TG582n dropping ASDL connection

[#1]

OK, don't like the look of the SNRM graphs, those regular humps indicates an issue of some sort.
I suspected looking at the graphs the sample rate was long - that confirmed by the log file - samples are 30 seconds. Can you please change that to 10 seconds 720 points per page (please see my guidance in reply #68), also right click the SNRM graph a select/tick Plot Upstream Noise.
Won't normally want to see the RS log files, the graphs usually tell all.
(As an aside what exactly did you mean by a "pattern box", also what brand is ADSL2+ plate and what was fitted when you plotted the SNRM graphs?).
This may be a bit of a nuisance but we need to try and eliminate anything local from the issue. (Don't forget to do graceful disconnects etc).
Can you plug a standard rat's tail filter into the test socket and connect the 582n to that.
Can you turn off and unplug your DECT phone (base station and handsets) and plug in a corded phone instead for the test.
Con you confirm if you have any extension phone sockets in the house at all, and if so that they are all dead when you are connected to the test socket.
Can you turn off your Sky boxes and unplug them from any phone or router connection.
Can you turn off your Belkin AP and unplug from the mains, and any other wifi equipment.
Ideally connect your computer to the 582n via ethernet and turn off the 582n wireless leaving only your computer connected to the 582n, anything else disconnected and switched off.
Can you ensure that any remote Smart Meter Monitoring devices are as far away as possible from the phone line, the 582n and your computer, ideally switch them off.
Ensure that any mobile phones are nowhere near the phone line or 582n, ideally switch them off if not needed during the testing.
Having changed the RSL settings as above run RSL for 2 hours daytime ideally asap, but well before sunset.
After the test post the SNRM graph.
Edit: sorry forget to remind you to do periodic Quiet Line Tests 17070 option 2. There should be no pops or crackles or hum, if anything just a barely perceptible hiss/mush sound.

[#2]

Sorry chart settings had changed; I installed the full routerstats program and those settings I missed: it's back as you requested now.
The ADSL filter (pattern) was the one shown in:    plug-in-faceplate-adsltelephone-filter.jpg (attached)
The one I used for the stats/charts was the ADSL 2+ shown in:  pressac adsl-nteface-1.png (attached)
I should be able to run the test during the late morning tomorrow.

[#3]

As you've probably discovered, the sample frequency for the full RS and TG582n is on the Bitloading/Options tab.

[#4]

2 hr test done
rx and tx noise charts attached
Ping chart attached
Router Stats attached

[#5]

After dark update.
Still connected via test socket
Reconnected everything individually: only the DECT station had an effect on the stats: rx noise margin dropped and ping went up from c22 to c26.
I rearranged the equipment to optimise performance and it was reasonably stable during daylight hours.
Just after 18:06 the router recycled and since then CRCs have climbed dramatically and ping is showing a large number of dropped packets.
I have attached some charts for illustration and will continue monitoring the router.
the DECT phone on quiet test, has a quiet, pulsing background buzz; it is the same on all 3 DECT handsets. I tried a quiet test on my elderly analogue handset; the line was quieter although their was evidence of some noisy crystals in the earpiece.

[#6]

Quote from: andyw

.......... their was evidence of some noisy crystals in the earpiece.

If it's that old a one, probably carbon granules in the microphone!
Anyway this doesn't look good. Your SNRM graphs from earlier today certainly suggest maybe an external issue.

Quote from: andyw

Just after 18:06 the router recycled .....

Can you elaborate on that please. Do you mean you rebooted it, or it rebooted itself?
Can you look at the 582n Event Log for the entries just before and after 1805.  Look at the Telnet log if the GUI log doesn't go back that far.
Looking at those recent stats, there's the possibility that DLM has raised the Target SNRM.
It also looks as though towards the end of the latest plots, RS has stalled, which it can do periodically. If you spot it do that just click Pause/Stop and then Start again.
Tomorrow in daylight hours, I would do a Graceful Disconnect and refit whichever Filtered Face plate appeared to give the best results and plug your Rat's-tail Filter into the phone socket. Connect the 582n to the Faceplate socket and plug your DECT into the Rat's-tail so the DECT ends up being double filtered.
My suspicion is that you have an Earth fault on the Line (or possibily a Battery Fault) we'll have to ask CRT to do a CLT tomorrow.
However there is one other possibility we should explore and that's make sure it's not the 582n or it's PSU itself. Do you happen to have a spare modem/router or can you borrow one?
Edit: Sorry, forgot to ask, as you are running the full RS, can you post a copy of the Bits/Tone graph set at 0-512.

[#7]

Thanks for getting back to me, I really appreciate it.
The analogue phone is 15 years old (I keep it for power outages) it may not have granules but it sounds like that. I tried another quiet line test with it and there was no sound at all so I have been running with it for the last half an hour or so; it doesn't seem to have improved very much. I will attach the latest rx noise margin and ping charts.
The router recycled itself; I've checked the event log, it's confusing as the system clock started at 00:00 after the recycle and just after it corrected itself the router recycled again: the event log is appended below.
I do have another 582n which I couldn't set up from new and it was replaced; I also have a Belkin N1 that I may be able to resurrect - I'll have a look

Event Logging
This page summarizes the last events that have been recorded on your Technicolor Gateway. Choose a display filter...

Category:
Recorded Events

Time Message

Info Mar 23 22:55:47 LOGIN User admin logged in on [HTTP] (from 192.168.1.69)

Info Mar 23 22:42:43 LOGIN User rstats logged in on [HTTP] (from 192.168.1.69)

Info Mar 23 22:13:52 SNTP Synchronised to server: 212.159.13.50

Info Mar 23 21:13:52 SNTP Synchronised again to server: 212.159.6.10

Info Mar 23 20:15:22 LOGIN User rstats logged in on [HTTP] (from 192.168.1.69)

Info Mar 23 20:13:47 SNTP Synchronised to server: 212.159.6.10

Info Mar 23 20:04:16 LOGIN User rstats logged in on TELNET (192.168.1.69)

Info Mar 23 19:58:58 LOGOUT User rstats logged out on TELNET (192.168.1.69)

Info Mar 23 19:55:37 LOGIN User rstats logged in on [HTTP] (from 192.168.1.69)

Info Mar 23 19:39:12 LOGIN User rstats logged in on [HTTP] (from 192.168.1.69)

Info Mar 23 19:38:54 LOGIN User rstats logged in on TELNET (192.168.1.69)

Info Mar 23 19:38:30 LOGOUT User rstats logged out on TELNET (192.168.1.69)

Info Mar 23 19:35:14 LOGIN User admin logged in on [HTTP] (from 192.168.1.69)

Info Mar 23 19:30:40 LOGIN User rstats logged in on [HTTP] (from 192.168.1.69)

Info Mar 23 19:13:46 SNTP Synchronised to server: 212.159.13.50

Info Mar 23 19:12:54 LOGIN User rstats logged in on [HTTP] (from 192.168.1.69)

Warning Mar 23 18:15:19 PPP link up (Internet) [80.229.143.132]

Info Mar 23 18:15:19 FIREWALL event (1 of 2): deleted rules

Info Mar 23 18:15:19 PPP CHAP Receive success (Internet)

Info Mar 23 18:15:18 PPP CHAP Receive challenge from rhost PCL-AG03 (Internet)

Info Mar 23 18:15:18 PPP CHAP Receive challenge from rhost bras-red11.pe (Internet)

Info Mar 23 18:15:05 FIREWALL event (1 of 1): modified rules

Info Mar 23 18:15:05 FIREWALL event (1 of 1): created rules

Warning Mar 23 18:15:04 PPP link down (Internet) [80.229.143.132]

Info Mar 23 18:13:46 SNTP Synchronised to server: 212.159.6.10

Error Mar 23 18:13:46 SNTP Systemtime update: time setting 00:05:57 > new time setting: 18:13:46

Info Jan 1 00:05:30 LOGIN User rstats logged in on [HTTP] (from 192.168.1.69)

Info Jan 1 00:03:38 FIREWALL event (1 of 32): modified rules

Info Jan 1 00:03:38 FIREWALL event (1 of 50): created rules

Warning Jan 1 00:03:38 PPP link up (Internet) [80.229.143.132]

Info Jan 1 00:03:38 FIREWALL event (1 of 20): deleted rules

Info Jan 1 00:03:37 PPP CHAP Receive success (Internet)

Info Jan 1 00:03:37 PPP CHAP Receive challenge from rhost PCL-AG02 (Internet)

Info Jan 1 00:03:37 PPP CHAP Receive challenge from rhost bras-red11.pe (Internet)

Info Jan 1 00:03:29 LOGIN User admin logged in on [HTTP] (from 192.168.1.69)

Info Jan 1 00:03:15 LOGIN User rstats logged in on TELNET (192.168.1.69)

Error Jan 1 00:02:58 IDS dos parser : land attack (1 of 1) : 192.168.1.68 192.168.1.68 78 128 UDP 137->137

Info Jan 1 00:01:50 xDSL linestate up (ITU-T G.992.5; downstream: 7214 kbit/s, upstream: 1192 kbit/s; output Power Down: 20.4 dBm, Up: 12.7 dBm; line Attenuation Down: 41.5 dB, Up: 22.8 dB; snr Margin Down: 9.1 dB, Up: 6.3 dB)

Info Jan 1 00:01:24 xDSL linestate down

[#8]

Try the Telnet log, it's the bits before     Info    Jan 1 00:01:24   xDSL linestate down   that I really could do with seeing.
Edit: meant to remind you Telnet in, the command is      syslog msgbuf show hist=enabled
That Bit loading is attrocious, either there's a lot of interference, or a line fault or the 582n.
I tell you what I would do tomorrow, hopefully when the DS SNRM will be 9dB or better, because I suspect the Target SNRM might be at 9dB, otherwise you'll end up with a slower sycn speed.
Graceful Disconnect. Refit the best Faceplate.and plug the rat's-rail into the phone socket ready. Get your other 582n and it's PSU (whatever you do don't get them or the PSUs mixed up), I would log into it off-line and enter your username and password manually. Then connect it to the Faceplate without any phone for now.
As soon as you have sync, grab a copy of the stats ready to post later. Set RS running.You should be able to tell after about an hour if you still have these repetitive humps on the DS. Once you know what it's doing you could connect your DECT to the Rat's-tail.
Then later post up the stats from just after sync, current stats and some SNRM graphs and a Bits/Tone.
The advantage of using another 582n, you already have RS set up, forgot to say if you created an additional user for RS on your old pone don't forget to do that on the new one. Diagnostics/stats on the 582n are better than most.

[#9]

Heres the syslog as far back as it goes
------------------------------------------------------------------------
{admin}=>:env set var=SESSIONTIMEOUT value=0
{admin}=>syslog msgbuf show hist=enabled
<85> Jan  1 00:00:08 FIREWALL has been enabled
<4> Jan  1 00:00:16 KERNEL Warm restart
<85> Jan  1 00:00:25 FIREWALL event (1 of 1): created rules
<85> Jan  1 00:00:25 FIREWALL event (1 of 1): modified rules
<82> Jan  1 00:00:32 UPnP framework disabled
<85> Jan  1 00:00:32 FIREWALL has been enabled
<82> Jan  1 00:00:34 UPnP framework disabled
<82> Jan  1 00:00:36 UPnP framework disabled
<82> Jan  1 00:00:42 UPnP framework enabled
<85> Jan  1 00:00:46 FIREWALL level changed to Standard.
<85> Jan  1 00:00:46 FIREWALL event (1 of 1): deleted rules
<148> Jan  1 00:00:46 DHCS server up
<173> Jan  1 00:00:54 xDSL linestate down
<3> Jan  1 00:01:02 [ANTI_SPOOFD] (err) llist_getobject_at for ENV.ANTI_SPOOFD's
value failed
<173> Jan  1 00:01:24 xDSL linestate down
<173> Jan  1 00:01:50 xDSL linestate up (ITU-T G.992.5; downstream: 7214 kbit/s,
upstream: 1192 kbit/s; output Power Down: 20.4 dBm, Up: 12.7 dBm; line Attenuat
ion Down: 41.5 dB, Up: 22.8 dB; snr Margin Down: 9.1 dB, Up: 6.3 dB)
<80> Jan  1 00:02:58 IDS dos parser : land attack (1 of 1) : 192.168.1.68    192
.168.1.68    78    128 UDP 137->137
<37> Jan  1 00:03:15 LOGIN User rstats logged in on TELNET (192.168.1.69)
<37> Jan  1 00:03:29 LOGIN User admin logged in on [HTTP] (from 192.168.1.69)
<38> Jan  1 00:03:37 PPP CHAP Receive challenge from rhost bras-red11.pe (Intern
et)
<38> Jan  1 00:03:37 PPP CHAP Receive challenge from rhost PCL-AG02 (Internet)
<38> Jan  1 00:03:37 PPP CHAP Receive success (Internet)
<85> Jan  1 00:03:38 FIREWALL event (1 of 20): deleted rules
<132> Jan  1 00:03:38 PPP link up (Internet) [80.229.143.132]
<85> Jan  1 00:03:38 FIREWALL event (1 of 50): created rules
<85> Jan  1 00:03:38 FIREWALL event (1 of 32): modified rules
<37> Jan  1 00:05:30 LOGIN User rstats logged in on [HTTP] (from 192.168.1.69)
<99> Mar 23 18:13:46 SNTP Systemtime update: time setting 00:05:57 > new time se
tting: 18:13:46
<102> Mar 23 18:13:46 SNTP Synchronised to server: 212.159.6.10
<132> Mar 23 18:15:04 PPP link down (Internet) [80.229.143.132]
<85> Mar 23 18:15:05 FIREWALL event (1 of 1): created rules
<85> Mar 23 18:15:05 FIREWALL event (1 of 1): modified rules
<38> Mar 23 18:15:18 PPP CHAP Receive challenge from rhost bras-red11.pe (Intern
et)
<38> Mar 23 18:15:18 PPP CHAP Receive challenge from rhost PCL-AG03 (Internet)
<38> Mar 23 18:15:19 PPP CHAP Receive success (Internet)
<85> Mar 23 18:15:19 FIREWALL event (1 of 2): deleted rules
<132> Mar 23 18:15:19 PPP link up (Internet) [80.229.143.132]
<37> Mar 23 19:12:54 LOGIN User rstats logged in on [HTTP] (from 192.168.1.69)
<102> Mar 23 19:13:46 SNTP Synchronised to server: 212.159.13.50
<37> Mar 23 19:30:40 LOGIN User rstats logged in on [HTTP] (from 192.168.1.69)
<37> Mar 23 19:35:14 LOGIN User admin logged in on [HTTP] (from 192.168.1.69)
<37> Mar 23 19:38:30 LOGOUT User rstats logged out on TELNET (192.168.1.69)
<37> Mar 23 19:38:54 LOGIN User rstats logged in on TELNET (192.168.1.69)
<37> Mar 23 19:39:12 LOGIN User rstats logged in on [HTTP] (from 192.168.1.69)
<37> Mar 23 19:55:37 LOGIN User rstats logged in on [HTTP] (from 192.168.1.69)
<37> Mar 23 19:58:58 LOGOUT User rstats logged out on TELNET (192.168.1.69)
<37> Mar 23 20:04:16 LOGIN User rstats logged in on TELNET (192.168.1.69)
<102> Mar 23 20:13:47 SNTP Synchronised to server: 212.159.6.10
<37> Mar 23 20:15:22 LOGIN User rstats logged in on [HTTP] (from 192.168.1.69)
<102> Mar 23 21:13:52 SNTP Synchronised again to server: 212.159.6.10
<102> Mar 23 22:13:52 SNTP Synchronised to server: 212.159.13.50
<37> Mar 23 22:42:43 LOGIN User rstats logged in on [HTTP] (from 192.168.1.69)
<37> Mar 23 22:55:47 LOGIN User admin logged in on [HTTP] (from 192.168.1.69)
<102> Mar 23 23:13:52 SNTP Synchronised to server: 212.159.6.10
<37> Mar 23 23:19:53 LOGIN User rstats logged in on [HTTP] (from 192.168.1.69)
,          deleted line for security reasons
<37> Mar 23 23:51:30 LOGIN User admin logged in on TELNET (192.168.1.69)
<37> Mar 23 23:56:24 LOGOUT User admin logged out on TELNET (192.168.1.69)
<37> Mar 24 00:07:41 LOGIN User admin logged in on TELNET (192.168.1.69)
{admin}=>syslog msgbuf send dest=192.168.1.254
{admin}=>

[#10]

Hmm, that's the line I was looking for, something caused it to reboot and it wasn't a power dip otherwise it would have said KERNEL Cold Restart.
<4> Jan  1 00:00:16 KERNEL Warm restart
<173> Jan  1 00:01:50 xDSL linestate up (ITU-T G.992.5; downstream: 7214 kbit/s,
upstream: 1192 kbit/s; output Power Down: 20.4 dBm, Up: 12.7 dBm; line Attenuat
ion Down: 41.5 dB, Up: 22.8 dB; snr Margin Down: 9.1 dB, Up: 6.3 dB)
That does clearly imply the Target SNRM has gone to 9dB unfortunately. No matter, when this is all sorted it can be reset.

[#11]

Glad you found what you wanted
I'm off-line in a few minutes to give the other 582n a try

[#12]

second 582n set up and synced BUT no WiFi; I remember now that that was why I had it replaced.
Nevertheless I have let it run for the hour and recorded the stats you asked for
I have attached the charts/stats for the initial sync to this message.
for clarity I will attach stats/charts after one hour to a second message
As I require WiFi I am going to re-instate my original 582n but connect it using the PSU from the second 582n
I will also collect sync stats/charts and follow-ups should you need to see them.

[#13]

Here are the stats/charts between 1 and 2 hours after 582n changeover
I have no WiFi as I said before  and I have lost Skype and incoming mail from Plus Net
next step is to revert to original 582n (to get WiFi back) but run it on the second PSU

[#14]

Well the repetitive hump is still there and the Tx SNRM is also very spikey up and down, all suggesting some sort of line issue, also confirmed by the sharp fall off on the Bits/Tone. As I mentioned last night, we'll see if someone from CRT could do a CLT for you.

[#15]

That sounds good.
The sooner the better.