Release 370-e LTS (17 Feb 2022)#

Starburst Enterprise platform (SEP) 370-e is the follow up release to the 369-e STS release and the 364-e LTS release.

This release is a long term support (LTS) release.

The 370-e release includes all improvements from the following Trino project releases:

It contains all improvements from the Starburst Enterprise releases since 364-e LTS:

Highlights since 364-e#

Revamp of the Starburst Enterprise web UI, including Starburst Enterprise query editor overview and Insights.
Add Data products.

Breaking changes since 364-e#

Remove distinction between system and user memory. This simplifies cluster configuration. The configuration property query.max-total-memory-per-node is removed, query.max-memory-per-node should be used instead.
Remove unnecessary spilling configuration properties spill-order-by and spill-window-operator.
Remove catalog configuration property allow-drop-table from the following connectors:
- Starburst IBM Db2 connector
- Starburst Generic JDBC connector
- Starburst Greenplum connector
- Starburst Netezza connector
- Starburst SAP HANA connector
- Starburst Snowflake connector
- Starburst Stargate connectors
- Starburst Synapse connector
- Starburst Teradata connectors
- Vertica connector
- Many Trino-based connectors and their Starburst extensions

370-e initial changes#

General#

Use UBI8 minimal base image for the Cache Service Docker image.

Security#

Add built-in access control.

Redshift connector#

Add support for timestampz and timestamp with time zone Redshift data types.

370-e.1 changes (23 Feb 2022)#

Fix data corruption when multiple queries DELETE from, INSERT into, or UPDATE a Delta Lake table.
Fix incorrect date result the value is <= 1582-10-14 in MySQL, SingleStore, and PostgreSQL connectors.
Migrate Docker image for Starburst Cache Service to UBI8 minimal base image.
Fix IndexOutOfBoundsException with accelerated parquet reader.
Improve query performance by removing redundant predicates above table scan.
Added separate properties for built-in access control to set authorized users and groups.
Fix potential query failures for queries writing data to tables backed by S3 in Hive and Iceberg connectors.
Fix problem when memory wasn’t always released after query completion.
Fix performance regression in internal communication authentication processing.
Fix failure when casting to decimal(38, 38)

370-e.2 changes (18 Mar 2022)#

Fix query failures due to exhausted file system resources after DELETE or UPDATE.
Fix issue where the length of log file names grow indefinitely upon log rotation.
Add INT to BIGINT coercion in accelerated Parquet reader in Hive, Delta Lake, and Iceberg connectors.
Fix failure reading timestamp data with accelerated Parquet reader.
Accept only lowercase letters for role names in built-in access control to ensure SQL compatibility.
Fix metastore impersonation for Avro tables.
Fix a bug that occurred when Parquet file schema and Trino table definition did not match.

370-e.3 changes (25 Apr 2022)#

Fix CVE-2022-23848 in Alluxio client.
Fix RPM to include MapR Hive. libraries.
Fix failure of the sync_partition_metadata procedure when partition names differ from partition paths on the file system.
Fix query failure when selecting partition when the name corresponds with a reserved keyword.
Fix performance regression for GROUP BY queries.
Fix built-in access control (BIAC) UI to support large amounts of schemas, tables, and columns during privilege creation.
Fix certain queries failing due to dictionary compacting error.
Fix performance regression when planning large IN lists.
Fix bug in accelerated parquet reader.
Fix failure reading timestamp data with accelerated parquet reader.
Fix Delta Lake queries with predicates that are not expressible by tuple domain for nullable partition columns.

370-e.4 changes (20 May 2022)#

Ignore non-redirected, non-Delta Lake tables or views when querying information_schema.columns.
Add support to the Hive connector for date type partition names with timestamp formatting.

370-e.5 changes (8 Jun 2022)#

Fix potential query failure when metastore caching is enabled.
Fix typo in the configuration property name to remove an unsupported underscore and replace it with the appropriate dash DELEGATED-OAUTH2.
Fix sync_partition_metadata procedure failure when table has a large number of partitions.
Fix incorrect results for queries where aggregation is pushed down for a remote database to execute and the aggregation function result is not needed to evaluate the query. Applies to the ClickHouse, MariaDB, MySQL, Oracle, SQL Server, PostgreSQL, and SingleStore connectors

370-e.6 changes (1 Jul 2022)#

Allow canceling a query on a transactional table if it is waiting for a lock.
Avoid errors when attempting to query tables that exist in multiple Snowflake databases with role impersonation enabled. The errors were a result of multiple tables matching the same schema/table name.

370-e.7 changes (1 Aug 2022)#

Important: In this release, we’ve patched a bug that causes the potential for incorrect results from certain queries that have had joins reordered in specific ways, in combination with certain data. A small percentage of queries are affected, but due to the complexity of the conditions that trigger the bug, it is not possible to predict which queries will be affected and thus we recommend that _all_ customers upgrade all clusters.

Fix incorrect results for certain join queries containing filters involving explicit or implicit casts.
Add Varbinary handling to Kafka Protobuf deserializer.
Fix incorrect query results when reading a Delta Lake table with a cached representation of its active data files that are outdated.
Fix certain complex queries that involve joins and aggregations.
Fix incorrect results when using the Glue metastore and queries contain IS NULL with additional filters. Applicable to Hive, Iceberg, and Delta connectors.
Fix incorrect pushdown of expression below join.

370-e.8 was skipped.

370-e.9 changes (9 Sep 2022)#

Fix writing incorrect results in the Delta Lake connector when the order of partition columns is different from the order in the table definition
Fix incorrect table already exists error in the Delta Lake connector caused by a client timeout when creating a new table.

370-e.10 changes (26 Sep 2022)#

Fix bug that prevented cluster metrics for Insights from being persisted.
Fix query failure when renaming or dropping columns that should be quoted. Applies to the ClickHouse, MariaDB, MySQL, Oracle, Phoenix, PostgreSQL, Redshift, SingleStore, and SQL Server connectors.
Fix query failure when adding a column that shoud be quoted. Applies to the Phoenix connector.
Fix query failure when adding a column with a column comment that has special characters which require it to be escaped. Applies to the ClickHouse connector.
Fix query failure when creating a table with a table or column comment that has special characters which require it to be escaped. Applies to the ClickHouse, MariaDB, and MySQL connectors.
Fix query failure when setting a table comment that has special characters which require it to be escaped. Applies to the ClickHouse, MariaDB, and MySQL connectors.
Fix query failure when setting a column comment that has special characters which require it to be escaped. Applies to the ClickHouse, Oracle, PostgreSQL, and Redshift connectors.

370-e.11 changes (6 Oct 2022)#

Fix potential SQL injection when querying BigQuery tables.
Fix inability to set the AWS STS endpoint and region when using a Glue metastore. Applies to the Delta, Hive, and Iceberg connectors.

370-e.12 changes (27 Oct 2022)#

Prevent coordinator out-of-memory failure when querying a large number of tables in a short period of time. Applies to the Delta Lake connector.
Fix error when using PREPARE with DROP VIEW where the view name is quoted.
Fix network issues during the data transfer for tables with a large number of columns. Applies to the Teradata Direct connector.

370-e.13 changes (17 Nov 2022)#

Fix failure for certain queries involving join operations over partitioned tables.
Fix potential query failure or incorrect results when reading from a table with the avro.schema.literal Hive table property set. Applies to the Hive connector.
Fix possible query failures for certain queries with join pushdown enabled on the Teradata Direct connector.
Fix failure when reading duplicated column statistics in the Hive connector.

370-e.14 changes (1 Dec 2022)#

Fix a correctness bug for queries with certain window operators used in sequence.
Suppress access denied exception in the Hive connector when listing all tables/views in a Glue database.
Fix failure for certain queries involving joins over partitioned tables.

370-e.15 changes (8 Dec 2022)#

Fix bug in parquet reader for arrays spanning multiple parquet pages. Applies to the Hive, Delta Lake, and Iceberg connectors.

370-e.16 changes (20 Jan 2023)#

Fix an issue when creating materialized view is denied because of missing permissions to HMS but replacing materialized fiew is allowed when using HMS impersonation.
Fix ArrayIndexOutOfBoundsException from accelerated parquet reader when reading string columns. Applies to the Hive, Delta Lake, and Iceberg connectors.
Disallow performing UPDATE or DELETE on Hive ACID transactional tables to prevent correctness issues when the operation modifies a large number of rows. These operations can be re-enabled using the hive.acid-modification-enabled catalog configuration property or the acid_modification_enabled catalog session property.
Fix parquet read failure where column indexes do not include a null count.

370-e.17 changes (14 Feb 2023)#

Fix incorrect results for large negative values stored in DECIMAL or TIME type columns with DELTA_BINARY_PACKED encoding in parquet files. Applies to the Hive, Iceberg, and Delta Lake connectors.
Fix query failure for some outer join operations.

370-e.18 changes (21 Feb 2023)#

Fix incorrect results for queries involving an equality predicate in a WHERE clause that is equal to a term of a SELECT clause in one of the branches of a JOIN.
Fix INSERT failing on DATE columns with multiple values containing NULL. Applies to the Teradata connectors.
Fix failure potential failure for queries involving joins and implicit or explicit casts of NULL to a concrete type.