OTN Discussion Forums : redo log switch ...

Reply to this Thread

Search Forum

Back to Thread List

Replies: 10 - Pages: 1 - Last Post: Oct 29, 2007 10:47 PM Last Post By: Aman....

user507271

Posts: 105
Registered: 05/02/06

redo log switch
Posted: Oct 29, 2007 7:14 AM

The sgbd has redo log size configuration of 25M but it generates 18M of archives every 3 minuetes on my system,

Those parameters are configured,

log_checkpoint_timeout = 1800
fast_start_mttr_target = 0
log_checkpoint_internal = 0

I do not understand why the system does not generate 25M archive redo log.

Thanks for your help

Fabrice Chapuis

FeNiCrC_Neil

Posts: 1,593
Registered: 02/21/00

Re: redo log switch
Posted: Oct 29, 2007 8:07 AM

in response to: user507271

If a transaction cannot 'fit' into a redo log, the associated arch file is smaller than the redo log file. 25M redo logs are too small.

Anantha R

Posts: 2,757
Registered: 08/02/07

Re: redo log switch
Posted: Oct 29, 2007 8:28 AM

in response to: user507271

Take a look at the following:

http://forums.oracle.com/forums/thread.jspa?messageID=2155019&#2155019

rahulras

Posts: 442
Registered: 11/23/98

Re: redo log switch
Posted: Oct 29, 2007 9:36 AM

in response to: user507271

Hi,

The size of your redo-log buffer is 25M.

What is the size of redo log files ? You can see the file names in v$logfile. Is it 18M?

Regards

yingkuan

Posts: 10,896
Registered: 10/08/98

Re: redo log switch
Posted: Oct 29, 2007 10:37 AM

in response to: user507271

double check the size of your redo logfiles, and some related parameters,

select group#, bytes from v$log;

---

show parameter archive_lag_target

burleson

Posts: 2,343
Registered: 05/06/98

Re: redo log switch
Posted: Oct 29, 2007 1:43 PM

in response to: user507271

Hi,

redo log size configuration of 25M but it generates 18M of archives every 3 minuetes on my system

I try to size my redo logs never to get more than 5 switches per hour, at peak load.

In a busy production environment, it is important to ensure that the frequency of redo log switches is not more than 5 per hour. As Oracle offloads redo log images from the redo log filesystem onto the archived redo log filesystem, excessive I/O can occur and the archived redo log must be promptly written to tape to keep the free space in the filesystem. If the archived redo log filesystem becomes full, the database grinds to a halt.

Not many DBA’s are aware of the v$log_history view, and how it can be used to plot the frequency of online redo log switches.

Hope this helps. . .

Don Burleson
Oracle Press author
Author of “Oracle Tuning: The Definitive Reference”
http://www.dba-oracle.com/bp/s_oracle_tuning_book.htm

Hans Forbrich

Posts: 9,948
Registered: 03/13/99

Re: redo log switch
Posted: Oct 29, 2007 1:59 PM

in response to: burleson

Not many DBA’s are aware of the v$log_history view,
and how it can be used to plot the frequency of
online redo log switches.

Just like not many DBAs are aware that ARCHIVE_LAG_TARGET can be used to force the log file switch when the load is low - allowing the DBA to set the log switch frequency to become quite predictable.

Aman....

Posts: 8,910
Registered: 05/20/01

Re: redo log switch
Posted: Oct 29, 2007 9:42 PM

in response to: Hans Forbrich

hi sir,
Can you in brief explain this parameters working?I am reading it here
http://download-uk.oracle.com/docs/cd/B19306_01/server.102/b14237/initparams009.htm

but its somehow related to dataguard I guess.Can you please explain it for singe db working without DG?
Thanks and regards,

yingkuan

Posts: 10,896
Registered: 10/08/98

Re: redo log switch
Posted: Oct 29, 2007 9:54 PM

in response to: Aman....

Check [url=http://forums.oracle.com/forums/thread.jspa?messageID=2126130&#2126130]This Thread [/url]

Where Howard had good explanation of the use of this parameter.

orawarebyte

Posts: 3,644
Registered: 10/02/00

Re: redo log switch
Posted: Oct 29, 2007 10:19 PM

in response to: yingkuan

yingkuan as you referred this link which nice one but as you and hjr agreed that

it is wrong loss of all current redo logfiles are protected by frequent checkpoint.

what i think if we set aside the performance for a while that frequent checkpoint reduce recovery time (i.e downtime concerned) cause less redo need to be applied which will ultimately/indirectly fall the current redo logfiles protected isnt?

Khurram

Aman....

Posts: 8,910
Registered: 05/20/01

Re: redo log switch
Posted: Oct 29, 2007 10:20 PM

in response to: yingkuan

thanks Ying.I shall read it.
Aman....

Pages: 1

Back to Thread List

Legend
	Guru : 2500 - 1000000 pts
	Expert : 1000 - 2499 pts
	Pro : 500 - 999 pts
	Journeyman : 200 - 499 pts
	Newbie : 0 - 199 pts
	Oracle ACE Director
	Oracle ACE Member
	Oracle Employee ACE
	Helpful Answer (5 pts)
	Correct Answer (10 pts)