assertion failed

Hello,

Noticed below error in application logs,

SQL Server Assertion: File: <"logk.cpp">, line=1235 Failed Assertion = 'lfh->lfh_startOffset == lfcb->lfcb_startOffset'. This error may be timing-related. If the error persists after rerunning the statement, use DBCC CHECKDB to check the database for structural integrity, or restart the server to ensure in-memory data structures are not

can you please advise the reason and remedy for the same error.

asked Feb 18 '11 at 01:13 AM in Default

pits
830 ● 47 ● 78 ● 88

(comments are locked)

3 answers:

Follow the error messages suggestion and run DBCC CHECKDB on the database.

answered Feb 18 '11 at 01:19 AM

Kev Riley ♦♦
46.1k ● 39 ● 43 ● 69

ok,but what is the issue and how the dbcc checkdb will help?

Feb 18 '11 at 01:26 AM pits

The issue is something within the internals of SQL - it has detected an issue using assertions - an assertion is a check or presumption that SQL relies on, if the assertion fails, the engine raises an error. DBCC CHECKDB will run various integrity checks on your database - the output from that will determine what, if anything, you need to do.

Feb 18 '11 at 01:31 AM Kev Riley ♦♦

(comments are locked)

Since it could be "... timing-related ...", it might be caused by an external event like an IO burst. In that case, it's not structural. But I do recommend a DBCC CHECKDB action (actually you should run this on a regularly base)

answered Feb 18 '11 at 02:34 AM

Wilfred van Dijk
549 ● 13 ● 18 ● 20

+1 - for sure on all points. DBA == paranoid

Feb 18 '11 at 02:48 AM Fatherjack ♦♦

(comments are locked)

aside from the cause/remedy I would recommend:
1 - preserving all your existing backups - dont overwrite or delete any until this is all wrapped up and you have a healthy database/server.
2 - making a backup of all your databases (see note 1, dont overwrite any existing backups)
3 - READING about CHECKDB in BOL, understand its function, options and effects.
4 - run CHECKDB
5 - celebrate that all databases are fine, back them all up again as per normal backup procedures

It is likely to be nothing but if you have encountered bad pages on a disk you need to ready for anything with as many good backups as you can put your hands on.

answered Feb 18 '11 at 02:26 AM