[FLINK-11034] Provide "rewriting config” to file system factory

• Type: Improvement
• Status: Open
• Priority: Not a Priority
• Resolution: Unresolved
• Component/s: Connectors / FileSystem, FileSystems
• Labels: auto-deprioritized-major auto-deprioritized-minor
• Assignee: Unassigned
• Reporter: Wei-Che Wei
• Created: 30/Nov/18 03:31
• Updated: 30/Nov/21 22:38

Description

In the discussion in this mailing thread http://apache-flink-user-mailing-list-archive.2336050.n4.nabble.com/how-to-override-s3-key-config-in-flink-job-td24606.html, it showed that it is not able to overwrite config in file system factory when submit a flink job.

That means we will share the same config for multiple jobs in a session cluster. Or user can't use different configuration for checkpointing and file sink. For example, user might have different s3 buckets for checkpointing and file sink, but each of the s3 bucket might have different s3 access key for some management concerns.

We might need to provide a way to overwrite configuration when calling file system factory "get" method, and let those user facing components, like checkpointing or file sink, will be able to get overwriting config from user and create a file system with those changes in the new config.