What’s NEW

The following sections summarize the most important new features and bug fixes included with this release.

For the previous release notes please navigate to Latest release section "Previous releases".

TABLE OF CONTENTS

• Installation scripts

• Installation process

• Connectors

• File connectors

• Google Big Query

• Oracle

• Parquet store

• Data Catalog

• Query Editor

• Conduit Backup

• Service Management

• Miscellaneous

• Logging information

Installation scripts

On premise (click to download)

• conduit-2.5.2.tar.gz

• conduit-transformless-install-onprem.sh

Installation process

For a complete guide to installing Conduit please navigate to:

• Installing on VM (any cloud or on-premise services)

• Installing from Azure Marketplace

• Post Installation Configuration

[CONDUITV3-1581] - increase default BDE_SERVER_MEMORY_OPTS from "-Xms1G -Xmx2G" to "-Xms2G -Xmx4G"
[CONDUITV3-1715] - expose default spark parameters in the conduit-transformless-install-onprem.sh
[CONDUITV3-1637] - sync keystore password with provisioning token on every update
[CONDUITV3-1440] - support configurable thrift server port for Conduit Console Management (UI)
[CONDUITV3-1372] - deployment script prompts the full path of the management console

Connectors

[CONDUITV3-1712] - add validation for columns if caching is enabled. At the moment, Conduit cannot cache columns that contain space in the name.
[CONDUITV3-1618] - in Connector creation wizard, show info message if 'cache now' is enabled so that the user can navigate to "Parquet Store" for more details
[CONDUITV3-1208] - fix failure to load Metadata default.connector in data_source_metadata for a connector which has more than two “_” sign in name

[CONDUITV3-1635] - improve connector name validation on cloned connector to check for names exceeding length limitations

[CONDUITV3-1488] - fix queries including the time columns from postgresql sample that failed

File connectors

[CONDUITV3-XYZ] - fix cache expiration for flat file connectors
[CONDUITV3-1544] - support distribution by flattened columns in file connectors when caching to parquet store

Google Big Query

   
[CONDUITV3-1570] - remove ActiveDirectory Pass Through authentication option. Not supported by Big Query.

Oracle

   
[CONDUITV3-216] - fix some issues with back-ticks, case sensitivity when using Query Editor

Parquet store

[CONDUITV3-1641] - add endpoint to trigger cache refresh from parquet store

[CONDUITV3-1634] - Handle cache jobs as separate groups for separate tables

[CONDUITV3-1644] - add absolute path and jobId to cloud store for each parquet store entry
[CONDUITV3-1601] - add more details about Parquet cache entries, such as cache path in storage

[CONDUITV3-1602] - add Parquet store auto refresh mechanism. Silently refresh web interface to update information.

[CONDUITV3-1650] - rename correlationId to Log correlationId when viewing details for a cache entry[CONDUITV3-1585] - Add metadata file marker for Parquet file system storage. Conduit will create a .txt file to allow better administration and tracking of configurations when using cloud storage.

[CONDUITV3-1623] - add parquet store help button to allow navigating directly to support documentation[CONDUITV3-1645] - support columns sorting in Parquet store table view[CONDUITV3-1701] - fix no second entry in parquet after first virtual dataset deletion for the same connector[CONDUITV3-1688] - fix expected Expiration time column that displayed the Current System time.

Related articles:

• Configure parquet store file system

• Caching 101

Data Catalog

[CONDUITV3-1457] - navigate from the catalog to the query editor via an asset view

Related articles:

• Data Catalog

Query Editor

[CONDUITV3-1592] - display correlationId for queries

Related articles:

• Query Editor

Conduit Backup

[CONDUITV3-1673] - fix issue when GCS backup can't be triggered on clean server
[CONDUITV3-1545] - add validation for GoogleCloudStorage backup location that it must contain absolute path starting with "/"" for folder
[CONDUITV3-1328] - make mandatory fields visible by default. save button disabled all the time until mandatory fields are filled-in

Related articles:

• Configure your backup policy

• Restore your backup

Service Management

[CONDUITV3-1579] - improve error handling for Conduit server so that the container restarts properly on OutOfMemory exceptions
[CONDUITV3-1500] - add disk space monitoring in conduit_agent
[CONDUITV3-1505] - add support for send warning email to all admins when one VM has low disk_space

Related articles:

• Conduit Services

Miscellaneous

[CONDUITV3-1513] - set TimeToLive (TTL) for Zookeper snapshots and update docker image so that Zookeepr service does not run out of space by not removing old files.
[CONDUITV3-1571] - display friendlier error message for OData endpoint over BigQuery
[CONDUITV3-1700] - fix reset password modal not displayed
[CONDUITV3-646] - add mechanisms to avoid spark taking up entire disk leading to "No space left on device" errors

Logging information

[CONDUITV3-1707] - increase default kibana time filter from 60min to 30days for correlation ids when navigating from an error message popup or correlation id details link
[CONDUITV3-793] - capture user account who created, updated and deleted connector

For more information regarding advanced logging please navigate to Conduit logs .