• Status: Solved
  • Priority: Medium
  • Security: Public
  • Views: 1949
  • Last Modified:

How can i get a dump of entire DB2 schema?

Hi All,

We are using a DB2/400 database. For a migration requirement, we are required to dump all data in all tables of a schema into a text file or into a excel file (or CSV file). Is there an easy way to do it preferably using the iSeries Navigator GUI?

The team that will be working on this is not too familiar with AS400 commands so I will have to do with using only the GUI.

Regards
Ali.
0
bhagatali
Asked:
bhagatali
1 Solution
 
Gary PattersonVP Technology / Senior Consultant Commented:
There is no "push button" way to do this for all tables in a schema at once.  You can use Client Access File Transfer function to download tables one at a time in CSV, Excel, or any of a number of additional formats.

Personally, I'd just write a short script (I prefer CL for these tasks) to dump a list of tables for each schema to a file, and then read through the file, executing the CPYTOIMPF commmand for each table to generate a CSV for each table.

The program below is an example - it has not been tested, and may not work as expected.  Use it at your own risk.

Cut and paste source code into SEU, save, and compile using the CRTCLPGM command (or use PDM).
Run program: CALL yourlib/yourprogramname PARM('SCHEMA_NAME_TO_DUMP')

Program will compile a list of all table in the named schema, and the convert them one by one into CSV files, storing them in the AS/400 IFS under the /CSVFILES folder.  IF your system already has a folder with that name, modify the program to use a different folder.

- Gary Patterson


PGM PARM(&SCHEMA)                                                    
                                                                     
DCLF QAFDBASI            /* System outfile for DSPFD OPTION(*BASATR) */                                                      
                                                                     
DCL &SCHEMA *CHAR 10     /* SCHEMA (LIBRARY) TO DUMP  */             
DCL &ROOTPATH *CHAR 106  VALUE('/CSVFILES') 
DCL &PATH *CHAR 117      /* AS/400 IFS PATH FOR CSV FILES */         
DCL &FILE *CHAR 128      /* STREAM FILE PATH AND NAME */             
                                                                     
             MD &ROOTPATH             /* Create the base folder undr the root of the IFS */
                 MONMSG CPF0000

             CHGVAR     VAR(&PATH) VALUE(&ROOTPATH *TCAT '/' *CAT &SCHEMA)                                   
             MD         DIR(&PATH)      /* Create subdirectory under CSVFILES for this schema */                             
               MONMSG CPF0000                                        

             /* Dump list of all physical files (tables) in this lib (schema) */
             DSPFD      FILE(&SCHEMA/*ALL) TYPE(*BASATR) +           
                          OUTPUT(*OUTFILE) FILEATR(*PF) +            
                          OUTFILE(QTEMP/QAFDBASI)                    

             /* Override to newly-created outfile */
             OVRDBF     FILE(QAFDBASI) TOFILE(QTEMP/QAFDBASI)        
                                                                     
READ:        RCVF                                                    
                   MONMSG   CPF0864  EXEC(GOTO EOF) /* Monitor for EOF */                    
                                                                        
             /* Build CSV path and filename */
              CHGVAR &FILE (&PATH *TCAT '/' *TCAT &ATFILE *TCAT '.CSV') 

             /* Export to CSV */
              CPYTOIMPF  FROMFILE(&ATLIB/&ATFILE *ALL) +                
                           TOSTMF(&FILE) MBROPT(*REPLACE) +             
                           STMFCODPAG(*PCASCII) RCDDLM(*CRLF)           
              GOTO READ                                                 
 EOF: ENDPGM

Open in new window

0
 
bhagataliAuthor Commented:
Thanks Gary.I have used this approach before so I can write up a CL for our requirement using your instructions as a guide. Just wanted to check with the experts if there was an easier "push" way to get this done.
0
 
theo kouwenhovenCommented:
Hi bhagatali,

Yes it can be done:

- Open the iSeries navigator.
  - Open Databases
    - Open your-DB
      - Open Schemas
        - Open the required library
          - right-click the requierd objecs (e.g. tables or indexes)
          - Click 'Generate SQL'
          - Select the objects you need and generate the SQL.

Good Luck

0
Concerto Cloud for Software Providers & ISVs

Can Concerto Cloud Services help you focus on evolving your application offerings, while delivering the best cloud experience to your customers? From DevOps to revenue models and customer support, the answer is yes!

Learn how Concerto can help you.

 
bhagataliAuthor Commented:
Hi murphey2,

The steps that you mentioned would only give the structure of the tables that I need. It would give me the SQL statement that was used to create that table.I am looking for a way to dump the data from the tables.

Regards
Ali.
0
 
tliottaCommented:
And don't forget that the .CSV files are going to take up space -- potentially more than the actual 'schema' itself, though most likely not.

Migrations sometimes have space issues as a contributing factor. It's just something to keep in mind.

Tom
0
 
bhagataliAuthor Commented:
Thanks for the pointer Tom. Appreciate it.
0

Featured Post

VIDEO: THE CONCERTO CLOUD FOR HEALTHCARE

Modern healthcare requires a modern cloud. View this brief video to understand how the Concerto Cloud for Healthcare can help your organization.

Tackle projects and never again get stuck behind a technical roadblock.
Join Now