5 Replies Latest reply: Jan 20, 2009 1:14 AM by 643952 RSS

    ORA-00333: redo log read error block 283081 count 8192

    643952
      I am starting the database..it mount but after that it gives me this error

      ORA-00333: redo log read error block 283081 count 8192

      Below are the contents of alert Log.._Please advice_


      Completed: ALTER DATABASE MOUNT
      Tue Jan 20 10:24:45 2009
      ALTER DATABASE OPEN
      Tue Jan 20 10:24:45 2009
      Beginning crash recovery of 1 threads
      parallel recovery started with 2 processes
      Tue Jan 20 10:24:45 2009
      Started redo scan
      Tue Jan 20 10:25:00 2009
      Errors in file /d01/oracle/PROD1/db/tech_st/10.2.0/admin/PROD1_prod1/udump/prod1_ora_32356.trc:
      ORA-00333: redo log read error block 283081 count 8192
      ORA-00312: online log 2 thread 1: '/d01/oracle/PROD1/db/apps_st/data/log02a.dbf'
      ORA-27072: File I/O error
      Linux-x86_64 Error: 2: No such file or directory
      Additional information: 4
      Additional information: 283081
      Additional information: 257536
      Tue Jan 20 10:25:16 2009
      Errors in file /d01/oracle/PROD1/db/tech_st/10.2.0/admin/PROD1_prod1/udump/prod1_ora_32356.trc:
      ORA-00333: redo log read error block 283081 count 8192
      ORA-00312: online log 2 thread 1: '/d01/oracle/PROD1/db/apps_st/data/log02a.dbf'
      ORA-27091: unable to queue I/O
      ORA-27072: File I/O error
      Linux-x86_64 Error: 2: No such file or directory
      Additional information: 4
      Additional information: 283081
      Additional information: 257536
      Tue Jan 20 10:25:31 2009
        • 1. Re: ORA-00333: redo log read error block 283081 count 8192
          174414
          Hi,
          you can try this...
          1. Recover database:

          sql>recover database until cancel

          If "recovery complete" is not returned and you are prompted for an
          archivelog which DOES NOT exist, then apply the online redologs.

          At the prompt enter the full path name to a member of your oldest
          online redolog group. Apply a member of all other groups if prompted
          for further recovery.

          If "recovery complete" is not returned and you are prompted for an
          archivelog which DOES exist, then apply the archivelogs and if
          necessary, the online redologs as above.

          At the prompt press return.

          2. Open database:

          sql> alter database open resetlogs
          • 2. Re: ORA-00333: redo log read error block 283081 count 8192
            643952
            Db is running in noarchivelog mode

            can i recover to log sequence 60 or something..currently its 61
            there were no log switches after the error
            • 3. Re: ORA-00333: redo log read error block 283081 count 8192
              150505
              Hi looks like your online redo log is corrupt.
              First place take backup of what ever is available.
              If previous shutdown was normal then I think you dont have to worry. Startup mount, take trace of control file.
              Recreate controlfile and open database with reset logs.
              I would suggest to check the disk on which concerned logfile is there. Also check for other datafiles if there are any on this mount point.
              And implement backup policy is data is important
              regards
              Pravin
              • 4. Re: ORA-00333: redo log read error block 283081 count 8192
                Naveed Yameen Khakoo
                Hi,

                Have checked the content(s) of (/d01/oracle/PROD1/db/tech_st/10.2.0/admin/PROD1_prod1/udump/prod1_ora_32356.trc) or have you verified if the file actually exists?

                Can you query v$log (select * from v$log) and post the contents here? The database should be in mounted state to do this.

                Regards,
                Naveed.
                • 5. Re: ORA-00333: redo log read error block 283081 count 8192
                  643952
                  I did what Prabhu told me..But i recovered using backup controlfile and when i was asked to apply logs..i applied my oldest logs and it worked


                  Like i have two groups with two members each
                  I applied log1a.dbf and it said media recovery completee
                  I opened the database But then it started giving me errors for undo tablespace

                  I made another undo tablespace..tried dropping the old one but it did not permitted as it had some segments with status "needs recovery"

                  Than i added this parameter in pfile with the correupted segment and than tried to drop the segment but it still did not permit

                  corruptedrollback_segments =(corrupted_undo segment_name)


                  Next what i did was I mounted the database, ran another session of media recovery and opened the databse using resetlogs

                  Than i dropped the old undo and it went through successfully..

                  If you think anything i did wrong than please advice..
                  I hope this action plans helps you in case you come across same errors some day

                  I would be very thankful if you can refer me a document of recovery which covers all kinds of recoveries and scenarios and commands too

                  Thanks alot