Oracle Redo Logs
To for a log switch to put the current redo log into the next group:alter system switch logfile;
Although current log is moved to the next group, previously current logs are still active and cannot be safely manipulated. Logs are active because a checkpoint has been added to the checkpoint queue, but the changes are still in buffer cache, not yet committed to datafiles. To force a checkpoint immediately:
alter system checkpoint;
This will force oracle to process the checkpoint queue immediately, commit all dirty buffers to the datafiles. Once that is done, ACTIVE redo logs are now INACTIVE. In this state, one can drop a redo log group, add /delete members to/from a redo log group.
backup and delete archive logs
Archive logs may be stored on ASM and sometimes, you do not want these logs to take up ASM space. One can backup archive logs with rman and then delete them from the ASM storage:RMAN> backup archivelog all delete input; SQL> select name, archived, status from v$archived_log;
