3 Replies Latest reply on May 31, 2012 10:59 AM by Fran

    Checkpoint not complete error in alert log

    james_p
      version : 10.2.0.5 , 11.2.0.2

      This is what I found about "Checkpoint not complete" from google

      "+This message indicates that Oracle wants to reuse a redo log file, but the current checkpoint position is still in that log. In this case, Oracle must+
      +wait until the checkpoint position passes that log. Because the incremental checkpoint target never lags the current log tail by more than 90%+
      +of the smallest log file size, this situation may be encountered if DBWR writes too slowly, or if a log switch happens before the log is completely full,+
      +or if log file sizes are too small.+ "


      Question1.
      What do they mean by Oracle reusing a redo log file ? When one ORL is full , oracle will switch to the next one ; it won't reuse it. Right ?

      Question2.
      "the current checkpoint position is still in that log" . What is checkpoint position ?


      Question3. How often does a checkpoint occur ? (frequency as per time)
        • 1. Re: Checkpoint not complete error in alert log
          Helios-GunesEROL
          Hi;

          Please see:

          Checkpoint Tuning and Troubleshooting Guide [ID 147468.1]
          Checkpoint Not Complete In Alert.log Due To Setting Of Archive_lag_target [ID 435780.1]

          For your other question check Oracle concept guide and also below link:
          http://www.orafaq.com/wiki/Checkpoint
          http://www.adp-gmbh.ch/ora/notes.html

          Regard
          Helios
          • 2. Re: Checkpoint not complete error in alert log
            Karan
            YOUR Question1.
            What do they mean by Oracle reusing a redo log file ? When one ORL is full , oracle will switch to the next one ; it won't reuse it. Right ?

            Answer is that lgwr overwrites the log group only when the corresponding changes from buffer cache has been written to datafiles by DBWR.

            YOUR Question2.
            "the current checkpoint position is still in that log" . What is checkpoint position ?

            A checkpoint position is that position above which whose system change number (SCN) changes made in data blocks less than or equal to that SCN are known to be written out to the data files.

            Your Question 3
            How often does a checkpoint occur ? (frequency as per time)

            Conventionally it happens at every log switch, and also when fast start checkpointing is enabled with fast_start_mttr_target

            Regards
            Karan
            1 person found this helpful
            • 3. Re: Checkpoint not complete error in alert log
              Fran
              Question1. 
              What do they mean by Oracle reusing a redo log file ? When one ORL is full , oracle will switch to the next one ; it won't reuse it. Right ?
              You have some groups of redo log, each redo log has members with same information. Redo log run in a circular way, when a group is full, changes to the next group of redo log and when the last group is written continue overwritten the first group.

              http://docs.oracle.com/cd/B28359_01/server.111/b28310/onlineredo001.htm
              Question2.
              "the current checkpoint position is still in that log" . What is checkpoint position ?
              copy and paste from:
              http://docs.oracle.com/cd/E14072_01/server.112/e10713/startup.htm#BABGDACG
              "A data structure that indicates the checkpoint position, which is the SCN in the redo stream where instance recovery must begin"

              Question3. How often does a checkpoint occur ? (frequency as per time)
              Checkpoints occur in diferents situations, not per time, please check:
              http://docs.oracle.com/cd/E11882_01/server.112/e16508/startup.htm#CNCPT89045
              1 person found this helpful