Alter system archive logthread 1 cannot allocate new log, sequence 2594 checkpoint not complete is actually a common error. Generally, it is because when the log is fully written, the log group will be switched, and a checkpoint will be triggered. dbwr will write dirty blocks in the memory to the data file, this log Group will not be released as long as it is not completed. If the archive mode is enabled, the archive write process is also accompanied by arch. If the redo log is generated too fast, when CPK or archiving is not completed, lgwr has filled the remaining log groups and needs to write redo logs into the current log group, in this case, a conflict occurs and the database is suspended. The preceding error message is always written to alert. log. Add a log group: 1. select group #, sequence #, bytes, members, status from V $ log; view the status group # sequence # bytes members status ---------- ------------ ------------------ 1 16946 52428800 1 inactive # idle 2 16947 52428800 1 inactive # idle 3 16948 1 current # In use 2. Alter database add logfile Group 4 ('/opt/oradata/orclbj/redo04.log ') size: 200 MB; add one group of logs to increase the size of the log group as needed. Alter database add logfile group 5 ('/opt/oradata/orclbj/redo05.log') size 200 m; Alter database add logfile group 6 ('/opt/oradata/orclbj/redo06.log ') size 200 m; 3. Alter system switch logfile; switch log group 4; Alter database drop logfile group 1; Delete log group 1 when adding a log group online, when deleting a log group, you can only delete a log group in the inactive status.
References