rt61 on a brand new gutsy install not working

Live forum: http://rt2x00.serialmonkey.com/viewtopic.php?t=4426

hogbender

10-11-2007 11:04:44

Hello all,

I had some problems with changing a wifi card over on an existing ubuntu gutsy install, so I decided that I'd reinstall the os and see if my problems were caused by my config.

On a stock ubuntu gutsy machine (can't connect to the net over wireless, so I KNOW it's got nothing else -) my wireless isn't working. Not with WPA or with straight WEP.

My box is an amd64 machine with a D-Link DWL-G510 network card. The details of lspci and lsmod are below

[code3cbzhaor]$ lsmod |grep rt
rt61pci 28288 0
rt2x00pci 13184 1 rt61pci
rt2x00lib 21760 2 rt61pci,rt2x00pci
rfkill 9616 1 rt2x00lib
mac80211 196104 4 rc80211_simple,rt61pci,rt2x00pci,rt2x00lib
input_polldev 6672 1 rt2x00lib
crc_itu_t 3456 1 rt2x00lib
parport_pc 41896 1
parport 44172 3 ppdev,lp,parport_pc
eeprom_93cx6 3712 1 rt61pci[/code3cbzhaor]
[code3cbzhaor]
$ lspci |grep RT61
03:05.0 Network controller: RaLink RT2561/RT61 rev B 802.11g[/code3cbzhaor]

Any ideas?

Spy84464

10-11-2007 11:19:01

Hello,
Is it the driver shipped by default with Ubuntu?

Regards,
Romain

hogbender

10-11-2007 11:29:46

Hello,
Is it the driver shipped by default with Ubuntu?

Regards,
Romain[/quote2kvfjl6n]

Hey Romain,

Yes, it is the default driver. The only source I had was the Gutsy CD - I didn't use anything external to that.

Regards,

hogbender

AdamBaker

10-11-2007 17:21:22

In that case there are 2 likely problems and we'd need to see dmesg output to see which it is.

1) The CD may not include the firmware - I know Gutsy does install firmware if it has a network connection but I don't know if it is on the CD.

2) You might be suffering from the "non-free entry in the non-full queue" problem. This seems to affect some machines more than others but it usually occurs pretty quickly for me with either the stock gutsy drivers or the latest git kernel. If I unplug the card and plug it back in a couple of times I sometimes get a connection that works for a while. There is a patch around to fix this (since Wednesday) but it probably isn't easy to apply to the Ubuntu kernel.

hogbender

11-11-2007 02:55:47

Hey there,

Filtered dmesg output is below. Is this enough info? Do you want my whole dmesg dump? This is all that relates to the wifi card afaik.

[code1q2alfta][ 32.156706] ACPI: PCI Interrupt 0000:03:05.0[A] -> GSI 20 (level, low) -> IRQ 20
[ 32.227053] ieee80211_init: failed to initialize WME (err=-17)
[ 32.238584] iwlwifi_rc80211_simple: Unknown symbol iwlwifi_ieee80211_rate_control_unregister
[ 32.238649] iwlwifi_rc80211_simple: Unknown symbol iwlwifi_sta_info_put
[ 32.238703] iwlwifi_rc80211_simple: Unknown symbol iwlwifi_sta_info_get
[ 32.238779] iwlwifi_rc80211_simple: Unknown symbol iwlwifi_ieee80211_rate_control_register
[ 32.242814] wmaster0: Selected rate control algorithm 'simple'
[ 32.299679] ACPI: PCI Interrupt 0000:00:14.2[A] -> GSI 16 (level, low) -> IRQ 16
[/code1q2alfta]

Also, it's worth noting that I can do an iwlist scan without issues and ifconfig reports that the interface is up... it's detected as wlan0.

Regards,

hogbender

IvD

11-11-2007 12:22:31

You seem to be using the iwlwifi driver instead of rt61pci...

hogbender

11-11-2007 13:35:16

This is really strange... I've blacklisted the iwlwifi modules, I've confirmed that they aren't loaded through lsmod, and (most importantly) I don't have an Intel wireless card.

I must admit that when I read the iwlwifi lines I thought that it was a generic reference to wireless, not another module. Hrm.

Looking through the web, there are other references to people with rt61 cards on gutsy with the iwlwifi error
https//bugs.launchpad.net/ubuntu/+bug/144239

Still trying to nut this one out. anyone got any suggestions of output I could go through that would make this easier to figure out?

iwlist scan returns access points...
iwconfig essid AP-NAME works ok...
iwconfig report the access point associated...
zero response from a dhclient...

I've opened up the access point to no encryption. My windows laptop works without any issues. Weird.

hogbender

11-11-2007 13:56:56

I've removed the directory with the iwlwifi modules. Rebooted, and have now got the following dmesg output. Any ideas?

[quote11bxu2uj][ 23.758160] ACPI PCI Interrupt 00000014.1[A] -> GSI 16 (level, low) -> IRQ 16
[ 23.758171] SB600_PATA chipset revision 0
[ 23.758174] SB600_PATA not 100% native mode will probe irqs later
[ 23.758182] ide0 BM-DMA at 0xf900-0xf907, BIOS settings hdapio, hdbpio
[ 23.758193] Probing IDE interface ide0...
[ 23.872217] Attempting manual resume
[ 23.872221] swsusp Resume From Partition 85
[ 23.872223] PM Checking swsusp image.
[ 23.872406] PM Resume from disk failed.
[ 23.925483] kjournald starting. Commit interval 5 seconds
[ 23.925497] EXT3-fs mounted filesystem with ordered data mode.
[ 24.339021] usb 1-2 new low speed USB device using ohci_hcd and address 3
[ 24.351197] ACPI PCI Interrupt 00000307.0[A] -> GSI 22 (level, low) -> IRQ 22
[ 24.402329] ACPI PCI Interrupt 00000013.3 -> GSI 17 (level, low) -> IRQ 17
[ 24.402384] ohci_hcd 00000013.3 OHCI Host Controller
[ 24.402440] ohci_hcd 00000013.3 new USB bus registered, assigned bus number 5
[ 24.402460] ohci_hcd 00000013.3 irq 17, io mem 0xfe02b000
[ 24.404548] ohci1394 fw-host0 OHCI-1394 1.1 (PCI) IRQ=[22] MMIO=[fdfff000-fdfff7ff] Max Packet=[2048] IR/IT contexts=[4/8]
[ 24.462968] usb usb5 configuration #1 chosen from 1 choice
[ 24.462996] hub 5-01.0 USB hub found
[ 24.463007] hub 5-01.0 2 ports detected
[ 24.551104] usb 1-2 configuration #1 chosen from 1 choice
[ 24.570772] ACPI PCI Interrupt 00000013.4[C] -> GSI 18 (level, low) -> IRQ 18
[ 24.570826] ohci_hcd 00000013.4 OHCI Host Controller
[ 24.570880] ohci_hcd 00000013.4 new USB bus registered, assigned bus number 6
[ 24.570899] ohci_hcd 00000013.4 irq 18, io mem 0xfe02a000
[ 24.634744] usb usb6 configuration #1 chosen from 1 choice
[ 24.634772] hub 6-01.0 USB hub found
[ 24.634784] hub 6-01.0 2 ports detected
[ 25.684995] ieee1394 Error parsing configrom for node 0-001023
[ 25.685083] ieee1394 Host added IDBUS[0-011023] GUID[0011d8000106b759]
[ 31.172095] pci_hotplug PCI Hot Plug PCI Core version 0.5
[ 31.173901] shpchp Standard Hot Plug PCI Controller Driver version 0.4
[ 31.969062] piix4_smbus 00000014.0 Found 00000014.0 device
[ 32.032778] usbcore registered new interface driver hiddev
[ 32.037306] input Logitech USB Receiver as /class/input/input1
[ 32.037353] input USB HID v1.10 Keyboard [Logitech USB Receiver] on usb-00000013.0-2
[ 32.045040] /build/buildd/linux-source-2.6.22-2.6.22/drivers/hid/usbhid/hid-core.c Fixing up Logitech keyboard report descriptor
[ 32.045632] input Logitech USB Receiver as /class/input/input2
[ 32.045753] input,hiddev96 USB HID v1.10 Mouse [Logitech USB Receiver] on usb-00000013.0-2
[ 32.045770] usbcore registered new interface driver usbhid
[ 32.045774] /build/buildd/linux-source-2.6.22-2.6.22/drivers/hid/usbhid/hid-core.c v2.6USB HID core driver
[ 32.425867] input PC Speaker as /class/input/input3
[ 32.546849] parport_pc 000a reported by Plug and Play ACPI
[ 32.546885] parport0 PC-style at 0x378, irq 7 [PCSPP,TRISTATE,EPP]
[ 32.586105] ACPI PCI Interrupt 00000305.0[A] -> GSI 20 (level, low) -> IRQ 20
[ 32.642852] wmaster0 Selected rate control algorithm 'simple'
[ 32.822552] ACPI PCI Interrupt 00000014.2[A] -> GSI 16 (level, low) -> IRQ 16
[ 33.033102] hda_codec Unknown model for ALC883, trying auto-probe from BIOS...
[ 33.941666] lp0 using parport0 (interrupt-driven).
[ 34.018611] Adding 6040400k swap on /dev/sda5. Priority-1 extents1 across6040400k
[ 34.443901] EXT3 FS on sda1, internal journal
[ 35.428128] input Power Button (FF) as /class/input/input4
[ 35.432752] ACPI Power Button (FF) [PWRF]
[ 35.460307] input Power Button (CM) as /class/input/input5
[ 35.464892] ACPI Power Button (CM) [PWRB]
[ 35.475490] No dock devices found.
[ 35.787807] powernow-k8 Found 1 AMD Sempron(tm) Processor 3200+ processors (version 2.00.00)
[ 35.787841] powernow-k8 0 fid 0xa (1800 MHz), vid 0x6
[ 35.787844] powernow-k8 1 fid 0x2 (1000 MHz), vid 0x12
[ 36.815167] ppdev user-space parallel port driver
[ 36.941732] audit(1194788666.2403) type=1503 operation="inode_permission" requested_mask="a" denied_mask="a" name="/dev/tty" pid=4819 profile="/usr/sbin/cupsd"
[ 37.016016] r8169 eth0 link down
[ 37.246585] Failure registering capabilities with primary security module.
[ 37.424461] Bluetooth Core ver 2.11
[ 37.424561] NET Registered protocol family 31
[ 37.424563] Bluetooth HCI device and connection manager initialized
[ 37.424567] Bluetooth HCI socket layer initialized
[ 37.436071] Bluetooth L2CAP ver 2.8
[ 37.436076] Bluetooth L2CAP socket layer initialized
[ 37.481115] Bluetooth RFCOMM socket layer initialized
[ 37.481187] Bluetooth RFCOMM TTY layer initialized
[ 37.481189] Bluetooth RFCOMM ver 1.8
[ 40.860647] Marking TSC unstable due to cpufreq changes
[ 40.862867] Time hpet clocksource has been installed.
[ 44.340313] hda-intel Invalid position buffer, using LPIB read method instead.
[ 77.813213] NET Registered protocol family 17
[ 78.321166] wlan0 Initial auth_alg=0
[ 78.321173] wlan0 authenticate with AP 00184d5e7bf0
[ 78.321856] wlan0 RX authentication from 00184d5e7bf0 (alg=0 transaction=2 status=0)
[ 78.321861] wlan0 authenticated
[ 78.321865] wlan0 associate with AP 00184d5e7bf0
[ 78.323064] wlan0 RX AssocResp from 00184d5e7bf0 (capab=0x461 status=0 aid=2)
[ 78.323068] wlan0 associated
[ 78.323126] wlan0 WMM queue=2 aci=0 acm=0 aifs=3 cWmin=15 cWmax=1023 burst=0
[ 78.323130] wlan0 WMM queue=3 aci=1 acm=0 aifs=7 cWmin=15 cWmax=1023 burst=0
[ 78.323133] wlan0 WMM queue=1 aci=2 acm=0 aifs=2 cWmin=7 cWmax=15 burst=30
[ 78.323136] wlan0 WMM queue=0 aci=3 acm=0 aifs=2 cWmin=3 cWmax=7 burst=15
[ 78.358769] wlan0 CTS protection enabled (BSSID=00184d5e7bf0)
[ 103.598746] wlan0 disassociate(reason=3)[/quote11bxu2uj]

AdamBaker

11-11-2007 22:55:32

It looks like the gutsy driver is built without debugging support so the firmware loading messages aren't in the dmesg output but the fact that scanning and association work indicate that it has loaded the firmware.

There was a bug in rt2x00 a while back that caused problems with DHCP packets, it is possible the Gutsy driver dates from that time - have you tried manual IP address configuration?

The disassociation code given in that log doesn't help much, reason code 3 is "Deauthenticated because sending STA is leaving (or has left) IBSS or ESS".

hogbender

12-11-2007 23:58:06

It looks like this is partially fixed... I updated my gutsy install with the online repositories and also installed the RutilT app. network-manager can't get it up, but rutilt does it like a charm. Wicd doesn't work either.

So there's still got to be something awry with things since all reports on gutsy and rt61 are that it usually works with the supplied tools. I'm mystified why my install would be any different to the hundreds of other people out there running the same card on the same distro.

I'll update the drivers through the crazy git update mojo I've read about. Maybe that'll fix it proper.

Thanks to all those that responded!

Regards,

hogbender.

Jo2003

20-11-2007 11:03:05

This was the same for me. I installed gutsy and the network manager only shows WEP encrypted networks but not my own WPA encrypted. Using the legacy rt61 driver (a fresh CVS compile) my network was shown ... but I can't get it up. After I installed Rutilt all works well. I now placed the rutilt call inside the /etc/network/interfaces file and my network comes up on boot without any problem.

The only tricky thing was to get the packages onto the PC because network not worked ... A binary of Rutilt would be nice ... but there are that much distros out there, that this will be a problem ...

Best regards,
Jo

hogbender

20-11-2007 13:19:04

Hey list,

I am not usually a person that whinges.... honestly! -) I really felt that I gave this combination of hardware/drivers/distro a go, but it just isn't working. I don't know if it's the Ubuntu people that have stuffed up a perfectly good driver, or if the rt61pci driver isn't ready for prime time, but I gave up.

Partially out of stubbornness, and partially out of morbid curiosity, I swapped my rt61pci-based card with another rt2x00 chipset. The rt2500 driver works a lot better. All I had to do was disable the network-manager and hard-code a static entry into the interfaces file to make it work. God help me if I want to connect to multiple wireless networks conveniently...

Is it just me, or are the new rt2x00 drivers not exactly ready for the mainstream? I'm no slouch with linux and I had real issues. Do you really think that distros like ubuntu should be shipping this driverset yet?

Regards,

Hogbender.

IvD

20-11-2007 19:03:06


Is it just me, or are the new rt2x00 drivers not exactly ready for the mainstream? I'm no slouch with linux and I had real issues. Do you really think that distros like ubuntu should be shipping this driverset yet?
[/quote293x7nsg]

In my opinion several distributions including Ubuntu decided to ship the rt2x00 drivers. This results in the fact that those distributions are now using quite old versions of the drivers and cannot upgrade to newer version which should work better because of kernel incompatibility.

On the other hand, rt2x00 is being pushed into 2.6.24 to make it more accessible and allow for a broader testing audience. Even though the rt2x00 version inside 2.6.24 will not be too great either. (

Spy84464

25-11-2007 20:14:14

Hello,
A binary of Rutilt would be nice[/quoteo59vh40a]
There is a package in both Ubuntu and Debian repositories. I don't know about the others.

Regards,
Romain

Daniel_Vasseur

26-11-2007 13:11:02

Hello,

no need of RutilT rpm for Fedora 8 because rt61pci 2.0.12 driver is working fine "out of the box" with my Hercules HWGPCI-54 PCI adapter (even firmware is included).

Regards,
Daniel

Spy84464

26-11-2007 23:42:31

Yes, RutilT is not compulsory at all, it just here to make it easier to configure the device. )