DataStage migration part2
I have given the environment details for the migration. Now will move to the technical part of the migration.
1. The achievement is moving the DataStage project from oracle to DB2 environment.
In the old DataStage 7.5 environment, we don’t have information server concept where the DataStage repository data will be sorted in the oracle/db2 database as a separate database.
So the data will be storage in the same DataStage server. So we don’t need to worry about the DataStage software. It will be up and running fine. But the database connection to the country DB should be defined here.
We have two type of DB here.
Usually the DataStage will be collecting the data from different source like
1. Text format
2. Excel format
3. Database (Sybase,SQL,oracle,DB2,Teradata..etc)
And again it will put the final processed data in the any format like text or database (Sybase, SQL, oracle, DB2, Teradata..etc) which is called as Datawarehouse. Now for the migration we need to consider the below steps. Make sure the DataStage 7.5 or 8.0to 8.7 is up and running .i will different work flow for both 7.5 and 8*
DataStage migration for 7.5
1. install the DataStage software in new server(server A)
2. Take the DataStage project backup in old server(server B)
3. Create the same project as same as OLD server B in new server A
4. Import the project backup taken from server B to server A
5. Configure the dsenv file for oracle/db2 entries same as in OLD servers
6. Create new odbc configuration for orcle/DB2 country DB entries or update tnsname.ora file for oracle entries for country DB.
7. Test the same jobs in new DataStage server.
Note: The main achievement here is. The Oracle/DB2 schema concept. Since whenever we login to the DataStage jobs will fallow the below
1.login with unix/datastage user user1
2.connect to the oracle using user2/passwd from datastage job and create the table. The table name as per the Database concept will be “scheam name.table Name”. so in oracle it will be user2.table1 will be the table name
User1.table1(here user1 is the user name and schema name where user will login to oracle using user1 )
3.conenct to DB2 using user3/passwd from DataStage job. But here DB2 login will be as unix login as
db2admin and the schema will be named as per the user
2.explicit function means use the separate schema name as u defined
User2.table1(here user2 is the schema name but not the user to login to DB2 database)
As I mentioned before, so for this major problem in the DataStage job migration. All the developers need to sit and edit all the jobs for the table name after the migration from oracle to DB2 or They can simple create the same user1 as schema name in DB2 and can run the jobs without editing :)
DataStage 8.0 or 81 to 8.5 or 8.7 migration
1. Install oracle/DB2 database in server A
2. Create the schema iauser/xmeta with DB privileges and with create DB link privileges
3. Create dsadm user and dstage group in DataStage server B.
4. Install the datastage 8* version (where the Datastage metadata repository will be in server A)
5. Configure the dsadm user in webconsole (map the unix dsadm user with webconsole dsadm user)
6. Configure the dsenv file for oracle/DB2 country database entries and also the tnsname.ora file for oracle entries.
7. Make sure DB2/Oracle client is installed in DataStage server B.
8. Create the DataStage projects as same as OLD DataStage server.
9. Import the DataStage project backup from OLD DataStage server.
10. Test the jobs and compile and run.
Please let me know if any confusion in DataStage migration from oracle to DB2