Mailinglist Archive: opensuse (4570 mails)

[JvdW <r2d2@xxxxxxxxxx>]

Hi all

I have a box(Dell) running SLES9 that started reporting the following errors:

--cut--

Nov  7 08:08:59 kernel: SCSI error : <0 0 0 0> return code = 0x20000
Nov  7 08:08:59 kernel: end_request: I/O error, dev sda, sector 273350823
Nov  7 08:08:59 kernel: Buffer I/O error on device sda1, logical block 
34168845
Nov  7 08:08:59 kernel: lost page write due to I/O error on sda1
Nov  7 08:09:21 kernel: SCSI error : <0 0 0 0> return code = 0x20000
Nov  7 08:09:21 kernel: end_request: I/O error, dev sda, sector 273350831
Nov  7 08:09:21 kernel: Buffer I/O error on device sda1, logical block 
34168846
Nov  7 08:09:21 kernel: lost page write due to I/O error on sda1
Nov  7 08:09:43 kernel: SCSI error : <0 0 0 0> return code = 0x20000
Nov  7 08:09:43 kernel: end_request: I/O error, dev sda, sector 273350839
Nov  7 08:09:43 kernel: Buffer I/O error on device sda1, logical block 
34168847
Nov  7 08:09:43 kernel: lost page write due to I/O error on sda1
Nov  7 08:10:05 kernel: SCSI error : <0 0 0 0> return code = 0x20000
Nov  7 08:10:05 kernel: end_request: I/O error, dev sda, sector 273350847
Nov  7 08:10:05 tappsdev kernel: Buffer I/O error on device sda1, logical 
block 34168848
Nov  7 08:10:05 kernel: lost page write due to I/O error on sda1
Nov  7 08:10:27 kernel: SCSI error : <0 0 0 0> return code = 0x20000
Nov  7 08:10:27 kernel: end_request: I/O error, dev sda, sector 273350855
Nov  7 08:10:27 kernel: Buffer I/O error on device sda1, logical block 
34168849
Nov  7 08:10:27 kernel: lost page write due to I/O error on sda1
Nov  7 08:10:49 kernel: SCSI error : <0 0 0 0> return code = 0x20000
Nov  7 08:10:49 kernel: end_request: I/O error, dev sda, sector 273350863
Nov  7 08:10:49 kernel: Buffer I/O error on device sda1, logical block 
34168850
Nov  7 08:10:49 kernel: lost page write due to I/O error on sda1

--cut--

The oracle alert file reported "IO error writing block".

A dmesg reports the following:

--cut--
JBD: barrier-based sync failed on sda1 - disabling barriers
--cut--

The box runs a Ext3 FS and the mountpoint on sda1 is on a remote CX storage 
connected with fiber and a QLogic fiber card.

A df -k for the partition in question:

/dev/sda1            347993768 133129208 197187440  41% /oracle

Can anyone tell me what this error means? Does it simply mean that the machine 
lost connectivity with the CX or is the problem more serious and possibly 
corrupt fs related?

Any help would be greatly appreciated.

Regards
JvdW