Description
The retry policy for the AWS 400/BadArgument response is currently "treat as a connectivity error" on the basis that sometimes it works again.
It doesn't, not normally, and by using the connectivity retry policy, unrecoverable failures can take time to surface.
Proposed: switch to a fail fast policy for BadArgumentException
Attachments
Attachments
Issue Links
- is depended upon by
-
HADOOP-15364 Add support for S3 Select to S3A
- Resolved