cancel
Showing results for 
Search instead for 
Did you mean: 

Redo log buffer 2228 < 4000 - 2,228 < 4000 redo entries

former_member196171
Participant
0 Kudos

Hi experts ,

I am getting following error in Oracle ( 11.2.0.3)

Redo log buffer 2228 < 4000 - 2,228 < 4000 redo entries

When I run sql for redo buffer allocation retries, I get as below

SELECT NAME, VALUE FROM V$SYSSTAT WHERE NAME = 'redo buffer allocation retries';

NAME                                                                  VALUE

---------------------------------------------------------------- ----------

redo buffer allocation retries                                        74955

Everyday it increases by 200

Current LOG_BUFFER size is 20MB

show parameter LOG_BUFFER;

NAME                                 TYPE        VALUE

------------------------------------ ----------- ------------------------------

log_buffer                           integer     19955712

How should I resolve this issue ? by increase LOG_BUFFER size ? How do I calculate optimal size ?

Accepted Solutions (0)

Answers (2)

Answers (2)

0 Kudos

Hello Abu Zahir,

This is cused when the active log file has filed up and Oracle is waiting for disk space to be allocated for the redo log entries. Space is created by performing a log switch.

Increase the number of redologs groups or increase their sizes.

Br

Tomislav

0 Kudos

I have to say that I never saw the mentioned error.

are you sure it is an error?, where do you see it? can you paste it here with some lines before and after?

It seems that you have the redo log set to 19 MB.

Do you have set the parameter or it is the default value?

How many CPUs Has your system.

In an SAP system, usually we recommend not to set this parameter (check note 1627481).

former_member196171
Participant
0 Kudos

Hi Fidel,

Thanks for your reply ,

I see the error in DB02 as below , looks like showing as RZ20 threshold value set

We have 16 CPU ,log_buffer is set to 20MB and redolog file size is 500 MB which are default value

-Al Mamun

0 Kudos

Abu Zahir Al-Mamun wrote:

We have 16 CPU ,log_buffer is set to 20MB and redolog file size is 500 MB which are default value

-Al Mamun

Hi,

1.- How many redo logs groups do you have?

2.- What is the size of the redo log (you indicate 500 but then you talk about the default value which is 50, unless SAP has changed it recently.

3.- what is the logswitch frequency?

4.- When was the DB restarted?

5.- what are the values for the Oracle statistics "redo log space requests", "redo entries" and "redo log space wait time"

That seems to be a "ratio" and as all ratios it might be useless.

stefan_koehler
Active Contributor
0 Kudos

Hi Fidel,


Fidel Vales wrote:

That seems to be a "ratio" and as all ratios it might be useless.

i want to give you 5000 thumbs up for that, but unfortunately SCN allows only one. The only thing i still wonder is why SAP (CCMS tools, etc.) does no get the rid of it.

Regards

Stefan

P.S.: By the way the redo buffer handling is much more complicated than just setting parameter log_buffer (or adding multiple redo log groups), e.g. private redo / IMU, multiple public redo buffers / threads, dynamic switching between old (one change, one record, one allocation) and new private / public redo handling. However troubleshooting this is impossible through a forum and without system access.

0 Kudos

"I've seen things you wouldn't believe"

The only thing i still wonder is why SAP (CCMS tools, etc.) does no get the rid of it.

probably a mix of dinosaurs and power,