Implemented full-featured support of Postgres Pro Shardman destination, including data transfer from Oracle and PostgreSQL to local Postgres Pro Shardman tables and routing the data across shards using filtering conditions.

Added validation of the procopy_options.batch_bytes configuration parameter value, which must be greater than 0, and fixed an error in processing of the default value.

Added the --enable-problem-rows flag to control saving of problematic rows. It is enabled by default. Refer to the section called “Error Handling” for details.

Improved error messages in case of different encodings of the source and destination databases.

Implemented the output of all table mapping errors. Previously, the utility exited when the first error in table or field mapping was found.

Fixed an issue that caused the utility to return a fatal error despite the completion of all the tasks.

Added the high-availability (HA cluster) mode, which includes support for heartbeats, leadership and leadership reset, query timeouts, and prohibition to complete in case of a lock.

Added routing of the data across shards of a Postgres Pro Shardman destination through filtering by conditions specified in the tasks.table.dst_shard_filter_mapping configuration parameter.

Implemented the resolution strategy for row conflicts, which is based on the prosync_options.conflict_resolve_strategy configuration parameter.

Updated generation of prosync_options.transfer_id to no longer write default configuration values to the configuration file.

Added support for the Oracle destination in an experimental mode. Refer to Table 1.1 for details.

Fixed errors that occurred during migration from Oracle Real Application Clusters (RAC) and inverse migration PostgreSQL -> Oracle.

Added automatic recognition of the source and destination database encoding to avoid data discrepancies caused by different encodings.

Added the authentication as well as security events.

Fixed a crash that could occur if Oracle geographic types were missing.

Fixed a segmentation violation that could occur when comparing custom types.

Fixed a failure on Oracle INVISIBLE columns.

Fixed the processing of hidden columns for XML and geometry types.

Added the access verification for connecting to each table during the introspection.

Changed the minimum password length to 10 characters; the default number of attempts to enter the password is limited to 4.

Added an email validation. Now specifying the email is required when a user is created/updated.

Added a possibility to update the connection without specifying a password.

Added processing of a conflict that could occur when trying to change the connection type.

Added a requirement to change the user password after changing the password of that user by an administrator.

Implemented logging of the version and configuration at the start.

Updated generation of the configuration to include an extended set of agent parameters.

Added support for case-sensitive identifiers.

Fixed the error code 404 that the API returned instead of 500 when an attempt to access a non-existing user was made.

Implemented the current user card with the profile and password change.

Optimized the operation for schemas with a large number of tables by fixing the hanging that could occur during browsing or matching tables.

Added clear error messages for errors in SQL queries (in Russian and English).

Added task parameters at the table level for mapping of schemas.

Added switches for an empty password to be used when editing PostgreSQL and Oracle connections.

Implemented global processing of the 403 error (access forbidden).

Added a possibility to exclude tables with errors in the schema–schema task type.

Fixed filtering in access groups and user groups.

Fixed the display of columns with case-sensitive names.