[parisc-linux] a b180 panic with 2.6.11-rc3-pa3

[parisc-linux] a b180 panic with 2.6.11-rc3-pa3

[Joel Soete]

Hi all,

Don't know if somebody else also emcountered such panic:
Backtrace:
 [<101bb588>] journal_add_journal_head+0xdc/0x13c
 [<101b4dd8>] journal_dirty_data+0x64/0x1b0
 [<101a6be4>] ext3_journal_dirty_data+0x1c/0x60
 [<101a69a8>] walk_page_buffers+0xe0/0xf0
 [<101a72b8>] ext3_ordered_writepage+0x168/0x1fc
 [<1018d770>] mpage_writepages+0x28c/0x3a0
 [<1018bae4>] __sync_single_inode+0x5c/0x278
 [<1018bd68>] __writeback_single_inode+0x68/0x168
 [<1018c034>] sync_sb_inodes+0x1cc/0x310
 [<1018c1e8>] writeback_inodes+0x70/0xd4
 [<10149088>] background_writeout+0xb4/0x10c
 [<10149e40>] __pdflush+0x138/0x204
 [<10149f2c>] pdflush+0x20/0x2c
 [<1013c564>] kthread+0xdc/0xe4
 [<1010ac5c>] ret_from_kernel_thread+0x1c/0x24


Kernel Fault: Code=3D15 regs=3D17c84840 (Addr=3D20702352)

     YZrvWESTHLNXBCVMcbcbcbcbOGFRQPDI
PSW: 00000000000001101111111100001111 Not tainted
r00-03  00000000 10201800 101bb588 20702352
r04-07  122be660 122be660 122be660 00000000
r08-11  00000000 00000000 00001000 1230c3e8
r12-15  101a7120 17d1a990 00002000 00000000
r16-19  101a7150 17c84508 00000000 00000002
r20-23  00000000 00000000 20702352 00000000
r24-27  00000030 00000000 20702352 10375010
r28-31  20702352 100c74f4 17c84840 100c74fc
sr0-3   00000000 00000000 00000000 0000534b
sr4-7   00000000 00000000 00000000 00000000

IASQ: 00000000 00000000 IAOQ: 10201f1c 10201f20
 IIR: 0ed91222    ISR: 00000000  IOR: 20702352
 CPU:        0   CR30: 17c84000 CR31: 103d0000
 ORIG_R28: 17c84980
 IAOQ[0]: memset+0x24/0xa8
 IAOQ[1]: memset+0x28/0xa8
 RP(r2): journal_add_journal_head+0xdc/0x13c
Kernel panic - not syncing: Kernel Fault

Here on a b180 running 2.6.11-rc3-pa3 (while the system should be idle)

Any idea about known ext3 pb?

Joel



_______________________________________________
parisc-linux mailing list
parisc-linux@lists.parisc-linux.org
http://lists.parisc-linux.org/mailman/listinfo/parisc-linux

Re: [parisc-linux] a b180 panic with 2.6.11-rc3-pa3

[Max Grabert]

On Mon, 14 Feb 2005 11:30:43 +0000, Joel Soete <soete.joel@tiscali.be> wrote:
> Hi all,
> 
> Don't know if somebody else also emcountered such panic:
> Backtrace:

> [<101bb588>] journal_add_journal_head+0xdc/0x13c
> [<101b4dd8>] journal_dirty_data+0x64/0x1b0
> [<101a6be4>] ext3_journal_dirty_data+0x1c/0x60
> [<101a69a8>] walk_page_buffers+0xe0/0xf0
> [<101a72b8>] ext3_ordered_writepage+0x168/0x1fc
> [<1018d770>] mpage_writepages+0x28c/0x3a0
> [<1018bae4>] __sync_single_inode+0x5c/0x278
> [<1018bd68>] __writeback_single_inode+0x68/0x168
> [<1018c034>] sync_sb_inodes+0x1cc/0x310
> [<1018c1e8>] writeback_inodes+0x70/0xd4
> [<10149088>] background_writeout+0xb4/0x10c
> [<10149e40>] __pdflush+0x138/0x204
> [<10149f2c>] pdflush+0x20/0x2c
> [<1013c564>] kthread+0xdc/0xe4
> [<1010ac5c>] ret_from_kernel_thread+0x1c/0x24
>
> [...]
> 
> IASQ: 00000000 00000000 IAOQ: 10201f1c 10201f20
> IIR: 0ed91222    ISR: 00000000  IOR: 20702352
> CPU:        0   CR30: 17c84000 CR31: 103d0000
> ORIG_R28: 17c84980
> IAOQ[0]: memset+0x24/0xa8
> IAOQ[1]: memset+0x28/0xa8
> RP(r2): journal_add_journal_head+0xdc/0x13c
> Kernel panic - not syncing: Kernel Fault
> 
> Here on a b180 running 2.6.11-rc3-pa3 (while the system should be idle)

Long time ago I have seen a couple of those. I can't remember what
exact kernel version(s) it was, but it happened with the earlier 2.6
kernels.
Pretty much the same output (I think I once posted it on the mailing
list), but I only got this when I tried out some graphics card (ATI,
Voodoo2 or Matrox?) and shortly after xdm started (and not every
time). I think 64bit kernels triggered it more easily.

But I haven't gotten one of those lately (at least not with 2.6.10-rc1
and later, with Matrox graphics card).

> Any idea about known ext3 pb?

Good question. I have a different problem than the one above: In very
rare occations, when transferring very big files onto a USB-hdd
(ext3), the files are being silently corrupted (no kernel log
messages). But I don't know whether that problem is related to ext3 or
usb-storage or a faulty h/w.

Back to your problem: I don't know/think that there are any severe
problems with ext3, at least I was using it on PA for a long time and
it worked quite well. But there is always the possibility that there
are some hidden bugs that are triggered in rare situations however ...


Greetings,
  Max
_______________________________________________
parisc-linux mailing list
parisc-linux@lists.parisc-linux.org
http://lists.parisc-linux.org/mailman/listinfo/parisc-linux

Re: [parisc-linux] a b180 panic with 2.6.11-rc3-pa3

[Joel Soete]

Hello Max,

>
> On Mon, 14 Feb 2005 11:30:43 +0000, Joel Soete <soete.joel@tiscali.be>
wrote:
> > Hi all,
> >
> > Don't know if somebody else also emcountered such panic:
> > Backtrace:
>
> > [<101bb588>] journal_add_journal_head+0xdc/0x13c
> > [<101b4dd8>] journal_dirty_data+0x64/0x1b0
> > [<101a6be4>] ext3_journal_dirty_data+0x1c/0x60
> > [<101a69a8>] walk_page_buffers+0xe0/0xf0
> > [<101a72b8>] ext3_ordered_writepage+0x168/0x1fc
> > [<1018d770>] mpage_writepages+0x28c/0x3a0
> > [<1018bae4>] __sync_single_inode+0x5c/0x278
> > [<1018bd68>] __writeback_single_inode+0x68/0x168
> > [<1018c034>] sync_sb_inodes+0x1cc/0x310
> > [<1018c1e8>] writeback_inodes+0x70/0xd4
> > [<10149088>] background_writeout+0xb4/0x10c
> > [<10149e40>] __pdflush+0x138/0x204
> > [<10149f2c>] pdflush+0x20/0x2c
> > [<1013c564>] kthread+0xdc/0xe4
> > [<1010ac5c>] ret_from_kernel_thread+0x1c/0x24
> >
> > [...]
> >
> > IASQ: 00000000 00000000 IAOQ: 10201f1c 10201f20
> > IIR: 0ed91222    ISR: 00000000  IOR: 20702352
> > CPU:        0   CR30: 17c84000 CR31: 103d0000
> > ORIG_R28: 17c84980
> > IAOQ[0]: memset+0x24/0xa8
> > IAOQ[1]: memset+0x28/0xa8
> > RP(r2): journal_add_journal_head+0xdc/0x13c
> > Kernel panic - not syncing: Kernel Fault
> >
> > Here on a b180 running 2.6.11-rc3-pa3 (while the system should be idl=
e)
>
mmm today on this same system with rc4-pa1 now:
Backtrace:
 [<101bb644>] journal_add_journal_head+0xdc/0x13c
 [<101b4e94>] journal_dirty_data+0x64/0x1b0
 [<101a6ca0>] ext3_journal_dirty_data+0x1c/0x60
 [<101a6a64>] walk_page_buffers+0xe0/0xf0
 [<101a6ddc>] ext3_ordered_commit_write+0x68/0x144
 [<10144db8>] generic_file_buffered_write+0x23c/0x68c
 [<10145474>] __generic_file_aio_write_nolock+0x26c/0x450
 [<101458c0>] generic_file_aio_write+0x88/0x164
 [<101a3858>] ext3_file_write+0x34/0xd0
 [<101658b4>] do_sync_write+0xb4/0xf4
 [<101659d4>] vfs_write+0xe0/0x158
 [<10165b50>] sys_write+0x5c/0xa8
 [<1010c120>] syscall_exit+0x0/0x14
...
IASQ: 00000000 00000000 IAOQ: 10201fe0 10201fe4
 IIR: 0ed91222    ISR: 00000000  IOR: d2b304d2
 CPU:        0   CR30: 108e8000 CR31: 103d0000
 ORIG_R28: 108e8900
 IAOQ[0]: memset+0x24/0xa4
 IAOQ[1]: memset+0x28/0xa4
 RP(r2): journal_add_journal_head+0xdc/0x13c
Kernel panic - not syncing: Kernel Fault

This time while 'apt-get dist-update' on a chroot disk :-(

I tried to reboot but pdc didn't find any more boot devices:
...
Main Menu: Enter command > sea
...
   Path Number        Device Path              Device Type
   -----------        -----------------        -----------
Error detected on single-ended FWSCSI interface.
No devices found.
...

even after a reset.
I have to cycle power off/on the external disk (a fw scsi disk connected
to the fw ctrlr) to be able to reboot then complete the apt update withou=
t
any pb??

...
> Pretty much the same output (I think I once posted it on the mailing
> list), but I only got this when I tried out some graphics card (ATI,
> Voodoo2 or Matrox?) and shortly after xdm started (and not every
> time). I think 64bit kernels triggered it more easily.
>
Here not additional pci card
...
>
> > Any idea about known ext3 pb?
>
> Good question. I have a different problem than the one above: In very
> rare occations, when transferring very big files onto a USB-hdd
> (ext3), the files are being silently corrupted (no kernel log
> messages).
mmm no such device available here to make test (sorry) (on a b2k obviousl=
y
;-) )

> But I don't know whether that problem is related to ext3 or
> usb-storage or a faulty h/w.
>
Considering the strange status in which was the external disk, it should
be possible to be some h/w failure demonstration occuring from time to ti=
me??

> Back to your problem: I don't know/think that there are any severe
> problems with ext3, at least I was using it on PA for a long time and
> it worked quite well.
The same for me ;-)

> But there is always the possibility that there
> are some hidden bugs that are triggered in rare situations however ...
>
Thanks for feedback,
    Joel


_______________________________________________
parisc-linux mailing list
parisc-linux@lists.parisc-linux.org
http://lists.parisc-linux.org/mailman/listinfo/parisc-linux

Re: [parisc-linux] a b180 panic with 2.6.11-rc3-pa3

[Thibaut VARENE]

On Mon, 14 Feb 2005 22:13:59 +0000
Max Grabert <parisc@gmail.com> wrote:

> On Mon, 14 Feb 2005 11:30:43 +0000, Joel Soete <soete.joel@tiscali.be>
> wrote:
> > Hi all,
> > 
> > Don't know if somebody else also emcountered such panic:
> > Backtrace:
> 
> > [<101bb588>] journal_add_journal_head+0xdc/0x13c
> > [<101b4dd8>] journal_dirty_data+0x64/0x1b0
> > [<101a6be4>] ext3_journal_dirty_data+0x1c/0x60
> > [<101a69a8>] walk_page_buffers+0xe0/0xf0
> > [<101a72b8>] ext3_ordered_writepage+0x168/0x1fc

Got hit by a similar panic during the night (the kernel uptime was < 5h)

Kernel Fault code 15
Top stack was:

cache_grow
cache_alloc_refill
kmem_cache_alloc
journal_alloc_journal

and the rest was identical to your dump.

IAOQ was alloc_slabmgmt and gr2 was cache_grow

This was 2.6.11-rc3-pa5 freshly built with Kernel Hacking turned off.

HTH

Thibaut VARENE
The PA/Linux ESIEE Team
http://www.pateam.org/
_______________________________________________
parisc-linux mailing list
parisc-linux@lists.parisc-linux.org
http://lists.parisc-linux.org/mailman/listinfo/parisc-linux

Re: [parisc-linux] a b180 panic with 2.6.11-rc3-pa3

[Joel Soete]

...
> > 
> > > [<101bb588>] journal_add_journal_head+0xdc/0x13c
> > > [<101b4dd8>] journal_dirty_data+0x64/0x1b0
> > > [<101a6be4>] ext3_journal_dirty_data+0x1c/0x60
> > > [<101a69a8>] walk_page_buffers+0xe0/0xf0
> > > [<101a72b8>] ext3_ordered_writepage+0x168/0x1fc
> 
> Got hit by a similar panic during the night (the kernel uptime was < 5h=
)
> 
> Kernel Fault code 15
> Top stack was:
> 
> cache_grow
> cache_alloc_refill
> kmem_cache_alloc
> journal_alloc_journal
> 
> and the rest was identical to your dump.
> 
> IAOQ was alloc_slabmgmt and gr2 was cache_grow
> 
I just come back because after a second read, it looks like the b2k pb re=
lated
here:
<http://lists.parisc-linux.org/pipermail/parisc-linux/2005-February/02587=
0.html>

mmm hard to say if there are actualy the same?

> This was 2.6.11-rc3-pa5 freshly built with Kernel Hacking turned off.
> 
That said, today same panic (as related above) on b2k  32bit  but not on
b180 with the same 2.6.11-rc4-pa1 (cvs dated 20050217).

By chance, I can save a top info at the crash moment:
top - 06:25:27 up 16:33,  3 users,  load average: 0.91, 0.32, 0.16
Tasks:  97 total,   2 running,  95 sleeping,   0 stopped,   0 zombie
Cpu(s):  4.9% us, 13.2% sy,  4.9% ni,  0.0% id, 75.7% wa,  0.7% hi,  0.7%=

si
Mem:    252696k total,   249768k used,     2928k free,    46300k buffers
Swap:   255928k total,        0k used,   255928k free,   133572k cached

  PID USER      PR  NI  VIRT  RES  SHR S %CPU %MEM    TIME+  COMMAND    
                                                          
 9304 nobody    26  10  1816  844  644 R 14.8  0.3   0:00.89 find       
                                                           
...

The time stamp seems to tell me to look into cron daily ...

I will try to locate this 'find' task (probably locate db update) and see=

if I can reproduce ;-)

Thanks,
    Joel


_______________________________________________
parisc-linux mailing list
parisc-linux@lists.parisc-linux.org
http://lists.parisc-linux.org/mailman/listinfo/parisc-linux