Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

DSL actually working? #6

Open
majuss opened this issue Jun 11, 2018 · 83 comments
Open

DSL actually working? #6

majuss opened this issue Jun 11, 2018 · 83 comments

Comments

@majuss
Copy link

majuss commented Jun 11, 2018

Hi there!

When I enter all my DSL credentials via PPPoE, hit save and apply and reconnect the dsl0 interface, nothing happens and the box can't login into the DSL connection.

Should this be working right now or is it still work in progress?

@majuss
Copy link
Author

majuss commented Jun 25, 2018

ohne titel1
ohne titel3
ohne titel4

So I figured that the configuration of the virtual switch (with the VLAN ID 7) is somewhat the problem. But I'am not experienced enough to set this switch up. Could somebody please help?

@majuss
Copy link
Author

majuss commented Jul 3, 2018

So this is the output of ifconfig on a original FW easybox. I think with that information someone could reconstruct the Switch settings for openWRT.

What is also wierd: The data rate shows only 20 Mbit... but it should be 100/40 :(

root@easy:~# ifconfig
br-lan    Link encap:Ethernet  HWaddr XXX
          inet addr:192.168.2.1  Bcast:192.168.2.255  Mask:255.255.255.0
          UP BROADCAST RUNNING MULTICAST  MTU:1500  Metric:1
          RX packets:624541 errors:0 dropped:0 overruns:0 frame:0
          TX packets:370049 errors:0 dropped:0 overruns:0 carrier:0
          collisions:0 txqueuelen:0
          RX bytes:56594044 (53.9 MiB)  TX bytes:81366278 (77.5 MiB)

br-lan1   Link encap:Ethernet  HWaddr YYY
          inet addr:192.168.3.1  Bcast:192.168.3.255  Mask:255.255.255.0
          UP BROADCAST RUNNING MULTICAST  MTU:1500  Metric:1
          RX packets:0 errors:0 dropped:0 overruns:0 frame:0
          TX packets:24690 errors:0 dropped:0 overruns:0 carrier:0
          collisions:0 txqueuelen:0
          RX bytes:0 (0.0 B)  TX bytes:1482544 (1.4 MiB)

eth0      Link encap:Ethernet  HWaddr XXX
          UP BROADCAST RUNNING MULTICAST  MTU:1500  Metric:1
          RX packets:4391218 errors:0 dropped:117 overruns:0 frame:0
          TX packets:13245353 errors:0 dropped:0 overruns:0 carrier:0
          collisions:0 txqueuelen:1000
          RX bytes:556435907 (530.6 MiB)  TX bytes:113289512 (108.0 MiB)

eth0.66   Link encap:Ethernet  HWaddr XXX
          UP BROADCAST RUNNING PROMISC MULTICAST  MTU:1500  Metric:1
          RX packets:978680 errors:0 dropped:0 overruns:0 frame:0
          TX packets:370052 errors:0 dropped:0 overruns:0 carrier:0
          collisions:0 txqueuelen:0
          RX bytes:184983722 (176.4 MiB)  TX bytes:82846724 (79.0 MiB)

eth0.71   Link encap:Ethernet  HWaddr YYY
          UP BROADCAST RUNNING PROMISC MULTICAST  MTU:1500  Metric:1
          RX packets:0 errors:0 dropped:0 overruns:0 frame:0
          TX packets:24692 errors:0 dropped:0 overruns:0 carrier:0
          collisions:0 txqueuelen:0
          RX bytes:0 (0.0 B)  TX bytes:1581480 (1.5 MiB)

imq1      Link encap:UNSPEC  HWaddr 00-00-00-00-00-00-00-00-00-00-00-00-00-00-00-00
          UP RUNNING NOARP  MTU:16000  Metric:1
          RX packets:329062 errors:0 dropped:0 overruns:0 frame:0
          TX packets:329058 errors:0 dropped:0 overruns:0 carrier:0
          collisions:0 txqueuelen:11000
          RX bytes:33462241 (31.9 MiB)  TX bytes:33462033 (31.9 MiB)

lo        Link encap:Local Loopback
          inet addr:127.0.0.1  Mask:255.0.0.0
          UP LOOPBACK RUNNING  MTU:16436  Metric:1
          RX packets:57415 errors:0 dropped:0 overruns:0 frame:0
          TX packets:57415 errors:0 dropped:0 overruns:0 carrier:0
          collisions:0 txqueuelen:0
          RX bytes:8897588 (8.4 MiB)  TX bytes:8897588 (8.4 MiB)

ppp50     Link encap:Point-to-Point Protocol
          inet addr:AAA  P-t-P:AAA  Mask:255.255.255.255
          UP POINTOPOINT RUNNING NOARP MULTICAST  MTU:1492  Metric:1
          RX packets:144027 errors:0 dropped:0 overruns:0 frame:0
          TX packets:328961 errors:0 dropped:0 overruns:0 carrier:0
          collisions:0 txqueuelen:3
          RX bytes:63227971 (60.2 MiB)  TX bytes:33426084 (31.8 MiB)

ptm0      Link encap:Ethernet  HWaddr ZZZ
          UP BROADCAST RUNNING MULTICAST  MTU:1500  Metric:1
          RX packets:12817835 errors:0 dropped:0 overruns:0 frame:0
          TX packets:3774190 errors:0 dropped:12 overruns:0 carrier:0
          collisions:0 txqueuelen:1000
          RX bytes:547722283 (522.3 MiB)  TX bytes:54648004 (52.1 MiB)

ptm0.7    Link encap:Ethernet  HWaddr ZZZ
          UP BROADCAST RUNNING MULTICAST  MTU:1492  Metric:1
          RX packets:160123 errors:0 dropped:0 overruns:0 frame:0
          TX packets:345062 errors:0 dropped:0 overruns:0 carrier:0
          collisions:0 txqueuelen:0
          RX bytes:65696950 (62.6 MiB)  TX bytes:42526629 (40.5 MiB)

ra0       Link encap:Ethernet  HWaddr XXY
          UP BROADCAST RUNNING MULTICAST  MTU:1500  Metric:1
          RX packets:0 errors:0 dropped:0 overruns:0 frame:0
          TX packets:0 errors:0 dropped:0 overruns:0 carrier:0
          collisions:0 txqueuelen:1000
          RX bytes:0 (0.0 B)  TX bytes:0 (0.0 B)

ra1       Link encap:Ethernet  HWaddr XXY
          UP BROADCAST RUNNING MULTICAST  MTU:1500  Metric:1
          RX packets:0 errors:0 dropped:0 overruns:0 frame:0
          TX packets:0 errors:0 dropped:0 overruns:0 carrier:0
          collisions:0 txqueuelen:1000
          RX bytes:0 (0.0 B)  TX bytes:0 (0.0 B)

rai0      Link encap:Ethernet  HWaddr XXZ
          UP BROADCAST RUNNING MULTICAST  MTU:1500  Metric:1
          RX packets:0 errors:0 dropped:0 overruns:0 frame:0
          TX packets:0 errors:0 dropped:0 overruns:0 carrier:0
          collisions:0 txqueuelen:1000
          RX bytes:0 (0.0 B)  TX bytes:0 (0.0 B)

rai1      Link encap:Ethernet  HWaddr XYZ
          UP BROADCAST RUNNING MULTICAST  MTU:1500  Metric:1
          RX packets:0 errors:0 dropped:0 overruns:0 frame:0
          TX packets:0 errors:0 dropped:0 overruns:0 carrier:0
          collisions:0 txqueuelen:1000
          RX bytes:0 (0.0 B)  TX bytes:0 (0.0 B)

@Quallenauge
Copy link
Owner

Quallenauge commented Jul 3, 2018

Hi there.
Sadly the ifconfig doesn't show the switch settings.
Also some settings are set within the vendor kernel.
At least I was able to connect to the T-DSL VDSL provider with low speed (20MB/s).
I'm not sure, but I guess the speed is a limitation from the used vdsl modem. Maybe a newer release would help, but I don't know. lantiq-vdsl-fw: update to provide recent vectoring firmware
I have a look how the setting was made for this, I think it was close to your settings.
Try to set also the CPU port 0 to "tagged", see here: https://telekomhilft.telekom.de/t5/Telefonie-Internet/Wie-OpenWRT-Router-am-FTTH-Anschluss-betreiben/td-p/2283926

@majuss
Copy link
Author

majuss commented Jul 3, 2018

How did you manage to login?! I tried every combination with and without # etc. it never worked...
Could you implement the new modem firmware to your openWRT release?

@Quallenauge
Copy link
Owner

I think, exactly as described within the post I previously linked.
Newer builds should contain the vdsl firmware, but I haven't tried that.
https://app.box.com/s/hvqg535dnubt4r2ontpmtodpvt6ydf00/folder/36913951101

@majuss
Copy link
Author

majuss commented Jul 4, 2018

I tried the latest image, but no luck. Still 20 mbit and the version is very buggy. Always crashes when you try edit the config. Still couldn't log into my DSL, even with all of the switch settings like in the forum post.

@Quallenauge
Copy link
Owner

Sorry, I don't develop LuCI so I can't help with this. Maybe a newer version might be better.
Also the kind how the switches are connected together isn't and straight forward. (You can easily lock out yourself from the device and have to use the recovery.)

@majuss
Copy link
Author

majuss commented Jul 7, 2018

I tried the latest version from 4th of july but no luck, can't log the ppoe in. I cant add a new VLAN id, when I click the add button nothing happens. And its still at 20 mbit.

These are all VLAN settings from a working EB conencted with telekom VDSL with the original image.

enable_vlan@vlan = 1
lan_netdevice@vlan = eth0
wan_netdevice@vlan = eth1
switch_port_num@vlan = 8
logical_port_num@vlan = 7
logical_total_port_num@vlan = 8
logical_cpu_port@vlan = 6
port_phyconf@vlan = 0 0 0 0 5 4 6 7
max_vlan@vlan = 4
lan_vlan_id@vlan = 66
lan_port@vlan = 0 1 2 3 4
wan_port@vlan = 5
wlan_port@vlan = 4
active_vlan@vlan = 3
vlan1_name@vlan = default LAN
vlan2_name@vlan =
vlan3_name@vlan =
vlan4_name@vlan =
vlan1_id@vlan = 71
vlan2_id@vlan = 66
vlan3_id@vlan = 66
vlan4_id@vlan = 4
vlan1_tagged@vlan = 1
vlan2_tagged@vlan = 0
vlan3_tagged@vlan = 1
vlan4_tagged@vlan = 0
vlan1_port@vlan = 4
vlan2_port@vlan = 0 1 2 3
vlan3_port@vlan = 4
vlan4_port@vlan =
vlan1_ssid@vlan = 0
vlan2_ssid@vlan =
vlan3_ssid@vlan =
vlan4_ssid@vlan =
vlan1_brwan@vlan = 0
vlan2_brwan@vlan = 0
vlan3_brwan@vlan = 0
vlan4_brwan@vlan = 0
vlan1_bind_rtwan@vlan = 0
vlan2_bind_rtwan@vlan = 0
vlan3_bind_rtwan@vlan = 0
vlan4_bind_rtwan@vlan = 0
wan_lan_share@vlan = 0
vlan_tagged@lan0 = 0
vlan_1p@lan0 = 0
vlan_id@lan0 = 0
vlan_tagged@lan1 = 0
vlan_1p@lan1 = 0
vlan_id@lan1 = 0
vlan_tagged@wan100 = 1
vlan_id@wan100 = 132
vlan_tagged@wan101 = 1
vlan_id@wan101 = 232
vlan_tagged@wan102 = 1
vlan_id@wan102 = 332
vlan_tagged@wan000 = 0
vlan_id@wan000 = 0
vlan_tagged@wan001 = 0
vlan_id@wan001 = 0
vlan_tagged@wan002 = 0
vlan_id@wan002 = 0
vlan_tagged@wan050 = 1
vlan_id@wan050 = 7
vlan_tagged@wan051 = 1
vlan_id@wan051 = 232
vlan_tagged@wan052 = 1
vlan_id@wan052 = 332
vlan_tagged@wan300 = 0
vlan_id@wan300 = 5
vlan_tagged@wan301 = 1
vlan_id@wan301 = 3
vlan_tagged@wan302 = 1
vlan_id@wan302 = 5
vlan_tagged@wan303 = 0
vlan_id@wan303 = 5
vlan_tagged@wan200 = 0
vlan_id@wan200 = 0

@Quallenauge
Copy link
Owner

Quallenauge commented Jul 9, 2018

I just tried to get DSL running, with success.
Try the following:
nano /etc/config/network

Add the following

config switch_vlan
        option device 'switch0'
        option vlan '7'
        option ports '4t'
        option vid '7'

Update the dsl section from

config interface 'wan'
        option ifname 'dsl0'
        option proto 'pppoe'
        option ipv6 '1'
...

to

config interface 'wan'
        option ifname 'dsl0.7'
        option proto 'pppoe'
        option ipv6 '1'
...

Reboot. As DSL firmware I used the vr9_dsl_fw_annex_b.bin (20MBit/s Downstream only).
There's also a firmware version which can be downloaded with vdsl_fw_install.sh it connects after a while with ~100MBit/s Downstream but reconnects after a few seconds. Maybe another FW will help.

@majuss
Copy link
Author

majuss commented Jul 16, 2018

Thanks for the help with the VLAN setup!
I finally got everything working also with 100 Mbit, connection was stable for 6 minutes, intense download, tomorrow I will test it in the long run. What I did:

wget http://download.avm.de/fritzbox/fritzbox-7490/deutschland/fritz.os/FRITZ.Box_7490.113.06.93.image
7z e FRITZ.Box_7490.113.06.93.image -r filesystem.image
7z e filesystem.image filesystem_core.squashfs
unsquashfs4-avm-be filesystem_core.squashfs -e lib/modules/dsp_vr9/

And then use the vr9-B-dsl.bin in openWRT by copying it into lib/firmware/ and then enter the full path in the webGUI. Note that you will need the binary for unsquashfs4-avm-be you can get it by get the freetz source and run make tools.

So this uses the Firmware of a FRITZ!Box 7490 FRITZ!OS 6.93.

@majuss
Copy link
Author

majuss commented Jul 17, 2018

So after some intense testing I can say that it works, but you're not able to get 100 Mbit only ~76 Mbit/s. The sirq and ksoftirqd are then eating 100 % CPU, which indicates the CPU can't handle the interrupts fast enough. Thats a pity because everything else seems fine. Maybe I will try a different firmware.

@Quallenauge
Copy link
Owner

Maybe we can unleash the dual core for this at the cost of that asterisk won't work anymore.

@majuss
Copy link
Author

majuss commented Jul 17, 2018

The Asterisk is all the call handling? This would be not a big deal. How can I archieve it? Could you build a version and upload it to appbox? Otherwise I will setup an openWRT build environment :)

https://forum.lede-project.org/t/how-can-we-make-the-lantiq-xrx200-devices-faster/9724
This thread is also quite interesting... They state that the original vendors can use other hw acc. features to get the full 100 Mbit.

@Quallenauge
Copy link
Owner

I can try to provide an experimental one...

@majuss
Copy link
Author

majuss commented Jul 18, 2018

That would be awesome :) I tested 100 Mbit with the Vodafone firmware and the sirq raises to 70%.

But another big problem is the WLAN of the easybox... It's unusable with the webinterface. You can only set WEP and nothing else, changig the password also is not working therefore WLAN is completely unusable for normal users. Can a working WLAN network be setted up via ssh?

@henning-schild
Copy link

Can this issue be closed? I managed to get DSL working with VLAN as well, this box is not any different than any other OpenWRT router that does pppoe on a VLAN. I could post my /etc/config/network in case anyone still cares.

The high load on a fast DSL could be another issue, same is true for the second core, and the wifi interface to luci.

@majuss
Copy link
Author

majuss commented Aug 1, 2018

I will maybe open two other issues for the high CPU usage and one for the WLAN. And yes this can get closed. Is there somekind of official documentation for the VLAN pppoe?

@ghost
Copy link

ghost commented Aug 1, 2018 via email

@henning-schild
Copy link

This is for a dslite setup so it looks different to what you will need.
Some things might not be required. The key is the ifname in wan .
Note that i did not have to add the vlanid (40) to the switch. I just touched "dsl" "wan" "wan6" "wan4"

config dsl 'dsl'
	option annex 'b'
	option tone 'av'
	option xfer_mode 'ptm'
	option line_mode 'vdsl'
	option firmware '/lib/firmware/lantiq-vrx200-b.bin'
...
config interface 'wan'
	option ifname 'dsl0.40'
	option proto 'pppoe'
	option ipv6 '1'
	option username '..'
	option password '..'
	option mtu '1492'

config device 'wan_dev'
	option name 'dsl0'
	option macaddr '..'

config interface 'wan6'
	option proto 'dhcpv6'
	option ifname '@wan'

config interface 'wan4'
	option proto 'dslite'
	option peeraddr '..'
	option mtu '1444'
	option encaplimit 'ignore'
...

@majuss
Copy link
Author

majuss commented Sep 14, 2018

Any news regardign the sirq problem? I'am ready to test some build, or build them by myself if somebody can guide me a little.

@Quallenauge
Copy link
Owner

Any news regardign the sirq problem? I'am ready to test some build, or build them by myself if somebody can guide me a little.

At least I added some builds with "smp" as prefix, which enables the second core.

@icke68746
Copy link

I would also test the new version.
I had the bootloop, the last two versions of
https://app.box.com/s/hvqg535dnubt4r2ontpmtodpvt6ydf00/folder/36913951101 Hans Mueller. Furthermore, I can't use the WLAN and the DSL, it can also be due to a wrong DSL setting, the original firmware did a lot by itself, I still have to get used to it.
@Quallenauge I cannot compile the image. Could you provide the image?

Translated with www.DeepL.com/Translator

Ich würde die neue Version auch testen.
Bei mir kam es zum Bootloop, bei den letzten beiden Versionen von
https://app.box.com/s/hvqg535dnubt4r2ontpmtodpvt6ydf00/folder/36913951101 Hans Mueller. Des Weiteren kann ich das WLAN und das DSL nicht in Betrieb nehmen, es kann auch an einer falschen Einstellung bei DSL liegen, die Original-Firmware hat vieles selbst gemacht, da muß ich mich noch einarbeiten.
@Quallenauge Ich kann das Image nicht kompilieren. Könntest Du das Image zur Verfügung stellen?

@majuss
Copy link
Author

majuss commented Sep 24, 2018

Since 30.8. the image contains smp, see the files in the directory. Maybe thats what causes the bootloop. We would need to connect a serial console and look into the bootlog.

I quickly tested the newest build from 2. of september. Not working at all, you can't reboot the device, without it resetting itself. Even ssh thinks it's a new host, so it does a factory reset at every boot.

@icke68746 I didn't got any bootloops or anything.

@majuss
Copy link
Author

majuss commented Oct 8, 2018

So after some intense testing hours I can say that you first have to flash the stock image, then the openWrt recovery image, upload the NON-SMP image and flash it via the webinterface. Boot into the NON-SMP fw and then flash the SMP firmware from there. Everything else is creating a zombie-Wrt.
I will publish a guide for that exhausting procedure in my repo very soon. 100 Mbit, WLAN etc. is working fine!

@henning-schild
Copy link

@majuss it might be more useful to understand why the smp image can not be flashed directly and fix that, maybe bisection could give a clue

@Quallenauge
Copy link
Owner

That's really weird that this doesn't work :-(
I thought I switched from normal to smp and vice versa in the past without such problems.
Sure Asterisk won't work due missing vpe cababilities but the other things should work, since the image contains all needed components (kernel + drivers). The remaining components doesn't rely on a smp/no-smp compiled kernel.

@majuss
Copy link
Author

majuss commented Oct 25, 2018

I think it has something to do with the partitioning. The SMP image can't create the partitions correctly and if you first flashed the non-SMP one the partitions are already set up in some wierd way (I hope that makes sense somehow).

Are you sure that the asterisk (so fixed line telephony?) is not working anymore? For me it seemed pretty alive :D I will maybe test it.

@majuss
Copy link
Author

majuss commented Nov 3, 2018

Still not able to get a reproducible working openWrt with smp -.-

root@OpenWrt:~# dmesg | grep mount
[    3.972340] Waiting 7 sec before mounting root device...
[   19.161224] mount_root: loading kmods from internal overlay
[   21.034874] [<801ce084>] ubifs_mount+0x1034/0x1850
[   21.039678] [<8012a06c>] mount_fs+0x20/0xc4
[   21.043839] [<8014f114>] vfs_kern_mount.part.9+0x60/0x170
[   21.049230] [<80152ab0>] do_mount+0x9dc/0xe58
[   21.053587] [<801532d4>] SyS_mount+0xb8/0xfc
[   21.149069] [<801ce084>] ubifs_mount+0x1034/0x1850
[   21.153857] [<8012a06c>] mount_fs+0x20/0xc4
[   21.158036] [<8014f114>] vfs_kern_mount.part.9+0x60/0x170
[   21.163435] [<80152ab0>] do_mount+0x9dc/0xe58
[   21.167779] [<801532d4>] SyS_mount+0xb8/0xfc
[   21.480181] CPU: 1 PID: 638 Comm: mount_root Not tainted 4.14.72 #0
[   21.571560] [<801ce084>] ubifs_mount+0x1034/0x1850
[   21.576356] [<8012a06c>] mount_fs+0x20/0xc4
[   21.580519] [<8014f114>] vfs_kern_mount.part.9+0x60/0x170
[   21.585917] [<80152ab0>] do_mount+0x9dc/0xe58
[   21.590264] [<801532d4>] SyS_mount+0xb8/0xfc
[   21.609096] CPU: 1 PID: 638 Comm: mount_root Not tainted 4.14.72 #0
[   21.686183] [<801ce084>] ubifs_mount+0x1034/0x1850
[   21.690979] [<8012a06c>] mount_fs+0x20/0xc4
[   21.695148] [<8014f114>] vfs_kern_mount.part.9+0x60/0x170
[   21.700544] [<80152ab0>] do_mount+0x9dc/0xe58
[   21.704896] [<801532d4>] SyS_mount+0xb8/0xfc
[   21.728806] mount_root: failed to mount -t ubifs /dev/ubi0_1 /tmp/overlay: Bad message
[   23.182035] [<801ce084>] ubifs_mount+0x1034/0x1850
[   23.186839] [<8012a06c>] mount_fs+0x20/0xc4
[   23.191002] [<8014f114>] vfs_kern_mount.part.9+0x60/0x170
[   23.196402] [<80152ab0>] do_mount+0x9dc/0xe58
[   23.200748] [<801532d4>] SyS_mount+0xb8/0xfc
[   23.296232] [<801ce084>] ubifs_mount+0x1034/0x1850
[   23.301024] [<8012a06c>] mount_fs+0x20/0xc4
[   23.305197] [<8014f114>] vfs_kern_mount.part.9+0x60/0x170
[   23.310593] [<80152ab0>] do_mount+0x9dc/0xe58
[   23.314944] [<801532d4>] SyS_mount+0xb8/0xfc
[   23.340512] mount_root: unable to set filesystem state
[   23.344460] mount_root: switching to ubifs overlay
[   23.349311] mount_root: switching to ubifs failed - fallback to ramoverlay
root@OpenWrt:~# mount
/dev/root on /rom type squashfs (ro,relatime)
proc on /proc type proc (rw,nosuid,nodev,noexec,noatime)
sysfs on /sys type sysfs (rw,nosuid,nodev,noexec,noatime)
tmpfs on /tmp type tmpfs (rw,nosuid,nodev,noatime)
tmpfs on /tmp/root type tmpfs (rw,noatime,mode=755)
overlayfs:/tmp/root on / type overlay (rw,noatime,lowerdir=/,upperdir=/tmp/root/upper,workdir=/tmp/root/work)
tmpfs on /dev type tmpfs (rw,nosuid,relatime,size=512k,mode=755)
devpts on /dev/pts type devpts (rw,nosuid,noexec,relatime,mode=600,ptmxmode=000)
debugfs on /sys/kernel/debug type debugfs (rw,noatime)

The overlay fs cant mount and so the config is getting lost after reboot.

@majuss
Copy link
Author

majuss commented Nov 8, 2018

@Quallenauge could you look into the issue or hint me in a direction? I'am somewhat lost right now.

@Quallenauge
Copy link
Owner

Not really. But can you let me know the steps in detail how to to reach your current state; what steps are needed (in detail please). Actually I use the non-smp version.

@majuss
Copy link
Author

majuss commented Dec 14, 2018

Ah thank you. So I should just try every annex and see if one yields 100 mbit?

@Quallenauge
Copy link
Owner

Maybe this helps. Please share your results. Sadly I can't verify and test atm.

@majuss
Copy link
Author

majuss commented Dec 14, 2018

I tried testing it, no success it kills the connection every few seconds, totally bugged out.

@kovz
Copy link
Contributor

kovz commented Dec 14, 2018

@majuss what do you mean by "kill connection"? Do you hear relay switching? This patch shouldn't impact on dsl modem, because it always connected to phone line directly.
I'm little bit confusing by your screenshot. Available attenuation DL speed is more then 100Mb.

@majuss
Copy link
Author

majuss commented Dec 14, 2018

I will test it tomorrow and will upload my config and logs.

@majuss
Copy link
Author

majuss commented Dec 15, 2018

Sat Dec 15 11:11:35 2018 daemon.notice netifd: Interface 'wan6' is now down
Sat Dec 15 11:11:35 2018 daemon.notice netifd: Network alias '' link is down
Sat Dec 15 11:11:35 2018 daemon.notice netifd: Interface 'wan6' has link connectivity loss
Sat Dec 15 11:11:35 2018 daemon.notice netifd: Interface 'wan6' is disabled
Sat Dec 15 11:11:35 2018 daemon.warn dnsmasq[3501]: no servers found in /tmp/resolv.conf.auto, will retry
Sat Dec 15 11:11:35 2018 daemon.info pppd[2978]: Terminating on signal 15
Sat Dec 15 11:11:35 2018 daemon.info pppd[2978]: Connect time 2.8 minutes.
Sat Dec 15 11:11:35 2018 daemon.info pppd[2978]: Sent 4325 bytes, received 4682 bytes.
Sat Dec 15 11:11:35 2018 daemon.notice netifd: Network device 'pppoe-wan' link is down
Sat Dec 15 11:11:35 2018 daemon.notice pppd[2978]: Connection terminated.
Sat Dec 15 11:11:35 2018 daemon.info pppd[2978]: Sent PADT
Sat Dec 15 11:11:35 2018 daemon.info pppd[2978]: Exit.
Sat Dec 15 11:11:35 2018 daemon.notice netifd: Interface 'wan' is now down
Sat Dec 15 11:11:35 2018 kern.info kernel: [  339.232293] IPv6: ADDRCONF(NETDEV_UP): dsl0.7: link is not ready
Sat Dec 15 11:11:35 2018 daemon.notice netifd: Interface 'wan' is disabled
Sat Dec 15 11:11:35 2018 daemon.notice netifd: Interface 'wan' is enabled
Sat Dec 15 11:11:35 2018 daemon.notice netifd: Interface 'wan' is setting up now
Sat Dec 15 11:11:35 2018 daemon.err insmod: module is already loaded - slhc
Sat Dec 15 11:11:35 2018 daemon.err insmod: module is already loaded - ppp_generic
Sat Dec 15 11:11:35 2018 daemon.err insmod: module is already loaded - pppox
Sat Dec 15 11:11:35 2018 daemon.err insmod: module is already loaded - pppoe
Sat Dec 15 11:11:35 2018 daemon.info pppd[4573]: Plugin rp-pppoe.so loaded.
Sat Dec 15 11:11:35 2018 daemon.info pppd[4573]: RP-PPPoE plugin version 3.8p compiled against pppd 2.4.7
Sat Dec 15 11:11:36 2018 daemon.notice pppd[4573]: pppd 2.4.7 started by root, uid 0
Sat Dec 15 11:11:38 2018 user.info adblock-3.5.5-4[4647]: adblock is currently disabled, please set adb_enabled to '1' to use this service
Sat Dec 15 11:11:46 2018 daemon.info pppd[4573]: PPP session is 28
Sat Dec 15 11:11:46 2018 daemon.warn pppd[4573]: Connected to 54:4b:8c:ec:bb:b6 via interface dsl0.7
Sat Dec 15 11:11:46 2018 kern.info kernel: [  349.839028] pppoe-wan: renamed from ppp0
Sat Dec 15 11:11:46 2018 daemon.info pppd[4573]: Using interface pppoe-wan
Sat Dec 15 11:11:46 2018 daemon.notice pppd[4573]: Connect: pppoe-wan <--> dsl0.7
Sat Dec 15 11:11:46 2018 daemon.info pppd[4573]: Remote message: SRU=21657#SRD=56495#
Sat Dec 15 11:11:46 2018 daemon.notice pppd[4573]: PAP authentication succeeded
Sat Dec 15 11:11:46 2018 daemon.notice pppd[4573]: peer from calling number 54:4B:8C:EC:BB:B6 authorized
Sat Dec 15 11:11:46 2018 daemon.notice pppd[4573]: local  IP address 93.214.162.161
Sat Dec 15 11:11:46 2018 daemon.notice pppd[4573]: remote IP address 62.155.240.98
Sat Dec 15 11:11:46 2018 daemon.notice pppd[4573]: primary   DNS address 217.0.43.65
Sat Dec 15 11:11:46 2018 daemon.notice pppd[4573]: secondary DNS address 217.0.43.81
Sat Dec 15 11:11:46 2018 daemon.notice netifd: Network device 'pppoe-wan' link is up
Sat Dec 15 11:11:46 2018 daemon.notice netifd: Interface 'wan6' is enabled
Sat Dec 15 11:11:46 2018 daemon.notice netifd: Network alias 'pppoe-wan' link is up
Sat Dec 15 11:11:46 2018 daemon.notice netifd: Interface 'wan6' has link connectivity
Sat Dec 15 11:11:46 2018 daemon.notice netifd: Interface 'wan6' is setting up now
Sat Dec 15 11:11:46 2018 daemon.notice netifd: Interface 'wan' is now up
Sat Dec 15 11:11:46 2018 daemon.info dnsmasq[3501]: reading /tmp/resolv.conf.auto
Sat Dec 15 11:11:46 2018 daemon.info dnsmasq[3501]: using local addresses only for domain test
Sat Dec 15 11:11:46 2018 daemon.info dnsmasq[3501]: using local addresses only for domain onion
Sat Dec 15 11:11:46 2018 daemon.info dnsmasq[3501]: using local addresses only for domain localhost
Sat Dec 15 11:11:46 2018 daemon.info dnsmasq[3501]: using local addresses only for domain local
Sat Dec 15 11:11:46 2018 daemon.info dnsmasq[3501]: using local addresses only for domain invalid
Sat Dec 15 11:11:46 2018 daemon.info dnsmasq[3501]: using local addresses only for domain bind
Sat Dec 15 11:11:46 2018 daemon.info dnsmasq[3501]: using local addresses only for domain lan
Sat Dec 15 11:11:46 2018 daemon.info dnsmasq[3501]: using nameserver 217.0.43.65#53
Sat Dec 15 11:11:46 2018 daemon.info dnsmasq[3501]: using nameserver 217.0.43.81#53
Sat Dec 15 11:11:46 2018 daemon.err odhcp6c[4796]: Failed to send RS (Permission denied)
Sat Dec 15 11:11:47 2018 user.notice firewall: Reloading firewall due to ifup of wan (pppoe-wan)
Sat Dec 15 11:11:47 2018 daemon.err odhcp6c[4796]: Failed to send DHCPV6 message to ff02::1:2 (Permission denied)
Sat Dec 15 11:11:47 2018 daemon.info privoxy[3393]: Restart request due to 'ifup' of interface 'wan'
Sat Dec 15 11:11:48 2018 daemon.err odhcp6c[4796]: Failed to send DHCPV6 message to ff02::1:2 (Permission denied)
Sat Dec 15 11:11:48 2018 daemon.notice privoxy[3393]: Shutdown successfully
Sat Dec 15 11:11:50 2018 user.info adblock-3.5.5-4[5027]: adblock is currently disabled, please set adb_enabled to '1' to use this service
Sat Dec 15 11:11:50 2018 daemon.notice privoxy[5021]: Started successfully
Sat Dec 15 11:11:50 2018 daemon.err odhcp6c[4796]: Failed to send DHCPV6 message to ff02::1:2 (Permission denied)
Sat Dec 15 11:11:50 2018 daemon.err odhcp6c[4796]: Failed to send RS (Permission denied)
Sat Dec 15 11:11:54 2018 daemon.err odhcp6c[4796]: Failed to send RS (Permission denied)
Sat Dec 15 11:11:55 2018 daemon.err odhcp6c[4796]: Failed to send DHCPV6 message to ff02::1:2 (Permission denied)
Sat Dec 15 11:11:58 2018 daemon.err odhcp6c[4796]: Failed to send RS (Permission denied)
Sat Dec 15 11:12:04 2018 daemon.err odhcp6c[4796]: Failed to send DHCPV6 message to ff02::1:2 (Permission denied)
Sat Dec 15 11:12:22 2018 daemon.err odhcp6c[4796]: Failed to send DHCPV6 message to ff02::1:2 (Permission denied)

This is the log when I hit the restart button on my WAN interface via the gui. For me it is stable at 60 Mbit, but yesterday with the connection of a friend, same Telekom 100 Mbit, the box got 100 Mbit but restarted randomly and kicked out the WAN interface, it thought that no cable is connected after all.

My etc/config/network:

config interface 'wan'
	option ifname 'dsl0.7'
	option proto 'pppoe'
	option username 'abc#[email protected]'
	option password 'qwe'
	option ipv6 '0'

config switch_vlan
	option device 'switch0'
	option vlan '7'
	option ports '4t'
	option vid '7'

config dsl 'dsl'
	option annex 'a'
	option tone 'av'
	option ds_snr_offset '0'
	option firmware '/lib/fw.bin'

@arnysch
Copy link
Contributor

arnysch commented Dec 15, 2018

Hi majuss, just curious, could you answer these questions:

  • Is there a reason that you configured "option annex 'a'" instead of "option annex 'b'".
    Wondering, because the DSL status says that you are using Annex-B.
  • what is the output of this command: grep "" /sys/class/gpio/*/value
    I assume you have /sys/class/gpio/annexj/value switched on, which I guess*) disables the internal DSL highpasses used for Annex-A or Annex-B. Don't know if this affects the data rate in your case.
    Anyway, the latest commit ("Add missing space...") should correct this.
    *)= kovz explained the internal connections of the relais, but to me this still leaves room for speculation about some details

@majuss
Copy link
Author

majuss commented Dec 15, 2018

The output:

/sys/class/gpio/annexa/value:1
/sys/class/gpio/annexb/value:0
/sys/class/gpio/annexj/value:1
/sys/class/gpio/dsl_en/value:1
/sys/class/gpio/hc595_0/value:1
/sys/class/gpio/hc595_1/value:1
/sys/class/gpio/hc595_3/value:1

I did not wrote annex a in there, it was preconfigured that way. The GUI say "Annex a + b + ..."

@kovz
Copy link
Contributor

kovz commented Dec 15, 2018

@majuss And could you please post here output for grep "" /sys/class/gpio/*/active_low
And for Germany, in case of VDSL connection, annex should be set to B.

@majuss
Copy link
Author

majuss commented Dec 15, 2018

/sys/class/gpio/annexa/active_low:1
/sys/class/gpio/annexb/active_low:0
/sys/class/gpio/annexj/active_low:0
/sys/class/gpio/dsl_en/active_low:1
/sys/class/gpio/hc595_0/active_low:0
/sys/class/gpio/hc595_1/active_low:0
/sys/class/gpio/hc595_3/active_low:0

@kovz
Copy link
Contributor

kovz commented Dec 15, 2018

Ok. Let me check with my connection. I can compare speed with speedport hybrid from Telekom.

@arnysch
Copy link
Contributor

arnysch commented Dec 15, 2018

Hi majuss, I guess he GUI does not say "Annex a + b + ...", but "Annex A + L + M...". Unless you have another GUI version than me.
To our current knowledge (kovz, anyone, please correct if I am wrong) we suggest these values:
/sys/class/gpio/annexa/value:0
/sys/class/gpio/annexb/value:1
/sys/class/gpio/annexj/value:0
To change this, I propose these actions:

  1. Switch DSL config settings to Annex B. For me (provider O2 using Deutsche Telekom line), I use these VDSL Annex B settings in the GUI:
  • Annex: Annex B (all)
  • Tone: auto
  • Encapsulation mode: PTM/EFM (Packet Transfer Mode)
  • DSL Line Mode: VDSL
  • Downstream SNR offset: 0.0 dB
  • Firmware File: /lib/...
  1. Use the latest file /etc/init.d/dsl_control, which got corrected in the master-lede branch today.
    No need to install a new OpenWrt; simply use an editor and insert a space character in your eb904 in the current /etc/init.d/dsl_control file: locate the line first listed below, and make it look like the 2nd line:
    if [ -d "${annexgpio}a" ] && [ -d "${annexgpio}b" ] && [-d "${annexgpio}j" ]; then
    if [ -d "${annexgpio}a" ] && [ -d "${annexgpio}b" ] && [ -d "${annexgpio}j" ]; then

  2. Then reboot and recheck the DSL values, and let's hope the best. Would be good to hear if/if not there was an improvement. You could then also use the grep "" /sys/class/gpio/*/value command again to check if the relay/gpio settings are now correct.

@majuss
Copy link
Author

majuss commented Dec 16, 2018

Still 60 Mbit:

/sys/class/gpio/annexa/value:0
/sys/class/gpio/annexb/value:1
/sys/class/gpio/annexj/value:0
/sys/class/gpio/dsl_en/value:1
/sys/class/gpio/hc595_0/value:1
/sys/class/gpio/hc595_1/value:1
/sys/class/gpio/hc595_3/value:1

config dsl 'dsl'
        option firmware '/lib/fw.bin'
        option ds_snr_offset '0'
        option annex 'b'
        option xfer_mode 'ptm'

@arnysch
Copy link
Contributor

arnysch commented Dec 16, 2018

Thanks for testing. Sorry, I cannot tell you the solution. But your reports are instructive. For example it looks like in your case it does not matter if the internal splitter (i.e. highpass for DSL) is/is not used.
I believe that the settings I suggested are correct. If it is no problem for you, you could add a picture of your new DSL Status like you did here, just so we can see if for example the noise margin improved.

Also interesting would be to hear if your 60Mb/s connection is stable with the new settings.

As kovz already indicated, your DSL values reported earlier are confusing. Downstream data rate is 60Mb/s, though it claims that max attainable rate were 113 Mb/s. In case you have access to another box, which also shows you the data rate (is this possible e.g. with Fritz!OS? I have no idea), you could check if this other box manages to get a better rate.

Maybe your provider limits the downstream rate to 60 Mb/s? For example my DSL as agreed on with my provider is limitted to 50Mb/s. My DSL status shows a data rate of 51 Mb/s, while the max attainable rate is 128 Mb/s.

Or lacking better ideas: maybe the non-smp kernel would help?

@majuss
Copy link
Author

majuss commented Dec 16, 2018

Here are my new DSL values:
screenshot 2018-12-16 at 18 12 06

Other boxes get 100 Mbit, the images before the DSL relay switch fix also got 100 Mbit. SMP is a must because the kernel interrupts eat so much CPU that you can't ever reach more than 70 Mbit.

@arnysch
Copy link
Contributor

arnysch commented Dec 16, 2018

My interpretation of the new DSL status:

  • Attenuation values increased a little bit - lower is better (attenuation caused by splitter)
  • SNR Signal-to-Noise ratio increased - higher is better (some line noise removed by splitter)

Result: higher max. attainable data rates, now 128.6/42.2 instead of 113.7/35.9 Mb/s.
But this does not explain the data rate limitation to 60/23 Mb/s.

@arnysch
Copy link
Contributor

arnysch commented Dec 16, 2018

Hi majuss,

the images before the DSL relay switch fix also got 100 Mbit

Because this looks so wierd, could you double-check this? Could you comment out the driver load line in /etc/modules.d/gpio-nxp-74hc164 by prefixing a '#' character, i.e. change the content from gpio-74x164 to #gpio-74x164, then reboot, and then report the DSL status?

@majuss
Copy link
Author

majuss commented Dec 17, 2018

Nope exactly the same.

@arnysch
Copy link
Contributor

arnysch commented Dec 17, 2018

the images before the DSL relay switch fix also got 100 Mbit

Guess not loading the driver is like the situation before the relay switch.

Other boxes get 100 Mbit

Guess this is still true for you.

Sorry, I am at a loss now.

@majuss
Copy link
Author

majuss commented Dec 19, 2018

Hm you guess it :D? Can't I revert all chages to fbb617f but implement the disabling of the subpages, so I can check if anything changes?

Btw. I tried the latest Fritz!Box Lab 7490 modem-firmware and it yields in these results:
screenshot 2018-12-19 at 10 22 46

Which modem-firmware are you using?

@arnysch
Copy link
Contributor

arnysch commented Dec 19, 2018

Reverting changes or cloning from fbb617f state, and then apply this patch: guess ;-) this is possible. And it is a good idea, because this could strengthen or dampen the impression that the latest changes indeed broke something. BTW, you also could omit the lcd display driver to make sure it does not impair DSL.

The usage of some of the new gpios which are controlled by kovz' changes is unknown. Currently these are all set to value 1. I don't know what their default value after box init is. Maybe it is 0 and one of these also affect DSL processing. So before you reflash an older version, you could try out
echo 0 > /sys/class/gpio/hc595_0/value
echo 0 > /sys/class/gpio/hc595_1/value
echo 0 > /sys/class/gpio/hc595_3/value
and check if DSL data rate improves (maybe unplug/wait/replug DSL line so re-synchronization is done).
And you could also toggle the /sys/class/gpio/annex/value value; maybe it makes a difference.

BTW, I am using this one. Think it is version 90D02 / 591407 and this is the latest.

@majuss
Copy link
Author

majuss commented Dec 19, 2018

So as a first result, your linked firmware will crash the whole box, as soon as it tries to sync :D

@arnysch
Copy link
Contributor

arnysch commented Dec 20, 2018

Strange. MD5 hash is ade291d69f28f2fdb6b1a4551d6eae9b. Permanently running here in two places in an o2 Box 6431 (ADSL2 annex-b 16 Mb/s, and in a bt homehub 5a (VDSL2 annex-b 50 Mb/s with vectoring).

@majuss
Copy link
Author

majuss commented Dec 27, 2018

Same hash, so it seems that it's not usable with the EB.

How can I disable the LCD driver? Didn't found something fitting in modules.d.

I only have these Pins in gpio when the driver is disabled: export, gpiochip462, unexport.
Datarate did not improved when I enabled the driver and toggled the GPIOs. Even if I set annexa and annexj to 1 and b to 0.
I reverted the dsl_control back to the old state and commented out the GPIO driver, but this did nothing.. So I guess I really need to compile an image?

My self compiled image is crashing after few minutes :( Do I also need to compile an own recovery image?

@majuss
Copy link
Author

majuss commented Jan 2, 2019

So I can contribute two new aspects:

The image from 18.12.2018 is completely broken. I flashed the stock firmware, then the 18.10.2018 recovery and after that the smp image. But this crashes just after 4-5 seconds screen time. @Quallenauge please look into the issue.

I prepared my EB with a modified stock image with enabled ssh and interesting enough, this image is also getting 54/32 Mbit. So maybe my connection has some kind of error, I will call Telekom tomorrow.

@majuss
Copy link
Author

majuss commented Jan 12, 2019

So Telekom slowed my connection down because the DSLAM "sensed" some errors. Now I'am back to 110/37. But the latest image is still causing openWrt to crash directly after boot. It tries to load the screen and then it crashes. After 3-4 crashes the screen goes red and the EB won't boot anymore with the usual message.

I compiled my own image yesterday without all the LCD stuff (because the EB crashed right when it tries to render the network stats) and without lualanes (because it gives a, for me, unsolvable dependency error). And this image is just working flawless.

@suppenkasper0815
Copy link

Hey majuss,

Would you share your Image resource and what do you think about the DSL driver Problem?

White just tried to download the official Firmware of the Vodafone site for this device and it was possible to extract all folders and files so maybe you can't just take this original driver which works on the 100/40 MBit/s Internet accesses well.

I have two boxes one is still running der Original Firmware with stable 95/27, but that's a limitation of my provider. It had one day also 105/36.

I would also love to get the Openwrt-wiki updated how to configure the device correctly.

@majuss
Copy link
Author

majuss commented Jan 29, 2019

Which driver problem do you mean? You can use the DSL-Modem firmware from a FritzBox 7490 which has the exact same DSL-Modem and reaches 100/40 but you can't use the fixed telephone anymore. For the image, an install guide and configuration options see my repo: https://github.com/majuss/easybox904

@suppenkasper0815
Copy link

Hey Majuss, would you share your newest image and development state of this eb904xdsl project? It seems to be a bit dead in the openwrt forum and there is none who wants to update the "TOH-Page" of the EB904. Thank you for your great work! I guess you use the device still with satisfying results. I saw there should be some improvements in 19.07 also for RALINK devices. So maybe the wifi performance and usabilty will improve. ;)

@majuss
Copy link
Author

majuss commented Apr 9, 2020

I was not able to compile an image a week ago and I also don't use the EB anymore, since I upgraded to 250 Mbit 35b.

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
None yet
Projects
None yet
Development

No branches or pull requests

7 participants