Ext3 users - Feb 2002 - errors

Hello,
I've been using ext3 on some of my production servers, I am now
having some serious issues.  I'm not sure where to pin point this
to.   

1.  I have success using this with a Mylex Raid controller and have
no problem.  setup as RAID 1.
2.  I have success using this with a standalone at all levels 
single ended SCSI, IDE Drives.  
3.  ""  This is where my problem starts  ""
i'm using a AMI MegaRaid on a JetStore with approximately 8 drives
that make on big fat 500Gigabyte drive. 
I have just upgraded this box 3 weeks ago to RedHat 7.2 and converted
the ext2 filesystem to ext3.  

These are my errors on the /var/log/messages.

Please tell me what you think about these.  Harddrive problem ?
etc..

Thanks.

--> snip
Feb  3 04:03:02 reallyreal syslogd 1.4.1: restart.
Feb  3 04:05:00 reallyreal kernel: scsi : aborting command due to timeout 
: pid 0, scsi2, channel 1, id 0, lu
n 0 Read (10) 00 19 72 fd 57 00 00 38 00
Feb  3 04:05:01 reallyreal kernel: scsi : aborting command due to timeout 
: pid 0, scsi2, channel 1, id 0, lu
n 0 Read (10) 00 0d 0d 38 77 00 00 20 00
Feb  3 04:05:01 reallyreal kernel: scsi : aborting command due to timeout 
: pid 0, scsi2, channel 1, id 0, lu
n 0 Read (10) 00 0d 0d 39 d7 00 00 68 00
Feb  3 04:05:02 reallyreal kernel: scsi : aborting command due to timeout 
: pid 0, scsi2, channel 1, id 0, lu
n 0 Read (10) 00 0d 1e 04 17 00 00 28 00
Feb  3 04:05:02 reallyreal kernel: scsi : aborting command due to timeout 
: pid 0, scsi2, channel 1, id 0, lu

Feb  3 04:05:28 reallyreal kernel: SCSI disk error : host 2 channel 1 id 0 
lun 0 return code = 25040001
Feb  3 04:05:28 reallyreal kernel:  I/O error: dev 08:11, sector 380222456
Feb  3 04:05:29 reallyreal kernel: SCSI disk error : host 2 channel 1 id 0 
lun 0 return code = 25040001
Feb  3 04:05:29 reallyreal kernel:  I/O error: dev 08:11, sector 380222456
Feb  3 04:05:29 reallyreal kernel: SCSI disk error : host 2 channel 1 id 0 
lun 0 return code = 25040001
Feb  3 04:05:29 reallyreal kernel:  I/O error: dev 08:11, sector 379675384
Feb  3 04:05:29 reallyreal kernel: SCSI disk error : host 2 channel 1 id 0 
lun 0 return code = 25040001
Feb  3 04:05:29 reallyreal kernel:  I/O error: dev 08:11, sector 379675384
Feb  3 04:05:29 reallyreal kernel: SCSI disk error : host 2 channel 1 id 0 
lun 0 return code = 25040001
Feb  3 04:05:29 reallyreal kernel:  I/O error: dev 08:11, sector 265940064
Feb  3 04:05:29 reallyreal kernel: SCSI disk error : host 2 channel 1 id 0 
lun 0 return code = 25040001
Feb  3 04:05:29 reallyreal kernel:  I/O error: dev 08:11, sector 265940064
Feb  3 04:05:29 reallyreal kernel: SCSI disk error : host 2 channel 1 id 0 
lun 0 return code = 25040001
Feb  3 04:05:29 reallyreal kernel:  I/O error: dev 08:11, sector 265940064
Feb  3 04:05:29 reallyreal kernel: SCSI disk error : host 2 channel 1 id 0 
lun 0 return code = 25040001
Feb  3 04:05:29 reallyreal kernel:  I/O error: dev 08:11, sector 379693264

Hi,

On Mon, Feb 04, 2002 at 11:15:38AM -0800, Steve Lee wrote:
> I've been using ext3 on some of my production servers, I am now
> having some serious issues.  I'm not sure where to pin point this
> to.   
> 
> 1.  I have success using this with a Mylex Raid controller and have
> no problem.  setup as RAID 1.
> 2.  I have success using this with a standalone at all levels 
> single ended SCSI, IDE Drives.  
> 3.  ""  This is where my problem starts  ""
> i'm using a AMI MegaRaid on a JetStore with approximately 8 drives
> that make on big fat 500Gigabyte drive. 
> I have just upgraded this box 3 weeks ago to RedHat 7.2 and converted
> the ext2 filesystem to ext3.  
Well, it's clearly a scsi-layer problem, not a filesystem problem.
> Feb  3 04:05:00 reallyreal kernel: scsi : aborting command due to timeout 
> : pid 0, scsi2, channel 1, id 0, lu
> n 0 Read (10) 00 19 72 fd 57 00 00 38 00
What version of firmware is in the megaraid?  147 was known to be bad.
161 is recommended.

Cheers,
 Stephen