In Oracle formal documentation, redo switch should be between 6-8 in 1 hour. If you have a number than this( its also related what you are running) you need to change your redolog file size.

In this post, I try to explain how we can resize redolog files on RAC instance while we have ASM.

1. First check existing status wiht below query:

select l.group#, l.thread#,
f.member,
l.archived,
l.status,
(bytes/1024/1024) fsize
from
v$log l, v$logfile f
where f.group# = l.group#
order by 1,2;

Sample output can be like below:

GROUP# THREAD# MEMBER                                        ARCHIVED    STATUS      MB
—— ——- ———————————             ——– ——— —
1    2    +ORADATA/test/onlinelog/group_8.882.743077403        NO        CURRENT     100
2    2    +ORADATA/test/onlinelog/group_9.610.743077405        YES       INACTIVE    100
3    1    +ORADATA/test/onlinelog/group_10.871.743077407       YES       INACTIVE    100
4    1    +ORADATA/test/onlinelog/group_11.886.743078145       YES       INACTIVE    100

2. Now our purpose is increasing redosize from 100Mb to 500 Mb

3. We are going to add our new redo file by below command:

alter database add logfile group 5 ‘+ORADATA’ size 500M;      << run this command on node 2

alter database add logfile group 6 ‘+ORADATA’ size 500M;    << run this command on node 2

alter database add logfile group 7 ‘+ORADATA’ size 500M;    << run this command on node 1

alter database add logfile group 8 ‘+ORADATA’ size 500M;    << run this command on node 1

4. We run below query again

select l.group#, l.thread#,
f.member,
l.archived,
l.status,
(bytes/1024/1024) fsize
from
v$log l, v$logfile f
where f.group# = l.group#
order by 1,2
/

Sample output can be like below:

GROUP# THREAD# MEMBER                                        ARCHIVED           STATUS        MB
——      ——-    ———————————                                    ——–                 ———           —
1    2    +ORADATA/test/onlinelog/group_1.882.743077403        NO                CURRENT        100
2    2    +ORADATA/test/onlinelog/group_2.610.743077405        YES               INACTIVE       100
3    1    +ORADATA/test/onlinelog/group_3.871.743077407        YES               INACTIVE       100
4    1    +ORADATA/test/onlinelog/group_4.886.743078145        YES               INACTIVE       100
5    2    +ORADATA/test/onlinelog/group_4.1728.743078149       NO                UNUSED         500
6    2    +ORADATA/test/onlinelog/group_6.1728.743077542       NO                UNUSED         500
7    1    +ORADATA/test/onlinelog/group_7.1728.743078149       NO                UNUSED         500
8    1    +ORADATA/test/onlinelog/group_5.1728.743077852       NO                UNUSED         500

5. Switch until we are into log group 5,6,7,8 so we can drop log groups 1, 2,3 and 4:

SQL> alter system switch logfile;
** repeat as necessary until group 5 first is CURRENT **

6. Run the query again to verify the current log group is group 4, if its INACTIVE than you can drop related group safely

SQL> select group#, status from v$log;

7. Now drop redo log groups 1, 2, and 3:

SQL> alter database drop logfile group 1;
SQL> alter database drop logfile group 2;
SQL> alter database drop logfile group 3;
SQL> alter database drop logfile group 4;

Verify the groups were dropped, and the new groups’ sizes are correct.

select l.group#, l.thread#,
f.member,
l.archived,
l.status,
(bytes/1024/1024) fsize
from
v$log l, v$logfile f
where f.group# = l.group#
order by 1,2
/
Advertisements