test backup Oracle

Similar Questions

• Backup Oracle DB fails with errors.

  Hi all
  
  My backup oracled via Netbackup fail with error 6 (Netbackup) and I have checked that the logs please find the log RMAN so file.
  
  Please help me to solve this error.
  
  
  Script /opt/rman_script/st72_oracle_full.sh
  = began may 28 Mar 15:27:46 SGT 2013 =.
  
  
  RMAN: / OraBase/V10203/bin/rman
  ORACLE_SID: ST72
  ORACLE_USER: oracle
  ORACLE_HOME: / OraBase/V10203
  
  NB_ORA_FULL: 1
  NB_ORA_INCR: 0
  NB_ORA_CINC: 0
  NB_ORA_SERV: zsswmasb
  NB_ORA_POLICY: bsswst72_oracle_daily
  
  Full backup asked
  Sun Microsystems Inc.     SunOS 5.10 generic January 2005
  You have new messages.
  RMAN > 2 > 3 > 4 > 5 > 6 > 7 > 8 > 9 > 10 > 11 > 12 > 13 > 14 > 15 > 16 > 17 > 18 > 19 > 20 > 21 > 22 > 23 > 24 > 25 > 26 > 27 > 28 > 29 > 30 > 31 > 32 > 33 > RMAN > ED.
  
  connected to target database: ST72 (DBID = 990954548)
  connected to the recovery catalog database
  
  RMAN > 2 > 3 > 4 > 5 > 6 > 7 > 8 > 9 > 10 > 11 > 12 > 13 > 14 > 15 > 16 > 17 > 18 > 19 > 20 > 21 > 22 > 23 > 24 > 25 > 26 > 27 > 28 > 29 > 30 > 31 > 32 > 33 >
  allocated channel: ch00
  channel ch00: sid = 149 devtype = SBT_TAPE
  channel ch00: VERITAS NetBackup for Oracle - version 6.0 (2007070504)
  
  command sent on the channel: ch00
  
  From backup may 28, 13
  output channel: ch00
  RMAN-00571: ===========================================================
  RMAN-00569: = ERROR MESSAGE STACK FOLLOWS =.
  RMAN-00571: ===========================================================
  RMAN-03002: failure of the backup to 28/05/2013 15:27:51 command
  RMAN-03014: implicit recovery catalog Resync failed
  RMAN-06004: error ORACLE recovery catalog database: RMAN-20035: invalid high recid
  
  RMAN >
  
  Complete recovery manager.
  
  Script /opt/rman_script/st72_oracle_full.sh
  = ended by mistake may Mar 28 15:27:51 SGT 2013 =.
  
  =========================================================== ==================================================
  
  
  
  
  
  Script /opt/rman_script/st72_oracle_full.sh
  = began may 28 Mar 15:24:28 SGT 2013 =.
  
  
  RMAN: / OraBase/V10203/bin/rman
  ORACLE_SID: ST72
  ORACLE_USER: oracle
  ORACLE_HOME: / OraBase/V10203
  
  NB_ORA_FULL: 1
  NB_ORA_INCR: 0
  NB_ORA_CINC: 0
  NB_ORA_SERV: zsswmasb
  NB_ORA_POLICY: bsswst72_oracle_daily
  
  Full backup asked
  Sun Microsystems Inc.     SunOS 5.10 generic January 2005
  You have new messages.
  RMAN > 2 > 3 > 4 > 5 > 6 > 7 > 8 > 9 > 10 > 11 > 12 > 13 > 14 > 15 > 16 > 17 > 18 > 19 > 20 > 21 > 22 > 23 > 24 > 25 > 26 > 27 > 28 > 29 > 30 > 31 > 32 > 33 > RMAN > ED.
  
  connected to target database: ST72 (DBID = 990954548)
  connected to the recovery catalog database
  
  RMAN > 2 > 3 > 4 > 5 > 6 > 7 > 8 > 9 > 10 > 11 > 12 > 13 > 14 > 15 > 16 > 17 > 18 > 19 > 20 > 21 > 22 > 23 > 24 > 25 > 26 > 27 > 28 > 29 > 30 > 31 > 32 > 33 >
  allocated channel: ch00
  channel ch00: sid = 143 devtype = SBT_TAPE
  channel ch00: VERITAS NetBackup for Oracle - version 6.0 (2007070504)
  
  command sent on the channel: ch00
  
  From backup may 28, 13
  output channel: ch00
  RMAN-00571: ===========================================================
  RMAN-00569: = ERROR MESSAGE STACK FOLLOWS =.
  RMAN-00571: ===========================================================
  RMAN-03002: failure of the backup to 28/05/2013 15:24:34 command
  RMAN-03014: implicit recovery catalog Resync failed
  RMAN-06004: error ORACLE recovery catalog database: RMAN-20035: invalid high recid
  
  RMAN >
  
  Complete recovery manager.
  
  Script /opt/rman_script/st72_oracle_full.sh
  = ended by mistake may Mar 28 15:24:34 SGT 2013 =.
  
  
  =========================================================== ===================================================
  
  
  
  
  
  
  Script /opt/rman_script/st72_oracle_full.sh
  = began may 28 Mar 14:59:30 SGT 2013 =.
  
  
  RMAN: / OraBase/V10203/bin/rman
  ORACLE_SID: ST72
  ORACLE_USER: oracle
  ORACLE_HOME: / OraBase/V10203
  
  NB_ORA_FULL: 1
  NB_ORA_INCR: 0
  NB_ORA_CINC: 0
  NB_ORA_SERV: zsswmasb
  NB_ORA_POLICY: bsswst72_oracle_daily
  
  Full backup asked
  Sun Microsystems Inc.     SunOS 5.10 generic January 2005
  You have new messages.
  RMAN > 2 > 3 > 4 > 5 > 6 > 7 > 8 > 9 > 10 > 11 > 12 > 13 > 14 > 15 > 16 > 17 > 18 > 19 > 20 > 21 > 22 > 23 > 24 > 25 > 26 > 27 > 28 > 29 > 30 > 31 > 32 > 33 > RMAN > ED.
  
  connected to target database: ST72 (DBID = 990954548)
  connected to the recovery catalog database
  
  RMAN > 2 > 3 > 4 > 5 > 6 > 7 > 8 > 9 > 10 > 11 > 12 > 13 > 14 > 15 > 16 > 17 > 18 > 19 > 20 > 21 > 22 > 23 > 24 > 25 > 26 > 27 > 28 > 29 > 30 > 31 > 32 > 33 >
  allocated channel: ch00
  channel ch00: sid = 136 devtype = SBT_TAPE
  channel ch00: VERITAS NetBackup for Oracle - version 6.0 (2007070504)
  
  command sent on the channel: ch00
  
  From backup may 28, 13
  output channel: ch00
  RMAN-00571: ===========================================================
  RMAN-00569: = ERROR MESSAGE STACK FOLLOWS =.
  RMAN-00571: ===========================================================
  RMAN-03002: failure of the backup to 28/05/2013-14:59:36 command
  RMAN-03014: implicit recovery catalog Resync failed
  RMAN-06004: error ORACLE recovery catalog database: RMAN-20035: invalid high recid
  
  RMAN >
  
  Complete recovery manager.
  
  Script /opt/rman_script/st72_oracle_full.sh
  = ended by mistake may Mar 28 14:59:36 SGT 2013 =.
  
  
  
  ===============================================================================================================
  
  
  
  Thank you
  Noelie

  Hello;

  Note on this here:

  http://pythianpang.WordPress.com/2007/11/20/RMAN-20035-invalid-high-RecId/

  Best regards

  mseberg

• Backup Oracle using AIX cmds

  Hi all
  
  I need a clarification about the Oracle DB hope that you guys can shed some light on my question.
  
  One of my clients wants to take backup of oracle with AIX - OS commands (file oradata) data file path, when the database is running, but
  database is in fact planned in RMAN backup regularly.
  
  Still images that he insists to make a backup of the folder the database oradata without even stopping. I told tat backup would be incomplete, incorrect and
  It cannot be used to restore but customer said that he had visited there previously with tat type backup so he needs a backup.
  
  Can you please clarify for me, if I take a backup of the oradata folder using AIX commands such as tar and gunzip when the database is running and db is in log mode archive.
  I was able to restore will be or is the backup will be valid.
  
  Thanks in advance...
  
  Nonuday

  Nonuday wrote:

  Can you tell me if I take a backup of the data files and archive the log while DB is running can tat be restored and recovered.

  Thank you
  Nonuday

  The answer is NO, NO, a thousand times NO!

  Unless you use the backup software is 'Oracle abreast' (in which case under the covers it is in fact a wrapper for RMAN or a managed user backup hot) you trash, for the reasons that you have already explained to your customer.

  I think you should show your client on a test... computer with his boss of bosses present bosses.

  Hold him (or another job!)

  Good luck

  Dave

• Test for oracle RAC workload. Ideas?

  Hello world
  
  I'm looking for somethnig who test the database with big load
  Any ideas how to achieve this. I would like to give my customers something
  They can count it.
  
  Thanks in advance.
  
  
  Best regards
  Mohd

  Hi Mohd,

  I always use Swingbench.
  Swingbench is a test of generator (benchmarks) designed to highlight the load free Oracle (9i, 10 g, 11 g) data.
  SwingBench consists of a load generator, a coordinator and an overview of the cluster. The software allows a charge to be generated and transactions/response times to be traced.
  Swingbench can be used to demonstrate and test technologies such as Real Application Clusters, rebuilt table online, Eve, backup recovery etc. and online databases.

  http://www.dominicgiles.com/swingbench.html

  Kind regards
  Levi Pereira

• 'The applet bad class name' exception when you try to save functional forms/EBS test with Oracle Open Script 12.2.0.1 in JRE 1.6.0_45

  Hello

  I have a problem when you try to save functional forms/EBS test with OpenScript in 12.2.0.1 and also 12.1.0.2 version.

  The problem has been verified in two environments:

  1. Windows XP 32-bit, Internet Explorer 7, 1.6.0_45 JRE.

  2. Windows 7 32 bit, Internet Explorer 8, 1.6.0_45 JRE.

  
  The exception that I can see in the Java Console is:

  ...

  SSV dialogue is removed

  Loading of ATS Oracle Forms Module functional test [5.5] [JRE_1.6.0_45 Oracle]

  Temporary directory FormsFT Helper: * \Temp\OpenScript_FormsFT

  ...

  Loading cached forms pots...

  ....

  oracle.forms.engine.Main cracked

  Base: exception: the applet Bad class name.

  exception: the applet Bad class name.

  java.lang.SecurityException: applet Bad class name

  at sun.plugin2.applet.Plugin2Manager.createApplet (unknown Source)

  to sun.plugin2.applet.Plugin2Manager$ AppletExecutionRunnable.run (unknown Source)

  at java.lang.Thread.run (unknown Source)

  Exception: java.lang.SecurityException: applet Bad class name

  I think that I already tried all possible combinations with options and preferences of OpenScript.

  I also reinstalled several times to JRE.

  What is interesting is that everything works fine when I use the 1.6.0_38 JRE. However, all versions up to it give me the above exception.

  Could you please give me some advice what I can do or check? Maybe you can provide me with some information about the things that have changed in JRE 1.6 update 38 which can affect the load java applets by OpenScript?

  Everything I could find in the release notes was that some security problems have been repaired.

  Thanks in advance for any help.

  Because of security changes introduced in JRE, use NO build longer than 38 for JRE 6 (1.6.0_38) and does not exceed the 11 for JRE (1.7.0_11) 7.

  REDA

• ShadowProtect backup Oracle databases

  Hi all
  
  I would like to hear from someone if you use ShadowProtect to backup your Oracle databases.
  What are your experiences and the advantages and disadvantages in using ShadowProtect like the one and only tools to backup and restore for Oracle databases?
  Can the potential of ShadowProtect replaces it commonly use Oracle backup/restore tools?
  
  The reason why I bring this up is because my boss has the idea that ShadowProtect is THE best backup and restore tools. Personally I don't think, because there may be cases where we only need to recover the database and not the whole OS. So need to hear more from you guys.
  
  
  Thank you.
  
  Published by: moslee on March 22, 2013 02:24

  moslee wrote:
  Hi all

  I would like to hear from someone if you use ShadowProtect to backup your Oracle databases.
  What are your experiences and the advantages and disadvantages in using ShadowProtect like the one and only tools to backup and restore for Oracle databases?
  Can the potential of ShadowProtect replaces it commonly use Oracle backup/restore tools?

  The reason why I bring this up is because my boss has the idea that ShadowProtect is THE best backup and restore tools. Personally I don't think, because there may be cases where we only need to recover the database and not the whole OS. So need to hear more from you guys.

  Thank you.

  Published by: moslee on March 22, 2013 02:24

  Given the complexity of the backup and recovery of an Oracle database, I wouldn't trust anything, but the solution of the Oracle, rman. BTW, you already paid for rman. BTW, I've been known to quietly to apply my own backups rman, then let the solution of "business" save my rman backups. When feces enter the ventilation system, your final responsibility go/no-go is to recover this database. And 'they' (management, storage management team, anyone) will never accept that their backup solution has been insufficient or fatally defective.

  Published by: EdStevens on March 22, 2013 06:18

• Test the oracle insert very very slow

  I just created a 11.2.0.3 db based on data warehouse model and tested 10 000 000 insert records through sysbench and took 1 hour.
  
  IT is extremely slow. I did not tune the db parameters still, I guess that the parameters are good for datawarehouse.
  
  Operating system is linux redhat.
  
  Anyone has any advice on this?
  Thanks in advance.

  If the tool generates separate SQL statements that do not bind variables, it is not terribly surprising that Oracle is much slower than MySQL (especially based on things like what engine you use for MySQL). It is a very bad way to test a data warehouse in general and it is a very bad way to design an application that interacts with a database of Oracle in particular. You need to watch a Statspack / report AWR to be sure, but my guess is that most of the time was spent hard - analysis of 10 million separate queries.

  If you really want to optimize a type very non - DW workload that doesn't use bind variables, you can to some extent. But there will be a way of poor Oracle reference or a reference data warehouse loading process.

  Justin

• Backup Oracle database files

  Quick question... If I put the data base oracle in Archivelog mode, can I use such as Symantec or ARCServe Backup software to back up the directory where all database files to include archiveredo newspapers? It would be an acceptable means of backup of the database without using RMAN? Reason I ask is if incase we have a total breakdown of the system, I would like to rebuild the server, install Oracle with the same instance name and installation directory, and simply restore the database files and be operational. Thanks to produce for your help.

  974680 wrote:
  Quick question... If I put the data base oracle in Archivelog mode, can I use such as Symantec or ARCServe Backup software to back up the directory where all database files to include archiveredo newspapers? It would be an acceptable means of backup of the database without using RMAN? Reason I ask is if incase we have a total breakdown of the system, I would like to rebuild the server, install Oracle with the same instance name and installation directory, and simply restore the database files and be operational. Thanks to produce for your help.

  The right way to achieve your goal would be to use rman to back up the database, and then use other tools to backup the backups that have been created by rman.

• Error when trying to backup oracle rac to another location

  Hello
  
  I'm trying to backup of database to another location to a 11 GR Oracle RAC database version 2. Here is my script:
  
  #! / bin/ksh
  export ORACLE_SID = vvsms1
  ORACLE_BASE = / u01/app/oracle; export ORACLE_BASE
  ORACLE_HOME=$ORACLE_BASE/product/11.2.0/dbhome_1; Export ORACLE_HOME
  BASE_PATH = / usr/sbin: $PATH; export BASE_PATH
  PATH = $ORACLE_HOME/bin: $BASE_PATH; export PATH
  /U01/app/Oracle/product/11.2.0/dbhome_1/bin/RMAN target sys/viviet@vvsms journal /home/oracle/log_rman/vvsms.log append < < EOF
  RUN {}
  CROSS-CHECKING OF BACKUP;
  OVERLAP ARCHIVELOG ALL;
  FORMAT of DISC TYPE ALLOCATE CHANNEL channel 1 ' / home/oracle/backup/vvsms/backup_%U';
  TAG 'incr_vvsms' DATABASE INCREMENTAL LEVEL 0 BACKUP;
  ARCHIVELOG BACKUP ALL.
  REMOVE OBSOLETE;
  CHANNEL1 CHANNEL OUTPUT;
  }
  EXIT;
  EXPRESSIONS OF FOLKLORE
  
  I write it to a file .sh and set crontab to run it. But when he does I get error like this:
  
  From 22 October 12 backup
  channels channel 1: start the incremental level 0 datafile backup set
  channels channel 1: specification of the datafile (s) in the backup set
  Enter a number of file datafile = name=+DISK2/vvsms/datafile/sysaux.289.794242439 00002
  Enter a number of file datafile = name=+DISK2/vvsms/datafile/ts_service.dbf 00006
  Enter a number of file datafile = name=+DISK2/vvsms/datafile/ts_viviet.dbf 00007
  Enter a number of file datafile = name=+DISK2/vvsms/datafile/viviet.dbf 00008
  Enter a number of file datafile = name=+DISK2/vvsms/datafile/ts_vivietct_primary.dbf 00009
  Enter a number of file datafile = name=+DISK2/vvsms/datafile/undotbs1.290.794242445 00003
  Enter a number of file datafile = 00001 name=+DISK2/vvsms/datafile/system.288.794242429
  Enter a number of file datafile = name=+DISK2/vvsms/datafile/undotbs2.292.794242453 00004
  Enter a number of file datafile = 00005 name=+DISK2/vvsms/datafile/users.293.794242455
  channels channel 1: from room 1 to October 22, 12
  output channel: channel 1
  RMAN-00571: ===========================================================
  RMAN-00569: = ERROR MESSAGE STACK FOLLOWS =.
  RMAN-00571: ===========================================================
  RMAN-03009: failure of the backup command on the channel 1 channel at 22/10/2012 01:50:16
  ORA-19504: cannot create the file ' / home/oracle/backup/vvsms/backup_2anobqu5_1_1.
  ORA-27040: create file error, cannot create the file
  Linux-x86_64 error: 2: no such file or directory
  
  I don't know what I am wrong? The location is OK ("/ home/oracle/backup/vvsms /").
  
  Please suggest me some things about it. What should I do?
  
  Thanks in advance!
  
  P/s: sorry for my bad English.

  /u01/app/oracle/product/11.2.0/dbhome_1/bin/rman target sys/viviet@vvsms  log /home/oracle/log_rman/vvsms.log append
  

  This line may be your problem.
  As this database is a RAC and you are using a service with NLB 'vvsms' and ' / home/oracle/backup/vvsms ' is not a shared location. RMAN is starting a session on the other node where ' / home/oracle/backup/vvsms ' does not exist.

  Try changing:

  /u01/app/oracle/product/11.2.0/dbhome_1/bin/rman target sys/viviet@vvsms  log /home/oracle/log_rman/vvsms.log append
  

  This (using the easy connection method):

  /u01/app/oracle/product/11.2.0/dbhome_1/bin/rman target sys/viviet@localhost:1521/VVSMS  log /home/oracle/log_rman/vvsms.log append
  

  Where:
  localhost: your local node
  VVSMS: is at the service of database

  Also check if ' / u01/app/oracle/product/11.2.0/dbhome_1/network/admin/sqlnet.ora ' set up the line "NAMES. DIRECTORY_PATH = (TNSNAMES, EZCONNECT).

  PS: When RMAN start a session it shows where is recorded, control over the newspapers which RMAN instance has been registered.

  HTH,
  Levi Pereira

• issue backup Oracle

  Hello
  
  I do backups rman of the enterprise Web site manager and I use the below script that I found on http://www.oracle.com/technology/pub/articles/havewala-rman-grid.html
  
  device type disk tag 'TAG' % database;
  backup disc tag '% TAG' type device archivelog all not saved delete all entries;
  allocate channel for maintenance of type disk;
  cross-checking of backup;
  delete expired backup noprompt.
  delete noprompt obsolete device type disk;
  overlap archivelog all;
  delete noprompt expired archivelog all;
  output channel;
  
  Once the backup is complete, I take a look at the newspapers and it shows:
  
  (...)
  
  RMAN > overlap archivelog all;
  specification does not match any newspaper archive in the recovery catalog
  
  RMAN > delete noprompt expired archivelog all;
  specification does not match any newspaper archive in the recovery catalog
  
  (...)
  
  All the time.
  
  My question is: are the lines ' intersection archivelog all; "and" delete noprompt expired archivelog all; "really necessary?
  
  I use only FRA my backups and archives logs destination.

  Yes, you can ignore it. This means that there are no deleted archivelogs apart from RMAN. This is absolutely normal and ok.

  Werner

• Test the Oracle Instance on the smaller box (RAM problems)

  I got the task of taking an existing database of production and creation of a database of test on a much smaller box. The production box has 48 GB of RAM and eight CPUS (Itanium running HP - UX 11.23). The test box has four gigabytes of RAM and a processor. We had a process in place in the past to do this for Oracle 9i. All the data files, redo logs and Oracle 10g binaries are hosted on an HP EVA SAN, so we use snapshot or mirror of the cloning technology during the production on the test server instance.
  
  On Oracle 9i, we used to copy the data, delete the control files, then run a script to create new files of control for the test instance. I don't know if the old 9i writing work and have not yet tested this part. Mainly because I have problems to get the test instance to start reliably in the first place. When I copied all the data on, I changed the tnsnames.ora to reflect the host name of the new machine. I made a few adjustments in the file init.ora for production DB as well. At the beginning, I was able to start the database and everything seemed fine. But when I tried to stop him, he gave me the following error message:
  
  [07:32] [oracle@OracleC:~]$sqlplus "/ as sysdba"]
  
  SQL * more: Release 10.2.0.1.0 - Production Wed Jul 22 07:32:11 2009
  
  Copyright (c) 1982, 2005, Oracle. All rights reserved.
  
  Connect to an instance is idle.
  
  SQL > startup nomount;
  ORA-03113: end of file on communication channel
  
  
  This error is broad enough that I am certain that what he means is a bad server configuration of database for me probably a problem in the init.ora or SPFILE with a parameter that is calibrated for 48 GB box, not the box of 4 GB.
  
  Here's what I have in my init.ora file:
  
  ===========================================================================
  
  ###########################################
  IO and cache memory #.
  ###########################################
  DB_BLOCK_SIZE = 8192
  
  ###########################################
  # The library Cache and sliders
  ###########################################
  open_cursors = 300
  
  ###########################################
  # Diagnostic and statistical
  ###########################################
  background_dump_dest = / oracle10g/admin/prod/bdump
  core_dump_dest = / oracle10g/admin/prod/cdump
  TIMED_STATISTICS = TRUE
  user_dump_dest = / oracle10g/admin/prod/udump
  
  ###########################################
  # Distributed replication and snapshot
  ###########################################
  db_domain =""
  Remote_login_passwordfile = EXCLUSIVE lock
  
  ###########################################
  # Configuration file
  ###########################################
  control_files = ("/ oradata10g/prod/control02.ctl",)
  "/ oradata10g/prod/control03.ctl"
  "" / oradata10g/prod/control04.ctl ")"
  
  ###########################################
  # MTS
  ###########################################
  # Uncomment the following line when your listener is configured for SSL
  # (listener.ora and sqlnet.ora)
  dispatchers # = "(PROTOCOL = CST)(SER=MODOSE) ', ' (PROTOCOL = CST)(PRE=oracle.aurora.server.GiopServer) ', ' (PROTOCOL =)(PRE=oracle.aurora.server.SGiopServer) CST.
  "Dispatchers ="(Protocol=TCP) (SER = MODOSE) "," (PROTOCOL = TCP)(PRE=oracle.aurora.server.GiopServer) ","(PROTOCOL=TCP) (PRE = oracle.aurora.server.SGiopServer) "
  
  ###########################################
  # Various
  ###########################################
  compatible = 10.2.0
  db_name = prod
  
  ###########################################
  # Network registration
  ###########################################
  instance_name = prod
  
  ###########################################
  # Pools
  ###########################################
  JAVA_POOL_SIZE = 24 M
  #shared_pool_size = 1610612736
  # is moved to 128M by George Lenzer to the TEST instance on 06/07/2009
  shared_pool_size = 134217728
  
  ###########################################
  # Process and Sessions
  ###########################################
  process = 500
  
  ###########################################
  Recovery and # redo Log
  ###########################################
  fast_start_mttr_target = 300
  
  ###########################################
  # Sort, hash joins, index Bitmap
  ###########################################
  #sort_area_size = 5242880
  
  ###########################################
  # System managed Undo and Rollback Segments
  ###########################################
  #undo_retention = 7200
  UNDO_RETENTION = 14400
  UNDO_MANAGEMENT = AUTO
  undo_tablespace = UNDOTBS
  
  UTL_FILE_DIR = *.
  
  DB_FILES = 256
  
  # The value db_cache_size 128M for the TEST by G instance. Lenzer 07-02-2009
  
  db_cache_size = 128 M
  db_file_multiblock_read_count = 8
  
  #log_archive_dest = / ora_archlog
  #log_archive_format='arch_%r_%t_%s.arc'
  
  # the recommendation is 1 per UC
  
  db_writer_processes = 1
  
  sort_area_size = 10485760
  
  ===========================================================================
  
  There is also a little info. When I try to start the DB and it's just sitting there, my only way out is to kill the process only oracle executes:
  
  Oracle 4370 4365 247 09:27:18? 09:59 oracleprod (DESCRIPTION =(LOCAL=YES) (ADDRESS = (PROTOCOL = DOB)))
  
  Then I run "ipcrm". But if I run "swapinfo", it always indicates that most of the system's RAM is eaten up:
  
  [09:38] [root@OracleC:~]$swapinfo
  Ko Ko Ko START PCT / Kb
  TYPE AVAIL USED FREE USED LIMIT RESERVE PRI NAME
  dev 4194304 0 4194304 0% 0-1/dev/vg00/lvol2
  dev 4194304 0 4194304 0% 0-2/dev/vg00/lvol10
  reserve - 8388608-8388608
  memory 4182760 4070604 112156 97%
  
  I can't to free up RAM (which is usually about 20% used) after the start of Oracle attempt. The kernel parameters are defined for Oracle9i recommendations, so that they should not be too far from what is recommended for Oracle 10 g.
  
  I worked on it for about two or three weeks and I am almost of ideas. So any suggestions are welcome.
  
  Published by: deckard2 on July 22, 2009 07:05

  Here's a quick explanation: http://download.oracle.com/docs/cd/B19306_01/install.102/b14316/ofa.htm#sthref872
  (this is for Windows, but also OK for Unix)

  You can put your SPFILE or INIT files in the directory admin/pfile but Oracle looking at this place unless you create a Unix link in $ORACLE_HOME/dbs /.

• Various problems related to the tests for Oracle Applications R12 PreInstall HCVE

  Hello world
  
  I am preparing a Linux machine for installation of Oracle Applications R12 and currently I have several problems with the HCVE tests that are part of the ANR ([Doc ID 250262.1| https://metalink.oracle.com/metalink/plsql/f?p=130:14:4000869102561861013:p14_database_id, p14_docid, p14_show_header, p14_show_help, p14_black_frame, p14_font:NOT, 250262.1, 1, 0, 0, helvetica]). Here are the first two problems:
  
  1. when I run the RDA (. rda.pl t hcve) and select a test (in this case, "Oracle Database 11 g R1 (11.1.0) PreInstall (Linux AMD64)"), the script wonder of the JDK 'House '. Why the script asks if the installation process will install JDK? If I jump it (i.e. just press enter), the JDK test shows up as failed in the results. What should I put here?
  
  2. the tests 'process limit' and 'Limit descriptors' fail both:
  
  A00120 doesn't have a process boundary less 16384
  A00130 FAILURE of lower limit to 65536 descriptors
  
  I put these two values in /etc/security/limits.conf according to the instructions here: [Doc ID 761566.1| https://metalink2.oracle.com/metalink/plsql/f?p=130:14:4000869102561861013:p14_database_id, p14_docid, p14_show_header, p14_show_help, p14_black_frame, p14_font:NOT, 761566.1, 1, 1, 1, helvetica]
  
  Here are the lines I added to limits.conf:
  
  # To install Oracle (MetaLink Doc ID: 761566.1)
  * hard nofile 65535
  * soft nofile 4096
  * nproc 16384 hard
  * soft nproc 2047
  
  Why so this test fails?
  
  -Kyle

  Kyle,

  You can submit feedback and send an email to Support of Oracle ([email protected]). The link 'Submit comments' is available in most of the documents of Metalink GDR (Diagnostics).

  Kind regards
  Hussein

• JDBC Oracle driver loading tests... Failed:

  Test the Oracle Home located at C:\app\oracle\product\11.2.0\instantclient-basic-nt-11.2.0.4.0\instantclient_11_2

  Test client directory... Ok

  JDBC Oracle driver loading tests... Failed:

  C:\app\oracle\product\11.2.0\instantclient-basic-NT-11.2.0.4.0\instantclient_11_2\jdbc\lib\ojdbc6.jar is not a valid jar

  I downloaded and extracted the zip instantclient-jdbc-nt - 11.2.0.4.0 Oracle Technology Network, several times.  Yet, receive the same error weird test Oracle SQL Developer Version 4.0.1.14.  The ojdbc6.jar file was extracted in the instantclient_11_2, not the subdirectory lib subdirectory.  There is no jdbc or lib subdirectory.  Where's configuration is defined jdbc and lib subdirectories?  I need Oracle SQL Developer 4 to look at in the instantclient_11_2 subdirectory.

  Hello

  .. .is not a file valid jar

  This message is quite misleading. It is not due to a corrupted download.  In the case of database > advanced > use Oracle Client > Configure... you select

  Type of customer: Oracle Home

  Location of the client:

  and then click test... this error will occur.  Instead, you need to select

  : Customer Type instant

  Kind regards

  Gary

  SQL development team

• Backup in Oracle procedures

  I'm trying to backup procedures for a number of research platforms and software and I was wondering if someone could help with the Oracle database with the following text:
  
  • For backups of Oracle databases, which is the default tool or utility that is used to back up the Oracle database and Oracle backup what file extension will have when the backup is created? Also, is there a default folder that will be written the backup too if the use of default tools?
  
  • Is more typical, an Oracle database will be saved using the default backup utilities in Oracle or do most people use a 3rd third party tool?
  
  • Too far do you go with backups of databases? You back up the database to another drive, then the backup to the backup media, once again? Extent to which channel are you going?
  
  • How do you feel about the use of backup utilities system including databases oracle on system within the backup volumes is common or it is essential to have a specific to the Oracle database backup procedure?
  
  • What audit connects or trails will be available to demonstrate to management or the Board of Auditors that the backup was performed in accordance with departmental policies? Where is this file and how much detail does show when you use the default oracle utilities?
  
  • As Oracle DBA are there any particular reasons why backups cannot be run on a specific day, that evening, if so could you give some details on what types of occurrence will prevent to be logical to run the backup that night? For example, if you saw, that she has been saved 6 days out of 7, why for the 7th day would be reasonable, as appropriate.
  
  Would prefer your comments and input as opposed to a load of links that the links I've read to date are not compatible.

  user599292 wrote:
  But the most popular / typical of utlitity used worldwide Oracle RMAN backup database? Are there other popular alternatives or is it used RMAN one more folk to backup Oracle databases?

  There is a very good thing that you should not forget. All that you can choose to use, if it is not supported by oracle, its of no use at all. And RMAN is the tool supported by oracle for making backups and recoveries. Another thing, is your own responsibility. In addition to RMAN, BCV backup are also very popular.

  The repository that you mention, which will provide us with 3 weeks backup days a, b, c but not d and z?

  Please excuse my ignorance, but I'm new on this.

  Please see th link mentioned by Centinul and read the books mentioned. It has tons of info.

  Aman...

• Oracle hot backup confusion

  Hello world
  This is a confusion on hot-backup Oracle.
  
  Suppose I have 3 online Redo Logs and I put a tablespace in backup Mode.
  So, whatever changes are made, Oracle will begin to generate redo for the entire block to avoid the fractured blocks.
  
  My Question is if, while doing this all my online Redo Logs files/members of Wat is full how do then Oracle manage to move forward and continue to save Transactions/changes to a Certain block in the Tablespace.
  
  
  1 > oracle runs a recovery from the archived Redo Logs once the Tablespace out of backup Mode.
  2 > or y at - it a fitness State/Re-sizing of the redo log file online space between other Segments.
  
  Thank you and best regards,
  Jean-Jacques Mukherjee.

  >
  I'm even not sure what #2 means! Anways.

  Suppose I have 3 online Redo Logs and I put a tablespace in backup Mode.

  No matter how many redo log groups you have and as I know it has no relationship to a tablespace in backup mode either.

  So, whatever changes are made, Oracle will begin to generate redo for the entire block to avoid the fractured blocks.

  Partially correct! The entire block is copied only for the first time . For subsequent changes, only agents of change go in the stamp newspaper, and from there, the redo log files. This is exactly what happens to a tablespace is not not in backup mode.

  My Question is if, while doing this all my online Redo Logs files/members of Wat is full how do then Oracle manage to move forward and continue to save Transactions/changes to a Certain block in the Tablespace.

  Does that mean exactly? If you complete the redo logs, switch journal will follow making the LGWR move from the current group of redo log to another, triggering the control point, which would make the DBWR write buffers in the data file. I don't know what is confusing?

  Edit: I apologize, I read this point a little too fast! If all the log files will be filled, database could be suspended. There will be transactions more because there is no place to write their vectors of change anywhere. Sorry, I realized in the first time you were talking about switching to the standard newspaper.

  1 > oracle runs a recovery of archived newspapers again once the data file is underwritten in backup Mode.

  What recovery? Before files go in the backup, there is a 'global' checkpoint which is executed for them, pushing them all buffers in the cache for them and then once the point control freezed for them, still let all read/write operations. Once they come on the backup mode, when the next time a checkpoint is made, then their number is synchronized with the rest of the database. Where the recovery comes into the picture here?

  2 > or y at - it a fitness State/Re-sizing of the redo log file online space between other Segments.

  As I said before, I didn't at all mean that point yet?

  HTH
  Aman...

  Published by: Edit added Aman... on October 24, 2009 21:19