SQL Server Failed Assertions

Thread view:

Enable EMail Alerts

Start New Thread

Thread rating:

Whit Gregg - 03 Jul 2008 17:02 GMT

I'm running SqlServer-2005 SP2 x64
I am getting the following 2 Failed Assertions:

SQL Server Assertion: File: "xact.cpp" line = 2879 Failed Assertion =
'!m_updNestedXactCnt' Trying to use the transaction while there are 1
parallel nested xacts outstanding.

SQL Server Assertion: File: <lckmgr.cpp><c/> line=9421 Failed Assertion =
'NULL == m_lockList.Head ()'.

is there a fix or work-around for this?

Thanks.

Signature

Whit Gregg
Lek Securities Corp
New York, NY

Reply to this Message

Eric Isaacs - 03 Jul 2008 19:50 GMT

Sounds like you may be starting a new transaction when one is already
outstanding on the same process. If that's the case, then using this
syntax before starting your transactions, might help...

IF @@TRANCOUNT = 0
BEGIN TRANSACTION

-Eric Isaacs

Reply to this Message

Whit Gregg - 03 Jul 2008 20:06 GMT

Eric,
Thanks for the reply.
I don't think any code changes at the T-SQL level will help.
these ASSERTs are inside the guts of the SQL Database-Engine.
I looks like an internal SQL-Server bug to me.
now these ASSERTs themselves may only be reporting an
internal data inconsistency, and the real problem began somewhere else
in the internals. But I was hoping that Microsoft had seen this problem
before,
and could help resolve it, behaps with a QFE

Signature

Whit Gregg
Lek Securities Corp
New York, NY

> Sounds like you may be starting a new transaction when one is already
> outstanding on the same process. If that's the case, then using this
[quoted text clipped - 4 lines]
>
> -Eric Isaacs

Reply to this Message

TheSQLGuru - 03 Jul 2008 22:22 GMT

If you open up a support request with MS and the issue is found to be a bug
I don't think they charge you for the support call. If it is a bug and it
isn't known, others could benefit from your pursuing this one with them.

Signature

Kevin G. Boles
Indicium Resources, Inc.
SQL Server MVP
kgboles a earthlink dt net

> Eric,
> Thanks for the reply.
[quoted text clipped - 15 lines]
>>
>> -Eric Isaacs

Reply to this Message

Linchi Shea - 03 Jul 2008 22:27 GMT

If it's critical, I'd open a support case. It's a SQL bug, and you wouldn't
be charged for teh case.

Linchi

> Eric,
> Thanks for the reply.
[quoted text clipped - 15 lines]
> >
> > -Eric Isaacs

Reply to this Message

JXStern - 04 Jul 2008 00:04 GMT

>Eric,
>Thanks for the reply.
>I don't think any code changes at the T-SQL level will help.

While I agree that this looks like an internal SQL bug, changing your
code might just work around it, even if you've done nothing wrong.

I assume this is repeatable. What is your code doing at that point?

J.

>these ASSERTs are inside the guts of the SQL Database-Engine.
>I looks like an internal SQL-Server bug to me.
[quoted text clipped - 3 lines]
>before,
>and could help resolve it, behaps with a QFE

Reply to this Message