I have been trying to understand the concept of redo log threads in a RAC environment. I understand how the redo logs work in a single instance database. But I cannot find proper documentation anywhere that clearly explains what a redo log thread is and how the whole thing works in a RAC setup.
e.g. Here is what I have on my 2 node RAC database:
Here is the formated output.
select inst_id,group#,thread#,status from gv$log
INST_ID GROUP# THREAD# STATUS
----- ----- ------- -------------
1 1 1 CURRENT *
1 2 1 INACTIVE
1 3 2 CURRENT *
1 4 2 ACTIVE
1 5 1 INACTIVE
2 1 1 CURRENT *
2 2 1 INACTIVE
2 3 2 CURRENT *
2 4 2 ACTIVE
2 5 1 INACTIVE
select inst_id,group#,member from gv$logfile order by member,inst_id
INST_ID GROUP# MEMBER
----- ----- ---------------------------------------------------
1 1 +DATA/rac11gr2/onlinelog/group_1.257.722950097 *
2 1 +DATA/rac11gr2/onlinelog/group_1.257.722950097 *
1 2 +DATA/rac11gr2/onlinelog/group_2.258.722950099
2 2 +DATA/rac11gr2/onlinelog/group_2.258.722950099
1 3 +DATA/rac11gr2/onlinelog/group_3.265.722965561
2 3 +DATA/rac11gr2/onlinelog/group_3.265.722965561
1 4 +DATA/rac11gr2/onlinelog/group_4.266.722965567
2 4 +DATA/rac11gr2/onlinelog/group_4.266.722965567
1 5 +DATA/rac11gr2/onlinelog/group_5.269.723713053
2 5 +DATA/rac11gr2/onlinelog/group_5.269.723713053
ASMCMD> pwd
+data/rac11gr2/onlinelog
ASMCMD> ls -l
Type Redund Striped Time Sys Name
------- ------ ----- ----------- --- -------------------------------
ONLINELOG UNPROT COARSE JUL 07 14:00:00 Y group_1.257.722950097
ONLINELOG UNPROT COARSE JUL 07 14:00:00 Y group_2.258.722950099
ONLINELOG UNPROT COARSE JUL 07 14:00:00 Y group_3.265.722965561
ONLINELOG UNPROT COARSE JUL 07 22:00:00 Y group_4.266.722965567
ONLINELOG UNPROT COARSE JUL 07 14:00:00 Y group_5.269.723713053
ASMCMD>
>
why are there 4 current log,shouldn't be 2 for database and only one for instance?
Can anybody help me?Note the line with '*'.
Edited by: liuxiang on 2010-7-7 下午6:27
Edited by: liuxiang on 2010-7-7 下午6:44