FAQ Search Today's Posts Mark Forums Read
» Video Reviews

» Linux Archive

Linux-archive is a website aiming to archive linux email lists and to make them easily accessible for linux users/developers.


» Sponsor

» Partners

» Sponsor

Go Back   Linux Archive > Debian > Debian User

 
 
LinkBack Thread Tools
 
Old 03-08-2011, 02:09 AM
Kyle Barbour
 
Default events/0 and events/1 maxing CPU to 100%, related to wireless or eth0 on a T61 Thinkpad

Hello all,

I upgraded to Debian Squeeze a week ago on my Lenovo Thinkpad T61.
Starting earlier today, after no trigger that I could determine, my
CPU (an Intel Core 2 Duo) began maxing out on one or both cores to the
point where doing anything on the computer was incredibly choppy. I
initially thought that this might have something to do with running
compiz, which had been problematic a year ago, but the choppiness and
CPU maxing was observed after running "$ metacity --replace &" and in
the gdm login screen after a reboot.

I then thought that this might have something to do with my wireless,
which is frequently dropping connections, having difficulty
connecting, and from which kerneloops has been sending off kernel
errors for the past few days (my kernel is 2.6.32-5-686 from the
Squeeze repos). I use wicd as a wireless manager, however, killing the
wicd process and "# /etc/init.d/wicd stop" failed to help. This still
seems like a possible problem source, as I've seen some posts that
suggest that removing wicd entirely and rebooting solves the problem
even when stopping the process doesn't (such as in the post cited
below). However, as removing wicd eliminates what is currently my only
way of accessing the internet, I haven't tried that. Further, what I'm
about to describe seems to possibly contradict this being a wireless
issue (although it might still be).

I eventually found this post,
https://bbs.archlinux.org/viewtopic.php?id=88781, which suggests
running shutting down the eth0 interface ("# ifconfig eth0 down").
Shockingly, that actually worked, although the CPU still frequently
runs high. Turning eth0 back on causes the CPU to max out again. This
fix is imperfect since I'd like to be able to use the eth0 interface,
but it works at the moment.

Any idea what's going on here and/or how to fix it?

Kyle


--
Kyle Barbour
kyle@kylebarbour.org
(415) 238-9953
San Francisco, CA, USA


--
To UNSUBSCRIBE, email to debian-user-REQUEST@lists.debian.org
with a subject of "unsubscribe". Trouble? Contact listmaster@lists.debian.org
Archive: AANLkTi=n+Z3VzwgxaBUEpuwN9jkBLwARbrCW_j-eUhfi@mail.gmail.com">http://lists.debian.org/AANLkTi=n+Z3VzwgxaBUEpuwN9jkBLwARbrCW_j-eUhfi@mail.gmail.com
 
Old 03-08-2011, 02:24 AM
Slicky Johnson
 
Default events/0 and events/1 maxing CPU to 100%, related to wireless or eth0 on a T61 Thinkpad

On Mon, 7 Mar 2011 19:09:25 -0800
Kyle Barbour <kyle@kylebarbour.org> wrote:

> Hello all,
>
> I upgraded to Debian Squeeze a week ago on my Lenovo Thinkpad T61.
> Starting earlier today, after no trigger that I could determine, my
> CPU (an Intel Core 2 Duo) began maxing out on one or both cores to the
> point where doing anything on the computer was incredibly choppy. I
> initially thought that this might have something to do with running
> compiz, which had been problematic a year ago, but the choppiness and
> CPU maxing was observed after running "$ metacity --replace &" and in
> the gdm login screen after a reboot.
>
> I then thought that this might have something to do with my wireless,
> which is frequently dropping connections, having difficulty
> connecting, and from which kerneloops has been sending off kernel
> errors for the past few days (my kernel is 2.6.32-5-686 from the
> Squeeze repos). I use wicd as a wireless manager, however, killing the
> wicd process and "# /etc/init.d/wicd stop" failed to help. This still
> seems like a possible problem source, as I've seen some posts that
> suggest that removing wicd entirely and rebooting solves the problem
> even when stopping the process doesn't (such as in the post cited
> below). However, as removing wicd eliminates what is currently my only
> way of accessing the internet, I haven't tried that. Further, what I'm
> about to describe seems to possibly contradict this being a wireless
> issue (although it might still be).
>
> I eventually found this post,
> https://bbs.archlinux.org/viewtopic.php?id=88781, which suggests
> running shutting down the eth0 interface ("# ifconfig eth0 down").
> Shockingly, that actually worked, although the CPU still frequently
> runs high. Turning eth0 back on causes the CPU to max out again. This
> fix is imperfect since I'd like to be able to use the eth0 interface,
> but it works at the moment.
>
> Any idea what's going on here and/or how to fix it?
>
> Kyle
>
>

I have the identical hardware and also use wicd. I've not experienced
any of the above. I've commented out eth0 from /etc/network/interfaces
as I don't even own a cable.

# The loopback network interface
auto lo
iface lo inet loopback

# The primary network interface

#allow-hotplug eth0
#iface eth0 inet dhcp

It's been that way since Lenny. Have you looked around
http://www.thinkwiki.org/wiki/ThinkWiki for possible problems? This
would be the first I've heard of it and any time something with a T61
comes up I pay pretty close attention.


--
To UNSUBSCRIBE, email to debian-user-REQUEST@lists.debian.org
with a subject of "unsubscribe". Trouble? Contact listmaster@lists.debian.org
Archive: 20110307222445.4f9c4f6e@t61.debian-linux">http://lists.debian.org/20110307222445.4f9c4f6e@t61.debian-linux
 
Old 03-08-2011, 02:29 AM
Ron Johnson
 
Default events/0 and events/1 maxing CPU to 100%, related to wireless or eth0 on a T61 Thinkpad

On 03/07/2011 09:09 PM, Kyle Barbour wrote:

Hello all,

I upgraded to Debian Squeeze a week ago on my Lenovo Thinkpad T61.
Starting earlier today, after no trigger that I could determine, my

[snip]


I eventually found this post,
https://bbs.archlinux.org/viewtopic.php?id=88781, which suggests
running shutting down the eth0 interface ("# ifconfig eth0 down").
Shockingly, that actually worked, although the CPU still frequently
runs high. Turning eth0 back on causes the CPU to max out again. This
fix is imperfect since I'd like to be able to use the eth0 interface,
but it works at the moment.

Any idea what's going on here and/or how to fix it?



Anything in syslog?

--
I prefer banana-flavored energy bars made from tofu.


--
To UNSUBSCRIBE, email to debian-user-REQUEST@lists.debian.org
with a subject of "unsubscribe". Trouble? Contact listmaster@lists.debian.org

Archive: 4D75A291.2030004@cox.net">http://lists.debian.org/4D75A291.2030004@cox.net
 
Old 03-08-2011, 06:39 AM
Kyle Barbour
 
Default events/0 and events/1 maxing CPU to 100%, related to wireless or eth0 on a T61 Thinkpad

O.K., so amazingly this is solved, at least on the practical end. I
have no idea why this works, and I've provided some more information
below if people want to pursue this or for others trying to solve this
problem, but the solution that worked for me was powering off the
computer, unplugging it, removing the battery, waiting 30 seconds,
plugging it back in, reattaching the battery, and powering back up.
I've been monitoring htop and CPU usage is normal with eth0 up. No
idea if eth0 works as I can't currently test it, but I have no reason
to believe that it doesn't.

Wireless is still wonky, but that's another issue, even if it's
possibly related.

This solution was found from here:
https://bugs.launchpad.net/ubuntu/+source/linux/+bug/53739

Thanks for all the help! More information is below for anyone who wants it.

Kyle

vvvvvvvvvvvvvvvvvvvvvvvvvvvvvvvvvvvvv

On Mon, Mar 7, 2011 at 7:24 PM, Slicky Johnson <slickyjohnson@gmail.com> wrote:
> I have the identical hardware and also use wicd. I've not experienced
> any of the above. I've commented out eth0 from /etc/network/interfaces
> as I don't even own a cable.
>
> # The loopback network interface
> auto lo
> iface lo inet loopback
>
> # The primary network interface
>
> #allow-hotplug eth0
> #iface eth0 inet dhcp

Huh. This is identical to my /etc/network/interfaces file, actually.

> It's been that way since Lenny. Have you looked around
> http://www.thinkwiki.org/wiki/ThinkWiki for possible problems? This
> would be the first I've heard of it and any time something with a T61
> comes up I pay pretty close attention.

That's a good idea, I hadn't checked there. Unfortunately, I don't see
anything there, either by searching around, by checking the
Category:T61 page (http://www.thinkwiki.org/wiki/Category:T61), the
page for my wireless card (which if I remember correctly is
http://www.thinkwiki.org/wiki/Intel_PRO/Wireless_3945ABG_Mini-PCI_Express_Adapter),
or the page for what they say is my ethernet card
(http://www.thinkwiki.org/wiki/Intel_Gigabit_Ethernet_(10/100/1000)_PCI-Express).
However, you're right that it may be a Thinkpad issue - the people on
the forum I quoted earlier had a Thinkpad X61, which has the same
wireless and ethernet cards.

However, after searching around, there's other reports of similar
behavior going back a while, such as here,
http://lkml.indiana.edu/hypermail/linux/kernel/0605.2/0212.html,
although I couldn't get much out of that.

On Mon, Mar 7, 2011 at 7:29 PM, Ron Johnson <ron.l.johnson@cox.net> wrote:
> Anything in syslog?

Well, if I bring up eth0 ("# ifconfig eth0 up"), wait for it to lag,
and then turn eth0 off again ("# ifconfig eth0 down"), this is what
shows up in syslog:

Mar 7 23:08:03 mjollnir kernel: [26127.412486] e1000e 0000:00:19.0:
irq 29 for MSI/MSI-X
Mar 7 23:08:03 mjollnir kernel: [26127.468269] e1000e 0000:00:19.0:
irq 29 for MSI/MSI-X
Mar 7 23:08:03 mjollnir kernel: [26127.468919] ADDRCONF(NETDEV_UP):
eth0: link is not ready

The only entry in /var/log/messages is fairly innocuous (I think):

Mar 7 23:08:03 mjollnir kernel: [26127.468919] ADDRCONF(NETDEV_UP):
eth0: link is not ready

And it also shows up in dmesg (via "dmesg | tail"):

[26127.412486] e1000e 0000:00:19.0: irq 29 for MSI/MSI-X
[26127.468269] e1000e 0000:00:19.0: irq 29 for MSI/MSI-X
[26127.468919] ADDRCONF(NETDEV_UP): eth0: link is not ready

I just Googled around for some of that stuff, but I'm way out of my
league and don't really know what that means or if it's normal or not.
Does that mean anything to you?

Now that things are working again, those entries do not show up after
using ifconfig to raise and lower the eth0 interface.

Kyle


--
Kyle Barbour
kyle@kylebarbour.org
(415) 238-9953
San Francisco, CA, USA


--
To UNSUBSCRIBE, email to debian-user-REQUEST@lists.debian.org
with a subject of "unsubscribe". Trouble? Contact listmaster@lists.debian.org
Archive: AANLkTinX6Nh60pgOSR_7cQqey2mPVi_zZsm2h9RurqpT@mail .gmail.com">http://lists.debian.org/AANLkTinX6Nh60pgOSR_7cQqey2mPVi_zZsm2h9RurqpT@mail .gmail.com
 
Old 03-08-2011, 06:40 AM
Kyle Barbour
 
Default events/0 and events/1 maxing CPU to 100%, related to wireless or eth0 on a T61 Thinkpad

> O.K., so amazingly this is solved, at least on the practical end. I
> have no idea why this works, and I've provided some more information
> below if people want to pursue this or for others trying to solve this
> problem, but the solution that worked for me was powering off the
> computer, unplugging it, removing the battery, waiting 30 seconds,
> plugging it back in, reattaching the battery, and powering back up.

<snip>

> This solution was found from here:
> https://bugs.launchpad.net/ubuntu/+source/linux/+bug/53739

Sorry, that was supposed to be
https://bugs.launchpad.net/ubuntu/+source/linux/+bug/537396 - note the
trailing 6.

Kyle


--
Kyle Barbour
kyle@kylebarbour.org
(415) 238-9953
San Francisco, CA, USA


--
To UNSUBSCRIBE, email to debian-user-REQUEST@lists.debian.org
with a subject of "unsubscribe". Trouble? Contact listmaster@lists.debian.org
Archive: AANLkTikuj_u8=DD3EbXWu-JGX_GyiNGhcj-LJoD36ng3@mail.gmail.com">http://lists.debian.org/AANLkTikuj_u8=DD3EbXWu-JGX_GyiNGhcj-LJoD36ng3@mail.gmail.com
 

Thread Tools




All times are GMT. The time now is 06:23 AM.

VBulletin, Copyright ©2000 - 2014, Jelsoft Enterprises Ltd.
Content Relevant URLs by vBSEO ©2007, Crawlability, Inc.
Copyright 2007 - 2008, www.linux-archive.org