Loss of IPv6

[#1]

According to my IPv6 TBB graph at about 10:00 this morning, Wednesday 12 October 2016, I lost IPv6 connectivity, the IPv4 graph is OK and my PPP session has not dropped. I am on gateway ptn-ag03.

Has anybody else on the trial noticed this?

[#2]

Seems to be a loss of routing.

>tracert -6 www.google.com

Tracing route to www.google.com [2a00:1450:4009:814::2004]
over a maximum of 30 hops:

  1     3 ms    <1 ms    <1 ms  2a02:16c8:2000:b00:6203:47ff:fe00:1baa
  2     9 ms     9 ms     8 ms  2a02:16c8:0:1::1b
  3     9 ms     8 ms     9 ms  2a02:16c8:1:2::11
  4     9 ms     8 ms     8 ms  2a02:16c8::a
  5     *        *        *     Request timed out.
  6     *        *        *     Request timed out.
  7     *        *        *     Request timed out.
  8     *        *        *     Request timed out.
  9     *        *        *     Request timed out.

tracert -4 www.google.com

Tracing route to www.google.com [216.58.204.68]
over a maximum of 30 hops:

  1    <1 ms    <1 ms    <1 ms  192.168.0.101
  2     9 ms     8 ms     9 ms  lo0-central10.ptn-ag03.plus.net [195.166.128.192]
  3     9 ms     8 ms     9 ms  link-a-central10.ptn-gw01.plus.net [212.159.2.136]
  4     8 ms     8 ms     8 ms  xe-5-3-0.ptw-cr01.plus.net [212.159.0.108]
  5     8 ms     9 ms     9 ms  72.14.222.97
  6     8 ms    10 ms     9 ms  72.14.232.193
  7     8 ms     9 ms     8 ms  108.170.238.123
  8     9 ms     8 ms    11 ms  lhr25s13-in-f4.1e100.net [216.58.204.68]

[#3]

I also lost IPv4 connectivity at 17:15, router is still connected and has a ppp session.

(Edit: Time changed from 16:45 to 17:15)

[#4]

I can't see how it would be related but I know we relocated some of the BNG endpoints to a different data centre this morning.

[#5]

IPv6 connectivity should still be functioning then Bob? I lost mine several weeks back after several years of uninterrupted operation and just assumed it had been abruptly abandoned (like the first trial!).

[#6]

I may be suffering from router problems.

The log has the following messages at the time I lost IPv4

Oct 12 17:14:46 192.168.0.101 kernel: À*#015ÔÀ*#015Ô#036}Ù÷
Oct 12 17:14:46 192.168.0.101 kernel: #036<82>Z#011#036<83>#032#014#036<85>#032#024
Oct 12 17:14:46 192.168.0.101 kernel: #036<87>Ú#037
Oct 12 17:14:46 192.168.0.101 kernel: #036<89>Ú'#036<8a><9a>*#036<8b>Z-

Following a power cycle on the router I am now connected to pcl-bng01 and have IPv4 back. Will look at gateway hopping tomorrow and see if I can land on a non BNG gateway.

[#7]

Managed to land on pcl-ag03 at the first attempt.

IPv4 OK

IPv6 no connectivity

tracert -6 www.google.com

Tracing route to www.google.com [2a00:1450:4009:803::2004]
over a maximum of 30 hops:

  1     1 ms    <1 ms    <1 ms  2a02:16c8:2000:b00:6203:47ff:fe00:1baa
  2    11 ms    19 ms    10 ms  2a02:16c8:0:1::13
  3    23 ms     9 ms    16 ms  2a02:16c8:1:2::17
  4    19 ms    11 ms     9 ms  2a02:16c8::c
  5     *        *        *     Request timed out.
  6     *        *        *     Request timed out.
  7     *        *        *     Request timed out.
  8     *        *        *     Request timed out.
  9     *        *        *     Request timed out.
 10     *        *        *     Request timed out.

[#8]

Will do some digging, but needless to say it's going to be pretty low priority

[#9]

That does seem to sum up PlusNet & IPv6 - it's low priority.

[#10]

Without any action on my part or loss of ppp session, it is now working again.

TBB graph showed it happened at about 23:15 last night (13 Oct 2016).

Still on gateway pcl-ag03

tracert -6 www.google.com

Tracing route to www.google.com [2a00:1450:4009:801::2004]
over a maximum of 30 hops:

1 1 ms 1 ms 1 ms 2a02:16c8:2000:b00:..........
2 10 ms 10 ms 9 ms 2a02:16c8:0:1::13
3 10 ms 10 ms 9 ms 2a02:16c8:1:2::17
4 10 ms 10 ms 10 ms 2a02:16c8::c
5 14 ms 10 ms 10 ms 2a00:2380:300c:1001::2
6 10 ms 10 ms 10 ms 2a00:2380:10c::1c
7 12 ms 11 ms 10 ms 2a00:2380:c::40
8 10 ms 10 ms 10 ms 2a00:2380:5::1e3
9 10 ms 10 ms 10 ms 2a00:2380:2001:1000::3
10 11 ms 11 ms 11 ms 2001:4860:0:1::1827
11 12 ms 11 ms 12 ms 2001:4860:0:1::259
12 11 ms 11 ms 10 ms lhr35s01-in-x04.1e100.net [2a00:1450:4009:801::2004]

Trace complete.

[#11]

Also started working again for me overnight, on ag06:

[#12]

Loss of IPv6 (again) at 02:20 19 October 2016.

Exactly the same as before, IPv4 OK, ppp session did not drop, still on gateway pcl-ag03.

[#13]

Ditto, exactly the same time. My PPP connection bounced overnight and IPv6 came back, this time I'm on ag01:

I figure it's good to see some disruption after months of no activity. This is a trial, and the loss and recovery of IPv6 shows that someone is at least doing something with the configuration. 

[#14]

My IPv6 came back at approx. 23:15 last night (22 Oct 2016), I lost it again at 07:45 the next morning when my PPP session dropped and I reconnected onto a BNG gateway.

I would hope that configuration changes are being made with a proper change control process.

Needed five attempts to land on a non-BNG gateway: pcl-ag08 and get IPv6 back.

[#15]

Lost IPv6 again at 17:30 this evening, no loss of PPP session.