database logs

Similar Questions

• Database log files

  Hi Experts,
  
  Can anyone tell me the location of the database log file?
  
  Thanks in advance

  Log files from database, 11g, would be in the folder called diag standing for the Dest diagnosis and is maintained by the diagnostic_dest parameter. To see where you have configured this setting to be, it would be in the Oracle default base location.

  HTH
  Aman...

• UCR database log files

  Hello

  Our RCU redologfiles reside on the file system, can they be moved to the ASM without restarting the database, I think we can run mix files in the file system and ASM.

  Thank you

  You must be running in logging mode archive, make one:

  SQL > change of check point system;

  and then try again.

• The database log switches statistical user name

  Hello

  is it possible to see the user witch / sql create a log switches (archivelog)?

  We have an oracle 11.2 database with multiple small schemas for many small applications.

  Some sql´s have unusual create many logswitches last night, we need to know what user / sql

  created the log switches.

  Thank you * T

  Newspapers do not belong to a specific user, so no user does cause a switching of newspaper. You can view statistics for a running session how many redo is generated, but it does not for the past.

• upgrade of database: log file error

  I run the database upgrade command and had error log below.
  
  SQL >
  SQL > select dbms_java.full_ncomp_enabled from double;
  Select double dbms_java.full_ncomp_enabled
  *
  ERROR on line 1:
  ORA-29558: JAccelerator (we'RE) not installed. See the Installation Guide for
  instructions.
  ORA-06512: at "SYS." DBMS_JAVA', line 236
  
  
  SQL > SELECT dbms_registry_sys.time_stamp ('JAVAVM') AS timestamp FROM DUAL;
  
  TIMESTAMP
  --------------------------------------------------------------------------------
  COMP_TIMESTAMP JAVAVM 2010-05-10 14:27:14
  DBUA_TIMESTAMP JAVAVM VALID ON 10-05 - 2010 14:27:14

  Please refer to:

  MOS Doc 293658.1 - 10.1 or 10.2 group Install patches to ORA - 29558 JAccelerator (ON) and ORA-06512 [VIDEO]
  MOS 452024.1 Doc - mandatory is WE (JAccelerator)?

  Re: upgrade 10.2.0.4.0!

  Concerning
  Rajesh

• The clause in the database log file duplicate

  I tried to duplicate a database (Oracle RAC 10 g ASM 2 node Windows 2003 Server R2) and in the document of the oracle, it went something like:
  
  DUPLICATE the TARGET DATABASE to dupdb
  PFILE = /dup/oracle/dbs/initDUPDB.ora
  LOGFILE
  ' / dup/oracle/oradata/trgt/redo01.log' SIZE 200 K,.
  ' / dup/oracle/oradata/trgt/redo02.log' SIZE 200 K,.
  ' / dup/oracle/oradata/trgt/redo03.log' SIZE 200 K;
  
  I thought as the source database RMAN backup already contained the definitions of redo log. Are only these redo log files used for the process during replication of the database? Is the logfile clause is required to use the command DUPLICATE TARGET DATABASE?
  
  Thank you much in advance.

  LOG file can be used to specify where the newspapers of recovery for the duplicate database should be placed. This clause is not mandatory and it is really necessary only if you intend to rename the redo log files. If you don't need/want to rename the files, you can use the NOFILENAMECHECK option.

• Standby database is down and redo logs is not transmitted

  Hi all
  
  I have a question about the primary recovery database logs transmition for the standby database.
  What happens if the transmition of redo logs was arrested for a period of time (1 month), if I want to re - build the database ensures what measures must be applied?
  The papers of recovery will be put on hold? and once the database pending is redo these logs will be sent again? or I have to rebuild my database Eve from scratch?
  
  Kind regards

  Hello;

  When I rebooted the month so I would change the parameter 'log_archive_dest_state_n' on the primary to postpone next it would stay like that.

  I would like to delete the database in waiting, and if the folder structure was not the same I correct it.

  When the month has increased I would use RMAN to duplicate a new sleep mode:

  http://www.Visi.com/~mseberg/duprman2.html

  And then I put "log_archive_dest_state_n" to Activate.

  Best regards

  mseberg

• TS4.1 database error

  Hello

  I was using my database.ini of my TS3.5 (on another PC) in my other PC with TS4.1.  I did it to be able to use the same schema.

  It comes to getting the error message:

  An error occurred calling "LogResults" in "ITSDBLog" of "ZNIUGOL TestStand database logging"
  An error has occurred by executing a statement.
  Schema: ATE_database_multiplePC
  Statement: UUT_RESULT.
  Description: The field ' UUT_RESULT. UUT_ERROR_MESSAGE'

  Research in the fields of the schema, I found that the expression of UUT_RESULT. UUT_ERROR_MESSAGE is 'Logging.UUTResult.Error.Msg'.

  I put a break before the stage of 'result of journal of database ' is executed and has looked at the settings.

  I do not see the parameter "'Logging.UUTResult.Error.Msg ' and that's why it failed."  There is, however, a similar setting called 'Logging.MainSequenceResult.Error.Msg'.

  My sequential model is also my TS3.5 system.  I use it instead of my sequential model TS4.1 since I have a little customized it.

  No explanation for this?  Can I use the parameter "Logging.MainSequenceResult.Error.Msg" in my scheme to replace the old "Logging.UUTResult.Error.Msg"? "'". "

  Thank you

  Rafi

  
  

• Changing result data Access to SQL Server database

  For many years, we have worked with the Access databases.
  Now, we started with SQL-Server.
  We only have a UutResultTable and a StepResultTable
  (because we have already started with TestStand 1.02).

  We have worked with integers as Primarykey (UutResultId, StepResultId)
  I get the following error when I work with this "integer as primary keys.

  An error occurred calling "LogResults" in "ITSDBLog" of "ZNIUGOL TestStand database logging"
  An error has occurred by executing a statement.
  Schema: E4DatabaseSchema
  Return: RESULT of the UUT.
  Description: The query with parameters ' (@P1 int,@P2 nchar (50), @P3 int,@P4 nchar (50), @P5 nchar (50), @P6 only)
  expects the parameter '@P1', which was not provided.
  Number :-2147217900
  NativeError: 8178
  SQLState: 42000
  Reported by: provider Microsoft OLE DB for SQL Server

  Source: TSDBLog

  My Questions:

  Is it necessary to work with the GUID is to use the generic schema insert (NOR)?
  Don't work with GUID does not reduce the performance of database?

  We use

  TestStand 4.2
  SQL Server 2008 R2
  CVI 2010
  Windows XP Service Pack 3

  Thank you very much for your help

  Paul

  Hello

  I think these 2 KBs should answer your question.

  http://digital.NI.com/public.nsf/allkb/904986636B5FC7D78625773E004E065F?OpenDocument

  http://digital.NI.com/public.nsf/allkb/5554E9522328ECCB8625772F004DA5C1?OpenDocument

  Reagards

• Why the connection to a local database becomes lost when the network is disconnected?

  I use TestStand 3.1. I have a configuration database on the local drive and I use the option "Log on the Fly.

  The test takes 48 hours and I want to assure you that it will not stop if the network connection is lost.

  If I unplug the network before starting the test, fine. I can plug the network again while the test runs without problem.

  But if the network is connected and I start the test and then disconnect the network of recording on the fly will give an error.

  I narrowed down it to "New USE for database Logging" step in the SequentialModel.seq (process model). If this step is performed with the network connection then the rest of the test needs a network connection.

  Is it possible to avoid this error?  The point of all the local database was to avoid network problems.

  Thank you for your help.

  TDOT-

  I connect you always locally, please try to change it to "localhost\SQLEXPRESS" to see if you can log in and then see if the same problem as you disconnect the LAN causes previous saw.

• TestStand database error - error converting data from nchar int type.

  Hello

  I used SQL standards database stored procedure options to store records of my test and I added the generated tables and stored procedures in a database and it seems to work OK in this form. However when I change the UUT_Results table to include a new column that contains the value of the primary key of another table in the database that I meet problems.

  I run a query that returns the primary key value (integer) for a given series number and stores the result as a global station (number format). The correct value is stored in the variable, but when I try to use this value in the stored procedure I get an error like below:

  An error occurred calling "LogResults" in "ITSDBLog" of "ZNIUGOL TestStand database logging"
  An error has occurred by executing a statement.
  Schema: SQL Server Stored Proc (NOR)
  Statement: UUT_RESULT.
  Description: Error converting data from nchar int type.
  Number:-2147217913
  NativeError: 8114
  SQLState: 22018
  Reported by: provider Microsoft OLE DB for SQL Server

  Source: TSDBLog

  My SQL statement for creating the table and the procedure for UUT_results are below:

  ~
  CREATE TABLE UUT_RESULT)
  Uniqueidentifier PRIMARY KEY ID,
  BoardDetailID int,
  STATION_ID varchar (255),
  BATCH_SERIAL_NUMBER varchar (255),
  TEST_SOCKET_INDEX int,
  UUT_SERIAL_NUMBER varchar (255),
  USER_LOGIN_NAME varchar (255),
  START_DATE_TIME is datetime,
  EXECUTION_TIME float,
  UUT_STATUS varchar (32),
  UUT_ERROR_CODE int,
  UUT_ERROR_MESSAGE varchar (255)
  CONSTRAINT UUT_RESULT_FK FOREIGN KEY (BoardDetailID) made REFERENCE BoardDetails)

  ~
  CREATE PROCEDURE InsertUUTRESULT
  @pID uniqueidentifier,
  @pBoardDetailID int,
  @pSTATION_ID varchar (255),
  @pBATCH_SERIAL_NUMBER varchar (255),
  @pTEST_SOCKET_INDEX int,
  @pUUT_SERIAL_NUMBER varchar (255),
  @pUSER_LOGIN_NAME varchar (255),
  DateTime @pSTART_DATE_TIME.
  float of @pEXECUTION_TIME.
  @pUUT_STATUS varchar (32),
  @pUUT_ERROR_CODE int,
  @pUUT_ERROR_MESSAGE varchar (255)
  AS
  INSERT INTO UUT_RESULT (ID, BoardDetailID, STATION_ID, BATCH_SERIAL_NUMBER, TEST_SOCKET_INDEX, UUT_SERIAL_NUMBER, USER_LOGIN_NAME, START_DATE_TIME, EXECUTION_TIME, UUT_STATUS, UUT_ERROR_CODE, UUT_ERROR_MESSAGE)
  VALUES)
  @pID,
  @pBoardDetailID,
  @pSTATION_ID,.
  @pBATCH_SERIAL_NUMBER,.
  @pTEST_SOCKET_INDEX,.
  @pUUT_SERIAL_NUMBER,.
  @pUSER_LOGIN_NAME,.
  @pSTART_DATE_TIME,.
  @pEXECUTION_TIME,.
  @pUUT_STATUS,.
  @pUUT_ERROR_CODE,.
  @pUUT_ERROR_MESSAGE)

  Any help would be appreciated,

  Thank you

  Stuart

  Stuart-

  What is the expression you are evaluating and the data type of the parameter that you defined in the Options database dialog box for the schema that you connect to data with?

• TestStand 4.2: error call "LogResults" in "ITSDBLog" of "ZNIUGOL TestStand database"

  I just installed 4.2 TestStand and SQL Server Express on a Windows XP computer brand new.  In TestStand, I implemented the "Database" property, so he tested connection successfully.  But when I open the Demo/C/computer.seq file to test, it gives me error in the result of logging.  On-the-Fly logging is not enabled.

  Here is the error message I get.

  ------------------------------------------------

  An error occurred calling "LogResults" in "ITSDBLog" of "ZNIUGOL TestStand database logging"
  An error has occurred by executing a statement.
  Figure: Insert generic (NOR)
  Statement: UUT_RESULT.
  Description: Invalid object name 'UUT_RESULT '.
  Number :-2147217900
  NativeError: 208
  SQLState: 42S 02
  Reported by: Microsoft SQL Server Native Client 10.0Description: instructions could not be prepared.
  Number :-2147217900
  NativeError: 8180
  SQLState: 42000
  Reported by: Microsoft SQL Server Native Client 10.0

  Source: TSDBLog

  Error :-2147217900; User-defined error code.

  Location: Step 'The journal of database results' sequence "Log to database" in "Database.seq".

  ------------------------------------------------

  Is there something I missed in configuration?

  Thank you.

  Peggy

  PeggyC,

  You can get detailed information on how to connect to the database in the TestStand reference manuall, they talk about script files SQL on page 113 which is consistent at 6-6.  Hope that helps.

• ORA-38760: this database instance could not turn on the flashback database

  Hi guys

  Without knowing it, I deleted all the files in the directory "fra" where "flashback" folder was there because of space issues. Now am facing the error.

  It's my practical database.

  SQL > startup

  ORACLE instance started.

  Total System Global Area 608174080 bytes

  Bytes of size 1268920 fixed

  184550216 variable size bytes

  419430400 of database buffers bytes

  Redo buffers 2924544 bytes

  Mounted database.

  ORA-38760: this database instance could not turn on the flashback database

  SQL > alter flashback database;

  Database altered.

  SQL > SHUTDOWN IMMEDIATE;

  SQL > STARTUP MOUNT

  SQL > select flashback_on from database v$.

  FLASHBACK_ON

  ------------------

  RESTORE POINT ONLY

  Am not able to extinguish the flame.

  SQL > select * from v$ restore_point;

  Select * from v$ restore_point

  *

  ERROR on line 1:

  ORA-38701: Flashback database log 1 seq 1 wire 1:

  "/ u01/fra/PROD/flashback/o1_mf_c1l5n3lq_.flb".

  ORA-27037: unable to get file status

  Linux error: 2: no such file or directory

  Additional information: 3

  SQL > select flashback_on from database v$.

  FLASHBACK_ON

  ------------------

  RESTORE POINT ONLY

  I also tried to create a restore point and the flashback there.

  SQL > create restore point guaranteed rspt1 flashback database;

  Restore point created.

  SQL > select * from v$ restore_point;

  Select * from v$ restore_point

  *

  ERROR on line 1:

  ORA-00600: internal error code, arguments: [krfg_get_keepseq_2], [1], [0], [0,]

  [0], [0], [0], [0]

  SQL > flashback data base to restore the point rspt1;

  flashback database to restore the rspt1 point

  *

  ERROR on line 1:

  ORA-38729: not enough flashback database data log do FLASHBACK.

  Please help me in this regard.

  Alert log file:

  ...

  ...

  ...

  Mount of redo thread 1, with mount id 305866694

  Sun 25 Oct 10:13:30 2015

  Bytes allocated of 3981204 in the shared pool for the generation of flashback buffer

  From RVWR background process

  RVWR started with pid = 16, OS id = 15486

  Sun 25 Oct 10:13:30 2015

  Errors in the /u01/oracle/product/10.2.0/db_1/admin/prod/bdump/prod_rvwr_15486.trc file:

  ORA-38701: seq 11 Flashback database log 11 1 thread: ' / u01/fra/PROD/flashback/o1_mf_c2oqypb0_.flb '.

  ORA-27037: unable to get file status

  Linux error: 2: no such file or directory

  Additional information: 3

  Sun 25 Oct 10:13:30 2015

  Database mounted in exclusive Mode

  Completed: ALTER DATABASE MOUNT

  Sun 25 Oct 10:13:30 2015

  ALTER DATABASE OPEN

  ORA-38760 marked during: ALTER DATABASE OPEN...

  Trace file:

  ...

  ...

  ...

  :() 2015-10-25 10:13:30.087 SERVICE NAME

  SESSION ID: (156.1) 2015-10-25 10:13:30.087

  ORA-38701: seq 11 Flashback database log 11 1 thread: ' / u01/fra/PROD/flashback/o1_mf_c2oqypb0_.flb '.

  ORA-27037: unable to get file status

  Linux error: 2: no such file or directory

  Additional information: 3

  1. Thank you.

  you experience serious problem... because usually it works very well.

  Try the following:

  SQL > shutdown immediate;

  SQL > bootable media.

  SQL > alter flashback database;

  SQL > alter database backup controlfile to trace noresetlogs;

  SQL > shutdown immediate;

  go to the trace file generated from running (alter database backup controlfile to trace noresetlogs ;)

  location of the trace file is located in the query:

  SQL > select * from v$ diag_info;

  It will contain the script with the SQL commands, you can rename the file with the extension ".sql".

  SQL > startup nomount

  SQL > @trace_file. SQL

  SQL > restore database;

  SQL > alter database open;

  So basically you are going to recreate the files of control again...

• Archive log list does not sequence of current journal is awaiting data base

  Hi all

  So I tried waiting for database physical installation in my local vm.

  I apparently completed the Setup, but I am facing a problem which I cannot understand.

  OS:-Linux 6.4

  Oracle version: 12.1.0.2.0

  primary data: ORA12C

  backup database: ORASTD

  Primary database: -.

  SYS@ORA12C > list of archive logs;

  Database log mode Archive Mode

  Active AutoArchive

  Archives/u01/app/oracle/fast_recovery_area of destination

  More old sequence of journal online 14

  Log sequence archive 16

  Current log sequence 16

  SYS@ORA12C > select name, database_role, controlfile_type from v$ database;

  NAME DATABASE_ROLE CONTROL

  --------- ---------------- -------

  ORA12C CURRENT PRIMARY

  SYS@ORA12C > select the sequence #, first_time, next_time, applied from v$ archived_log;

  FIRST_TIM SEQUENCE NEXT_TIME # APPLIED

  ---------- --------- --------- ---------

  4 15 APRIL 15 APRIL 16 15 NO.

  4 15 APRIL 15 APRIL 16, 15 YES

  5 16 APRIL 15 APRIL 16 15 NO.

  5 16 APRIL 15 APRIL 16, 15 YES

  6 16 APRIL 15 APRIL 16 15 NO.

  6 16 APRIL 15 APRIL 16, 15 YES

  7 16 APRIL 15 APRIL 16 15 NO.

  7 16 APRIL 15 APRIL 16, 15 YES

  8 16 APRIL 15 APRIL 16 15 NO.

  8 16 APRIL 15 APRIL 16, 15 YES

  9 16 APRIL 15 APRIL 16 15 NO.

  FIRST_TIM SEQUENCE NEXT_TIME # APPLIED

  ---------- --------- --------- ---------

  9-16 APRIL 15 APRIL 16, 15 YES

  10 16 APRIL 15 APRIL 16 15 NO.

  10-16 APRIL 15 APRIL 16, 15 YES

  11 16 APRIL 15 APRIL 16 15 NO.

  12 16 APRIL 15 APRIL 16 15 NO.

  13 16 APRIL 15 APRIL 16 15 NO.

  11-16 APRIL 15 APRIL 16, 15 YES

  12-16 APRIL 15 APRIL 16, 15 YES

  13-16 APRIL 15 APRIL 16, 15 YES

  14 16 APRIL 15 APRIL 16 15 NO.

  15 16 APRIL 15 APRIL 16 15 NO.

  FIRST_TIM SEQUENCE NEXT_TIME # APPLIED

  ---------- --------- --------- ---------

  15 16 APRIL 15 APRIL 16 15 NO.

  14-16 APRIL 15 APRIL 16, 15 YES

  Waiting for the database: -.

  Select the sequence #, first_time, next_time, applied from v$ archived_log;

  FIRST_TIM SEQUENCE NEXT_TIME # APPLIED

  ---------- --------- --------- ---------

  4 15 APRIL 15 APRIL 16, 15 YES

  5 16 APRIL 15 APRIL 16, 15 YES

  6 16 APRIL 15 APRIL 16, 15 YES

  7 16 APRIL 15 APRIL 16, 15 YES

  8 16 APRIL 15 APRIL 16, 15 YES

  9-16 APRIL 15 APRIL 16, 15 YES

  10-16 APRIL 15 APRIL 16, 15 YES

  11-16 APRIL 15 APRIL 16, 15 YES

  12-16 APRIL 15 APRIL 16, 15 YES

  13-16 APRIL 15 APRIL 16, 15 YES

  15 16 APRIL 15 16 APRIL 15 IN-MEMORY

  FIRST_TIM SEQUENCE NEXT_TIME # APPLIED

  ---------- --------- --------- ---------

  14-16 APRIL 15 APRIL 16, 15 YES

  SQL > select name, database_role, controlfile_type from v$ database;

  NAME DATABASE_ROLE CONTROL

  --------- ---------------- -------

  PHYSICAL STANDBY ORA12C STANDBY

  SQL > archive logs list;

  Database log mode Archive Mode

  Active AutoArchive

  Archives/u01/app/oracle/fast_recovery_area of destination

  More old sequence of journal online 0

  Log sequence to archive 0

  Current log sequence 0

  Now, here's my question is why the sequence of current newspapers shows 0. I tested the datas

  database primary eve happens again. I'm totally out of index.

  I tried to find it on google but nothing could help me until now.

  Any kind of help would be very appreciated.

  Thank you

  Soumya

  Hello

  I am also testing in Data Guard 12.1 configurations.

  Oracle support response is "it seems therefore that this information be removed.

  So is the best way to check on the database pending:

  SQL > select thread #, max(sequence#) ' last day before Seq received. "

  v $ archived_log val, v$ database vdb

  where val.resetlogs_change # = vdb.resetlogs_change #.

  Group thread # order by 1;

  SQL > select thread #, max(sequence#) ' last day before Seq applied.

  v $ archived_log val, v$ database vdb

  where val.resetlogs_change # = vdb.resetlogs_change #.

  and val.applied in ('YES', "IN-MEMORY")

  Group thread # order by 1;

  concerning

• Expectations extended on synchronization of log file while the parallel writing journal is fine

  We have 9.2.0.8 as experiences of long waits on the database log file sync (average waiting time = 46 ms) while waiting for the log file write Parallels is very good (average waiting time is less than 1 millisecond).

  The application is of type middleware, it connects to several other applications. A single user in a single application action train several requests to send back through this middleware, so he needs response time of db in milliseconds.

  The database is quite simple:

  -It has a few config tables that the application reads but rarely updated

  -She has table TRANSACTION_HISTORY: the application inserts records into this table using Insert einreihig (about 100 lines per second); each insert is followed by a validation.

  Records are kept for several months and then purged. The table has only column VARCHAR2/NUMBER/DATE, no LOBS, LONG, etc. The table has 4 non-unique single-column index.

  The average line length is 100 bytes.

  The load profile does not appear something unusual, the main figures: 110 transactions per second average transaction = 1.5 KB size.

  The data below are to 1 hour interval (purge wasn't running during this interval), physical reads or writes physical rate is low:

  Load profile

  ~ ~ ~ Per second per Transaction

  ---------------       ---------------

  Size: 160,164.75 1,448.42

  Logical reads: 521,58 57 675,25

  Block changes: 934,90 8.45

  Physical reads: 76,27 0.69

  Physical writings: 86,10 0.78

  Calls of the user: 491,69 4.45

  Analysis: 321,24 2.91

  Hard analysis: 0.09 0.00

  Kinds: 126.96 1.15

  Logons: 0.06 0.00

  Runs: 17.70 1 956,91

  Operations: 110,58

  
  

  Top 5 events are dominated by the synchronization of log file:

  Top 5 timed events

  ~~~~~~~~~~~~~~~~~~                                                     % Total

  Event expects Ela time (s)

  -------------------------------------------- ------------ ----------- --------

  401 608 18 448 59.94 file synchronization log

  db file parallel write 124 044 3 404 11.06

  CPU time                                                        3,097    10.06

  Enqueue 10 476 2 916 9.48

  DB file sequential read 261 947 2 435 7.91

  
  

  Section events:
  

  AVG

  Total wait wait wait

  Hour of wait time wait for the event (s) (ms) /txn

  ---------------------------- ------------ ---------- ---------- ------ --------

  Synchronize 0 401 608 46 18 448 1.0 file log

  db file parallel write 124 044 0 3 404 27 0.3

  Enqueue 10 476 277 2 916 278 0.0

  DB file sequential read 261 947 0 2 435 9 0.7

  buffer busy waits 11 467 67 173 15 0.0

  SQL * Net more data to the client 1 565 619 0 79 0 3.9

  lock row cache 2 800 0 52 18 0.0

  control file parallel write 1 294 0 45 35 0,0

  Log end of switch file 261 0 36 138 0.0

  latch free 2 087 1 446 24 12 0.0

  PL/SQL 1 1 20 19531 0,0 lock timer

  log file parallel write 0 143 739 17 0.4 0

  db file scattered read 1 644 0 17 10 0.0

  sequential log file read 636 0 8 13 0.0

  
  

  Log buffer is about 1.3 MB. We could increase the log buffer, but there is no log buffer space waits, so I doubt this will help.

  
  

  Newspapers in recovery have their own file systems, not shared with the data files. This explains the difference between waiting avg on parallel writing of log (less than 1 ms) file and db file parallel write (27 ms).

  Restoring logs is 100 MB, there are about 120 journal switches per day.

  
  

  What has changed: the pads/validations rate grew. Several months ago there were 25 inserts/validations per second in the TRANSACTION_HISTORY table, now get us 110 inserts/validation per second.

  
  

  What problem it causes application: due to slow down the reaction of the basis of the application (Java-based) requires discussions more and more.

  
  

  MOS documents on synchronization of log file (for example, 1376916,1 waits for troubleshooting "log file sync") recommend to compare the average waiting time on synchronization of log file and the log file parallel write.

  If the values are close (for example log file sync = 20 ms and log file parallel write = 10 ms) so expectations are caused by nits IO. However, it is not the case here.

  
  

  There was a bug (2669566) in 9.2 which resulted in underreporting lgwr parallel time of writing to the log file. I was talking about September 2005, during which the bug was present in 9.2.0.6, reported 10.1 fixed in: file parallel journal written (JL Comp) it is possible that your problem IS written to the log file.

  Concerning

  Jonathan Lewis