Oracle cannot allocate new log "Private strand flush not complete" or "Checkpoint not complete" in alert log.

It's not unusual to get "Private strand flush not complete" or "Checkpoint not complete" messages with cannot allocate new log in alert log. Both are relating redo log writing and they are not errors. But if you get these a lot then it is better to try to fix them.

1. Usually easiest way to fix these is add more Redo Logs and Redo Log groups. 

2. Other way is increase log files size.

3. Increasing the value for db_writer_processes (and use ASYNC I/O with Redo Logs ) can help with  these. These are increasing Redo writing speed.

4. You can also try to tune checkpoint to fix "Checkpoint not complete". Here you can find parameters to tune: Checkpoint Tuning and Troubleshooting Guide [ID 147468.1]

5. And also moving log files to faster disk is possible fix.


If you want more detailed info about these alert log messages read following MY Oracle Support (MOS) documents:

Alert Log Messages: Private Strand Flush Not Complete [ID 372557.1]

Can Not Allocate Log [ID 1265962.1]
Checkpoint Tuning and Troubleshooting Guide [ID 147468.1]