Oracle FAQ Your Portal to the Oracle Knowledge Grid
HOME | ASK QUESTION | ADD INFO | SEARCH | E-MAIL US
 

Home -> Community -> Usenet -> c.d.o.server -> Re: MS SQL 7.0 HANGS

Re: MS SQL 7.0 HANGS

From: David Owen <weedwiper_at_my-deja.com>
Date: 10 Aug 1999 09:54:58 -0600
Message-ID: <wkzozzfx65.fsf@CHY79.i-did-not-set--mail-host-address--so-shoot-me>


"Max Akbar" <Maximum2000_at_msn.com> writes:

> I don't think this is a bug!

Come on now, this is a Micro$oft product :-)

>
> Common sense tells anyone who is involved in DBA work that if there is any
> task that requires time like your task 30 min adequate time should be
> allocated or the task should be run off hours. If you have to run this query
> then you should think about multi processor server or a hardware that can
> handle such a query the fault may not be the software (again common sense).

I do not know how M$ SQL Server 7 handles its processes (I presume that it is still similar to Sybase and that it is thread based and non-pre-emptive (can a word have two hyphens?)). If that is so, then this is a very peculiar situation. Unless every last bit of data is cached and the process sends and receives *no* data, then the process will send itself to sleep and give control back to the kernel when it requires some form of I/O, at which point a separate process should be able to be scheduled. That is certainly how Sybase works, multi-engined or not. I know that, in theory, it would be possible for a piece of T-SQL to cause a server to appear hung, but it would have to be tight loop that did not require any interaction or any data from anywhere but cache. I suspect a bug.

Cha Ching!

dowen

--
David Owen 7, Stradbrooke Rise SW, Calgary, AB, Canada dso@<s.p.a.m-me-not>home.com
You needn't be well to wealthy, but you gotta be whole to be holy.

                                            Greg Lake, Works Vol 1
Received on Tue Aug 10 1999 - 10:54:58 CDT

Original text of this message

HOME | ASK QUESTION | ADD INFO | SEARCH | E-MAIL US