ndiswrapper fails to connect in Edgy 2.6.17-11 and Feisty Fawn Herd 4

Asked by TFrog

On my Compaq Presario R4125 US (Broadcom 4318 chipset) ndiswrapper (all versions - i.e. repository versions as well as source from project site) install fine on 2.6.17-11 Edgy and load the Windows driver properly. However, they fail to connect whatsoever. On Feisty Fawn Herd 4 which I finished downloading via bit torrent yesterday, I have the same issue with the exception of the fact that I can't compile ndiswrapper from source. Lots of errors. Either way, neither will connect for the moment. I've even removed the blacklisted the bcm43xx "native" driver and the acx modules to no avail. On 2.6.17-10 I can run just about any Broadcom Windows driver. Currently using the most current for this laptop from HP/Compaq. I am currently posting this via Edgy Eft 6.10 kernel 2.6.17-10. Should you find out why this is happening, I'm curious to know what exactly is going on.

Question information

Language:
English Edit question
Status:
Answered
For:
Ubuntu Edit question
Assignee:
No assignee Edit question
Last query:
Last reply:
Revision history for this message
williamts99 (williamts99) said :
#1

I believe this should be filed as a bug so that the developers see it.

Best Regards,
Williamts99

Revision history for this message
TFrog (stgparris) said :
#2

Thank you for your speedy reply. The bug has been submitted. It actually
affects both kernels in Edgy Eft but was able to compile source code version
1.27 and 1.37 from ndiswrapper project on both kernels for Edgy Eft. Can't
compile any ndiswrapper on a fresh install of Kubuntu 7.04 without errors
and failure to load. All versions of ubuntu ndiswrapper 1.1, 1.8, and 1.9
load and unload the drivers correctly but fail to connect. The bug was
reported with the above description with a little more detail.

>From: williamts99 <email address hidden>
>Reply-To: williamts99 <email address hidden>
>To: <email address hidden>
>Subject: Re: [Support #3786]: ndiswrapper fails to connect in Edgy
>2.6.17-11and Feisty Fawn Herd 4
>Date: Mon, 19 Feb 2007 23:07:07 -0000
>
>Your support request #3786 on Ubuntu changed:
>https://answers.launchpad.net/ubuntu/+ticket/3786
>
> Status: Open => Answered
>
>williamts99 proposed the following answer:
>I believe this should be filed as a bug so that the developers see it.
>
>Best Regards,
>Williamts99
>
>_______________________________________________________________________
>If this answers your request, please go to the following page to let us
>know that it is solved:
>https://answers.launchpad.net/ubuntu/+ticket/3786/+confirm?answer_id=0
>
>If you still need support, you can reply to this email or go to the
>following page to enter your feedback:
>https://answers.launchpad.net/ubuntu/+ticket/3786

_________________________________________________________________
http://homepage.msn.com/zune?icid=hmetagline

Revision history for this message
TFrog (stgparris) said :
#3

I was sent a link by your development team for a newer version of ndiswrapper compiled with 1.30 version of the source code. As of last night I was connected to the Internet for a brief time. This morning when I attempted to connect with Feisty, it yet again failed to connect. It reached a 28% level and froze there in an attempt to connect to an open (no WPA or WEP) network. Will attempt again later to see if there is any change in the way it's acting. I do understand that Feisty is still in "alpha" stages so it's possible this will be fixed by the time it's final.

Revision history for this message
kosson (kosson) said :
#4

I'll give it a try with Faisty Fawn Hurd 4. I hope for success...

Revision history for this message
Joost Ringoot (joost) said :
#5

I have the same wirelless broadcom, never used in feisty fawn, no need for.

But since last week, I get memory leaks. Causing heavy use of swap even when no application is doing something "big" on the CPU, CPU is free for 98%.
I got lots of lines in syslog about problems with bcm43xx.
Unloading the bcm43xx: "rmmod bcm43xx" appears to avoid the memory leaks. But that does not free the memory, it must be done before the leaking started.

I don't know how to properly permanently disable this driver. I was considering to put a small file in /etc/init.d with this line "rmmod bcm43xx" and then add a link to runlevel.Because
I can't find what replaces /etc/inittab, but I will RTFM ubuntu to find out.

suggestions are welcome.

Revision history for this message
Joost Ringoot (joost) said :
#6

Sorry I hit the wrong button, should have been "Add information request." Can I undo this?

Revision history for this message
Joost Ringoot (joost) said :
#7

I've found the "/etc/event.d" (as replacment for /etc/inittab) I've put the "rmmod bcm43xx" somwhere in "/etc/event.d/rc2".
Really ugly,...sigh...
See if it is sufficient to counter the memory leak(s).

Suggestions are still welcome, should I make a bugreport?

Revision history for this message
Joost Ringoot (joost) said :
#8

Bad news, the leak is back, even though the modification of "/etc/event.d/rc2" did what I would expect: unload the wireless driver.

At least /var/log/syslog stays clean :-}

Does anyone have an idea/suggestions?
Why the memory fills up so fast?

Reached now 445052k in use from 450472k total, after 38 minutes uptime. This without doing anything special: opened some files via nautilus with gedit, created and archive, emailed it with evolution.

System is up to date with automatic updater, problems did not appear before saturday.

Can you help with this problem?

Provide an answer of your own, or ask TFrog for more information if necessary.

To post a message you must log in.