[clug] Ethernet problem: NIC Link goes down regularly on ThinkPad X60s

[Boyd Wilding]

I'm not trying to take the piss here, so please don't hurt me.

In all seriousness, try this:

Install windows and do some testing, eg stitch the port at 100mb full duplex
and see what happens. Also under windows, try to set to Autonegotiate and
see what happens (AFAIK I don't believe you can stitch it/lock it hard to
gigabit).
This will tell you in short order where the problem lies - OS/driver, or
hardware.

Then again it might just be a link negotiation problem, but try it and see.

Heck, if you are dualbooting at the moment, even better, less hassle for
you...windows is already there for you to experiment with and more quickly
enable you to determine the root cause by quickly seeing where the
consistency of the problem lies.

Another possibility is to quickly borrow a HDD from the same model X60 at
the next CLUG meeting, that has windows on it already...if they are willing
to play around and if there is a network available to test with in the
meeting room where CLUG is. Laptop HDDs in thinkpads are fairly easy to swap
as long as you know how to operate a phillips screwdriver.

Failing that I have a T60 with a 60gb SATA disk with windows and suse on it
that might be free during the next CLUG meeting.

PS BIOS upgrades are done easy enough, punch in your machine/model type into
the IBM/Lenovo website and you should be able to find either windows based
packages or extractable DOS diskette versions of BIOS updates...which would
require a USB floppy drive to update, obviously, but those are the only 2
options they provide, I believe.

Hope this helps.

On 7/10/07, Lachlan Rogers <lachlan at rogers.name> wrote:
>
> Hi everybody,
>
> I am running Gentoo linux on my new IBM ThinkPad X60s and am having
> problems with the ethernet connection dropping out.  I've been along to two
> CLUG meetings now, and have seen at least 2 people with ThinkPad X60
> laptops.  I am wanting to determine if my problem is simply a Linux driver
> issue or whether it is hardware that should be fixed while under warranty.
>
> I have had this problem with numerous kernels (I keep trying new releases
> to see if it is fixed), and am currently using the e1000-7.5.5 driver from
> sourceforge with a 2.6.21.5 kernel.
>
> The relevant lspci output is:
>
>         02:00.0 Ethernet controller: Intel Corporation 82573L Gigabit
> Ethernet Controller
>
> When the connection drops out, I get the following dmesg output:
>
>         e1000: eth0: e1000_watchdog_task: NIC Link is Down
>
> and the green connection indicator on the ethernet plug goes out.  When
> the connection comes back up (just as randomly as dropping out) I get the
> following dmesg output:
>
>         e1000: eth0: e1000_watchdog_task: NIC Link is Up 1000 Mbps Full
> Duplex, Flow Control: RX/TX
>
> and the green indicator light comes back on.
>
>
> While trying to diagnose the problem, I found ethtool and installed
> it.  While the ethernet connection is operating correctly, I get the
> following ethtool output:
>
> Settings for eth0:
>         Supported ports: [ TP ]
>         Supported link modes:   10baseT/Half 10baseT/Full
>                                 100baseT/Half 100baseT/Full
>                                 1000baseT/Full
>         Supports auto-negotiation: Yes
>         Advertised link modes:  10baseT/Half 10baseT/Full
>                                 100baseT/Half 100baseT/Full
>                                 1000baseT/Full
>         Advertised auto-negotiation: Yes
>         Speed: 1000Mb/s
>         Duplex: Full
>         Port: Twisted Pair
>         PHYAD: 1
>         Transceiver: internal
>         Auto-negotiation: on
>         Supports Wake-on: umbg
>         Wake-on: g
>         Current message level: 0x00000007 (7)
>         Link detected: yes
>
>
> While the connection is down, I get the following ethtool output:
>
> Settings for eth0:
>         Supported ports: [ TP ]
>         Supported link modes:   10baseT/Half 10baseT/Full
>                                 100baseT/Half 100baseT/Full
>                                 1000baseT/Full
>         Supports auto-negotiation: Yes
>         Advertised link modes:  10baseT/Half 10baseT/Full
>                                 100baseT/Half 100baseT/Full
>                                 1000baseT/Full
>         Advertised auto-negotiation: Yes
>         Speed: Unknown! (65535)
>         Duplex: Unknown! (255)
>         Port: Twisted Pair
>         PHYAD: 1
>         Transceiver: internal
>         Auto-negotiation: on
>         Supports Wake-on: umbg
>         Wake-on: g
>         Current message level: 0x00000007 (7)
>         Link detected: no
>
>
> In my searches I found reference to a similar looking problem at
> http://lkml.org/lkml/2006/1/5/379, and the follow up emails in this thread
> suggest that heat is a problem.  Obviously, it is not possible to install
> additional cooling utilities in my laptop.
> Since finding this thread at LKML, I have tried to diagnose whether my
> problem is linked to the heat of my laptop.  I have not noticed any
> definitive trend.
>
>
> This problem first appeared as short periods of down time in the
> connection.  For unknown reasons it often now manifests itself as prolonged
> periods of down time interspersed with only short bursts of the link being
> up.  It is essentially unusable, and I would greatly appreciate any
> assistance in finding a solution.
>
> I sent this to the e1000 driver mailing list, and it was suggested that I
> upgrade BIOS.  However, I have not found a way to do this yet (any help here
> would be appreciated also).
>
> I'm keen to hear any ideas.
>
> Thanks,
> Lachlan
> --
> linux mailing list
> linux at lists.samba.org
> https://lists.samba.org/mailman/listinfo/linux
>