ReiserFS corruption with kernel 2.5.59

ReiserFS corruption with kernel 2.5.59

[venom]

HI,
I was using reiserFS for the root FS on a desktop, just to test kernel 2.5.59,
and I started to get those messages:

PAP-14030: direct2indirect: pasted or inserted byte exists in the tree [5094
5096 0x1001 IND]. Use fsck to repair.

Of course I repaired the FS.

System was under a really light I/O load, and stared to have those problems
after a couple of days of uptime.

System is:
Pentium III 933 Mhz, 256 KB L2 cache
512 MB RAM
i810 MB chipset

kernel is 2.5.59 compiled with gcc 3.2.1 and binutils 2.13.90.0.16

Disk is a Maxtor 32049H2, connected to an ATA66 intel 82801AA controller.


/proc/pci is:
PCI devices found:
  Bus  0, device   0, function  0:
    Host bridge: Intel Corp. 82815 815 Chipset Host Bridge and Memory Controller
Hub (rev 2).
      Prefetchable 32 bit memory at 0x44000000 [0x47ffffff].
  Bus  0, device   1, function  0:
    PCI bridge: Intel Corp. 82815 815 Chipset AGP Bridge (rev 2).
      Master Capable.  Latency=64.  Min Gnt=12.
  Bus  0, device  30, function  0:
    PCI bridge: Intel Corp. 82801AA PCI Bridge (rev 2).
      Master Capable.  No bursts.  Min Gnt=6.
  Bus  0, device  31, function  0:
    ISA bridge: Intel Corp. 82801AA ISA Bridge (LPC) (rev 2).
  Bus  0, device  31, function  1:
    IDE interface: Intel Corp. 82801AA IDE (rev 2).
      I/O at 0x2460 [0x246f].
  Bus  0, device  31, function  2:
    USB Controller: Intel Corp. 82801AA USB (rev 2).
      IRQ 19.
      I/O at 0x2440 [0x245f].
  Bus  0, device  31, function  5:
    Multimedia audio controller: Intel Corp. 82801AA AC'97 Audio (rev 2).
      IRQ 17.
      I/O at 0x2000 [0x20ff].
      I/O at 0x2400 [0x243f].
  Bus  1, device   0, function  0:
    VGA compatible controller: Matrox Graphics, Inc. MGA G400 AGP (rev 4).
      IRQ 18.
      Master Capable.  Latency=64.  Min Gnt=16.Max Lat=32.
      Prefetchable 32 bit memory at 0x42000000 [0x43ffffff].
      Non-prefetchable 32 bit memory at 0x40300000 [0x40303fff].
      Non-prefetchable 32 bit memory at 0x40800000 [0x40ffffff].
  Bus  2, device   8, function  0:
    Ethernet controller: 3Com Corporation 3c905C-TX/TX-M [Tornado] (rev 120).
      IRQ 16.
      Master Capable.  Latency=64.  Min Gnt=10.Max Lat=10.
      I/O at 0x1000 [0x107f].
      Non-prefetchable 32 bit memory at 0x40000000 [0x4000007f].

Hope this helps

Luigi

Re: ReiserFS corruption with kernel 2.5.59

[Oleg Drokin]

Hello!

On Mon, Jan 20, 2003 at 02:23:58PM +0100, venom@sns.it wrote:
> I was using reiserFS for the root FS on a desktop, just to test kernel 2.5.59,
> and I started to get those messages:
> PAP-14030: direct2indirect: pasted or inserted byte exists in the tree [5094
> 5096 0x1001 IND]. Use fsck to repair.
> Of course I repaired the FS.

Was the FS clean before you started to use 2.5.59?

Bye,
    Oleg

Re: ReiserFS corruption with kernel 2.5.59

[venom]

Yes, it was.

Before I started 2.5.59 I was running 2.4.20 with no problems at all.



On Mon, 20 Jan 2003, Oleg Drokin wrote:

> Date: Mon, 20 Jan 2003 16:43:44 +0300
> From: Oleg Drokin <green@namesys.com>
> To: venom@sns.it
> Cc: linux-kernel@vger.kernel.org
> Subject: Re: ReiserFS corruption with kernel 2.5.59
>
> Hello!
>
> On Mon, Jan 20, 2003 at 02:23:58PM +0100, venom@sns.it wrote:
> > I was using reiserFS for the root FS on a desktop, just to test kernel 2.5.59,
> > and I started to get those messages:
> > PAP-14030: direct2indirect: pasted or inserted byte exists in the tree [5094
> > 5096 0x1001 IND]. Use fsck to repair.
> > Of course I repaired the FS.
>
> Was the FS clean before you started to use 2.5.59?
>
> Bye,
>     Oleg
>

Re: ReiserFS corruption with kernel 2.5.59

[Oleg Drokin]

Hello!

On Mon, Jan 20, 2003 at 03:23:54PM +0100, venom@sns.it wrote:
> Yes, it was.
> Before I started 2.5.59 I was running 2.4.20 with no problems at all.

Hm. Interesting. I will look for the cause.
Let us know if you'd see this problem again, though (and if you'd be able
to find what sequence of events caused this to happen, that would be even
more good).

Thank you.

Bye,
    Oleg

Re: ReiserFS corruption with kernel 2.5.59

[venom]

It think it would be quite difficoult to find which sequence of events caused
this corruption.
The first message appeared saturday, while the desktop was just running X,
KDE 3.1 as desktop, (no users conencted or using it), and
the just running application was kslideshow.kss, to lock the desktop, showing
some nice good gif.
But /ust /opt and home of the user running this KDE was not the root partition
that has been corrupted.

also /var is in a different partition.

The only I/O that could be done on / partition was reading the so libraries in
/lib, and some configuration file in /etc.
Then every night updatedb has been runned, reading all partition.

The latest 2.5 kernel I tested on this desktop before 2.5.59 was 2.5.55, and
I am sure the filesystem was not corrupted with 2.5.55, (then I used 2.4.20 for
a while, and no corruption also in this case).


Bests
Luigi


On Mon, 20 Jan 2003, Oleg Drokin wrote:

> Date: Mon, 20 Jan 2003 17:28:26 +0300
> From: Oleg Drokin <green@namesys.com>
> To: venom@sns.it
> Cc: linux-kernel@vger.kernel.org
> Subject: Re: ReiserFS corruption with kernel 2.5.59
>
> Hello!
>
> On Mon, Jan 20, 2003 at 03:23:54PM +0100, venom@sns.it wrote:
> > Yes, it was.
> > Before I started 2.5.59 I was running 2.4.20 with no problems at all.
>
> Hm. Interesting. I will look for the cause.
> Let us know if you'd see this problem again, though (and if you'd be able
> to find what sequence of events caused this to happen, that would be even
> more good).
>
> Thank you.
>
> Bye,
>     Oleg
>