funky tyan s2510

funky tyan s2510

[josh]

I have a tyan s2510 with a single pIII 800Mhz cpu and 1GB of RAM.
I have been having problems with this system from the get go and
cant seem to narrow down what the problem is.  I have tried running
2.4.6, but the system usually doesnt last more than a day.  With 
2.4.2 i get a variety of kernel messages:
#############################################
vs-5150: search_by_key: invalid format found in block 0. Fsck?
vs-13050: reiserfs_update_sd: i/o failure occurred trying to update [59906829 0x0 SD] stat datavs-13060: reiserfs_update_sd: stat data of object [10616 10762 0x0 SD] (nlink == 1) not found (pos 6)
vs-13060: reiserfs_update_sd: stat data of object [8766 9018 0x0 SD] (nlink == 1) not found (pos 0)
vs-13060: reiserfs_update_sd: stat data of object [8766 8959 0x0 SD] (nlink == 1) not found (pos 12)
vs-13060: reiserfs_update_sd: stat data of object [8766 8959 0x0 SD] (nlink == 1) not found (pos 12)
memory.c:303: bad pmd 8524468b.
memory.c:83: bad pmd 831074c0.
memory.c:83: bad pmd 8524468b.
memory.c:83: bad pmd 831074c0.
memory.c:83: bad pmd 8524468b.
memory.c:83: bad pmd 831074c0.
memory.c:83: bad pmd 8524468b.
memory.c:83: bad pmd 831074c0.
kernel BUG at inode.c:79!
invalid operand: 0000
CPU:    0
EIP:    0010:[<c01418fe>]
EFLAGS: 00010282
eax: 0000001a   ebx: f5e3c2a0   ecx: d99ae000   edx: 00000000
esi: c027e120   edi: dc12a620   ebp: d99affa4   esp: d99aff44
ds: 0018   es: 0018   ss: 0018
Process rm (pid: 13400, stackpage=d99af000)
Stack: c02301a5 c0230225 0000004f f5e3c2a0 c01428b6 f5e3c2a0 dc12a620 f5e3c2a0
       c01413c6 f5e3c2a0 00000000 f5d1ea00 c013b1fc dc12a620 dc12a620 dc12a620
       d82e0000 c013b2d3 f5d1ea00 dc12a620 d99ae000 00000000 bffffe82 bffffc48
Call Trace: [<c01428b6>] [<c01413c6>] [<c013b1fc>] [<c013b2d3>] [<c0108d5f>]

Code: 0f 0b 83 c4 0c a1 90 09 2d c0 53 50 e8 d9 65 fe ff 83 c4 08
############################################################

gcc never gets all the way through a make... it will die with a
sig11, misc asm errors, or random crap.

This is a serverworks chipset... i have always thought that they were
a bit, you know, funny.  :)   

Any ideas?

                          www.mammoth.org/~skulcap
**********************************************BEGIN GEEK CODE BLOCK************
"Sometimes, if you're perfectly      * GCS d- s: a-- C++ ULSC++++$ P+ L+++ E--- 
still, you can hear the virgin       * W+(++) N++ o+ K- w--(---) O- M- V- PS-- 
weeping for the savior of your will."* PE Y+ PGP t+ 5 X+ R !tv b+>+++ DI++ D++  
 --DreamTheater, "Lines in the Sand" * G e h+ r-- y-   (www.geekcode.com)
**********************************************END GEEK CODE BLOCK**************

Re: funky tyan s2510

[Alan Cox]

> gcc never gets all the way through a make... it will die with a
> sig11, misc asm errors, or random crap.

If its doing that at random then suspect hardaware

> This is a serverworks chipset... i have always thought that they were
> a bit, you know, funny.  :)   

Serverworks have an obscure MTRR bug in a few chips (which we handle) but
quite honestly they don't show up a lot in kernel bug reports.



Alan

Re: funky tyan s2510

[josh]

On Fri, 6 Jul 2001, Alan Cox wrote:

> > gcc never gets all the way through a make... it will die with a
> > sig11, misc asm errors, or random crap.
> 
> If its doing that at random then suspect hardaware

Thats what I thought at first, but after going through three different 
brands of memory and a new (slower) cpu, im not sure what to think.

> > This is a serverworks chipset... i have always thought that they were
> > a bit, you know, funny.  :)   
> 
> Serverworks have an obscure MTRR bug in a few chips (which we handle) but
> quite honestly they don't show up a lot in kernel bug reports.

I actually like SW chipsets... they do some neat things... more than
me at least. 


                          www.mammoth.org/~skulcap
**********************************************BEGIN GEEK CODE BLOCK************
"Sometimes, if you're perfectly      * GCS d- s: a-- C++ ULSC++++$ P+ L+++ E--- 
still, you can hear the virgin       * W+(++) N++ o+ K- w--(---) O- M- V- PS-- 
weeping for the savior of your will."* PE Y+ PGP t+ 5 X+ R !tv b+>+++ DI++ D++  
 --DreamTheater, "Lines in the Sand" * G e h+ r-- y-   (www.geekcode.com)
**********************************************END GEEK CODE BLOCK**************

Re: funky tyan s2510

[Ignacio Vazquez-Abrams]

On Fri, 6 Jul 2001, josh wrote:

> Thats what I thought at first, but after going through three different
> brands of memory and a new (slower) cpu, im not sure what to think.

It could also be other peripherals like video or the NIC. I personally had
Linux (2.0.x kernel, WAAAY back when...) flake out on me because of a bad
D-Link DE-220 ISA NIC.

-- 
Ignacio Vazquez-Abrams  <ignacio@openservices.net>

Re: funky tyan s2510

[arjan]

In article <Pine.LNX.4.21.0107061109400.20154-100000@hannibal.mammoth.org> you wrote:

> I have a tyan s2510 with a single pIII 800Mhz cpu and 1GB of RAM.
> I have been having problems with this system from the get go and
> cant seem to narrow down what the problem is.  I have tried running
> 2.4.6, but the system usually doesnt last more than a day.  With 
> 2.4.2 i get a variety of kernel messages:
> #############################################
> vs-5150: search_by_key: invalid format found in block 0. Fsck?

IDE on a serverworks chipset ? If so try Alan's latest patches for an
updated driver

Re: funky tyan s2510

[Alan Cox]

> > 2.4.2 i get a variety of kernel messages:
> > #############################################
> > vs-5150: search_by_key: invalid format found in block 0. Fsck?
> 
> IDE on a serverworks chipset ? If so try Alan's latest patches for an
> updated driver

The serverworks IDE fixes dont fit the reports , and the gcc one so says
'hardware problem'

Re: funky tyan s2510

[ognen]

I have had this happen to me on a very old machine whose CPU fan died at
one point without notice and something burned out - the machine was
working but crashing at random with all sorts of messages and in all sorts
of situations, usually some time after startup (but not too long after
startup).

Best regards,
Ognen

On Fri, 6 Jul 2001, josh wrote:

>
> I have a tyan s2510 with a single pIII 800Mhz cpu and 1GB of RAM.
> I have been having problems with this system from the get go and
> cant seem to narrow down what the problem is.  I have tried running
> 2.4.6, but the system usually doesnt last more than a day.  With
> 2.4.2 i get a variety of kernel messages:
> #############################################
> vs-5150: search_by_key: invalid format found in block 0. Fsck?
> vs-13050: reiserfs_update_sd: i/o failure occurred trying to update [59906829 0x0 SD] stat datavs-13060: reiserfs_update_sd: stat data of object [10616 10762 0x0 SD] (nlink == 1) not found (pos 6)
> vs-13060: reiserfs_update_sd: stat data of object [8766 9018 0x0 SD] (nlink == 1) not found (pos 0)
> vs-13060: reiserfs_update_sd: stat data of object [8766 8959 0x0 SD] (nlink == 1) not found (pos 12)
> vs-13060: reiserfs_update_sd: stat data of object [8766 8959 0x0 SD] (nlink == 1) not found (pos 12)
> memory.c:303: bad pmd 8524468b.
> memory.c:83: bad pmd 831074c0.
> memory.c:83: bad pmd 8524468b.
> memory.c:83: bad pmd 831074c0.
> memory.c:83: bad pmd 8524468b.
> memory.c:83: bad pmd 831074c0.
> memory.c:83: bad pmd 8524468b.
> memory.c:83: bad pmd 831074c0.
> kernel BUG at inode.c:79!
> invalid operand: 0000
> CPU:    0
> EIP:    0010:[<c01418fe>]
> EFLAGS: 00010282
> eax: 0000001a   ebx: f5e3c2a0   ecx: d99ae000   edx: 00000000
> esi: c027e120   edi: dc12a620   ebp: d99affa4   esp: d99aff44
> ds: 0018   es: 0018   ss: 0018
> Process rm (pid: 13400, stackpage=d99af000)
> Stack: c02301a5 c0230225 0000004f f5e3c2a0 c01428b6 f5e3c2a0 dc12a620 f5e3c2a0
>        c01413c6 f5e3c2a0 00000000 f5d1ea00 c013b1fc dc12a620 dc12a620 dc12a620
>        d82e0000 c013b2d3 f5d1ea00 dc12a620 d99ae000 00000000 bffffe82 bffffc48
> Call Trace: [<c01428b6>] [<c01413c6>] [<c013b1fc>] [<c013b2d3>] [<c0108d5f>]
>
> Code: 0f 0b 83 c4 0c a1 90 09 2d c0 53 50 e8 d9 65 fe ff 83 c4 08
> ############################################################
>
> gcc never gets all the way through a make... it will die with a
> sig11, misc asm errors, or random crap.
>
> This is a serverworks chipset... i have always thought that they were
> a bit, you know, funny.  :)
>
> Any ideas?

-- 
Ognen Duzlevski
Plant Biotechnology Institute
National Research Council of Canada
Bioinformatics team