Yes, at 1:3068 you are above the magic 1:5000. However, it is unlikely that you
need to enlarge your log_buffer. The most common cause of 'redo log space
requests' is slow switching between log files, rather than a lack of buffer
space. This is treated in several places on the Ixora web site. However, the
bottom line is that you need to get a breakdown of these waits from
V$SYSTEM_EVENT. In general, if the number of 'log buffer space' waits is less
than half of the number of 'log file switch completion' waits, then the space
waits can be regarded as a secondary problem.


-----Original Message-----
Sent: Thursday, 24 August 2000 14:20


I have a question on the Ratio of the redo log buffers.
I have the following from my OLTP database

SQL> select name,value from v$sysstat where name in('redo log space
requests','redo entries');

NAME                                                                  VALUE
---------------------------------------------------------------- ----------
redo entries                                                        8649145
redo log space requests                                                2819

And i have read from the manual that the ratio is the no. of redo log space
requests to the number of redo entries that should not be greater than 1 in
5000. What does this statement means, is the above ratio good enough, do i have
to add more redo log buffer space?