New line - connection unstable

[#1]

Hey all,

I've just recently had my line updated to Fibre (torture while we waited!)
Since the line has gone live we've had a problem with it being highly unstable.
Now I know to expect some instability due to new line configuration and such but my connection would drop when even the slightlest load was placed on it, 500mb download? Reset, low quality video stream from Youtube? Reset...
Now I've had a play with my router settings and increased the stability adjustment to 10 dB which seems to have helped but of course it's had an impact on my speed.

Is there any advice anyone can give for this and is it possible to check the status of the line as well? (really hoping OR haven't done a rush job)

[#2]

Changing the SNRM on a Fibre connection has no effect at all so you must have something else going on. Can you provide any connection stats from the router?

[#3]

Which service have you gone for? Unlimited Fibre or Unlimited Fibre Extra?

There is a known problem with the Unlimited Fibre, more details on this thread it might be worth posting there and asking @Bob Pullen to make the mod on your router.

[#4]

I've gone for the Unlimited Fibre Extra.

Sorry I probably should have mentioned earlier on that we're using an Asus AC68U as well (although the PN router also has issues being reported by other's in the house)

As for stats for my router (hoping this pastes in correctly)

[#5]

@deansmith02 i have the DSL AC68U this is what i have

[#6]

Hey Dean, welcome to the community forums firstly.

I am sorry to hear you're having problems with your connection. Things seemed pretty unstable last week:

Testing your line, your router is still in sync below your estimates:

Test Outcome	Pass
Test Outcome Code	GTC_FTTC_SERVICE_0003
Description	No problem found, OAM test is not currently supported on this line.
Main Fault Location	OK
Sync Status	In Sync
Downstream Speed	27.0 Mbps
Upstream Speed	11.7 Mbps
Appointment Required	N
Fault Report Advised	N
NTE Power Status	PowerOn
Voice Line Test Result	Pass
Bridge Tap	Not Detected
Radio Frequency Ingress	Not Detected
Repetitive Electrical Impulse Noise	Not Detected
Cross Talk	Not Detected
Estimated Line Length In Metres	649.5
Upstream Rate Assessment	Reasonable
Downstream Rate Assessment	Low
Interference Pattern	Regular Interference Observed Daily
Service Impact	Retrains Observed
Interference Duration Longest Occurrence	From 09:15 to 09:30
Interference Location	Customer Premise
Home Wiring Problem	Not Detected
Downstream Policing Discard Rate	0.0
Customer Traffic Level	Upstream and Downstream Traffic Detected
Technology	VDSL
Profile Name	0.128M-80M Downstream, Error Protection Off - 0.128M-20M Upstream, Error Protection Off

If you haven't already I've advise plugging your router into the test socket, as explained here: https://community.plus.net/t5/Library/Testing-From-The-Master-Socket/ba-p/1322242 and if you still experience problems then to raise a fault to us at http://faults.plus.net as it's likely to need an engineer to investigate this.

[#7]

Hey all, thanks for the replies.

Ok, so I've got the faceplate of the master socket off and router (and only the router) plugged in to the test socket.
My router stats seem to be a little more in line with a stable connection and the speed and load testing that I've done haven't caused a drop out as yet and so far it's been a stable connection for the last 40 minutes.

Speed is still a little slower than expected but I understand it can take a little while for this to settle on a new line anyway and the router seems to be picking up the correct line speed now (within reason) so I'd say this is a settling down issue with the speed.

To give you an idea, router plugged in to the test socket and no stability adjustment in place (I could get it stable with a value of 13db in this before).

So this leaves me wondering if there's a problem with the faceplate connector on my master socket as the connection seems fine in the test socket?
Any ideas here?

Once again thank you all for the help.

Edit:

My current DSL stats:

Update:
It's been a little over an hour since I reconnected the faceplate and i've been running speed tests and downloading to put it to the test under pressure.
So far I've had no drop outs and the speed is for the most part remaining constant with a few spikes in the right direction.
While I was reconnecting the faceplate I noticed that the socket itself was actually a little loose (seemed fine with the faceplate on?) so I tightened up the screws, could it be that this was the cause of the issues? I'm pretty sure it's cabling that connects to the back though which would rule this out wouldn't it? Any insights would be appreciated.

[#8]

Are there any extension sockets (which would normally be disconnected when you expose the test socket)?

Do you have a filtered faceplate?

Also, are you using any powerline network adapters?

[#9]

No extensions

The faceplate requires a filter to be plugged in to it

Yes I use powerline adaptors but the problems come no matter how I'm connected be it wired over powerline or direct to router (with no powerlines connected) or wireless.

That being said I've had no stability issues since re-connecting the faceplate, a solid 4 hours of uninterrupted connection so far but I'm about to put it through a painful stress test by streaming a game, so far all previous levels of load that would have killed the connection have managed to go through without a hitch so I'm hoping by some strange reason the faceplate being loose was the problem and that it's now sorted, one way to find out!

Any other suggestions on putting this to the test?

Update - 07-05-17

Line became unstable again last night around 9pm, left it to "settle" over night without any load, now my router is scheduled for an automatic restart cycle every Sunday at 2am so one of the down time cycles I expected, the rest that I'm getting are issues relating to either "ethernet disconnected / link down" or "connection reset by peer".

Going to get the report logged with the fault team today as my hopes of the loose faceplate being the issue turned out to be wrong. Dang.