backup incremental update feature

Similar Questions

• Software update feature

  I have a Mac OS 10.6 and a C4450 all-in-one printer, would like to know how to stop the software update feature?

  I decided that I don't require additional updates

  Really?  You require no additional updates?  I don't think you can predict it.

  Are you eager to cut all software updates for the HP or all updates for your Mac?

• Backup incremental level 2,3,4 RMAN.

  Hello guys,.

  Can someone let me know what is the meaning / usage of incremental backup of level 2,3,4.

  Kind regards.

  Wikipedia can have an easy to understand the reason for several additional levels:

  A backup of level 1 decision Monday would only include changes made since Sunday. A backup of level 2 decision Tuesday would only include changes made since Monday. A backup of level 3 Wednesday would only include changes made since Tuesday. If a level 2 backup taken Thursday, it would include all changes made since Monday because Monday was the most recent level n - 1 backup.

  https://en.Wikipedia.org/wiki/Incremental_backup

  What is RMAN?

  The Oracle documentation refers only to additional 0 year 1. What I understand, RMAN never supported incremental backups multilevel at 3 levels (0-4). Apparently, people often had trouble understanding the logic for levels 3 and 4. RMAN provides the clause 'cascading', which also existed in 9i and is usually an easier to understand the backup strategy. RMAN also provides or has provided an option to perform incremental backups cumulative at level 1 or higher. For simplicity, the la documentation documentation since 10 g mentions only level 0 only a 1 an other levels is deprecated.

  BACKUP INCREMENTIELLE from LEVEL 1 (differential by default, changes made since the last level 0 or 1)

  BACKUP INCREMENTIELLE LEVEL 1 CUMULATIVE (changes since the last last level 0)

  Below some info from http://docs.oracle.com/cd/E11882_01/backup.112/e10643/obs_comm.htm#RCMRF910

  RMAN version 10.0.1: levels other than 0 and 1 are deprecated.

  If you make an additional workforce 1 backup, incremental or cumulative, and there is no previous level 0, RMAN automatically performs an incremental backup of level 0. BTW, a full backup is identical to a level 0 incremental backup but excluded a strategy of backup incremental level.

• What should I use the incremental update for IKM Oracle?

  IM missing some conditions before using the incremental update

  I have a simple integration of an Oracle source data source table directly to another target Oracle data source table

  I'd like to run a simple merger on a periodic basis

  Source table that contains data, the target table is created and set in the model

  Project created, put two tables in mapping

  LKM specified LKM SQL for Oracle (I won't use DBLink)

  Key of the target table are defined

  I go to set the IKM but cannot choose the incremental update option

  

  Probably a new problem user but can't seem to find a solution

  So it's a stale issue that I have long found the answer and thought I would post it here just in case someone else is beginner and difficulties as well

  For the incremental update be visible the following must be in place:

  (1) the IKM that supports the update must be visible in the project, as shown in the diagram above - that's what our ODI comes with

  (2) in the logical scheme of the model to the target in the context table to see its properties

  (3) set a goal on the properties of incremental update

  (4) now put target in context in the physical tab

  (5) IKM Oracle Fusion is now visible in the knowledge of integration Module

• Load with the incremental update of the IKM Oracle

  Hi Experts,

  According to my understanding, incremental load is that the new data (with insert append or load incremental (update/insert or merged with or without behavior SCD)).

  While peek into the code of the IKM KM here it is my understanding:

  Incremental update: given the PK defined in the target data store, that KM will check line by line and put insert/update (change data capture).

  Append: Blindly block for the table insertion target and any changed data will be captured. It will not truncate and insert so that to avoid duplicates in the data, have a PK defined in the target table, so whenever the duplicate data comes it can be prevented or go for CKMs.

  
  

  Now my doubt is,

  
  

  When you use the incremental km update: the scenario is I have an incremental load today, inserted for example 200000 files today and tomorrow other 200000 records added (may include updates of certain lines in the previous loaded 2,00,000 documents. Now it will scan 4,00,000 (yesterday + today) and seek changes, I mean to update or insert

  Because according to my understanding will treat this KM row by row all records (my understanding is correct?). If it reads each record and are looking for a change or not change it seems to me that his time and performance issues?  CDC is right approach in this scenario or the implementation of SDC on all columns in the table?

  
  

  Regarding the large number of records coming daily, updated incremental if IKM checking all records for update or insert or no change, in my opinion, this isn't a performance wise and time to compare source and target value. This KM eliminate comparing itself the Chronogram of the source to target those who does not charge any change in any of the previous column value?

  
  

  
  Sorry if this is a silly question. Just trying to figure which can be better charge strategy, especially when I have millions of records entering source daily.

  
  

  SSX I remember ealier JeromeFr our expert member in the community, said Partioned Exchange to treat only to process the given month data when you manage tables partitioned at the database level.
  

  
  

  Best regards

  ASP.
  

  
  

  
  

  
  

  
  

  
  

  
  

  
  

  Hi ASP_007,

  Charge, as opposed to full reload, does indeed only new (and possibly changed) data. There are 3 main ways to do this:

  • Set up a filter in your mapping interface/load only the data including the upper date to a variable (which holds the last loading date).
  • Use the framework of the CDC in ODI. There are several JKMs. The solution optimal is probably the Golden Gate, one, but you must purchase this additional license. mRainey wrote about this several times: http://www.rittmanmead.com/2014/05/goldengate-odi-perfect-match-12c-1/
  • Retrieve all the data in the source and allow an incremental update of the IKM define what already exists.

  Of course, the first two still will take a little more time to develop, but it will be much faster in terms of performance because you treat data less.

  It is for the part "Extract", get data from the source.

  Now, you must know how to "integrate" into your target. There are different strategies as insert Append, incremental update, Type 2 SCD...

  • Indeed insert Append won't update. It will only insert lines. It is a good approach for a full charge, or for an additional charge when you want to insert data. There is an option in most of the IKMs Append (control) to truncate the table before inserting (or delete all the lines if you do not have the privileges to truncate).
  • Incremental update: there are different IKMs for this and some may have better performance than others depending on your environment. I recommend you to try a few and see which is more fast for you. For example ' IKM Oracle incremental update (MERGE) "could be faster than 'IKM Oracle incremental update. I personally often use a slightly modified version of ' IKM Oracle incremental update (MERGE) for Exadata ' to avoid using a work table (I$ _) and perform the merge directly into the target table. The last approach works well with the CDC when you know that all data are new or changed and needs to be treated.
  • SCD2: To maintain your dimensions needing SCD2 behavior.

  So in answer to your questions:

  Because according to my understanding will treat this KM row by row all records (my understanding is correct?). If it reads each record and are looking for a change or not change it seems to me that his time and performance issues?

  Some of the IKMs will do it line by line, others will do it based on a game. This is why it is important to check what he does and he spots.

  CDC is right approach in this scenario or the implementation of SDC on all columns in the table?

  Yes certainly, you will have less data to be processed.

  Regarding the large number of records coming daily, updated incremental if IKM checking all records for update or insert or no change, in my opinion, this isn't a performance wise and time to compare source and target value. This KM eliminate comparing itself the Chronogram of the source to target those who does not charge any change in any of the previous column value?

  Yes, by using ' IKM Oracle incremental update (MERGE) for Exadata ' with the strategy of 'NONE '. This means that he will not try to see the rows from the source is already in the target.

  PS; I am remember ealier JeromeFr our expert member in the community, said Partioned Exchange to treat only to process the given month data when you manage tables partitioned at the database level.

  It is a good approach when you want to reload an entire partition (if you have a monthly charge in a monthly partition or a daily load in a daily score for example). It is easier to set up to load the new lines only. But if you need to update things in the source, you can use incremental update strategy on an intermediate table that contains the same data that your partition and then create the swap partition in a further step.

  Thanks for the mention.

  Be sure to close your other discussions.

  It will be useful.

  Kind regards

  JeromeFr

• Why the format does not work on my backup incremental level 1 the difference?

  I use following script to make a backup of difference incremental level 1.
  ------------------------------
  Run {}
  allocate channels ch1 disc type;
  Set order id "Monday_INC_1_df";
  backup as compressed backupset database incremental level 1
  format ' / opt/oracle/rman/bk/incr1_df_%d_%U'
  tag "Monday_INC_1_df".
  database more entered archivelog delete;
  output channel ch1.
  }
  -------------------------------

  According to this scenario, all the rman output files are placed in "/ opt/oracle/rman/bk / '...  And his label is 'Monday_INC_1_df'.

  But at the bottom of the newspaper, I find the archived log rman output files located in "/ opt/oracle/flash_recovery_area/TEST/backupset/2014_12_02 / '. And his label is "TAG20141202T210529".
  Also, at the bottom of the newspaper, the output data files rman files was partly on the right path "/ opt/oracle/rman/bk / '. And it seems that the data files has been supported for 2 times. The second time datafile rman output files was badly located in the path "/ opt/oracle/flash_recovery_area/TEST/backupset/2014_12_02 / '. And his label is "TAG20141202T215627".

  
  ================================================ Detail Log ===================================================================================
  channel ch1: compressed boot archived log backup set
  channel ch1: specifying the newspapers archived in the backup set
  archived journal entry thread = 1 sequence = 36706 RECID = 2773 STAMP = 865282473
  archived journal entry thread = 1 sequence = 36707 RECID = 2774 STAMP = 865282476
  archived journal entry thread = 1 sequence = 36708 RECID = 2775 STAMP = 865282478
  archived journal entry thread = 1 sequence = 36709 RECID = 2776 STAMP = 865282482
  archived journal thread = 1 sequence = 36710 entry RECID = STAMP 2777 = 865282486
  archived journal entry thread = 1 sequence = 36711 RECID = 2778 STAMP = 865285528
  channel ch1: starting piece 1 at 2014-12-02 21:56:20
  channel ch1: finished piece 1 at 2014-12-02 21:56:27
  piece handle=/opt/oracle/flash_recovery_area/TEST/backupset/2014_12_02/o1_mf_annnn_TAG20141202T210529_b7vk8408_.bkp tag = TAG20141202T210529 comment = NONE
  channel ch1: complete set of backups, time: 00:00:07
  channel ch1: archived deletion or newspapers
  name=/opt/oracle/flash_recovery_area/TEST/archivelog/2014_12_02/o1_mf_1_36706_b7vc98xc_.arc RECID archived log file = 2773 STAMP = 865282473
  name=/opt/oracle/flash_recovery_area/TEST/archivelog/2014_12_02/o1_mf_1_36707_b7vc9cz7_.arc RECID archived log file = 2774 STAMP = 865282476
  name=/opt/oracle/flash_recovery_area/TEST/archivelog/2014_12_02/o1_mf_1_36708_b7vc9gn1_.arc RECID archived log file = 2775 STAMP = 865282478
  name=/opt/oracle/flash_recovery_area/TEST/archivelog/2014_12_02/o1_mf_1_36709_b7vc9l3f_.arc RECID archived log file = 2776 STAMP = 865282482
  RECID = STAMP 2777 = 865282486 name=/opt/oracle/flash_recovery_area/TEST/archivelog/2014_12_02/o1_mf_1_36710_b7vc9plj_.arc archived log file
  name=/opt/oracle/flash_recovery_area/TEST/archivelog/2014_12_02/o1_mf_1_36711_b7vg8qz5_.arc RECID archived log file = 2778 STAMP = 865285528
  A backup over the 2014-12-02 21:56:27

  
  From backup 2014-12-02 21:56:27
  channel ch1: from compressed data file backup set incremental level 1
  channel ch1: specification-datafile in the backup set (s)
  Enter a number of file datafile = name=/opt/oracle/oradata/test/undotbs01.dbf 00003
  Enter a number of file datafile = 00005 name=/opt/oracle/oradata/test/hp01.dbf
  Enter a number of file datafile = name=/opt/oracle/oradata/test/cmtspace01.dbf 00035
  Enter a number of file datafile = name=/opt/oracle/oradata/test/sysaux01.dbf 00002
  Enter a number of file datafile = 00001 name=/opt/oracle/oradata/test/system01.dbf
  Enter a number of file datafile = name=/opt/oracle/oradata/test/users01.dbf 00004
  channel ch1: starting piece 1 at 2014-12-02 21:56:28
  channel ch1: finished piece 1 at 2014-12-02 23:48:06
  piece handle tag = / opt/oracle/rman/bk/incr1_df_TEST_1gpp6gcc_1_1 MONDAY_INC_1_DF = comment = NONE
  channel ch1: complete set of backups, time: 01:51:38
  channel ch1: from compressed data file backup set incremental level 1
  channel ch1: specification-datafile in the backup set (s)
  Enter a number of file datafile = name=/opt/oracle/oradata/test/undotbs01.dbf 00003
  Enter a number of file datafile = 00005 name=/opt/oracle/oradata/test/hp01.dbf
  Enter a number of file datafile = name=/opt/oracle/oradata/test/cmtspace01.dbf 00035
  Enter a number of file datafile = name=/opt/oracle/oradata/test/sysaux01.dbf 00002
  Enter a number of file datafile = 00001 name=/opt/oracle/oradata/test/system01.dbf
  Enter a number of file datafile = name=/opt/oracle/oradata/test/users01.dbf 00004
  channel ch1: starting piece 1 at 2014-12-02 23:48:07
  channel ch1: finished piece 1 at 2014-12-03 01:17:22
  piece handle=/opt/oracle/flash_recovery_area/TEST/backupset/2014_12_02/o1_mf_nnnd1_TAG20141202T215627_b7vqsq8y_.bkp tag = TAG20141202T215627 comment = NONE
  channel ch1: complete set of backups, time: 01:29:15
  A backup over the 2014-12-03 01:17:22

  From backup 2014-12-03 01:17:23
  Current archived log
  channel ch1: compressed boot archived log backup set
  channel ch1: specifying the newspapers archived in the backup set
  archived journal entry thread = 1 sequence = 36712 RECID = 2779 STAMP = 865288890
  archived journal entry thread = 1 sequence = 36713 RECID = 2780 STAMP = 865288965
  archived journal entry thread = 1 sequence = 36714 RECID = 2781 STAMP = 865288971
  channel ch1: starting piece 1 at 2014-12-03 01:17:24
  channel ch1: finished piece 1 at 2014-12-03 01:17:49
  piece handle=/opt/oracle/flash_recovery_area/TEST/backupset/2014_12_03/o1_mf_annnn_TAG20141203T011723_b7vx145d_.bkp tag = TAG20141203T011723 comment = NONE
  channel ch1: complete set of backups, time: 00:00:25
  channel ch1: archived deletion or newspapers
  name=/opt/oracle/flash_recovery_area/TEST/archivelog/2014_12_02/o1_mf_1_36712_b7vkkts7_.arc RECID archived log file = 2779 STAMP = 865288890
  name=/opt/oracle/flash_recovery_area/TEST/archivelog/2014_12_02/o1_mf_1_36713_b7vkn5v0_.arc RECID archived log file = 2780 STAMP = 865288965
  name=/opt/oracle/flash_recovery_area/TEST/archivelog/2014_12_02/o1_mf_1_36714_b7vkncv2_.arc RECID archived log file = 2781 STAMP = 865288971

  =============================================================================================================

  
  When I here is the backup incremental level 0, everything seems good.

  ----------------------------------
  Run {}
  allocate channels ch1 disc type;
  Set order id "Sunday_INC_0";
  backup incremental level 0 compressed backupset
  format ' / opt/oracle/rman/bk/incr0_%d_%U'
  tag "Sunday_INC_0".
  database more entered archivelog delete;
  output channel ch1.
  }
  --------------------------------

  But why my backup of difference in level 1 increntmental would not have good correct rman output path?

  
  

  You have specified the keyword 'database' twice in the backup command.  If execution is in the specified format, the second with the default format (to the FRA)

  Hemant K Collette

• Invisible signatures by the incremental update does not have any signs?

  Hi all!

  Sorry if here is not the right place for my question. If this is the case, is - can someone inform in the right place?

  Well, I am trying to build a tool to sign a PDF of incremental update using Delphi [].

  I am beginner in PDF structure and the brevity of the time made me perhaps a bit inattentive and desperate.

  After 2 months of hard fighting, I managed to get Acrobat Reader know my PDF (initially created by a java tool called SignServer) has a signature and recognize the certificate but it says that the certification is not valid because 'the byte range is invalid.

  I analyzed the PDF file in a hex editor and the byte range signature so any other object compensates seems correct.

  My PDF test is here: https://drive.Google.com/open?ID=0B0KKmaB-a0Z4SDZfNXdZRWZEem8

  Can you tell me what the problem with this PDF?

  Many thanks for any help!

  Specifications and PDF language

• Incremental update

  Hello

  Can someone tell me please detailed steps how to do an incremental update in ODI.

  I'm able to append, but incremental update gives expression error missing.

  Hello

  Perform the steps below to add expression -:

  1. go in the properties of the table target, and then in the target, you will see integration Type - select Incremetal update here.

  2. in the properties of the target table, under attributes , it has columns named Key, activate the option (check the attribute that you want to use in the expression).

  For example, if you check this option for the column, then A column will be used in expressions.

• difference between the incremental update of the IKM oracle and incremental update IKM oracle (PL - SQL)

  Hello

  What is the difference between the incremental update of the IKM oracle and updated incremental IKM oracle (PL - SQL) and incremental update IKM oracle (line by line).

  Thank you

  Papai

  The only difference is that the second using plsql for incremental update. He also to manage clob issues well enough.

  If you need to know more you can read the description of each KM section.

  Incremental update IKM Oracle (PL-SQL)

  -------

  Description:

  -Knowledge integration module

  -Integrates data into an Oracle table from target in incremental update mode using PL/SQL.

  -Non-existent rows are inserted. already existing lines are updated.

  -Data can be controlled. Data invalid are isolated in the error Table and can be recycled.

  -This KM uses PL/SQL to perform the inserts and updates until and blob columns are supported. Please see the restrictions.

  -When you use this module with a source table logged, it is possible to synchronize the deletions.

  Restrictions:

  -When working with the logged data, if the "synchronize destruction of the newspaper" are executed, the lines deleted on the target are engaged

  -The data are updated even if not changed (upgrade from any (e)

  -The number of lines (number of inserts/changes) is not available because the transactions are performed using PL/SQL

  -Comparison of the data is performed using the key to update defined in the interface. It must be set.

  -L'option TRUNCATE does not work if the target table is referenced by another table (foreign key)

  -Options FLOW_CONTROL, and STATIC_CONTROL call the Module knowledge check to isolate invalid data (if no CKM is defined, an error occurs). These two options should be set to NO in the case where an integration Interface meets a TEMPORARY target data store.

  -L' FLOW_TABLE_OPTION option is set by default to NOLOGGING. Set it to a space if the interface is running on an Oracle 7 database

  -Deletions are committed regardless of the VALIDATION option

  Incremental update of the IKM Oracle

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

  DESCRIPTION:

  -Integrates data into an Oracle table from target in incremental update mode.

  -Non-existent rows are inserted. already existing lines are updated.

  -Data can be controlled. Data invalid are isolated in the error Table and can be recycled.

  -When you use this module with a source table logged, it is possible to synchronize the deletions.

  REQUIREMENTS:

  -The update key defined in the interface is required.

  RESTRICTIONS:

  -When working with the logged data, if the "synchronize destruction of the newspaper" are executed, the lines deleted on the target are engaged

  -L'option TRUNCATE does not work if the target table is referenced by another table (foreign key)

  -Options FLOW_CONTROL, and STATIC_CONTROL call the Module knowledge check to isolate invalid data (if no CKM is defined, an error occurs).

  These two options should be set to NO in the case where an integration Interface meets a TEMPORARY target data store.

  -L' FLOW_TABLE_OPTION option is set by default to NOLOGGING. Set it to a space if the interface is running on an Oracle 7 database

  -Deletions are committed regardless of the VALIDATION option

  -L' ANALYZE_TARGET option will allow to assess correct statistics only if the VALIDATION is set to Yes. Otherwise, the IKM gather statistics based on old data.

  -Default UPDATE option is TRUE, which means by default it is assumed that there is at least one column nonkey specified in a target data store.

• Problem of incremental update in the treatment of updates

  Hi all

  I use ODI 11.1.1 and IKM Oracle incremental update. I have the following problem:

  My target table contains the following keys: NAME EVENT_ID, location_id and following columns, VALUE

  And assume that my source table that contains the same columns and a TIME column.

  If the source table is as follows:

  EVENT_ID LOCATION_ID NAME TIME VALUE

  1                1                   T         1          1

  1                1                   T         2          2

  If these two lines are loaded into the staging table I$ at the same time and the target table does not contain already the (1,1) combination for event_id, location_id 2 new lines will be inserted into the fact table. I would not have only one line with the most recent value. The problem is that step 'make update' will never report the second row as 'U' because there is no such thing as the combined update key in the fact table.

  How can I get the desired behavior?

  Hello

  There are several ways to do so. It depends mainly on your cardinality.

  APPROACH 1 (VERY BAD): filter

  • Place a filter with this text
    • EVENT_ID | LOCATION_ID | TIME IN (SELECT EVENT_ID |) LOCATION_ID | MAX (TIME) FROM TABLE GROUP BY EVENT_ID, LOCATION_ID)

  APPROACH 2: make a subselect interface

  1. create a temporary with interface
     • EVENT_ID = EVENT_ID
     • LOCATION_ID = LOCATION_ID
     • TIME = MAX (TIME)
  2. then put this interface (yellow) in your source and mark as subselect
  3. make an inner join

  METHOD 3: VIEW

  1. give an opinion on your db
     • create view MYVIEW AS SELECT EVENT_ID, location_id, MAX (TIME) FROM TABLE GROUP BY EVENT_ID location_id
  2. reversing the trend
  3. inner join

  APPROACH 4: FUNCTION ANALYTICS (good)

  Use an analytic function. See for example Re: 2nd Max Sal. But what will little customization.

  However, a variable is not a solution.

  Let us know

• Using synonym as data store can use Incremental update IKM

  Hello
  
  can we use synonymous as the target data store, if yes can we use incremental update IKM on it.
  
  Please let me know.
  
  Thank you

  Yes, you can use.
  Thank you

• Question about backups gradually updated

  Hello
  
  I have two issues with this script
  

  RUN
  {
    RECOVER COPY OF DATABASE 
      WITH TAG 'incr_update';
    BACKUP 
      INCREMENTAL LEVEL 1
      FOR RECOVER OF COPY WITH TAG 'incr_update'
      DATABASE;
  }

  taken from http://docs.oracle.com/cd/E11882_01/backup.112/e10642/rcmbckba.htm#CHDEHBFF here
  
  (1) no backup archivers? Otherwise, how to add them?
  (2) can this script be executed every day forever?
  
  Thank you.

  Hello;

  It archiveurs backup? N °

  Otherwise, how to add them? (not tested)

  RUN
  {
    RECOVER COPY OF DATABASE
      WITH TAG 'incr_update'
      UNTIL TIME 'SYSDATE - 7';
    BACKUP
      INCREMENTAL LEVEL 1
      FOR RECOVER OF COPY WITH TAG 'incr_update'
      DATABASE PLUS ARCHIVELOG;
  }
  

  This script be usable every day forever? I would go with something with a large capacity of recovery window.

  I like some like this:

  Sunday

  INCREMENTAL LEVEL 0 DATABASE BACKUP MORE ARCHIVELOG ALL;

  Rest of the week

  INCREMENTAL LEVEL 1 DATA BACKUP MORE ARCHIVELOG ALL;

  After that I configured my daily RMAN on a test system, I found as much chess as I could and tested them.

  Best regards

  mseberg

  Published by: mseberg on June 7, 2012 08:41

• Modify the incremental update of the IKM Oracle

  Hello
  
  For loads of our data, we use the IKM above (incremental update of Oracle).
  
  We found that it was running a bit slow on our target system (Oracle DB). And now we would like to change the IKM by setting the following ODI session setting, IE.
  
  ALTER session set "_always_semi_join" = "SELECT";
  
  who works wonders for the sql that the IKM runs and helps the programming of schedules.
  
  Question is how better add this line and merge into the existing IKM.
  
  In the "Details" section of the IKM, would I have a new order there?
  
  Thank you.

  >

  In the "Details" section of the IKM, would I have a new order there?

  Yes, you must add a new order in the detail section.

• Production signed PDF files without using the incremental updates

  Hello

  I know that it is possible to sign existing PDF files by adding new objects (such as the dictionary of the Signature) and gradually change catalogue, Pages objects etc accordingly. But what about programs produce the PDF file? I mean, they must also implement signature by adding a few items after the trailer and using incremantal update or they produce objects related to the signature inside the document directly and don't use incremental updates? Is there a difference? Is implemented with the help of incremental updates easier?

  This is a design decision of the software according to their needs and requirements surrounding the signing process.

• RMAN backup incremental level 1 without level 0-10 gr 2

  Hello
  
  I'm a bit confused after a few tries.
  
  The issue: It is mandatory to perform incremental level 0 backup before level 1 using Oracle 10 g 2?
  
  On Oracle 11.2 is mandatory, but I don't know about Oracle 10.2.0.5.
  
  11.2 test. If there is no level 0 rman take care on this subject and automatically perform level 0 before level1
  

  RMAN> backup incremental level 1 database;
  
  Starting backup at 29-APR-11
  using channel ORA_DISK_1
  no parent backup or copy of datafile 1 found
  ...
  channel ORA_DISK_1: starting incremental level 0 datafile backup set
  channel ORA_DISK_1: specifying datafile(s) in backup set
  ...

  Docs said:
  Incremental backups capture only blocks that change between backups in each data file.
  In a typical incremental backup strategy, an incremental backup of level 0 is used as a starting point. A level 0 backup captures all the blocks in the data file.
  
  So, on Oracle 10.2.0.5 happens not like 11.2:
  
  Parade level 1 without level 0 backup:

  RMAN> list backup;
  
  
  RMAN> backup incremental level 1 database;
  
  Starting backup at 29-APR-11
  using channel ORA_DISK_1
  channel ORA_DISK_1: starting incremental level 1 datafile backupset
  channel ORA_DISK_1: specifying datafile(s) in backupset
  input datafile fno=00001 name=+DG_ORCL/db10g/datafile/system.260.749756975
  ...
  channel ORA_DISK_1: starting piece 1 at 29-APR-11
  channel ORA_DISK_1: finished piece 1 at 29-APR-11
  piece handle=+DG_FRA/db10g/backupset/2011_04_29/nnndn1_tag20110429t190340_0.260.749761421 tag=TAG20110429T190340 comment=NONE
  channel ORA_DISK_1: backup set complete, elapsed time: 00:00:25
  channel ORA_DISK_1: starting incremental level 1 datafile backupset
  channel ORA_DISK_1: specifying datafile(s) in backupset
  including current control file in backupset
  including current SPFILE in backupset
  channel ORA_DISK_1: starting piece 1 at 29-APR-11
  channel ORA_DISK_1: finished piece 1 at 29-APR-11
  piece handle=+DG_FRA/db10g/backupset/2011_04_29/ncsnn1_tag20110429t190340_0.262.749761449 tag=TAG20110429T190340 comment=NONE
  channel ORA_DISK_1: backup set complete, elapsed time: 00:00:05
  Finished backup at 29-APR-11
  
  RMAN> backup archivelog all delete input;
  
  Starting backup at 29-APR-11
  current log archived
  using channel ORA_DISK_1
  ...
  Finished backup at 29-APR-11
  
  RMAN> shutdown abort;
  Oracle instance shut down

  Delete all (except SPFILE) ASM files.
  

  ASMCMD> cd +DG_ORCL/DB10g
  ASMCMD> ls
  PARAMETERFILE/
  spfiledb10g.ora

  So we will perfom a database restore
  

  oracle@butao:/home/oracle> rman target /
  Recovery Manager: Release 10.2.0.5.0 - Production on Fri Apr 29 19:06:52 2011
  Copyright (c) 1982, 2007, Oracle.  All rights reserved.
  connected to target database (not started)
  RMAN> startup nomount
  Oracle instance started
  
  Total System Global Area     293601280 bytes
  Fixed Size                     2095872 bytes
  Variable Size                 92275968 bytes
  Database Buffers             192937984 bytes
  Redo Buffers                   6291456 bytes
  
  
  RMAN> restore controlfile from '+DG_FRA/db10g/backupset/2011_04_29/ncsnn1_tag20110429t190340_0.262.749761449';
  
  Starting restore at 29-APR-11
  using channel ORA_DISK_1
  
  channel ORA_DISK_1: restoring control file
  channel ORA_DISK_1: restore complete, elapsed time: 00:00:05
  output filename=+DG_ORCL/db10g/controlfile/current.263.749761699
  output filename=+DG_FRA/db10g/controlfile/current.263.749761699
  Finished restore at 29-APR-11
  
  RMAN> startup mount
  
  database is already started
  database mounted
  released channel: ORA_DISK_1
  
  RMAN> restore database;
  
  Starting restore at 29-APR-11
  Starting implicit crosscheck backup at 29-APR-11
  allocated channel: ORA_DISK_1
  channel ORA_DISK_1: sid=156 devtype=DISK
  Crosschecked 1 objects
  Finished implicit crosscheck backup at 29-APR-11
  
  Starting implicit crosscheck copy at 29-APR-11
  using channel ORA_DISK_1
  Finished implicit crosscheck copy at 29-APR-11
  
  searching for all files in the recovery area
  cataloging files...
  cataloging done
  
  List of Cataloged Files
  =======================
  File Name: +dg_fra/DB10G/BACKUPSET/2011_04_29/ncsnn1_TAG20110429T190340_0.262.749761449
  File Name: +dg_fra/DB10G/BACKUPSET/2011_04_29/annnf0_TAG20110429T190442_0.264.749761485
  
  using channel ORA_DISK_1
  
  channel ORA_DISK_1: starting datafile backupset restore
  channel ORA_DISK_1: specifying datafile(s) to restore from backup set
  restoring datafile 00001 to +DG_ORCL/db10g/datafile/system.260.749756975
  restoring datafile 00002 to +DG_ORCL/db10g/datafile/undotbs1.261.749757085
  restoring datafile 00003 to +DG_ORCL/db10g/datafile/sysaux.262.749757095
  restoring datafile 00004 to +DG_ORCL/db10g/datafile/users.264.749757107
  channel ORA_DISK_1: reading from backup piece +DG_FRA/db10g/backupset/2011_04_29/nnndn1_tag20110429t190340_0.260.749761421
  channel ORA_DISK_1: restored backup piece 1
  piece handle=+DG_FRA/db10g/backupset/2011_04_29/nnndn1_tag20110429t190340_0.260.749761421 tag=TAG20110429T190340
  channel ORA_DISK_1: restore complete, elapsed time: 00:00:26
  Finished restore at 29-APR-11
  
  RMAN> recover database;
  
  Starting recover at 29-APR-11
  using channel ORA_DISK_1
  
  starting media recovery
  
  archive log thread 1 sequence 27 is already on disk as file +DG_FRA/db10g/onlinelog/group_3.259.749756971
  archive log thread 1 sequence 28 is already on disk as file +DG_FRA/db10g/onlinelog/group_1.257.749756963
  archive log filename=+DG_FRA/db10g/onlinelog/group_3.259.749756971 thread=1 sequence=27
  archive log filename=+DG_FRA/db10g/onlinelog/group_1.257.749756963 thread=1 sequence=28
  media recovery complete, elapsed time: 00:00:02
  Finished recover at 29-APR-11
  
  RMAN> alter database open resetlogs;
  
  database opened
  
  RMAN>

  See that I don't need level 0 backup to restore the level 1.
  
  Thank you
  Levi Pereira

  Levi,

  Please visit: http://download.oracle.com/docs/cd/B19306_01/backup.102/b14192/bkup004.htm

  If no level 0 backup is available, then the behavior depends on the setting of compatibility mode. If compatibility is > = 10.0.0, copies all blocks changed since the file was created RMAN and stores the results as a level 1 backup. In other words, the CPN at the time the incremental backup is taken is the creation of SCN file. If compatibility<10.0.0, rman="" generates="" a="" level="" 0="" backup="" of="" the="" file="" contents="" at="" the="" time="" of="" the="" backup,="" to="" be="" consistent="" with="" the="" behavior="" in="" previous="">

  Same info is here:

  http://www.Pythian.com/news/287/level-1-incremental-backup-in-Oracle-10G/

  Concerning

  Grosbois