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.
| Author |
Topic |
|
AskSQLTeam
Ask SQLTeam Question
0 Posts |
 Posted - 2006-01-30 : 13:48:06
|
| Rob writes "We have had an on-going problem in our production database. It appears to all be index corruption. We've opened a PSS with Microsoft, but I figured I'd post here also. We're getting 604,605 errors as well as DBCC CHECKDB errors. All reports are that either REPAIR_REBUILD or REPAIR_FAST will fix the problem. Since putting the database into single_user mode is not an option at this time, we've just been rebuilding the problem indexes with DBCC DBREINDEX. This typically fixes the problem. Microsoft PSS says they believe that it's hardware problems. We're in the process of updating firmware on our storage device. (MSA 1000) This is a clustered machine. Tonight, our index rebuilds completed successfully, but a DBCC CHECKTABLE failed with 8951 and 8955 errors with the REPAIR_FAST option being supplied. I've rebuilt all the indexes on the table, but the DBCC CHECKTABLE still reports with errors. As a test, I DTS'd a copy of this table to make sure that I would at least be able to run a select * on it. That worked okay. My question is the following: Does the repair option for DBCC CHECKDB/TABLE do something that DBCC DBREINDEX doesn't? Thanks in advance! Rob P.S. This is a Siebel environment. :(" |
|
|
paulrandal
Yak with Vast SQL Skills
899 Posts |
Posted - 2006-02-02 : 13:26:55
|
| Can you post:- the actual CHECKDB output (using the WITH NO_INFOMSGS option please)- the exact version number you're using- the index rebuild statements you're using- the PSS case number (starts with SR) and the MS contactAnd I'll take a look. I'm skeptical of the h/w diagnosis. Are the errors always about the same key values?ThanksPaul RandalLead Program Manager, Microsoft SQL Server Storage Engine(Legalese: This posting is provided "AS IS" with no warranties, and confers no rights.) |
 |
|
|
fdewy
Starting Member
2 Posts |
Posted - 2009-09-09 : 10:41:06
|
| I have had similar issues with recurring index corruption with just one particualr table in our database. This same database structure exists on multiple servers, different sql server versions, etc. in each case it's always this same table. So i do not believe that this is a hardware issue. My solution has been to select data out to a temp table, drop and re-create the corrupted table, then insert the records back in. this fixes the corruption every time with no loss of data. I'm investigating how this data is entering the database (data uploaded from handheld devise via xml post). My theory is that there is a problem in transmission but I don't have proof.Below is the result from dbcc checkdb showing the corruption. If any one has any ideas please let me know. thanks.Msg 8929, Level 16, State 1, Line 1Object ID 1749541962, index ID 0, partition ID 72057594460504064, alloc unit ID 72057594477477888 (type In-row data): Errors found in off-row data with ID 91042152448 owned by data record identified by RID = (1:54952:1)Msg 8970, Level 16, State 1, Line 1Row error: Object ID 1749541962, index ID 0, partition ID 72057594460504064, alloc unit ID 72057594477477888 (type In-row data), page ID (1:54952), row ID 1. Column 'SigPts' was created NOT NULL, but is NULL in the row.Msg 8964, Level 16, State 1, Line 1Table error: Object ID 1749541962, index ID 0, partition ID 72057594460504064, alloc unit ID 72057594477543424 (type LOB data). The off-row data node at page (1:55118), slot 13, text ID 91042152448 is not referenced.Msg 8964, Level 16, State 1, Line 1Table error: Object ID 1749541962, index ID 0, partition ID 72057594460504064, alloc unit ID 72057594477543424 (type LOB data). The off-row data node at page (1:55248), slot 1, text ID 91042217984 is not referenced. |
|
|
|
GilaMonster
Master Smack Fu Yak Hacker
4507 Posts |
|
|
|
|
|
|
|
index corruption not getting fixed by DBREINDEX