Problems with Dlink DWL-G510 [SuSE 11.0]

bjohanns · September 27, 2008, 1:15pm

Dear reader,

I am still somewhat new to linux and could not solve my problems completely with “search” and “google”.

I still have some problems with my DWL-G510 which I installed some days ago.

Some basic system data:

uname -r = 2.6.25.16-0.1-default

lspci -vv (partial)
03:02.0 Network controller: RaLink RT2561/RT61 rev B 802.11g
        Subsystem: D-Link System Inc DWL-G510 Rev C
        Control: I/O+ Mem+ BusMaster+ SpecCycle- MemWINV+ VGASnoop- ParErr- Stepping- SERR+ FastB2B- DisINTx-
        Status: Cap+ 66MHz- UDF- FastB2B- ParErr- DEVSEL=slow >TAbort- <TAbort- <MAbort- >SERR- <PERR- INTx-
        Latency: 32, Cache Line Size: 64 bytes
        Interrupt: pin A routed to IRQ 22
        Region 0: Memory at febf8000 (32-bit, non-prefetchable) [size=32]
        Capabilities: [40] Power Management version 2
                Flags: PMEClk- DSI- D1- D2- AuxCurrent=0mA PME(D0-,D1-,D2-,D3hot-,D3cold-)
                Status: D0 PME-Enable- DSel=0 DScale=0 PME-
        Kernel driver in use: rt61
        Kernel modules: rt61pci, rt61

iwconfig (partial)

ra0       RT61 Wireless  ESSID:"WLAN-001C4AA4B091"  Nickname:""
          Mode:Managed  Frequency:2.427 GHz  Access Point: 00:1C:4A:A4:B0:91
          Bit Rate=54 Mb/s
          RTS thr:off   Fragment thr:off
          Encryption key:xxxx-xxxx-xxxx-xxxx-xxxx-xxxx-xxxx-xxxx
          Link Quality=87/100  Signal level:-52 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

ifconfig ra0

ra0       Link encap:Ethernet  Hardware Adresse 00:1E:58:AC:57:49
          inet Adresse:192.168.2.104  Bcast:192.168.2.255  Maske:255.255.255.0
          inet6 Adresse: fe80::21e:58ff:feac:5749/64 Gültigkeitsbereich:Verbindung
          UP BROADCAST MULTICAST  MTU:1500  Metric:1
          RX packets:824358 errors:0 dropped:0 overruns:0 frame:0
          TX packets:347029 errors:1069 dropped:1069 overruns:0 carrier:0
          collisions:6434 Sendewarteschlangenlänge:1000
          RX bytes:482873513 (460.5 Mb)  TX bytes:24178554 (23.0 Mb)
          Interrupt:22

lsmod | grep rt

parport_pc             59592  0
rtc_cmos               28728  0
parport                60448  2 ppdev,parport_pc
rtc_core               41012  1 rtc_cmos
rt61pci                42496  0
rtc_lib                20224  1 rtc_core
rt2x00pci              28800  1 rt61pci
rt2x00lib              43136  2 rt61pci,rt2x00pci
firmware_class         27776  1 rt2x00lib
rfkill                 25888  1 rt2x00lib
input_polldev          22416  1 rt2x00lib
crc_itu_t              18944  2 udf,rt2x00lib
mac80211              220568  2 rt2x00pci,rt2x00lib
eeprom_93cx6           19328  1 rt61pci
rt61                  266340  1

rpm -q -a | grep -i -e ralink -e wireless -e network -e lan | grep -v lang

wireless-tools-29-24.1
NetworkManager-kde-0.7r821737-0.3
kdenetwork3-news-3.5.9-39.1
kdenetwork4-filesharing-4.0.4-19.1
phalanx-22-620.1
libredland0-1.0.7-16.1
kdenetwork3-3.5.9-39.1
irqbalance-0.55-88.1
vlan-1.9-90.1
xalan-j2-2.7.0-188.1
ralink-firmware-1.0-25.1
NetworkManager-0.7.0.r3685-7.2
kdenetwork3-wireless-3.5.9-39.1
yast2-network-2.16.48-3.1
NetworkManager-glib-0.7.0.r3685-7.2

after a long fight with SuSE which did not recognize the wlan card - i compiled the rt61 driver from ralink and put it into the system according to ralinks README.

This finally brought the wlan (ra0) up and I was able to configure it through yast and connect to google and others.
I notice, that the “original” ralink description feels somewhat “out of place” for SuSE as it seems that SuSE uses some other locations and filenames to provide the stuff for the whole thing.
But somehow all fell into place.

But now I am experiencing trouble with transfer stability and speed. Ping reports a packet loss ratio of 30% and downloading data from the internet is tedious. I face transmission rates down to 300 B/s (according to yast update download dialog).

There are some system messages which I do not understand:

Sep 27 02:11:07 tommy kernel: 46:3e:80:1b:60:a2:95:3c:58:40:91:2a:17:6b:53:eb:
Sep 27 02:11:07 tommy kernel: 6d:8c:28:54:c8:2c:0b:07:
Sep 27 02:11:07 tommy kernel: ba:88:84:3d:d1:a4:db:a6:

and there are several firewall error messages in dmesg:


SFW2-OUT-ERROR IN= OUT=ra0 SRC=192.168.2.104 DST=195.135.221.130 LEN=52 TOS=0x00 PREC=0x00 TTL=64 ID=55642 DF PROTO=TCP SPT=7070 DPT=80 WINDOW=57 RES=0x00 ACK FIN URGP=0 OPT (0101080A00E292E251E307B5)

I have an XP-box side by side with my unix which works fine (other WLAN-Card which I could not get to work with unix). I even swapped place for both machines to make sure that the location is not the primary factor for the different behaviour.

The wlan router is only ~5 m away and plain in sight.

Could anybody give some hints what could be wrong?

Greetings
Bernd[/size]

cookdav · September 27, 2008, 1:56pm

Where are you located, and what are you pinging?
[e.g. pinging Google is always flawless for me, but beyond that, all bets are off.]

>>Ping reports a packet loss ratio of 30%.
That can be normal (e.g. when I ping a site in netherlands (from here
in Florida), I get 50% loss, yet their website works flawlessly)

>>I face transmission rates down to 300 B/s (according to yast update download dialog).

Those are totally normal…those sites just get loaded down. (Try different times
of day.)

Try comparing numbers from a site such as:
Speakeasy - Speed Test

Try turning off IPV6, as that can slow down browsers!

You may well HAVE some sort of issue, but your evidence is unconvincing
from those two examples.

cookdav · September 27, 2008, 2:10pm

[Sorry for double-post…editing responses sux.]

Where are you located, and what are you pinging?
[e.g. pinging Google is always flawless for me, but beyond that, all bets are off.]

>>Ping reports a packet loss ratio of 30%.
That can be normal (e.g. when I ping a site in netherlands (from here
in Florida), I get 50% loss, yet their website works flawlessly)
Linux wireless LAN support http://linux-wless.passys.nl

>>I face transmission rates down to 300 B/s (according to yast update download dialog).

Those are totally normal…those sites just get loaded down. (Try different times
of day.)

Try comparing numbers from a site such as:
Speakeasy - Speed Test

You may well HAVE some sort of issue, but your evidence is unconvincing
from those two examples.

deltaflyer44 · September 27, 2008, 4:04pm

Try changing the channel on the router,sometimes interference causes this & changing channels helps

Andy

bjohanns · September 28, 2008, 8:08pm

Hi cookdav,
the 30% packet loss occurs for pings to the local wlan router(!) [192.168.2.1 in my case]. So there can be no other issues than local problems between the “router and me”.

I kind of “measured” the transfer rate through a connect to an internet site of a friend. As this site is hosted by a big provider there might be some bottleneck effects - but my xp box showed the pages in ~ 2 seconds whereas my linux box needed ~4 minutes to present the data.

I am still a puzzled penguine but I still don’t want to install XP or the like on my linux box just to test if there are hardware issues. I still hope that someone could help me to find some misconfiguration.

Thanks for your comments.
Bernd*

bjohanns · September 28, 2008, 8:12pm

Hi deltaflyer44
I have a nearby router on channel 11 - so I changed my channel to 4 (which has been reported as beeing “far enough away” from 11).

There are no other detectable routers nearby.

Thanks for your replies.
Bernd

lwfinger · September 28, 2008, 8:40pm

bjohanns wrote:
> Hi deltaflyer44
> I have a nearby router on channel 11 - so I changed my channel to 4
> (which has been reported as beeing “far enough away” from 11).

As a general rule, you should use channels 1, 6 or 11. The reason is
that the transmission for 6 outputs considerable radiation in channels
4 - 8. Your use of 4 will suffer interference from 6. It may not
matter now, but if someone installs a new router and doesn’t change
the defaults, it will likely be on channel 6 and interfere with you.
As you want to avoid 11, I would suggest channel 1.

Larry

bjohanns · September 28, 2008, 10:38pm

Maybe I found the solution to my problem:

after many tries (channels, position of router, different configurations through yast) I came back to /etc/Wireless/rt61sta/rt61sta.dat which I did not expect to contain any useful information as the relevant SSID and WPA-Key information was stored in /etc/sysconfig/network/ifcfg-ra0 (the settings in rt61sta.dat for these seems to be irrelevant).
So I was mislead to the assumption that the /etc/Wireless directory was some kind of “no SuSE relevant fragment” and did not influence anything.
But I was wrong. Renaming rt61sta.dat to something different caused ra0 to fail and so I started to play around with some of the parameter (namely txBurst, PktAggregate, NativeWpa) and found that throughput increased considerably with
TxBurst=1 (was 0), PktAggregate=0 (was 1) and NativeWpa=0 (was 1).
I do not know the meaning of all those parameter or if the settings are sensible and sound but my packet loss rate dropped down to ~7% and I am able to happily surf again :-).

Well I should prove things now by switching anything back and test again (to exclude the possibility that someone switched of something noisy thus causing the sudden improvement) - but not tonight.

greetings and thanks for your suggestions.
Bernd