SymmetricDS 2.2.0 is a minor release that brings 15 bug fixes, 8 improvements, and 5 new features. Some highlights include:
- The incoming data loader was changed to use a single connection to eliminate contention with a nested transaction and reduce the connection pool requirements.
- The purge job that cleans up the runtime tables is now three separate jobs that can be scheduled independently.
- A new query plan is being used to route and batch data changes that is more efficient.
This release also includes a new project organization and reduced library dependencies in preparation for small footprint clients. We've created a new Mobile Development forum to discuss plans for an upcoming mobile implementation.
For the full release notes, click through to read more.
Release Notes - SymmetricDS - Version 2.2.0
Bug
- [SYMMETRICDS-352] - Trigger Hist row is inserted, even if the trigger generation failed.
- [SYMMETRICDS-362] - Documentation isn't being packaged during a release
- [SYMMETRICDS-367] - Nodes actually pull their configuration data 2 times during registration. It should only happen once.
- [SYMMETRICDS-368] - Possible to extract the wrong target schema when two tables of the same name are syncing
- [SYMMETRICDS-369] - SymmetricDS doesn't start up because of xml column in postgres
- [SYMMETRICDS-378] - Segmentation fault in sym_hex UDF for Firebird
- [SYMMETRICDS-382] - Oracle timestamp columns should synchronize milliseconds
- [SYMMETRICDS-384] - Authentification failed when using custom INodePasswordFilter
- [SYMMETRICDS-389] - Fix IOfflineClientListener. Add a few more possible exceptions that might cause offline.
- [SYMMETRICDS-394] - During the extract process, if there is an error, then the extract file does not get deleted.
- [SYMMETRICDS-397] - When initial.load.delete.first is set to true and a table that is being synchronized is missing, the initial load fails
- [SYMMETRICDS-399] - 22 million row initial load blew up at the very end because the stats in sym_incoming_batch became too big for the column
- [SYMMETRICDS-402] - If there is an ER on the reload channel, then other channels have to be sent to correct the error
- [SYMMETRICDS-403] - symmetric-ddl conflicts with apache ddlutils
- [SYMMETRICDS-404] - Should always check for duplicate trigger names, not just when the trigger name is > max length
Improvement
- [SYMMETRICDS-212] - Add option on sym_trigger to not record data in sym_data if none of the columns have changed
- [SYMMETRICDS-346] - Build a stripped down client only distribution of SymmetricDS
- [SYMMETRICDS-365] - Refactor JobManager so that a list of jobs can be retrieved and reported on
- [SYMMETRICDS-373] - XmlPublisherDataRouter should only publish one message, not a message per node.
- [SYMMETRICDS-374] - Only use 1 database connection while loading data.
- [SYMMETRICDS-388] - Improve the push and pull api to return a status code instead of a boolean.
- [SYMMETRICDS-400] - Log the fact that a table is not found by the data loader.
- [SYMMETRICDS-405] - Break the purge job out into 3 different jobs to match how the job is locked.
New Feature
- [SYMMETRICDS-268] - Support array data types on Postgres
- [SYMMETRICDS-342] - Add statistics at node level and at the job level. This is follow up to the new statistics in 2.1
- [SYMMETRICDS-366] - Add a sym_registration_request table that captures when a node both attempts and successfully registers
- [SYMMETRICDS-380] - Add batchid element to XML published by the XML data loader and the XML data router.
- [SYMMETRICDS-395] - Use dbms_lob.substr during an initial load on oracle if contains_big_lob column of sym_channel is set to false
Task
- [SYMMETRICDS-383] - Turn on routing.data.reader.type=gap
Eric is a software developer and technology enthusiast with a background in developing custom applications for Information Technology. As a long-time Linux user, he strongly believes in open source. He focuses on product solutions and spreading the word for JumpMind.
