Hi there - I am unable to save the LDM after making a simple change that I have made many times due to a new error that I received. Nothing else in the LDM has changed and it will not enable me to save.

Hi Max, it looks like an issue with data type and the expected entry. Can you please confirm the steps you have made and and if the column type in your source data matches the data type?

The change I made was in SQL in the Users table (a change I have made many times to add new users) and that save worked, but to save the overall LDM I receieved this error which is stating that a different table that was unchanged (Zendesk_users) and "users" has different data types?

Hi @Max Donsky, we have recently introduced additional validation of data model. The columns which contain references to primary keys from different datasets shall use the same data type so that datasets can be joined on equality between the two columns (equi-join). Join of columns with different types would either fail or result in sub-optimal performance if e.g. one of the columns is a numeric data type and other column is a string.

Hi @Jakub Sterba I was able to remove the connections as they were not relevant and the save function worked... but now in the dashboards when I filter by the partner names - this partner is not appearing? Any ideas? I followeed the same process I have done successfully many times

It is hard to tell without knowing more details about the case. I suggest to unify types in the database and refresh list of tables in the modeler while editing model. This will detect type changes and these changes can be saved with data model. You can also convert dataset to SQL dataset to apply explicit type casting if you cannot change type in database. It may be good to clear caches as well after this change. Are you filtering data in the same dataset? Relation between datasets is needed if you use for filtering attribute from different dataset than from which metric is computed.

@Jakub Sterba Please see DM's!

Hi @Max Donsky please let’s try to keep the conversation here. We work on different shifts and it’s better if we keep the conversation in 1 place so we can help you better. Thanks

Hi Daniela - this issue was resolved and is not the issue I am looking for support on. Please see the latest message that was posted in this channel