Re: LGWR / CKPT Spin Out of Control

Redo Logs	3 Groups / 2 Members each / 5MB each
db_block_size	8192
log_buffer	163840
log_checkpoint_interval	10000
log_checkpoint_timeout	0

The database is running in NOARCHIVELOG mode. Also, this is a development database, so the log switches are generally very sporadic, but with occassional bursts of activity. When the LGWR / CKPT problem has occurred, there have typically been very few log switches preceding the problem (1 or 2 in an hour).

Any ideas?
Rich

In article <80refg$5av$1_at_news2.tor.accglobal.net>,   "John Chiu" <johnymc_at_netscape.net> wrote:
> What is your redo log size, db_block_size, log_checkpoint_interval
and
> log_check_point_timeout in your init.ora?
>
> If you size the log_check_point_interval too small, it will have
multiple
> checkpoint happened between log switchs.
>
> John Chiu
>
> Rich Goldkamp <RJGoldka_at_dcss.com> wrote in message
> news:80qs52$bl9$1_at_nnrp1.deja.com...
> > Help -- I've got 2 instances running (Oracle 8.0.5) on Solaris 2.6.
> > Occassionally, the LGWR and CKPT processes for one of the instances
> > will start consuming massive amounts of CPU such that the system
> > becomes quite unresponsive. The alert log shows no activity (very
> > limited amount of log switches), and no background or user trace
files
> > are created. When we use Top Sessions, we can see no run away
queries
> > contributing to the problem.
> >
> > Has anyone ever encountered strange behavior like this before?
Where
> > should I be looking to root out the ultimate cause of this?
> >
> > Thanks,
> > Rich
> > --
> > Rich Goldkamp, Consultant, OCP RJGoldka_at_dcss.com
> > Digital Consulting & Software Services (504) 523-5005 x2736
> >
> >
> > Sent via Deja.com http://www.deja.com/
> > Before you buy.
>
>

--

Rich Goldkamp, Consultant, OCP		RJGoldka_at_dcss.com
Digital Consulting & Software Services	(504) 523-5005 x2736

Sent via Deja.com http://www.deja.com/
Before you buy. Received on Tue Nov 16 1999 - 11:02:55 CST