Bug#677880: iwlwifi: flakey and slow connection, high "invalid misc" count ( Same problem here.)
Hi,
On Sun, Jun 17, 2012 at 3:44 PM, Jonathan Nieder <jrnieder@gmail.com> wrote:
> notfound 677880 3.2.12-1
> found 677880 linux-2.6/3.2.16-1
> retitle 677880 iwlwifi: flakey and slow connection, high "invalid misc" count
> quit
>
> Hi,
>
> Olivier Berger wrote:
>
>> I think you didn't experience exactly the same problem explained in the
>> original report. My wifi wasn't disconnecting, but I was just annoyed by the
>> syslog pollution linked to cfg80211.
>>
>> So cloning to another bug report hoping you don't mind.
>
> Victor, your bug is now #677880.
>
> Victor Pablos Ceruelo wrote:
>
>> When it does not connect it is usual to see in the logs
> [...]
>> wlan0: authenticated
>> wlan0: associate with 88:25:2c:92:3d:aa (try 1)
>> wlan0: RX AssocResp from 88:25:2c:92:3d:aa (capab=0x421 status=0 aid=1)
>> wlan0: associated
> [...]
>> but it seems to be that dhcp does not receive any package.
>> using static ip does not fix this, nor using a different dhcp client
>> (I've tried pump and dhclient, now I'm using dhcpcd)
>>
>> I think the problem is related to the "Invalid misc" field.
>> Now it's 384, but it is usually over 10.000
>> (although I have not seen 40.000, it disconnects before always).
>
> Interesting.
>
> [...]
>> Same problem with zd1211rw (this time it is even worst, it says it's
>> connected but it is not).
>
> Ok, so either both drivers have a similar bug (or common code they
> both call is broken) or there is something in the environment that is
> responsible for both sets of symptoms. *Either seems possible.
>
> [...]
>> This is reportbug info from the other machine.
> [...]
>> It switches between 11Mb/s and 1Mb/s very often,
>> as the other machine.
>
> Now I'm getting lost. *Can you please summarize:
>
> *- what the setup is like --- what are the relevant machines, what
> * kind of wireless router, etc
>
> *- what kind of card each machine uses
>
> *- what kernels you have tried on each machine and what happened with
> * each
>
> Then for concreteness, let's focus on getting this fixed on a single
> machine and card (e.g., the first with the iwlwifi adapter) and go
> from there.
>
> Hope that helps,
> Jonathan
>
3 machines:
Asus 2552 HUH with with 3.2.0-2 (no amd64) and b43-legacy wifi driver
Lenovo W500 with 3.3.4-1~experimental.1 (amd64) and iwlwifi agn 5300
Desktop with 3.2.0-2-686-pae (Debian 3.2.19-1) (no amd64) and pci rt2500,
[ 14.938382] rt2500pci 0000:01:0a.0: PCI INT A -> GSI 21 (level,
low) -> IRQ 21
2 usb dongles with Zydas zd1211rw driver
[ 2931.665411] usb 1-1: New USB device found, idVendor=0ace, idProduct=1215
[ 2931.665421] usb 1-1: New USB device strings: Mfr=16, Product=32,
SerialNumber=0
[ 2931.665428] usb 1-1: Product: USB2.0 WLAN
[ 2931.665433] usb 1-1: Manufacturer: ZyDAS
...
[ 2971.033011] usb 1-1: New USB device found, idVendor=083a, idProduct=4505
[ 2971.033022] usb 1-1: New USB device strings: Mfr=16, Product=32,
SerialNumber=0
[ 2971.033029] usb 1-1: Product: USB2.0 WLAN
[ 2971.033034] usb 1-1: Manufacturer: SMC
and 2 routers:
arcadyan arv7518 with wifi bgn
Prestige 660HW-61 with wifi bg
I was using the first one until a bug I reported some years ago.
Since it was an "n" issue I am now using the second one just for wifi
(the first one hangs and has some other problems when I enable its
wifi support,
although only with linux).
The wifi problem I reported has dissapear (it occurs once a week or so)
after I upgraded the Lenovo W500 to 3.3.4-1~experimental.1
and decided to install
compat-wireless-3.4-rc3-1.tar.bz2
from
http://linuxwireless.org/en/users/Download/stable/#compat-wireless_3.4_stable_releases
At first it was an small improvement (it disconnected 2~3 times each hour)
but then I played with options and with
options iwlagn ack_check=1 fw_restart=1 11n_disable=1 bt_coex_active=0
options iwlwifi ack_check=1 fw_restart=1 11n_disable=1 bt_coex_active=0
options cfg80211 ieee80211_regdom=EU
it now disconnects once or twice a week.
I think Olivier found the same solution for his problem in
http://bugs.debian.org/cgi-bin/bugreport.cgi?bug=668547#40
The strange thing is that I've made the options change only in the
Lenovo machine,
the other machines disconnect once or twice too without any change.
I should have tested more, but since it works and I do not know which
is the problem
I decided to stop making changes.
Just let me know if you want me to test something else.
If not I think bug should be closed.
Regards,
Victor.
--
--
(Groucho Marx) Those are my principles, and if you don't like them ...
well, I have others.
---------------------------------------------------------------------------------------------
Víctor Pablos Ceruelo - Software Engineer, Phd student.
Web page: http://babel.ls.fi.upm.es/~vpablos/
---------------------------------------------------------------------------------------------
--
To UNSUBSCRIBE, email to debian-kernel-REQUEST@lists.debian.org
with a subject of "unsubscribe". Trouble? Contact listmaster@lists.debian.org
Archive: http://lists.debian.org/CAF1aHE=iGnEheqfhhvh6yPYKGQg3cQdy9DHBawivM=7VMe4A9 Q@mail.gmail.com
|
