20.35. Release 0.214
Fix history leak in coordinator for failed or canceled queries.
Fix memory leak related to query tracking in coordinator that was introduced in Release 0.213.
Fix planning failures when lambdas are used in join filter expression.
Fix responses to client for certain types of errors that are encountered during query creation.
Improve error message when an invalid comparator is provided to the
Improve performance of lookup operations on map data types.
Improve planning and query performance for queries with
Fix issue where queries containing distributed
ORDER BYand aggregation could sometimes fail to make progress when data was spilled.
Make top N row number optimization work in some cases when columns are pruned.
Add session property
optimize-top-n-row-numberand configuration property
optimizer.optimize-top-n-row-numberto toggle the top N row number optimization.
ngrams()function to generate N-grams from an array.
Add functionality to delay query execution until a minimum number of workers nodes are available. The minimum number of workers can be set with the
query-manager.required-workersconfiguration property, and the max wait time with the
Remove experimental pre-allocated memory system, and the related configuration property
Add functionality to refresh the configuration of file-based access controllers. The refresh interval can be set using the
JDBC Driver Changes
Clear update count after calling
Web UI Changes
Show query warnings on the query detail page.
Allow selecting non-default sort orders in query list view.
Hive Connector Changes
Prevent ORC writer from writing stripes larger than the maximum configured size.
hive.s3.upload-acl-typeconfiguration property to specify the type of ACL to use while uploading files to S3.
Add Hive metastore API recording tool for remote debugging purposes.
Add support for retrying on metastore connection errors.
Handle SQL execution timeouts while rewriting queries.