1 - Client connectivity

You can now set the client label for vsql connections at the start of the session with the --label option.

Client labels

You can now set the client label for vsql connections at the start of the session with the --label option. Client sessions and their labels are listed in the SESSIONS system table. This option supplements the existing SET_CLIENT_LABEL function, which sets the client label for existing sessions.

JDBC database name

The database name parameter in DriverManager.getConnection() is now optional.

2 - Client drivers

You can now disable hostname validation of the OAuth identity provider with the boolean oauthvalidatehostname in OAuthJsonConfig.

OAuth: disable hostname validation

You can now disable hostname validation of the OAuth identity provider with the boolean oauthvalidatehostname in OAuthJsonConfig.

OAuthValidateHostname is set to true by default, requiring the identity provider's certificate to list the IP or hostname of the identify provider. For details, see JDBC connection properties and ODBC DSN connection properties.

ODBC request timeout

You can now specify the timeout for requests sent by ODBC clients with the ConnectionTimeout parameter. For details on this and other parameters, see ODBC DSN connection properties.

3 - Complex types

ARRAY_FIND and CONTAINS can now search for elements that satisfy a predicate by using Lambda Functions.

ARRAY_FIND and CONTAINS can now search for elements that satisfy a predicate by using Lambda functions. The new FILTER function applies a predicate to an array and returns an array containing only elements that satisfy the predicate. See Searching and Filtering.

Parquet export

EXPORT TO PARQUET now supports the ARRAY and ROW types in any combination.

4 - Containers and Kubernetes

You can authenticate to Amazon Elastic Kubernetes Service (EKS) with an Identity and Access Management (IAM) profile.

IAM profile authentication to Amazon EKS

You can authenticate to Amazon Elastic Kubernetes Service (EKS) with an Identity and Access Management (IAM) profile. For details, see Configuring communal storage.

TLS certificate management for helm charts

The VerticaDB operator can generate and manage TLS certificates with the webhook.certSource Helm chart parameter. The operator can generate certificates internally, or it can manage existing certificates.

For details about setting the webhook.certSource parameter, see Helm chart parameters.

For details about managing TLS certificates for Helm charts, see Installing the Vertica DB operator.

CA bundle management for helm charts

You can include your certificate authority (CA) bundle in the Secret provided to the webhook.tlsSecret Helm chart parameter. This parameter stores the CA bundle with its TLS certificate and keys.

Previously, you had to use the webhook.caBundle parameter to provide access to the CA bundle. This parameter is now deprecated.

For details, see Helm chart parameters.

Custom catalog path

You can create a custom catalog path in the container filesystem to store the catalog in a different location than the local data. Set the local.catalogPath Custom Resource Definition (CRD) parameter to define the custom catalog location.

For details, see Custom resource definition parameters.

Role and RoleBinding creation

The skipRoleAndRoleBindingCreation Helm chart parameter determines whether the Helm chart creates Roles and RoleBindings when you install the Helm chart. When set to true, you can install the Helm chart as a user that does not have privileges to create Roles or RoleBindings.

For details about the parameter, see Helm chart parameters.

For details about granting operator privileges, see Installing the Vertica DB operator.

5 - Documentation updates

Documentation about setting up Vertica in on-premises and cloud environments has been combined into a new Setup section.

New setup section

Documentation about setting up Vertica in on-premises and cloud environments has been combined into a new Setup section. The section contains three parts:

Information about setting up Vertica in Kubernetes remains in Containerized Vertica.

6 - Eon Mode

The RESHARDING_EVENTS system table provides information about historic and ongoing resharding operations, such as the current status of a resharding process and the timestamps for resharding events.

RESHARDING_EVENTS system table

The RESHARDING_EVENTS system table provides information about historic and ongoing resharding operations, such as the current status of a resharding process and the timestamps for resharding events.

7 - Installation and upgrade

The new --parallel-no-prompts flag tells the install script to install the Vertica binaries on hosts in parallel rather than one at a time.

Install binary packages in parallel

The new --parallel-no-prompts flag tells the install script to install the Vertica binaries on hosts in parallel rather than one at a time. Using this flag reduces the time it takes to install Vertica, especially in large clusters. See --parallel-no-prompts.

8 - Licensing and auditing

Vertica now rejects AutoPass licenses on builds that don't have Autopass License Server and returns an error code and notice, which details why the license failed to parse.

Improved handling of AutoPass licenses

Vertica now rejects AutoPass licenses on builds that don't have Autopass License Server and returns an error code and notice, which details why the license failed to parse.

9 - Loading data

The JSON and Avro parsers now produce warnings if the data being loaded contains fields that are not part of the table definition.

Warning on unmatched data fields

The JSON and Avro parsers now produce warnings if the data being loaded contains fields that are not part of the table definition. New fields are logged in the UDX_EVENTS table and can be added to the target table definition using ALTER TABLE.

For more information, see Schema evolution and the parser reference pages, FJSONPARSER and FAVROPARSER.

10 - Machine learning

You can now train autoregression models with the Yule-Walker algorithm.

Autoregression models support the yule-walker algorithm

You can now train autoregression models with the Yule-Walker algorithm. The AUTOREGRESSOR training function provides a method parameter by which you can specify either the Ordinary Least Squares or Yule-Walker training algorithm.

Jaro-winkler distance

You can now calculate the Jaro Distance and Jaro-Winkler Distance between two strings.

11 - Management Console

You can localize user interface (UI) text using language files in /opt/vconsole/temp/webapp/resources/i18n/lang.

Localize user interface text

You can localize user interface (UI) text using language files in /opt/vconsole/temp/webapp/resources/i18n/lang. The language files include the locales.json file, a list of supported languages, and a resource bundle, a directory that stores files that contain UI text strings that you can translate. By default, Vertica provides resource bundles for the following:

  • American English

  • Simplified Chinese

For details, see Localizing user interface text.

Support for r6i and c6i instance types on AWS

Management Console supports the following c6i and r6i EBS instance types on Amazon Web Services (AWS) for both Eon and Enterprise Mode:

  • c6i.4xlarge

  • c6i.8xlarge

  • c6i.12xlarge

  • c6i.16xlarge

  • r6i.4xlarge

  • r6i.8xlarge

  • r6i.12xlarge

  • r6i.16xlarge

You can select these instance types when you perform any of the following operations with Management Console:

  • Provision a database

  • Revive a database

  • Scale up a database

  • Add a new subcluster

For details about EBS volume defaults, see Eon Mode volume configuration defaults for AWS and Enterprise Mode volume configuration defaults for AWS.

Scale down and terminate subclusters with stopped nodes

You can scale down or terminate subclusters with stopped nodes in the following cloud environments:

  • Amazon Web Services (AWS)

  • Google Cloud Platform (GCP)

  • Microsoft Azure

For details, see Scaling subclusters in MC and Terminating subclusters in MC.

12 - Query optimization

Previously, event series joins allowed you to directly compare values from two series using INTERPOLATExa0PREVIOUS.

Interpolate next value

Previously, event series joins allowed you to directly compare values from two series using INTERPOLATE PREVIOUS. This feature has now been expanded, allowing you to use INTERPOLATE NEXT as well.

13 - Security and authentication

You can now specify a TLS Configuration for CREATE NOTIFIER and ALTER NOTIFIER.

TLS configuration support

Notifiers

You can now specify a TLS Configuration for CREATE NOTIFIER and ALTER NOTIFIER. This parameter supersedes the existing TLSMODE, CA BUNDLE, and CERTIFICATE notifier parameters.

To maintain backward compatibility, Vertica automatically generates TLS Configurations for notifiers that use the deprecated parameters.

CONNECT TO VERTICA

You can now specify a TLS Configuration for CONNECT TO VERTICA. This parameter supersedes the existing TLSMODE PREFER parameter.

14 - SQL functions and statements

You can now drop custom TLS Configurations.

DROP TLS CONFIGURATION

You can now drop custom TLS Configurations.

Inverse hyperbolic functions

Vertica now supports the following functions:

Directed query enhancements

The meta-function SAVE_PLANS lets you create multiple optimizer-generated directed queries from the most frequently executed queries. New columns have been added to the system table DIRECTED_QUERIES to store SAVE_PLANS-generated metadata.

Several directed query statements—ACTIVATE DIRECTED QUERY, DEACTIVATE DIRECTED QUERY, and DROP DIRECTED QUERY—now support WHERE clauses that can resolve to multiple directed queries. For example:

DEACTIVATE DIRECTED QUERY WHERE save_plans_version = 21;

15 - Vertica on the cloud

Vertica AMIs can now use the AWS Instance Metadata Service Version 2 (IMDSv2) to authenticate to AWS services, including S3.

Vertica AMIs support IMDSv2

Vertica AMIs can now use the AWS Instance Metadata Service Version 2 (IMDSv2) to authenticate to AWS services, including S3.