Google Groups no longer supports new Usenet posts or subscriptions. Historical content remains viewable.
Dismiss

UML error - apps killed 50% of the time

4 views
Skip to first unread message

Adam Nielsen

unread,
Feb 15, 2010, 2:00:02 AM2/15/10
to
Hi all,

From my other message I am trying to get a UML kernel running again after an
OS upgrade, but I am unable to get it working again (on the original machine
this time, everything is 32-bit.)

When I boot the kernel normally I get tons of error messages, but if I boot it
with "init=/bin/bash" I can get a command prompt up. Unfortunately most of
the time my apps break:

$ ls
bin boot dev etc home lib lost+found mnt opt proc root sbin sys
tmp usr var
$ ls
Killed
$ ls
bin boot dev etc home lib lost+found mnt opt proc root sbin sys
tmp usr var
$ ls
Killed
$ ls
Killed
$ ls
bin boot dev etc home lib lost+found mnt opt proc root sbin sys
tmp usr var

What's going on? My host dmesg says things like this:

vmlinux[6300] general protection eip:b7ea5124 esp:1785bdfc error:0

Why doesn't UML work any more? It's running the same filesystem and apps as
before. Any help would be much appreciated, I've wasted hours on this now :-(

Thanks,
Adam.

(please CC)
--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majo...@vger.kernel.org
More majordomo info at http://vger.kernel.org/majordomo-info.html
Please read the FAQ at http://www.tux.org/lkml/

Américo Wang

unread,
Feb 15, 2010, 4:00:01 AM2/15/10
to

http://patchwork.kernel.org/patch/79358/

Does this patch help?

Thanks.

Adam Nielsen

unread,
Feb 15, 2010, 4:30:03 AM2/15/10
to
>> $ ls
>> Killed
>> $ ls
>> Killed
>>
>> vmlinux[6300] general protection eip:b7ea5124 esp:1785bdfc error:0
>>
>> Why doesn't UML work any more?
>

Sort of - ls doesn't get killed any more and I can get about half way through
the boot process before things start getting killed (as opposed to them
getting killed right away.)

Whatever this patch does, if it could do it twice as much then my problem
would be solved ;-)

Thanks,
Adam.

Américo Wang

unread,
Feb 15, 2010, 7:30:02 AM2/15/10
to
On Mon, Feb 15, 2010 at 07:21:46PM +1000, Adam Nielsen wrote:
>>> $ ls
>>> Killed
>>> $ ls
>>> Killed
>>>
>>> vmlinux[6300] general protection eip:b7ea5124 esp:1785bdfc error:0
>>>
>>> Why doesn't UML work any more?
>>
>> http://patchwork.kernel.org/patch/79358/
>>
>> Does this patch help?
>
> Sort of - ls doesn't get killed any more and I can get about half way
> through the boot process before things start getting killed (as opposed
> to them getting killed right away.)
>
> Whatever this patch does, if it could do it twice as much then my problem
> would be solved ;-)
>

So what are the errors now? Could you please provide your boot log?

Thanks.

Adam Nielsen

unread,
Feb 15, 2010, 4:30:03 PM2/15/10
to
> So what are the errors now? Could you please provide your boot log?

Just more random tasks being killed:

* Calculating module dependencies .../sbin/modules-update: line 73: 967
Broken pipe sed -ne 1p "${x}"
968 Killed | egrep -q "^${HEADER}"
* Mounting local filesystems ... [ ok ]
/etc/init.d/localmount: line 9: 982 Killed modprobe usbcore
>&/dev/null
* Cleaning /tmp directory ... [ ok ]
/sbin/depscan.sh: line 21: 1055 Killed mkdir -p -m 0755
"${svcdir}/${x}" 2>/dev/null
* Could not create needed directory '/var/lib/init.d/snapshot'!
INIT: Entering runlevel: 3
/lib/rcscripts/sh/rc-services.sh: line 35: 1096 Killed
/sbin/depscan.sh -u

* Error running '/sbin/depscan.sh'!
* Please correct any problems above.
/sbin/functions.sh: line 184: 1098 Killed /usr/bin/logger -p
"${pri}" -t "${tag}" -- "$*"

Cheers,
Adam.

Américo Wang

unread,
Feb 20, 2010, 4:40:01 AM2/20/10
to
On Tue, Feb 16, 2010 at 5:28 AM, Adam Nielsen <a.ni...@shikadi.net> wrote:
>> So what are the errors now? Could you please provide your boot log?
>
> Just more random tasks being killed:
>
>  * Calculating module dependencies .../sbin/modules-update: line 73:   967
> Broken pipe             sed -ne 1p "${x}"
>       968 Killed                  | egrep -q "^${HEADER}"
>  * Mounting local filesystems ... [ ok ]
> /etc/init.d/localmount: line 9:   982 Killed                  modprobe
> usbcore >&/dev/null
>  * Cleaning /tmp directory ... [ ok ]
> /sbin/depscan.sh: line 21:  1055 Killed                  mkdir -p -m 0755
> "${svcdir}/${x}" 2>/dev/null
>  *  Could not create needed directory '/var/lib/init.d/snapshot'!
> INIT: Entering runlevel: 3
> /lib/rcscripts/sh/rc-services.sh: line 35:  1096 Killed /sbin/depscan.sh -u
>
>  * Error running '/sbin/depscan.sh'!
>  * Please correct any problems above.
> /sbin/functions.sh: line 184:  1098 Killed                  /usr/bin/logger
> -p "${pri}" -t "${tag}" -- "$*"
>

Oh, probably caused by the same commit....

I will check it tonight.

Thank you.

Américo Wang

unread,
Feb 20, 2010, 9:50:02 AM2/20/10
to
On Tue, Feb 16, 2010 at 07:28:06AM +1000, Adam Nielsen wrote:
>> So what are the errors now? Could you please provide your boot log?
>
> Just more random tasks being killed:
>
> * Calculating module dependencies .../sbin/modules-update: line 73:
> 967 Broken pipe sed -ne 1p "${x}"
> 968 Killed | egrep -q "^${HEADER}"
> * Mounting local filesystems ... [ ok ]
> /etc/init.d/localmount: line 9: 982 Killed modprobe
> usbcore >&/dev/null
> * Cleaning /tmp directory ... [ ok ]
> /sbin/depscan.sh: line 21: 1055 Killed mkdir -p -m 0755
> "${svcdir}/${x}" 2>/dev/null
> * Could not create needed directory '/var/lib/init.d/snapshot'!
> INIT: Entering runlevel: 3
> /lib/rcscripts/sh/rc-services.sh: line 35: 1096 Killed /sbin/depscan.sh
> -u
>
> * Error running '/sbin/depscan.sh'!
> * Please correct any problems above.
> /sbin/functions.sh: line 184: 1098 Killed
> /usr/bin/logger -p "${pri}" -t "${tag}" -- "$*"
>

I tested it, it is fixed by the updated patch:

http://linux.derkeiler.com/Mailing-Lists/Kernel/2010-02/msg05620.html

Américo Wang

unread,
Feb 20, 2010, 9:20:01 PM2/20/10
to
On Sun, Feb 21, 2010 at 10:11 AM, Adam Nielsen <a.ni...@shikadi.net> wrote:
>>> Just more random tasks being killed:
>>> /sbin/functions.sh: line 184:  1098 Killed
>>> /usr/bin/logger -p "${pri}" -t "${tag}" -- "$*"
>>
>> I tested it, it is fixed by the updated patch:
>>
>> http://linux.derkeiler.com/Mailing-Lists/Kernel/2010-02/msg05620.html
>
> That patch won't apply - if I am reading it correctly it looks like it was
> already part of the original patch
> (http://patchwork.kernel.org/patch/79358/)
>

No, they are different. Sorry that the above page mangles the patch,
please try this one:

http://lkml.org/lkml/diff/2010/2/15/61/1

Thanks.

Adam Nielsen

unread,
Feb 20, 2010, 9:20:01 PM2/20/10
to
>> Just more random tasks being killed:
>> /sbin/functions.sh: line 184: 1098 Killed
>> /usr/bin/logger -p "${pri}" -t "${tag}" -- "$*"
>
> I tested it, it is fixed by the updated patch:
>
> http://linux.derkeiler.com/Mailing-Lists/Kernel/2010-02/msg05620.html

That patch won't apply - if I am reading it correctly it looks like it was

already part of the original patch (http://patchwork.kernel.org/patch/79358/)

Thanks,
Adam.

Adam Nielsen

unread,
Feb 20, 2010, 10:20:01 PM2/20/10
to
>> That patch won't apply - if I am reading it correctly it looks like it was
>> already part of the original patch
>> (http://patchwork.kernel.org/patch/79358/)
>
> No, they are different. Sorry that the above page mangles the patch,
> please try this one:
>
> http://lkml.org/lkml/diff/2010/2/15/61/1

Oh my mistake, I thought it was to be applied on top of the original patch. I
reverted the original patch and applied this one, and it seems to have fixed
the problem!

I can now fully boot into the UML and run programs again - thanks for your help!

Cheers,
Adam.

0 new messages