Task for Cleanup of Database

[Satish Mullapudi]

Hi All,
First of all am happy to announce that the scripts which I have developed for the cleanup of a database have successfully been deployed into production.
These scripts are windows batch scripts. I want you ppl to work out & develop the same as Unix shell scripts. As a matter of fact unix is much more user friendly when coming to scripting when compared to windows. I'll jus give the online of the task. I want you ppl. to work on it.

1. Take the backup of the existing db
2. Verify whether the backup operation is successful of not using the appropraite cmds
3. Once it is confirmed that the backup is successful
4. Run Emp_Org cleanup script
   a. From current tables, export rows we want to keep
    b.Drop old tables
    c.ReCreate required tables
    d.Import the exported rows
5.Verify the entire cleanup operation
6.Backup new database
If at any stage the cleanup is unsuccessful, RESTORE the database to its old state (before cleanup)

I am giving a sample datamodel to work on & the conditions . Do create a database as required & insert sample data into it. The data preferably should be huge because that tests the performance of your scripts.

Ravi: It is preferrable that you update me with the status of the task being done there , so that I can guide you ppl, if you are going wrong at any point.
Varma Sir/Ravi: Pls. see to it that all of our DB team works on this bcoz it is really a good task to work on both on DBADMIN & UNIX shell scripting.

If you have any queries feel free to contact me.

Thanks & Regards,

Satish Mullapudi,
IBM Certified Database Associate,
IBM Information Management (DB2 UDB),  
Miracle Software Systems Inc.
email: smull...@miraclesoft.com
Mobile: +919923020321

[Venkata Ravi Kumar Koppala]

Gokul,

 

    Go through the Data Modelling Doc given by Satish & Validate by EOD today.

 

Thanks & Regards,
Venkata Ravi Kumar Koppala,
Team Lead,
Database/BI/ETL,
Miracle Software Systems, Inc.
Phone :- 248-233-1854
WWW.MIRACLESOFT.COM
A Premier Business Partner of IBM.
A National Partner of SAP.
A National Partner of Siebel.
An Associate Partner of WebMethods.

[Gokul Sagi]

Hi Ravi,

 

Here’s the status of my task (Data modeling Doc)

 

We have to create 6 tables (employee, Department, Project, sales, Staff, Emp_proj) and a view (emp_dept_view) based upon Department, sales.

 

I had a problem here

 

In the creation of tables there is a foreign key specified for column PROJ_DEPT in EMP_PROJ table and it’s related to DEPARTMENT table where such column does not exist.

 

Similarly there is a foreign key specified for column PROJ_MGR in table PROJECT and it’s related to employee where there’s no column in EMPLOYEE table.

 

Just check it once and mail me. Then I shall create tables and populate the data into them.

 

Thanks & Regards,

Gokul Sagi

color=black>

[Satish Mullapudi]

Hi Gokul,

I have got your question. Even if the names doesn't match in Parent & Child Tables, you can have a Primary - Foreign Key relationship. The reason why we go for common names in both (parent-child tables) is , not to get confused between both of them.

For further queries feel free to contact me,

Thanks & Regards,

Satish Mullapudi,
IBM Certified Database Associate,
IBM Information Management (DB2 UDB),  
Miracle Software Systems Inc.
email: smull...@miraclesoft.com
Mobile: +919923020321