Changelog for us.qubole.com

Date and time of release Version Change type Change
22nd April, 2019 (08:13 AM PST) 55.0.48 Bug fix AN-2056: Uploading objects or files through the Qubole UI is now supported for all AWS regions. KMS-enabled buckets are also supported.
02nd April, 2019 (06:31 PM PST) 55.0.37 Enhancement

PRES-1350: Qubole supports configuring the required number of worker nodes during autoscaling. It is a cluster configuration override, query-manager.required-workers. You can set it to denote the number of worker nodes that must be in the cluster before a query is scheduled to be run on the cluster. This enhancement is only supported with Presto 0.193 and later versions. For more information, see Configuring the Required Number of Worker Nodes.

PRES-2397: Qubole supports escaping newline \n and carriage return (\r) characters in data for correctly parsing on the QDS UI. This enhancement is not available by default and it is only supported with Presto 0.193 and later versions. Create a ticket with the Qubole Support to enable it on the QDS account.

PRES-2417: Presto clusters do not terminate while actively running Presto notebook paragraphs. The enhancement is not available by default. Create a ticket with Qubole Support to use it.

PRES-2474: The optimization to speed up queries on system.jdbc.tables with filter on a single table name. This speeds up the extract operation in Business Intelligence tools such as DBeaver, which query system.jdbc.tables with filter on a single table name.

Bug fix

PRES-1373: This is a fix for an issue where the retry operation was not working on the failed Presto queries. Retries work on the failed Presto queries now. As part of the fix, retries are configurable for Presto queries that run on the QDS platform.

Caution

Configuring retries will just do a blind retry of a Presto query. This may lead to data corruption for non-Insert Overwrite Directory (IOD) queries.

PRES-2306: Fixed a bug where the Usage page on the QDS UI did not show the Presto query’s bytes read statistics.

PRES-2493: Fixed the data loss issue in the Hive connector when writing bucketed sorted table in Qubole Presto 0.208 queries.

PRES-2567: Fixed a bug in the user-overridden IAM Role feature where stale credentials were being used in a long-running cluster.

PRES-2571: Fixed a concurrency related issue in autoscaling on Qubole Presto 0.193 and 0.208 versions due to which the cluster’s size permanently exceeds its maximum size.

Bug fix

HIVE-3740: The Hive metastore API call has been replaced with a less expensive API call to resolve the issue of false alerts that were triggered on Datadog.

HIVE-4040: It is a fix for an issue in DelimitedJSONSerde that caused NullPointerException when hive.qubole.escape.crlf.in.result was enabled.

HIVE-4053: It is a fix for an issue where the hive.qubole.drop.table.metastore.client.socket.timeout configuration was not honored when a Hive query containing the DROP Table statement was run.

HIVE-4280: Hive jobs were failing with a Metaexception that included a timeout message. To resolve this issue, the SSH configuration parameters’ maximum value is increased on QDS servers.

QTEZ-335: It resolves an issue where the Tez Offline UI was inaccessible when yarn.timeline-service.store-class was set to RollingLevelDBTimelineStore in the cluster’s Hadoop overrides.

27th March, 2019 (2:01 PM PST) 55.0.32 Bug fix SPAR-3391: Some paragraphs ran successfully but with a warning message during the first run. However, those paragraphs ran successfully without the warning on a re-run. This issue is fixed.
25th March, 2019 (05:21 AM PST) 55.0.25 Bug fix TOOLS-1139: The previous docker version had security vulnerabilities, which could allow malicious containers to gain root-level privileges on the host. To resolve security vulnerabilities of the previous docker version, the latest patch version of docker-18.06.1ce is installed.
13th March, 2019 (04:45 AM PST) 55.0.22 Enhancement SPAR-3005: For the offline Spark clusters, only the event log files that are less than 400 MB are processed in the offline Spark History Server (SHS). This prevents high CPU utilization on the internal servers due to SHS.
Bug fix SPAR-3371: Dynamic filtering now works for join keys of type Int, String, Long, and Short.
13th March, 2019 (04:45 AM PST) 55.0.22 Bug fix

ZEP-3247: Zeppelin failed to start due to the interpreter.json._COPYING_ file in defloc. As a result, the notebooks failed to load. With this fix, Zeppelin starts successfully and the interpreter.json._COPYING_ file is not seen on defloc.

ZEP-3252: On a cluster restart, the latest dashboards are displayed after the scheduled runs.

ZEP-3230: Zeppelin server in Java 7 clusters was failing due to some ciphers that were disabled for security. Now, the Zeppelin server starts successfully as the required ciphers are added when starting the Zeppelin server.

ZEP-3243: A notebook with the ERROR status on a cluster failed to associate back to the cluster even after switching to another cluster. This issue is fixed.

ZEP-3234: Notebooks failed to load in case of higher latencies when fetching notebook permissions. This issue is fixed.

04th March, 2019 (09:30 PM PST) 55.0.14 R55 Upgrade
15th January, 2019 (7:18 AM PST) 54.0.53 Bug fix SCHED-321: In the Cron expression, the scheduler calculates the Next Materialized Time (NMT)/Start time considering the current time as the base time and Cron expression passed. Start time is not honored in the Cron expression. It resolves the issue where the first instance of the scheduler did not honor the scheduled time.
11th January, 2019 (12:30 PM PST) 54.0.52 Bug fix AN-1407: While filtering commands on the Analyze page, you must now provide the date range (start/end date) when searching for cluster labels.
7th January, 2019 (1:30 AM PST) 54.0.46 Enhancement AD-1834: The list of email IDs configured in Account Settings will now receive notifications related to releases, account configuration and feature changes.
21st December, 2018 (8:35 AM PST) 54.0.45 Bug fix SPAR-3207: The Spark shuffle cleanup feature is not supported in Spark 2.2.1. However, the Spark shuffle cleanup feature continues to be supported in Spark 2.3.1 and later versions.
17th December, 2018 (9:51 AM PST) 54.0.41 Bug fix ZEP-3109: Fixed the issue that caused notebooks to freeze frequently due to the huge content, such as large number of clusters, users, and schemas.
13th December, 2018 (3:45 PM PST) 54.0.40 Bug fix RUB-102: It is a fix to avoid unnecessary HTTP GET calls to Ganglia for fetching the cluster metrics due to which the master node was overloaded.
12th December, 2018 (12:00 AM PST) 54.0.39 Enhancement

ACM-3436: QDS now supports c5n, m5a, and r5a instances.

ACM-3545: Qubole now provides a feature to avoid the cluster from starting whenever the master node’s Elastic IP address fails. This feature is not enabled by default. Contact Qubole Support to enable this feature.

6th December, 2018 (10:56 PM PST) 54.0.38 Bug fix EAM-1502: Fixed an issue in the Automatic Statistics Collection framework due to which the fresh statistics were not collected for a few tables.
29th November, 2018 (12:56 AM PST) 54.0.33

Major release

For the list of changes until 29th November, 2018, see Changelog for api.qubole.com.