June 19, 2003 at 2:47 am
Yesterday evening the error log on my production server included the following lines:
2003-06-18 21:15:39.60 spid95 Using 'dbghelp.dll' version '4.0.5'
*Dump thread - spid = 95, PSS = 0x4bc7d218, EC = 0x32da0098
*Stack Dump being sent to F:\Program Files\Microsoft SQL Server\MSSQL$ISCO4SQL\log\SQLDump0003.txt
* ***********
2003-06-18 21:15:45.49 spid95 Stack Signature for the dump is 0xE6954E0C
2003-06-18 21:15:45.50 spid95 SQL Server Assertion: File: <recbase.cpp>, line=1378 Failed Assertion = 'm_offBeginVar < m_SizeRec'.
2003-06-18 21:15:45.50 spid95 Error: 3624, Severity: 20, State: 1.
I can't find anything in BOL about error 3624. Can anyone help?
As it happens, I had a profiler trace running because I have been experiencing chain linkage mismatch errors in a pair of tables and have been trying to tie down the cause. The spid 95 referenced in the above error extract was accessing the tables involved in those errors at the time, but was only performing reads, not writes.
I would really appreciate some advice here, if anyone would care to make any suggestions.
June 19, 2003 at 4:29 am
It may be a clustered index corruption. Have a look at MS Knowledge Base Article - 317852
Cheers,
- Mark
June 21, 2003 at 1:09 pm
From Microsoft.
SYMPTOMS
A clustered index may be corrupted if all of the following conditions are met:
You try to add a column to an existing table and in the same transaction you add a foreign key reference from the new column to a column that is a unique clustered or unique nonclustered index key on the same table.
The add column fails and the transaction is rolled back.
After the transaction fails and is rolled back, an INSERT statement to the same table may result in an error message similar to one of the following:
Server: Msg 3624, Level 20, State 1, Line 1
Location: recbase.cpp:1378
Expression: m_offBeginVar < m_SizeRec
SPID: 51
Process ID: 948
Connection Broken
-or-
[Microsoft][ODBC SQL Server Driver][DBNETLIB]ConnectionCheckForData (CheckforData()).
Server: Msg 11, Level 16, State 1, Line 0
General network error. Check your network documentation.
Connection Broken
Additionally, if you try to reindex the table by using a DBCC DBREINDEX statement, the following error message occurs:
Server: Msg 2727, Level 16, State 48, Line 1
Cannot find index ''.
CAUSE
The clustered index of the table is corrupted after the failed add column attempt. DBCC CHECKDB does not detect or report this particular scenario.
RESOLUTION
To resolve this problem, obtain the latest service pack for Microsoft SQL Server 2000. For additional information, click the following article number to view the article in the Microsoft Knowledge Base:
290211 INF: How To Obtain the Latest SQL Server 2000 Service Pack
WORKAROUND
To work around this problem you can either:
Drop, and then re-create the index.
-or-
Use this statement to re-create the corrupted index: CREATE CLUSTERED INDEX...WITH DROP_EXISTING
STATUS
June 21, 2003 at 8:32 pm
I think the error is undocumented. You may want to send the erorlog together with the stack dump file to Microsoft.
June 24, 2003 at 11:54 am
About a month ago I started receiving 3624 errors. I finally discovered that I had a bad sectors on my hard drive by running chkdsk. After replacing the drive and re-installing, the errors are gone.
Bill
July 14, 2023 at 4:37 am
This reply has been reported for inappropriate content.
DBCC CHECKDB and the associated SQL queries are a general issue solver. If that's too technical, try restoring a previous version of the Server. Make sure it's the one before you get a system assertion check has failed alert. Also, update to the latest SQL patch released by Microsoft. You can get a better understanding with this tutorial to resolve SQL database error 3624 here.
Viewing 6 posts - 1 through 5 (of 5 total)
You must be logged in to reply to this topic. Login to reply