Intermittent WAN disconnects.

[#1]

@Thatwilllldo12 

This is from your first post

Assuming this is the same data capture, reading from today's post

The concern here is the Line Attenuation figure

The second post suggests your'e further from the exchange which makes no sense, this would explain the drop of speed, possible caused by the high restant fault.

After opening up my spare faceplate the data feed output does include a REIN (Repetitive Electrical Impulse Noise) filter and possible improved filtering of the broadband signal from the attached phone.

HTH

[#2]

Good spot on the attenuation figure, @Dan_the_Van . Interestingly enough, I observed an impossible change in my copper line length recently - that was down to, to quote the OR engineer that fixed it - 'cowboys' - he didn't say who they were, but I have a good idea who he meant. They'd disturbed a number of connections at the top of the first pole from the u/g DP.

[#3]

@Dan_the_Van  Interesting point on the line attenuation, good spot. Here's another shot of what it currently sits at after the engineers visit this evening.

As to the filtered extension - Everyday is a school day and I'm certainly learning new things thanks to you all.

Small update.  Several drops today with one large one before I went out. Skimming over the logs here it looks like that's been the case all afternoon.  Anyway, just after I got home there was a knock at the door. To my surprise an OR engineer stood, apologising for not notifying of his visit beforehand and that I may have experienced some outages as he was testing the line. He ran some tests, found a High resistance point within 10 meters of the socket, threw a new master socket base plate on and speeds immediately returned to what they were in November last year: 

BTW I need to give a massive thank you to @adamwalker for organising this so speedily. 

So now the question remains what to do from here. This engineer seems pretty confident this will be the end of the problems ( Fingers crossed this is the case) and I can put all my equipment back to how it was, utilising the data extension should I wish to. He saw no issues with the set up. However,  the two previous engineers never found any faults, never found any HR fault (certainly none were thrown up on the system) and you would think a simple fault on the master socket would be the first thing they would have look at?  So I have my doubts.

I'm tempted to leave the Modem/router at the Master socket for another week to see if the dropouts have been fixed.

Thank you all for the input.

[#4]

Interesting result, @Thatwilllldo12 ! You (possibly) got third time lucky, and an engineer who knew his job, this time.

I would be be tempted to leave it as is, as you suggest, for a while, and then try going back to your usual set-up.

[#5]

---

@jab1 wrote:

Interesting result, @Thatwilllldo12 ! You (possibly) got third time lucky, and an engineer who knew his job, this time.

I would be be tempted to leave it as is, as you suggest, for a while, and then try going back to your usual set-up.

---

Maybe. Had a couple of drops since his last night, which may just be outliers, but it also kinda of follows the pattern after every engineer visit.  I think I'll leave the set up as it is and keep monitoring for a few days. 

[#6]

@Thatwilllldo12 just to let you know I've not forgotten about you, I'm still waiting for the fault to drop into the right status to allow me to book an appointment. I'll chase it up with Openreach if that's still an issue later this afternoon. 

[#7]

@adamwalker   Cheers bud.

[#8]

Engineer number 4 has just left the building. Another unannounced, unexpected OR visit. Another new master socket/faceplate,- that's two replacements in less than 24 hours- Another disconnection of the "bridge tap" extension that wasn't being used.  Latest diagnosis....It's the router/modem. Recommendation: Go out and buy a cheap modem (my current one is less than 6 month old) to eliminate my current one. Thesis being: I don't "think this router likes the network"- even though it ran perfectly fine up until November last year and the drops mainly happen at night!

On top of this, I've just received an email of a price increase for my broadband which hasn't worked properly since November,

It's getting a bit silly the now.  According to this Engineer there are 20 of them covering the South East- I've now been visited by just under a quarter of them!

[#9]

@Thatwilllldo12 Another engineer with little knowledge - 'it's the router' is one of their favourite 'get-out' clauses, and what we have seen is clear evidence of OR-side network issues.

[#10]

@jab1  I must admit I'm starting to get the faint whiff that this has become a parts slinging exercise - Swap parts until the problems stops.  At this point I might just get a key cut for Open-reach to let themselves in. 

Crazy thing is we've now reached a point where work is being carried out at short intervals without giving enough time to see if the results are stable.

Four engineers since the 4th of January: three just this week. Four  engineers that can't find a reason for the drops and four different suggestions.     

[#11]

@Thatwilllldo12 What would be interesting to see is the RADIUS log for your line, but PN seem to be reluctant to post those these days.

[#12]

A quick update: longest drop free uptime in quite a while. 

 

[#13]

Awesome! That's really good to see. All being well it should stay that way but I'll check back in with you on Monday. 

If you need to message in over the weekend for anything one of my team mates should be able to pick it up. 

[#14]

Hi Adam,

Funnily enough I just messaged you on the CS portal-ma-bob.  I don't want to jinx it and call it fixed just yet, but is certainly looks promising.  Fingers crossed.

[#15]

As of this evening:

 

So far, so good. Beg's the question; What possibly fixed it? Eliminated Bridge tap as the problem (problem remained whether connected or disconnected). Modem is currently the one that's been used all along. Two new master sockets fitted, that I'm aware of < surely this wasn't the problem? If so why did it take four engineers and nearly three months to sort it.  I have a sneaky suspicion that the problem lay on the network somewhere and someone finally spotted it. Be nice to know what it was.