Hi all, could someone explain how Datahub manages Elasticsearch schema updates/migrations?

Elasticsearch schemas (mappings and settings) are inferred from the Searchable annotations on the models (pdl files). Every time there are updates to the model causing a schema change, it 1) creates a new index with the new schema, 2) reindexes existing index for the given entity into the new index, 3) deletes the old index and changes the alias to point to the new index

Thanks @early-lamp-41924! Looks like this is the file that creates and updates indices? https://github.com/datahub-project/datahub/blob/55357783f330950408e4624b3f1421594c98e3bc/metadata-io/src/main/java/com/linkedin/metadata/search/elasticsearch/indexbuilder/ESIndexBuilder.java

Yes!