connection lost after a while - ubuntu gutsy

Live forum: http://rt2x00.serialmonkey.com/viewtopic.php?t=4347

mrshark

08-10-2007 21:27:16

hi! I've a D-Link G122 Rev.C Usb wifi key, and I use your latest cvs drivers on ubuntu gutsy beta, updated to today... my connection is lost after a while, for example when doing a simple "apt-get update"... following, every info I can give you, hoping to solve soon... i blacklisted ipv6, not needing it, but even with it my connection gets lost... it's attached to a mediacenter, so i do not want other cables... wink

[codeye87vdj9]root@lmce:~# cat /etc/resolv.conf
#nameserver 62.94.0.1
#nameserver 62.94.0.2
nameserver 208.67.222.222
nameserver 208.67.220.220
nameserver 192.168.1.1[/codeye87vdj9]

[codeye87vdj9]root@lmce:~# chattr +i /etc/resolv.conf[/codeye87vdj9]

[codeye87vdj9]root@lmce:~# cat /etc/modprobe.d/aliases|grep ipv6
alias net-pf-10 off #ipv6
alias ipv6 off[/codeye87vdj9]

[codeye87vdj9]root@lmce:~# cat /etc/modprobe.d/blacklist|egrep "rt|ipv6"
blacklist rt73usb
blacklist rt2570
blacklist rt2x00lib
blacklist ipv6[/codeye87vdj9]

[codeye87vdj9]root@lmce:~# uname -a
Linux lmce 2.6.22-13-generic #1 SMP Thu Oct 4 17:18:44 GMT 2007 i686 GNU/Linux[/codeye87vdj9]

[codeye87vdj9]root@lmce:~# cat /etc/issue
Ubuntu 7.10 \n \l[/codeye87vdj9]

[codeye87vdj9]root@lmce:~# iwconfig wlan0
wlan0 RT73 WLAN ESSID:"Shark-WLAN"
Mode:Managed Frequency=2.457 GHz Access Point: 00:13:49:38:C5:6C
Bit Rate=54 Mb/s
RTS thr:off Fragment thr:off
Encryption key:AD32-68DC-AD9E-C359-906F-B39A-25B7-6E68-1B2E-903A-439D-BB32-B6AE-A3D0-B3F1-88D3
Link Quality=86/100 Signal level:-50 dBm Noise level:-79 dBm
Rx invalid nwid:0 Rx invalid crypt:0 Rx invalid frag:0
Tx excessive retries:0 Invalid misc:0 Missed beacon:0[/codeye87vdj9]

[codeye87vdj9]root@lmce:~# ifconfig wlan0
wlan0 Link encap:Ethernet HWaddr 00:1B:11:0A:E4:E0
inet addr:192.168.1.2 Bcast:192.168.1.255 Mask:255.255.255.0
UP BROADCAST RUNNING MULTICAST MTU:1500 Metric:1
RX packets:4969 errors:0 dropped:607 overruns:607 frame:607
TX packets:452 errors:0 dropped:0 overruns:0 carrier:0
collisions:0 txqueuelen:1000
RX bytes:1148903 (1.0 MB) TX bytes:254670 (248.7 KB)[/codeye87vdj9]

[codeye87vdj9]root@lmce:~# iwlist wlan0 scan
wlan0 Scan completed :
Cell 01 - Address: 00:13:49:38:C5:6C
ESSID:"Shark-WLAN"
Mode:Managed
Channel:10
Encryption key:on
Bit Rates:0 kb/s
Cell 02 - Address: 00:18:84:14:58:0A
ESSID:"VA_CCHIU_DE_NA_STIZZA"
Mode:Managed
Channel:11
Encryption key:on
Bit Rates:0 kb/s[/codeye87vdj9]

[codeye87vdj9]root@lmce:~# cat /etc/network/interfaces
auto lo
iface lo inet loopback

auto wlan0
#iface wlan0 inet dhcp
iface wlan0 inet static
address 192.168.1.2
netmask 255.255.255.0
network 192.168.1.0
gateway 192.168.1.1
#wireless-essid Shark-WLAN
pre-up ifconfig wlan0 up
pre-up iwconfig wlan0 essid Shark-WLAN
pre-up iwconfig wlan0 mode Managed
pre-up iwconfig wlan0 channel 10
pre-up iwpriv wlan0 set AuthMode=WPAPSK
pre-up iwpriv wlan0 set EncrypType=TKIP
pre-up iwpriv wlan0 set WPAPSK=xxxxxxxxxxxxxxxxxxxxxx
pre-up iwpriv wlan0 set SSID=Shark-WLAN[/codeye87vdj9]

[codeye87vdj9]root@lmce:~# route -n
Kernel IP routeing table
Destination Gateway Genmask Flags Metric Ref Use Iface
192.168.1.0 0.0.0.0 255.255.255.0 U 0 0 0 wlan0
169.254.0.0 0.0.0.0 255.255.0.0 U 1000 0 0 wlan0
0.0.0.0 192.168.1.1 0.0.0.0 UG 100 0 0 wlan0[/codeye87vdj9]

[codeye87vdj9]root@lmce:~# lspci
00:00.0 RAM memory: nVidia Corporation C51 Host Bridge (rev a2)
00:00.1 RAM memory: nVidia Corporation C51 Memory Controller 0 (rev a2)
00:00.2 RAM memory: nVidia Corporation C51 Memory Controller 1 (rev a2)
00:00.3 RAM memory: nVidia Corporation C51 Memory Controller 5 (rev a2)
00:00.4 RAM memory: nVidia Corporation C51 Memory Controller 4 (rev a2)
00:00.5 RAM memory: nVidia Corporation C51 Host Bridge (rev a2)
00:00.6 RAM memory: nVidia Corporation C51 Memory Controller 3 (rev a2)
00:00.7 RAM memory: nVidia Corporation C51 Memory Controller 2 (rev a2)
00:02.0 PCI bridge: nVidia Corporation C51 PCI Express Bridge (rev a1)
00:03.0 PCI bridge: nVidia Corporation C51 PCI Express Bridge (rev a1)
00:04.0 PCI bridge: nVidia Corporation C51 PCI Express Bridge (rev a1)
00:05.0 VGA compatible controller: nVidia Corporation C51PV [GeForce 6150] (rev a2)
00:09.0 RAM memory: nVidia Corporation MCP51 Host Bridge (rev a2)
00:0a.0 ISA bridge: nVidia Corporation MCP51 LPC Bridge (rev a3)
00:0a.1 SMBus: nVidia Corporation MCP51 SMBus (rev a3)
00:0a.2 RAM memory: nVidia Corporation MCP51 Memory Controller 0 (rev a3)
00:0b.0 USB Controller: nVidia Corporation MCP51 USB Controller (rev a3)
00:0b.1 USB Controller: nVidia Corporation MCP51 USB Controller (rev a3)
00:0d.0 IDE interface: nVidia Corporation MCP51 IDE (rev a1)
00:10.0 PCI bridge: nVidia Corporation MCP51 PCI Bridge (rev a2)
00:10.1 Audio device: nVidia Corporation MCP51 High Definition Audio (rev a2)
00:14.0 Bridge: nVidia Corporation MCP51 Ethernet Controller (rev a3)
00:18.0 Host bridge: Advanced Micro Devices [AMD] K8 [Athlon64/Opteron] HyperTransport Technology Configuration
00:18.1 Host bridge: Advanced Micro Devices [AMD] K8 [Athlon64/Opteron] Address Map
00:18.2 Host bridge: Advanced Micro Devices [AMD] K8 [Athlon64/Opteron] DRAM Controller
00:18.3 Host bridge: Advanced Micro Devices [AMD] K8 [Athlon64/Opteron] Miscellaneous Control
04:05.0 FireWire (IEEE 1394): Texas Instruments TSB43AB22/A IEEE-1394a-2000 Controller (PHY/Link)
04:08.0 Multimedia controller: Philips Semiconductors SAA7133/SAA7135 Video Broadcast Decoder (rev d1)[/codeye87vdj9]

[codeye87vdj9]root@lmce:~# lsusb
Bus 001 Device 001: ID 0000:0000
Bus 002 Device 001: ID 0000:0000
Bus 002 Device 003: ID 0aec:3260 Neodio Technologies Corp. 7-in-1 Card Reader
Bus 002 Device 002: ID 07d1:3c03 D-Link System[/codeye87vdj9]

[codeye87vdj9]root@lmce:~# dmesg |grep -i rt
[ 34.940951] rtusb init ====>
[ 35.152798] usbcore: registered new interface driver rt73
[ 35.188317] rt73 driver version - 1.0.3.6 CVS
[ 35.540556] ***rt73***: Interface goes up for the first time, activating permanent MAC
[ 35.540562] ***rt73***: Active MAC is: 00:1b:11:0a:e4:e0.
[ 526.353984] rt73 driver version - 1.0.3.6 CVS
[ 526.508791] ***rt73***: net_device supplies MAC, activating this one
[ 526.508798] ***rt73***: Active MAC is: 00:1b:11:0a:e4:e0.
[ 566.296295] rt73 driver version - 1.0.3.6 CVS
[ 566.452330] ***rt73***: net_device supplies MAC, activating this one
[ 566.452338] ***rt73***: Active MAC is: 00:1b:11:0a:e4:e0.[/codeye87vdj9]

[codeye87vdj9]root@lmce:~# modinfo rt73
filename: /lib/modules/2.6.22-13-generic/extra/rt73.ko
license: GPL
description: Ralink RT73 802.11abg WLAN Driver 1.0.3.6 CVS 2007100814
author: http://rt2x00.serialmonkey.com
srcversion: 5E2A8500E365E3A7E0ED91E
alias: usb:v1740p7100d*dc*dsc*dp*ic*isc*ip*
alias: usb:v2019pAB50d*dc*dsc*dp*ic*isc*ip*
alias: usb:v2019pAB01d*dc*dsc*dp*ic*isc*ip*
alias: usb:v0769p31F3d*dc*dsc*dp*ic*isc*ip*
alias: usb:v0DF6p90ACd*dc*dsc*dp*ic*isc*ip*
alias: usb:v0DF6p9712d*dc*dsc*dp*ic*isc*ip*
alias: usb:v18E8p6238d*dc*dsc*dp*ic*isc*ip*
alias: usb:v18E8p6229d*dc*dsc*dp*ic*isc*ip*
alias: usb:v18E8p6196d*dc*dsc*dp*ic*isc*ip*
alias: usb:v148Fp2671d*dc*dsc*dp*ic*isc*ip*
alias: usb:v148Fp2573d*dc*dsc*dp*ic*isc*ip*
alias: usb:v0DB0pA874d*dc*dsc*dp*ic*isc*ip*
alias: usb:v0DB0pA861d*dc*dsc*dp*ic*isc*ip*
alias: usb:v0DB0p6874d*dc*dsc*dp*ic*isc*ip*
alias: usb:v0DB0p6877d*dc*dsc*dp*ic*isc*ip*
alias: usb:v13B1p0023d*dc*dsc*dp*ic*isc*ip*
alias: usb:v13B1p0020d*dc*dsc*dp*ic*isc*ip*
alias: usb:v06F8pE020d*dc*dsc*dp*ic*isc*ip*
alias: usb:v06F8pE010d*dc*dsc*dp*ic*isc*ip*
alias: usb:v1472p0009d*dc*dsc*dp*ic*isc*ip*
alias: usb:v1044p800Ad*dc*dsc*dp*ic*isc*ip*
alias: usb:v1044p8008d*dc*dsc*dp*ic*isc*ip*
alias: usb:v15A9p0004d*dc*dsc*dp*ic*isc*ip*
alias: usb:v07D1p3C04d*dc*dsc*dp*ic*isc*ip*
alias: usb:v07D1p3C03d*dc*dsc*dp*ic*isc*ip*
alias: usb:v14B2p3C22d*dc*dsc*dp*ic*isc*ip*
alias: usb:v1371p9032d*dc*dsc*dp*ic*isc*ip*
alias: usb:v1371p9022d*dc*dsc*dp*ic*isc*ip*
alias: usb:v1631pC019d*dc*dsc*dp*ic*isc*ip*
alias: usb:v050Dp905Bd*dc*dsc*dp*ic*isc*ip*
alias: usb:v050Dp705Ad*dc*dsc*dp*ic*isc*ip*
alias: usb:v050Dp7050d*dc*dsc*dp*ic*isc*ip*
alias: usb:v0411p00F4d*dc*dsc*dp*ic*isc*ip*
alias: usb:v0B05p1724d*dc*dsc*dp*ic*isc*ip*
alias: usb:v0B05p1723d*dc*dsc*dp*ic*isc*ip*
alias: usb:v1690p0722d*dc*dsc*dp*ic*isc*ip*
alias: usb:v07B8pB21Dd*dc*dsc*dp*ic*isc*ip*
depends: usbcore
vermagic: 2.6.22-13-generic SMP mod_unload 586
parm: debug:Debug mask: n selects filter, 0 for none (int)
parm: firmName:Permit to load a different firmware: (default: rt73.bin) (charp)[/codeye87vdj9]

[codeye87vdj9]root@lmce:~# lsmod|grep rt73
rt73 214400 0
usbcore 138248 6 usb_storage,libusual,rt73,ehci_hcd,ohci_hcd[/codeye87vdj9]

alex-it

11-10-2007 16:30:30

I report the same issue.

Ubuntu Gutsy and dlink g122 C1.

Debug info similar to mrshark.

Attached a piece of debug trace, hope it helps.

diesel_here

14-10-2007 12:32:20

Same here too. I have a Belkin USB stick, all was fine until a recent update a few days ago.

I have tried to make with the latest driver & kernel but get errors, so I have to use a previous kernel ***.12 as opposed to ****.14 but still get problems with disconnection.

I get GCC errors when trying to make with the latest kernel.

I know this is vague but I'm having trouble staying connected.

herrib

17-10-2007 19:32:46

[b84uua4bu]configuration[/b84uua4bu]

Attached file.

[b84uua4bu]symptoms[/b84uua4bu]

The connection is correctly set up and established. It shuts down about 10 / 15 mn later. No way to start it (boot is then required).

The messages are

[code84uua4bu]Oct 17 20:05:13 herrib-desktop kernel: [ 956.045282] phy0 -> rt2x00usb_vendor_request: Error - Vendor Request 0x06 failed for offset 0x308c with error -110.
Oct 17 20:05:13 herrib-desktop kernel: [ 956.145141] phy0 -> rt2x00usb_vendor_request: Error - Vendor Request 0x07 failed for offset 0x308c with error -110.
---repeated -----
Oct 17 20:05:18 herrib-desktop kernel: [ 960.638263] wlan0: No ProbeResp from current AP 00:11:50:35:f6:9b - assume out of range
Oct 17 20:05:15 herrib-desktop kernel: [ 957.243484] phy0 -> rt2x00usb_vendor_request: Error - Vendor Request 0x07 failed for offset 0x308c with error -110.
Oct 17 20:05:15 herrib-desktop kernel: [ 957.343344] phy0 -> rt2x00usb_vendor_request: Error - Vendor Request 0x06 failed for offset 0x308c with error -110.

---repeated -----
Oct 17 20:05:19 herrib-desktop kernel: [ 961.468963] phy0 -> rt2x00usb_vendor_request: Error - Vendor Request 0x07 failed for offset 0x308c with error -110.
Oct 17 20:05:19 herrib-desktop kernel: [ 961.469072] phy0 -> rt73usb_bbp_read: Error - PHY_CSR3 register busy. Read failed.

--- repeated -----

Oct 17 20:06:58 herrib-desktop kernel: [ 1060.324558] phy0 -> rt73usb_enable_radio: Error - Register initialization failed.[/code84uua4bu]

Vern

18-10-2007 19:34:59

Hi alex-it,

I finally got around to looking at your log. Basically, there's a few iterations of where we successfully authenticate and associate, then respond to the first EAPOL handshake message from our AP, followed by a de-authentication frame from the AP. This happens 14 times. Finally, about 1 1/2 minutes into the log (see line 9661), we get a replay counter mismatch, initiate a disassociation, then apparently get a successful connection(!).

Getting a de-authentication in that context can be taken to mean that the AP does not like something about the second EAPOL message - i.e. our response to EAPOL message 1. So I've attached a small patch that prints the reason code from the deauthentication frame received from the AP. Could you apply it, compile with debug enabled, and run with debug=15? That may give us a little information as to where to look next.

In the meantime, are we dead bang certain that the authentication scheme, encryption scheme, and PSK key value correspond to what the AP has? (Sorry, but I had to ask.)

Also, could you relate what happens around the apparently successful connection? i.e. what behavior are you observing?

(herrib The messages you posted look like they're coming from the rt2x00 driver I have no expertise there.)

Thanks,

Heart

18-10-2007 20:36:58

I also have often have the same errors/problems as herrib mentions... (rt73 chip usb wlan stick on openSUSE 10.3)

alex-it

19-10-2007 23:27:35

First of all, thank you Vern for your help.

I've followed your directive and I'm waiting the error happens again to post the log you've requested.

Meanwhile[quote32cg3kl8]are we dead bang certain that the authentication scheme, encryption scheme, and PSK key value correspond to what the AP has?[/quote32cg3kl8]Yes, I'm sure of that. They worked perfectly under Feisty... roll
[quote32cg3kl8]Also, could you relate what happens around the apparently successful connection? i.e. what behavior are you observing? [/quote32cg3kl8]It's a strange behavior, in effect...
It disconnect suddenly with apparently no reason...
Then I make a couple of "/etc/init.d/networking restart" and then everthing works again... for minutes.

I hope the log I'll post later could help you more...

Thank you!

herrib

20-10-2007 07:45:25

Bugs have been reported in https//bugs.launchpad.net/ubuntu/ #153718, #154651 and #133486.

Failures seem to be correlated with high downloads (such circumstances result in a lost of synchronization ...).

alex-it

20-10-2007 10:13:55

Bugs have been reported in https//bugs.launchpad.net/ubuntu/ #153718, #154651 and #133486.

Failures seem to be correlated with high downloads (such circumstances result in a lost of synchronization ...).[/quote1bvuqe9x]These bug are related to rt2x00 driver (not legacy one), which is included in Ubuntu kernel by default.
This issue is already reported here http//rt2x00.serialmonkey.com/phpBB2/v ... php?t=4364

erlend

24-10-2007 15:55:56

I can report that I have the same problem with rt73usb (1.0.3.6 CVS) drivers compiled on Gentoo against the 2.6.22 kernel.

Sometimes the network goes very very slow and signal strength (which works fine on my laptop in the same room) drops to zero. Normal signal strength is about 80%.

This may or may not help but dmesg sometimes contains the line
wlan0 unable to signal thread


The usb card is a Peak Hardware Peak II Lan Usb adapter.

lsusb gives
Bus 002 Device 002 ID 07b8b21d D-Link Corp

Thanks,

Erlend

Spy84464

24-10-2007 20:54:27

Hello,
Are you using rt2x00 on a 2.6.22 kernel? This would mean you're running an old version, rt2x00 is moving a lot.
Have you tried the legacy driver?

Regards,
Romain

erlend

26-10-2007 09:35:53

[quote2tnazw1m]Are you using rt2x00 on a 2.6.22 kernel?[/quote2tnazw1m]
I'm running a 2.6.22 kernel. I am not using the drivers from that kernel. Instead I downloaded the cvs daily tarball from
http//rt2x00.serialmonkey.com/rt73-cvs-daily.tar.gz

I assume these are the legacy drivers?

Thanks,

Erlend

hennichodernich

26-10-2007 15:40:53

[quote1ln8lhj7]Are you using rt2x00 on a 2.6.22 kernel?[/quote1ln8lhj7]
I'm running a 2.6.22 kernel. I am not using the drivers from that kernel. Instead I downloaded the cvs daily tarball from
http//rt2x00.serialmonkey.com/rt73-cvs-daily.tar.gz

I assume these are the legacy drivers?
[/quote1ln8lhj7]

Yes, but the rt73usb you mentioned definitively is a rt2x00 module. You'll have to blacklist it to explicitely use the legacy driver.

regards
Henning

erlend

26-10-2007 19:04:27

[quote106emab0]Yes, but the rt73usb you mentioned definitively is a rt2x00 module. You'll have to blacklist it to explicitely use the legacy driver. [/quote106emab0]

I don't understand... I install the rt73usb module - i didn't install rt2x00. I just downloaded the file I mentioned earlier and make, make install'ed it against my kernel source tree in Gentoo.

Then i modprobe rt73 and use my internet... it works except for being unreliable and slow.

If I blacklist rt73 I won't be able to use the driver at all...

Starcrasher

26-10-2007 20:38:47

There are three different drivers. The legacy from Ralink, the rt2x00 legacy driver and a new rt2x00 driver.

The module for the rt2x00 legacy driver is named rt73. That's the one you want to use.

Modules for the new rt2x00 driver are named rt73usb rt2x00usb rt2x00lib. You need to blacklist these modules to avoid them to load and conflict.

Is it more clear ?

If you didn't install yourself the new rt2x00 driver, it was certainly already compiled in the kernel shipped by your distribution.

erlend

27-10-2007 01:40:08

There are three different drivers. The legacy from Ralink, the rt2x00 legacy driver and a new rt2x00 driver.

The module for the rt2x00 legacy driver is named rt73. That's the one you want to use.

Modules for the new rt2x00 driver are named rt73usb rt2x00usb rt2x00lib. You need to blacklist these modules to avoid them to load and conflict.

Is it more clear ?

If you didn't install yourself the new rt2x00 driver, it was certainly already compiled in the kernel shipped by your distribution.[/quote30ct9vev]

Ok thanks. I'm running Gentoo, therefore I don't get any drivers with my distribution. I'm running the module named rt73. Downloaded from here http//rt2x00.serialmonkey.com/rt73-cvs-daily.tar.gz

hennichodernich

27-10-2007 08:29:49

I don't understand... I install the rt73usb module - i didn't install rt2x00. I just downloaded the file I mentioned earlier and make, make install'ed it against my kernel source tree in Gentoo.

Then i modprobe rt73 and use my internet... it works except for being unreliable and slow.

If I blacklist rt73 I won't be able to use the driver at all...[/quote94bghinm]

Once more rt73.ko and rt73usb.ko are [b94bghinm]not[/b94bghinm] the same! rt73.ko is the module that is being built when compiling rt73-cvs-daily.tar.gz (or the RaLink version).
rt73usb.ko is the driver from the "new" rt2x00 package. Many distributions ship this driver, so you'll have to blacklist rt73usb if you want to use the self compiled legacy driver rt73.ko!

regards
Henning

erlend

27-10-2007 10:25:16

I don't understand... I install the rt73usb module - i didn't install rt2x00. I just downloaded the file I mentioned earlier and make, make install'ed it against my kernel source tree in Gentoo.

Then i modprobe rt73 and use my internet... it works except for being unreliable and slow.

If I blacklist rt73 I won't be able to use the driver at all...[/quote1bvnkemg]

Once more rt73.ko and rt73usb.ko are [b1bvnkemg]not[/b1bvnkemg] the same! rt73.ko is the module that is being built when compiling rt73-cvs-daily.tar.gz (or the RaLink version).
rt73usb.ko is the driver from the "new" rt2x00 package. Many distributions ship this driver, so you'll have to blacklist rt73usb if you want to use the self compiled legacy driver rt73.ko!

regards
Henning[/quote1bvnkemg]

Thanks for all your patience. I am already using the rt73 (not rt73usb) driver
[code1bvnkemg]
lsmod | grep rt73
rt73 211200 0
firmware_class 7424 2 snd_emu10k1,rt73
[/code1bvnkemg]

I do not have the rt73usb.ko on my system... I only have rt73.ko...

[code1bvnkemg]locate rt73 | grep /lib/modules
/lib/modules/2.6.21-gentoo-r4-4/extra/rt73.ko
/lib/modules/2.6.22-gentoo-r8-1/extra/rt73.ko
[/code1bvnkemg]

erlend

27-10-2007 15:15:59

[quote2ei8rob4]want to use the self compiled legacy driver rt73.ko! [/quote2ei8rob4]
This is the driver I am using, and it is the one I am experiencing problems with.

For example sporadic disconnects and poor connection speed (sometimes, not always e.g. it's fine right now).

Neville

30-10-2007 05:03:53

Hi -

I've been trying to get your drivers to work for weeks, on both a D-link WUA-1340 and an Airlink101 AWLL5025. I just rebuilt and tried to rerun the 10/29 sources.

For both devices, the behavior is the same comes up instantly, looks good, dies the moment I try to transfer more than a few bytes across the link. At that point, I have to bring the interface down quickly or it'll probably take the kernel with it within a few seconds or minutes.

I'm running a Via C3 Nehemiah. It should be x86 "enough" for your drivers in source form, shouldn't it?

Right now, the only reliable wireless link I have is my crummy 802.11b bridge. At least the Rhine driver is solid enough that it runs without any issues.

Should I just get a 54g bridge and say the heck with USB wireless adapters?

(Should mention I'm running Ubuntu Gutsy/7.10)

BlackEdder

29-11-2007 13:58:20

Any news on this? I am using the newest kernel (2.6.24_rc3 on gentoo) with the build in rt73usb driver and have the same problem that the connection gets lost after a while. Also sometimes it is impossible to get a connection at all.

zero_knowledge

30-11-2007 00:08:28

Hi Vern,

just like alex-it and others i experience connection loss with a D-Link DWL-G122.

(...)
Getting a de-authentication in that context can be taken to mean that the AP does not like something about the second EAPOL message - i.e. our response to EAPOL message 1. So I've attached a small patch that prints the reason code from the deauthentication frame received from the AP. Could you apply it, compile with debug enabled, and run with debug=15? That may give us a little information as to where to look next.
(...)[/quote13374mhz]

So i tried to apply your patch but it is rejected. A quick look at auth_rsp.c gave me the impression that this patch is no longer needed. Is this true? Should i just compile with debug enabled and set the debug level to 15?

Regards,
Lothar

Vern

30-11-2007 18:38:13

Hi Zero,

Right. That patch has been put into CVS. So just do the debug=15 bit, and we'll take it from there.

Thanks,

zero_knowledge

02-12-2007 22:17:27

Hi Vern,

thanks for your answer. I was away this weekend, so please excuse the delayed answer. But here it is, along with a brief summary of my situation

I've got a D-Link DWL-G122, hardware version C1, firmware version 3.10. I use it to connect a standard Debian Etch system to the net via WPA. It basically works, but sometimes the connection is lost and can only be re-established after some time or after reboot. The connection loss seems to appear randomly. Sometimes it takes hours, sometimes minutes until the connection is lost. It doesn't seem to be related to any of my actions.

I've compiled the rt73 module with "make debug" and insmodded it with "insmod rt73.ko debug=15". See attached my /var/log/debug, starting from insmodding and ending in the middle of a "connection lost" episode.

uname -a reports
[code3eqe5ro6]Linux apothekerknecht 2.6.18-5-k7 #1 SMP Wed Oct 3 00:47:27 UTC 2007 i686 GNU/Linux[/code3eqe5ro6]

modinfo rt73 reports
[code3eqe5ro6]filename: /lib/modules/2.6.18-5-k7/extra/rt73.ko
author: http://rt2x00.serialmonkey.com
description: Ralink RT73 802.11abg WLAN Driver 1.0.3.6 CVS 2007112916
license: GPL
vermagic: 2.6.18-5-k7 SMP mod_unload K7 REGPARM gcc-4.1
depends: usbcore,firmware_class
alias: usb:v07B8pB21Dd*dc*dsc*dp*ic*isc*ip*
alias: usb:v1690p0722d*dc*dsc*dp*ic*isc*ip*
alias: usb:v0B05p1723d*dc*dsc*dp*ic*isc*ip*
alias: usb:v0B05p1724d*dc*dsc*dp*ic*isc*ip*
alias: usb:v0411p00F4d*dc*dsc*dp*ic*isc*ip*
alias: usb:v050Dp7050d*dc*dsc*dp*ic*isc*ip*
alias: usb:v050Dp705Ad*dc*dsc*dp*ic*isc*ip*
alias: usb:v050Dp905Bd*dc*dsc*dp*ic*isc*ip*
alias: usb:v050Dp905Cd*dc*dsc*dp*ic*isc*ip*
alias: usb:v1631pC019d*dc*dsc*dp*ic*isc*ip*
alias: usb:v1371p9022d*dc*dsc*dp*ic*isc*ip*
alias: usb:v1371p9032d*dc*dsc*dp*ic*isc*ip*
alias: usb:v14B2p3C22d*dc*dsc*dp*ic*isc*ip*
alias: usb:v07D1p3C03d*dc*dsc*dp*ic*isc*ip*
alias: usb:v07D1p3C04d*dc*dsc*dp*ic*isc*ip*
alias: usb:v15A9p0004d*dc*dsc*dp*ic*isc*ip*
alias: usb:v1044p8008d*dc*dsc*dp*ic*isc*ip*
alias: usb:v1044p800Ad*dc*dsc*dp*ic*isc*ip*
alias: usb:v1472p0009d*dc*dsc*dp*ic*isc*ip*
alias: usb:v06F8pE010d*dc*dsc*dp*ic*isc*ip*
alias: usb:v06F8pE020d*dc*dsc*dp*ic*isc*ip*
alias: usb:v13B1p0020d*dc*dsc*dp*ic*isc*ip*
alias: usb:v13B1p0023d*dc*dsc*dp*ic*isc*ip*
alias: usb:v0DB0p6877d*dc*dsc*dp*ic*isc*ip*
alias: usb:v0DB0p6874d*dc*dsc*dp*ic*isc*ip*
alias: usb:v0DB0pA861d*dc*dsc*dp*ic*isc*ip*
alias: usb:v0DB0pA874d*dc*dsc*dp*ic*isc*ip*
alias: usb:v148Fp2573d*dc*dsc*dp*ic*isc*ip*
alias: usb:v148Fp2671d*dc*dsc*dp*ic*isc*ip*
alias: usb:v18E8p6196d*dc*dsc*dp*ic*isc*ip*
alias: usb:v18E8p6229d*dc*dsc*dp*ic*isc*ip*
alias: usb:v18E8p6238d*dc*dsc*dp*ic*isc*ip*
alias: usb:v0DF6p9712d*dc*dsc*dp*ic*isc*ip*
alias: usb:v0DF6p90ACd*dc*dsc*dp*ic*isc*ip*
alias: usb:v0769p31F3d*dc*dsc*dp*ic*isc*ip*
alias: usb:v2019pAB01d*dc*dsc*dp*ic*isc*ip*
alias: usb:v2019pAB50d*dc*dsc*dp*ic*isc*ip*
alias: usb:v1740p7100d*dc*dsc*dp*ic*isc*ip*
parm: firmName:Permit to load a different firmware: (default: rt73.bin) (charp)
parm: ifname:Network device name (default wlan%d) (charp)
parm: debug:Debug mask: n selects filter, 0 for none (int)
[/code3eqe5ro6]

Anyway, thanks to you in advance and also thanks to the team for working on this driver! Please let me know if you need more information (and, of course, if you found the solution D)

Best regards,
Lothar

Vern

03-12-2007 00:05:28

Hi zero,

This thread seems to have at least two problems in it. Wrt yours

Looking at your log, it seems things start out with a "legitimate" problem the AP's beacon is lost for more than a second. However, then the driver decides to zap all the keys - for reasons I don't really understand - and the recovery reassociation attempts fail.

Before I just change it so it doesn't do that, I need to make sure I understand the rationale, or lack thereof.

In the meantime, does simply providing the key again help (trying to verify that I've got the right understanding, here)?

Thanks,

zero_knowledge

03-12-2007 01:35:07

Hi Vern,

[...]
In the meantime, does simply providing the key again help (trying to verify that I've got the right understanding, here)?[/quote3pg07hr6]

The interface is brought up via ifup (Debian specific?). In /etc/network/interfaces i have some lines like
[code3pg07hr6] pre-up ifconfig wlan0 up
pre-up iwconfig wlan0 mode managed
pre-up iwconfig wlan0 essid "bar"
pre-up iwconfig wlan0 ap 00:15:0C:6E:E4:05
pre-up iwpriv wlan0 set AuthMode=WPAPSK
pre-up iwpriv wlan0 set WPAPSK="foooooooooo"[/code3pg07hr6]
Right now i cannot tell exactly what is in there because i'm not sitting at the computer in question. It should be like this, though, but note that the AP listed above is [b3pg07hr6]not[/b3pg07hr6] the one the G122 is trying to talk to, just in case the AP should be the problem and someone tried to identify it by looking at the MAC.

But what i remember is that in the past i tried to do "ifdown wlan0" and "ifup wlan0" over and over again, also "/etc/init.d/networking restart" but most of the time it didn't work. One rare time it actually did work i had put the line with "pre-up iwconfig wlan0 ap 00150C6EE405" (with a different MAC, of course) in the interfaces file before ifupping it. But at the next connection loss ifdown/ifup didn't bring the connection back although i had not changed anything.

So... should have all this ifdown/ifup set the key again and thus the answer to your question is (mostly) no? Or should i only issue "iwpriv wlan0 set WPAPSK=..." without any ifdown/ifup when i experience a connection loss?

Regards,
Lothar

Vern

03-12-2007 02:43:17

Hi zero,

[quoterjlcjwmj]the AP listed above is not the one the G122 is trying to talk to,[/quoterjlcjwmj]
It looks like there may be a conflict in your setup, then. The command
[coderjlcjwmj]iwconfig wlan0 ap 00:15:0C:6E:E4:05[/coderjlcjwmj]
should force the card to register to that access point (see "man iwconfig").

I think
[coderjlcjwmj]iwpriv wlan0 set AuthMode=WPAPSK
iwpriv wlan0 set WPAPSK="foooooooooo"[/coderjlcjwmj]
will do the trick if it's going to work at all for forcing a reassociation.

Thanks,

zero_knowledge

03-12-2007 09:36:34

Hi Vern,

ah, no, the config shown in my previous post belongs to a different computer, with which i tested the rt73 usb stick. The config at the computer which uses the rt73 right now was modified slightly and definitely has the right MAC in its config. I wanted to force it to use the AP.

What i wanted to say is that i think it could be the AP (see below), at least to some part. And as one can see which company this MAC belongs to, i just wanted to make sure that no one thinks that this special MAC/AP in the config in my previous post is causing the problem. I will check which AP model is used tonight.

Maybe i should also mention that in the past, there was a prism2.5 card in use. Everything was fine with the hostap_pci driver. Then my girlfriend moved. At her new (and current) flat with a different AP, the prism2.5 was loosing connection all the time. dmesg reported "replay detected".

So i thought to get rid of this old 11b card and bought the D-Link DWL-G122. Here we are. I chose this topic because you found replays in the log of alex-it and i thought it might be the same for me.

Regards,
Lothar

zero_knowledge

03-12-2007 22:17:02

Hi Vern,

i tried what you said and found strange things. Here is what i did

1. Startet the computer with the wlan stick unplugged
2. [code1waxexyl]rmmod rt73[/code1waxexyl]
(the debug starts from here)
3. [code1waxexyl]modprobe rt73 debug=15[/code1waxexyl]
4. Plugged in the wlan stick
5. [code1waxexyl]apothekerknecht:~# ifup wlan0
Internet Systems Consortium DHCP Client V3.0.4
Copyright 2004-2006 Internet Systems Consortium.
All rights reserved.
For info, please visit http://www.isc.org/sw/dhcp/

Listening on LPF/wlan0/00:1b:11:20:48:62
Sending on LPF/wlan0/00:1b:11:20:48:62
Sending on Socket/fallback
DHCPDISCOVER on wlan0 to 255.255.255.255 port 67 interval 4
DHCPDISCOVER on wlan0 to 255.255.255.255 port 67 interval 7
DHCPDISCOVER on wlan0 to 255.255.255.255 port 67 interval 21
DHCPDISCOVER on wlan0 to 255.255.255.255 port 67 interval 11
ip length 314 disagrees with bytes received 534.
accepting packet with data after udp payload.
DHCPOFFER from 192.168.1.1
DHCPREQUEST on wlan0 to 255.255.255.255 port 67
ip length 314 disagrees with bytes received 534.
accepting packet with data after udp payload.
DHCPACK from 192.168.1.1
bound to 192.168.1.5 -- renewal in 36263 seconds.
[/code1waxexyl]
This ip length disagreeing was also the case with the former prism2.5 and this AP.

6. Tried to ping www.heise.de, worked
7. Waited for some minutes
8. Tried to ping www.heise.de, worked
9. Opened firefox (or should i say iceweasel, it's the version that comes with Debian Etch)
10. Opened http//rt2x00.serialmonkey.com/, it didn't load the whole page, some pictures were missing!
11. Tried to ping www.heise.de, did [b1waxexyl]not[/b1waxexyl] work, connection is lost
12. Tried to provide the WPA key as you said[code1waxexyl]apothekerknecht:~# date && iwpriv wlan0 set AuthMode=WPAPSK && iwpriv wlan0 set WPAPSK="<my correct key here>"
Mo 3. Dez 21:53:56 CET 2007
apothekerknecht:~# dhclient wlan0
Internet Systems Consortium DHCP Client V3.0.4
Copyright 2004-2006 Internet Systems Consortium.
All rights reserved.
For info, please visit http://www.isc.org/sw/dhcp/

Listening on LPF/wlan0/00:1b:11:20:48:62
Sending on LPF/wlan0/00:1b:11:20:48:62
Sending on Socket/fallback
DHCPDISCOVER on wlan0 to 255.255.255.255 port 67 interval 5
DHCPDISCOVER on wlan0 to 255.255.255.255 port 67 interval 12
DHCPDISCOVER on wlan0 to 255.255.255.255 port 67 interval 14
DHCPDISCOVER on wlan0 to 255.255.255.255 port 67 interval 14
DHCPDISCOVER on wlan0 to 255.255.255.255 port 67 interval 16
No DHCPOFFERS received.
No working leases in persistent database - sleeping.
apothekerknecht:~#
[/code1waxexyl]
I put the date command in so you can find this event in my debug somewhat faster.

13. Of course, pinging servers didn't work now

So that's it... i could reproduce this. Connection works, but is lost after visiting some sites. My favorite news site works perfectly, but serialmonkey.com triggers a connection loss. My girlfriend reported that this also happened with some ebay pages.

I was pretty sure that this phenomenon can also be found in a different thread in this forum, but when i searched it now, i could not find it, sorry.

Here is some more info. My /etc/network/interfaces looks like this[code1waxexyl]auto lo wlan0
iface lo inet loopback

iface wlan0 inet dhcp
pre-up ifconfig wlan0 up
pre-up iwconfig wlan0 mode managed
pre-up iwconfig wlan0 essid "WirelessGemeinschaft"
pre-up iwconfig wlan0 ap 00:1B:2F:61:E8:94
pre-up iwpriv wlan0 set AuthMode=WPAPSK
pre-up iwpriv wlan0 set WPAPSK="<my correct key here>"
pre-up iwpriv wlan0 set EncrypType=TKIP
[/code1waxexyl]

The AP MAC is correct, i extra checked this.

The AP is a Netgear WGR614v7 with firmware version V1.1.14_1.1.14GR.

I attached the output of /var/log/debug.

Hope this helps?

Best regards,
Lothar

zero_knowledge

03-12-2007 22:38:37


I was pretty sure that this phenomenon can also be found in a different thread in this forum, but when i searched it now, i could not find it, sorry.
[/quote1szk5or9]

I was wrong. It's here
http//forum.ubuntuusers.de/topic/124431/

Unfortunately, this is in German and it doesn't apply to the rt73 driver but ndiswrapper.

Regards,
Lothar

Vern

04-12-2007 02:31:19

Hi zero,

Looks like I didn't catch the whole story both your logs show a second EAPOL group key message from the AP before the BEACON lost message occurs. In both logs, the second key message occurs 14 minutes or more after the first one.

The 802.11i spec recognizes rekeying as a potoential problem area, but nevertheless, its a valid operation as far as the standard is concerned. While it looks like we respond OK, we stop getting beacons from the AP after a while. (I don't know why.)

While rekeying is a valid operation in the standard, I suspect the driver doesn't update everything it needs to when it happens. I need to look into that some more.

In the meantime, do you know the manufacturer and model of the AP? Does the problem happen with just this AP? This info may help me get a better handle on things.

Wrt. the German language thread wpa_supplicant 0.6.0 may work with the driver; it has the ralink "plugin" included. If you're feeling brave, you might also try that. Looking at the English snippets in the referenced thread, it seems you could set up wpa_supplicant as indicated there.

Thanks,

zero_knowledge

04-12-2007 08:55:57

Hi Vern,

thank you for your answer and the time you spend looking into this!

[quote3cmytctj]In the meantime, do you know the manufacturer and model of the AP?[/quote3cmytctj]
It's a Netgear WGR614v7 with firmware version V1.1.14_1.1.14GR.

[quote3cmytctj]Does the problem happen with just this AP? This info may help me get a better handle on things.[/quote3cmytctj]
I'm not sure. I can check this with a Fritz!Box, but not until the weekend.

[quote3cmytctj]Wrt. the German language thread wpa_supplicant 0.6.0 may work with the driver; it has the ralink "plugin" included. If you're feeling brave, you might also try that. Looking at the English snippets in the referenced thread, it seems you could set up wpa_supplicant as indicated there.[/quote3cmytctj]
Ok, i can try that, but it will also take some time till i get to do this. Debian testing has the 0.6.0 version, so i will try to backport this to Etch. Config should not be a problem.

Best regards,
Lothar

Vern

05-12-2007 02:23:13

Hi zero,

[quoteamcxu2w5]Debian testing has the 0.6.0 version, so i will try to backport this to Etch.[/quoteamcxu2w5]
You may find it simpler to purge the Debian package, download the tarball from the wpa supplicant site, and buld that.

When you type "make install", it puts wpa_supplicant, wpa_passphrase, and wpa_cli into /usr/local/sbin; so its not very intrusive.

Just one man's opinion.

zero_knowledge

06-12-2007 22:34:33

Hi Vern,

thank you for your tip, but as the necessary wpa_supplicant is in Debian testing, it's not that hard. The build dependencies are necessary for both ways, and provided that one has the tools installed and a deb-src line with lenny in the sources.list, it just needs to
[code38n7jplw]apt-get -t lenny --download-only source wpasupplicant
dpkg-source -x wpasupplicant_0.6.0-4.dsc
cd wpasupplicant-0.6.0/[/code38n7jplw]
[EDIT] By default, the Debian package of wpa_supplicant doesn't include the ralink driver. To include it, open the file wpasupplicant-0.6.0/debian/dot.config.mk with your favourite editor and put a line with[code38n7jplw] echo CONFIG_DRIVER_RALINK=y >> $@[/code38n7jplw]
inside, for example after the line starting with "echo CONFIG_DRIVER_NDISWRAPPER=y >> $@". Keep the formatting the same as the NDISWRAPPER line in your RALINK line. Then you can proceed with
[/EDIT]
[code38n7jplw]dpkg-buildpackage -rfakeroot -uc -us
[/code38n7jplw]
as normal user and then dpkg -i the packages. I took this recepy out of the great book "The Debian System" by Martin F. Krafft, so it's not me being a master of Debian magic (and, if anyone chooses to follow this recepy, one should know that i might have missed an important point). If the package had not already been in lenny or sid, i would have done it the make install way.

I've already build the package and tested it with my wlan card ([b38n7jplw]not[/b38n7jplw] rt73), it works. On saturday or sunday i will test it with my girlfriend's rt73 based stick. Then i will also test how the driver behaves when encountered with a different ap than the Netgear.

Do you want me to do this with the same driver version as before or should i download the latest cvs version?

Regards,
Lothar

Vern

07-12-2007 04:44:16

Hi zero,

Using the latest CVS version is probably best, if you can do that. The reason is it corresponds to what I have. Note that now, when you compile to a 2.6 target, "make install" puts the alias into /etc/modprobe.d/ralink.

Wrt the Fritz!Box We had a problem with this about a year ago, which I never was able to satisfactorily resolve (just ask poor MadMax); the manuacturer - AVS, I think - wasn't willing to provide a profile for me to work on, and they could not reproduce the problem we were having. So I guess, go ahead and try it, but I think caution is warranted.

Thanks very much for the lengths you're willing to go to, here,

zero_knowledge

09-12-2007 19:22:03

Hi Vern,

i was finally able to do some tests. I did all tests with my own ap (FRITZ!Box WLAN 3030 (UI), Firmware-Version 21.04.34) instead of the Netgear. Here is what i found

1. Changing the ap didn't help. I experience connection loss in exactly the same fashion.

2. wpa_supplicant 0.6.0 (the Debian Lenny/Testing) version

The package in its original setup didn't include the ralink driver, so i had to repackage/recompile with the ralink driver enabled. I will edit my previous post and tell there how i achieved this.

The debug attached shows the /var/log/debug output created during this experiment. I did the following[codelbto9esf]modprobe rt73 debug=15[/codelbto9esf]
After this the rt73 stick was plugged in, then i started wpa_supplicant. I did not tell wpa_supplicant to go to background so i could see its output[codelbto9esf]apothekerknecht2:~# wpa_supplicant -iwlan0 -Dralink -w -cwpa_supplicant-home.conf
ioctl[SIOCGIWAP]: Transport endpoint is not connected
ioctl[SIOCGIWAP]: Transport endpoint is not connected
ioctl[SIOCGIWAP]: Transport endpoint is not connected
ioctl[SIOCGIWAP]: Transport endpoint is not connected
Trying to associate with SSID 'secret garden'
WPA: Could not find AP from the scan results
WPA: Key negotiation completed with 00:15:0c:6e:e4:05 [PTK=TKIP GTK=TKIP]
CTRL-EVENT-CONNECTED - Connection to 00:00:00:00:00:00 completed (auth) [id=0 id_str=][/codelbto9esf]
Now i was connected and could obtain an ip by issuing[codelbto9esf]dhclient wlan0[/codelbto9esf]
Then i visited some pages with firefox/iceweasel. In [blbto9esf]all[/blbto9esf] cases (meaning also #3 and #4 below) i could browse my favourite news site ([urllbto9esf]http://www.tagesschau.de[/urllbto9esf]) without problems, but as soon as i visited [urllbto9esf]http://rt2x00.serialmonkey.com[/urllbto9esf], the connection was lost. I did this at least ten times with different amounts of browsing before visiting serialmonkey.com, it's absolutely reproducible. rt2x00.serialmonkey.com shows up almost complete, but the links, pictures and colors of the link bar on top are missing.

Soon after the connection loss this appeared in the wpa_supplicant xterm
[codelbto9esf]CTRL-EVENT-DISCONNECTED - Disconnect event - remove keys
ioctl[SIOCGIWAP]: Transport endpoint is not connected
ioctl[SIOCGIWAP]: Transport endpoint is not connected
ioctl[SIOCGIWAP]: Transport endpoint is not connected
ioctl[SIOCGIWAP]: Transport endpoint is not connected
Trying to associate with SSID 'secret garden'
Authentication with 00:00:00:00:00:00 timed out.
Trying to associate with SSID 'secret garden'[/codelbto9esf]
My wpa_supplicant-home.conf looks like this[codelbto9esf]ctrl_interface=/var/run/wpa_supplicant
ctrl_interface_group=0
eapol_version=1
ap_scan=2
fast_reauth=1
network={
ssid="secret garden"
scan_ssid=1
proto=WPA
key_mgmt=WPA-PSK
pairwise=TKIP
psk="<my key here>"
priority=2
}[/codelbto9esf]

3. wpa_supplicant 0.6.1

As it might be that something is different in wpa_supplicant version 0.6.1, i decided to try this version also. So i did the make install you suggested ) Unfortunately, everything was the same as in #2 above. Nevertheless i also attached the debug from this one.

4. without wpa_supplicant

As a last test i didn't use wpa_supplicant, but let the driver handle the wpa stuff. Therefore i did it the Debian way with ifup, here is the relevant section in /etc/network/interfaces[codelbto9esf]iface wlan0 inet dhcp
pre-up ifconfig wlan0 up
pre-up iwconfig wlan0 mode managed
pre-up iwconfig wlan0 essid "secret garden"
pre-up iwpriv wlan0 set AuthMode=WPAPSK
pre-up iwpriv wlan0 set WPAPSK="<my key here>"
pre-up iwpriv wlan0 set EncrypType=TKIP
[/codelbto9esf]
I also attached a debug log. I did this because my ifup way debugs from previous posts ([blbto9esf]not[/blbto9esf] the other debugs in this post) were using the Netgear ap, this time it's the FRITZ!Box WLAN 3030.

Ok, that was it for now. Hope you can find something, and thank you for working on this!

Best regards,
Lothar

Vern

10-12-2007 01:47:52

Hi zero,

Thanks for the regression testing.
The package in its original setup didn't include the ralink driver, so i had to repackage/recompile with the ralink driver enabled. I will edit my previous post and tell there how i achieved this. [/quote2ey5ebmg]
The tarballs I download from [url2ey5ebmg]http://hostap.epitest.fi/releases[/url2ey5ebmg] include the ralink driver in both 0.6.0 and 0.6.1. I looked at the Debian package, and it seems to be a pretty stripped down subset of the "real" wpa_supplicant offering.

Anyway, it sounds like the problem is the same with and without using wpa_supplicant. So maybe just to keep things simple, just choose the setup that is easiest for you to work with, and we'll stick with that.

In the meantime, I'll take a look at your log data and see what shows up.

Thanks again for the work,

Vern

10-12-2007 03:41:05

Hi zero,

I took a look at your latest logs. It looks like the Fritzbox has a different failure signature than the Netgear. Is it possible to use the Netgear box and either the "Debian way" or wpa supplilcant - whichever is more convenient - for future testing?

Before you run more tests, though, let me look at your previous logs again to see if there's more info I've overlooked.

Thanks,

zero_knowledge

10-12-2007 08:29:21

Hi Vern,

The tarballs I download from [url17hz9erq]http://hostap.epitest.fi/releases[/url17hz9erq] include the ralink driver in both 0.6.0 and 0.6.1. I looked at the Debian package, and it seems to be a pretty stripped down subset of the "real" wpa_supplicant offering.[/quote17hz9erq]
The Debian package includes the ralink source, but it is not configured for building. This can be changed by adding a line to a config file, see my previous post about building a backported package.

The 0.6.1 i tested was from [url17hz9erq]http://hostap.epitest.fi/releases[/url17hz9erq]. Thanks to checkinstall i now have it ready as a Debian package also.

So i can easily test any way (ifup, wpa_supplicant 0.6.0 and 0.6.1).

Is it possible to use the Netgear box and either the "Debian way" or wpa supplilcant - whichever is more convenient - for future testing?[/quote17hz9erq]

Yes, i can test with the Netgear box. But maybe i won't be able to do much during the week, because the Netgear box is at my girlfriend's place. Quite understandable, she gets a little bit annoyed when i first sit at the computer to run some tests every time i visit her. So just tell me what to do, i will prepare it and do the actual test at the weekend.

Regards,
Lothar

Vern

22-12-2007 00:56:53

Hi zero,

Looks like I dropped the ball, here.

I've noticed that when the AP does send a new group key, its specifying a different key index value. The resulting presence of more than one group key may be causing problems.

Most of what the attached patch changes is just trying to get more debug messages to actually go to the debug log. The main change is to change the group key message handler to clear out previous group key value(s) when a new one arrives.

Its attached as eapol1.patch.gz. Could you try it?

Merry Christmas!

Skylight

29-12-2007 08:59:11

hi there

I am experiencing similar problems like mentioned in this thread. I am using an Asus WL167-g on Ubuntu 7.10

Didn't work out of the box, but I've tried several tribers including the current and older serialmonkeys. I successfully got a connection and everything runs fine until random connection losses. Then the USB Wifi isn't able to find any access point until I do an ifdown/ifup. I am using WEP by the way.

my config at the moment

modinfo rt73
[codece51mbe8]filename: /lib/modules/2.6.22-14-generic/extra/rt73.ko
license: GPL
description: Ralink RT73 802.11abg WLAN Driver 1.0.3.6 CVS 2007080606
author: http://rt2x00.serialmonkey.com
srcversion: 74B23D4DCB1F2C723945F0D
alias: usb:v2019pAB50d*dc*dsc*dp*ic*isc*ip*
alias: usb:v2019pAB01d*dc*dsc*dp*ic*isc*ip*
alias: usb:v0769p31F3d*dc*dsc*dp*ic*isc*ip*
alias: usb:v0DF6p90ACd*dc*dsc*dp*ic*isc*ip*
alias: usb:v0DF6p9712d*dc*dsc*dp*ic*isc*ip*
alias: usb:v18E8p6229d*dc*dsc*dp*ic*isc*ip*
alias: usb:v18E8p6196d*dc*dsc*dp*ic*isc*ip*
alias: usb:v148Fp2671d*dc*dsc*dp*ic*isc*ip*
alias: usb:v148Fp2573d*dc*dsc*dp*ic*isc*ip*
alias: usb:v0DB0pA874d*dc*dsc*dp*ic*isc*ip*
alias: usb:v0DB0pA861d*dc*dsc*dp*ic*isc*ip*
alias: usb:v0DB0p6877d*dc*dsc*dp*ic*isc*ip*
alias: usb:v13B1p0023d*dc*dsc*dp*ic*isc*ip*
alias: usb:v13B1p0020d*dc*dsc*dp*ic*isc*ip*
alias: usb:v06F8pE020d*dc*dsc*dp*ic*isc*ip*
alias: usb:v06F8pE010d*dc*dsc*dp*ic*isc*ip*
alias: usb:v1472p0009d*dc*dsc*dp*ic*isc*ip*
alias: usb:v1044p800Ad*dc*dsc*dp*ic*isc*ip*
alias: usb:v1044p8008d*dc*dsc*dp*ic*isc*ip*
alias: usb:v15A9p0004d*dc*dsc*dp*ic*isc*ip*
alias: usb:v07D1p3C04d*dc*dsc*dp*ic*isc*ip*
alias: usb:v07D1p3C03d*dc*dsc*dp*ic*isc*ip*
alias: usb:v14B2p3C22d*dc*dsc*dp*ic*isc*ip*
alias: usb:v1371p9032d*dc*dsc*dp*ic*isc*ip*
alias: usb:v1371p9022d*dc*dsc*dp*ic*isc*ip*
alias: usb:v1631pC019d*dc*dsc*dp*ic*isc*ip*
alias: usb:v050Dp905Bd*dc*dsc*dp*ic*isc*ip*
alias: usb:v050Dp705Ad*dc*dsc*dp*ic*isc*ip*
alias: usb:v050Dp7050d*dc*dsc*dp*ic*isc*ip*
alias: usb:v0411p00F4d*dc*dsc*dp*ic*isc*ip*
alias: usb:v0B05p1724d*dc*dsc*dp*ic*isc*ip*
alias: usb:v0B05p1723d*dc*dsc*dp*ic*isc*ip*
alias: usb:v1690p0722d*dc*dsc*dp*ic*isc*ip*
alias: usb:v07B8pB21Dd*dc*dsc*dp*ic*isc*ip*
depends: usbcore
vermagic: 2.6.22-14-generic SMP mod_unload 586
parm: debug:Debug mask: n selects filter, 0 for none (int)
parm: firmName:Permit to load a different firmware: (default: rt73.bin) (charp)[/codece51mbe8]

uname -a
[codece51mbe8]Linux christoph-desktop 2.6.22-14-generic #1 SMP Tue Dec 18 08:02:57 UTC 2007 i686 GNU/Linux[/codece51mbe8]

/etc/network/interfaces
auto lo
iface lo inet loopback

[codece51mbe8]auto wlan0
iface wlan0 inet static
address 192.168.0.6
netmask 255.255.255.0
network 192.168.0.0
gateway 192.168.0.1
pre-up ifconfig wlan0 up
pre-up iwconfig wlan0 essid yyyyyyyy
pre-up iwconfig wlan0 mode Managed
pre-up iwconfig wlan0 channel 8
pre-up iwpriv wlan0 set AuthMode=WEPAUTO
pre-up iwpriv wlan0 set EncrypType=WEP
pre-up iwpriv wlan0 set Key1=xxxxxxx
pre-up iwpriv wlan0 set SSID=yyyyyyy[/codece51mbe8]

cat /etc/modprobe.d/blacklist|egrep "rt|ipv6"
[codece51mbe8]#added when rt73 module was installed
blacklist rt73usb
blacklist rt2570
blacklist rt2x00lib
blacklist rt2500usb[/codece51mbe8]

lsusb
[codece51mbe8]Bus 005 Device 007: ID 0f30:010a Jess Technology Co., Ltd
Bus 005 Device 005: ID 0a12:0001 Cambridge Silicon Radio, Ltd Bluetooth Dongle (HCI mode)
Bus 005 Device 006: ID 0f30:001c Jess Technology Co., Ltd
Bus 005 Device 002: ID 05e3:0608 Genesys Logic, Inc.
Bus 005 Device 003: ID 0b05:1723 ASUSTek Computer, Inc.
Bus 005 Device 001: ID 0000:0000
Bus 002 Device 005: ID 1038:0310 Ideazon, Inc.
Bus 002 Device 001: ID 0000:0000
Bus 001 Device 001: ID 0000:0000
Bus 003 Device 001: ID 0000:0000
Bus 004 Device 001: ID 0000:0000 [/codece51mbe8]

dmesg |grep -i rt
[codece51mbe8][ 0.000000] ACPI: PM-Timer IO Port: 0x808
[ 0.000000] Allocating PCI resources starting at 50000000 (gap: 40000000:bfb80000)
[ 0.000000] Enabling unmasked SIMD FPU exception support... done.
[ 26.587028] virtual kernel memory layout:
[ 30.505304] Booting paravirtualized kernel on bare hardware
[ 30.521441] ACPI: (supports S0 S1 S3 S4 S5)
[ 30.529832] ACPI: PCI Interrupt Routing Table [\_SB_.PCI0._PRT]
[ 30.529941] ACPI: PCI Interrupt Routing Table [\_SB_.PCI0.P0P4._PRT]
[ 30.536209] Linux Plug and Play Support v0.97 (c) Adam Belay
[ 30.540321] PCI: If a device doesn't work, try "pci=routeirq". If it helps, post a report
[ 30.544229] pnp: 00:0b: ioport range 0x680-0x6ff has been reserved
[ 30.544232] pnp: 00:0b: ioport range 0x290-0x297 has been reserved
[ 31.537530] Serial: 8250/16550 driver $Revision: 1.90 $ 4 ports, IRQ sharing enabled
[ 31.542273] serio: i8042 KBD port at 0x60,0x64 irq 1
[ 31.542281] serio: i8042 AUX port at 0x60,0x64 irq 12
[ 31.542738] Using IPI No-Shortcut mode
[ 33.344634] hub 1-0:1.0: 2 ports detected
[ 33.448710] hub 2-0:1.0: 2 ports detected
[ 33.552583] hub 3-0:1.0: 2 ports detected
[ 33.656485] hub 4-0:1.0: 2 ports detected
[ 33.857145] hub 1-1:1.0: 4 ports detected
[ 34.756421] ehci_hcd 0000:00:1d.7: debug port 1
[ 34.756430] PCI: cache line size of 128 is not supported by device 0000:00:1d.7
[ 34.867576] ehci_hcd 0000:00:1d.7: USB 2.0 started, EHCI 1.00, driver 10 Dec 2004
[ 34.867772] hub 5-0:1.0: 8 ports detected
[ 34.997704] sd 2:0:0:0: [sda] Write cache: enabled, read cache: enabled, doesn't support DPO or FUA
[ 34.997812] sd 2:0:0:0: [sda] Write cache: enabled, read cache: enabled, doesn't support DPO or FUA
[ 35.337190] hub 1-1:1.0: hub_port_status failed (err = -71)
[ 35.338188] hub 1-1:1.0: hub_port_status failed (err = -71)
[ 35.339183] hub 1-1:1.0: hub_port_status failed (err = -71)
[ 35.340186] hub 1-1:1.0: hub_port_status failed (err = -71)
[ 35.343234] swsusp: Resume From Partition 8:4
[ 35.381129] kjournald starting. Commit interval 5 seconds
[ 35.969066] hub 5-1:1.0: 4 ports detected
[ 37.392395] hub 5-1.2:1.0: 3 ports detected
[ 43.240049] parport_pc 00:0a: reported by Plug and Play ACPI
[ 43.240149] parport0: PC-style at 0x378 (0x778), irq 7, dma 3 [PCSPP,TRISTATE,COMPAT,EPP,ECP,DMA]
[ 43.276544] iTCO_vendor_support: vendor-support=0
[ 43.322643] Linux agpgart interface v0.102 (c) Dave Jones
[ 43.324919] agpgart: Detected an Intel 865 Chipset.
[ 43.328432] agpgart: AGP aperture is 64M @ 0xf8000000
[ 43.571028] iTCO_wdt: initialized. heartbeat=30 sec (nowayout=0)
[ 43.579233] gameport: EMU10K1 is pci0000:02:09.1/gameport0, io 0xdfe0, speed 1147kHz
[ 43.740088] rtusb init ====>
[ 43.994152] usbcore: registered new interface driver rt73
[ 44.313149] rt73 driver version - 1.0.3.6 CVS
[ 44.540267] ***rt73***: Interface goes up for the first time, activating permanent MAC
[ 44.540275] ***rt73***: Active MAC is: 00:1b:fc:df:2b:6b.
[ 45.042043] lp0: using parport0 (interrupt-driven).
[ 48.729488] ppdev: user-space parallel port driver
[ 52.366934] agpgart: Found an AGP 3.0 compliant device at 0000:00:00.0.
[ 52.366952] agpgart: Putting AGP V3 device at 0000:00:00.0 into 8x mode
[ 52.366984] agpgart: Putting AGP V3 device at 0000:01:00.0 into 8x mode[/codece51mbe8]

I would be glad for any help, and please tell me if you need any more information

thx

Spy84464

29-12-2007 15:39:53

Hello,
It would be useful if you could post a debug trace right after a disconnection, so we can see why it happened. You'll find the relevant information in the "TESTING" file, coming with the driver.

Regards,
Romain

Skylight

30-12-2007 13:51:42

thanks for your answer

i did that, but my /var/log/ directory has now 5 GB and the debug file about 700 MB
I will try to get the necessary information out of it
is there a way to reduce the amount of storage?

Skylight

30-12-2007 17:12:33

do you need just the /var/log/debug or kern.log, messages and syslog also?

the zipped debug file still has 30MB which i can not upload here

is that file size normal?

Spy84464

01-01-2008 13:37:53

It is normal, but you don't need to upload the whole file. The last, say 1000, lines would suffice I guess.

mattaw

07-01-2008 02:15:35

I have the same problem with rt73-cvs-2008010618 and your EAPOL patch applied (it applied cleanly) on Mythdora 4 (2.6.20).

The wifi environment is a pair of 3com 3CRWDR100A-72 ADSL router/AP b/g doing Wireless Distribution (WDS) with WPA PSK.

The card is an Edimax USB key. I get connectivity with Apple and edimax under windows.

Before the patch I got connectivity in the order of 30s to 3mins. After the patch this seems to have improved a bit.

I load the connection by a 200kbps download, which should be trivial.

Attached is the debug log from the intial modprobe of rt73 debug=31 to the loss of connection.

Thanks for your efforts!

Matthew

zero_knowledge

07-01-2008 21:30:19

Hi Vern and others,

finally i too was able to run the test with eapol1.patch applied. I attached 3 logs, two with debug=15 and one with debug=31 (which one should we use?).

debug-15--2 shows what happens right after the connection loss in debug-15--1. I just rmmodded rt73 and loaded it again, followed by ifdown/ifup (dhclient got no ip address).

Best regards,
Lothar

mattaw

10-01-2008 08:27:03

Is there anything more we can do while we wait for you to have enough free time to look?

What is the difference in the debug=31 and the debug=15 logs?

Matthew

alex-it

10-01-2008 10:52:42

Hi guys,
I've to tell you I've finally found a way to have rt73 working on ubuntu gutsy (and kernel < 2.6.24).

I just upgraded rt73usb module (not the legacy, but the [u2nwd0q9j]rt2x00[/u2nwd0q9j] one) compiling and installing the "[i2nwd0q9j]linux wireless compatibility package[/i2nwd0q9j]" found there http//linuxwireless.org/en/users/Download

Now I'm using version 2.0.14 and problems are gone wink

Skylight

10-01-2008 17:35:19

I finally was able to get a log of a disconnection

this one happens regularly to me when unmounting and disconnecting an external usb disk

it happens around 1828

thanks for your help

Vern

10-01-2008 20:48:58

Hi Skylight,

Thanks for stepping up with a debug log. Unfortunately, it looks like something is whacky there almost every line repeats. Could you check your /etc/syslog.conf settings and make sure messages with debug priority are correctly routed to /var/log/debug?

That said, after stripping out the duplicates, it looks like we're filling up the transmit queue. That this happens in the neighborhood of disconnecting a usb disk is very interesting.

First, what symptoms do you observe (in terms of console messages, etc.) when this happens?

Second, can you provide your usb topology - i.e. how the disk and wifi adapter are hooked up to your PC?

Finally, because this is really a different problem than what is being dealt with in this topic, and because it's *definitely* non- trivial to solve, could you answer by starting a new topic, with an appropriately descriptive title, so that we can deal with this problem by itself?

Thanks in advance (and - I'm afraid - don't expect quick results),

Skylight

11-01-2008 17:33:35

thanks Vern!

I think I have found the bug for the duplicates

at first I got interested in this thread because I also had random disconnects
but then by using the latest serialmonkey drivers, editing the /etc/network/interfaces and using Rutilt, I guess they disappeared and left me with the disconnects when unplugging the usb disk

I will start a new thread the the necessary information soon

thx