[BUG] VPC Flow Log Integration Issue with Glue Catalog Database Name

[rafael-gumiero]

VPC Flow Log Integration Issue with Glue Catalog Database Name

Environment

• OpenSearch Version: 2.17
• Integration Type: VPC Flow Log
• External Source: Amazon S3
• AWS Services: AWS Glue Data Catalog
• Catalog Name: test_zero_etl_aos
• DataBase Name: db_test_zero_etl_aos
• Table Name: table_test_zero_etl_aos_flow

Description

When attempting to configure VPC Flow Log integration with OpenSearch using AWS Glue Data Catalog as the metadata store, the system fails to properly handle the database name configuration. The integration expects the default database name in the Glue Catalog, but fails even when this is explicitly specified.

Current Behavior

• Integration setup fails when attempting to configure VPC Flow Log integration
• Error occurs even when explicitly specifying the expected default database name
• Integration does not properly recognize the Glue Catalog database configuration

Steps to Reproduce

1. Configure Amazon S3 as external source in OpenSearch 2.17
2. Attempt to set up VPC Flow Log integration
3. Specify different database name in the Glue Catalog configuration
4. Execute the integration setup

Error Messages

Table Sample

CREATE TABLE 
test_zero_etl_aos.db_test_zero_etl_aos.table_test_zero_etl_aos_flow (version INT, account_id STRING, interface_id STRING, 
srcaddr STRING, dstaddr STRING, srcport INT, dstport INT, protocol INT, packets BIGINT, 
bytes BIGINT, start BIGINT, end BIGINT, action STRING, log_status STRING, 
`aws-account-id` STRING, `aws-service` STRING, `aws-region` STRING, year STRING, 
month STRING, day STRING, hour STRING) 

USING parquet PARTITIONED BY (`aws-account-id`, `aws-service`, `aws-region`, year, month, 
day, hour) 

LOCATION "s3://xxxxx/flow/AWSLogs/"
```

[noCharger]

@rafael-gumiero I would call it a feature gap rather than a bug. Based on the source code it will only take the table name rather than database name.