Google Groups no longer supports new Usenet posts or subscriptions. Historical content remains viewable.
Dismiss
Bug#659521: ath9k: ATH9K driver will not connect to my adhoc network but works with managed (router) connections
7 views
Skip to first unread message
Ankman
Feb 11, 2012, 3:10:02 PM2/11/12
to
Package: linux-headers-3.2.0-1-686-pae
Version: 3.2.4-1
Severity: important
File: ath9k
Dear Maintainer,
*** Please consider answering these questions, where appropriate ***
* What led up to the situation?
Connecting to an adhoc (IBSS) node.
* What exactly did you do (or not do) that was effective (or
ineffective)?
Connection to adhoc won't work. Booting the previous (3.1.0-1-686-pae) and it works again. Although in 3.2.0-1-686-pae managed (router bound) connection DO work.
* What was the outcome of this action?
Adhoc connections (at least to my other computer, which worked before since months) do not work with the ATH9K, but managed connection are not affected by this bug (if it's a bug).
* What outcome did you expect instead?
Espected to connect in adhoc mode.
-- System Information:
Debian Release: wheezy/sid
APT prefers testing-proposed-updates
APT policy: (500, 'testing-proposed-updates'), (500, 'testing')
Architecture: i386 (i686)
Kernel: Linux 3.2.0-1-686-pae (SMP w/2 CPU cores)
Locale: LANG=en_CA.utf8, LC_CTYPE=en_CA.utf8 (charmap=UTF-8)
Shell: /bin/sh linked to /bin/dash
Versions of packages linux-headers-3.2.0-1-686-pae depends on:
ii gcc-4.6 4.6.2-12
ii linux-headers-3.2.0-1-common 3.2.4-1
ii linux-kbuild-3.2 3.2.1-2
linux-headers-3.2.0-1-686-pae recommends no packages.
linux-headers-3.2.0-1-686-pae suggests no packages.
-- no debconf information
--
To UNSUBSCRIBE, email to debian-bugs-...@lists.debian.org
with a subject of "unsubscribe". Trouble? Contact listm...@lists.debian.org
Version: 3.2.4-1
Severity: important
File: ath9k
Dear Maintainer,
*** Please consider answering these questions, where appropriate ***
* What led up to the situation?
Connecting to an adhoc (IBSS) node.
* What exactly did you do (or not do) that was effective (or
ineffective)?
Connection to adhoc won't work. Booting the previous (3.1.0-1-686-pae) and it works again. Although in 3.2.0-1-686-pae managed (router bound) connection DO work.
* What was the outcome of this action?
Adhoc connections (at least to my other computer, which worked before since months) do not work with the ATH9K, but managed connection are not affected by this bug (if it's a bug).
* What outcome did you expect instead?
Espected to connect in adhoc mode.
-- System Information:
Debian Release: wheezy/sid
APT prefers testing-proposed-updates
APT policy: (500, 'testing-proposed-updates'), (500, 'testing')
Architecture: i386 (i686)
Kernel: Linux 3.2.0-1-686-pae (SMP w/2 CPU cores)
Locale: LANG=en_CA.utf8, LC_CTYPE=en_CA.utf8 (charmap=UTF-8)
Shell: /bin/sh linked to /bin/dash
Versions of packages linux-headers-3.2.0-1-686-pae depends on:
ii gcc-4.6 4.6.2-12
ii linux-headers-3.2.0-1-common 3.2.4-1
ii linux-kbuild-3.2 3.2.1-2
linux-headers-3.2.0-1-686-pae recommends no packages.
linux-headers-3.2.0-1-686-pae suggests no packages.
-- no debconf information
--
To UNSUBSCRIBE, email to debian-bugs-...@lists.debian.org
with a subject of "unsubscribe". Trouble? Contact listm...@lists.debian.org
Jonathan Nieder
Feb 11, 2012, 6:40:02 PM2/11/12
to
reassign 659521 src:linux-2.6 3.2.4-1
quit
Hi,
Ankman wrote:
> Connection to adhoc won't work. Booting the previous
> (3.1.0-1-686-pae) and it works again. Although in 3.2.0-1-686-pae
> managed (router bound) connection DO work.
Thanks for reporting it. Could you elaborate? In particular:
- please attach output from
"reportbug --template linux-image-$(uname -r)", so we can get to
know your configuration and hardware
- what commands do you use to set up an ad hoc connection? See [1]
for hints on ruling out problems from networkmanager and other
higher-level tools when working on wireless bugs, and for hints on
getting an event log. How do you know it failed? Any other hints
or weird symptoms?
- please attach full "dmesg" output from booting up and attempting to
connect in ad-hoc mode.
If you have time to test the attached patch[2], that might also be
useful.
Hope that helps,
Jonathan
[1] http://linuxwireless.org/en/users/Documentation/Reporting_bugs
[2] see http://kernel-handbook.alioth.debian.org/ch-common-tasks.html#s4.2.2
or the corresponding page in the debian-kernel-handbook package for
instructions.
quit
Hi,
Ankman wrote:
> Connection to adhoc won't work. Booting the previous
> (3.1.0-1-686-pae) and it works again. Although in 3.2.0-1-686-pae
> managed (router bound) connection DO work.
- please attach output from
"reportbug --template linux-image-$(uname -r)", so we can get to
know your configuration and hardware
- what commands do you use to set up an ad hoc connection? See [1]
for hints on ruling out problems from networkmanager and other
higher-level tools when working on wireless bugs, and for hints on
getting an event log. How do you know it failed? Any other hints
or weird symptoms?
- please attach full "dmesg" output from booting up and attempting to
connect in ad-hoc mode.
If you have time to test the attached patch[2], that might also be
useful.
Hope that helps,
Jonathan
[1] http://linuxwireless.org/en/users/Documentation/Reporting_bugs
[2] see http://kernel-handbook.alioth.debian.org/ch-common-tasks.html#s4.2.2
or the corresponding page in the debian-kernel-handbook package for
instructions.
Jonathan Nieder
Feb 11, 2012, 7:40:01 PM2/11/12
to
Ankman wrote:
> This (attachment "report.txt") will be from my currently booted kernel
> (3.1.0-1-686-pae)
Thanks! That's fine. Forwarded to the bug log.
[...]
> I solely rely on a script I wrote (key and essid were replaced by me),
> no network manager involved:
>
> ifconfig wlan0 down
> iwconfig wlan0 mode ad-hoc
> iwconfig wlan0 my-essid
> iwconfig wlan0 key 12345
> ifconfig wlan0 192.168.198.251
> route add default gw 192.168.198.78
>
> Above works reliable for me since months. Until yesterday when the
> kernel-update was installed
Ok, please send a summary of this problem upstream (that's
ath9k...@lists.ath9k.org, cc-ing linux-w...@vger.kernel.org,
linux-...@vger.kernel.org, and either me or this bug log so we can
track it).
Be sure to mention the steps to reproduce, expected result, actual
result, and versions tested, and what kind of testing you would and
would not be able to do to track this down, just as you have here, and
include a link to this bug log for the full story.
>> If you have time to test the attached patch[2], that might also be
>> useful.
>
> Sorry, I don't have much time for that.
That's fine. If we're lucky, someone more familiar with these things
upstream can come up with some way to confirm or rule out that cause
without having to build a new kernel.
Thanks for the quick feedback.
Jonathan
> This (attachment "report.txt") will be from my currently booted kernel
> (3.1.0-1-686-pae)
Thanks! That's fine. Forwarded to the bug log.
[...]
> I solely rely on a script I wrote (key and essid were replaced by me),
> no network manager involved:
>
> ifconfig wlan0 down
> iwconfig wlan0 mode ad-hoc
> iwconfig wlan0 my-essid
> iwconfig wlan0 key 12345
> ifconfig wlan0 192.168.198.251
> route add default gw 192.168.198.78
>
> Above works reliable for me since months. Until yesterday when the
> kernel-update was installed
Ok, please send a summary of this problem upstream (that's
ath9k...@lists.ath9k.org, cc-ing linux-w...@vger.kernel.org,
linux-...@vger.kernel.org, and either me or this bug log so we can
track it).
Be sure to mention the steps to reproduce, expected result, actual
result, and versions tested, and what kind of testing you would and
would not be able to do to track this down, just as you have here, and
include a link to this bug log for the full story.
>> If you have time to test the attached patch[2], that might also be
>> useful.
>
That's fine. If we're lucky, someone more familiar with these things
upstream can come up with some way to confirm or rule out that cause
without having to build a new kernel.
Thanks for the quick feedback.
Jonathan
Jonathan Nieder
Feb 12, 2012, 12:50:02 PM2/12/12
to
Hi,
Ankman wrote:
> Am I supposed to send all I sent to you previously to the addresses
> given in this mail? So the dmesg log and my setup? Can't you do it
> instead? I don't want to make a mistake.
More important than the logs (which they can get e.g. through a link to
<http://bugs.debian.org/659519>) is a succinct summary of the problem.
I usually ask bug reporters to make the first contact with upstream
so they can add additional information if appropriate and choose an
appropriate time to be able to reply to followups if busy.
> I also might later (cannot guarantee though) be able to apply the
> patch you provided. But I didn't apply pages since quite some time and
> have not much time. If you are interested that I try this can you
> please provide me with all console commands how I do this?
Thanks! Here's a summary of the page[1] I mentioned before:
# as root:
apt-get source linux-2.6
apt-get install build-essential fakeroot devscripts
apt-get build-dep linux-2.6
# as a normal user:
cd linux-2.6-<version>
bash debian/bin/test-patches <path to patch>
# or, for parallel build:
bash debian/bin/test-patches -j <num> <path to patch>
# as root:
dpkg -i ../<name of package>
reboot
[...]
with the "ip" and "iw" tools from the iproute and iw packages. Something
like this:
ip link set wlan0 down
ip link set wlan0 up
iw dev wlan0 interface add ah0 type ibss
ip link set ah0 up
iw dev ah0 ibss join my-essid 2412
ip address add 192.168.198.251 dev wlan0
ip route add default via 192.168.198.78 dev wlan0
...
iw dev ah0 ibss leave
ip link set ah0 down
ip link set wlan0 down
Running "iw event -t" in another terminal while doing this can give
interesting progress information. And "dmesg" tends to report errors,
as usual.
Hope that helps,
Jonathan
[1] http://kernel-handbook.alioth.debian.org/ch-common-tasks.html#s4.2.2
Ankman wrote:
> Am I supposed to send all I sent to you previously to the addresses
> given in this mail? So the dmesg log and my setup? Can't you do it
> instead? I don't want to make a mistake.
More important than the logs (which they can get e.g. through a link to
<http://bugs.debian.org/659519>) is a succinct summary of the problem.
I usually ask bug reporters to make the first contact with upstream
so they can add additional information if appropriate and choose an
appropriate time to be able to reply to followups if busy.
> I also might later (cannot guarantee though) be able to apply the
> patch you provided. But I didn't apply pages since quite some time and
> have not much time. If you are interested that I try this can you
> please provide me with all console commands how I do this?
Thanks! Here's a summary of the page[1] I mentioned before:
# as root:
apt-get source linux-2.6
apt-get install build-essential fakeroot devscripts
apt-get build-dep linux-2.6
# as a normal user:
cd linux-2.6-<version>
bash debian/bin/test-patches <path to patch>
# or, for parallel build:
bash debian/bin/test-patches -j <num> <path to patch>
# as root:
dpkg -i ../<name of package>
reboot
[...]
>>> ifconfig wlan0 down
>>> iwconfig wlan0 mode ad-hoc
>>> iwconfig wlan0 my-essid
>>> iwconfig wlan0 key 12345
>>> ifconfig wlan0 192.168.198.251
>>> route add default gw 192.168.198.78
As David Goodenough hinted, upstream might be more comfortable working
>>> iwconfig wlan0 mode ad-hoc
>>> iwconfig wlan0 my-essid
>>> iwconfig wlan0 key 12345
>>> ifconfig wlan0 192.168.198.251
>>> route add default gw 192.168.198.78
with the "ip" and "iw" tools from the iproute and iw packages. Something
like this:
ip link set wlan0 down
ip link set wlan0 up
iw dev wlan0 interface add ah0 type ibss
ip link set ah0 up
iw dev ah0 ibss join my-essid 2412
ip address add 192.168.198.251 dev wlan0
ip route add default via 192.168.198.78 dev wlan0
...
iw dev ah0 ibss leave
ip link set ah0 down
ip link set wlan0 down
Running "iw event -t" in another terminal while doing this can give
interesting progress information. And "dmesg" tends to report errors,
as usual.
Hope that helps,
Jonathan
[1] http://kernel-handbook.alioth.debian.org/ch-common-tasks.html#s4.2.2
Jonathan Nieder
Feb 12, 2012, 2:00:02 PM2/12/12
to
(Resending to the correct bug. Sorry for the duplicate messages, Ankman
and David.)
Jonathan
[1] http://kernel-handbook.alioth.debian.org/ch-common-tasks.html#s4.2.2
and David.)
>>> ifconfig wlan0 down
>>> iwconfig wlan0 mode ad-hoc
>>> iwconfig wlan0 my-essid
>>> iwconfig wlan0 key 12345
>>> ifconfig wlan0 192.168.198.251
>>> route add default gw 192.168.198.78
>>> iwconfig wlan0 mode ad-hoc
>>> iwconfig wlan0 my-essid
>>> iwconfig wlan0 key 12345
>>> ifconfig wlan0 192.168.198.251
>>> route add default gw 192.168.198.78
As David Goodenough hinted, upstream might be more comfortable working
with the "ip" and "iw" tools from the iproute and iw packages. Something
like this:
ip link set wlan0 down
ip link set wlan0 up
iw dev wlan0 interface add ah0 type ibss
ip link set ah0 up
iw dev ah0 ibss join my-essid 2412
ip address add 192.168.198.251 dev wlan0
ip route add default via 192.168.198.78 dev wlan0
...
iw dev ah0 ibss leave
ip link set ah0 down
ip link set wlan0 down
Running "iw event -t" in another terminal while doing this can give
interesting progress information. And "dmesg" tends to report errors,
as usual.
Hope that helps,
with the "ip" and "iw" tools from the iproute and iw packages. Something
like this:
ip link set wlan0 down
ip link set wlan0 up
iw dev wlan0 interface add ah0 type ibss
ip link set ah0 up
iw dev ah0 ibss join my-essid 2412
ip address add 192.168.198.251 dev wlan0
ip route add default via 192.168.198.78 dev wlan0
...
iw dev ah0 ibss leave
ip link set ah0 down
ip link set wlan0 down
Running "iw event -t" in another terminal while doing this can give
interesting progress information. And "dmesg" tends to report errors,
as usual.
Jonathan
[1] http://kernel-handbook.alioth.debian.org/ch-common-tasks.html#s4.2.2
Ankman
Feb 12, 2012, 2:00:03 PM2/12/12
to
Hello,
as this is my first bug report ever I'm not sure if I understand.
Am I supposed to send all I sent to you previously to the addresses
given in this mail? So the dmesg log and my setup? Can't you do it
instead? I don't want to make a mistake.
as this is my first bug report ever I'm not sure if I understand.
Am I supposed to send all I sent to you previously to the addresses
given in this mail? So the dmesg log and my setup? Can't you do it
instead? I don't want to make a mistake.
I also might later (cannot guarantee though) be able to apply the
patch you provided. But I didn't apply pages since quite some time and
have not much time. If you are interested that I try this can you
please provide me with all console commands how I do this?
With kind regards, Ankman
patch you provided. But I didn't apply pages since quite some time and
have not much time. If you are interested that I try this can you
please provide me with all console commands how I do this?
Jonathan Nieder
Feb 12, 2012, 7:20:02 PM2/12/12
to
Jonathan Nieder wrote:
> Forwarding to the bug log.
Oh, dear. Now I've reread your message. Sorry for publishing your
email address (in message ID <20120213000954.GB2256@burratino>, sent
to bug 659521). Don, I assume it is too late to remove that message
from the archive? :/
Unhappily,
> Forwarding to the bug log.
Oh, dear. Now I've reread your message. Sorry for publishing your
email address (in message ID <20120213000954.GB2256@burratino>, sent
to bug 659521). Don, I assume it is too late to remove that message
from the archive? :/
Unhappily,
Jonathan Nieder
Feb 12, 2012, 7:20:02 PM2/12/12
to
Forwarding to the bug log. Please use your mailer's reply-to-all
feature in the future so others can pick up the thread if I end up too
busy or miss important details.
feature in the future so others can pick up the thread if I end up too
busy or miss important details.
Jonathan Nieder
Feb 12, 2012, 7:40:01 PM2/12/12
to
Jonathan Nieder wrote:
> Oh, dear. Now I've reread your message. Sorry for publishing your
> email address (in message ID <20120213000954.GB2256@burratino>, sent
> to bug 659521). Don, I assume it is too late to remove that message
> from the archive? :/
Same question for listmaster (for http://lists.debian.org/debian-kernel).
> Oh, dear. Now I've reread your message. Sorry for publishing your
> email address (in message ID <20120213000954.GB2256@burratino>, sent
> to bug 659521). Don, I assume it is too late to remove that message
> from the archive? :/
I realize the mistake cannot be taken back and the message is already
public, but maybe it is possible to mitigate the damage by removing the
message from public archives.
I've contacted Lars from gmane separately. mail-archive.com already
munges email addresses. I'm not aware of any other public archives
of debian-kernel@ and debian-bugs-dist@, though that doesn't mean
they don't exist. I've read
http://www.debian.org/MailingLists/disclaimer
which makes it clear I cannot *demand* that you remove my message, but
maybe I am lucky and politely asking it can work so another does not
suffer for my mistake. (Information on that page either way about
whether old messages can be removed and under what circumstances would
be welcome, by the way.)
Thanks and sorry,
Jonathan Nieder
Feb 14, 2012, 7:20:01 PM2/14/12
to
tags 659927 + patch moreinfo
quit
Hi,
Ankman wrote:
> as this is my first bug report ever I'm not sure if I understand.
Sorry for the roller coaster ride.
If I understand correctly, your ad hoc network uses WEP. That's very
good news, since it means there's a better patch to try.
See [*] for instructions if you'd like to test it. The patch is
queued for inclusion in v3.2.7 and could be worth including in the
next upload for unstable (which might be based on 3.2.6), too.
Thanks again,
Jonathan
[*] http://bugs.debian.org/cgi-bin/bugreport.cgi?msg=71;bug=659521
quit
Hi,
Ankman wrote:
> as this is my first bug report ever I'm not sure if I understand.
If I understand correctly, your ad hoc network uses WEP. That's very
good news, since it means there's a better patch to try.
See [*] for instructions if you'd like to test it. The patch is
queued for inclusion in v3.2.7 and could be worth including in the
next upload for unstable (which might be based on 3.2.6), too.
Thanks again,
Jonathan
[*] http://bugs.debian.org/cgi-bin/bugreport.cgi?msg=71;bug=659521
Jonathan Nieder
Feb 15, 2012, 4:30:02 PM2/15/12
to
tags 659927 - moreinfo
merge 659521 659927
quit
(cc-ing the bug log with permission)
Ankman wrote:
WPA should already work if I understand correctly --- the bug only
affects WEP.
>> [*] http://bugs.debian.org/cgi-bin/bugreport.cgi?msg=71;bug=659521
>
> I followed the instructions but had to download the git thingy and all
> from my working kernel, which worked. Then I booted the not working
> kernel (3.2.x) to apply to patch but failed as I seem not to be able
> to relocate the path to the repository. I get "... is outside
> repository". I searched for "git" and "repository" and cd'ed in to
> what I found, but no luck.
>
> Please let me know the path I have to cd in.
Thanks. Could you provide the exact error message?
The repository is a directory named "linux" under whichever directory
you were in when you cloned the repo before. It contains the Linux
source code.
Hopefully something like
cd linux
git apply --index <thepatch>
merge 659521 659927
quit
(cc-ing the bug log with permission)
Ankman wrote:
> On Tue, Feb 14, 2012 at 19:13, Jonathan Nieder <jrni...@gmail.com> wrote:
>> If I understand correctly, your ad hoc network uses WEP. That's very
>> good news, since it means there's a better patch to try.
>
> Yes, just WEP so far, but might want to use WPA in after this issue is solved.
>> If I understand correctly, your ad hoc network uses WEP. That's very
>> good news, since it means there's a better patch to try.
>
WPA should already work if I understand correctly --- the bug only
affects WEP.
>> [*] http://bugs.debian.org/cgi-bin/bugreport.cgi?msg=71;bug=659521
>
> I followed the instructions but had to download the git thingy and all
> from my working kernel, which worked. Then I booted the not working
> kernel (3.2.x) to apply to patch but failed as I seem not to be able
> to relocate the path to the repository. I get "... is outside
> repository". I searched for "git" and "repository" and cd'ed in to
> what I found, but no luck.
>
> Please let me know the path I have to cd in.
Thanks. Could you provide the exact error message?
The repository is a directory named "linux" under whichever directory
you were in when you cloned the repo before. It contains the Linux
source code.
Hopefully something like
cd linux
git apply --index <thepatch>
dpkg -i ../<name of package>
reboot
will work.
reboot
Jonathan Nieder
Feb 16, 2012, 1:40:02 PM2/16/12
to
(cc-ing the bug log with permission)
Ankman wrote:
Ankman wrote:
> On Wed, Feb 15, 2012 at 16:19, Jonathan Nieder <jrni...@gmail.com> wrote:
>> cd linux
>> git apply --index <thepatch>
>> make deb-pkg
>> cd linux
>> git apply --index <thepatch>
>> dpkg -i ../<name of package>
>> reboot
>
> I was starting this. Besides taking very long (need to postpone it to
>> reboot
>
> the weekend then, sorry) would the created and then installed package
> also get used by other kernels? I just fear that in case this fails
> (although I doubt) I won't be able to get online at all then. If this
> would apply to all kernels then I would rather not want to apply this
> now.
These commands build a new kernel image package, which would show up
in the GRUB menu as an option at boot time like the rest of them. It
does not affect any other kernel versions that might already be
installed.
The next upload of linux in Debian sid includes this patch and will
probably show up in the archive some time in the next few days.
Hope that helps,
Jonathan
Jonathan Nieder
Feb 16, 2012, 2:00:02 PM2/16/12
to
Ankman wrote:
> Does this mean it's approved already and you don't need me to test the
> patch at all? As I won't be able to test it until Saturday or Sunday
> anyway.
That's right. The patch was accepted upstream and we've had some
confirmation that it works.
Soon enough we'll see whether it works on your system, as the fix gets
uploaded and then makes it way through Debian mirrors and to your
machine when you upgrade.
Thanks again for your help,
> Does this mean it's approved already and you don't need me to test the
> patch at all? As I won't be able to test it until Saturday or Sunday
> anyway.
That's right. The patch was accepted upstream and we've had some
confirmation that it works.
Soon enough we'll see whether it works on your system, as the fix gets
uploaded and then makes it way through Debian mirrors and to your
machine when you upgrade.
Thanks again for your help,
Jonathan Nieder
Feb 16, 2012, 2:20:02 PM2/16/12
to
Ankman wrote:
> Thanks Jonathan, I appreciate all your and the other maintainers work.
>
> And sorry again for all the trouble I caused.
>
> I hope next time I should find a bug I will not be so stupid again.
No big problem. We all learn, and I never got the impression you were
stupid here.
Thanks for working to improve Debian.
Sincerely,
> Thanks Jonathan, I appreciate all your and the other maintainers work.
>
> And sorry again for all the trouble I caused.
>
> I hope next time I should find a bug I will not be so stupid again.
No big problem. We all learn, and I never got the impression you were
stupid here.
Thanks for working to improve Debian.
Sincerely,
0 new messages