DB Maintenance Corrupts Data

Hi all,

I'm new to the forum, and fairly new to SQL, so please forgive me if I say something that sounds stupid at any time. I have been looking around this forum (and others) while trying to diagnose my current issues, and there seems to be a really good amount of knowledge about here so I thought I'd try my luck with a post.

We've been running SQL 2000 for around 2 years now, and had very few problems until we suffered a torn page error around 6 months ago (just before the time I started looking after the server).

Since then we've been getting intermittent database corruption (I implemented a nightly DBCC job as part of DB maintenance) and it seems to occur after our weekly index maintenance has run (also a scheduled DBMaint job). I have just confirmed this by restoring nightly backups to a test database - the current round of corruption has occured sometime between the Saturday and Sunday night backups (indexing runs at 1 am Sunday morning, and is the only activity between these two backups)

We have 9 small (all less than 200MB) databases which are used faily lightly by our salesforce. The corruption is occurring on different databases each time.

We are running SQL 2000 (SP4) on a Windows 2003 clustered server (2 x DL380s and an MSA500 G1) running in failover mode (active-passive). The raid 5 array does utilise disk caching, and I have wondered if this is involved somehow.

Our hardware support comany have checked the server thoroughly for HW issues, but have come up blank.


Here is the DBCC output from one of our corrupt databases from this morning:

dbcc checkdb (casestatus) WITH ALL_ERRORMSGS, NO_INFOMSGS
-----------------------
Server: Msg 2533, Level 16, State 1, Line 1
Table error: Page (1:20040) allocated to object ID 117575457, index ID 0 was not seen. Page may be invalid or have incorrect object ID information in its header.
Server: Msg 8976, Level 16, State 1, Line 1
Table error: Object ID 117575457, index ID 1. Page (1:20040) was not seen in the scan although its parent (1:19808) and previous (1:20039) refer to it. Check any previous errors.
Server: Msg 8978, Level 16, State 1, Line 1
Table error: Object ID 117575457, index ID 1. Page (1:20041) is missing a reference from previous page (1:20040). Possible chain linkage problem.
Server: Msg 8964, Level 16, State 1, Line 1
Table error: Object ID 117575457. The text, ntext, or image node at page (1:12449), slot 1, text ID 1190656606208 is not referenced.
Server: Msg 8964, Level 16, State 1, Line 1
Table error: Object ID 117575457. The text, ntext, or image node at page (1:12449), slot 3, text ID 1190656671744 is not referenced.
Server: Msg 8964, Level 16, State 1, Line 1
Table error: Object ID 117575457. The text, ntext, or image node at page (1:12449), slot 5, text ID 1190656737280 is not referenced.
Server: Msg 8964, Level 16, State 1, Line 1
Table error: Object ID 117575457. The text, ntext, or image node at page (1:12449), slot 7, text ID 1190656802816 is not referenced.
Server: Msg 8964, Level 16, State 1, Line 1
Table error: Object ID 117575457. The text, ntext, or image node at page (1:12449), slot 9, text ID 1190656868352 is not referenced.
Server: Msg 8964, Level 16, State 1, Line 1
Table error: Object ID 117575457. The text, ntext, or image node at page (1:12449), slot 11, text ID 1190656933888 is not referenced.
Server: Msg 8964, Level 16, State 1, Line 1
Table error: Object ID 117575457. The text, ntext, or image node at page (1:12449), slot 13, text ID 1190656999424 is not referenced.
Server: Msg 8964, Level 16, State 1, Line 1
Table error: Object ID 117575457. The text, ntext, or image node at page (1:12449), slot 15, text ID 1190657064960 is not referenced.
CHECKDB found 0 allocation errors and 11 consistency errors in table 'Raw:LifeStatusFP' (object ID 117575457).
CHECKDB found 0 allocation errors and 11 consistency errors in database 'CaseStatus'.
repair_allow_data_loss is the minimum repair level for the errors found by DBCC CHECKDB (CaseStatus ).
-----------------------

The funny thing is that this particular database isn't even in use any more - I've had it in single user mode for about 2 months (and the users don't have any access to the back end stuff).

If anyone can shed any light on what's happening, I'd really appreciate this.

Thanks

Dave