How are the existing processes affected when new columns are added to the CDT an

RE: How are the existing processes affected when new columns are added to the CDT an

chandu — Thu, 24 Jan 2019 12:39:21 GMT

Hi Arun,

Please go through the link, docs.appian.com/.../Process_Upgrade.html

RE: How are the existing processes affected when new columns are added to the CDT an

Arun Theja Muthyalu — Thu, 24 Jan 2019 12:29:47 GMT

Hi,

Thanks for the reply. I am trying to understand how can I update the process variable with the latest version of CDT in all the active instances. Because we have tasks already generated from these active process instances before the deployment. So all these tasks point to the earlier version of the CDT which doesn’t have these two newly added fields. So when the task is submitted, NULLS are being written to these new fields.

RE: How are the existing processes affected when new columns are added to the CDT an

chandu — Thu, 24 Jan 2019 12:22:34 GMT

Hi Arun,
Suppose you have CDT and it has 4 fields and there are 200 active instances runing on it.

Now you added two more columns,untill you refer these two more columns in process instances all the active Instances will
not fail.

If you have redesigned the subprocess and used the newly added columns, In this case the new subprocess always use the
new version of the CDT and Existing Instances still continue to work.

If you have used any where in SAIL, then the sail code will fails in the active process instances since the SAIL updates directly take impact.

RE: How are the existing processes affected when new columns are added to the CDT an

Arun Theja Muthyalu — Thu, 24 Jan 2019 11:29:55 GMT

I have added two new columns to the existing CDT. I am facing no errors as of now. I have around 200 active instances.

RE: How are the existing processes affected when new columns are added to the CDT an

josep — Thu, 24 Jan 2019 06:18:47 GMT

Hello Arun,

Are you running into an issue?

From this link docs.appian.com/.../Generating_Database_Tables_from_CDTs.html
"... running processes cannot be updated to use the latest version of a CDT. Therefore, changes to a CDT, its underlying database structure, or related artifacts such as rules and sub-processes may cause the running processes to behave unexpectedly."

That is why it is important to try to keep the process to be short living instances and restartable.

If you already have the CDT's updated and running process instances running and facing an issue please tell us more to help you better. for example how much instances? how was the change of your CDT?

Jose

RE: How are the existing processes affected when new columns are added to the CDT an

Arun Theja Muthyalu — Thu, 24 Jan 2019 00:53:00 GMT

Hi Eduardo Fuentes ,

Then how to update the process variable type to the latest version of the CDT in the active instances?

RE: How are the existing processes affected when new columns are added to the CDT an

Eduardo Fuentes — Tue, 02 Dec 2014 08:33:34 GMT

This was designed that way so running instances don't break when changes are made to the CDT. Keep in mind that any sub-process node which hasn't started the process yet will launch a new instance using the updated CDT if Impact Analysis is executed.

RE: How are the existing processes affected when new columns are added to the CDT an

Eduardo Fuentes — Tue, 02 Dec 2014 08:32:41 GMT

To update a CDT you first need to delete it and then upload the new version. When you do this the original one is versioned and all running instances will keep using that original version. The new version (with the additional columns) will only be used by new instances after you run impact analysis to update the relevant objects.

RE: How are the existing processes affected when new columns are added to the CDT an

ashleyb — Tue, 02 Dec 2014 06:20:20 GMT

**Also to mention all additional fields are nillable