Default DB table for a process in a different DB schema or different DB?

Sorina Grave

I wonder if this is currently possible in Metastorm 9.2. We have several business processes with fairly large amounts of global variables which Metastorm stores implicitly in the default process tables (e.g. if the process is Overtime Request, the table dbo.Overtime_Request). There are some concerns (security, performance, etc) about storing the process data for various business processes in the default dbo schema in Metastorm DB.

 

It would be a good idea to store Metastorm specific tables e.g. eFolder in one place (dbo) and process specific tables elsewhere e.g. [ovt].[Overtime_Request]. Is there a way of achieving this in Metastorm ? I have tried changing the object's owner after a deployment but it isn't as obvious as that. I have also noticed that any custom business objects other than the default process table can be pointed to other schemas.

Thank you.

Jerome_Pearce

Rather odd, as your table naming implies a version 7 (or before) model, yet this is in version 9 forum. Is it really version 9?

 

I ask because in version 9 you can use any table you want. It can be in a completely diffrent database if you wished. You merely have to manage the creation of the appropriate record, and (most probably) just use the Folder Id as the key. There is no actual need to have a Process BO & table at all in version 9, although it can make adding variables a lot easier as these can be added via the 'manage variables' dialog and directly from fields.

 

In version 7, no, you cannot do this, at least not in a supported way.

Sorina Grave

Yes, it is version 9 - as I've posted in my original message  we've got version 9.2 here. The process has been converted from version 7 therefore the confusing naming.

 

I realise I can use any table I want for CUSTOM business objects, however, my query was specific to the PROCESS business object - according to the designer guide "Process Business Objects are created when a process is created. Every process contains it own Process Business Objects that cannot be deleted and remains permanently associated to the process." What this guide is not saying is that you might not have control where this business object will be saved.

 

The process I'm working on has most of its variables stored in the default process business object table and it would be great if this table could be stored elsewhere.

 

The difference between custom and process business objects is not only the ease in maintaining variables, you'll find that when wanting to write to these business objects, Metastorm creates a default record for the process business object but it won't do the same with a custom business object - yes, it's not the end of the world, a simple INSERT will do it but it would be nice if that was consistent with the process business object table.

 

Brian Mellert

Create your own table and supporting BOs and don't create the variables as part of the process itself.  You have to drag the applicable BO into each form anyway and you can drag it into the the process itself for data access.  Your connections/BO would handle the rest as Jerome suggests.

 

Since it sounds like it you migrated from V7 to V9, or developed using a similar method, which by default would move then to the process BO.  A bit of rework, though perhaps painful, could separate the data elsewhere as desired.