DBCC SHRINKDATABASE fails on TempDB with Error 8909

[Gus]

We have a rather large TempDB (11+ GB). I have a nightly job that runs DBCC SHRINKDATABASE (TempDB) to keep it from taking over the Raid Array. Last night, the job failed, repeating the following error message many times:


[SQLSTATE 42000] (Error 8909) Table Corrupt: Object ID 0, index ID 0, page ID (4:1273382). The PageId in the page header = (0:0).


I tried to run it in Query Analyzer and got this message:


Server: Msg 8909, Level 16, State 1, Line 0 Table Corrupt: Object ID 0, index ID 0, page ID (5:245247). The PageId in the page header = (0:0).


After I run it in QA, however, it proceeds to display the normal result set and shrink the DB, which the scheduled job did not do.


I searched the Microsoft KB and only found one entry for this error number (8909) referring to running CHECKDB concurrently with AuthShrink or SHRINKDATABASE, but that doesn't apply here. There are no references to this error number in either SP2 (which we are currently running) or SP3 (which we are testing on another server). Tonight I am going to restart the SQL Services in order to re-create the TempDB and see if that works. Does anyone have any insight into this? Thanks.


Ciao,
Gus

stop and start sql server. that's the best way to keep temp from getting too big. if you can't do this often, then first check your code that makes temp so big, that's the real issue


------------
Gus at 6/14/01 8:56:31 AM

We have a rather large TempDB (11+ GB). I have a nightly job that runs DBCC SHRINKDATABASE (TempDB) to keep it from taking over the Raid Array. Last night, the job failed, repeating the following error message many times:


[SQLSTATE 42000] (Error 8909) Table Corrupt: Object ID 0, index ID 0, page ID (4:1273382). The PageId in the page header = (0:0).


I tried to run it in Query Analyzer and got this message:


Server: Msg 8909, Level 16, State 1, Line 0 Table Corrupt: Object ID 0, index ID 0, page ID (5:245247). The PageId in the page header = (0:0).


After I run it in QA, however, it proceeds to display the normal result set and shrink the DB, which the scheduled job did not do.


I searched the Microsoft KB and only found one entry for this error number (8909) referring to running CHECKDB concurrently with AuthShrink or SHRINKDATABASE, but that doesn't apply here. There are no references to this error number in either SP2 (which we are currently running) or SP3 (which we are testing on another server). Tonight I am going to restart the SQL Services in order to re-create the TempDB and see if that works. Does anyone have any insight into this? Thanks.


Ciao,
Gus