User Error 0x3 and 0x4 - frequent drops OpenReach given up

[#1]

@Dan_the_Van 

Thank you I had come across the RouterStats but assumed it was the BT version of the router only and as I'm running a Mac and have no Windows BootCamp option (let alone I.E.) I'm a bit stuck.

I do have Edge on my Mac and a x86 Linux box (plus a NAS and a few Pi boxes) but the HomeHub does not seem to support SNMP or telnet and I'm not up to creating a screen scrape with Python and Beautiful Soup at the moment - maybe in a few weeks when my mind settles again.

[#2]

@jab1 

The package was 40/10 when I signed up but described as 'Unlimited Fibre' on the bills and account pages.

Like I say, the connection speed is slowly dropping at the moment as I keep changing kit - it's like the system is going through its training again and again but some days goes up others down.

[#3]

@MisterW 

Would agree except the last time OR came in the house they were reporting the ISP was running tests before the line dropped and he said these '0x3 and 0x4' reports are the router logging the test.  These 'tests' occurred anything up to 3 minutes before the router dropped the DSL connection (assuming times are correct - the TP-Link was sync'd to an external 3rd party NTP server).

Hence my original question - does anyone know what these actually are?

SNR drifts in and out depending on the day of the week - I did have a record of them and was seeing changes every hour but I've binned my records out of frustration TBH and am surprised I can put these notes up here.

I have OR due to call tomorrow BUT they have cancelled one call already.

[#4]

---

@adbeasley wrote:

@jab1 

The package was 40/10 when I signed up but described as 'Unlimited Fibre' on the bills and account pages.

Like I say, the connection speed is slowly dropping at the moment as I keep changing kit - it's like the system is going through its training again and again but some days goes up others down.

---

So you are on a 40/10 service - the 80/20 is 'Unlimited Fibre Extra'. While ever you are connecting/disconnecting equipment, the DLM equipment in the cab will think there are even more problems, and continue to try and compensate by increasing the SNR and lowering your connection speed.

I've never had a PN Hub or your other modem/router, but if there is a connect/disconnect button on the control panel, I would always use that before physically removing the Hub.

[#5]

Would agree except the last time OR came in the house they were reporting the ISP was running tests before the line dropped and he said these '0x3 and 0x4' reports are the router logging the test.

Complete and utter rubbish! ISP's don't run routine tests. There are tests that they run to investigate a fault. These are required by Openreach ( they host the testing systems ) before a fault can be reported.

As I said above, those errors are reported by the PPP protocol losing contact with the gateway because the underlying DSL connection was lost.

[#6]

Hello @adbeasley, thanks a lot for reaching out and I'm really sorry to hear about the problems that you're having with your connection. At this point, I'm not going to make any recommendations / assessments as you have an engineer visit booked in for this afternoon. 

Let us know how you get on following the visit and we'll be happy to follow up.  

[#7]

@jab1 

---

...

So you are on a 40/10 service - the 80/20 is 'Unlimited Fibre Extra'. While ever you are connecting/disconnecting equipment, the DLM equipment in the cab will think there are even more problems, and continue to try and compensate by increasing the SNR and lowering your connection speed.

I've never had a PN Hub or your other modem/router, but if there is a connect/disconnect button on the control panel, I would always use that before physically removing the Hub.

---

I normally use a soft reboot, then a power cycle and finally only swap when the desk say to do so (often these have been their first option). My router is a lot more powerful and configurable than the PN basic one and gives decent logging to my syslog server rather than the screen shots / copy paste I am doing at the moment.

At some-point this may get swapped out for commercial grade kit with mesh WiFi but that's on hold for now.

Big thanks for all the support and help so far - I did manage to get around 5 hours sleep last night but on edge with OR coming today - not taking any support tablets though as I have to get used to folk coming to me!

[#8]

Well not sure what OpenReach did yesterday but they now need to visit the house. Why they did not come yesterday I do not know...

So another day shall pass.

Looks like the possible max speed has gone up, noise margin changed and we binged on Prime / Netflix / YouTube:

6. Data rate: 9879 / 33478
7. Maximum data rate: 9879 / 61723
8. Noise margin: 6.4 / 17.9
9. Line attenuation: 33.1 / 21.6
10. Signal attenuation: 33.0 / 19.9
11. Data sent/received: 3.5 GB / 67.2 GB

[#9]

Hi there,

I'm really sorry to see an engineer hasn't shown up.

It looks like the engineers are incredibly busy in the area and have struggled to get through all their outstanding jobs so an engineer couldn't take a look at your connection today.

Let me know how it goes tomorrow. 

[#10]

Well the engineering visit was interesting.

His call log showed the pile of previous visits though he was new to the house so I explained the fault is intermittent from my point of use and explained what had been changed previously. He also had a screen based report that showed lots of drops and he questioned the power on / off of the router - as normal.

He ran a couple of test but it showed nothing (do I need to use another word than 'intermittent') - surprise surprise

He then kindly changed the face plate back to the inbuilt filter - yes a minor win 🙂

Then he could not get a PPP session going. A rebuild of his test kit, a full switch off / back on (yup loved that), the odd tap and wiggle and a remote reset of something in the exchange later, still no PPP session...

He then left site saying it would be between 10 minutes and an hour or so to organise a new laser module in the cabinet (I assume he means a fibre to copper transceiver) and he will ring me when its changed.

Internet sprung into life a couple of minutes before he rang back and so far no drops.

I have had one pause (YouTube via Amazon buffer rang out and no web pages would load on two devices) but this may have been a WiFi issue (no Ethernet devices handy) though no messages where in the event log around this time.

I'm 'constantly' (well 8-10 an hour) getting messages such as:

20:13:36, 25 Mar. (411110.610000) Device disconnected: Hostname: Unknown-​5c-​96-​9d-​67-​63-​ff IP: 192.168.1.5 MAC: 5c:96:9d:67:63:ff
20:13:36, 25 Mar. (411110.610000) Device disconnected: Hostname: Unknown-​6c-​70-​9f-​d8-​49-​51 IP: 192.168.1.4 MAC: 6c:70:9f:d8:49:51

but I do not know exactly what these are:

192.168.1.4 and .5 are my APs

5C:96:9D and 6C:70:9F are both Apple prefixes for MAC addresses and we have 6 or 7 Apple devices around (depending who is here with what) BUT some will have 'WiFi privacy' (random MAC addresses on the WiFi link) on them so need to check and lock them down to see what's dropping in / out like this.

Link wise:

5. DSL uptime: 1 days, 04:22:06
6. Data rate: 9999 / 38171
7. Maximum data rate: 10439 / 58540
8. Noise margin: 6.7 / 13.8
9. Line attenuation: 33.8 / 21.0
10. Signal attenuation: 33.6 / 19.6
11. Data sent/received: 4.8 GB / 103.5 GB

I did notice the max data rate peak at 614xx earlier today.

We will see what happens over the next couple of weeks - Louise Grayson (ee ?!?) has kindly agreed to have the link monitored for 24 hours and be a contact point if it starts happening again.

Meanwhile - I'll canvas the folks in the village and see how things are in general - its amazing how common the issues happen to others but our ISPs do not know as we have at least 5 suppliers just in this little road and friends are on another. Before the pole was replaced we did consider trying a conference call with all the ISPs but thought they may be a little scared and just quote 'data protection' as they ran away.

Decentralisation of market (free market forces) are not always better than monopolies for customers.

Bring back benevolent dictators 😂 🤣 😂

[#11]

Thanks a lot for getting back to us @adbeasley and I'm glad to read that the engineer appeared to grasp the situation a little better. Let's see how the line performs over the next few days, and please let us know if you continue to experience any further issues at all.  

[#12]

Well looks like the support team do not give a monkeys unless it hits a KPI as despite a set of drops I'm now being quoted:

I have tested the line today and can see the fault now seems to be resolved I can see from our systems that your router is online and that your connection speeds are all performing within expectations. 1 or 2 drops are actually fine. When the connection has 3 or more drops within 24 hours, then that is a concern. 

---

and being passed details of the ombudsman.

I am convinced that the infrastructure monopoly do not care about rural communities and have no understanding of the issues I have with an unstable connection (my chronic anxiety leaves me on edge every time I use the internet  and I panic during every drop).

[#13]

Hiya @adbeasley, thanks so much for your time on the phone. I do really appreciate it!

I'm really sorry for the continued issues you've been having with your service and for the frustration this must be causing you! As we discussed I am going to take ownership of this for you going forward and I'll certainly do what I can to help get this sorted for you

I have provided an outline of our chat on the open ticket here: https://www.plus.net/wizard/?p=view_question&id=222863528 where you can get back to me at any point and I will follow up with you tomorrow. 

Have a lovely evening

[#14]

After a two plus hour visit by a senior engineer it seems to be a REIN/SHINE issue that impacts my line but the cause is currently out of my control.  The OR engineer was hunting through some data and the physical area but I've not heard anything back - I've not seen a drop (family do not know a line drop from a general application or site issue) but I've cut down on my use a great deal at the mo and am unsure when I turned the full logging on the router!

I drilled a hole through the wall to run a network cable and moved the router (this house is a real messy layout) and now using a shielded cable between the router and master socket and asked the help desk to check with OR as to their findings.

If I was on my own I would have just cancelled the whole internet to be honest - between these problems and my mental health somethings going to give!

Lauren has been a great support since picking up this issue.

[#15]

If it is indeed REIN or SHINE then as mentioned previously, routerstats for the BTHH5a work well with the Plusnet Hub One.  You note that you do not have Windows platforms, however that tool will run on Linux with black magic - see the KITZ link from the page link below.

If you have good relationships with others in the area, there would be merit in getting them to also run router stats logging.  If you all see similar issues and you all make the same complaints to your individual ISPs you might collectively get BT Openreach to get their digits out!