Alert Log Report checkpoint not complete error how to resolve

Peak Alert Log Report checkpoint not complete more frequently, what needs to be adjusted redo?

When Oracle wants to reuse one of your redo logs, it finds that the checkpoint in the redo log is still in, and Oracle will report this warning in alter log

The adjustments associated with this alert are as follows:

1, the system IO performance problems, DBWR process written too slow

2, Log_checkpoint_timeout,fast_start_mttr_target, Log_checkpoint_interval set unreasonable, caused CHECKPOINT too frequent

3, the log file is too small

4. Too many data files

There are also several aspects of solving nature:

1, to find SA evaluation of IO performance, you can appropriately increase the number of DBWR process + open Asynchronous IO

2, reasonable set these several parameters and log file size, Oracle's official recommendation is 20 minutes to switch log

3, if there are too many file files, in the checkpoint data file header is frozen, not conducive to checkpoint performance and database performance, so the data file is not recommended too small

More Wonderful content: http://www.bianceng.cnhttp://www.bianceng.cn/database/Oracle/