April 4, 2008 at 9:55 pm
getting the following errors on one of my databases.. MS SQL Server 2000 ENT edition (SP4)
Trying to figure out how to handle this.. can anyone help????
I/O error (bad page ID) detected during read at offset 0x000000cee08000 in file and Error: 823, Severity: 24, State: 2
DBCC CHECKDB, DBCC CHECKCATALOG and DBCC CHECKALLOC
CHECKALLOC found 0 allocation errors and 0 consistency errors in database
CHECKDB found 0 allocation errors and 5492 consistency errors in database 'Orion'. (see attached)
April 5, 2008 at 3:41 am
Have you checked your hardware because your database has quite some errors.
Other suggestions:
http://www.sqlservercentral.com/Forums/Topic477115-5-1.aspx#bm477258
April 5, 2008 at 9:43 am
hardware guys fixed the hardware problem. I still have databases kicking errors on a dbcc checkdb inconsistency errors.
is a reboot in order or should I try and repair the datasbases??? what's the best way to repair???
April 5, 2008 at 11:20 am
A reboot isn't going to fix anything - these are corruptions in the database files on disk. The only way for you to repair without losing data is to restore from your backups. Do you have backups? If not you're going to have to run repair through DBCC CHECKDB which *will* cause you to lose data, and depending on the actual errors, may not be able to repair absolutely everything.
If you don't have any backups, my advice is to take a full backup of the database before running repair.
Thanks
Paul Randal
CEO, SQLskills.com: Check out SQLskills online training!
Blog:www.SQLskills.com/blogs/paul Twitter: @PaulRandal
SQL MVP, Microsoft RD, Contributing Editor of TechNet Magazine
Author of DBCC CHECKDB/repair (and other Storage Engine) code of SQL Server 2005
April 5, 2008 at 11:52 am
We do have backups but our servers guys control them.. they are to tape... tracking them down now.. best to drop the DBs before restore??? Do I need to go to a backup from BEFORE the errors?
what is the syntax for the repair with dataloss of I want to try that first?????
April 5, 2008 at 2:40 pm
Don't drop the databases - the restore may fail and then you're left with nothing. Restore to a different database name in case there are problems with the backups.
Yes, of course you need to go back to a backup from before the corruptions - otherwise you'll be restoring a corrupt database.
Don't try repair first as that is guaranteed to lose data. You may be able to have zero data loss if you have full and log backups up to the point of failure.
No offense but I seriously suggest you call Product Support to help you through this. It sounds like you've never done anything like this before and could use some live help.
Thanks
Paul Randal
CEO, SQLskills.com: Check out SQLskills online training!
Blog:www.SQLskills.com/blogs/paul Twitter: @PaulRandal
SQL MVP, Microsoft RD, Contributing Editor of TechNet Magazine
Author of DBCC CHECKDB/repair (and other Storage Engine) code of SQL Server 2005
April 8, 2008 at 8:38 am
ONce they FINALLY fixed the SAN problems, I ran DBCC DBCHECK on the databases again and it repaired some but not all of the consistency errors. At that point, had the databases restored to the closest point in time backup BEFORE the crash. thanks for your help and support
April 9, 2008 at 3:27 pm
This error may be hardware related, but not always true. See the link below.
April 9, 2008 at 4:45 pm
823s are always caused by a problem in the IO subsystem - something underneath SQL Server. I've never seen a case of a SQL Server 2000 corruption causing an 823. The KB article title is worded badly in that it could imply that the issue is in SQL Server.
Thanks
Paul Randal
CEO, SQLskills.com: Check out SQLskills online training!
Blog:www.SQLskills.com/blogs/paul Twitter: @PaulRandal
SQL MVP, Microsoft RD, Contributing Editor of TechNet Magazine
Author of DBCC CHECKDB/repair (and other Storage Engine) code of SQL Server 2005
April 10, 2008 at 5:36 am
Thank you everyone!! This was my first time dealing with corrupted databases caused by hardware problems and your support was MUCH appreciated~!!!!
Ended up restoreing two databases on one Instance and totally rebuilding another because some of the the binaries got hosed and the Instance would not start!!!! reinstalling and then attaching the databases did the trick. thanks again!
Viewing 10 posts - 1 through 9 (of 9 total)
You must be logged in to reply to this topic. Login to reply