[HIVE-18552] Split hive.strict.checks.large.query into two configs - ASF JIRA

Log work

Agile Board

Rank to Top

Rank to Bottom

Bulk Copy Attachments

Bulk Move Attachments

Voters

Watch issue

Watchers

Create sub-task

Convert to sub-task

Move

Link

Clone

Labels

Update Comment Author

Replace String in Comment

Update Comment Visibility

Delete Comments

XML

Word

Printable

JSON

Details

Type: Improvement
Status: Closed
Priority: Major
Resolution: Fixed
Affects Version/s: None
Fix Version/s: 3.0.0
Component/s: Hive
Labels:
None

Target Version/s:

3.0.0
Hadoop Flags:

Incompatible change
Release Note:

Hide
This change splits {{hive.strict.checks.large.query}} into two new configs: {{hive.strict.checks.no.partition.filter}} and {{hive.strict.checks.orderby.no.limit}}

Previously, {{hive.strict.checks.large.query}} controlled whether or not (1) order bys were allowed without limits, and (2) scans on partitioned tables were allowed without filters.

After this change, {{hive.strict.checks.no.partition.filter}} controls the partitioned table check and {{hive.strict.checks.orderby.no.limit}} controls the order by check.

Show
This change splits {{hive.strict.checks.large.query}} into two new configs: {{hive.strict.checks.no.partition.filter}} and {{hive.strict.checks.orderby.no.limit}} Previously, {{hive.strict.checks.large.query}} controlled whether or not (1) order bys were allowed without limits, and (2) scans on partitioned tables were allowed without filters. After this change, {{hive.strict.checks.no.partition.filter}} controls the partitioned table check and {{hive.strict.checks.orderby.no.limit}} controls the order by check.

Description

hive.strict.checks.large.query controls the strict checks for restricting order bys with no limits, and scans of a partitioned table without a filter on the partition table.

While both checks prevent "large" queries from being run, they both control very different behavior. It would be better if users could control these restrictions separately.

Furthermore, many users make the mistake of abusing partitioned tables and often end up in a situation where they are running queries that are doing full-table scans of partitioned tables. This can lead to lots of issues for Hive - e.g. OOM issues because so many partitions are loaded in memory. So it would be good if we enabled this restriction by default.