Tracking down a Linux crash

Tracking down a Linux crash

[John Ackermann N8UR]

Hi all --

I realize this may be a bit off-topic, but my rationale is that (a) I'm
among knowledgeable friends, and (b) this machine is destined to be my
hamshack system, displacing a Windows box.

I've been building Linux systems since back in the 0.99pl4 days, and for
the first time I'm confronted with a system that regularly crashes.  There
are unfortunately quite a few variables in the equation, so I doubt anyone
here can give me an outright answer.  But perhaps with your collective
experience, I can zero in on the problem.  Please bear with the length of
this message as I want to provide as much detail as I can.

The box is an Athlon XP2000+ on an Elite Group L-7VMM motherboard with VIA
chipset.  On-board S3 SavagePro video, Realtek 8139 LAN, and VIA8233A
sound.  512MB DDR, 40GB drive.  Expansion cards are an M-Audio Delta44
sound card, and a Boca 8 port serial card.

I've been loading Debian 3.0 (woody) with KDE[2,3].  I'm using a
home-compiled kernel, both 2.4.18 and 2.4.19.  I'm not turning on any
exotic features in the kernel (other than AX.25 :-) ) and I'm setting the
CPU type to Athlon/K7.

I've installed ALSA sound drivers and utils.  I've had some problems with
the onboard VIA8233A sound, so I'm not currently enabling it.  I do have
the ICE1712 ALSA module installed for the Delta44 card, and the sound
system seems to be working OK with that card.  To get the S3 SavagePro 
video working in X
I had to download a later version of the XFree-4 driver for that card; I
was able to find a .deb file for it and it seems to install and work OK.
I'm using the ext3 filesystem.

The crash occurs after the system is fully booted and the KDE environment
is up and running -- in fact, it may occur as much as a couple of days
after bootup.  Several of the crashes have occurred when I was using
Mozilla (version 1.0, installed from Debian), immediately after typing in a
URL and hitting the return key -- but it's also crashed when Mozilla wasn't
running.  I've had both KDE2 (standard for Debian 3.0) and KDE3 (by
upgrading the system from woody to sarge and installing .debs from kde.org)
installed and had crashes under both environments.  The crash does not seem
to be related to sound card activity.

The crash itself is unusual.  The symptom is that the X screen freezes,
though usually the mouse pointer continues to move with the mouse. However,
there's no response to keyboard or mouse-clicks, with one exception.  That
is that if I use ctr-alt-F1 to go to a virtual console, the screen will
switch to VGA mode, but remains completely black, with no login banner or
anything.  After that point, nothing I do with the keyboard or the mouse
yields a response. The keyboard LEDs stop working (at least at this point;
I'm not sure if they stop before switching consoles).

The network subsystem remains at least partially active.  I can ping the
box from a remote host, but attempting to ssh to it fails with no reponse.
I don't have a telnet server installed, but attempting to telnet from a
remote machine to the normal port yields a "connection refused" message. If
I telnet to port 25 (smtp), I get a connect, but exim does not respond to
any commands sent to it.

There's no trace left in the standard logs that I can tie to the crash.

After rebooting by a power cycle or hardware reset, there are often some
corrupted files left behind, and they're hosed in a pretty dramatic way --
the ownership and dates are changed to impossibly large numbers, and it's
often impossible to delete them without extraordinary effort -- you get an
"unlink: operation not permitted" error and only using the filesystem
editor to edit the inode entry lets you get to the point where you can
delete the file.

A web search hasn't shown any mention of problems using Linux with this
motherboard or chipset.

Well, that's a lot of info, but I'm hoping that someone out there will
recognize something here that will point me in the right direction.  Thanks
for any help!

73,

John N8UR
jra@febo.com

----
John Ackermann    N8UR         jra@febo.com     http://www.febo.com
President, TAPR                n8ur@tapr.org    http://www.tapr.org

Re: Tracking down a Linux crash

[Kelly Black]

On Wed, 2002-10-16 at 08:08, John Ackermann N8UR wrote:
> Hi all --
> 
> I realize this may be a bit off-topic, but my rationale is that (a) I'm
> among knowledgeable friends, and (b) this machine is destined to be my
> hamshack system, displacing a Windows box.
> 
---Snip long description---

Could the problem be heat related?  What is the CPU temp when the
lockups occur (or just after a reboot).  Does anything show up in the
logs (kernel panics or anything suspicious?).

Kelly Black
KB0GBJ

Re: Tracking down a Linux crash

[Kelly Black]

> On Wed, 2002-10-16 at 08:08, John Ackermann N8UR wrote:
> > Hi all --
> > 
> > I realize this may be a bit off-topic, but my rationale is that (a) I'm
> > among knowledgeable friends, and (b) this machine is destined to be my
> > hamshack system, displacing a Windows box.
> > 
> ---Snip long description---
> 

Could the problem be heat related?  What is the CPU temp when the
lockups occur (or just after a reboot).  Does anything show up in the
logs (kernel panics or anything suspicious?).
 
 Kelly Black
 KB0GBJ
 

Re: Tracking down a Linux crash

[John Ackermann N8UR]

Hi Kelly --

I don't think it's heat as the crash sometimes occurs relatively quickly 
after powerup, and other times after days of running.  The CPU is not being 
overclocked, and has an original AMD heatsink/fan, plus there's a fan on 
the case.  However, I'm not monitoring CPU temperature in the system (don't 
have that software installed yet) so I can't say for sure what the temp is.

Re the logs, there's nothing -- in particular, no sign of a kernel panic.

Thanks for the quick reply!

73,
John

--On Wednesday, October 16, 2002 8:20 AM -0500 Kelly Black 
<kelly.black@penguinpackets.com> wrote:

> On Wed, 2002-10-16 at 08:08, John Ackermann N8UR wrote:
>> Hi all --
>>
>> I realize this may be a bit off-topic, but my rationale is that (a) I'm
>> among knowledgeable friends, and (b) this machine is destined to be my
>> hamshack system, displacing a Windows box.
>>
> ---Snip long description---
>
> Could the problem be heat related?  What is the CPU temp when the
> lockups occur (or just after a reboot).  Does anything show up in the
> logs (kernel panics or anything suspicious?).
>
> Kelly Black
> KB0GBJ
>
>
>
>



----
John Ackermann    N8UR         jra@febo.com     http://www.febo.com
President, TAPR                n8ur@tapr.org    http://www.tapr.org

Re: Tracking down a Linux crash

[Pär Flitt]

    Hi all --
    
    I realize this may be a bit off-topic, but my rationale is that (a) I'm
    among knowledgeable friends, and (b) this machine is destined to be my
    hamshack system, displacing a Windows box.
    ---Snip long description---


I've had a similar problem and it completely vanished when I replaced
the graphics adapter. I only saw the problem when in X, mostly when
surfing the web.

Pär Flitt
SM0RWO




-
To unsubscribe from this list: send the line "unsubscribe linux-hams" in
the body of a message to majordomo@vger.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html

Re: Tracking down a Linux crash

[jbennett]

Hi John.

I have experienced the same thing, but no where near the same hardware. Same
symptoms, same deal with the logs. I have not found a fix.  However, I
firmly feel the problem is X or KDE related since it has never crashed in
the character mode (run level 3). Application does not seem to matter. I've
had the GUI lock up in Kate, K-IDE, Konquerer, and a few others I don't
recall (because I rarely use them). I've also had a friend experience the
same thing. The machine that has done it the most here has an Intel SE440BX
motherboard with a P3-500 installed w/256 MB RAM. Video is an ATI 128 Rage.
My laptop, a Dell Inspiron 8100 w/a P3-850 has done it on occasion. It uses
the Nvidia chipset. I believe my friend's machine is a Dell 5100 w/a P3-550.
I believe his video chipset is ATI. My gut instinct says it's in the X
driver, but that's just a guess. Like you, have been unable to find anything
on the Internet.  Will be interesting to see what turns up.

73,
John Bennett
n4xi


----- Original Message -----
From: "John Ackermann N8UR" <jra@febo.com>
To: <linux@lists.tapr.org>; <linux-hams@vger.kernel.org>
Sent: Wednesday, October 16, 2002 8:08 AM
Subject: Tracking down a Linux crash


> Hi all --
>
> I realize this may be a bit off-topic, but my rationale is that (a) I'm
> among knowledgeable friends, and (b) this machine is destined to be my
> hamshack system, displacing a Windows box.
>
> I've been building Linux systems since back in the 0.99pl4 days, and for
> the first time I'm confronted with a system that regularly crashes.  There
> are unfortunately quite a few variables in the equation, so I doubt anyone
> here can give me an outright answer.  But perhaps with your collective
> experience, I can zero in on the problem.  Please bear with the length of
> this message as I want to provide as much detail as I can.
>
> The box is an Athlon XP2000+ on an Elite Group L-7VMM motherboard with VIA
> chipset.  On-board S3 SavagePro video, Realtek 8139 LAN, and VIA8233A
> sound.  512MB DDR, 40GB drive.  Expansion cards are an M-Audio Delta44
> sound card, and a Boca 8 port serial card.
>
> I've been loading Debian 3.0 (woody) with KDE[2,3].  I'm using a
> home-compiled kernel, both 2.4.18 and 2.4.19.  I'm not turning on any
> exotic features in the kernel (other than AX.25 :-) ) and I'm setting the
> CPU type to Athlon/K7.
>
> I've installed ALSA sound drivers and utils.  I've had some problems with
> the onboard VIA8233A sound, so I'm not currently enabling it.  I do have
> the ICE1712 ALSA module installed for the Delta44 card, and the sound
> system seems to be working OK with that card.  To get the S3 SavagePro
> video working in X
> I had to download a later version of the XFree-4 driver for that card; I
> was able to find a .deb file for it and it seems to install and work OK.
> I'm using the ext3 filesystem.
>
> The crash occurs after the system is fully booted and the KDE environment
> is up and running -- in fact, it may occur as much as a couple of days
> after bootup.  Several of the crashes have occurred when I was using
> Mozilla (version 1.0, installed from Debian), immediately after typing in
a
> URL and hitting the return key -- but it's also crashed when Mozilla
wasn't
> running.  I've had both KDE2 (standard for Debian 3.0) and KDE3 (by
> upgrading the system from woody to sarge and installing .debs from
kde.org)
> installed and had crashes under both environments.  The crash does not
seem
> to be related to sound card activity.
>
> The crash itself is unusual.  The symptom is that the X screen freezes,
> though usually the mouse pointer continues to move with the mouse.
However,
> there's no response to keyboard or mouse-clicks, with one exception.  That
> is that if I use ctr-alt-F1 to go to a virtual console, the screen will
> switch to VGA mode, but remains completely black, with no login banner or
> anything.  After that point, nothing I do with the keyboard or the mouse
> yields a response. The keyboard LEDs stop working (at least at this point;
> I'm not sure if they stop before switching consoles).
>
> The network subsystem remains at least partially active.  I can ping the
> box from a remote host, but attempting to ssh to it fails with no reponse.
> I don't have a telnet server installed, but attempting to telnet from a
> remote machine to the normal port yields a "connection refused" message.
If
> I telnet to port 25 (smtp), I get a connect, but exim does not respond to
> any commands sent to it.
>
> There's no trace left in the standard logs that I can tie to the crash.
>
> After rebooting by a power cycle or hardware reset, there are often some
> corrupted files left behind, and they're hosed in a pretty dramatic way --
> the ownership and dates are changed to impossibly large numbers, and it's
> often impossible to delete them without extraordinary effort -- you get an
> "unlink: operation not permitted" error and only using the filesystem
> editor to edit the inode entry lets you get to the point where you can
> delete the file.
>
> A web search hasn't shown any mention of problems using Linux with this
> motherboard or chipset.
>
> Well, that's a lot of info, but I'm hoping that someone out there will
> recognize something here that will point me in the right direction.
Thanks
> for any help!
>
> 73,
>
> John N8UR
> jra@febo.com
>
> ----
> John Ackermann    N8UR         jra@febo.com     http://www.febo.com
> President, TAPR                n8ur@tapr.org    http://www.tapr.org
> -
> To unsubscribe from this list: send the line "unsubscribe linux-hams" in
> the body of a message to majordomo@vger.kernel.org
> More majordomo info at  http://vger.kernel.org/majordomo-info.html
>

Re: Tracking down a Linux crash

[jbennett]

Oops...

Forgot to mention in my first reply--- running RH 7.3. Happens with
kernel-as-shipped, or custom w/AX25. Have upgraded kernel from RH and also
from kernel.org. No difference.

73,
John Bennett
n4xi

----- Original Message -----
From: "John Ackermann N8UR" <jra@febo.com>
To: <linux@lists.tapr.org>; <linux-hams@vger.kernel.org>
Sent: Wednesday, October 16, 2002 8:08 AM
Subject: Tracking down a Linux crash

Re: Tracking down a Linux crash

[terry]

On Wed, 16 Oct 2002, John Ackermann N8UR wrote:

... ciao:

> with VIA chipset

    i've had nothing but trouble with via chipsets.  however, i've
declined to upgrade to the 2.4 series kernel, so can not speak to its
accommodation to them.

    was your choice of motherboard driven by the HW HOWTO, or is the board
'supposed' to run under linux ...

-- 
73 , de WD0FPC
    aka: terry

Re: Tracking down a Linux crash

[Tomi Manninen OH2BNS]

On Wed, 16 Oct 2002, John Ackermann N8UR wrote:

> I don't think it's heat as the crash sometimes occurs relatively quickly 
> after powerup, and other times after days of running.  The CPU is not being 
> overclocked, and has an original AMD heatsink/fan, plus there's a fan on 
> the case.  However, I'm not monitoring CPU temperature in the system (don't 
> have that software installed yet) so I can't say for sure what the temp is.

Have you tried beating your system hard with something like Cerberus
(search google with key words "cerberus" and "ctcs") with the system
running in text mode? Cerberus was the only program to reliably report
memory failures on my new box half a year ago. All the others (including
memtest86) showed no errors, but still I had occasionally lockups and
programs crashing. A new memory module fixed everything.

-- 
Tomi Manninen           Internet:  oh2bns@sral.fi
OH2BNS                  AX.25:     oh2bns@oh2rbi.fin.eu
KP20ME04                Amprnet:   oh2bns@oh2rbi.ampr.org

Re: Tracking down a Linux crash

[w9ya]

On Wednesday 16 October 2002 08:08 am, John Ackermann N8UR wrote:
> Hi all --
>
> I realize this may be a bit off-topic, but my rationale is that (a) I'm
> among knowledgeable friends, and (b) this machine is destined to be my
> hamshack system, displacing a Windows box.
>
> I've been building Linux systems since back in the 0.99pl4 days, and for
> the first time I'm confronted with a system that regularly crashes.  There
> are unfortunately quite a few variables in the equation, so I doubt anyone
> here can give me an outright answer.  But perhaps with your collective
> experience, I can zero in on the problem.  Please bear with the length of
> this message as I want to provide as much detail as I can.
>
> The box is an Athlon XP2000+ on an Elite Group L-7VMM motherboard with VIA
> chipset.  On-board S3 SavagePro video, Realtek 8139 LAN, and VIA8233A
> sound.  512MB DDR, 40GB drive.  Expansion cards are an M-Audio Delta44
> sound card, and a Boca 8 port serial card.
>
> I've been loading Debian 3.0 (woody) with KDE[2,3].  I'm using a
> home-compiled kernel, both 2.4.18 and 2.4.19.  I'm not turning on any
> exotic features in the kernel (other than AX.25 :-) ) and I'm setting the
> CPU type to Athlon/K7.
>
> I've installed ALSA sound drivers and utils.  I've had some problems with
> the onboard VIA8233A sound, so I'm not currently enabling it.  I do have
> the ICE1712 ALSA module installed for the Delta44 card, and the sound
> system seems to be working OK with that card.  To get the S3 SavagePro
> video working in X
> I had to download a later version of the XFree-4 driver for that card; I
> was able to find a .deb file for it and it seems to install and work OK.
> I'm using the ext3 filesystem.
>
> The crash occurs after the system is fully booted and the KDE environment
> is up and running -- in fact, it may occur as much as a couple of days
> after bootup.  Several of the crashes have occurred when I was using
> Mozilla (version 1.0, installed from Debian), immediately after typing in a
> URL and hitting the return key -- but it's also crashed when Mozilla wasn't
> running.  I've had both KDE2 (standard for Debian 3.0) and KDE3 (by
> upgrading the system from woody to sarge and installing .debs from kde.org)
> installed and had crashes under both environments.  The crash does not seem
> to be related to sound card activity.
>
> The crash itself is unusual.  The symptom is that the X screen freezes,
> though usually the mouse pointer continues to move with the mouse. However,
> there's no response to keyboard or mouse-clicks, with one exception.  That
> is that if I use ctr-alt-F1 to go to a virtual console, the screen will
> switch to VGA mode, but remains completely black, with no login banner or
> anything.  After that point, nothing I do with the keyboard or the mouse
> yields a response. The keyboard LEDs stop working (at least at this point;
> I'm not sure if they stop before switching consoles).
>
> The network subsystem remains at least partially active.  I can ping the
> box from a remote host, but attempting to ssh to it fails with no reponse.
> I don't have a telnet server installed, but attempting to telnet from a
> remote machine to the normal port yields a "connection refused" message. If
> I telnet to port 25 (smtp), I get a connect, but exim does not respond to
> any commands sent to it.
>
> There's no trace left in the standard logs that I can tie to the crash.
>
> After rebooting by a power cycle or hardware reset, there are often some
> corrupted files left behind, and they're hosed in a pretty dramatic way --
> the ownership and dates are changed to impossibly large numbers, and it's
> often impossible to delete them without extraordinary effort -- you get an
> "unlink: operation not permitted" error and only using the filesystem
> editor to edit the inode entry lets you get to the point where you can
> delete the file.

I have had lock-ups from time tot ime while doing several video related items 
in KDE at once. I rather suspect the video driver or associated code. I have 
NEVER lost a file though, as I am running ext3 and it recovers gracefully 
enough without my doing anything except rebooting.

Bob Finch

>
> A web search hasn't shown any mention of problems using Linux with this
> motherboard or chipset.
>
> Well, that's a lot of info, but I'm hoping that someone out there will
> recognize something here that will point me in the right direction.  Thanks
> for any help!
>
> 73,
>
> John N8UR
> jra@febo.com
>
> ----
> John Ackermann    N8UR         jra@febo.com     http://www.febo.com
> President, TAPR                n8ur@tapr.org    http://www.tapr.org
> -
> To unsubscribe from this list: send the line "unsubscribe linux-hams" in
> the body of a message to majordomo@vger.kernel.org
> More majordomo info at  http://vger.kernel.org/majordomo-info.html

Re: Tracking down a Linux crash

[Haines Brown]

I don't have much to offer except commiseration ;-)

It would help to define just what happens to your system. If you
really mean "crash," in that the screen goes black, and your drive
access stops, and a reboot begins, etc., then the usual culprit may be
flakey RAM.

Had it happen to me twice over the years, where RAM replacement solved
the problem. Manufacturers usually pretty willing to replace RAM,
questions unasked, if covered by warranty.

If you've got RAM kicking around, try a substitution. If you have
plenty of RAM and can take out a bank, try that. Real RAM testers are
prohibitively expensive, but software testers I'm told work OK
(apparently better than they did ten years ago). If that's the route
for you, consider memtester-2.93.1. It compiles easily. The only catch
is that it runs and runs until you stop it. I didn't know that and
spent a week with it running in the background before it dawned on me
it would not complete by itself ;-(

If instead your system freezes (no keyboard our mouse inputs), forcing
you to RESET, you might have a problem with your video card. I'm not
positive, but I believe I once had this kind of problem because the
heat sink on the card had fallen off. When I put it back on with
thermal glue, the problem stopped (so far)..

Haines KB1GRM 

Re: Tracking down a Linux crash

[jjgpg]

Hi,

Have check all your memory. I had a simular prob. running Gnome 1. I was
using some old simms 70ns. But when I swiched to 133mhz dimms the prob
went ayway. 

Jeremiah KD7DMP

On Wed, 16 Oct 2002, John Ackermann N8UR wrote:

> Hi Kelly --
> 
> I don't think it's heat as the crash sometimes occurs relatively quickly 
> after powerup, and other times after days of running.  The CPU is not being 
> overclocked, and has an original AMD heatsink/fan, plus there's a fan on 
> the case.  However, I'm not monitoring CPU temperature in the system (don't 
> have that software installed yet) so I can't say for sure what the temp is.
> 
> Re the logs, there's nothing -- in particular, no sign of a kernel panic.
> 
> Thanks for the quick reply!
> 
> 73,
> John
> 
> --On Wednesday, October 16, 2002 8:20 AM -0500 Kelly Black 
> <kelly.black@penguinpackets.com> wrote:
> 
> > On Wed, 2002-10-16 at 08:08, John Ackermann N8UR wrote:
> >> Hi all --
> >>
> >> I realize this may be a bit off-topic, but my rationale is that (a) I'm
> >> among knowledgeable friends, and (b) this machine is destined to be my
> >> hamshack system, displacing a Windows box.
> >>
> > ---Snip long description---
> >
> > Could the problem be heat related?  What is the CPU temp when the
> > lockups occur (or just after a reboot).  Does anything show up in the
> > logs (kernel panics or anything suspicious?).
> >
> > Kelly Black
> > KB0GBJ
> >
> >
> >
> >
> 
> 
> 
> ----
> John Ackermann    N8UR         jra@febo.com     http://www.febo.com
> President, TAPR                n8ur@tapr.org    http://www.tapr.org
> 
> -
> To unsubscribe from this list: send the line "unsubscribe linux-hams" in
> the body of a message to majordomo@vger.kernel.org
> More majordomo info at  http://vger.kernel.org/majordomo-info.html
> 

Re: Tracking down a Linux crash

[John Ackermann N8UR]

--On Wednesday, October 16, 2002 10:40 AM -0400 Haines Brown 
<brownh@hartford-hwp.com> wrote:


> It would help to define just what happens to your system. If you
> really mean "crash," in that the screen goes black, and your drive
> access stops, and a reboot begins, etc., then the usual culprit may be
> flakey RAM.

Good point -- this may be more of a "freeze" than a "crash" -- the system 
does not reboot, and remains minimally responsive (eg, you can ping it). 
My original message describes the damage in more detail.

Thanks to all for the replies.  First, I'm going to do some memory tests 
and then possibly try swapping out the video (as I have a spare card I can 
plug in).  After that, I'm not sure...

73,
John

----
John Ackermann    N8UR         jra@febo.com     http://www.febo.com
President, TAPR                n8ur@tapr.org    http://www.tapr.org

Re: Tracking down a Linux crash

[Curt Mills, WE7U]

On 16 Oct 2002, [ISO-8859-1] Pär Flitt wrote:

> I've had a similar problem and it completely vanished when I replaced
> the graphics adapter. I only saw the problem when in X, mostly when
> surfing the web.

Roger on that one.  I had one machine (dual-boot Win95/SuSE Linux)
that was exhibiting similar problems.  I tracked it down to both the
video adapter _and_ a bad bank of memory.  Darned-near replaced
every piece of that computer before I found the two problems.

-- 
Curt Mills, WE7U                    hacker_NO_SPAM_@tc.fluke.com
Senior Methods Engineer/SysAdmin
"Lotto:    A tax on people who are bad at math!"
"Windows:  Microsoft's tax on computer illiterates!" -- WE7U
"The world DOES revolve around me:  I picked the coordinate system!"

-
To unsubscribe from this list: send the line "unsubscribe linux-hams" in
the body of a message to majordomo@vger.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html

Re: Tracking down a Linux crash

[Curt Mills, WE7U]

On Wed, 16 Oct 2002, jbennett wrote:

> motherboard with a P3-500 installed w/256 MB RAM. Video is an ATI 128 Rage.
> My laptop, a Dell Inspiron 8100 w/a P3-850 has done it on occasion. It uses
> the Nvidia chipset. I believe my friend's machine is a Dell 5100 w/a P3-550.
> I believe his video chipset is ATI. My gut instinct says it's in the X
> driver, but that's just a guess. Like you, have been unable to find anything
> on the Internet.  Will be interesting to see what turns up.

The video card I had the lockup problems with was an older ATI PCI
card.  I don't remember the model.  I replaced it with an S3 PCI
card and all is well.  Not that it assures the problem is with the
video or the ATI drivers, but it's another data point.

-- 
Curt Mills, WE7U                    hacker_NO_SPAM_@tc.fluke.com
Senior Methods Engineer/SysAdmin
"Lotto:    A tax on people who are bad at math!"
"Windows:  Microsoft's tax on computer illiterates!" -- WE7U
"The world DOES revolve around me:  I picked the coordinate system!"

RE: Tracking down a Linux crash

[Ben Gelb]

I had some problems with X freezing and virtual terminals disappearing
when using the Kernel framebuffer device. You might want to make sure
that kernel framebuffer support is not enabled in the kernel and that
your XConfig doesn't attempt to use the Kernel Framebuffer device.

-Ben



-----Original Message-----
From: linux-hams-owner@vger.kernel.org
[mailto:linux-hams-owner@vger.kernel.org] On Behalf Of John Ackermann
N8UR
Sent: Wednesday, October 16, 2002 9:09 AM
To: linux@lists.tapr.org; linux-hams@vger.kernel.org
Subject: Tracking down a Linux crash

Hi all --

I realize this may be a bit off-topic, but my rationale is that (a) I'm
among knowledgeable friends, and (b) this machine is destined to be my
hamshack system, displacing a Windows box.

snip

RE: Tracking down a Linux crash

[Ben Gelb]

I had some problems with X freezing and virtual terminals disappearing
when using the Kernel framebuffer device. You might want to make sure
that kernel framebuffer support is not enabled in the kernel and that
your XConfig doesn't attempt to use the Kernel Framebuffer device.

-Ben



-----Original Message-----
From: linux-hams-owner@vger.kernel.org
[mailto:linux-hams-owner@vger.kernel.org] On Behalf Of John Ackermann
N8UR
Sent: Wednesday, October 16, 2002 9:09 AM
To: linux@lists.tapr.org; linux-hams@vger.kernel.org
Subject: Tracking down a Linux crash

Hi all --

I realize this may be a bit off-topic, but my rationale is that (a) I'm
among knowledgeable friends, and (b) this machine is destined to be my
hamshack system, displacing a Windows box.

<Snip>

Re: Tracking down a Linux crash

[Nate Bargmann]

* terry <wd0fpc@aniota.com> [2002 Oct 16 16:37 -0500]:
> On Wed, 16 Oct 2002, John Ackermann N8UR wrote:
> 
> ... ciao:
> 
> > with VIA chipset
> 
>     i've had nothing but trouble with via chipsets.  however, i've
> declined to upgrade to the 2.4 series kernel, so can not speak to its
> accommodation to them.

I'm not running AX25 on this box, but I do have a Soyo KTV7A with a VIA
chipset here and an Athlon 1300.  Video card is an Abit board with the
GeForce 2 chipset.  Solid as a rock running Debian Testing and a custom
2.4.18 kernel.  I did some homework and found rave reviews on the Soyo
board and my experience has vindicated the the reviews,  Some other
boards I read about were not so stable.

John, have you tried dropping back to a simple window manager like
IceWm?  I use IceWm here, so I can't comment on KDE.

73, de Nate >>

-- 
 Wireless | Amateur Radio Station N0NB          | "We have awakened a
 Internet | n0nb@networksplus.net               | sleeping giant and
 Location | Bremen, Kansas USA EM19ov           | have instilled in him
  Amateur radio exams; ham radio; Linux info @  | a terrible resolve".
             http://www.qsl.net/n0nb/           | - Admiral Yamomoto

Re: Tracking down a Linux crash

[Bob Nielsen]

On Wed, Oct 16, 2002 at 07:18:05AM -0700, terry wrote:
> On Wed, 16 Oct 2002, John Ackermann N8UR wrote:
> 
> ... ciao:
> 
> > with VIA chipset
> 
>     i've had nothing but trouble with via chipsets.  however, i've
> declined to upgrade to the 2.4 series kernel, so can not speak to its
> accommodation to them.

There are a number of VIA chipsets.  I had an older one with a K6/2-350
and it worked pretty well, except for some problems with DMA and an IDE
(non-UDMA) drive (it worked fine with DMA disabled).

Currently I am running the same drive (without problems) in my shack
computer which has a MSI motherboard with a 1.0 GHz Celeron and the VIA
686 chipset running 2.4.19.  I haven't tried the onboard audio yet, but
it seems pretty solid otherwise.  I don't know offhand if the onboard
video is part of the VIA chipset or separate--it is reported as a
Trident CyberBlade/i1 and seems pretty crummy running X.  I'm about to
try an old S3 PCI video card to see if that helps, since I plan to use
this machine for PSK31.

Bob, N7XY

-- 
Bob Nielsen, N7XY                          n7xy@n7xy.net
Bainbridge Island, WA  
IOTA NA-065, USI WA-028S 

Re: Tracking down a Linux crash

[Bill Walton]

Hi John -

The others are much more qualified than me to comment on your problem but I do
have some information that I would like to contribute.  Recently I have built a
prototype system (RH7.2)(2.4.19)(Duron 1300)(MSI KT3 Ultra2)(250MB DDR
SDRAM)(30gig Maxtor 7200 HD)(TNT2 PCI Video Card)(D-Link Lan cards).
I had the same type of failure that your having and it was driving me nuts.  My
servers are in a 19" rack cabinet out in the garage.  As you can imagine during the
hot summer months the servers take a beating out there.  However I do run every
type of cooler gadget I can get my hands on...even HD coolers.  Believe it or not
they work great out there inspite of the heat.

In my case I had to determine if the CPU was locking up, or if the MB, or something
else was overheating.  I ran a heat test on the DURON 1300.  I put a variable
powersupply on the CPU fan and slowly starved it for air to see if I could duplicate
the problem that I was having.  The Duron 1300 locked up at 60 degrees C.  Even
on the hottest days I was not exceeding a CPU temperature of 50 degrees C max.
To make a long story short, my problem was due to a motherboard failure.  I traced
it down by using some component cooler.  I replaced the MB .... problem solved.

I don't know about the Athlon processor but the Duron runs extremely hot.  The
run of the mill CPU cooler down at the computer store just will not cut it.  I am
running  a copper heatsink CPU cooler now with a high speed fan, heat pipes, silver
grease, etc.  Even on the hottest days I don't exceed around 42 degrees C.

It would be interesting to know what temperature your CPU is running at!  I bet that
the Athlon would lock up at 60 degrees C.  If you have a good CPU cooler that
would be almost impossible to accomplish inside your house with the AC on.

Good Luck,

Bill KJ6EO

Re: Tracking down a Linux crash

[Volker Schroer]

John,
some months ago I had a similar problem due to dma problems with me hard
disk.

The system seemed to be frozen, but I found an error message by pressing
strg- alt -F10. If I remember well, I couldn't find the message in the logs.

73, Volker, DL1KSV

-- 
+++ GMX - Mail, Messaging & more  http://www.gmx.net +++
NEU: Mit GMX ins Internet. Rund um die Uhr für 1 ct/ Min. surfen!

-
To unsubscribe from this list: send the line "unsubscribe linux-hams" in
the body of a message to majordomo@vger.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html

Re: Tracking down a Linux crash

[Haines Brown]

> I put a variable powersupply on the CPU fan and slowly starved it
> for air to see if I could duplicate the problem that I was having.
> The Duron 1300 locked up at 60 degrees C.  Even on the hottest days
> I was not exceeding a CPU temperature of 50 degrees C max.  To make
> a long story short, my problem was due to a motherboard failure.  I
> traced it down by using some component cooler.  I replaced the MB
> .... problem solved.

Bill,

Could I ask for a little more detail? How were you measuring CPU
temperature? What do you mean by a "component cooler?" Did you select
areas of the MB to cool and then await the result?

Haines Brown KB1GRM

Re: Tracking down a Linux crash

[Bill Walton]

Hi Haines -

I was using a temperature probe wrapped around the bottom of the CPU
cooling fan.  The probe plugs into a Digital Volt Meter that I bought down
at Sears.  I put the system into a "kernel compile" then slowed down the
speed of the CPU cooling fan via the adjustable power supply.  When the
CPU would reach 60C (plus or minus a degree sometimes), it would just
sieze up "freeze".  The CPU would not recover so I would have to cool it off
and do a cold boot on the entire system.  During this test I would spray some
"component cooler" (freeze spray) lightly on the motherboard to keep it
cool.  You can get the freeze spray at almost any electronics store.  You can
also turn those cans of "compressed air" into freeze spray by holding them
upside down while spraying them.  Be advised that some of the compressed
air (canned) stuff is flammable.  Should you spray it into say some high
voltage it will ignite (you may have known that already).

Since I determined the sieze up temperature of the CPU all I had to do was
to stop cooling down the motherboard with the freeze spray.  And sure
enough at around 50 to 51C the system hung up.  I went down to the local
computer store and exchanged the motherboard....problem resolved.

In my opinion nobody should intentionally overheat their CPU.  In my case
the CPU was "inexpensive".  Furthermore I asked my friends down at the
computer store what type of heat the "Duron" could take before it would
sieze.  I work in the electronics industry, I own a small business and we
service high resolution X-Ray equipment down to component level.  I can't
begin to tell you how many electronics problems I have isolated by
cooling suspected components down with freeze spray (>:  It's just a trick
of the trade.

Regards,

Bill KJ6EO

Haines Brown wrote:

> > I put a variable powersupply on the CPU fan and slowly starved it
> > for air to see if I could duplicate the problem that I was having.
> > The Duron 1300 locked up at 60 degrees C.  Even on the hottest days
> > I was not exceeding a CPU temperature of 50 degrees C max.  To make
> > a long story short, my problem was due to a motherboard failure.  I
> > traced it down by using some component cooler.  I replaced the MB
> > .... problem solved.
>
> Bill,
>
> Could I ask for a little more detail? How were you measuring CPU
> temperature? What do you mean by a "component cooler?" Did you select
> areas of the MB to cool and then await the result?
>
> Haines Brown KB1GRM

Re: Tracking down a Linux crash (update)

[John Ackermann N8UR]

Hi --

Thanks to all for the many responses to my plea for help about my crashing 
Linux box.  I set off quite a discussion.

The majority of responses pointed the finger at a hardware problem, mainly 
faulty memory or possibly CPU overheating.  Tomi suggested "cerberus" which 
is a burn-in testing tool that VA-Linux used on their hardware products. 
It's available at http://sourceforge.net/projects/va-ctcs/ .  It does a 
pretty extreme job of exercising the CPU, memory, and hard disk, and since 
it runs under Linux, it can find some kernel instabilities as well.  I've 
been running it for about 24 hours now, and have had no failures.  I also 
checked the CPU temperature, and got 44 degrees C with both the CPU and the 
case fan apparently working properly, but the CPU not working very hard. 
That's way under the 90 degree maximum rating for the Athlon XP.  So, I 
don't think there's a CPU or memory problem.

(One note about cerberus -- under 2.4.x kernels, the memory test requires 
swap space of 2.5 to 4 times the physical memory.  When I tried to test my 
512MB of RAM with 1GB of swap, things crashed in a very ugly way.  But 
testing each 256MB stick separately works just fine.)

Right now I'm suspecting a video driver problem.  The S3 ProSavage chip on 
the motherboard is a relatively new one, and the XFree86-4 driver for it is 
quite new (it's not even included in the Debian xfree-server package).  I'm 
going to install an outboard AGP video card to see what difference that 
might make.

A couple of people have suggested trouble with AMD chipsets, and that might 
be the case, but changing mobos is kind of a last resort... I'll go there 
if nothing else works.

I had one suggestion that there's a bug in Mozilla when it encounters Flash 
pages and accesses the sound card.  I haven't been able to track down 
anything definitive about that, so I'm not sure if that bug causes a full 
system freeze like I'm seeing, or just a Mozilla crash.

Finally, one suggestion was to turn off the artsd sound server in KDE. 
I'll give that a try as well; the sound server doesn't really do any good 
for the applications I'm using, anyway.

Thanks to all for the suggestions; if I ever find out what's going on, I'll 
let you know.

73,
John

Re: Tracking down a Linux crash (update)

[Ray Heasman]

Hi John,

One more thing you should consider with an AMD CPU - make sure you have
"mem = nopentium" appended to your kernel boot line. This is usually set
in your lilo.conf using the "append" keyword.

This resolves (mostly) some nasty conflicts you can get between AMD
cache memory and AGP memory. This stopped my AMD system from randomly
freezing. Since then, the only reboots I have had have been to upgrade
kernels.

Oh, and the Mozilla-Flash freeze only freezes Mozilla, not your whole
system.

Cheers,
Ray
ZR6RAY

Re: Tracking down a Linux crash (update)

[John Ackermann N8UR]

Hi Ray --

Thanks for that tip; someone else mentioned it without providing any 
context but I just found more detail while looking at nvidea's website (I'm 
going to try using a Geforce card).

According to nvidea, kernel 2.4.19 is supposed to incorporate the fix for 
this problem.  Since that's what I'm running, I suspect that this isn't the 
cause (though my next step if I still crash with the new card will be to 
use the nopentium setting).

Thanks!

John

--On Friday, October 18, 2002 10:41 AM -0700 Ray Heasman <nurf@spamcop.net> 
wrote:

> Hi John,
>
> One more thing you should consider with an AMD CPU - make sure you have
> "mem = nopentium" appended to your kernel boot line. This is usually set
> in your lilo.conf using the "append" keyword.
>
> This resolves (mostly) some nasty conflicts you can get between AMD
> cache memory and AGP memory. This stopped my AMD system from randomly
> freezing. Since then, the only reboots I have had have been to upgrade
> kernels.
>
> Oh, and the Mozilla-Flash freeze only freezes Mozilla, not your whole
> system.
>
> Cheers,
> Ray
> ZR6RAY
>
>
>
>
>
>



----
John Ackermann    N8UR         jra@febo.com     http://www.febo.com
President, TAPR                n8ur@tapr.org    http://www.tapr.org