Cloud CMS Release 3.2.83

The following are the new features, enhancements and fixed issues for Cloud CMS 3.2.83 release.

Upgrade Considerations

Please review the upgrade considerations provided here:
https://www.gitana.io/documentation/gitana/3.2/guide/guide/docker/upgrades/3-2-83.html

Enhancements

Branches

  • Branch merge operations now use less memory and run more quickly, allowing for larger merges to execute without incurring a large impact to allocated heap memory.

Command Line Client

  • Removed commands related to repository compression as those facilities are being moved out into MongoDB-level scripts.
  • Additional commands for delete branch and delete snapshot.

Dictionary

  • Reduction in runtime memory requirements for storing compiled dictionary state per branch, resulting in less heap allocation.

Indexing

  • Improvements to default MongoDB collection indexes that result in much faster query performance.

Java Driver

  • The Cluster.waitForJobCompletion() method will now re-attempt to read the job if an unexpected error occurs (potentially due to a network issue). Upon a second failure, null will be returned.

JavaScript Driver (v1)

  • The cluster.waitForJobCompletion() method now provides a trap handler.
  • The logic for domain ID extraction within extractPrincipalIdentifiers() is now more precise.

JavaScript Driver (v2)

  • Updates to dependency libraries, latest version of Axios.

Multilingual Feature

  • Improved the error logging for the multilingual feature so that the Node ID is included in the report.

Python Driver

  • Added new methods for CRUD operations around deployment targets.

Query

  • Added query optimization pre-parsing stage to automatically optimize some common queries for faster performance.

Relator Properties

  • Relator associations now resolve much more quickly with code paths that have been optimized to be much faster.

Transactions

  • Branch transactions now run much more efficiently and utilize less heap memory. They automatically passivate state to disk to allow for much larger transsaction sizes to run with lower memory requirements.

User Interface

  • Faster performance and lower latency with the introduction of a faster reverse HTTP proxy. This offers better socket handling and greater throughput for customers.
  • DNS caching has been improved to lower latency further by reducing the number of DNS calls during routing.
  • All internal HTTP communication has been refactored fully to use the Axios client.

Bug Fixes

API

  • Fix so that thread locals bound during the request are released via the fallback servlet listener as intended.
  • Fix so that retrieval of node version history from an archived branch works by forgoing the inclusion of attachment information if the branch view is not ready.

Branch Changes

  • Fix so that the accuracy of branch differences for cases where the source node was deleted is correct.
  • Fix so that branch changes are invalidated (instead of rebuilt) if the branch is archived.

Copy / Deploy

  • Fix so that Copy and Deploy jobs are more resilient by having them default to pulling down the full binary as part of the payload (avoiding the “preserve archive output directory” optimization).

Dictionary

  • Fix for a subtle case where the compiled flag could be prematurely set for some compiled associations.

Merge

  • Fix so that patch paths are always correctly normalized into path format ahead of calculating index adjustments during merges of array properties.

Transfer

  • Fix so that Transfer Import jobs correctly handle the import of Node Lists.
  • Fix so that branch reloads are handled more gracefully, preventing early errant termination of transfer workers.

User Interface

  • Fix so the pagination of team members works as intended within the Administration Console.
  • Fix so that direct JSON edits to documents refreshes the screen as intended.
  • Fix so that the workflow task, documents and instance screens show more information. New fields are shown by default, such as purpose.
  • Fix so that the PDF viewer consistently formats the filename of retrieved and saved documents.
  • Fix so that static files are served back with the proper MIME type (resolving issue with errant .appcache file detection).
  • Fix so branch custom indexing pages to remove pagination and filter box.









Ready to Get Started?

Unlock your data with smart content services and real-time deployment


Try Gitana for free Ask for a demo