I love new DB nodes. At least, they perform far much faster than the legacy ones. They brought some limitations too. I used to employ a configurable Wrapped Metanode in my workflows, which was responsible for providing DB connection to a particular database I need to work with, in a particular part of the workflow. With new DB nodes, it stopped working as I described in my [ Recommended way of configuring a database connectors ]( http://Recommended way of configuring a database connectors) forum topic. Well, I got learnt that I can’t get DB connections from wrapped metanodes or components anymore but I’m expected to add appropriate type of DB Connector right to my workflow. It usually looks as follows: [image] As I stated in the topic I referred, I had to use the Credentials Configuration Node in order to credentials in the PostgreSQL Connector because I was not able to fill a database password with a Flow Variable value. The problem is if I open the workflow that contains this sequence nodes in executed state I cant continue with the workflow since a problem KNIME describes as: “Status: DataLoadError: Loading model internals failed: The server requested password-based authentication, but no password was provided.” Please keep in mind the sequence can be a part of loop or have a number of already executed following nodes so it’s not easy to re-execute the nodes with no impact to workflow. This particular issue makes unavailable one of the crutial features we as a company have chosen KNIME as an integration platform, that was an ability to load, investigate and continue with a forkflow that has failed. For this reason, I’d be glad if I have the option to continue with partially executed workflow back.

DB nodes - can't continue with workflow after it is re-loaded

jan_lender March 25, 2020, 2:01pm 14

I’d like the KNIME team would address this. In a number of topics here we complain “new” DB nodes is hard to use unless we use they a very simple way.

Some of the related topics are:

1 Like