How often should a FTTP connection drop?

I'm with Zen Internet on an Openreach 500M/70M fibre service. I've noticed that the connection rarely stays up for more than a couple of days. Looking in the Fritzbox logs I see multiple blocks of:-

24.09.26 05:00:32 IPv6 prefix obtained successfully. New prefix: 2a02:8012:xxxx::/48
24.09.26 05:00:32 IPv6 internet connection established successfully. IP address: 2a02:8011:d017:xxxx::1
24.09.26 05:00:31 Internet connection established successfully. IP address: 77.104.176.xxx, DNS server: 212.23.3.100 and 212.23.6.100, gateway: 51.148.77.128, broadband PoP: BNG1.IXN-LON-RE1
24.09.26 05:00:20 Timeout during PPP negotiation.
24.09.26 05:00:20 Internet connection cleared.
24.09.26 05:00:20 IPv6 internet connection was cleared; prefix no longer valid.

Since 16th Sept I've had 13 drops and reconnects. Is this normal or do I need to raise a case with Zen?

Thanks


Richard

Why do you want to use IPv6?

I have a Plusnet IPV6 service which I have never found much use for.

Can you disable the ipv6 service in the first instance, and see if that resolves things?

It's the default setting, I'm not knowledgeable enough to know what the optimum setting is for my home. Perhaps this would make a good new post topic. "Is IPV6 worth enabling on a home router"

Yes, I can do this and see the impact. I think its the fact that the PPP drops that then causes the ipv6 message rather than the otherway round, but lets see.

In reply to a post by RichG63:

I'm with Zen Internet on an Openreach 500M/70M fibre service. I've noticed that the connection rarely stays up for more than a couple of days. Looking in the Fritzbox logs I see multiple blocks of:-

24.09.26 05:00:32 IPv6 prefix obtained successfully. New prefix: 2a02:8012:xxxx::/48
24.09.26 05:00:32 IPv6 internet connection established successfully. IP address: 2a02:8011:d017:xxxx::1
24.09.26 05:00:31 Internet connection established successfully. IP address: 77.104.176.xxx, DNS server: 212.23.3.100 and 212.23.6.100, gateway: 51.148.77.128, broadband PoP: BNG1.IXN-LON-RE1
24.09.26 05:00:20 Timeout during PPP negotiation.
24.09.26 05:00:20 Internet connection cleared.
24.09.26 05:00:20 IPv6 internet connection was cleared; prefix no longer valid.

Since 16th Sept I've had 13 drops and reconnects. Is this normal or do I need to raise a case with Zen?

Thanks


Richard

Its not IPv6 at fault / cause here. Your PPP session is being dropped.

In other words you're barking up the wrong tree. Disabling IPV6 won't help. You need to determine why the PPP connection is being dropped.

In reply to a post by Pheasant:

Its not IPv6 at fault / cause here. Your PPP session is being dropped.

In other words you're barking up the wrong tree. Disabling IPV6 won't help. You need to determine why the PPP connection is being dropped.

Are there any logs from the Openreach ONT that I can see, like is it losing sync? I cant see anyway of getting more detail from the Fritzbox.

Edited by RichG63 (Fri 25-Sep-26 17:26:55)

No. There is no customer access to anything on the ONT other than the LEDs you can see.

In any event the ONT is a Layer 2 device for all intents, so all of this is transparent higher layer stuff. If the connection had a physical issue then the LOS lamp would be lit.

Swapping the Fritz for another router or ethernet gateway or device that has a PPP client would enable you to see if the problem travels or disappears.

Are the lights on the ONT visible … keep an eye on them, see if anything is dipping out.

Swap out the Ethernet cable between the ONT and router, another simple thing that you can tick off your list .

I’d call Zen and see if remote tests show anything.

Mine hasn’t dropped connection once since I got it many months ago

Yeah. A stable PPP connection can / should / oughta really stay up for many weeks if not months, or at least until when you reboot the router or the service provider resets connections on the BNG. It really should not be flapping.

In reply to a post by RichG63:

Are there any logs from the Openreach ONT that I can see, like is it losing sync?

Nope! All ONTs from Openreach is locked down!

i may have read that log from top to bottom, not bottom to top...........

😅

The next step is a support ticket with Zen with those log records. They can check with Openreach whether your ONT is dropping or if they're seeing your router terminating the PPP session.

As the log is only to the nearest second, and I am unsure whether the FritzBox guarantees the logs are in the correct order within a given second, there are two possible interpretations.

One possibility is that the IPv6 prefix delegation expired, causing the FritzBox to tear down the PPP session.

The second possibility is that Zen has torn down the PPP session.

Given that this drop happened at almost exactly 0500, when Zen sometimes tears down sessions for load balancing or maintenance, Zen may have torn down the session. However, 13 drops in 9 days is abnormal, and I would contact Zen support.

I am on Zen via Openreach FTTP, but my router is a pfSense virtual machine running on a Proxmox box, so I don't have experience with the Fritz routers Zen supplies. I can say that Zen's IPv6 is rock solid on pfSense.

OP has only emailed a snippet of the entire log, would be in their interest to check the timings of the other disconnections. Something is dropping the session, the IPv6 delegation being lost is just a side effect.

It’s either Zen forcing a reconnection from their end - which would strongly correlate to this occurring each day around 5am. Or it’s more random which would indicate a fault of some sort, possibly the router itself or a random network fault.

I agree with everything you say. While the posted log snippet is consistent with a planned session drop by Zen, I cannot believe that all the drops fall into this category.

As been said, have a word with Zen, have you tried resetting the router or just restarting it?

I had a problem months ago where things we not working too well and I restarted my router and things were fine. How long have you been on the service?

My fibre only drops when my network provider do a softweare update and that is early in the mornings.

As also been said, nothing to do with IPv6.

Thanks everyone for the replies and ideas. I have now restarted both the ONT and Fritzbox and will monitor over the next week. If it still keeps dropping then I'll contact Zen to see if they can offer any insights.

do monitor the ont as well, if you notice a drop in real time check that ont los light.

In reply to a post by Taras:

do monitor the ont as well, if you notice a drop in real time check that ont los light.

I usually setup a temporary indoor camera in these situations to capture the lights, its easy to then review the logs and lights at the same time.

In reply to a post by PCJM40:

In reply to a post by Taras:

do monitor the ont as well, if you notice a drop in real time check that ont los light.

I usually setup a temporary indoor camera in these situations to capture the lights, its easy to then review the logs and lights at the same time.

that would obviously be the best option especially if your out of the property often

In reply to a post by Taras:

In reply to a post by PCJM40:

In reply to a post by Taras:

do monitor the ont as well, if you notice a drop in real time check that ont los light.

I usually setup a temporary indoor camera in these situations to capture the lights, its easy to then review the logs and lights at the same time.

that would obviously be the best option especially if your out of the property often

That would be useful as the ONT is not next to the router.

Makes sense to give the full history, when the FTTP was installed the engineer was not happy locating the ONT where the NTE was as the drop wire ran through the loft down a wall and then under some floorboards before popping up in the middle of the house. In the end the neatest solution for both of us was to put the ONT in the porch and then I ran a cat5 from there to the router location. I had a Zen 100Mb solution and I never noticed any drops. A month ago my contract was renewed and Zen offered the 500Mb service for a £1 less than the 100Mb so I switched. I don't actually notice the disconnects but I was looking through the logs to dignose a wifi connection issue and noticed all the drops. I'll keep better records now of exactly when the drops happen as it will help Zen too if I need to contact them.

Hmm. How much do you trust that Cat5e connection? Was it ever tested - I’m presuming not.

In reply to a post by Pheasant:

Hmm. How much do you trust that Cat5e connection? Was it ever tested - I’m presuming not.

The Openreach engineer helped me do it as he had to leave the broadband in a fully working state, i.e. connected to the router. He used a tester at both ends after we had finished. It was lucky that he had the time, but their records didn't show that the pole feeding my property was a 'do not climb' so he had to wait for a cherry picker.

However, it would be prudent to see if I can borrow a tester to double check.

Did you install a pre-made long flylead or did you crimp on RJ45 plugs (or jacks)?

I don't actually notice the disconnects but I was looking through the logs to dignose a wifi connection issue and noticed all the drops.

Hmmm 🤔
I would make that call to Zen now … if their tests aren’t seeing these drops .. it seems like you might be chasing a red herring here.

In reply to a post by PCJM40:

I usually setup a temporary indoor camera in these situations to capture the lights, its easy to then review the logs and lights at the same time.

^^ this proved more evidence! Cheap tapo indoor camera will do it (make sure u buy microsd card for recording storage)

In reply to a post by RichG63:

Thanks everyone for the replies and ideas. I have now restarted both the ONT and Fritzbox and will monitor over the next week. If it still keeps dropping then I'll contact Zen to see if they can offer any insights.

ONT light loss tester is needing by an engineer if it kept dropping out!

This sounds a lot like the issues I had with my PlusNet FF service.

See the thread here: https://forums.thinkbroadband.com/plusnet/t/4788018-...

The end result was the dropouts were caused by a faulty fibre splice from the original installation and a duff ONT. Openreach replaced both and it has been rock solid ever since.

In my case, PlusNet blamed the router, my cabling and anything else they could to avoid admitting there was a problem.

Notably, whenever they tested, they could see nothing wrong, because when they tested, the connection was up and running.

At least with Zen you don't have the problem of them blaming the router, since they provided it.

In reply to a post by Pheasant:

Hmm. How much do you trust that Cat5e connection? Was it ever tested - I’m presuming not.

I'm seconding this. Rich have you got a spare cat5e cable.......... thats long enough?

if not a £5 cat 7 flat cable will work as a testing Guinea pig !

In reply to a post by Axolotl:

The end result was the dropouts were caused by a faulty fibre splice from the original installation and a duff ONT. Openreach replaced both and it has been rock solid ever since.

A faulty splice, can cause intermittent issues along with a duff ont as you found out.

In reply to a post by Axolotl:

In my case, PlusNet blamed the router, my cabling and anything else they could to avoid admitting there was a problem.

Notably, whenever they tested, they could see nothing wrong, because when they tested, the connection was up and running.

Sadly when its intermittent issues, the above can happen, its frustrating for you because you are witness to it and the isp is not .

In reply to a post by RichG63:

Thanks everyone for the replies and ideas. I have now restarted both the ONT and Fritzbox and will monitor over the next week. If it still keeps dropping then I'll contact Zen to see if they can offer any insights.

Did you capture or save the full logs before you reset the router by any chance? Shame if you didn’t as that would have been very useful for fault diagnosis.

You will now need to wait for the faults to reoccur. Just something to be mindful of in future.

In reply to a post by PCJM40:

In reply to a post by Taras:

do monitor the ont as well, if you notice a drop in real time check that ont los light.

I usually setup a temporary indoor camera in these situations to capture the lights, its easy to then review the logs and lights at the same time.

That is a good idea,

Fritz!Box defaults to “clearing the internet connection” every night, dropping and reconnecting PPP.

To disable, open the UI and:

- Navigate to Internet > Account Information
- Expand “Change Connection Settings”
- Uncheck “Delay disconnection by provider to the period between”
- Click Apply

The connection should remain active 24x7. I typically have many months of uptime, usually only bouncing after a firmware update .

Edited by DougM (Sun 27-Sep-26 08:35:12)

....or change that setting to force this to happen daily at a time when it won't affect you (eg in the small hours of the morning). I have had mine (bought, not provided by Zen, and a different ISP for service) doing this since new ~5 years ago. Mine is set to between 1am and 2am. Never had any other disconnects to press.

The OP should be able to conclude from the logs if this is happening the same time each day, or if not then its something else…

There really should be no need these days to tear down a PPP session every 24 hours. The BNG can easily keep the session alive for months on end. Hopefully the router software is stable and suffers from no memory leaks for this to be required.

Edited by Pheasant (Sun 27-Sep-26 10:56:17)

In reply to a post by DougM:

Fritz!Box defaults to “clearing the internet connection” every night, dropping and reconnecting PPP.

To disable, open the UI and:

- Navigate to Internet > Account Information
- Expand “Change Connection Settings”
- Uncheck “Delay disconnection by provider to the period between”
- Click Apply

The connection should remain active 24x7. I typically have many months of uptime, usually only bouncing after a firmware update .

For what reason? Seems a strange thing to do, I have never known that from any router i have had or seen.

I checked and my Fritzbox didn't default to that mode of operation so not that.

I have never known that from any router i have had or seen.

My Fritz!Box 3940 (3490? dunno some collection of those numbers ) used to do it for FTTC... i don't think it does now its an AP for the ISP supplied FTTP router.

In reply to a post by RichG63:

I checked and my Fritzbox didn't default to that mode of operation so not that.

Have you logged the fault with Zen support?

What are the router logs showing - stable or any more (random) session disconnects?

In reply to a post by zyborg47:

For what reason? Seems a strange thing to do, I have never known that from any router i have had or seen.

It's a thing in Germany https://de.wikipedia.org/wiki/Zwangstrennung (run it though a translator) but more applicable to DSL & ISDN connections rather than FTTP

I may have solved it, however, before I share my findings I need to give it at least 5 days to confirm it stays up. I may well be counting my chickens....

You going to give us a little clue? 😎🤣

Well I don't believe anyone mentioned the issue which made me rather sceptical but it's been up for 3 days now so still looking good....

In reply to a post by RichG63:

Well I don't believe anyone mentioned the issue which made me rather sceptical but it's been up for 3 days now so still looking good....

So what was the issue? Glad it is working now.

Thats impressive as even ONT updates, router updates or maintenance usually knock out a connection for a few mins overnight

In reply to a post by bobble_bob:

Thats impressive as even ONT updates, router updates or maintenance usually knock out a connection for a few mins overnight

I guess… but I’ve not noticed it.

In reply to a post by bobble_bob:

Thats impressive as even ONT updates, router updates or maintenance usually knock out a connection for a few mins overnight

ONT updates, for me on Openreach and Hyperoptic at least (can't say for other networks) - are pretty infrequent. So infrequent I can't remember the last time the connection dropped for one.

Another benefit of running your own router is that you can choose the update window and timing.

I have one every few month, usually around 12.30am so assume its some ONT update/maintenance. Also run my own router so its not that

In reply to a post by bobble_bob:

Thats impressive as even ONT updates, router updates or maintenance usually knock out a connection for a few mins overnight

15 days 21 hours 59 minutes mine have been up and the only reason it disconnected was because I updated the router.
Still no idea what the isuues was, he seems to scared to tell us.

BT hubs do a reboot every fortnight or so, usually during the night but occasionally while I am using the computer. Ethernet light goes off but PON light stays solid and it negotiates a new network connection when it comes back. Maybe this is what the OP is seeing? Otherwise my connection has been solid for the past three years.

Hahaha. I clearly tempted the gods!

This morning just before 7am my Aquiss FTTP connection flapped the PPPoE connection about 6 times in the space of 2.5 minutes, then settled and reconnected itself. The base latency on the service has now doubled, so something has gone awry. Ticket has been raised.

Prior it had been stable since the router was last updated a few weeks ago.

Edit: Flapped again now at 08:03 and has reconnected once more. This time the base latency seems to have returned to normal.

Edit 2: second flap, was actually not a flap, but a manual boot to get my session back on the primary LNS, so Martin tells me. There was apparently a BTW issue going on and I was initially brought back on the backup LNS, which explains the base latency shift.

Edited by Pheasant (Thu 01-Oct-26 08:21:52)

In reply to a post by dsergeant:

BT hubs do a reboot every fortnight or so, usually during the night but occasionally while I am using the computer. Ethernet light goes off but PON light stays solid and it negotiates a new network connection when it comes back. Maybe this is what the OP is seeing? Otherwise my connection has been solid for the past three years.

That is the problem with using an ISP router, now and again Zzoomm will do a network update at night, but not done one for a few weeks.

The OP connection seems to break too often to be a router update.

Oh no, I hope it gets sorted soon.
Mine is still fine, touch wood, I got wood here somewhere

i did not think FTTP had latency as such or do you mean ping?

I got wood here somewhere

TMI Adrian.

In reply to a post by zyborg47:

Oh no, I hope it gets sorted soon.
Mine is still fine, touch wood, I got wood here somewhere

i did not think FTTP had latency as such or do you mean ping?

Yeah it’s resolved itself I think. This connection is usually solid as a rock. You can see from the BQM where it’s failed over to the backup LNS cluster just before 7am and then Martin manually reset the connection after 8am and it came back to the primary LNS cluster….

https://www.thinkbroadband.com/broadband/monitoring/...

Unfortunately it's not, it worked fine for 4 days which is the longest it's been up for weeks and then last night it dropped 4 times.

To help diagnose I had routed the WAN connection through my Netgear managed switch (VLAN of course). This allowed me to monitor the Ethernet traffic and see if the Ethernet was experiencing any CRC errors (Zero). What I did notice was that the link speed was dropping to 100Mbps which I presume was due to IEEE 802.3az and then back up to 1G when the circuit was being used. I noticed that most of the PPP drops were at the same time this rate was changing. A search brought up a few other reports of this causing the Openreach ONT to drop connection, so I prohibited EEE on the switch and kept the link at 1Gbps and it was stable so I had hoped that I had found it, but obviously not. Routing the connection through the switch will also allow me to snoop on the data and run a packet capture which would provide some more accurate timings, but I'm a bit short of time at the moment...

I've also contacted Zen but unfortunately they don't appear to have any further diagnostics, they can see the PPP drops but that's all. So they've asked me to factory reset the ONT, and if that doesn't fix it I will get sent another router to try, I'd hoped for more. They didn't believe it would be the fibre, it either works or it doesn't...

If the ONT factory reset doesn't work then I'll also move the router to be next to the ONT with a short Ethernet cable, I don't think this is a contributing factor but it will eliminate it. Trouble is waiting 4-5 days every time is slow...

Good thinking with the network switch / VLAN for diagnostics.

The hint is the link speed dropping to 100Mbps / co-incidental with the PPP drops.

So if you can I'd move the router now to be next to the ONT with a short (known good) Cat5e/6 patch lead. Then just run the long connection back to the rest of your network. Removing the long network connection between the ONT and router, eliminates a potentially cause of error.

If the problem persists after a few days then you have a flappy GbE port on either the ONT or the router.

Annoyingly I'd also set a camera up to monitor the ONT lights and it crashed 30 minutes before the drops started, argh! I new it was flakey that's why I'd stopped using it but talk about bad timing...

How’s the stability been the past few days?

The factory reset of the ONT made no difference, dropped after a few hours. So I followed the advice given here, I pulled a new Ethernet cable in parallel with the original. (Probably should have pulled 2 in when I did it). I added a new punchdown outlet at each end and used 2 new patch leads, one at each end. I also replaced the ONT PSU so everything is new. Been up for 3 days so far so time will tell. If it stays up for 10 days I'll tempted ro start swapping bits back in as I'd like to know which part is faulty.

I wasn’t very hopeful on the ONT reset. If there is a fault with these it’s generally hardware and not firmware/software.

See how it goes with the new cabling. If it drops again you’re back to the ONT and router ports. Either one will likely be the culprit.

Assuming your camera didn’t work / spot any drops on the fibre side of the ONT / LOS light…

Thanks for reminding me, yes the camera caught the LAN LED pulsing hence the cable replacement. If anyone wants to lose a minute of their life you can watch it here.

Video

In reply to a post by RichG63:

Thanks for reminding me, yes the camera caught the LAN LED pulsing hence the cable replacement. If anyone wants to lose a minute of their life you can watch it here.

Video

The LAN light WILL flash intermittently with data transfer, even if nothing appears to be going on.

Flashing/blinking is fine as you say, but completely off is not good. Should default back to solid green when link is up (no data passing).

No red visible on 2nd LED (LOS) so that is good but I don’t think Ethernet is completely in the clear…

The ethernet led will go off if the router is rebooting which may have absolutely nothing to do with the ONT and fibre line.

Will wait for the OP to elaborate, but there was no suggestion of a router reboot. What was clear from the log snippet was that the PPP session was dropping, but it did not look like there was any sort of reboot going on.

On the other hand a port flapping on either the router or ONT could cause a loss of link. Lower layer events like this are typically not logged in consumer routers like this. It’s just not clear from the grainy video necessarily either.

Edited by Pheasant (Tue 06-Oct-26 07:47:40)

In reply to a post by Pheasant:

Flashing/blinking is fine as you say, but completely off is not good. Should default back to solid green when link is up (no data passing).

No red visible on 2nd LED (LOS) so that is good but I don’t think Ethernet is completely in the clear…

Not sure how a Ethernet cable can go bad unless it is treated badly. I have cables here that are years old and still working, one of them was rolled up in a way it should not be and stored for a couple of years and yet when i got it out, it still worked.

A mate have one that goes outside the house and is not protected and been out there for over 10 years and still works.

Edited by zyborg47 (Tue 06-Oct-26 08:30:59)

Any cable can be borderline on performance and still appear OK. Hardware faults do happen. It’s all within the realms of possibility.

In reply to a post by Pheasant:

Any cable can be borderline on performance and still appear OK. Hardware faults do happen. It’s all within the realms of possibility.

also borderline cables, may work fine on one device and not on another ......

I’ve got Madonna in my head now. Ear worm. Borderline. 🤣

Correct, the pulsing LAN light happened at the exact time as the PPPoE dropped no router reboot. Now whether the router drops the Ethernet before it starts to resync I have no idea but it seems unlikely?

The only time I have had an Ethernet connection fail previously it was the wall socket prongs not making a reliable contact to the plug pins. Unplugging and replugging a new times cleared the issue, but that didn't work in this instance. Or a well used patch lead but again they were new out of the packet. I was also suspicious as it seemed to start when my FTTP was upgraded from 100 to 500 but that may just have been coincidence.

In reply to a post by Pheasant:

I’ve got Madonna in my head now. Ear worm. Borderline. 🤣

i do like that song lol Dammit .............. its in my head now

Seems very unlikely that the router would initiate a full link reset after a PPP connection drop. The other way around however makes sense.

In the first instance it will be easier for them to replace the router than getting Openreach out to replace the ONT.

can be checked obviously with a direct connection from pc to ont

Good point. However OP would beed to be prepared to leave that machine connected and wait for any flaps to happen on the link which may not happen for a few days (as evidenced).

So this arrangement could be disruptive if, with say general internet access, without some further reconfig. to share the connection.

Also that machine would need a strong / reliable / configured firewall given its directly facing the open internet with no other firewall protection and also needs PPPoE client configured - a Linux box would be ideal. A Windows box probably not the one for this job.

In reply to a post by Pheasant:

Any cable can be borderline on performance and still appear OK. Hardware faults do happen. It’s all within the realms of possibility.

I am not saying it is not possible, just seems unlikely, but stranger things have happened.

In reply to a post by Taras:

can be checked obviously with a direct connection from pc to ont

Umm, not really advisable, unless as said above a Linux box is used and one that is not important and can be reinstalled if needed.

In reply to a post by zyborg47:

In reply to a post by Pheasant:

Any cable can be borderline on performance and still appear OK. Hardware faults do happen. It’s all within the realms of possibility.

I am not saying it is not possible, just seems unlikely, but stranger things have happened.

I've had a Huawei ONT ethernet port fall out of gigabit capability and only connect on 100BaseT. Was a few years ago and did need an 'act of god' for good measure 😂 but did happen

Cabling issues - how long have you got? 😉

In reply to a post by zyborg47:

In reply to a post by Taras:

can be checked obviously with a direct connection from pc to ont

Umm, not really advisable, unless as said above a Linux box is used and one that is not important and can be reinstalled if needed.

huh!!!!!!!!!!!!! that really makes no sense.

you plug a cat 5e or greater into the pc, then into the ont. configure - test and remove.

https://support.aquiss.net/en/knowledgebase/article/...

In reply to a post by Taras:

In reply to a post by zyborg47:

In reply to a post by Taras:

can be checked obviously with a direct connection from pc to ont

Umm, not really advisable, unless as said above a Linux box is used and one that is not important and can be reinstalled if needed.

huh!!!!!!!!!!!!! that really makes no sense.

you plug a cat 5e or greater into the pc, then into the ont. configure - test and remove.

https://support.aquiss.net/en/knowledgebase/article/...

Adrian was making the point that doing it is risky if left for a long time unprotected

Correct. The flap (if it is that) doesn’t seem to appear for a few days…

i said test and remove.

Unfortunately in this case, that wouldn’t prove much other than the initial link was connecting at GbE line rate. To observe any flapping the connection would need to remain active until / if the flapping was observed. That could be days, judging by how long it’s been between PPP session drops.

In reply to a post by Taras:

i said test and remove.

Are you skim reading again and missing the detail?

I had a bizarre hardware fault years ago with my router. Thing worked fine, but as soon as anyone made a Facetime/Whatsapp video call, the wifi would drop 2 mins into the call on all devices attached to the router, and require a reboot to bring it back up

Wired worked fine the whole time, just wifi that dropped and only after a video call was in progress

Edited by bobble_bob (Wed 07-Oct-26 18:36:10)

In reply to a post by Pheasant:

I've had a Huawei ONT ethernet port fall out of gigabit capability and only connect on 100BaseT. Was a few years ago and did need an 'act of god' for good measure 😂 but did happen

Cabling issues - how long have you got? 😉

The only issue I have had with Ethernet cables is the tabs that hold the things in.

In reply to a post by Taras:

huh!!!!!!!!!!!!! that really makes no sense.

you plug a cat 5e or greater into the pc, then into the ont. configure - test and remove.

https://support.aquiss.net/en/knowledgebase/article/...

It make loads of sense, you have a unprotected computer directly connected to the internet. Well almost unprotected, The windows firewall and security may protect if for a bit.
I would not risk it.

In reply to a post by bobble_bob:

I had a bizarre hardware fault years ago with my router. Thing worked fine, but as soon as anyone made a Facetime/Whatsapp video call, the wifi would drop 2 mins into the call on all devices attached to the router, and require a reboot to bring it back up

Wired worked fine the whole time, just wifi that dropped and only after a video call was in progress

It is strange how some things work or don't.
Like on FTTC here, when the only thing that would connect was a Huawei modem. It all went strange when I had a fault on my land line and when they fixed it, the ECI modem would not connect, so I thought it had gone pop. Borrowed a Huawei modem from my next door neighbours and that worked, so Plusnet sent me a all in one hub, that would not work, then a Zyxel router, but again nope. I connected them to the Huawei modem I had borrowed from my neighbour, that was the only way i could get connected.

Plusnet sent Openreach down, they checked every thing here and at the cabinet, even their own equipment would not pick up a connection here. They gave up and told me to keep the Huawei modem inline, I got one from ebay and i had that and Zyxel connected for a few years. At some a few years later i decided to reconnect the Zyxel directly and it worked . then a year later it started to fail and stopped working, which is when i bought the TP-link I am using now and connected it up to the Huawei modem until I changed to FTTP.

To this day, I have no idea what happened and i don't think Openreach do.

Just strange how things go. A mate did look at the Zyxel, a capacitor had gone, he did replace it, so it is there for a spare.

In reply to a post by zyborg47:

It make loads of sense, you have a unprotected computer directly connected to the internet. Well almost unprotected, The windows firewall and security may protect if for a bit.
I would not risk it.

Surely in 2026 you connect ethernet to the router which is giving exactly the same protection as WiFi. This isn't 1999 when I had cable modem with ethernet into Win98 PC and no router, no firewall, and Steve Gibson was telling everyone to install ZoneAlarm. (some people had the USB ADSL modem that caused problems).

Or you connect Ethernet from computer to computer (mac->mac, PC->PC, mac->PC etc) and configure private IP addressing and then you can transfer files at 1Gbps.

Edited by jchamier (Thu 08-Oct-26 09:55:18)

Did Zen indicate they would replace the router (or ONT) if the sessions kept randomly disconnecting, after the resets?

They would first sent a new router and if that didn't fix the issue then they would contact Openreach. It still seems odd to me that they can't put a request in to Openreach for ONT data/logs. Surely Openreach must have some form of diagnostics of their fibre network?

We've been up 7 days now though so it looks like it was one of the removed Ethernet components....

I've been trying to find someone with a proper tester but no luck so far. I've also been looking for an open source home build version (not just a continuity checker), does anyone know of one?

In reply to a post by jchamier:

Surely in 2026 you connect ethernet to the router which is giving exactly the same protection as WiFi. This isn't 1999 when I had cable modem with ethernet into Win98 PC and no router, no firewall, and Steve Gibson was telling everyone to install ZoneAlarm. (some people had the USB ADSL modem that caused problems).

Or you connect Ethernet from computer to computer (mac->mac, PC->PC, mac->PC etc) and configure private IP addressing and then you can transfer files at 1Gbps.

Taras is saying about connecting the computer directly to the ONT to test that it is not a router problem. That is connecting directly to the net like we used to do many years agao before we had routers. Even in those days it could be iffy, these days it would be worse, unless the Windows firewall protect you.

I had a ADSL usb modem, the Stingray thing or frog as some people called it, in the loft I think somewhere.
I remember ZoneAlarm.

i would never connect my computer directly to the net these days.

In reply to a post by RichG63:

They would first sent a new router and if that didn't fix the issue then they would contact Openreach. It still seems odd to me that they can't put a request in to Openreach for ONT data/logs. Surely Openreach must have some form of diagnostics of their fibre network?

We've been up 7 days now though so it looks like it was one of the removed Ethernet components....

I've been trying to find someone with a proper tester but no luck so far. I've also been looking for an open source home build version (not just a continuity checker), does anyone know of one?

That is good news,
A tester to test the cable? May as well just buy a new cable.

In reply to a post by RichG63:

They would first sent a new router and if that didn't fix the issue then they would contact Openreach. It still seems odd to me that they can't put a request in to Openreach for ONT data/logs. Surely Openreach must have some form of diagnostics of their fibre network?

We've been up 7 days now though so it looks like it was one of the removed Ethernet components....

I've been trying to find someone with a proper tester but no luck so far. I've also been looking for an open source home build version (not just a continuity checker), does anyone know of one?

Glad you seem to have nailed it.

p.s. Do I win anything for having suggested the solution as the 1st reply ?

https://forums.thinkbroadband.com/fibre/t/4790368-re...

I’ve never seen anything resembling a DIY / home brew ‘proper’ cabling tester - that is one that will test all the parameters including length, attenuation, near and far-end cross talk, ACR, powersum ACR and then certify to either an EIA/TIA or ISO based cabling standard. All calibrated and up to date.

The proper ones are specialist hybrid between a 4-channel time domain reflectometer, a digital scope and a PC that has a built-in database of standards and cable specs.

We’ve just sent our Fluke DSX8000 test set in for calibration and the service alone is well over a grand. They are north of 12 grand to buy new, so that gives you some idea of the complexity of the beasts.

It’s not going to be worth it to get someone out to test it / nor to rent a test set. It would be cheaper to get them to run new fixed cabling in and test that. But that’s neither practical or proportionate in the circumstances.

For a single link / cable I’d just live with it.

A short run length, keeping twists up in the pairs all the way to the back of the jack (don’t over twist either to compensate) and a simple 4-pair continuity test / conductor order will be sufficient and a clean connection onto the IDC block is good enough.

In reply to a post by zyborg47:

In reply to a post by RichG63:

They would first sent a new router and if that didn't fix the issue then they would contact Openreach. It still seems odd to me that they can't put a request in to Openreach for ONT data/logs. Surely Openreach must have some form of diagnostics of their fibre network?

We've been up 7 days now though so it looks like it was one of the removed Ethernet components....

I've been trying to find someone with a proper tester but no luck so far. I've also been looking for an open source home build version (not just a continuity checker), does anyone know of one?

That is good news,
A tester to test the cable? May as well just buy a new cable.

It’s not a long patch lead, out of a packet - chucked in under the floorboard.

As I understand it, he’s installed solid core cabling onto jacks / into keystones / faceplates and then patch leads either end…

So probs a bit more work than just opening a packet from Amazon.

Only if the ISP doesn’t use an encapsulation like PPPoE. Each ISP is different, yours might be DHCP.

The Windows Firewall or macOS firewall IS sufficient for this, otherwise people with laptops who take them to coffee shops, or airports, or hotels, would be getting hacked on a daily basis. We know that is not the case.

In reply to a post by Pheasant:

In reply to a post by zyborg47:

That is good news,
A tester to test the cable? May as well just buy a new cable.

It’s not a long patch lead, out of a packet - chucked in under the floorboard.

As I understand it, he’s installed solid core cabling onto jacks / into keystones / faceplates and then patch leads either end…

So probs a bit more work than just opening a packet from Amazon.

That's correct, I thought I had done a decent job which is why I'd like to know which bit has failed...

In my experience it’s almost always the ends: that is the terminations.

It’s preferable to match the manufacturer of the cable to the connectors for a start (a single manufacturer system), but I realise that this is a domestic install rather than commercial - so mix and match is pretty normal and still works - but that is the first thing behind a good installation.

Then it’s down to good workmanship - score the sheath, don’t cut it all the way though so you don’t damage the individual pairs/conductors, their insulation is super thin. A gentle bend once scored and the sheath should come away. Peal back the pairs so they’re radiating outwards. Trim any centre strength / pair separator out of the way. Leave the twists as they are. Feed the pairs in through the jack (each manufacturer has their own way). Open the twists just enough to feed the conductors onto the respective IDC position. Use a 110 punch down tool / blade facing out to seat and trim. Move to the next conductor in the pair. Rinse and repeat. Put any retaining caps on. Job done.

Very rarely do you get a fault mid way along a cable - and it’s usually a broken conductor if that’s the case, so quite easy to spot on a test.

Cable performance (as in does it meet the EIA/TIA category or ISO class) in a permanent link is invariably down to how well the ends have been terminated.

Are the sessions stable now?