[KUDU-1563] Add support for INSERT/UPDATE/DELETE IGNORE

• Type: New Feature
• Status: Resolved
• Priority: Major
• Resolution: Fixed
• Fix Version/s: 1.14.0
• Labels: backup impala roadmap-candidate
• Assignee: Grant Henke
• Reporter: Dan Burkert
• Created: 16/Aug/16 17:50
• Updated: 30/Apr/22 05:01
• Resolved: 13/Jan/21 15:20

Description

The Java client currently has an option to ignore duplicate row key errors, which is implemented by filtering the errors on the client side. If we are going to continue to support this feature (and the consensus seems to be that we probably should), we should promote it to a first class operation type that is handled on the server side. This would have a modest perf. improvement since less errors are returned, and it would allow INSERT IGNORE ops to be mixed in the same batch as other INSERT, DELETE, UPSERT, etc. ops.

Issue Links

• blocks: KUDU-1636 Deprecate KuduSession#setIgnoreAllDuplicateRows
• is duplicated by: KUDU-968 Allow double-deletes of rows if requested.
• is related to: KUDU-1577 Spark insert-ignore is significantly slower that upsert
• is related to: IMPALA-10465 Improve Kudu DML error logging in Impala
• is related to: KUDU-2061 Java Client Not Honoring setIgnoreAllDuplicateRows When Inserting Duplicate Values
• is related to: KUDU-2482 Add "delete ignore" capability to spark
• is related to: NIFI-7987 Support ignore operations in the PutKudu processor
• relates to: IMPALA-4416 Return warning msgs/counts to clients for ignored Kudu warnings/conflict violations
• relates to: KUDU-3211 Add a cluster supported features request