A.1. Tungsten Clustering 5.1.1 GA (23 May 2017)

Bug Fixes

  • Tungsten Manager

    • A memory leak in the manager could cause the manager to restart after exhausting memory. The issue was most often seen when monitoring the system where the frequent update of status information.

      Issues: CT-211

Tungsten Clustering 5.1.1 Includes the following changes made in Tungsten Replicator 5.1.1

Tungsten Replicator 5.1.1 is a minor bugfix release that addresses some bugs found in the previous 5.1.0 release. It is a recommended upgrade for all users.

Bug Fixes

  • Command-line Tools

    • The dsctl command has been updated:

      • The -ascmd option has been added to output the current position as a command that you can use verbatim to reset the status. For example:

        shell> dsctl get -ascmd
        dsctl set -seqno 17 -epoch 11 -event-id "mysql-bin.000082:0000000014031577;-1" -source-id "ubuntu"
      • The -reset option has been added so that the current position can be reset and then set using dsctl set -reset without having to run two separate commands.

      Issues: CT-24

    • The availability and default configuration of some filters has been changed so that certain filters are now available in all configurations. This does not effect existing filter deployments.

      Issues: CT-84

    • The tungsten_provision_slave command could fail to complete properly due to a problem with the threads created during the provision process.

      Issues: CT-202

  • Backup and Restore

    • The trepctl backup operation could fail if the system ran out of disk space, or the storage.index file could not be written or become corrrupted. The backup system will now recreate the file if the information could be read properly.

      Issues: CT-122

  • Heterogeneous Replication

    • When creating DDL from an Oracle source for Hadoop using ddlscan, the template that is used to create the metadata file was missing.

      Issues: CT-206