ORA-27063: number of bytes read/written is incorrect之二

In this Document

 Goal Fix

---

Applies to:

Oracle Server - Enterprise Edition - Version 9.2.0.8 to 11.2.0.3 [Release 9.2 to 11.2]
Oracle Solaris on SPARC (64-bit)
***Checked for relevance on 27-Apr-2012***

Goal

On database startup, the following error is reported;

SQL> startup

ORACLE instance started.

Total System Global Area  514328528 bytes

Fixed Size                   731088 bytes

Variable Size             218103808 bytes

Database Buffers          285212672 bytes

Redo Buffers               10280960 bytes

Database mounted.

ORA-01110: data file n: 'datafile.DBF'

ORA-01115: IO error reading block from file n (block # 1)

ORA-27063: skgfospo: number of bytes read/written is incorrect

Additional information: 8192

Additional information: 16384

DBVerify reports the following;

DBV-00102: File I/O error on FILE (datafile.DBF) during verification read operation (-2)

plus, at the same time the DBVerify was run, the /var/adm/messages file contains warnings, similar to;

Dec 27 14:28:19 node scsi: [ID 107833 kern.warning] WARNING: /pci@1c,600000/scsi@2/sd@1,0 (sd1):

Dec 27 14:28:19 node Error for Command: read(10)                Error Level: Retryable

Dec 27 14:28:19 node scsi: [ID 107833 kern.notice] Requested Block: 2910944                   Error Block: 2911043

Dec 27 14:28:19 node scsi: [ID 107833 kern.notice] Vendor: FUJITSU                            Serial Number: nnnn

Dec 27 14:28:19 node scsi: [ID 107833 kern.notice] Sense Key: Media Error

Dec 27 14:28:19 node scsi: [ID 107833 kern.notice] ASC: 0x11 (<vendor unique code 0x11>), ASCQ: 0x1, FRU: 0x0

 

What do these findings mean, and how do I overcome them?

Fix

These messages indicate a hardware problem with the disk, where the Media Error's reported in the messages log need to be referred to the hardware vendor.

To be able to open the database, the disk problems need to be resolved first, then the file will need to be restored from backup and/or recovered.

---