Thread 1 cannot allocate new log - Checkpoint not complete caused by small
727854Jul 19 2010 — edited Jul 23 2010Hello, one of my customers databases (setup by the customer) is running in archivelog mode. We find every day in the archivelog more than 10x the error "...cannot allocate new log - Checkpoint not complete "
The system has theese parameters:
select group#,members,bytes from v$log;
GROUP# MEMBERS BYTES
---------------------- ---------------------- ----------------------
1 2 8388608
2 2 8388608
3 2 8388608
4 2 8388608
5 2 8388608
6 2 8388608
7 2 8388608
8 2 8388608
9 2 8388608
10 2 8388608
I asked the formder admin, why the redologs has only 8mb size. The answer was: "To find out, when there was big activity in the database. The count of archivelogs are bigger if you has small redologs. If you use some 250m redologs, we cannot see when tha database has big load. But this is important in case of recovery to recover the database to the point in time before the big load was has taken place." ... ???
Okay, if the customer wants to have 8 MB redologs ... I don´t care. But how can i prevent now the error message? The only way for me is to setup more redolog groups. But it would be very confusing to have 50 or 70 groups of redologs.
What can I do else to prevent the error?