4.10-stable self eating

[Harald Schmalzbauer]

Am Samstag, 10. Juli 2004 03:53 schrieb Doug White:
> On Sat, 10 Jul 2004, Harald Schmalzbauer wrote:
> > Dear all,
> >
> > today my 4.10-stable box practillay died with lightning HDD-LED. I could
> > "ping" the box with occasional 5000ms replys, but no ssh-login nor
> > acpi-power-button shutdown was possible.
> > Pressing the num-lock button on the keyboard resulted in a 30sec lag
> > between pressing the button and changing the LED.
> > I'm running 4.10-stable from 4th July with HighPoint kernel module for
> > the HPT372 (declared for 4.9 by HighPoint), for more infos find links at
> > the bottom.
> >
> > Here is what I saw on the console:
> >
> > spec_getpages:(#da/0x20000) I/O read failure: (error=5) bp 0xcbb0 vp
> > 0xcd3bdec0
>
> looks like your disk went to lunch. You might check connections, then
> consider how much you like the data on that disk since it might be dying.

Well, I love that data, it's my IMAP server.
But if you're convinced that this failure is because of bad disk then the 
HighPoint module must be bad.
da0 is a HPT372 RAID1 array. I extensively simulated disk failures and array 
rebuilds before taking this box productive (I just suffered badly from the 
last ATAng commit from Søren to -current which prevented the system from 
detecting any raid configuration (and is still unrepl(y|ie)d btw.) with my 
old sil0680 controller).

Let's hope HighPoint will continue FreeBSD support!

Thanks for the hint,

-Harry
-------------- next part --------------
A non-text attachment was scrubbed...
Name: not available
Type: application/pgp-signature
Size: 187 bytes
Desc: signature
Url : http://lists.freebsd.org/pipermail/freebsd-stable/attachments/20040710/0c66f9ee/attachment.bin