January 25, 2016 at 3:22 am
Hi
I have a Cluster server with 2 nodes running a SQL Server 2008 R2 Enterptrise edition failover cluster.
The instance has 102 user databases and they are all stored on Dell compellent storage, each database has the same 800 tables.
In December one of the databases reported corruption on one of the tables. In order to fix the issue I ran a SELECT * INTO to copy the whole table, then I ran DBCC CHECKTABLE with repair_allow_data_loss, I then inserted the missing entries and everything seemed to be OK.
Last week the same table reported the same kind of corruption, there were less errors and it removed fewer entries that I had to reinsert.
The same table is reporting corruption again, I have been running the checks every 24 hours for that database and I have narrowed it down to one 24 hour period, but I cannot see what has caused the corruption. When I have posted about this before it was suggested to look into the disks but we even had Dell check them over and they see no issues.
Can someone please help? Should I drop the table and recreate it?
What else could be the issue?
The error is below.
DBCC results for 'Table'.
Msg 8961, Level 16, State 1, Line 1
Table error: Object ID 1355151873, index ID 1, partition ID 72057640837644288, alloc unit ID 72057594106478592 (type LOB data). The off-row data node at page (1:3534404), slot 0, text ID 389905679056896 does not match its reference from page (1:3985465), slot 27.
Msg 8961, Level 16, State 1, Line 1
Table error: Object ID 1355151873, index ID 1, partition ID 72057640837644288, alloc unit ID 72057594106478592 (type LOB data). The off-row data node at page (1:5814100), slot 0, text ID 389905679056896 does not match its reference from page (1:3985465), slot 27.
Msg 8929, Level 16, State 1, Line 1
Object ID 1355151873, index ID 1, partition ID 72057640837644288, alloc unit ID 72057661876666368 (type In-row data): Errors found in off-row data with ID 108430702346240 owned by data record identified by RID = (1:3985465:27)
There are 580597 rows in 7361 pages for object "Table".
CHECKDB found 0 allocation errors and 3 consistency errors in table 'Table' (object ID 1355151873).
Thanks
Sam
January 25, 2016 at 3:58 am
Corruption, especially repeated corruption, is very likely an IO subsystem problem. It's not necessarily the disks, it could be anything from SQL all the way down to the disks, switch, controller, HBAs, cache, drivers, firmware, etc, etc, etc.
SQL does not corrupt its own database.
Gail Shaw
Microsoft Certified Master: SQL Server, MVP, M.Sc (Comp Sci)
SQL In The Wild: Discussions on DB performance with occasional diversions into recoverability
February 28, 2016 at 6:11 am
I am moving my storage for SQL 2012 from Equallogic SAN to a Compellent SAN, and I have noticed more than one case of DB corruption in the last month. I can't see any errors on the SAN, so I am looking for tools to test deeper the SAN to find the problem.
Viewing 3 posts - 1 through 2 (of 2 total)
You must be logged in to reply to this topic. Login to reply