AWS Glue is a serverless data integration service. There can be a condition where you can expect new column in JSON file regularly . There can be a case such that tables partitions contain different schemas. When you try to read from AWS Athena , you will get the schema mismatch error as HIVE_PARTITION_SCHEMA_MISMATCH. In situation like this if we have multiple files, then the schema can get by reading all the source files together from S3 location , as schema of each partition can varry. To handle this AWS Glue crawler have a specific configuration options as “Update all new and existing partitions with metadata from the table.” You need to check the check box to handle this.
Configure crawler as below
Reference
