Home All Groups Group Topic Archive Search About


Author
17 Mar 2006 4:46 PM
Marek Wierzbicki
Hi

In my server (MSSQL 2000 sp4) very often I see long time of executing query
with waittype=WRITELOG

Recovery is set to SIMPLE and there is no autoShrink.

Auto increment file size (log) is set to 10%

What can be the reason of this problem?

Marek

Author
17 Mar 2006 5:58 PM
David Browne
"Marek Wierzbicki" <marek.wierzbickii***@azymuttttt.pl> wrote in message
news:dvep4k$1loj$1@news2.ipartners.pl...
> Hi
>
> In my server (MSSQL 2000 sp4) very often I see long time of executing
> query with waittype=WRITELOG
>
> Recovery is set to SIMPLE and there is no autoShrink.
>
> Auto increment file size (log) is set to 10%
>
> What can be the reason of this problem?


This wait type means that a session is has finished a transaction or
autocommit statement and it's waiting for the results to be flushed to the
log file.

At the end of a transaction or statement that changes data the changes must
be written to the log and the log committed to disk.  This is the only part
of the change that has to hit the physical disk, and it's a major
performance consideration.

In SIMPLE recovery mode the transaction log is written just the same, except
it's truncated when the database is checkpointed.  At checkpoint the
commited data is flushed to the database files and the log records are no
longer needed in SIMPLE mode.


David
Are all your drivers up to date? click for free checkup

Author
20 Mar 2006 12:26 PM
Marek Wierzbicki
> This wait type means that a session is has finished a transaction or
> autocommit statement and it's waiting for the results to be flushed to the
> log file.

OK


> At the end of a transaction or statement that changes data the changes
> must be written to the log and the log committed to disk.  This is the
> only part of the change that has to hit the physical disk, and it's a
> major performance consideration.
>
> In SIMPLE recovery mode the transaction log is written just the same,
> except it's truncated when the database is checkpointed.  At checkpoint
> the commited data is flushed to the database files and the log records are
> no longer needed in SIMPLE mode.

Is the data must be deleted from log file?

Marek
Author
17 Mar 2006 11:18 PM
Andrew J. Kelly
Make sure your log file is on a different physical drive than the data,
preferably a Raid 1 or 10.

--
Andrew J. Kelly  SQL MVP


Show quoteHide quote
"Marek Wierzbicki" <marek.wierzbickii***@azymuttttt.pl> wrote in message
news:dvep4k$1loj$1@news2.ipartners.pl...
> Hi
>
> In my server (MSSQL 2000 sp4) very often I see long time of executing
> query with waittype=WRITELOG
>
> Recovery is set to SIMPLE and there is no autoShrink.
>
> Auto increment file size (log) is set to 10%
>
> What can be the reason of this problem?
>
> Marek
>
>
Author
20 Mar 2006 12:23 PM
Marek Wierzbicki
> Make sure your log file is on a different
> physical drive than the data,
> preferably a Raid 1 or 10.

I know it, but today its on the same disk as data file

Marek
Author
20 Mar 2006 12:29 PM
Dan Guzman
> I know it, but today its on the same disk as data file

Hence, the waits on log I/O.  Also, it's better to size the log to
accommodate needs (largest transaction) and let growth kick in only as a
safety net.

--
Hope this helps.

Dan Guzman
SQL Server MVP
Author
20 Mar 2006 3:56 PM
Marek Wierzbicki
>> I know it, but today its on the same disk as data file
>
> Hence, the waits on log I/O.
> Also, it's better to size the log to
> accommodate needs (largest transaction)
> and let growth kick in only as a
> safety net.

OK. So I need to large my matrix

Marek

Bookmark and Share