iwconfig bitrate is not set properly
Affects | Status | Importance | Assigned to | Milestone | |
---|---|---|---|---|---|
wireless-tools (Fedora) |
Invalid
|
Medium
|
|||
wireless-tools (Ubuntu) |
Triaged
|
Medium
|
Unassigned | ||
Bug Description
Binary package hint: wireless-tools
I'm using Xubuntu 9.04 Jaunty Jackalope, and the bug is in wireless-tools 29-1.1ubuntu2 (but not limited to it; this bug has also appeared in other distributions such as Fedora). The hardware I'm using is a Ralink RT2500 wireless card, and my internet connection security is WPA/WPA2. I'm using the latest NetworkManager to connect to the internet, and NetworkManager-
The issue is that the internet connection speed is not set to what it should be. All the connections always get set to 1Mb/s, no matter what the actual connection speed should be, and thus it greatly limits the connection. Although I'm not sure, but I believe that the auto bitrate in iwconfig is malfunctioning in that it always sets the lowest possible speed.
I've also found a workaround: if you manually set the connection speed, it works properly (and NetworkManager-
ifconfig wlan0 up
iwconfig wlan0 rate 54M
...assuming the connection is wlan0. This needs to be done each time the connection is established, though, so I have added the lines to /etc/rc.local. Though I have read that the same can be achieved by entering "iwconfig wlan0 rate 54M fixed" instead.
In Red Hat Bugzilla #469120, Mark (mark-redhat-bugs) wrote : | #16 |
I have a similar problem, only I can't connect to an unencrypted network.
The last two kernels I've gotten from updates have been unable to connect to the wireless network (2.6.27.
Other packages:
NetworkManager-
wpa_supplicant-
This machine uses an old Microsoft MN-520 wireless card which has worked flawlessly on various linux distributions (including the last several Fedora releases) for the last 4 years. It has an Orinoco chipset.
The end of my dmesg output:
eth2: no IPv6 routers present
fuse init (API version 7.9)
wifi0: CMD=0x0121 => res=0x7f, resp0=0x0004
wifi0: hfa384x_set_rid: CMDCODE_
wifi0: LinkStatus=2 (Disconnected)
wifi0: LinkStatus: BSSID=44:
wifi0: LinkStatus=2 (Disconnected)
wifi0: LinkStatus: BSSID=44:
eth2: Preferred AP (SIOCSIWAP) is used only in Managed mode when host_roaming is enabled
wifi0: LinkStatus=1 (Connected)
wifi0: LinkStatus: BSSID=00:
wifi0: LinkStatus=2 (Disconnected)
wifi0: LinkStatus: BSSID=00:
wifi0: LinkStatus=1 (Connected)
wifi0: LinkStatus: BSSID=00:
ifup output:
# ifup wifi0
Error for wireless request "Set Mode" (8B06) :
SET failed on device eth2 ; Operation not supported.
Determining IP information for eth2...Nothing to flush.
Then it just hangs there until I hit ctrl-C, after which dmesg says
wifi0: LinkStatus: BSSID=00:
wifi0: LinkStatus=2 (Disconnected)
wifi0: LinkStatus: BSSID=44:
wifi0: LinkStatus=2 (Disconnected)
wifi0: LinkStatus: BSSID=44:
wifi0: LinkStatus=1 (Connected)
wifi0: LinkStatus: BSSID=00:
wifi0: invalid skb->cb magic (0x00000168, expected 0xf08a36a2)
wifi0: invalid skb->cb magic (0x00000168, expected 0xf08a36a2)
wifi0: invalid skb->cb magic (0x00000168, expected 0xf08a36a2)
wifi0: invalid skb->cb magic (0x00000168, expected 0xf08a36a2)
In Red Hat Bugzilla #469120, Gian (gian-redhat-bugs) wrote : | #17 |
I should have mentioned that I don't seem to be able to connect to unencrypted networks either. Just tried today at a local hotspot and while getting very good signal readings in Fedora Rawhide (and confirmed later in Vista), I couldn't connect to the network. Unencrypted network connection negotiation takes longer than protected networks, and both take a LONG time (1+ minute negotiation) before stating that the connection couldn't be stablished, or prompting for the Network passphrase/security token.
In Red Hat Bugzilla #469120, Mark (mark-redhat-bugs) wrote : | #18 |
For what it's worth, I tried compiling and installing the vanilla 2.6.27.4 kernel, and the problem occurred with that as well.
In Red Hat Bugzilla #469120, Anne (anne-redhat-bugs) wrote : | #19 |
More info, after many tests:
F10-Snapshot 3 installed on an Acer Aspire One - kernel 2.6.27.3-34rc1 - wireless worked, but there was no wired connection. At the next boot the wired connection worked, but wireless was lost, and has not worked since.
Typical event - First it offers me the chance to connect to hidden networks. I give it the ESSID, set it to WPA and give it the passphrase. It whirs for a while, then a
dialogue box opens asking me to give a WEP key. There is no option to set the WPA key this time.
Kernels tested since that date:
2.6.27.4-79
2.6.27.5-94
2.6.27.5-100
2.6.27.5-109 - all i686
/etc/sysconfig/
# Atheros Communications Inc. AR242x 802.11abg Wireless PCI Express Adapter
On several occasions NetworkManager has reported a successful link, but at 0%. Further investigation shows that it is to a different subnet. At no time has any
other network been listed by NetworkManager, and the nearest other wifi network does not use the IP range mentioned. Also, it reports that it is connected to my ESSID.
'service network status' gives
Configured devices:
l0 eth0 wlan0
Currently active devices:
lo eth1 wmaster0 wlan0
lsmod | grep ath gives
ath5k 17164 0
mac80211 112520 1 ath5k
cfg80211 23816 2 ath5k,mac80211
nm-tool
NetworkManager Tool
State: connected
- Device: eth1
-------
Type: Wired
Driver: r8169
State: connected
Default: yes
HW Address: 00:1E:68:BD:EF:73
Capabilities:
Supported: yes
Carrier Detect: yes
Speed: 100 Mb/s
Wired Settings
IPv4 Settings:
Address: 192.168.0.92
Prefix: 24 (255.255.255.0)
Gateway: 192.168.0.1
DNS: 192.168.0.1
- Device: wlan0
-------
Type: 802.11 WiFi
Driver: ath5k_pci
State: disconnected
Default: no
HW Address: 00:22:33:44:55:66 #obscured
Capabilities:
Supported: yes
Wireless Settings
WEP Encryption: yes
WPA Encryption: yes
WPA2 Encryption: yes
Wireless Access Points
---
iwlist scanning
lo Interface doesn't support scanning.
wmaster0 Interface doesn't support scanning.
wlan0 No scan results
eth1 Interface doesn't support scanning.
pan0 Interface doesn't support scanning.
---
ifconfig wlan0
wlan0 Link encap:Ethernet HWaddr
00:22:33:44:55:66
UP BROADCAST MULTICAST MTU:1500 Met...
In Red Hat Bugzilla #469120, Bug (bug-redhat-bugs) wrote : | #20 |
This bug appears to have been reported against 'rawhide' during the Fedora 10 development cycle.
Changing version to '10'.
More information and reason for this action is here:
http://
In Red Hat Bugzilla #469120, Gian (gian-redhat-bugs) wrote : | #21 |
I'm going to try to fresh-install F10 rather than "upgrade from rawhide to release" on this laptop and see what happens, if I have the same issue, I'll report back.
In Red Hat Bugzilla #469120, Mark (mark-redhat-bugs) wrote : | #22 |
Just installed the F10 release from the live CD. On the first boot, with kernel 2.6.27.
In Red Hat Bugzilla #469120, David (david-redhat-bugs) wrote : | #23 |
I think this one ought to be "high" severity because some systems are completely unable to connect to networks.
In Red Hat Bugzilla #469120, David (david-redhat-bugs) wrote : | #24 |
I have a fully-updated F10 as of right now, and these are the wireless problems that I am seeing (some of them must relate to this bug report):
1. The network-
2. knetwork-manager and the gnome network manager can see my wireless network, but fails to get DHCP to allocate it an IP address; it times-out.
Both 1 and 2 above occur whether I have WEP on or off on my Netgear router.
So, I think there are several bugs with connecting to wireless networks.
The title of this bug ought to be changed from 'Rawhide' to 'latest F10' or similar.
Hope that helps :)
In Red Hat Bugzilla #469120, Gian (gian-redhat-bugs) wrote : | #25 |
Indeed, installed F10 on the laptop I'm seeing this problem occur, and it has happened just like with Rawhide. Updates applied and still getting the same behavior. Tried some LiveCD images from other distributions with similar versions of components and there does not seem to be a problem with NM or WPASupplicant or DHCP there, only F10 shows these problems (thus far). Is there a way to help troubleshoot this? I realize that simply stating "cannot connect" is not as helpful as what messages are being printed to dmesg or the like, so what would you recommend doing to try and locate the origin of these problems?
In Red Hat Bugzilla #469120, Anne (anne-redhat-bugs) wrote : | #26 |
F10 on Acer Aspire One - temporarily turned NM of and configured with s-c-n. The result is the same. Possibly helpful notes:
boot message
ADDRCONF(
ifup wlan0
Determining IP information for wlan0...Nothing to flush.
failed
This seems to imply that NM is not the problem, but either the ath5k driver or wpa_supplicant?
In Red Hat Bugzilla #469120, Gian (gian-redhat-bugs) wrote : | #27 |
I would think that rather than drivers or NM as such, the problem seems to be either or both, WPA supplicant and the 802.11 kernel stack. I just tried the latest stable LiveCD image of another popular distro, with virtually equal software versions as F10, and there I experienced the problem as well, with all and the very same dmesg messages, which is to say the least, odd, and points towards either or a relation between the drivers, kernel 802.11 stack, WPA_Supplicant and Network Manager, since there seems to be a connection going on, but then *dropped* due to "local choice" (for whatever that may mean), "wlan0: disassociating by local choice (reason=3)", I would think that the problem lies within WPA_Supplicant rather than the other components, but then again, I'm not sure how to properly debug this.
PS: Summary changed from Rawhide to F10 to reflect the current state of affairs.
In Red Hat Bugzilla #469120, Gian (gian-redhat-bugs) wrote : | #28 |
As I said before, I tried to build from source the latest WPA Supplicant code under Linux, but have failed miserably (I'm told Makefile is missing a config file, but I can see no configure script anywhere, does it make use of Auto Tools at all?). Keeping this in mind I have tried a couple more LiveCD images of other distributions (some based on Fedora 10 even) and with all of them with recent versions of kernel (for the drivers), NM and WPA supplicant, I get the very same behavior. Now what I don't fully understand is why does this happen with only a seemingly limited array of WLAN chipsets/cards?
What about trying to detrmine which adapters DO exhibit the problem and try to address those?
In my case I'm using a Realtek 8187B (USB) internal WLAN adapter.
In Red Hat Bugzilla #469120, David (david-redhat-bugs) wrote : | #29 |
Just tried my RT2500 card with the latest F10 updates, latest kernel etc, and get this, but no connection:
Dec 12 17:11:23 stail00422 NetworkManager: <info> Found new 802.11 WiFi device'wlan0'.
Dec 12 17:11:23 stail00422 NetworkManager: <info> (wlan0): exported as /org/freedeskto
Dec 12 17:11:27 stail00422 NetworkManager: <info> (wlan0): device state change: 1 -> 2
Dec 12 17:11:27 stail00422 NetworkManager: <info> (wlan0): bringing up device.
Dec 12 17:11:27 stail00422 kernel: ADDRCONF(
Dec 12 17:11:27 stail00422 NetworkManager: <info> (wlan0): preparing device.
Dec 12 17:11:27 stail00422 NetworkManager: <info> (wlan0): deactivating device(reason: 2).
Dec 12 17:11:27 stail00422 NetworkManager: <info> (wlan0): device state change: 2 -> 3
Dec 12 17:11:27 stail00422 NetworkManager: <info> (wlan0): supplicant interface state: starting -> ready
Dec 12 17:12:06 stail00422 NetworkManager: <WARN> wait_for_
Dec 12 17:12:21 stail00422 NetworkManager: <info> (wlan0): device state change: 3 -> 2
Dec 12 17:12:21 stail00422 NetworkManager: <info> (wlan0): deactivating device(reason: 0).
Dec 12 17:12:21 stail00422 NetworkManager: <info> (wlan0): taking down device.
BTW: I'm getting this on a Pentium 3, so the platform should be changed from x86_64 to 'All' probably.
In Red Hat Bugzilla #469120, Anne (anne-redhat-bugs) wrote : | #30 |
At some point early this month and up to yesterday, I was able to connect after giving a gnome-keyring-
Finally, I installed knetworkmanager and removed NetworkManager-
I think this brings the problem fairly and squarely to NM, as
WPA is being handled correctly by knetworkmanager.
In Red Hat Bugzilla #469120, Anne (anne-redhat-bugs) wrote : | #31 |
Excerpt from log during time when gnome-keyring-pam was installed, and prior to changing to knetworkmanager:
kdm: :0: PAM adding faulty module: /lib/security/
Time(s)
kdm: :0: PAM unable to dlopen(
/lib/security/
file or directory: 1 Time(s)
kdm: :0: gnome-keyring-
Failed to contact configuration server; some possible causes are that you need
to enable TCP/IP networking for ORBit, or you have stale NFS locks due to a
system crash. See http://
(Details - 1: Not running within active session)
couldn't lookup ssh component setting: Failed to contact configuration server;
some possible causes are that you need to enable TCP/IP networking for ORBit,
or you have stale NFS locks due to a system crash. See
http://
running within active session)
component setting: Failed to contact configuration server; some possible
causes are that you need to enable TCP/IP networking for ORBit, or you have
stale NFS locks due to a system crash. See
http://
- 1: Not running within active session): 5 Time(s)
In Red Hat Bugzilla #469120, Dan (dan-redhat-bugs) wrote : | #32 |
Are people still having problems with rtl8187 and latest kernel updates? The issue with establishing connections is highly driver dependent.
Next, what type of connections are having the problem; all, or just WEP, or just WPA, etc?
Gian: if WEP, what type of WEP key is it (ASCII, Hex, or Passphrase), and how long is it?
In Red Hat Bugzilla #469120, Brent (brent-redhat-bugs) wrote : | #33 |
bug 457441 is tracking rt2500 problems for those of you interested
In Red Hat Bugzilla #469120, Gian (gian-redhat-bugs) wrote : | #34 |
(In reply to comment #17)
> Are people still having problems with rtl8187 and latest kernel updates? The
> issue with establishing connections is highly driver dependent.
>
> Next, what type of connections are having the problem; all, or just WEP, or
> just WPA, etc?
>
> Gian: if WEP, what type of WEP key is it (ASCII, Hex, or Passphrase), and how
> long is it?
Well, this issue doesn't seem to be driver-dependent, but rather Network Manger is messing up for some reason. I can establish the connection once I "hack" the correct key into the keyring... What I have seen that happens (with any new network that I register) is the following:
1.- When trying to establish the connection with Netowrk Manager, after introducing the key (WPA/WEP, any kind [hex, ascii, etc) the first attempt fails.
2.- When the prompt pops back up, and I select "show key", the key that NM presents is completely different from the one I actually entered (especially true for WEP-HEX).
3.- I have to resort to use either Seahorse or gnome-keyring-
I have seen this happening with either RealTek or RaLink hardware. I do not know if other chipsets have similar issues, or if this might be a problem of Network Manager + hardware type.
In Red Hat Bugzilla #469120, Brent (brent-redhat-bugs) wrote : | #35 |
I have also seen the corrupted HEX KEY when "show key" is clicked using rt2500pci
In Red Hat Bugzilla #469120, Dan (dan-redhat-bugs) wrote : | #36 |
(In reply to comment #19)
> (In reply to comment #17)
> > Are people still having problems with rtl8187 and latest kernel updates? The
> > issue with establishing connections is highly driver dependent.
> >
> > Next, what type of connections are having the problem; all, or just WEP, or
> > just WPA, etc?
> >
> > Gian: if WEP, what type of WEP key is it (ASCII, Hex, or Passphrase), and how
> > long is it?
>
> Well, this issue doesn't seem to be driver-dependent, but rather Network Manger
> is messing up for some reason. I can establish the connection once I "hack" the
> correct key into the keyring... What I have seen that happens (with any new
> network that I register) is the following:
>
> 1.- When trying to establish the connection with Netowrk Manager, after
> introducing the key (WPA/WEP, any kind [hex, ascii, etc) the first attempt
> fails.
>
> 2.- When the prompt pops back up, and I select "show key", the key that NM
> presents is completely different from the one I actually entered (especially
> true for WEP-HEX).
That's some good info. How many characters is your key? It's definitely a WEP Hex key, not a 5 or 13 character WEP ASCII key, and not a 8 - 64 character WEP passphrase?
In Red Hat Bugzilla #469120, Brent (brent-redhat-bugs) wrote : | #37 |
WEP HEX
0x0011223344556
13 hex digits
In Red Hat Bugzilla #469120, Dan (dan-redhat-bugs) wrote : | #38 |
Are you entering the key with or without the leading "0x"? The 0x is not required (on any modern OS, Apple stopped using that a while ago even) and it shouldn't be needed when entering the password into NM.
If you enter the 0x, then it's 28 characters, and it's then a WEP passphrase, which is totally different.
WEP sucks. As you can tell. Let me know if not using the "0x" works for you.
In Red Hat Bugzilla #469120, Gian (gian-redhat-bugs) wrote : | #39 |
Indeed, WEP utterly sucks, but sadly is the only type of wireless security supported by many ISP-provided WiFi routers (at least down here in Mexico is pretty much the defacto standard, and 64-bit keys to top it all). Pretty much since NM was made standard in Fedora, and even before that (FC6 era, I believe) when you had to separately install it and enable it as a service, only a couple releases actually required that you used the 0x prefix for entering HEX keys. At any rate, I'll do a series of tests to change the key from WEP to WPA2 in my router (which is one of the few that actually does support that security scheme) and see what happens. Will report back any success or failure later today.
In Red Hat Bugzilla #469120, Dan (dan-redhat-bugs) wrote : | #40 |
Early release may have required the 0x prefix (which Apple used to use up until OS X 10.3 as well), but NM hasn't used the 0x prefix at least since 0.5 back in 2004/2005. Please let me know how the testing goes. Thanks!
In Red Hat Bugzilla #469120, Brent (brent-redhat-bugs) wrote : | #41 |
No, the key is just hex, no 0x in front ... just trying to make it more "clear" ... that backfired nicely. I will try to be more confusing in the future.
:>
In Red Hat Bugzilla #469120, Dainius (dainius-redhat-bugs) wrote : | #42 |
I have the same problem on vanilla FC10. NetworkManager-
I've also tried switching to KNetworkManager (actually, now both Gnome and K managers are installed, and they work as one :) ), and it's the same.
Finally I've attempted to make the connection manageable by the system-
lspci shows my network devices:
00:1f.6 Modem: Intel Corporation 82801DB/DBL/DBM (ICH4/ICH4-
02:00.0 Ethernet controller: Realtek Semiconductor Co., Ltd. RTL-8139/
03:00.0 Network controller: RaLink RT2500 802.11g Cardbus/mini-PCI (rev 01)
iwlist wlan0 scanning (when online, doesn't work offline):
wlan0 Scan completed :
Cell 01 - Address: 00:1D:7E:BC:C6:F4
ifconfig:
wlan0 Link encap:Ethernet HWaddr 00:02:44:AF:65:51
inet addr:192.168.1.101 Bcast:192.168.1.255 Mask:255.255.255.0
inet6 addr: fe80::202:
UP BROADCAST RUNNING MULTICAST MTU:1500 Metric:1
RX packets:11 errors:0 dropped:0 overruns:0 frame:0
TX packets:34 errors:0 dropped:0 overruns:0 carrier:0
RX bytes:2233 (2.1 KiB) TX bytes:5772 (5.6 KiB)
iwconfig:
wlan0 IEEE 802.11bg ESSID:"linksys"
Bit Rate=1 Mb/s Tx-Power=24 dBm
Retry min limit:7 RTS thr:off Fragment thr=2352 B
Power Management:off
Link Quality=69/100 Signal level:-62 dBm
Rx invalid nwid:0 Rx invalid crypt:0 Rx invalid frag:0
Tx excessive re...
In Red Hat Bugzilla #469120, Dainius (dainius-redhat-bugs) wrote : | #43 |
I've just downloaded a Live CD of Xubuntu (Jaunty Jackalope) to check what is distro-specific. And yes, it seems that it's exactly what I thought: this bug is actually two separate bugs. Xubuntu's NetworkManager-
The second problem is speed and reliability. Strangely, the speed is simply not set correctly, and a fix is pathetically easy. In a Terminal, you need to write this:
ifconfig wlan0 up
iwconfig wlan0 rate 56M
And that fixes the speed problem, I think on both Fedora and Xubuntu. This issue is obviously NetworkManager-
And the reliability issue remains - the card tends to disconnect at times, and then attempt to reconnect. I haven't found a fix for this yet, so I'm not sure what is the cause, too.
In Red Hat Bugzilla #469120, Brent (brent-redhat-bugs) wrote : | #44 |
I have two other PCMCIA wireless cards that work fine with NetworkManager and DHCP under F10.
The rt2500pci worked fine until a kernel update, the offending kernel has been identified and the developer has the data.
..."And that fixes the speed problem, I think on both Fedora and Xubuntu. This
issue is obviously NetworkManager-
automatically."...
would it be possible to get you to test the speed change on Fedora ...
In Red Hat Bugzilla #469120, Dainius (dainius-redhat-bugs) wrote : | #45 |
I've tested it with a Live CD (couldn't make Konqueror resolve URLs, but it just seems it needs it installed for that), and yes, that method works in Fedora 10 too. Although one correction, it should be like this:
ifconfig wlan0 up
iwconfig wlan0 rate 54M
As there is no speed of 56M. You will also need to either make the changes permanent (or else you'll have to reset it each time you launch your PC) by either putting both lines to the /etc/rc.local file or adding "fixed" at the end of the last command (didn't check this method, though):
ifconfig wlan0 up
iwconfig wlan0 rate 54M fixed
In Red Hat Bugzilla #469120, Dainius (dainius-redhat-bugs) wrote : | #46 |
Finally I managed to solve all the internet problems! And it really seems to be three separate bugs! Here is the solution for the last bug, where internet would randomly drop connection:
http://
This guide is only for Gentoo and Ubuntu, but something similar should work with Fedora as well. This bug is in the kernel itself - the system doesn't monitor the connection as long as it should, thus often declaring it dead. With a simple change in one file, the bug is solved! However, that means that you have to compile your own kernel, and there will be no auto updates for it.
So, to sum up, the three bugs in Fedora are:
1. Unknown cause - DHCP failure. The system doesn't acquire IPs like it should, making it possible to connect to the internet only by manually entering all the data and thus setting a static local IP. This is the only bug that is Fedora-specific.
2. iwconfig - wrong bitrate set. This is a generic bug, because it seems that auto bitrate always sets it to 1Mbps, which is intolerable. The solution is in the comment #30.
3. kernel - wrong IEEE monitoring interval set. This makes the connection unstable. The solution is higher in this comment.
Hopefully this will help others that are having these issues!
Changed in wireless-tools (Fedora): | |
status: | Unknown → Confirmed |
Dainius 'GreatEmerald' Masiliūnas (pastas4) wrote : | #1 |
This bug is also confirmed in wireless-tools question #30772 :
https:/
In Red Hat Bugzilla #469120, David (david-redhat-bugs) wrote : | #47 |
Created attachment 347080
/var/log/messages featuring NetworkManager and DHCP error messages with RT2500 failure to connect
Just to confirm that this is still a problem with F11 (released today, no updates.)
I have attached the /var/log/messages file so that a kind fixer can see the exact error messages from NetworkManager and DHCP. Look for wlan0 in the file, of course.
The problem occurs with WEP encryption.
I use a 10 hex digit key.
Can some-one please mark this as F11 as well as F9 and F10? Ta.
In Red Hat Bugzilla #469120, Dainius (dainius-redhat-bugs) wrote : | #48 |
How many of the three bugs are you getting? I've read that the latest prepatch kernel, 2.6.30-rc8, might solve the problem #3. Can someone confirm or deny this? And what kernel is used in Fedora 11, 2.6.29.4?
In Red Hat Bugzilla #469120, David (david-redhat-bugs) wrote : | #49 |
I'm getting #1 on F11.
#2 may be there too, but I can't tell because I'm not getting a connection.
Have certainly seen #3, i.e. instability on F9, but not sure about F10 and F11.
To try a later kernel, I would have to rip a PCI rt2500 card out of a machine that my kids use, so may not be able to check symptoms very easily just now.
In Red Hat Bugzilla #469120, David (david-redhat-bugs) wrote : | #50 |
Fc11 kernel is kernel-
Dainius 'GreatEmerald' Masiliūnas (pastas4) wrote : | #2 |
I've found out another fix: get compat-wireless from http://
In Red Hat Bugzilla #469120, Dainius (dainius-redhat-bugs) wrote : | #51 |
I've now tested 2.6.30 kernel, and here are the results:
1. The POWER LED on the card is now active!
2. Bug #3 is solved!
However, I seem to have overlooked one more bug (yes, this bug is actually FOUR in one!):
4. The connection is unstable not only due to monitoring interval (bug #3; the now fixed bug outputted this into dmesg: "No ProbeResp from current AP"), but it's also unstable because of something else that output this into dmesg: "no probe response from AP - disassociating" and "direct probe to AP timed out".
Solution, which also fixes bug #3 for older kernels and even to some degree bug #2: get and install compat-wireless from this page:
http://
So, as you can see, this looks like the ultimate solution that fixes almost all the bugs mentioned here. The only bug left now is #1, and is fedora-specific.
In Red Hat Bugzilla #469120, Dainius (dainius-redhat-bugs) wrote : | #52 |
Uhh, the highlighter in bugzilla linked to wrong bugs... I meant the subbugs of this bug, #469120.
In Red Hat Bugzilla #469120, David (david-redhat-bugs) wrote : | #53 |
I think some-one should change this one from F10 to F11.
In Red Hat Bugzilla #469120, Anne (anne-redhat-bugs) wrote : | #54 |
Still using F10 on the netbook. By now NM seems more stable, but I still occasionally have the problem of the passphrase not being accepted. I've no idea why, but doing an ifdown command causes NM to ask again, and this time it works. This seems to be consistent. Does that help?
John Stiles (john-isabelle) wrote : | #3 |
Atheros AR9170 usb repeats this issue (and work arounds).
In Red Hat Bugzilla #469120, philby (philby-redhat-bugs) wrote : | #55 |
(In reply to comment #39)
> Still using F10 on the netbook. By now NM seems more stable, but I still
> occasionally have the problem of the passphrase not being accepted.
Its lost that stability again now with...
NetworkManager-
NetworkManager-
2.6.27.
Atheros Communication Inc. AR242x 802.11abg Wireless PCI Express Adapter on a Toshiba Satellite A215
AP: Netgear
Error: wlan0: disassociating by local choice (reason=3)
FC10 pops up "Wireless Network Authentication required" dialogue repeatedly and if it does succeed by other means would show 0% signal strength.
In Red Hat Bugzilla #469120, philby (philby-redhat-bugs) wrote : | #56 |
Confirming that the problem is confined to NetworkManager-
With knetworkmanager, we get the same error on an FC10 2.6.27.
wlan0: associated
ADDRCONF(
wlan0: disassociating by local choice (reason=3)
In Red Hat Bugzilla #469120, Christian (christian-redhat-bugs) wrote : | #57 |
With the current F11 kernel update to
kernel-
I cannot get a WLAN connection anymore, so the problem got worse for me.
Going back to
kernel-
solves the problem for me (note that I too have to make several connection retries and sometimes I have also to reconnect my USB stick to get a wireless connection)
I am also using the following:
NetworkManage
Bus 001 Device 004: ID 0846:4260 NetGear, Inc. WG111v3 54 Mbps Wireless [realtek RTL8187B]
Device Descriptor:
bLength 18
bDescriptorType 1
bcdUSB 2.00
bDeviceClass 0 (Defined at Interface level)
bDeviceSubClass 0
bDeviceProtocol 0
bMaxPacketSize0 64
idVendor 0x0846 NetGear, Inc.
idProduct 0x4260 WG111v3 54 Mbps Wireless [realtek RTL8187B]
bcdDevice 2.00
iManufacturer 1 Manufacturer_
iProduct 2 NETGEAR WG111v3
iSerial 3 001E2AC1E6C1
bNumConfigura
Configuration Descriptor:
bLength 9
bDescriptorType 2
wTotalLength 81
bNumInterfaces 1
bConfigurat
iConfiguration 4 Wireless Network Card
bmAttributes 0x80
(Bus Powered)
MaxPower 500mA
Interface Descriptor:
bLength 9
bDescript
bInterfac
bAlternat
bNumEndpoints 9
bInterfac
bInterfac
bInterfac
iInterface 2 NETGEAR WG111v3
Endpoint Descriptor:
bLength 7
Transfer Type Bulk
Synch Type None
Usage Type Data
bInterval 0
Endpoint Descriptor:
bLength 7
Transfer Type Bulk
Synch Type None
Usage Type Data
bInterval 0
Endpoint Descriptor:
bLength 7
Transfer Type Bulk
Synch Type None
Usage Type Data
bInterval 0
Endpoint Descriptor:
bLength 7
Transfer Type Bulk
Synch Type None
Usag...
In Red Hat Bugzilla #469120, David (david-redhat-bugs) wrote : | #58 |
Problem still present in F12 beta i386.
In Red Hat Bugzilla #469120, David (david-redhat-bugs) wrote : | #59 |
This bug affects rt2500 wireless cards and is all about them.
It is also now in Fedora 12.
I cannot see what INFO is needed, so am ticking the info box.
Please would somebody
a) improve the title so that it includes all the Fedoras from 8 through to 12, and
b) improve the title so that it says rt2500 or Ralink or similar.
Ta.
In Red Hat Bugzilla #469120, philby (philby-redhat-bugs) wrote : | #60 |
(In reply to comment #44)
> This bug affects rt2500 wireless cards and is all about them.
All about them? The problem is seem in numerous other cards and is not confined to rt2500.
> It is also now in Fedora 12.
> I cannot see what INFO is needed, so am ticking the info box.
>
> Please would somebody
> a) improve the title so that it includes all the Fedoras from 8 through to 12,
> and
Good idea.
> b) improve the title so that it says rt2500 or Ralink or similar.
Should stay clear of any particularism cause of the concern that others might get ignored when a fix is found for a specific hardware.
In Red Hat Bugzilla #469120, Dan (dan-redhat-bugs) wrote : | #61 |
Can people respond with the kernel version causing the issue, and the specific piece of wlan hardware they have the issue with? All that's somewhat buried in the comments above, and you may have updated your kernel since you posted the comment. Thanks!
In Red Hat Bugzilla #469120, David (david-redhat-bugs) wrote : | #62 |
All the kernels in F8, F9, F10, F11 and F12beta that I have tried are involved with this bug except for the very first F8 kernel. I can probably get the version numbers, but its lots of different versions continually from F8 to F12 as far as I can tell.
Hardware affected = rt2500 pci and rt2500 PCMCIA.
In Red Hat Bugzilla #469120, Christian (christian-redhat-bugs) wrote : | #63 |
Regarding the affected hardware: again, I have a
usb 1-9: New USB device found, idVendor=0846, idProduct=4260
usb 1-9: New USB device strings: Mfr=1, Product=2, SerialNumber=3
usb 1-9: Product: NETGEAR WG111v3
usb 1-9: Manufacturer: Manufacturer_
usb 1-9: SerialNumber: 001E2AC1E6C1
usb 1-9: configuration #1 chosen from 1 choice
which is a realtek RTL8187B according to lsusb.
The latest kernel which works for me is
2.6.29.
With every other kernel since then, I cannot establish a connection, no matter what I do. By using any newer kernel even the 'trick' with pluging the USB-device out and back in does not work for me.
In Red Hat Bugzilla #469120, philby (philby-redhat-bugs) wrote : | #64 |
(In reply to comment #48)
> Regarding the affected hardware: again, I have a
>
> usb 1-9: New USB device found, idVendor=0846, idProduct=4260
> usb 1-9: New USB device strings: Mfr=1, Product=2, SerialNumber=3
> usb 1-9: Product: NETGEAR WG111v3
> usb 1-9: Manufacturer: Manufacturer_
> usb 1-9: SerialNumber: 001E2AC1E6C1
> usb 1-9: configuration #1 chosen from 1 choice
>
> which is a realtek RTL8187B according to lsusb.
>
> The latest kernel which works for me is
>
> 2.6.29.
Wow! that's cool info, this is a regression. Now if only someone could bisect and find the faulty commit ;-)
In Red Hat Bugzilla #469120, Christian (christian-redhat-bugs) wrote : | #65 |
It would be even cooler if someone could confirm this with similar hardware ;)
In Red Hat Bugzilla #469120, Dan (dan-redhat-bugs) wrote : | #66 |
Just for reference:
2.6.30.
works fine with rt73usb.
In Red Hat Bugzilla #469120, Bug (bug-redhat-bugs) wrote : | #67 |
This message is a reminder that Fedora 10 is nearing its end of life.
Approximately 30 (thirty) days from now Fedora will stop maintaining
and issuing updates for Fedora 10. It is Fedora's policy to close all
bug reports from releases that are no longer maintained. At that time
this bug will be closed as WONTFIX if it remains open with a Fedora
'version' of '10'.
Package Maintainer: If you wish for this bug to remain open because you
plan to fix it in a currently maintained version, simply change the 'version'
to a later Fedora version prior to Fedora 10's end of life.
Bug Reporter: Thank you for reporting this issue and we are sorry that
we may not be able to fix it before Fedora 10 is end of life. If you
would still like to see this bug fixed and are able to reproduce it
against a later version of Fedora please change the 'version' of this
bug to the applicable version. If you are unable to change the version,
please add a comment here and someone will do it for you.
Although we aim to fix as many bugs as possible during every release's
lifetime, sometimes those efforts are overtaken by events. Often a
more recent Fedora release includes newer upstream software that fixes
bugs or makes them obsolete.
The process we are following is described here:
http://
In Red Hat Bugzilla #469120, Dainius (dainius-redhat-bugs) wrote : | #68 |
Someone should bump this to version 12 and change the title to reflect the change.
However, the only Fedora-specific bug here for me is DHCP not working properly. Others are NetworkManager and Kernel specific and can be worked around.
In Red Hat Bugzilla #469120, David (david-redhat-bugs) wrote : | #69 |
Definitely affects F11, F12 and Rawhide.
In Red Hat Bugzilla #469120, Dan (dan-redhat-bugs) wrote : | #70 |
*** Bug 472183 has been marked as a duplicate of this bug. ***
In Red Hat Bugzilla #469120, cornel (cornel-redhat-bugs) wrote : | #71 |
i can confirm the bug is present in updated f12 with internal pci card 10ec:8172 (realtek), using w2k drivers with ndiswrapper
Charlie Kravetz (charlie-tca) wrote : | #4 |
Thanks for reporting this bug and any supporting documentation. Since this bug has enough information provided for a developer to begin work, I'm going to mark it as confirmed and let them handle it from here. Thanks for taking the time to make Ubuntu better!
Changed in wireless-tools (Ubuntu): | |
status: | New → Confirmed |
importance: | Undecided → Medium |
In Red Hat Bugzilla #469120, Rogerio (rogerio-redhat-bugs) wrote : | #72 |
Ok so I am looking for this message I get when I try to connect to wireless:
[ 127.228213] wlan0: deauthenticating from 00:21:04:1a:21:f6 by local choice (reason=3)
And I found your bug ... I followed the discussion BUT I have something to add:
I am using WICD , no NetworkManager whatsoever, I am using a Debian Testing machine with kernel 2.6.32 , this error was reproductible in Mandriva 2010.0 64bit , Ubuntu 9.10 64bit and Debian 64bit
My card is a Realtek rtl8187b internal USB
Works fine in MS-Win , but has a STRANGE bahavior in Linux.
Have a post on LinuxQuestions.org about my problem with all the logs:
Thanks for letting me know I am not insane :)
Rogerio
In Red Hat Bugzilla #469120, Paulo (paulo-redhat-bugs) wrote : | #73 |
I had the same problem but i solved just deleting the files "ifup-wlan0" of my
system and after my network card has worked.
The files i found at directories:
/etc/sysconfig/
/etc/sysconfig/
/etc/sysconfig/
I think the same should solve problem with other cards..
In Red Hat Bugzilla #469120, Stanislaw (stanislaw-redhat-bugs) wrote : | #74 |
This bug report contains too many comments describing different problems. Please open separate bug reports for rt2500 and rtl8xxx hardware, against current fedora kernels (but only if you have kernel/driver problem, for example comment 58 does not describe kernel problem).
In Red Hat Bugzilla #469120, cornel (cornel-redhat-bugs) wrote : | #75 |
afaict, in recent f13 kernels, rt2500 works.
In Red Hat Bugzilla #469120, philby (philby-redhat-bugs) wrote : | #76 |
For the Atheros Communications Inc. AR242x 802.11abg Wireless PCI Express Adapter, I have long stopped using WAP and have configured my WiFi router to use WPA Personal/
AndyCee (andrewozear) wrote : | #5 |
Any progress on this?
I've tripled the speed of my wireless network now that I've read the solution to this bug.
What's doing?
Charlie Kravetz (charlie-tca) wrote : | #6 |
The issue that you reported is one that should be reproducible with the live environment of the Desktop CD of the development release - Natty Narwhal. It would help us greatly if you could test with it so we can work on getting it fixed in the next release of Ubuntu. You can find out more about the development release at http://
This bug was reported in Ubuntu 9.04. Which release are you using? Are the symptoms you experience exactly the same as the original report?
Changed in wireless-tools (Ubuntu): | |
status: | Confirmed → Incomplete |
Osik (stefan-tollkuehn) wrote : | #7 |
Hi,
I'm on 10.10 x64 and experiencing quite similar behaviour with my wifi device, except that my connection speed seems to be 11MB/s instead of 1MB/s.
lspci:
12:00.0 Network controller: Broadcom Corporation BCM4313 802.11b/g LP-PHY (rev 01)
iwconfig eth1:
eth1 IEEE 802.11 ESSID:"MY-WiFi"
Bit Rate=11 Mb/s Tx-Power:24 dBm
Retry min limit:7 RTS thr:off Fragment thr:off
Power Management:off
Link Quality=5/5 Signal level=-42 dBm Noise level=-89 dBm
Rx invalid nwid:0 Rx invalid crypt:216 Rx invalid frag:0
Tx excessive retries:82 Invalid misc:0 Missed beacon:0
If I try to set the rate manually (i.e. iwconfig eth1 rate 48M) nothing works. I'm using the proprietary driver from repository installed through jockey.
Any ideas how to fix this?
Charlie Kravetz (charlie-tca) wrote : | #8 |
Osik: We appreciate the difficulties you are facing, but it would make more sense to raise problems you are having in the support tracker at https:/
AndyCee (andrewozear) wrote : | #9 |
Apologies for the lack of detail.
I encountered the exact symptoms of the original report, except:
* I am running 10.10
* I'm using an Intel wireless card (4965)
* I resolved the problem by adding the following line to /etc/rc.local:
iwconfig wlan0 rate 54M auto
Will test the liveCD for Natty when I can.
Jamie Kitson (jamie-kitson) wrote : | #10 |
I have found that if I do a rmmod wl (I am using the compiled broadcom sta driver) and then a modprobe wl that will fix the problem for me.
Cheers, Jamie
Jamie Kitson (jamie-kitson) wrote : | #11 |
Sorry I think my above comment can be disregarded, it seems to have just started working for no particular reason :-/
Jonathan Challinger (mr-challinger) wrote : | #12 |
Would love to see a fix for this. I'm willing to provide any information necessary to get this fixed.
Charlie Kravetz (charlie-tca) wrote : | #13 |
Since this bug has enough information provided for a developer to begin work, I'm going to mark it as confirmed and let them handle it from here. Thanks for taking the time to make Ubuntu better!
Changed in wireless-tools (Ubuntu): | |
status: | Incomplete → Triaged |
AndyCee (andrewozear) wrote : | #14 |
Tested with liveUSB of Natty & problem appears fixed.
Without connection, iwconfig lists an "unknown"-type bitrate speed
Upon first connecting to a 54Mb wireless network, iwconfig reported 2Mb/s bitrate.
After running "sudo apt-get update" the bitrate setting increased to 36 Mb/s.
Seems 1Mb/s connection restriction is no longer in place. Have not tested on full install, and am personally unlikely to install until final release.
Changed in wireless-tools (Fedora): | |
importance: | Unknown → Medium |
status: | Confirmed → Invalid |
Description of problem:
When connecting to a security enabled network (WEP/WPA), the computer cannot stablish a connection using the right key and security protocol.
Version-Release number of selected component (if applicable): 0.7.0-0. 11.svn4229. fc10.x86_ 64 0.6.4-2. fc10.x86_ 64 2.6.27. 4-58.fc10. x86_64
NetworkManager-
wpa_supplicant-
kernel-
How reproducible:
Always, since last update.
Steps to Reproduce:
1. Try to establish a connection to a network with security enabled.
2. Enter the key according to the security protocol.
Actual results:
Connection fails
Expected results:
Establish a connection.
Additional info:
NetworkManager does not seem to store the right key or protocol in the connections when editing the stored connection settings. When attempting to connect dmesg shows this:
[dmesg]
wlan0: authenticate with AP 00:40:10:10:00:03
wlan0: authenticate with AP 00:40:10:10:00:03
wlan0: authenticated
wlan0: associate with AP 00:40:10:10:00:03
wlan0: RX AssocResp from 00:40:10:10:00:03 (capab=0x411 status=0 aid=1)
wlan0: associated
wlan0: disassociating by local choice (reason=3)
[/dmesg]
When negotiating, I can see both dots green, then the prmpt again.
Hardware: Realtek 8187B