Click here to monitor SSC
SQLServerCentral is supported by Red Gate Software Ltd.
 
Log in  ::  Register  ::  Not logged in
 
 
 
        
Home       Members    Calendar    Who's On


Add to briefcase 12»»

Dbcc checkdb says 0 errors but SQL error 823 produced Expand / Collapse
Author
Message
Posted Thursday, April 19, 2012 1:29 PM
SSC-Enthusiastic

SSC-EnthusiasticSSC-EnthusiasticSSC-EnthusiasticSSC-EnthusiasticSSC-EnthusiasticSSC-EnthusiasticSSC-EnthusiasticSSC-Enthusiastic

Group: General Forum Members
Last Login: Thursday, August 7, 2014 7:15 AM
Points: 125, Visits: 408
The db is working as normal, but in the SQL logs I see these 3 messages

The operating system returned error incorrect checksum (expected: 0x1b0a0fbe; actual: 0x1b0a0fbe) to SQL Server during a read at offset 0x00000ae7e9c000 in file 'D:\Data\MyData.mdf'. Additional messages in the SQL Server error log and system event log may provide more detail. This is a severe system-level error condition that threatens database integrity and must be corrected immediately. Complete a full database consistency check (DBCC CHECKDB). This error can be caused by many factors; for more information, see SQL Server Books Online.

Error: 823, Severity: 24, State: 7.

Operating system error 'incorrect checksum (expected: 0x1b0a0fbe; actual: 0x1b0a0fbe)' resulted from attempt to read the following: sort run page (3:5717838), in file 'D:\Data\MuData_data.mdf', in database with ID 23. Sort is retrying the read.

I've run dbcc checkdb(MyData) but that said 0 errors found 0 errors repaired.

This is a sql 2005 instance and the DB page verify is set to checksum

Any ideas from a SQL viewpoint? whilst I also ask the infrastructure team to check the D: drive
Post #1286664
Posted Thursday, April 19, 2012 1:54 PM


SSC-Forever

SSC-ForeverSSC-ForeverSSC-ForeverSSC-ForeverSSC-ForeverSSC-ForeverSSC-ForeverSSC-ForeverSSC-ForeverSSC-ForeverSSC-ForeverSSC-ForeverSSC-ForeverSSC-ForeverSSC-Forever

Group: General Forum Members
Last Login: Today @ 1:53 PM
Points: 42,771, Visits: 35,870
It's a sort page, so temporarily allocated for a sort operation, probably an index rebuild, and deallocated afterwards. Hence by the time CheckDB ran, the page had long since been deallocated. Deallocated pages can't be checked with checkDB as they are not part of the consistent database structure.

That said, something's up with the IO subsystem if that could happen at all, check carefully.



Gail Shaw
Microsoft Certified Master: SQL Server 2008, MVP
SQL In The Wild: Discussions on DB performance with occasional diversions into recoverability

We walk in the dark places no others will enter
We stand on the bridge and no one may pass

Post #1286678
Posted Friday, April 20, 2012 3:00 AM
SSC-Enthusiastic

SSC-EnthusiasticSSC-EnthusiasticSSC-EnthusiasticSSC-EnthusiasticSSC-EnthusiasticSSC-EnthusiasticSSC-EnthusiasticSSC-Enthusiastic

Group: General Forum Members
Last Login: Thursday, August 7, 2014 7:15 AM
Points: 125, Visits: 408
thanks for the reply.
Yes the times this has been noted is during an maintenance task that rebuilds indexes and when new application objects are deployed/compiled.

So my understanding is that we have no immediate danager as dbcc checkdb says all ok, but a 'small' problem on the disk somewhere so does need to be investigate as quickly as possible.
Can you elaborate on 'check carefully', I mean is there anything from SQL I can do or just ask our hardware team to fully test the D: drive.

Thanks
Post #1286944
Posted Friday, April 20, 2012 3:41 AM


SSC-Forever

SSC-ForeverSSC-ForeverSSC-ForeverSSC-ForeverSSC-ForeverSSC-ForeverSSC-ForeverSSC-ForeverSSC-ForeverSSC-ForeverSSC-ForeverSSC-ForeverSSC-ForeverSSC-ForeverSSC-Forever

Group: General Forum Members
Last Login: Today @ 1:53 PM
Points: 42,771, Visits: 35,870
Not necessarily a small problem on the disk, you can't conclude that. Could be a huge problem that's only shown up once so far.

No, nothing really from SQL. Check windows logs, check raid logs, san logs, etc.



Gail Shaw
Microsoft Certified Master: SQL Server 2008, MVP
SQL In The Wild: Discussions on DB performance with occasional diversions into recoverability

We walk in the dark places no others will enter
We stand on the bridge and no one may pass

Post #1286962
Posted Tuesday, April 24, 2012 7:30 AM
SSC-Enthusiastic

SSC-EnthusiasticSSC-EnthusiasticSSC-EnthusiasticSSC-EnthusiasticSSC-EnthusiasticSSC-EnthusiasticSSC-EnthusiasticSSC-Enthusiastic

Group: General Forum Members
Last Login: Thursday, August 7, 2014 7:15 AM
Points: 125, Visits: 408
Hi Gail,
I have an update from my hardware guys.
The D: drive is very fragmented and chdkdsk failed, so their solution is to create a new LUN on clean disks format as X: then take SQL offline, copy the entire contents of D: (just MDf/NDF files really) to X: then drop D: rename X: to D: then bring SQL back online.

I wanted to use detach/attach as we have many Db's for different countries so downtime would be just for that country being moved rather than server down until all are copied, although my way would mean we cannot use D: drive

Thanks for youe help.
Post #1288911
Posted Tuesday, April 24, 2012 8:30 AM


SSC-Forever

SSC-ForeverSSC-ForeverSSC-ForeverSSC-ForeverSSC-ForeverSSC-ForeverSSC-ForeverSSC-ForeverSSC-ForeverSSC-ForeverSSC-ForeverSSC-ForeverSSC-ForeverSSC-ForeverSSC-Forever

Group: General Forum Members
Last Login: Today @ 1:53 PM
Points: 42,771, Visits: 35,870
Make sure you have backups of all of those databases first.


Gail Shaw
Microsoft Certified Master: SQL Server 2008, MVP
SQL In The Wild: Discussions on DB performance with occasional diversions into recoverability

We walk in the dark places no others will enter
We stand on the bridge and no one may pass

Post #1289014
Posted Tuesday, April 24, 2012 8:33 AM
SSC-Enthusiastic

SSC-EnthusiasticSSC-EnthusiasticSSC-EnthusiasticSSC-EnthusiasticSSC-EnthusiasticSSC-EnthusiasticSSC-EnthusiasticSSC-Enthusiastic

Group: General Forum Members
Last Login: Thursday, August 7, 2014 7:15 AM
Points: 125, Visits: 408
Hi,
Yes I am asking/insisting for this (as their is a fussy line as to responsilbilites due to geography and where its a SQL or infrastructure issue) the last reply I had from them was this

As we copy the data and do not write anything to the LUN in question, I see no need for (extra)backups – the risk of the old LUN to fail ultimately is not bigger as right now in this moment or any other.

He misses the whole point of a backup.

I shall insist though.
Post #1289019
Posted Tuesday, April 24, 2012 8:40 AM


SSC-Forever

SSC-ForeverSSC-ForeverSSC-ForeverSSC-ForeverSSC-ForeverSSC-ForeverSSC-ForeverSSC-ForeverSSC-ForeverSSC-ForeverSSC-ForeverSSC-ForeverSSC-ForeverSSC-ForeverSSC-Forever

Group: General Forum Members
Last Login: Today @ 1:53 PM
Points: 42,771, Visits: 35,870
Ask him if he wants to bet his job on there being no additional problems (because without backups, that's what you would be doing)


Gail Shaw
Microsoft Certified Master: SQL Server 2008, MVP
SQL In The Wild: Discussions on DB performance with occasional diversions into recoverability

We walk in the dark places no others will enter
We stand on the bridge and no one may pass

Post #1289022
Posted Saturday, May 19, 2012 11:00 AM


SSCommitted

SSCommittedSSCommittedSSCommittedSSCommittedSSCommittedSSCommittedSSCommittedSSCommitted

Group: General Forum Members
Last Login: Friday, August 15, 2014 5:01 PM
Points: 1,613, Visits: 1,539
GilaMonster (4/24/2012)
Ask him if he wants to bet his job on there being no additional problems (because without backups, that's what you would be doing)


+1000 Nobody has ever gotten fired for having backups and not needing them!

But then why are they even involved in the backup process. That should be the DBA's duty.




My blog: SQL Soldier
Twitter: @SQLSoldier
My book: Pro SQL Server 2008 Mirroring
Microsoft Certified Master: SQL Server 2008
Principal DBA: Outerwall, Inc.
Also available for consulting: SQL DBA Master
Post #1303043
Posted Wednesday, February 5, 2014 1:28 PM
Forum Newbie

Forum NewbieForum NewbieForum NewbieForum NewbieForum NewbieForum NewbieForum NewbieForum Newbie

Group: General Forum Members
Last Login: Monday, July 14, 2014 2:14 PM
Points: 4, Visits: 74
sotn ,
I'm curious if moving your data to a new LUN has solved your issue. I'm running into the same issue with checksums on my tempdb and like you my expected and actual checksums are matching in the output. I'm using a SAN so wondering if I just create a new drive partition from my datastore and move all my dbs over. Thanks!
Post #1538357
« Prev Topic | Next Topic »

Add to briefcase 12»»

Permissions Expand / Collapse