MVSFORUMS.com

[sumithar]

OK guys, this has to be a stupid question.

Let's say I have a situation.

I am developing a web frontend app that is inquiry only. The data I need to display is in DB2 tables on a z/OS box.

I have a choice.

I can code JDBC calls in my app (for purposes of simplicity let us mix up presentation and data access into the same program for now to go directly against the DB2 tables.

OR

I don't know exactly what the benefits of bringing CICS into the equation but I have been told that it is better to go thru CICS to get to DB2 than direct calls to the database.
So, being an ex COBOL programmer, I can write some trivial CICS transactions to dump the required data on the screen using SEND TEXT (no need to format, I can do that in my Java app) and then use the HTTP server built into CICS to invoke these trans, retrieve the data into my Java app and display formatted as HTML.

is there any reason not to go with direct ODBC calls?

[Bill Dennis]

1. Will your app be executing 24x7 waiting for users to enter a query, "triggered" as needed or would an occasional CICS transaction be better?

2. How will you handle user authentication? Could CICS userID/password security do this better?
_________________
Regards,
Bill Dennis

Disclaimer: My comments on this foorum are my own and do not represent the opinions or suggestions of any other person or business entity.

[sumithar]

1)The intent is user can go to the URL any time but expectation is s/he will do it only during the working day. But you raise an interesting point, since our CICS does come down 9pm-7AM option 2 would cause an unavailability situation tho' I doubt it will be an issue
2) If I chose option 2, we already have a CICS user created for any web based access which would need to be authorised for this new transaction i.e. all web access would supply this user id/pwd as credential. But the input data stream would include the actual userid of the person typing the URL and verification would be done against DB2 database.