Checkpoint not complete

[pd123]

Hi,

I am getting 'Checkpoint Not Complete' messages in the alert log. What is causing this? There is plenty of disk space so it hasn't run out of space. Other than that the db seems to be running fine.

Thanks,
pd123

Below is an extract from the log file:


Tue Oct 29 05:48:32 2002
ARC0: Beginning to archive log# 1 seq# 211379
ARC0: Completed archiving log# 1 seq# 211379
Tue Oct 29 05:48:37 2002
Thread 1 cannot allocate new log, sequence 211381
Checkpoint not complete
Current log# 2 seq# 211380 mem# 0: /oracle/redogroup2/redo01.log
Current log# 2 seq# 211380 mem# 1: /oracle/redogroup2/redo02.log
Thread 1 advanced to log sequence 211381
Current log# 1 seq# 211381 mem# 0: /oracle/redogroup1/redo01.log
Current log# 1 seq# 211381 mem# 1: /oracle/redogroup1/redo02.log
Tue Oct 29 05:48:39 2002
ARC0: Beginning to archive log# 2 seq# 211380
ARC0: Completed archiving log# 2 seq# 211380
Tue Oct 29 05:48:44 2002
Thread 1 cannot allocate new log, sequence 211382
Checkpoint not complete
Current log# 1 seq# 211381 mem# 0: /oracle/redogroup1/redo01.log
Current log# 1 seq# 211381 mem# 1: /oracle/redogroup1/redo02.log
Thread 1 advanced to log sequence 211382
Current log# 2 seq# 211382 mem# 0: /oracle/redogroup2/redo01.log
Current log# 2 seq# 211382 mem# 1: /oracle/redogroup2/redo02.log
Tue Oct 29 05:48:46 2002
ARC0: Beginning to archive log# 1 seq# 211381
ARC0: Completed archiving log# 1 seq# 211381
Tue Oct 29 05:48:51 2002
Thread 1 cannot allocate new log, sequence 211383
Checkpoint not complete
Current log# 2 seq# 211382 mem# 0: /oracle/redogroup2/redo01.log
Current log# 2 seq# 211382 mem# 1: /oracle/redogroup2/redo02.log
Thread 1 advanced to log sequence 211383
Current log# 1 seq# 211383 mem# 0: /oracle/redogroup1/redo01.log
Current log# 1 seq# 211383 mem# 1: /oracle/redogroup1/redo02.log
Tue Oct 29 05:48:52 2002
ARC0: Beginning to archive log# 2 seq# 211382
ARC0: Completed archiving log# 2 seq# 211382

 

[rcurva]

It means that the checkpoint is still not complete but a new log switch is being performed (that triggers another checkpoint).

There are several parameters that can be used to delay the checkpoints like LOG_CHECKPOINT_TIMEOUT, LOG_CHECKPOINT_INTERVAL.

But I would strongly suggest that you increase the size of your logfiles.
Robbie

"The rule is, not to besiege walled cities if it can possibly be avoided" -- Art of War

 

[pd123]

Thanks for that.

Is it possible to increase the log file size in one command, or must I drop the existing log file and re-create it with a new size (and doing this for each one)?

Thanks,
pd123

 

[karluk]

You have to drop and recreate the existing redo log to resize it. Please see thread186-109022 for more details.