 daemon warn radvd[8382]: poll error: Interrupted system call
Jan 24 19:15:34 daemon warn radvd[8382]: Exiting, sigterm received.
Jan 24 19:15:34 daemon info radvd[8390]: version 1.8 started
Jan 24 19:15:34 daemon warn radvd[8390]: poll error: Interrupted system call
Jan 24 19:15:34 daemon warn radvd[8390]: Exiting, sigterm received.
Jan 24 19:15:34 daemon info radvd[8394]: version 1.8 started
Jan 24 19:15:34 daemon warn radvd[8394]: poll error: Interrupted system call
Jan 24 19:15:34 daemon warn radvd[8394]: Exiting, sigterm received.
Jan 24 19:15:34 daemon info radvd[8396]: version 1.8 started
Jan 24 19:15:35 daemon warn radvd[8396]: poll error: Interrupted system call
Jan 24 19:15:35 daemon warn radvd[8396]: Exiting, sigterm received.
Jan 24 19:15:35 daemon info radvd[8459]: version 1.8 started

A disconnect reconnect cycle brought back the 2a00::::::: address

The Router log covering the reconnect was very similar except for a dhcp6c restart.

Jan 24 19:23:06 daemon notice syslog: pppd:Terminating connection due to link down.
Jan 24 19:23:06 daemon crit syslog: Clear IP addresses. PPP connection DOWN.
Jan 24 19:23:06 daemon crit syslog: Clear IP addresses. Connection DOWN.
Jan 24 19:23:06 daemon info radvd[8459]: attempting to reread config file
Jan 24 19:23:06 daemon notice syslog: pppd:Couldn't increase MTU to 1500.
Jan 24 19:23:06 daemon err syslog: pppd:Couldn't increase MRU to 1500
Jan 24 19:23:06 daemon info radvd[8459]: resuming normal operation
Jan 24 19:23:06 daemon notice syslog: pppd:Connection terminated.
Jan 24 19:23:06 daemon notice syslog: pppd:Connect time 2151.5 minutes.
Jan 24 19:23:06 daemon notice syslog: pppd:Sent 589002663 bytes, received 1668066257 bytes.
Jan 24 19:23:06 daemon info radvd[8459]: attempting to reread config file
Jan 24 19:23:06 daemon info radvd[8459]: resuming normal operation
Jan 24 19:23:06 daemon notice syslog: pppd:Doing disconnect
Jan 24 19:23:12 daemon warn radvd[8459]: poll error: Interrupted system call
Jan 24 19:23:12 daemon warn radvd[8459]: Exiting, sigterm received.
Jan 24 19:23:12 daemon info radvd[11117]: version 1.8 started
Jan 24 19:23:12 daemon warn radvd[11117]: poll error: Interrupted system call
Jan 24 19:23:12 daemon warn radvd[11117]: Exiting, sigterm received.
Jan 24 19:23:12 daemon info radvd[11124]: version 1.8 started
Jan 24 19:23:12 daemon err dhcp6c[11023]: client6_send: transmit failed: Network is unreachable
Jan 24 19:23:15 daemon notice syslog: PPP: Start to connect ...
Jan 24 19:23:20 daemon crit syslog: PPP server detected.
Jan 24 19:23:20 daemon crit syslog: PPP session established.
Jan 24 19:23:20 daemon info radvd[11124]: attempting to reread config file
Jan 24 19:23:20 daemon info radvd[11124]: resuming normal operation
Jan 24 19:23:20 daemon warn kernel: netdev path : ppp1.1 -> ptm0.1 -> ptm0
Jan 24 19:23:20 daemon notice syslog: pppd:Using interface ppp1.1
Jan 24 19:23:20 daemon notice syslog: pppd:Connect: ppp1.1 <--> ptm0.1
Jan 24 19:23:20 daemon notice syslog: pppd:Couldn't increase MTU to 1500.
Jan 24 19:23:20 daemon err syslog: pppd:Couldn't increase MRU to 1500
Jan 24 19:23:20 daemon err syslog: pppd:Couldn't increase MRU to 1500
Jan 24 19:23:20 daemon crit syslog: PPP LCP UP.
Jan 24 19:23:20 daemon notice syslog: pppd:Remote message: CHAP authentication success
Jan 24 19:23:20 daemon info radvd[11124]: attempting to reread config file
Jan 24 19:23:20 daemon notice syslog: pppd:local LL address fe80::11cd:8b88:e784:5d40
Jan 24 19:23:20 daemon notice syslog: pppd:remote LL address fe80::0221:05ff:feae:0424
Jan 24 19:23:20 daemon notice syslog: pppd:local IP address 109.153.139.232
Jan 24 19:23:20 daemon notice syslog: pppd:remote IP address 172.16.14.2
Jan 24 19:23:20 daemon notice syslog: pppd:primary DNS address 81.139.57.100
Jan 24 19:23:20 daemon notice syslog: pppd:secondary DNS address 81.139.56.100
Jan 24 19:23:20 daemon crit syslog: Received valid IP address from server. Connection UP.
Jan 24 19:23:20 daemon info radvd[11124]: resuming normal operation
Jan 24 19:23:21 daemon info syslog: Hop limit : 64
Jan 24 19:23:21 daemon info syslog: Stateful address conf. : Yes
Jan 24 19:23:21 daemon info syslog: Stateful other conf. : No
Jan 24 19:23:21 daemon info syslog: Router preference : medium
Jan 24 19:23:21 daemon info syslog: Router lifetime : 4500 seconds
Jan 24 19:23:21 daemon info syslog: Reachable time : 0 milliseconds
Jan 24 19:23:21 daemon info syslog: Retransmit time : 0 milliseconds
Jan 24 19:23:24 daemon info radvd[11124]: attempting to reread config file
Jan 24 19:23:24 daemon warn kernel: ^[[0;36;44mBroadcom Packet Flow Cache flushing the flows^[[0m
Jan 24 19:23:24 daemon info radvd[11124]: resuming normal operation
Jan 24 19:23:24 daemon info syslog: dhcp6c restart..
Jan 24 19:23:36 daemon warn radvd[11124]: poll error: Interrupted system call
Jan 24 19:23:36 daemon warn radvd[11124]: Exiting, sigterm received.
Jan 24 19:23:37 daemon info radvd[11636]: version 1.8 started
Jan 24 19:23:37 daemon warn radvd[11636]: poll error: Interrupted system call
Jan 24 19:23:37 daemon warn radvd[11636]: Exiting, sigterm received.
Jan 24 19:23:37 daemon info radvd[11648]: version 1.8 started
Jan 24 19:23:37 daemon warn radvd[11648]: poll error: Interrupted system call
Jan 24 19:23:37 daemon warn radvd[11648]: Exiting, sigterm received.
Jan 24 19:23:37 daemon info radvd[11668]: version 1.8 started
Jan 24 19:23:37 daemon warn radvd[11668]: poll error: Interrupted system call
Jan 24 19:23:37 daemon warn radvd[11668]: Exiting, sigterm received.
Jan 24 19:23:37 daemon info radvd[11690]: version 1.8 started

Does this give any clues as to what is going on?

The rest of the log seems to be just Ipv4 DHCP OFFERs and ACKs

I think the issue is related to BT, I remember when I was testing the R2 version I had a similar issue, and after a week or so it sorted it self out.

Attached is latest firmware, try loading this firmware and see how you get on

[elwisty]

Thanks downloaded, I will see how it goes without the keep alive - as it used to go away quite quickly.

One thing I still don't understand. In the Home screen of the router (the new Basic one in this release - I don't remember seeing it in the last one) the two lines for
Default IPv6 Gateway :
IPv6 Address :
are both blank - but I clearly have an IPv6 address.

[billion_fan]

Thanks downloaded, I will see how it goes without the keep alive - as it used to go away quite quickly.

One thing I still don't understand. In the Home screen of the router (the new Basic one in this release - I don't remember seeing it in the last one) the two lines for
Default IPv6 Gateway :
IPv6 Address :
are both blank - but I clearly have an IPv6 address.

I think this is because BT don't issue a WAN IPv6 address, they assign the IPv6 to the LAN, from packet captures BT doesn't provide the RA information so router can't show it.

I wouldn't worry to much as long as you pass the IPv6 test, or go to the advanced page setup page of the router, you can check the LAN IPv6 address (I think BT and Sky do the same thing)