These are errors on the line, not router errors.
There are a really high number of errors showing on the line at the moment. Please let us know how you get on following the engineer visit.
Ok, thank you for your reply.
If there are errors on the line, why do these not show as errors on the router stats, the router is on the internet via the line ?
Are you saying that the router only records errors when they are on the back haul network i.e. to a website, and not on your actual line ?
Surely the router is connected to the DSLAM and it is this section that it is monitoring or is it something else please ?
I would like to understand this point.
Good to know you are seeing the errors, I have to type this lots of times to get a message through. I hope they stay that way until the BT Openreach comes tomorrow, and that he does not 'knock off' the fault when he is testing , like before !
The line history is certainly showing very high errored seconds every hour from around 3:30am today. To be honest I don't know why the errors aren't shown in the router logs, could you post a screen shot of the logs please (please make sure that you to post any personal information)
Could it be you are seeing the physical layer and the HG633 router is showing the Network layer to me.
You would think that line errors would also cause network errors, but maybe the Network Layer corrects them or just asks for a repeat, so the only result is that it is very slow data rate ?
Or something else ?
DLM changed your profile in the early hours of this morning which has resulted in the error count dropping to zero since then
Our line history show errored seconds per hour, these are seconds containing at least one CRC error, they can affect speed performance as they require retransmission of data
I'll try to find out a bit more information about the errors in the router logs.
Many thanks for your reply.
It seems to me that the DLM is reducing the SNR ( by inserting extra line loss) to force an error free line to give the customer some kind of service, but at a lower speed. However when it does this on my line it reduces the speed to such a low amount ( I normally get only 2.5Mbps DOWNlink nowadays, but 5 years ago it was 3.2Mbps ) at around 1.5Mbps that it causes problems in using the service, even for e-mails.
So much stuff is running in the background nowadays (marketing tools for talktalk and google etc) that it takes forever to do anything.
i would appreciate more info on the router logs as I cannot see how I can see the line errors at all.
I've asked our product team why there's a difference between the errors showing in the router logs and the errors we can see on our systems. I'll let you know when I have any more information
Is your router still connected to the test socket at the moment?
Thank you for your reply.
My router is connected to my test socket at present, confirmed.
Are there line errors now Chris please ?
System Up Time 00:25:03
Above are my router stats (netgear router) from 2013 proof of much better speed from TT. Note the line loss is the same as today, so why is the speed so poor ?
Can you confirm that VDSL2 does not use the same band as ADSL ?
Perhaps this could be crosstalk from VDSL2 systems being installed in the last 5 years ?
Looks as though DLM was optimised earlier today. Sync speed is now 2.5Mbps. There was one hour earlier this morning when the errors spiked again then couple of hours following the profile change that showed no connection data but multiple reconnections. The latest information that we have is from a couple of hours ago,
All I can say regarding potential cross talk is the line tests aren't picking up any issues
here is the latest router stats
DSL synchronization status:
Upstream line rate (kbit/s):
Downstream line rate (kbit/s):
Maximum upstream rate (kbit/s):
Maximum downstream rate (kbit/s):
Upstream noise safety coefficient (dB):
Downstream noise safety coefficient (dB):
Upstream interleave depth:
Downstream interleave depth:
Upstream line attenuation (dB):
Downstream line attenuation (dB):
Upstream output power (dBm):
Downstream output power (dBm):
The line went down again at 11:00, the BT eng had gone by then.
What do you think caused this please ?
And it is connected at 2.2Mbps, not 2.5Mbps now.
Also notice how the line loss has now increased to 44dB in the UPlink direction ??
What is causing that please ?
Are you sure that a line test will actually measure crosstalk ?
I don't think it will not show up on DC tests Chris.
It would only show on noise if the tester actually measured in the ADSL Downlink band.
Is it possible to check if the VDSL2's being installed have notching below 2.2MHz t.e. they will not transmit in the ADSL bands.
I ran a line test which completed at 11:09, this will have caused a disconnection
Line history shows that the upstream attenuation has been 42db then 44db since Monday night. It did appear to drop to 37db earlier this morning but has gone back to 44db - I don't know why it has varied.
"Is it possible to check if the VDSL2's being installed have notching below 2.2MHz t.e. they will not transmit in the ADSL bands."
Sorry, I have no way of checking this.
Thank you for your reply.
Is there any more line testing going on today or tomorrow please ?
What do you think might have caused the large drop in speed since 2013 then ?
Does the DLM change the line loss (UP or DOWN) as part of the DLM optimisation program i.e. to force the SNR down and thus to lower the speed for a more stable service ?
Sorry, I don't know about the line testing, it's possible that our network team may run further tests following the engineer visit.
"What do you think might have caused the large drop in speed since 2013 then ?"
We were on a different DLM system back in 2013 which may account for some difference, but lines deteriorate over time which can cause an increase in noise which can affect speed
DLM changes the target SNR to adjust speed (it can't change line attenuation)
DLM changed your profile, this will be the cause of the drop in connection and drop in speed.
The error count has been zero or low single figures every hour since around 7am yesterday, I think DLM has dropped the speed because there was a number of resyncs yesterday morning. The connection appears to have been OK since so DLM should move you to a faster profile if the connection remains stable (and error count remains low)