“Log file switch (checkpoint incomplete)” waiting events are seen in v$session_wait, and AWR report. This event is caused normally by :
1) Redo log files are too small.
2) More redo groups are needed.
3) Disk I/O is not fast enough.
Normally this issue can be addressed through adding more redo groups and/or increasing redo log file size.
Oracle recommends no more than five log switches per hour.
Unpublished and Unofficially switch the logfile of all the threads by one command
A. Switch the logfile of current instance only.
SQL> alter system switch logfile;
B. Archive the logfile of all the threads in RAC environment.
SQL> alter system archive log current;
C. Switch the logfile of all the threads in RAC environment (Unpublished , Unofficially ).
SQL>alter system switch all logfile;