more system failures with Severn.... (on an Athlon)

[Elton Woo]

I don't have kmail installed (this time) ... third *successful* install.
Instead I have to use Mozilla mail. Starting Evolution, and I get
this error message:
"  Cannot activate component OAFIID:GNOME_Evolution_Mail_ShellComponent :
The error from the activation system is:
Unknown CORBA exception id: 'IDL:omg.org/CORBA/INV_OBJREF:1.0'  "
Evolution then loads, but without it's default contents and folders.

Also, sendmail is VERY SLOW to load, so the load process (even the graphical
one) 'hangs' for about 2 full minutes when almost at the 80% mark .... I'm
gauging the "80 %" as a "guesstimate" of the visual completion bar.

bugs? you betcha! ... and to think that some in this list are already 
running the
2.6 prerelease kernel.  I guess this means serious problems for the AMD 
users
here.  Compared to most of 'em on the list, I'm running a fairly "old" 
system
(about 4 years): AMD Athlon,  nVidia (3 D Labs) card, and Viewsonic 17.

Before my last crash and burn, I tried installing the nVidia drivers, and it
could not create the needed module. Sounds logical though, since I imagine
there would be some need of updating or patching to support the present
kernel in Severn.

Elton.

-- 
   http://setiathome.ssl.berkeley.edu/stats/team/team_4504.html
  "You only live once, so let's make life EASIER for each other."
  LINUX Registered User #193975. AMD-K7 ATHLON CPU power on board.

File system corruption with Severn

[djh]

On Wed, 30 Jul 2003, Alan Cox wrote:
>  If you can duplicate this reliably can you install the beta, then
> swap the
>  kernel for a generic kernel (either kernel.org, -ac or for that matter the
>  RH9 kernel) and see if it occurs again ? 
I can't reproduce it with the RH9 errata kernel.  (2.4.20-19.9)
Bug #101357 has the details.

David.

File system corruption with Severn

[djh]

On 28 Jul 2003, Jean-Francois wrote:
>  
>  I've got file system corruption using Severn twice, some time after the
>  installation, I'm using a  PII-800 with a Intel 440ZX chipset based
>  board, and a Quantum 40Gig hd. After repaired the filesystem, I've got a
>  ton of error messages in the messages log:
>  
>  hda: task_no_data_intr: status 0x51
>  hda: task_no_data_intr: error 0x04 
I'm seeing that as well (errors and fs corruption) on an old Athlon system 
of mine, with and without ACPI.
(Via KT133, onboard Promise ata100 and VIA ata66, Quantum 40G hd)

>  With RedHat 5,2 to 9.0 never had any problems on this computer, it
> was
>  very stable, now very unstable.  I've got strange message too, when
>  trying to execute something, it said something like unable to execute
>  binary file when I tried to run some command, then I restarted the
>  system and the file executed perfectly. 
That happens after ext3 gives up on it.  ("Journal has aborted")

Will add something to bugzilla later today.

David.

[Gerald Henriksen]

On Mon, 28 Jul 2003 16:20:38 -0400, you wrote:

> Before my last crash and burn, I tried installing the nVidia drivers,
> and it
> could not create the needed module. Sounds logical though, since I imagine
> there would be some need of updating or patching to support the present
> kernel in Severn. 
Nope.  The current NVIDIA drivers work just fine with the kernel
included with Severn once you tell the NVIDIA installer to use the
compiler used to compile the kernel by doing export CC=gcc32 first.

[Pekka Pietikäinen]

On Mon, Jul 28, 2003 at 04:40:51PM -0400, Elton Woo wrote:
>  >>Before my last crash and burn, I tried installing the nVidia
> drivers,
>  >>and it could not create the needed module. Sounds logical though, since
>  >>I imagine there would be some need of updating or patching to support
>  >>the present kernel in Severn.
>  I took a clue from Bill Nottingham who was responding to a message regading
>  kernel oopses with AMD's and the Taroon beta. Looks like the same case for
>  AMD and Severn... 
Mine has been rock solid (Athlon XP 2000, KT400)... Well not
quite, I did
have some serious nVidia-related stability problems. The fan on my
GeForce 4 blew up, which made the chip blow up, which made the 
system blow up (well, just crash all the time and print random 
characters/pixels on the screen). Apart from that, no problems with a
AMD+Severn combination.

-- 
Pekka Pietikainen

File system corruption with Severn

[Jean-Francois Bélanger]

I've got file system corruption using Severn twice, some time after the
installation, I'm using a  PII-800 with a Intel 440ZX chipset based
board, and a Quantum 40Gig hd. After repaired the filesystem, I've got a
ton of error messages in the messages log:

hda: task_no_data_intr: status 0x51
hda: task_no_data_intr: error 0x04

With RedHat 5,2 to 9.0 never had any problems on this computer, it was
very stable, now very unstable.  I've got strange message too, when
trying to execute something, it said something like unable to execute
binary file when I tried to run some command, then I restarted the
system and the file executed perfectly.

My newer pc run perfectly with Severn.

Don't know what this is....

File system corruption with Severn

[Alan Cox]

>  I've got file system corruption using Severn twice, some time
> after the
>  installation, I'm using a  PII-800 with a Intel 440ZX chipset based
>  board, and a Quantum 40Gig hd. After repaired the filesystem, I've got a 
Does this occur if you run with acpi=off ?

[Ethan Bonick]

>  I don't have kmail installed (this time) ... third *successful*
> install.
>  Instead I have to use Mozilla mail. Starting Evolution, and I get
>  this error message:
>  "  Cannot activate component OAFIID:GNOME_Evolution_Mail_ShellComponent
>  : The error from the activation system is:
>  Unknown CORBA exception id: 'IDL:omg.org/CORBA/INV_OBJREF:1.0'  "
>  Evolution then loads, but without it's default contents and folders.
>
>  Also, sendmail is VERY SLOW to load, so the load process (even the
>  graphical one) 'hangs' for about 2 full minutes when almost at the 80%
>  mark .... I'm gauging the "80 %" as a "guesstimate" of the
> visual
>  completion bar.
>
>  bugs? you betcha! ... and to think that some in this list are already
>  running the
>  2.6 prerelease kernel.  I guess this means serious problems for the AMD
>  users
>  here.  Compared to most of 'em on the list, I'm running a fairly "old"
>  system
>  (about 4 years): AMD Athlon,  nVidia (3 D Labs) card, and Viewsonic 17.
>
>  Before my last crash and burn, I tried installing the nVidia drivers,
>  and it could not create the needed module. Sounds logical though, since
>  I imagine there would be some need of updating or patching to support
>  the present kernel in Severn.
>  
I installed 2.6.0-test1 one today on ath 700 over ssh and rebooted the
machine, it came up fine and i just ssh'd in to check it. Granted I am not
running anything on it yet. I also did to my ath 1800 and rebooted, but it
didnt come back up. So its not "old" hardware causing it. I also have
rage128 pro vid on the 700. Will try beta on my ibm 600E later and maybe I
can have sound again wiht 2.6 kernel.

-- 
Ethan Bonick
etbonick_AT_networkinggeeks.com
http://www.networkinggeeks.com

File system corruption with Severn

[Matt Whiteley]

On Wed, 2003-07-30 at 13:15, Jean-Francois Bélanger wrote:
>  I'll try, but this time with a backup my mp3 collection before
> trying to
>  reinstall on this buggy pc, maybe tomorrow. 
>  
>  I wish I could reproduce this. but with no precious data on it this
>  time.... :) 
Is mounting file systems read-only enough to insure that that certain
file system will remain intact?

thanks,

-- 
Matt Whiteley <matt-whiteley(a)comcast.net&gt;

File system corruption with Severn

[Stephen Smoogen]

Well in most cases.. but if you try an experimental kernel it might not. 
EG if the kernel or utilities do not respect the read only bits, or say 
its on a software raid array, etc.

On 30 Jul 2003, Matt Whiteley wrote:

> On Wed, 2003-07-30 at 13:15, Jean-Francois Bélanger wrote:
> > I'll try, but this time with a backup my mp3 collection before trying to
> > reinstall on this buggy pc, maybe tomorrow. 
> > 
> > I wish I could reproduce this. but with no precious data on it this
> > time.... :)
>
> Is mounting file systems read-only enough to insure that that certain
> file system will remain intact?
>
> thanks,
>
>  
-- 
Stephen John Smoogen            smoogen(a)lanl.gov
Los Alamos National Labrador  CCN-5 Sched 5/40  PH: 5-8058
Ta-03 SM-261  MailStop P208 DP 17U  Los Alamos, NM 87545
-- So shines a good deed in a weary world. = Willy Wonka --

File system corruption with Severn

[Jean-Francois Bélanger]

On Wed, 2003-07-30 at 06:49, Alan Cox wrote:
>  > I've just switched the hard disk with a smaller one. Seem to
> work fine
>  > for now.  I've experienced some lockup too, but that must be my old
>  > i740. 
>  > 
>  > The corruption appeared while doing a lot of file transfer, i tried to
>  > disable the dma for the drive but the corruption appeared anyway. 
>  
>  If you can duplicate this reliably can you install the beta, then swap the
>  kernel for a generic kernel (either kernel.org, -ac or for that matter the
>  RH9 kernel) and see if it occurs again ?
>  
>  
>  --
>  Rhl-beta-list mailing list
>  Rhl-beta-list(a)redhat.com
>  http://www.redhat.com/mailman/listinfo/rhl-beta-list 
I'll try, but this time with a backup my mp3 collection before trying to
reinstall on this buggy pc, maybe tomorrow. 

I wish I could reproduce this. but with no precious data on it this
time.... :)

[Pekka Pietikäinen]

On Tue, Jul 29, 2003 at 01:23:05AM +0300, Pekka Pietikainen wrote:
>  On Mon, Jul 28, 2003 at 04:40:51PM -0400, Elton Woo wrote:
>  > >>Before my last crash and burn, I tried installing the nVidia drivers,
>  > >>and it could not create the needed module. Sounds logical though, since
>  > >>I imagine there would be some need of updating or patching to support
>  > >>the present kernel in Severn.
>  > I took a clue from Bill Nottingham who was responding to a message regading
>  > kernel oopses with AMD's and the Taroon beta. Looks like the same case for
>  > AMD and Severn...
>  Mine has been rock solid (Athlon XP 2000, KT400)... Well not quite, I did
>  have some serious nVidia-related stability problems. The fan on my
>  GeForce 4 blew up, which made the chip blow up, which made the  
Surprisingly the
box started crashing with the Severn kernel yesterday
(pci=noacpi and acpi=off tried), vanilla 2.4.21 and 2.6.0-test2 are happy.
Somehow this started just after I upgraded nvidia drivers to the 
one released a few days ago (although this happened even in text mode and
without nvidia.o ever being loaded). SysRq was the only thing that worked, 
rpm -Uvh <a couple of packages> was a certain way of triggering it... No oops.

-- 
Pekka Pietikainen

File system corruption with Severn

[Alan Cox]

>  I've just switched the hard disk with a smaller one. Seem to work
> fine
>  for now.  I've experienced some lockup too, but that must be my old
>  i740. 
>  
>  The corruption appeared while doing a lot of file transfer, i tried to
>  disable the dma for the drive but the corruption appeared anyway.  
If you can duplicate this reliably can you install the beta, then swap the
kernel for a generic kernel (either kernel.org, -ac or for that matter the
RH9 kernel) and see if it occurs again ?

File system corruption with Severn

[Jean-Francois Bélanger]

On Tue, 2003-07-29 at 21:57, djh wrote:
>  On 28 Jul 2003, Jean-Francois wrote:
>  > 
>  > I've got file system corruption using Severn twice, some time after the
>  > installation, I'm using a  PII-800 with a Intel 440ZX chipset based
>  > board, and a Quantum 40Gig hd. After repaired the filesystem, I've got a
>  > ton of error messages in the messages log:
>  > 
>  > hda: task_no_data_intr: status 0x51
>  > hda: task_no_data_intr: error 0x04
>  
>  I'm seeing that as well (errors and fs corruption) on an old Athlon system 
>  of mine, with and without ACPI.
>  (Via KT133, onboard Promise ata100 and VIA ata66, Quantum 40G hd)
>  
>  > With RedHat 5,2 to 9.0 never had any problems on this computer, it was
>  > very stable, now very unstable.  I've got strange message too, when
>  > trying to execute something, it said something like unable to execute
>  > binary file when I tried to run some command, then I restarted the
>  > system and the file executed perfectly.
>  
>  That happens after ext3 gives up on it.  ("Journal has aborted")
>  
>  Will add something to bugzilla later today.
>  
>  David.
>  
>  
>  --
>  Rhl-beta-list mailing list
>  Rhl-beta-list(a)redhat.com
>  http://www.redhat.com/mailman/listinfo/rhl-beta-list 
While in rescue mode from the cd, I've tried to check the disk with the
command: dd if=/dev/hda | wc -c 
Everything worked fine, no error, no messages.

I've just switched the hard disk with a smaller one. Seem to work fine
for now.  I've experienced some lockup too, but that must be my old
i740. 

The corruption appeared while doing a lot of file transfer, i tried to
disable the dma for the drive but the corruption appeared anyway.

[Ethan Bonick]

>  Ethan Bonick wrote:
>
> >>2.6 prerelease kernel.  I guess this means serious problems for the
> >> AMD users
> >>here.  Compared to most of 'em on the list, I'm running a fairly
> "old"
> >> system
> >>(about 4 years): AMD Athlon,  nVidia (3 D Labs) card, and Viewsonic
> >> 17.
> >>
> >>Before my last crash and burn, I tried installing the nVidia drivers,
> >> and it could not create the needed module. Sounds logical though,
> >> since I imagine there would be some need of updating or patching to
> >> support the present kernel in Severn.
> >>
> >>
> >>
> >
> >I installed 2.6.0-test1 one today on ath 700 over ssh and rebooted the
> > machine, it came up fine and i just ssh'd in to check it. Granted I am
> > not running anything on it yet. I also did to my ath 1800 and rebooted,
> > but it didnt come back up. So its not "old" hardware causing it. I also
> > have rage128 pro vid on the 700. Will try beta on my ibm 600E later and
> > maybe I can have sound again wiht 2.6 kernel.
> >
>  I took a clue from Bill Nottingham who was responding to a message
>  regading kernel oopses with AMD's and the Taroon beta. Looks like the
>  same case for AMD and Severn...
>  
After getting home the kernel had booted, but my netowrk wasnt working. So
I have both a slot one 700 and a socket a 1800 working with 2.6 test1 rpm.

-- 
Ethan Bonick
etbonick_AT_networkinggeeks.com
http://www.networkinggeeks.com

[Robert P. J. Day]

On Mon, 28 Jul 2003, Elton Woo wrote:

>  Before my last crash and burn, I tried installing the nVidia drivers,
> and it
>  could not create the needed module. Sounds logical though, since I imagine
>  there would be some need of updating or patching to support the present
>  kernel in Severn. 
this part i think i can help with.  i built the nvidia driver for severn
and its stock kernel with their 4363 release.  it seemed to work fine.
you only need the patch at www.minion.de for the 2.5.xx and 2.6 kernels,
AFAIK.

rday

[Robert P. J. Day]

On Mon, 28 Jul 2003, Elton Woo wrote:

>  bugs? you betcha! ... and to think that some in this list are
> already
>  running the 2.6 prerelease kernel. 
time to toss in the towel, methinks.  i've been struggling for the last
day or so to try to get a working system of:

  dell inspiron 8100
  severn base install
  kernel-2.6.0 testx kernel
  nvidia driver for geforce2go card
  linksys wireless pcmcia card
  logitech optical PS/2 mouse

as it stands, i can generally get some subset of the above working
at any one time, but so far, have been unable to get it all going at
once.

if i can get wireless, for some reason, the nvidia driver fails to build.  
if i try to use the 2.6.0 kernel, i lose wireless.  my last attempt cost
me my optical PS/2 mouse. any attempt to use the 2.6.0-test2 kernel causes
a kernel panic.

now, i'm aware that, this being a severn list, use of a beta kernel is
outside the mandate of the list, but i'm pretty spoiled by the new
kernel, and i've been using it happily for several weeks now under
RH 9.  so it's frustrating to think that, to test severn, i have to
sacrifice my kernel.  

time to back off to RH 9 and wait for the next beta, i think.  bummer.

rday

[Robert P. J. Day]

On Mon, 28 Jul 2003, Elton Woo wrote:

>  bugs? you betcha! ... and to think that some in this list are
> already
>  running the 2.6 prerelease kernel. 
then there are those of us for whom the 2.6.0-test2 kernel explodes
into panic at boot time. :-(

rday

File system corruption with Severn

[Mark W. Jeanmougin]

On 28 Jul 2003, Jean-Francois wrote:
>  I've got file system corruption using Severn twice, some time
> after the
>  installation, I'm using a  PII-800 with a Intel 440ZX chipset based
>  board, and a Quantum 40Gig hd. After repaired the filesystem, I've got a
>  ton of error messages in the messages log:
>  
>  hda: task_no_data_intr: status 0x51
>  hda: task_no_data_intr: error 0x04 
Sounds hardware related to me.  Try executing

dd if=/dev/hda | wc -c

(That command will read your entire hard drive and then output a count of 
how many bytes were read)

Then, keep an eye on syslog for error messages.

>  With RedHat 5,2 to 9.0 never had any problems on this computer, it
> was
>  very stable, now very unstable.  I've got strange message too, when
>  trying to execute something, it said something like unable to execute
>  binary file when I tried to run some command, then I restarted the
>  system and the file executed perfectly. 
If it's been stable under other OS's, try booting to one of them and then 
trying the same command.  Or, boot off one of their install cd's (using 
'linux rescue') and run the same command.

It sounds to me like your drive may be going bad.  If you don't get any 
errors from that command, then you know that your drive is reading OK.

MJ

[Elton Woo]

Ethan Bonick wrote:

> >2.6 prerelease kernel.  I guess this means serious problems for
> the AMD
> >users
> >here.  Compared to most of 'em on the list, I'm running a fairly
> "old"
> >system
> >(about 4 years): AMD Athlon,  nVidia (3 D Labs) card, and Viewsonic 17.
> >
> >Before my last crash and burn, I tried installing the nVidia drivers,
> >and it could not create the needed module. Sounds logical though, since
> >I imagine there would be some need of updating or patching to support
> >the present kernel in Severn.
> >
> >    
> >
>
> I installed 2.6.0-test1 one today on ath 700 over ssh and rebooted the
> machine, it came up fine and i just ssh'd in to check it. Granted I am not
> running anything on it yet. I also did to my ath 1800 and rebooted, but it
> didnt come back up. So its not "old" hardware causing it. I also have
> rage128 pro vid on the 700. Will try beta on my ibm 600E later and maybe I
> can have sound again wiht 2.6 kernel.
>  
I took a clue from Bill Nottingham who was responding to a message regading
kernel oopses with AMD's and the Taroon beta. Looks like the same case for
AMD and Severn...

Elton.

-- 
   http://setiathome.ssl.berkeley.edu/stats/team/team_4504.html
  "You only live once, so let's make life EASIER for each other."
  LINUX Registered User #193975. AMD-K7 ATHLON CPU power on board.