You work for a web retailer where you need to analyze data produced for your company by an outside market data provider. You need to produce recommendations based on patterns in user preferences by demographic found in the supplied data. You have stored the data in one of your company’s S3 buckets. You have created a Glue crawler that you have configured to crawl the data on S3 and you have written a custom classifier. Unfortunately, the crawler failed to create a schema. Why might the Glue crawler have failed in this way?

Answer options:

A.You did not add an exclude pattern when you configured the data store.
B.The IAM role you assigned to the crawler has the AWSGlueServiceRole managed policy attached plus an inline policy that allows read access to your S3 bucket.
C.All the classifiers returned a certainty of 0.0
D.You chose to create a single schema for each S3 path.

Answer: C
Option A is incorrect. This configuration option is used to exclude objects from the crawler. From the help text on the Add a Data Store screen in the Add Crawler console flow: “The exclude pattern is relative to the include path. Objects that match the exclude pattern are not crawled. For example, with include path s3://mybucket/ and exclude pattern,mydir/**. Then all objects in the include path below the mydir directory are skipped. In this example, any object whose path matchess3://mybucket/mydir/** is not crawled. For more information about patterns, see Cataloging Tables with a Crawler”
Option B is incorrect. The IAM role assigned to your crawler needs exactly this managed policy and S3 bucket access. From the Choose an IAM Role screen on the Add Crawler console flow: “Create an IAM role named ‘AWSGlueServiceRole-rolename’ and attach the AWS managed policy, AWSGlueServiceRole, plus an inline policy that allows read access to: s3://yourbucketname”
Option C is correct. The data from the market data provider did not match with certainty any of the built-in classifiers that are part of Glue or your custom classifier. Therefore, Glue returned the default classification string of UNKNOWN. (See the Amazon Glue doc Adding Classifiers to a Crawler)
Option D is incorrect. This setting allows you to group compatible schemas. Choosing this option would not prevent the crawler from producing the schema. From the Configure the Crawler’s Output screen in the Add Crawler console flow: “This crawler configuration groups compatible schemas into a single table definition across all S3 objects under the provided include path. Other criteria will still be considered to determine proper grouping.”
Reference:
Please see the AWS developer guides AWS Glue: How It Works and AWS Glue Concepts.