ALTER table

Similar Questions

• ALTER TABLE SHRINK the SPACE COMPACT takes a long time

  Hello
  I run:
  ALTER TABLE TOM. TAB1 SHRINK SPACE COMPACT;
  ALTER TABLE TOM. TAB1 SHRINK SPACE WATERFALL;

  
  The "ALTER TABLE TOM. Tab1 SHRINK SPACE COMPACT"was completed in 2 hours, but"ALTER TABLE TOM. Tab1 SHRINK SPACE WATERFALL"is still underway after 8 hours, less than the size of the table TAB1 and index:

  NOM_SEGMENT... GB
  TAB1................................. 166
  IDX01................................ 83
  SYS_C00......................... 42
  IDX02............................. 142
  IDX03.............................. 66
  IDX04............................. 122

  
  Why 'SHRINK the SPACE COMPACT' takes a lot of time?
  Is it possible to check how much time should I take?

  Thanks in advance!

  Running

  ALTER INDEX TOM. TAB1_IDX03 SHRINK SPACE COMPACT;

  ALTER INDEX TOM. TAB1_IDX03 SHRINK SPACE;

  set is useless, as does 'SHRINK the SPACE' that means "SHRINK SPACE COMPACT" and much more (see the quote from the Doc Oracle Hemant).  2, 'SHRINK the SPACE' is the operation that will take more time.  So if 'SHRINK the SPACE' is what you wanted (reclaim space) and you can afford to wait, run 'SHRINK the SPACE' only.

  "SHRINK the COMPACT SPACE" is a part of the work that 'SHRINK the SPACE' is only.  Generally, use you it when time is at a premium: you run SHRINK SPACE COMPACT a weekday evening (because SHRINK the SPACE does not fit in one night and accrue hourly) and you run SHRINK SPACE on the weekend to finish following the recovery of your space.

  Regarding the fact that the entire segment or a partition after another, I have no experience - I both of their time and test methods.  Perhaps running several sessions at the same time on different partitions would save time (but be more heavy on the CPU and storage).

  Also note that SHRINK the AREA generates a lot of REDO (keep an eye on the free space of your ARCHIVELOG destination).  With your data volume, I would expect dozens (or even hundreds) of gigabytes of roll forward must be generated.

• ALTER TABLE RETRACTABLE is possible online?

  RDBMS Version: 11.2.0.4

  Platform: Oracle Linux 6.4

  We have a database of CARS of Production 2 nodes very critical. To recover the space, we want to cut some large tables. The first table we'll shrink is a 300 GB, which is not partitioned with no LOB table.

  Here are the steps that we'll run.

  Step1.alter mytable table allow the movement of the line;

  Step2.alter mytable table shrink space compact;

  Step3.alter mytable table shrink space;

  Step4.alter table mytable disable the movement of the line;

  STEP 5. COLLECT STATISTICS of the table (because the RETRACTABLE table will cancel all the sliders in the shared pool)

  The steps above will take at least 10 hours to complete a 300 GB table. We would like to know if all of these steps can be done online. That is to say. We do not want the SHRINK command to lock the table and prevents the table to be modified (any DML) by the application

  Doc below said this can be done online. But, MOS DOC 1357878.1 said that it is not and you must use the DBMS_REDEFINITION package if you want to do it online.

  https://docs.Oracle.com/CD/E18283_01/server.112/e17120/schema003.htm

  The MOS article you are talking about is not correct. Outside the DOF at the end of a SHRINK lock when the high tide is shifted, SHRINK the SPACE is an online operation. DBMS_REDEFNITION is actually worse because you have DDL locks at both the beginning and the end. However, in both cases, it is the DDL that is blocked by other operations DML (and not the other way around) so that your users will have no problem. If the narrowing or the redefinition is blocked, you can suspend the database for a few seconds. I described how to do here,

  Database Quiesce

  --

  John Watson

  Oracle Certified Master s/n

• ALTER table move online in an encrypted tablespace

  Hello

  Our DBA want to put all our data in encrypted storage.

  Once they create encrypted tablespaces, we will have to do «alter table move online...» "orders for encrypted storage space around the tables.

  Has anyone here done already?   Keeps the table moved really 'on-line' (available at query/insert/update) during the move?  Or will we have to wait some time stop/downtime while the tables are moved?
  What about the tables with long or CRAFT?

  Thank you!

  KSandberg and to add to the recall of John that only ITO tables can be moved online, if the table needs to be moved offline the indexes would be invalid until rebuilt.  If the table is an IOT I suggest you rebuild the index online after the move because even if the indexes are always used the logical ROWID is more pointing to the correct location and access of the index will be less effective that will you rebuild the index.

  - -

  The dbms_redefinition package could be an alternative solution if you need to move the tables no - ITO online.

  - -

  HTH - Mark D Powell.

• ALTER TABLE hf_altertest ADD COLUMN does not

  I am using ORACLE 11g and my FDI is Developer SQL 4.1.1

  It's all free PC version that has limits.

  I tried on:

  CREATE TABLE hf_altertest

  (

  contact_id INTEGER NOT NULL

  , name VARCHAR (25)

  , last_name VARCHAR (35)

  PRIMARY KEY (contact_id)

  );

  ALTER TABLE hf_altertest State to add a CHAR (2) COLUMN;

  But when I go to run the ALTER TABLE statement then I get following error:

  rror starting at line: 1 323 in command.

  ALTER TABLE hf_altertest State to add a CHAR (2) COLUMN

  Error report-

  SQL error: ORA-00904: invalid identifier

  00904, 00000 - '% s: invalid identifier '.

  * Cause:

  * Action:

  I can run the same code in MySQL without error.

  This leads me to think that maybe I'm just bumping into a limit in the free version of PC.

  Any thoughts?

  3003916 wrote:

  I am using ORACLE 11g and my FDI is Developer SQL 4.1.1

  It's all free PC version that has limits.

  ALTER TABLE hf_altertest State to add a CHAR (2) COLUMN;

  But when I go to run the ALTER TABLE statement then I get following error:

  rror starting at line: 1 323 in command.

  ALTER TABLE hf_altertest State to add a CHAR (2) COLUMN

  Error report-

  SQL error: ORA-00904: invalid identifier

  00904, 00000 - '% s: invalid identifier '.

  * Cause:

  * Action:

  I can run the same code in MySQL without error.

  This leads me to think that maybe I'm just bumping into a limit in the free version of PC.

  Any thoughts?

  There is no LIMIT to the 'free' version  If you read the error message he said THAT was not a valid identifier.  Just to show you that I tried your erroneous statement

  ALTER TABLE hf_altertest ADD COLUMN state2 CHAR(2);
  

  and here is the indicator error message in SQL +.

  SQL * more: Production of liberation 12.1.0.2.0 Tue Oct 15 14:34:57 2015

  Copyright (c) 1982, 2014, Oracle.  All rights reserved.

  Connected to:

  Oracle Database 11 g Enterprise Edition Release 11.2.0.4.0 - 64 bit Production

  With the partitioning, Real Application Clusters, Automatic Storage Management, OLAP

  and Data Mining options

  Add a COLUMN state2 TANK (2)

  *

  ERROR on line 2:

  ORA-00904: invalid identifier

  As you can see (and as the previous poster of response indicates), the 'COLUMN' KEYWORD is not necessary when adding new columns.  Only when 'RENAME' is used, then you add the keyword COLUMN.

  See the documentation for the syntax of ALTER TABLE here--> http://docs.oracle.com/database/121/SQLRF/statements_3001.htm#SQLRF01001

  See specific clause RENAME syntax here--> http://docs.oracle.com/database/121/SQLRF/statements_3001.htm#i2183480

• Invalid procedure after alter table add < column >

  Hello Experts,

  I was faced with a situation where I add a column in a table that is used in a procedure, immediately after the addition of the new procedure associated with this table column becomes invalid.

  Can someone explain please know why this happens.

  Database version:-11.2.0.4.

  I was faced with a situation where I add a column in a table that is used in a procedure, immediately after the addition of the new procedure associated with this table column becomes invalid.

  Can someone explain please know why this happens.

  Yes - the Oracle documentation explains this. I found this doc with a simple search of "column, add the oracle 11g invalid procedure.

  http://docs.Oracle.com/CD/B28359_01/server.111/b28318/dependencies.htm#CHDJIIFC

  Table 6-2 , shows how objects are affected by changes to other objects upon which they depend.

  Table 6-2 operations affects this object status

  Operation	Where the status of dependent objects
  ALTERTABLEtableADDcolumn	INVALIDWhen: Uses of the object (except a view)-dependent SELECT * on table . Dependent object uses table %rowtype . Dependent object performs INSERT on table without specifying a list of columns. Dependent object references table in the query that contains a join in SQL. Dependent object references table in the query that refers to a PL/SQL variable. Otherwise, no change.

  Read the WHOLE ARTICLE and see if it does not have to explain your problem.

• Online syntax to create Foreign key with Alter Table statement

  I am a pretty meticulous reader. That being said, from the 4 texts that I read to start to learn SQL, I do not remember and do more that probably not, find a syntax online for the creation of a CF with the statement alter table.

  Just something I thought while I was reading.

  Is there a such syntax?

  And if so someone can give me direction on my code?

  ALTER table emp4 change employee_id key foreign employees of references (employe_id);

  It gives me the option error invalid alter table.

  Any idea would be appreciated.

  Thank you.

  rp0428 wrote:

  A foreign key is a constraint, you must use ADD CONSTRAINT.

  Thare is more than one way to skin a cat:

  SQL > create table tbl (id number, name varchar2 (10))
  2.

  Table created.

  SQL > alter table tbl adds (foreign key references emp (empno))
  2.

  Modified table.

  SQL > select constraint_name,.
  2 constraint_type
  3 from user_constraints
  4 where table_name = 'TBL '.
  5.

  CONSTRAINT_NAME C
  ------------------------------ -
  SYS_C0021920 R

  SQL > drop table tbl purge
  2.

  Deleted table.

  SQL > create table tbl (id number, name varchar2 (10))
  2.

  Table created.

  SQL > alter table tbl change (foreign key references emp (empno))
  2.

  Modified table.

  SQL > select constraint_name,.
  2 constraint_type
  3 from user_constraints
  4 where table_name = 'TBL '.
  5.

  CONSTRAINT_NAME C
  ------------------------------ -
  SYS_C0021921 R

  SQL >

  SY.

• alter table drop column

  Hi all

  I create a simple table like this:

  create table x (a number, b varchar2 (20));

  ALTER table x modify a primary key constraint a_pk;

  insert into x values (1, 'first row');

  insert into x values (2, 'second row');

  commit;

  Then:

  SQL > select * from x;

  A AND B

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

  1 first row

  2 second row

  Then I run these commands:

  SQL > alter table drop column x.

  Modified table.

  SQL > select * from x;

  B

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

  first row

  second row

  Question: Why am I able to remove this column when a PRIMARY KEY for the table? I do not use the key words of CONSTRAINT CASCADE with the DROP COLUMN clause...

  I have Mr. Steve O'Hearn here saying this in his book 'Certified Expert Guide exam 1Z0-047:

  CREATE TABLE ORDER_RETURNS

  (NUMBER OF ORDER_RETURN_ID,

  NUMBER OF CRUISE_ORDER_ID

  DATE OF CRUISE_ORDER_DATE,

  CONSTRAINT PK_OR PRIMARY KEY (ORDER_RETURN_ID));

  .. We can not remove the column the table ORDER_RETURNS that

  is subject to the constraint PRIMARY KEY (ORDER_RETURN_ID)... The reasons are:

  ORDER_RETURN_ID is the PRIMARY KEY of this table...

  Isn't this correct?

  No - it's NOT correct - the book is bad.

  You could have just tested the FULL example in the book to verify that the statement is wrong.

  http://weirdoootamee.yolasite.com/resources/Oracle%20sql%20expert.PDF

  See page 432

  CREATE TABLE CRUISE_ORDERS
  (NUMBER OF CRUISE_ORDER_ID,
  ORDER_DATE DATE,
  PK_CO (CRUISE_ORDER_ID, ORDER_DATE) PRIMARY KEY CONSTRAINT);

  CREATE TABLE ORDER_RETURNS
  (NUMBER OF ORDER_RETURN_ID,
  NUMBER OF CRUISE_ORDER_ID
  DATE OF CRUISE_ORDER_DATE,
  CONSTRAINT PK_OR PRIMARY KEY (ORDER_RETURN_ID).
  FOREIGN KEY CONSTRAINT FK_OR_CO
  (CRUISE_ORDER_ID, CRUISE_ORDER_DATE)
  REFERENCES CRUISE_ORDERS (CRUISE_ORDER_ID, ORDER_DATE));

  create two tables - add data, if you want.

  Then try to delete the primary key column and see what happends

  ALTER table drop column order_return_id order_returns;

  The best way to learn is to actuall TRY of THINGS.

  Try it with your own table, but the FIRST thing that you should try was the real example of the book.

• ALTER table add constraint vs alter table change

  Hello

  could someone explain why we have two different statements when you add a constraint on an existing table?

  If I need to add a primary key constraint to a column from an existing table that I use:

  ALTER table tab1 add the key primary constraint tab1_id_cst (id)

  But if I want to add a NOT NULL constraint, the same syntax does not work, I need to use alter table change instead

  Why?

  Thank you!

  Hello

  The constraint not null is a constraint inline, average, it applies to the column level. So in order to change anything at the column level you must use "ALTER TABLE CHANGE" only.

  Remaining constraint is forced to outline, which are part of the table definition. The table definition can be modified by "ALTER TABLE... ADD '.

  In the framework of the above mentioned constraints, those are the two columns level constraints and table except constraint NOT NULL value. Change column level, you must use the syntax

  and for the level of the table, you need to use the respective syntax.

  Kind regards

  Bigot

• ALTER table, if not amended

  Hi, thank you very much in advance. How to add columns to a table only the columns not yet existed.

  ALTER table

  table-name

  Add

  (

  column1_name column1_datatype column1_constraint,

  column2_name column2_datatype column2_constraint,

  column3_name column3_datatype column3_constraint

  );

  
  

  If column1_name does not exist I have add the column in table_name. Otherwise, do not add the column.

  
  

  Best regards, ib
  

  Here the case, I give my client a script and if they add columns or not, I don't know. So, I want only the columns are added to the case where they do not exist. The script includes other ddl and dml so I don't want a mistake due to the change table make the script error. Thank you.

  Sorry-, you will have to wait for someone else to help.

  Just, that makes NO sense to me. If you are unsure when columns are added to a table, then you have lost ALL control over your data model. You can possibly write, test, or deploy code that uses tables and columns without knowing which columns a table has.

  You should just use separate scripts for these tables of problem so that if the script didn't it won't affect other scripts.

• Invalid option to TRUNCATE PARTITION in ALTER TABLE

  Below edit sql works fine when I run this one.

  ALTER TABLE table_name TRUNCATE partition partition_nm UPDATE GLOBAL INDEXES;

  But if I run this procedure I get the error message "14054. 00000 - option ALTER TABLE TRUNCATE PARTITION invalid"

  create or replace PROCEDURE TRUNCATE_WS_CLAIMS_REPORT

  as

  whole crsor;

  rval around edge;

  Start

  crsor: = dbms_sql.open_cursor;

  DBMS_SQL. Parse (crsor, ALTER TABLE table_name TRUNCATED partition partition_nm INDEXES', dbms_sql.v7 GLOBAL UPDATE);

  rval Tip: = dbms_sql.execute (crsor);

  DBMS_SQL.close_cursor (crsor);

  end;

  Can you guide me how to solve this error?

  Have you tried to run immediately instead of dbms_sql?

  Start

  run immediately "ALTER TABLE table_name partition TRUNCATE partition_nm UPDATE GLOBAL INDEXES";

  end;

• ALTER TABLE with the column NULL time consuming

  Hello

  I have a table with nearly 2 million documents (about 10 GB size table) when I perform the following change statement its run fast

  ALTER TABLE tbl_1 ADD (DEFAULT col_4 NUMBER 100 NOT NULL);

  but when I do not give the NOT NULL in the alter its taking a long time to complete this i.e tbl_1 of ALTER TABLE ADD (NUMBER col_4 by DEFAULT 100); once I do the column as NOT NULL.

  Can someone explain why is it happening like that regarding the execution point.

  Thanks in advance.

  I can reproduce the effect in 11.2.0.1 - but not in 12.1.0.1.

  drop table tbl_1;

  create table tbl_1

  as

  Select rownum id

  , lpad ('* ', 50,' *') padding

  of the double

  connect by level<=>

  ALTER TABLE tbl_1 ADD (NUMBER col_4 by DEFAULT 100);

  --> slow in 11.2.0.1; fast 12.1.0.1

  ALTER TABLE tbl_1 ADD (DEFAULT col_4 NUMBER 100 NOT NULL);

  --> Quick 11.2.0.1 and 12.1.0.1

  A SQL trace for two operations (11g) shows that the altar slow (without the constraint explicit non-null) made explicit (and tedious) update of any row in the table:

  Update 'TBL_1"set"COL_4"= 100

  call the query of disc elapsed to cpu count current lines

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

  Parse        1      0.00       0.02          0          1          0           0

  Run 1 11,63 33.56 8336 1016125 2053531 1000000

  Fetch        0      0.00       0.00          0          0          0           0

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

  Total 2 11,63 33,58 8336 1016126 2053531 1000000

  The other trace file does not contain the explicit Update. I guess in the optimized case that oracle stores somewhere that col_4 has the value 100 and updates only the line of the block after completing DML operations. But that's just a guess.

• grant alter table

  Hello
  on the 11g R2,.
  which should be granted to the user, so being able to modify the table, or add column?
  I published
  

  GrANT ALTER TABLE TO USER 

  but I got ORA-00990.
  Thank you and best regards.

  user522961 wrote:
  Hello
  on the 11g R2,.
  which should be granted to the user, so being able to modify the table, or add column?
  I published

  GrANT ALTER TABLE TO USER 
  

  but I got ORA-00990.
  Thank you and best regards.

  You try with non-existent, should be amended any table ?

  SQL > SELECT DEALER, PRIVILEGE OF DBA_SYS_PRIVS WHERE DEALER ='&user';

  You can see any user if already granted.

• ALTER table a enable row movement cause problem

  Hello
  
  I did allow the movement of the row in the table and then shrink the space. which causes the performance issue? If so, how? How to allow the movement of the line and shrink the work of space internally?
  
  can deposit in full table scan, then the idexes on this table is not used more?

  947721 wrote:

  can shrink space available internally to what extent?

  Use the following syntax in a google search: alter table shrink space site: docs.oracle.com

  Note, by the way: "allow the movement of the line" allows you to take advantage of the features of line-movement, but he made no movement.

  Concerning
  Jonathan Lewis

• ALTER table order by

  Dear Sir
  
  Could someone help, I created a table with the increase of records of tests (test_id)
  I want to have test_id on descending in to not have to click desc order all the time at the opening of the table to insert a new record.
  
  the query
  
  ALTER table test_cases test_id desc order gives me following error: invalid alter table option...
  
  
  Thank you

  845498 wrote:
  Dear Sir

  Could someone help, I created a table with the increase of records of tests (test_id)
  I want to have test_id on descending in to not have to click desc order all the time at the opening of the table to insert a new record.

  the query

  ALTER table test_cases test_id desc order gives me following error: invalid alter table option...

  Thank you

  A table is a SET of data. Relational databases have been designed originally based on the principles of mathematical sets of information and this is what SQL design and database is based on, so why we have operations such as UNION and LESS etc. Data sets do not have a prescription for them, they are just data. If you want an order of your data, you must specify an ORDER BY clause in your query, or if you use a type TOAD etc. the GUI interface allows you to choose the desired order (which is just apply an ORDER BY clause to the background)... and with the TOAD, he remembers your preference for each table in the schema for the future. Other tools may or may remember preferences.