What would cause TG582n to to lose WiFi on all devices at the same time

[#1]

Cheers Matt
Glad you picked this up, and again sorry your leaving
If I may
You said "If you plug a USB device into the port I very much doubt it'll pick up anything."
Answer:
What port there is no port on the NON USB Router ?
Therefore impossible to plug 1 in for at least 2 reasons
1 without cutting  hole in the side where the Controller Chip is
&
2 for violating T&C's warranties etc
Unless for clarification & to prevent miscommunication you was referring to applying the NON USB Firmware on the USB Router
Both Personally & Professionally I still feel that it is still incorrectly marked up especially whilst it clearly has the USB code turned on by default for an obvious NON USB Router
Whilst the USB Code is in the NON USB Firmware it is = Extra code that is not needed takes up space
Think of it as the difference of eating a yoghurt for breakfast as opposed to a full English since it creates unneeded bloat
Tidy markup practice = better quality coding and easier to spot any bugs glitches etc = Minimizing the chances for anything to go wrong

[#2]

Hi there, I'll do my best to answer but am not really an expert on this.

Quote from: TORPC

If I may You said "If you plug a USB device into the port I very much doubt it'll pick up anything." Answer: What port there is no port on the NON USB Router ?

You're correct when you say:

Quote from: TORPC

Unless for clarification & to prevent miscommunication you was referring to applying the NON USB Firmware on the USB Router

That is indeed what I meant - sorry if I got the wrong end of the stick there. If you install the non-USB firmware on a USB router and try to use the USB slot, I very much doubt anything will happen.

Quote from: TORPC

Both Personally & Professionally I still feel that it is still incorrectly marked up especially whilst it clearly has the USB code turned on by default for an obvious NON USB Router

As I said in my previous reply to you:

Quote from: Matt

1) That's because the controller chip still has the functionality enabled for the USB port, it's just that the hardware isn't installed. The power management bit still has the ability to control it, and I daresay it's still in the firmware because there was no real need to remove it.

That is to say, the controller chip still has the code embedded into it - in order to change that it'd mean getting a whole new batch of controller chips created with the code modifications afaik.

Quote from: TORPC

Whilst the USB Code is in the NON USB Firmware it is = Extra code that is not needed takes up space Think of it as the difference of eating a yoghurt for breakfast as opposed to a full English since it creates unneeded bloat Tidy markup practice = better quality coding and easier to spot any bugs glitches etc = Minimizing the chances for anything to go wrong

Agreed, I'll request that we mention this along with the copyright in the headers.

[#3]

Cheers Matt
I am glad I knew what you meant but others might not have twigged on
I didn't realise the code was embedded in the chip itself (sorry my bad  I should of known better)
Thank you, I wish you would still be around when they give the answer
As there are obvious issues that are clearly transparent with 2 NON USB Routers
Would I be able to order that replacement router as Adam previously discussed mentioned in the form of the USB flavour ?
as
It would test some theory's of mine IE: regarding USB turned on could be causing issues to the NON USB Router which I feel needs to be ruled in / out, plus I could also benefit from the USB port 

Quote from: _Adam_Walker_

That's not good, it's the first I've heard of that happening to be honest. Let me know if it keeps happening so I can sort you a replacement. Adam

[#4]

I don't think there's much software code embedded in the "chip". The USB controller chip is probably integrated into a single chip along with everything else (cpu, adsl modem, wifi too prossibly) - a "system on chip" or SoC. So the missing hardware on the non-USB variant probably comprises the physical USB socket and something to supply 5V DC power.
Most of the usb controller code would be part of the Linux kernel, which is part of the firmware (note lack of source code provided by Technicolor), so removing the usb related code might mean re-compiling the kernel or not including the usb related kernel modules, and not included various usb related userspace programs. Currently the non-usb software changes only involve disabling or hiding usb related telnet commands and web pages. In the 8.4 non-USB version some of the media sharing server programs weren't even turned off.

[#5]

After the firmware flash
No dropouts to report as yet and still early days
I have noticed that in the Embedded Help pages in the 10.2.2.B Firmware without USB PORT & just an observation

Quote

Firmware Embedded Help, R8.4.x, E-DOC-CTC-20100324-0012 v2.0

My theory's so far (nothing proved or disproved as yet)
Using OpenDNS servers (The TG582n struggles to cope when it tries to revert back to the default PlusNets DNS servers)
In am attempt to prove / disprove I flushed the DNS [code=Telnet file=Flush DNS server list]dns server forward dnsset flush[/code] prior to adding the OpenDNS servers
External interference (Very unlikely however still theoretically possible)
WiFi Channel set to No:-13 as it is the only non congested channel that is reported back by both inSSIDer for windows and Droids WiFi Anylizer (all other channels on various routers including TG582n have their issue(s) but never completely dropped WiFI to all devices) (Highly unlikely as on a few occasions has also knocked off the Ethernet connectivity)
Default to Annex A Previous ISP on all their routers routers including my own DGN2000 was Default to Annex M
[code=Telnet file=Physical Layer Statistics]Technicolor TG582n
Firmware Revision  10.2.2.B.EH
Copyright (c) 1999-2012, Technicolor
{admin}=>:xdsl info expand=enabled
Physical Layer Statistics:
  Modem state:                                  up
  Up time (Days hh:mm:ss):            0 days, 15:49:57
  xDSL Standard:                                ITU-T G.992.5
  xDSL Annex:                                    Annex A
  Channel Mode:                                  Interleaved
  Number of reset:                              0[/code]
Testing the DNS server Flush 1^st prior to Adding Annex M atm (again Unlikely but yet to be tested)
[code=Telnet file=Connection Modulation Mode All Modes Available]xdsl debug multimode config=t1.413issue2+g992.1_annex_a+g992.3_annex_a+g992.3_annex_l+g9
92.3_annex_m+g992.5_annex_a+g992.5_annex_m[/code]
Someone I spoke to yesterday thinks it could be an MTU issue (However I very much doubt it) Since the line in itself has proved to be very stable on 3db Interleaved default setting 1500 MTU)
Power Commands all set to Enabled, can not cope under stress (Yet to be Proved / Disproved)
[code=Telnet file=Power controls default => Enabled]Additional pwr config commands
Command                                             Comments
pwr config eco-manager=enabled / disabled The ECO manager
pwr config cpu-microsleep=enabled / disabled Allow the CPU to use low power instructions
pwr config cpu-lowspeed=enabled /disabled Allows the CPU to adjust it's clock speed.
pwr config usb-controller=enabled / disabled The USB controller --------------------------------> (Will not Disable)
pwr config wlan-pwrcontrol=enabled / disabled Wireless LAN power control
[/code]

[#6]

When the problem occurs, try checking the IP address etc. on a computer e.g. with the ipconfig / ifconfig commands.
I would have thought an obvious thing to try would be to reset the router to factory defaults (rather than changing more and more settings).
However, you might be able to change settings by saving the configuration, editing the user.ini file, then restoring the configuration from the edited user.ini file. There are also the wlan-pwr-options, perhaps disabling "single-rxchain" would stop people complaining about the router reporting 65Mbps rather than 130Mbps (65Mbps probably isn't the actual maximum speed, it's just the speed the router reports when the wifi is idle).

[#7]

Well you are now on the same firmware as me 10.2.2.b but mine has no EH after it. My router is also the non usb variant.  I also tried a factory reset that didn't cure it... its been stable now for a few days but I am sure that it will happen again shortly

[#8]

Quote from: ejs

(65Mbps probably isn't the actual maximum speed, it's just the speed the router reports when the wifi is idle).

I would agree with that
The reported wireless speed is also dependent on whatever is at the other end of the wireless connection.
If I use my win 8.1 laptop it always reports 65mbps
If I use my windows vista laptop it always reports either 117 or 130 mbps.
Just tried viewing the routers GUI from both laptops at the same time:
a) vista reported a wireless speed of 117mbps
b) win 8.1 reported 65mbps.
However, if I refresh the page on the vista machine followed quickly by refreshing the page in win 8.1 then that shows 117mbps.
After repeating this a few times I did get the win8.1 machine to show 130mbps but a second refresh brings it back to 65mbps.

[#9]

Quote from: ejs

When the problem occurs, try checking the IP address etc. on a computer e.g. with the ipconfig / ifconfig commands. I would have thought an obvious thing to try would be to reset the router to factory defaults (rather than changing more and more settings).

As clearly mentioned in posts #6 and #15

Quote from: TORPC

<snip> Will factory reset the router and edit report After a factory reset I can access the router GUI <snip>

Quote from: TORPC

It has happened again with a twist <snip> Quote from: ejs If it happens again you could check each device still has a correct IP address, subnet, default gateway address, and if not, see if setting those manually to some correct values on the computer changes anything. All devices checked as per your direction and their settings were and still all correct

Which also included doing an ipconfig on both the WiFI and Ethernet affected machines going down at the same time and
[code=Telnet file=Show Wireless Settings]wireless ifconfig[/code]
Note:
This has happened only on 2 yes 2 PlusNet TG582n's
I did factory reset 1 of them as reported in post #6 and the only changes I made at the time was to change DNS servers to OpenDNS (albeit without flushing PN's DNS servers ) Set devices to their own names,  to always on static IP addresses & change wireless channel to the non congested 13
in post#18

Quote from: TORPC

Router now changed over all is ok so far Now setup most of the devices Named & shamed them in the router and give them static always on addresses

Also would flashing the firmware not constitute as factory restored to that firmware ?
The only changes I made after flashing the routers firmware is to the DNS servers to OpenDNS (This time flushing PN's DNS servers)  and set all devices to their respective names & static always on IP addresses, & change wireless channel to the non congested 13

Quote from: ejs

<snip> would stop people complaining about the router reporting 65Mbps rather than 130Mbps (65Mbps probably isn't the actual maximum speed, it's just the speed the router reports when the wifi is idle).

Where have I ever Complained or was ever mentioned in any of my threads any reports of the reporting of the WiFi actual speed discrepancy  report across devices & please explain to aid better understanding, why you think it has to do with the issue(s) I have reported for the TG582n router(s)
Please note:
This is an observation & I am also  guilty of it myself
Some people either try to memorise the content of thread rather to fully recap and attempts to reply accordingly (thus making, which can be classed as noob mistakes) and results in inaccurate responses.
Speed reading is another downfall I have also fell foul of and I am not the only human being to have done so !!!
I am no way pointing the finger at any individual as I have stated I have also fell foul myself 

[#10]

It has gone again
This time back to the original fix of turning OFF / ON the WiFi Button on the router
Conclusion the TG582n Without USB is not the best

Quote

The only changes I made after flashing the routers firmware is to the DNS servers to OpenDNS (This time flushing PN's DNS servers)  and set all devices to their respective names & static always on IP addresses, & change wireless channel to the non congested 13

Is anyone from PlusNet going to recognise this at all ?
For testing purposes
Does anyone know the correct working Telnet command to turn off the USB on the NON USB firmware 10.2.2.B EH ?
Also would getting a 5Ghz router improve the situation of channels 1 through 12 being congested on 2.4Ghz ?

[#11]

Quote from: TORPC

Where have I ever Complained or was ever mentioned in any of my threads any reports of the reporting of the WiFi actual speed discrepancy  report across devices & please explain to aid better understanding, why you think it has to do with the issue(s) I have reported for the TG582n router(s)

I didn't say you did.

Quote from: TORPC

Does anyone know the correct working Telnet command to turn off the USB on the NON USB firmware 10.2.2.B EH ?

I would have thought that the "pwr config usb-controller" setting would enable or disable power management for the USB controller. With power management enabled, the (not connected to any physical usb sockets) USB controller should get switched to a low-power state. You might be able to bypass the disabled telnet commands by saving, editing, then restoring the user.ini config file. If you could access a busybox shell on the router it would be possibly to unload the usb driver Linux kernel modules, but no such access is possible.
Channel 13 half overlaps with channels 11 and 12, and inSSIDer doesn't show how heavily each network is used, nor non-wifi interference, so can be worse than useless for picking the best channel. It's not necessarily a problem to be on the same channel as another network - even some PDFs from the makers of inSSIDer say it's better to be on the same channel rather than half overlap. You could try using RTS/CTS protection always if the problem were due to wifi congestion, but I doubt that it is, technicolor routers have horrifically overcomplicated firmware and I prefer not to use mine.

[#12]

Again
In Post #28
[code=Telnet file=option not allowed]{admin}=>pwr config usb-controller=disabled
Option not allowed => usb-controller
Unknown command.[/code]

Quote from: ejs

You could try using RTS/CTS protection always if the problem were due to wifi congestion, but I doubt that it is, technicolor routers have horrifically overcomplicated firmware and I prefer not to use mine.

Would you explain why you do not think it is WiFi congestion & what you think it is ?
Would you explain more about the RTS/CTS protection & why you recommend trying it ?

[#13]

Ive given up.... I dont feel plusnet are taking the issue seriously.  I realise it's only 2 people who have noted this behaviour but I am sure there must be more. My setup is nothing out of the ordinary.  I have now bought a billion bipc 7300n and will see how that goes.  Happy to send this router back to plusnet if they want to do some testing themselves or let Thompson have a look.

[#14]

I assume you're telling me that editing this section of the user.ini file then restoring the config didn't work:

Quote

[ pwr.ini ] wlan-pwr-options t-on=5 t-off=10 t-ext=2 slow-radio-toggle=disabled fast-radio-toggle=disabled single-rxchain=enabled ethphy-pwr-options t-on="" t-off="" config eco-manager=enabled cpu-microsleep=enabled cpu-lowspeed=enabled usb-controller=enabled wlan-pwrcontrol=enabled eth-phy=disabled debug traceconfig state=disabled

I which case I don't know how to disable power management of the usb controller.
Wifi congestion wouldn't be fixed by switching the wireless on and off or rebooting the router. Instead perhaps something local to your network or environment causes some part of the router to crash. There are numerous commands to extract various diagnostic data from the router, although most of them will return a lot of data that is impossible to interpret. The Linux kernel messages can be viewed with debug dmesg, and a load of mostly meaningless wireless data can be view with debug exec cmd="wld alldata" .The 582n has a history of wireless problems (mostly people complaining about poor speeds), the previous model, the 585v8, also had a history of wireless problems.
RTS/CTS mode might improve performance in congested environments, but alternatively could just add extra overheads and reduce performance. It goes like this:
The device that wants to transmit something first sends an RTS packet and waits to receive a CTS packet. The receiving device first receives the RTS packet, then sends out a CTS packet. Anything else that receives the CTS packet should shut up and not interrupt the subsequent data transmission. So if equipment was physically arranged like this:
another router                      your computer                    your router
In that arrangement, the other router could be too far away to detect signals from your router. But it might output a stronger signal, so your router might detect signals from the other router. So the other router might keep transmitting over your router's transmissions, and have no idea it was doing it, whereas your router would check to see if the channel was clear, and would wait if it detects the other router transmitting. But if the other router receives a CTS packet from your computer, it should not interrupt the subsequent data transmission. I think that would only be applicable if the two routers were on the same channel, and not if the channels only partly overlap. And the 582n might enable RTS/CTS protection anyway depending on what 11g or older devices it detects, I've observed it doing RTS/CTS with default wireless settings (which list ctstoself for protection mode) - example: http://ejs1920.users.sourceforge.net/582n-wifi-packets.pcap
Wireshark running on Linux can capture the wireless level (802.11) packets in "monitor mode". I believe Microsoft Network Monitor can do the same on Windows (Vista or later, not XP), but haven't tried it.

[#15]

Its all to easy to give in & since PlusNet does not do the USB router on ADSL anymore
The average of 700,000+ PlusNets client base would not have done all the testing like we have done, or know what to look for, or for some they just persevere with the issue(s) if they are intermittent 
I am thinking of trying some routers from my previous ISP whilst I wait for all the responses about dual band / 5Ghz routers in my other thread 
(Those in the know should know who was my previous ISP by some of the router model(s) )
Also taking into consideration that in the no to distant future I will be building a NAS server using freeNAS
There will be ipad & iphone, Android, linux and Windows Operating systems in use, the ipad & iphone not so much as that's when their brought on a visit by pests oops I mean guests
Suggestions which would most likely see the best overall sync rate, performance, and why
They are as follows and in no particular order and they are all updated to the latest possible firmware with the exception of the BT Fusion
D-Link
2680
2780
3680
Huawei
HG532
HG533
Thompson
TG585v7 Flashed to the latest Plusnet firmware possible
Very old
BT Fusion
@ejs
No
As I have not attempted to edit the user.ini file as yet
I was referring to the Telnet command not working  with (Unknown Command)
Whats the best way to extract the user.ini

Can not recall seeing anything relating to RTS/CTS settings, is it via Telnet or GUI