[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: XFS/DISK crashing



Hello Walter Marchuk,

Once you wrote about "Re: XFS/DISK crashing":
> SCSI storage controller: Adaptec 7892P (rev 02)
> 
> I never had these crashing problems with ext2.
> 
> My question was if this error is xfs related.

No XFS specific prooblem.

Restart computer and enter to SCSI BIOS (CTRL+A (?)) on Adaptec SCSI
Controller.

Make "Verify" on scsi device. Verify process found error on your disk.
Make "remap bad block" (your lost 512byte on each 1 sector).

Or if your have backup, change disk.



> On Thu, 23 Aug 2001, Joshua Baker-LePain wrote:
> 
> > On Thu, 23 Aug 2001 at 11:49am, Walter Marchuk wrote
> > 
> > > I am running Linux 2.4.5 with xfs 1.0.1, the disk with XFS is a 150GB
> > > Raid.
> > 
> > A hardware RAID?  What type of controller?  What type of SCSI controller?
> > 
> > > Aug 22 09:46:06 marconi kernel: SCSI disk error : host 0 channel 0 id 2
> > > lun 0 return code = 8000002
> > > Aug 22 09:46:06 marconi kernel: Info fld=0x0, Current sd08:21: sns = f0  b
> > > Aug 22 09:46:06 marconi kernel: ASC=47 ASCQ= 0
> > > Aug 22 09:46:06 marconi kernel: Raw sense data:0xf0 0x00 0x0b 0x00 0x00
> > > 0x00 0x00 0x0a 0x00 0x00 0x00 0x00 0x47 0x00 0x00 0x00 0x00 0x00
> > > Aug 22 09:46:06 marconi kernel:  I/O error: dev 08:21, sector
> 123453963
> > 
> > These look like generic SCSI errors.  How sure are you of the hardware --
> > the RAID itself, the SCSI controller, etc?  Have you done the usual --
> > checked cables, termination, termination (yes, again ;), that you used the
> > proper color goat...
> > 
> > > Aug 22 09:46:06 marconi kernel: xfs_force_shutdown(sd(8,33),0x2) called
> > > from line 942 of file xfs_log.c.  Return address = 0xc01cf5d4
> > > Aug 22 09:46:06 marconi kernel: I/O Error Detected.  Shutting down
> > > filesystem: sd(8,33)
> > > Aug 22 09:46:06 marconi kernel: Please umount the filesystem, and rectify
> > > the problem(s)
> > 
> > This is XFS' standard response when it detects disk errors, in order to
> > prevent (further) corruption.
> > 
> > -- 
> > Joshua Baker-LePain
> > Department of Biomedical Engineering
> > Duke University
> > 
> 

-- 
bye.
Andrey Nekrasov, SpyLOG.