This is for other people's information as I think I might have worked this one out.

The Vigor was dropping and reconnecting the DSL line after about every 23 hours. I had been reading other people's posts on the subject and trying to get to the bottom of it. People had been changing modems, firewalls, cables, and all sorts. The last thing I wanted was to be dealing with a faulty router, as my D6400 had been working for seven years without a hitch.

I remembered, when I was programming up the firewall and watching packets, seeing some from my ISP that didn't make sense.

Slowly, the 23 hour-ish thing started to make sense. I was using the "Disable ping from internet" option.

I've disabled that option and now gone 24 hours without a drop.

I believe that my ISP is sending ICMP packets and if they doesn't get a response, that my ISP is resetting the DSL line. At least, that's how it looks from where I'm sitting.

If I wake up in the morning and it's dropped again, then at least I'll have the logs that are currently writing to a USB stick, but this might just have sorted it, so I thought I'd put it up for other people to consider.

Scratch that. I spoke too soon. It just went down again.

OK - I got my hands on the logs...

Code:

Jul 9 19:42:46 WAN WAN 1 is down. Jul 9 19:42:46 WAN DSL: Modem Shut Down from ADSL Phy Layer (0) Jul 9 19:42:47 WAN statistic: Session Usage: 144 (5 min average) Jul 9 19:45:07 WAN WAN1 PPPoE ==> V:1 T:1 PADT ID:4433 Jul 9 19:45:07 WAN WAN1 PPPoE ==> V:1 T:1 PADI ID:0 Jul 9 19:45:13 WAN WAN1 PPPoE ==> V:1 T:1 PADT ID:0 Jul 9 19:45:25 WAN WAN1 PPPoE ==> V:1 T:1 PADT ID:0 Jul 9 19:45:28 WAN WAN1 PPPoE <== V:1 T:1 PADO ID:0 Jul 9 19:45:28 WAN WAN1 PPPoE ==> V:1 T:1 PADR ID:0 Jul 9 19:45:28 WAN WAN1 PPPoE <== V:1 T:1 PADS ID:4561

There is an account of this here - https://community.plus.net/t5/Fibre-Broadband/Internet-connection-dropping-out-several-times-a-day/td-p/1608006 - so I'm going to contact Draytek UK support and see what they say.

Draytek UK support forwarded a firmware that isn't available on the web site.

That has been installed now and we'll see how it goes. Ironically, it is reporting a slightly faster upload speed, but I guess that this is probably going to be at a cost... possibly latency. We'll see how it goes.

I've just read your thread on the Plusnet site. Probably nothing to do with your issue but fyi in future as far as I am aware the BT test socket is behind the faceplate - the socket you plug into on the faceplate is already filtered inside the box behind the faceplate. Apologies if I've misunderstood and you knew that.

With that many disconnections I'm surprised that interleaving hasn't been applied and dropped your speed dramatically. If it hasn't maybe that's a clue?

Anyway, best of luck getting it sorted.

The plusnet thread was someone else, years ago, who experienced the same log errors as I have, and had noise on the line.

The faceplate, in my case, does not contain a filter. It does contain a capacitor which, I believe, helps with ringing signalling, but that's a potential wrong memory recall from years ago.

All was fine in my village until OpenReach installed FTTC in 2013. At that point, we started getting noise on the copper telephone line which we can hear on voice calls. That noise caused the broadband to drop and, believing that FTTP wasn't far away, I believe that Open Reach wouldn't put the money behind fixing the fault behind it, as the copper run in our case is a bit of a pain.

When an engineer came out to the fault, he took the faceplate off, plugged a handset into the master socket, dialled the quiet test and waited. His attitude led me to believe that he knew exactly what the problem was before he set foot in our home. He waited until the noise came on the line and happily signed off a report that said a fault existed outside our premises.

However, in the mean time a village in Wales that had been converted to FTTP was hit by a storm and lost power. Their telephones were out for weeks and they couldn't call emergency services because, obviously, no power at the receiving end meant no fibre optics. That caused protest from other villages and OpenReach is still chewing a brick over that because there is no technical solution to villages who are prone to power cuts, who have limited mobile phone coverage and rely on power from the telephone exchange to run telephony when the village loses power for days at a time.

If we report a noisy line, then an engineer... as far as I can tell... simply swaps our copper pair with another copper pair, which then moves the noise problem to someone else, who then complains and the cycle repeats. It seems that it would not be cost effective to repair the noise issue, when FTTP is still, yet again, supposed to be around the corner.

In the mean time, that leaves villages like mine stuck on copper with no date scheduled to translate to FTTP. (I queried my ISP again last week) And as we are also suffering a noise fault from which there is no relief, we're stuck.

I therefore don't blame Draytek, in fact I'm impressed that they have various different firmware versions that actually try and help in situations like this; because that's been the pitiful state of UK telephony for many years. It's just that in my case, the noise seems to be of an intensity that, regardless of the alteration to the balance of speed/latency/etc. of the DSL signal, it just reaches points where the router has no option other than to drop the DSL line.