Consider the following rules and guidelines when you migrate a mapping:
When you migrate a mapping, ensure that the metadata of the imported object matches the metadata fetched in an override.
When you migrate a mapping, do not specify the relative directory in the directory override field. When you migrate and run the mapping, the path used for importing the object during the design time is added to the specified file path while fetching the metadata at runtime.
When you configure a mapping to create a target at runtime and enable the dynamic schema handling option, and then upon migration, you change the source connection in the mapping for which the FileName field is deleted in the database, the FileName field is still written to the target even though it was deleted.
You must exclude the FileName field from the Target transformation to resolve this issue.
When you use a different object path for the migrated mapping, Data Integration uses the imported object metadata instead of the metadata fetched at runtime in the following cases:
Gzip compression is enabled in advance source properties.
FileName field is mapped.
When you enable the Recursive Directory Read option and the first level directory contains multiple folders but no files, after you migrate and run the mapping, Data Integration uses the imported object metadata instead of the metadata fetched at runtime.