February 2, 2004 at 1:26 pm
Help! I need ideas for a quick resolution! Here are the details.
Nightly, I run a job that calls an sp.
this sp updates a particular table several times.
Not always, but maybe 3 out of 5 nights, the job fails with the following error:
EventLog Error: 1203, Severity: 20, State: 1
Process ID 52 attempting to unlock unowned resource PAG: 9:1:11623.
Profiler indicates failure on the following steps:
1. ) UPDATE
tblHMCReports
SET
ResidenceType = PopupDesc
FROM
LCDB.dbo.Popups
WHERE
ResidenceType = PopupValue
AND PopupId = -20
2.)
Execution Plan Execution Tree
--------------
Clustered Index Update(OBJECT[LCDBReports].[dbo].[tblHMCReports].[PK_LoanNum]), SET[tblHMCReports].[ResidenceType]=[Expr1006]))
|--Compute Scalar(DEFINE[Expr1006]=Convert([PopUps].[PopupDesc])))
|--Top(ROWCOUNT est 0)
|--Stream Aggregate(GROUP BY[Bmk1000]) DEFINE[PopUps].[PopupDesc]=ANY([PopUps].[PopupDesc])))
|--Nested Loops(Inner Join, WHERE[PopUps].[PopupValue]=[tblHMCReports].[ResidenceType]))
|--Clustered Index Scan(OBJECT[LCDBReports].[dbo].[tblHMCReports].[PK_LoanNum]))
|--Table Spool
|--Clustered Index Seek(OBJECT[LCDB].[dbo].[PopUps].[PK_PopUps]), SEEK[PopUps].[PopupID]=-20) ORDERED FORWARD)
9 PSQL1 2724 SQLAgent - TSQL JobStep (Job 0x95425A7A6958F0468C3551640E9F9237 : Step 1) sa 52 2004-02-01 02:32:45.153 0X01 PSQL1
3. ) Exception Error: 1203, Severity: 20, State: 1 9 PSQL1 2724 SQLAgent - TSQL JobStep (Job 0x95425A7A6958F0468C3551640E9F9237 : Step 1) sa 52 2004-02-01 02:32:45.840 20 1203 0X01 PSQL1
Any ideas on causes and or resolutions? thanks in advance
February 2, 2004 at 1:34 pm
From BOL.
"
This error occurs when Microsoft® SQL Server™ is engaged in some activity other than normal post-processing cleanup and it finds that a particular page it is attempting to unlock is already unlocked. The underlying cause for this error may be related to structural problems within the affected database. SQL Server manages the acquisition and release of pages to maintain concurrency control in the multi-user environment. This mechanism is maintained through the use of various internal lock structures that identify the page and the type of lock present. Locks are acquired for processing of affected pages and released when the processing is completed.
Execute DBCC CHECKDB against the database in which the object belongs. If DBCC CHECKDB reports no errors, attempt to reestablish the connection and execute the command.
Important If executing DBCC CHECKDB with one of the repair clauses does not correct the index problem, or if you are unsure what effect DBCC CHECKDB with a repair clause has on your data, contact your primary support provider."
Viewing 2 posts - 1 through 1 (of 1 total)
You must be logged in to reply to this topic. Login to reply