Please start any new threads on our new site at https://forums.sqlteam.com. We've got lots of great SQL Server experts to answer whatever question you can come up with.

 All Forums
 SQL Server 2000 Forums
 SQL Server Administration (2000)
 I/O Error???

Author  Topic 

nbkr3bi
Starting Member

15 Posts

Posted - 2008-01-21 : 05:28:26
We have a database maintenance job that is scheduled to run everyday.
It backs up all the databases on the server. Additionally, Performance maintenence happens on saturdays in the same job.
The job has been failing since the last couple of days on Saturdays.
On further analysis, we found the following errors in the Sql server log. It looks to be a hardware I/O(hard disk) issue but we cant figure out why the failure is not happening on other days(besides saturday).
Could someone help us interpret the log?

2008-01-20 01:27:30.05 server SQL Server terminating because of system shutdown.

2008-01-20 01:27:31.48 spid53 BackupMedium::ReportIoError: write failure on backup device 'VDI_BECDF8AB-F059-45B4-8323-2A9F4DB23D62_0'. Operating system error 6(The handle is invalid.).

2008-01-20 01:27:31.51 spid53 LogEvent: Failed to report the current event. Operating system error = 1717(The interface is unknown.).

2008-01-20 01:27:31.51 backup BACKUP failed to complete the command BACKUP database [AAA] TO VIRTUAL_DEVICE=....

2008-01-20 01:27:31.51 spid53 Internal I/O request 0x1A757740: Op: Write, pBuffer:
0x0F020000, Size: 1048576,...

2008-01-20 01:27:31.51 spid53 BackupMedium::ReportIoError: write failure on backup device 'VDI_BECDF8AB-F059-45B4-8323-2A9F4DB23D62_3'. Operating system error 6(The handle is invalid.).

2008-01-20 01:27:31.52 spid53 Internal I/O request 0x4C8C3D08: Op: Write, pBuffer: 0x0EC20000, Size: 327680, ...

2008-01-20 01:27:33.10 spid53 Internal I/O request 0x4C1F2050: Op: Write, pBuffer: 0x0F520000, Size: 1048576,...

2008-01-20 01:27:33.10 spid53 BackupMedium::ReportIoError: write failure on backup device 'VDI_BECDF8AB-F059-...

2008-01-20 01:27:33.15 spid53 Internal I/O request 0x4C17BB60: Op: Write, pBuffer: 0x0E520000, Size: 1048576,...

2008-01-20 01:27:33.15 spid53 BackupMedium::ReportIoError: write failure on backup device 'VDI_BECDF8AB-F059-...

2008-01-20 01:27:33.57 spid53 BackupVirtualDeviceFile::RequestDurableMedia: Flush failure on backup device 'V...

2008-01-20 01:27:33.60 spid53 BackupVirtualDeviceFile::RequestDurableMedia: Flush failure on backup device 'V...

2008-01-20 01:27:33.63 spid53 BackupVirtualDeviceFile::RequestDurableMedia: Flush failure on backup device 'V...

2008-01-20 01:27:33.66 spid53 BackupVirtualDeviceFile::RequestDurableMedia: Flush failure on backup device 'V...


Thank you.

Haywood
Posting Yak Master

221 Posts

Posted - 2008-01-21 : 09:59:39
Check your backup solution, it looks like the problem lies there and not with the SQL server. See the error indicated by the failure of the backup to the VIRTUAL_DEVICE? That's either a tape backup or a SAN backup (and it doesn't look like a SAN snap/mirror/clone) which means the write is occurring on a remote storage system and not the local SQL server.

Can you post the rest of the error (the full text of it) that follows the VIRTUAL_DEVICE error?
Go to Top of Page

rmiao
Master Smack Fu Yak Hacker

7266 Posts

Posted - 2008-01-21 : 16:17:00
And check windows event logs for disk related errors.
Go to Top of Page

nbkr3bi
Starting Member

15 Posts

Posted - 2008-01-31 : 06:45:31
As it turns out, there were some consistency and reference errors in one of the tables which was causing the DBCC CHECKDB to fail and thereby failing the job too. Running

DBCC CHECKTABLE with ,REPAIR_ALLOW_DATA_LOSS fixed the problem.

Thanks to everyone who helped.
Go to Top of Page

eyechart
Master Smack Fu Yak Hacker

3575 Posts

Posted - 2008-01-31 : 12:38:13
quote:
Originally posted by nbkr3bi

As it turns out, there were some consistency and reference errors in one of the tables which was causing the DBCC CHECKDB to fail and thereby failing the job too. Running

DBCC CHECKTABLE with ,REPAIR_ALLOW_DATA_LOSS fixed the problem.

Thanks to everyone who helped.





that means you lost data btw. A better solution would be to restore from your last full backup that you have from before the corruption event and apply all transaction logs so you don't lose data.



-ec
Go to Top of Page
   

- Advertisement -