Hello Veera,
No, the work process can only search for the table in the SAP table buffers.
If the table is not found in the SAP table buffers, the Work process, through Database (DB) interface intimates its Shadow process (Oracle process) for this table.
The Shadow process then searches for the required table in the database buffer cache in the system global area (SGA).
If the data is not found in the database buffers, the shadow process then searches for the table in the data files, copies it into the database buffer and then sends it to the work process through the DB interface, the work process then populates the table in the SAP table buffers (provided table buffering is allowed for that table)
The point to understand here is that, When the SAP instance(application server) is started, each and every workl process, as soon as it is started, connects to the Database and Oracle (Database) creates a shadow process for each work process.
Also note that the work process needs a user ID for the database to connect to it. All the work processes use a shared used ID SAP<Schema ID>(SID) to connect to the DB. There is a complex mechanism by which the work processes retrieve the password for the SAP<Schema ID>.
They use an OPS$ user ID (windows level user) which is trusted by the Database. to login to the OPS$ schema. Here the password is retrieved from a table called SAPUSER.
Then the Work processes disconnect from this schema of the database and connect SAP<Schema ID> using the password retrieved.
If the work processes lose conenction to the data base, they re connect when the database is up again.
If they are not able to connect to the database then the system crashes.
Hope the explanation has helped.
Do reward points if this has been useful
Regards,
Prashant
Edited by: Prashant Pasala on Feb 23, 2008 3:17 PM
You must be a registered user to add a comment. If you've already registered, sign in. Otherwise, register and sign in.