Hello Group,

We are using DB2 8.1 on Win 2k server and we need to perform data porting from existing database (.nsf files) of lotus notes v5.x. The data in lotus notes is of both Double Byte Character Set and Single Byte Character Set.

We have set the DB2 database to UTF-8 at the time of creation in server and the client codepage is set to 1208 (UTF-8).

The issue is we have written lotus notes agents to push the data to DB2 as there are many virtual fields in notes which is a issue for Lotus Enterprise Integrator.

any help on this is appreciated.

Thanks

MDN

Would you mind indicating what your problem is?

The problem is the DBCS characters in the data are getting ported as JUNK.

It would help if you provided more details about your situation, such as:
- do you run Domino on the same server as db2?
- if not, what is the codepage on the Domino server?
- do you use server agents or local agents?
- if local, what is the codepage on the Notes client machine?
- how do you access db2 from Notes? (LC-LSX? Java? ODBC?)
- what fields do you update in db2 (char? clob? if both types, is the "junk" the same or different?)

for each codepage, there are max three ccsids(SBCS, MIXED, DBCS),
and min one ccsid(SBCS).
UTF-8 has three: 367, 1208, and 1200(DBCS). Hope this can help you.
Another issue you should concern is: what's the original encoding for your .nsf files, if it is not unicode, there will be a conversion mapping for it. You should check whether it is valid.

Please find the response for your queries below:

1. we run domino and db2 on different servers
2. the code page set on domino server is default but the db2 client codepage environment variable is set to 1208
3. we are using server agents
4. we are using odbc (DB2 ODBC Driver), but like to know if there are any alternatives and their approach
5. the firlds that we are updating are char and varchar the pattern of junk data is same in all fields.

In this case I would suggest looking at the ODBC driver configuration. This is where the (mis)conversion occurs, I believe. For example, the driver might think that the data that's being passed to it use the platform default code page while in fact they are in Unicode.

For IBM driver, look at the db2cli.ini file. I'm not sure how you configure MS DB2 ODBC driver, if that's what you use, but I think you'll be able to find some information on MSDN...

Below are the only param that I see in the db2cli.ini file under the user profile which I use and login.

DBALIAS=APPDB
UID=maheshdn

I can't give you a definitive answer; however if you search DB2 Knowledge Base on IBM web site for something like "odbc unicode" you may find something that will help you.