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 > Redhat > Crash Utility

 
 
LinkBack Thread Tools
 
Old 02-09-2009, 03:06 AM
Jun Koi
 
Default crash bug on 2.6.27.8

Hi,

When using the latest crash on 2.6.27.14 to analyze live memory
(running crash without any option), I got a bug like below. How can I
fix that?

(A similar problem is also seen on 2.6.27.8)

Thanks,
Jun

.....
GNU gdb 6.1
Copyright 2004 Free Software Foundation, Inc.
GDB is free software, covered by the GNU General Public License, and you are
welcome to change it and/or distribute copies of it under certain conditions.
Type "show copying" to see the conditions.
There is absolutely no warranty for GDB. Type "show warranty" for details.
This GDB was configured as "i686-pc-linux-gnu"...

crash: read error: kernel virtual address: c0488200 type: "cpu_possible_map"
WARNING: cannot read cpu_possible_map
crash: read error: kernel virtual address: c041b428 type: "cpu_present_map"
WARNING: cannot read cpu_present_map
crash: read error: kernel virtual address: c041ab00 type: "cpu_online_map"
WARNING: cannot read cpu_online_map
crash: read error: kernel virtual address: c04b1900 type: "xtime"

--
Crash-utility mailing list
Crash-utility@redhat.com
https://www.redhat.com/mailman/listinfo/crash-utility
 
Old 02-09-2009, 01:51 PM
Dave Anderson
 
Default crash bug on 2.6.27.8

----- "Jun Koi" <junkoi2004@gmail.com> wrote:

> Hi,
>
> When using the latest crash on 2.6.27.14 to analyze live memory
> (running crash without any option), I got a bug like below. How can I
> fix that?
>
> (A similar problem is also seen on 2.6.27.8)
>
> Thanks,
> Jun

The "read error" indications are most likely due to your kernel
being configured with CONFIG_STRICT_DEVMEM, which restricts
/dev/mem access to the first 256 pages of physical memory,
which makes it useless for the crash utility.

If you cannot reconfigure your kernel without it, then you
have two other options.

1. Port the RHEL/Fedora /dev/crash (drivers/char/crash.c) memory
driver, and insmod it prior to invoking crash. Red Hat kernels
have had the /dev/mem restriction in place by default since
RHEL4. It's only fairly recently that it's been imposed by
default in upstream kernels.
2. You can also write a simple kprobe module module that
modifies the kernel's devmem_is_allowed() function to
always return 1. That will fake the /dev/mem driver into
accepting all physical address requests.

I don't have the links off-hand, but if you search through
this mailing list's archives, say since the September 2008
timeframe, you can find threads that that describe how (1)
and (2) can be accomplished.

But if it all possible, the easiest way to go is by reconfiguring
your kernel with CONFIG_STRICT_DEVMEM turned off.

I'm going to be out of touch for a couple of days, so I won't be
of much further help until Wednesday. If anybody else on the list
has done this, please give this guy any pointers you might have.

Thanks,
Dave



>
> .....
> GNU gdb 6.1
> Copyright 2004 Free Software Foundation, Inc.
> GDB is free software, covered by the GNU General Public License, and
> you are
> welcome to change it and/or distribute copies of it under certain
> conditions.
> Type "show copying" to see the conditions.
> There is absolutely no warranty for GDB. Type "show warranty" for
> details.
> This GDB was configured as "i686-pc-linux-gnu"...
>
> crash: read error: kernel virtual address: c0488200 type:
> "cpu_possible_map"
> WARNING: cannot read cpu_possible_map
> crash: read error: kernel virtual address: c041b428 type:
> "cpu_present_map"
> WARNING: cannot read cpu_present_map
> crash: read error: kernel virtual address: c041ab00 type:
> "cpu_online_map"
> WARNING: cannot read cpu_online_map
> crash: read error: kernel virtual address: c04b1900 type: "xtime"
>
> --
> Crash-utility mailing list
> Crash-utility@redhat.com
> https://www.redhat.com/mailman/listinfo/crash-utility

--
Crash-utility mailing list
Crash-utility@redhat.com
https://www.redhat.com/mailman/listinfo/crash-utility
 
Old 02-10-2009, 07:45 AM
Jun Koi
 
Default crash bug on 2.6.27.8

On Mon, Feb 9, 2009 at 11:51 PM, Dave Anderson <anderson@redhat.com> wrote:
>
> ----- "Jun Koi" <junkoi2004@gmail.com> wrote:
>
>> Hi,
>>
>> When using the latest crash on 2.6.27.14 to analyze live memory
>> (running crash without any option), I got a bug like below. How can I
>> fix that?
>>
>> (A similar problem is also seen on 2.6.27.8)
>>
>> Thanks,
>> Jun
>
> The "read error" indications are most likely due to your kernel
> being configured with CONFIG_STRICT_DEVMEM, which restricts
> /dev/mem access to the first 256 pages of physical memory,
> which makes it useless for the crash utility.

Thanks a lot for the information - I can fix the kernel here.

But perhaps it is better to give some hints when we have such a error
report, so users know that they should turn off the above
configuration? I am sure that many people will have this problem, as
the option may be ON by default.

Thanks,
Jun

--
Crash-utility mailing list
Crash-utility@redhat.com
https://www.redhat.com/mailman/listinfo/crash-utility
 
Old 02-10-2009, 08:02 AM
Durga Prasad
 
Default crash bug on 2.6.27.8

You can find the kprobe module that Dave mentioned on the thread Date: Sat, 8 Mar 2008 00:34:27 +0530

Regards
Durga


On Tue, Feb 10, 2009 at 2:15 PM, Jun Koi <junkoi2004@gmail.com> wrote:

On Mon, Feb 9, 2009 at 11:51 PM, Dave Anderson <anderson@redhat.com> wrote:


>

> ----- "Jun Koi" <junkoi2004@gmail.com> wrote:

>

>> Hi,

>>

>> When using the latest crash on 2.6.27.14 to analyze live memory

>> (running crash without any option), I got a bug like below. How can I

>> fix that?

>>

>> (A similar problem is also seen on 2.6.27.8)

>>

>> Thanks,

>> Jun

>

> The "read error" indications are most likely due to your kernel

> being configured with CONFIG_STRICT_DEVMEM, which restricts

> /dev/mem access to the first 256 pages of physical memory,

> which makes it useless for the crash utility.



Thanks a lot for the information - I can fix the kernel here.



But perhaps it is better to give some hints when we have such a error

report, so users know that they should turn off the above

configuration? I am sure that many people will have this problem, as

the option may be ON by default.



Thanks,

Jun



--

Crash-utility mailing list

Crash-utility@redhat.com

https://www.redhat.com/mailman/listinfo/crash-utility



--

---------
Keep dreaming, for dreams are the precursors of things to come real.. @!#$

--
Crash-utility mailing list
Crash-utility@redhat.com
https://www.redhat.com/mailman/listinfo/crash-utility
 
Old 02-11-2009, 01:45 PM
Dave Anderson
 
Default crash bug on 2.6.27.8

----- "Jun Koi" <junkoi2004@gmail.com> wrote:

> On Mon, Feb 9, 2009 at 11:51 PM, Dave Anderson <anderson@redhat.com>
> wrote:
> >
> > ----- "Jun Koi" <junkoi2004@gmail.com> wrote:
> >
> >> Hi,
> >>
> >> When using the latest crash on 2.6.27.14 to analyze live memory
> >> (running crash without any option), I got a bug like below. How can
> I
> >> fix that?
> >>
> >> (A similar problem is also seen on 2.6.27.8)
> >>
> >> Thanks,
> >> Jun
> >
> > The "read error" indications are most likely due to your kernel
> > being configured with CONFIG_STRICT_DEVMEM, which restricts
> > /dev/mem access to the first 256 pages of physical memory,
> > which makes it useless for the crash utility.
>
> Thanks a lot for the information - I can fix the kernel here.
>
> But perhaps it is better to give some hints when we have such a error
> report, so users know that they should turn off the above
> configuration? I am sure that many people will have this problem, as
> the option may be ON by default.
>
> Thanks,
> Jun

Good idea -- I'll put something in place for the next release.

I'll also look into creating a web page with a "canned answer"
for this issue, including directions for going the crash-module-build
or kprobe routes for those who cannot rebuild their kernels.

Thanks,
Dave


--
Crash-utility mailing list
Crash-utility@redhat.com
https://www.redhat.com/mailman/listinfo/crash-utility
 
Old 02-12-2009, 03:20 AM
Jun Koi
 
Default crash bug on 2.6.27.8

On Wed, Feb 11, 2009 at 11:45 PM, Dave Anderson <anderson@redhat.com> wrote:
>
> ----- "Jun Koi" <junkoi2004@gmail.com> wrote:
>
>> On Mon, Feb 9, 2009 at 11:51 PM, Dave Anderson <anderson@redhat.com>
>> wrote:
>> >
>> > ----- "Jun Koi" <junkoi2004@gmail.com> wrote:
>> >
>> >> Hi,
>> >>
>> >> When using the latest crash on 2.6.27.14 to analyze live memory
>> >> (running crash without any option), I got a bug like below. How can
>> I
>> >> fix that?
>> >>
>> >> (A similar problem is also seen on 2.6.27.8)
>> >>
>> >> Thanks,
>> >> Jun
>> >
>> > The "read error" indications are most likely due to your kernel
>> > being configured with CONFIG_STRICT_DEVMEM, which restricts
>> > /dev/mem access to the first 256 pages of physical memory,
>> > which makes it useless for the crash utility.
>>
>> Thanks a lot for the information - I can fix the kernel here.
>>
>> But perhaps it is better to give some hints when we have such a error
>> report, so users know that they should turn off the above
>> configuration? I am sure that many people will have this problem, as
>> the option may be ON by default.
>>
>> Thanks,
>> Jun
>
> Good idea -- I'll put something in place for the next release.
>
> I'll also look into creating a web page with a "canned answer"
> for this issue, including directions for going the crash-module-build
> or kprobe routes for those who cannot rebuild their kernels.
>

Yes, it is nice to have a FAQ page.

Thanks,
Jun

--
Crash-utility mailing list
Crash-utility@redhat.com
https://www.redhat.com/mailman/listinfo/crash-utility
 

Thread Tools




All times are GMT. The time now is 01:09 PM.

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