OrientDB 2.2 Release Notes

Luigi Dell'Aquila edited this page Jun 13, 2018 · 225 revisions

Table of Contents

2.2.37 - DRAFT

2.2.36 - June 12, 2018

Core

  • Fixed security issue with restore of malicious zip files
  • Fixed logger name when publishing Log Record

ObjectDB

  • Fixed detaching of maps in ObjectDB
  • Fixed propagation of save() on lists and sets in ObjectDB

2.2.35 - May 10, 2018

Core

2.2.34 - April 12, 2018

Studio

  • Fixed issue with Cluster Selection Dropdown - Issue #8086

Core

  • Updated dependency to concurrentlinkedhashmap - Issue #6869
  • Fix OStorageException on db open - Issue #7998
  • Fix WAL problem with full backup - Issue #8005
  • Fix indexing issue with encryption at rest enabled - Issue #8201

2.2.33 - March 5, 2018

Core

  • Fix Page File index placed in wrong list Warning - Issue #7971

SQL

  • Fix OOM Exception on Fetching Records with the help of functions - Issue #8135

2.2.32 - February 5, 2018

Core

  • Automatic conversion of tree ridbags to embedded when switching to distributed mode
  • fix DROP INDEX with invalid index name - Issue #8007

SQL

  • Fix ArrayIndexOutOfBoundsException on shortestPath() - Issue #8013

Console

  • Fix case sensitivity on CREATE DATABASE - Issue #7923

2.2.31 - December 11, 2017

Core

  • Prevent Recoverable exceptions to be logged as ERROR - Issue #7905
  • Prevent StackOverflow while starting in OSGi environment - Issue #7882

Functionalities added

Core

  • Treat JVM Error as server crash - Issue #7908
  • Add servlet context listener to init/shutdown ODB gracefully - Issue #7906
  • Manage cgroup (container) limits when calculating memory limits - Issue #7812

2.2.30 - November 14, 2017

Bugs Fixed

Core

  • Fixed CancellationException on wal close. - Issue #7808
  • In previous versions of OrientDB to get the total amount of physical memory on PC we used reflection to call a method of a class which is placed inside "sun" package. In the new version, we merely requesting attribute value of JMX bean from JMX server avoiding any limitations of JDK 9 module system - Issue #7773
  • Fixed OSGi compatibility issue - Issue #7772

Lucene

  • Fixed issue with lucene in memory due to autoclose. Issue #7844

SQL

  • Fixed NPE on Dijkstra function - Issue #7869
  • Fixed parsing issue with ALTER PROPERTIES with two blank spaced in between - Issue #7809
  • Support LET x = RID - Issue #7633

Distributed

  • Inhibit usage of SBTree Ridbags in case of distributed execution - Issue #7825

Console

  • Fixed management of WINCH signal on Win systems - Issue #7863

Functionalities added

SQL

  • Added logging of running queries in case of OOM - Issue #7861
  • Added support for keys() and values() functions on embedded documents - Issue #7795

2.2.29 - October 5, 2017

Bugs Fixed

Core

  • Fixed issue on edge creation: in case of high load graphdb cache may contain records with diff content but the same version. This could lead to lost of edge pointers in vertices. - Issue #7786

Console

  • Fix problem with upper and lowercase js/JS jss/JSS and multi-line scripts - Issue #7782

Studio

2.2.28 - September 26, 2017

Bugs Fixed

Core

  • Fix overwrite of disk cache use when set by user API - Issue #7760
  • Fix WAL compatibility issue - Issue #7756
  • Add a check to block execution of database if data corruption on disk is detected - Issue #7540

Remote

Add outdate detection and reload of storage configuration in the case of new clusters

2.2.27 - September 16, 2017

Functionality Added or Changed

Studio

  • Added default query pagination configuration Issue #494

Bugs Fixed

Core

ETL

  • Fixed "skipDuplicates" doesn't work in debug mode Issue #7682

SQL

  • Fixed NullPointerException on MATCH statement when the filter involves @class attribute
  • Fixed LIKE operator to work on links with LET and subquery
  • Fixed usage of composite index with IS NULL condition

Studio

  • Fixed Issue with server soles in cluster management Issue #492

Distributed

  • Fixed issue during fix/rollback of transactions that caused to write a wrong record count.

Enterprise

  • Improved RAM consumed by the Enterprise Profiler Issue #7623

2.2.26 - August 16, 2017

Functionality Added or Changed

Distributed

  • Upgraded Hazelcast version to 3.8.4.
    • Warning: As a result of this change it is not possible to perform a hot (rolling) upgrade from versions <= 2.2.25 to versions > 2.2.25. Rolling upgrades are supported again for upgrades between v1 and v2, where v1 and v2 are equals or higher than 2.2.26

Studio

  • Implemented removal of permission in Role UI - Issue #420
  • Add a delete button to remove an entry on security panel - Issue #420

Bugs Fixed

Core

  • Preserve superclass order - Issue #7460
  • Fix deadlock on edge unique index - Issue #7589
  • OrientDB able to automatically correct direct memory consumption even if flag XX:MaxDirectMemorySize is not set. But because OrientDB allocates direct memory by the big amount of chunks in previous versions amount of memory to be allocated in the last chunk can exceed direct memory limits. This issue was fixed. Issue #7627
  • Fix multiple usages of the same sequence in the same TX Issue #7547
  • Fix possible updated information in cache state file after shutdown/startup Issue #7601

SQL

  • Fix DELETE statement parsing with RETURN COUNT Issue #7598

Lucene

  • Lucene Spatial indices were closed incorrectly - Issue #7585 & Internal Issue #9132

Studio

  • The databases that are populated in the connection window are now sorted alphabetically - Issue #486

Distributed

  • Under certain conditions, when a node was starting up, a database was mistakenly deleted. Now fixed - Internal Issue #9064

2.2.25 - August 02, 2017

Functionality Added or Changed

Studio

  • A new 'Neo4j to OrientDB Importer' plugin that connects to Neo4j via the Neo4j remote bolt protocol has now been integrated into both OrientDB Community and Enterprise Editions. This is in addition to the already existing command-line 'Neo4j to OrientDB Importer" tool, that made use of embedded connections via the Java API (and that is still available as an external package) - Issue #432

Distributed

  • Improved java client reconnection when Data Centers are used: two new values for the -Dnetwork.retry.strategy option have now been added: same-dc and auto- Issue #7509 - Documentation

ETL

  • Added support for algorithm parameter in index creation - Issue #7581

Bugs Fixed

Core

  • Fixed Bug on JSON import in multithreading environment - Internal Issue #9020
  • Fixed bug with database backup/restoration and cache state file, cache state file is now ignored - Issue #7601 & Internal Issue #9024
  • Improved performance of rid bag compare - Internal Issue #8803

SQL

  • Fixed query parsing issue when "in" clause occurs after "or" condition with parentheses - Issue #6834
  • Fixed exception in select from collection of DATETIME or DATE, using prepared statement - Issue #7574
  • Fixed comparison of Short values with IN and CONTAINS in SQL - Issue #7578
  • Fixed StackOverflow exception when a list of enums is passed as an SQL param - Issue #7418
  • Fixed usage of include function in Match statement - Issue #271
  • Fixed security checks in SQL to allow reading of own OUser record - Issue #7549

Server

  • Implemented GZip Compression for static resources - Issue #476

Distributed

  • When executing the HA SYNC CLUSTER command, part of the cluster was not migrated. Now fixed - Internal Issue #8982
    • Warning: HA SYNC CLUSTER <cluster-name> will now trigger an automatic rebuild of the indices defined on the Class the cluster <cluster-name> belongs to. Depending on the number of defined indices and amount of data included in the Class this operation can take some time. Furthermore it is important to wait that the sync cluster operation has finished before performing any database operations that involve that cluster.
  • Fixed issue related to local DC quorum management - Issue #7539 & Internal Issue #8981
  • Enlarged default timeout for database chunk deployment - Internal Issue #8923
  • Fixed issue that was causing record version to be different on different nodes - Internal Issue #9003
  • Fixed issue with distributed sequences

Lucene Spatial

  • Fixed lifecycle on the legacy spatial index - Issue #7585

Console

  • Fixed retrieval of custom conflict strategy from remote - Internal Issue #8931

2.2.24 - July 20, 2017

Functionality Added or Changed

Packaging

As new importer plugins will be integrated into OrientDB Community Edition (after the integration of Teleporter, in version 2.2.24), the package name of OrientDB Community has now been changed in "orientdb-community-importers-major.minor.hotfix.format", e.g. "orientdb-community-importers-2.2.24.zip".

Studio

SQL

  • Query timeout strategy is now configurable at a global level using the new global property query.timeout.defaultStrategy which accepts "EXCEPTION" or "RETURN" as value - Internal issue #8715

Bugs Fixed

Core

  • Server Log was filled with warnings 'SEVER OCachePointer.finalize: readers != 0 [OCachePointer]'. Now fixed - Issue #7388
  • Under certain conditions, an OIndexException was thrown on graph.getVertices - Issue #7484
  • Database import now continues even if some single record import fails - Issue #7550
  • In case backup of the disk cache state is broken, a warning is now show (previously the server could not be started). This has been implemented as broken backup of the disk cache state does not affect the functionality of database, only its initial performance - Issue #7558
  • -d64 switch has been added on server startup scripts (server and dsever). As a result, on systems where both a 32bit and a 64bit JVM is installed, the 64bit JVM will be used by OrientDB - Issue #7553
  • In early versions of OrientDB deleted files did not preserve their ids correctly, as result after upgrade to latest version and creation of the cluster with the same name all cluster files are assigned the same id, so one file overwrites another file in file system and created cluster contains incomplete information about its state. Now fixed - Issue #7546

SQL

  • Fixed parsing of MOVE VERTEX statement inside batch scrips - Internal Issue #8854
  • Fixed timeout on queries with expand() - Internal issue #8715
  • Fixed wrong tracking flag on select insert - Issue #7538
  • Fixed binary comparison of date/datetime Issue #7557

Lucene

  • Fixed issue with OR condition and list fields - Issue #7554 & Internal Issue #8929
  • 'Laziness' on Lucene indexes (fulltext and spatial): starting with 2.2.24 only after the first access (read or write) the index is opened. If the index is in idle mode for a while (the time is configurable, 20s by default) it is closed. This results in less threads, less file handlers and less used memory. Useful in case the database has an high number of fulltext or spatial indexes rarely used - Issues #7555 and #7545

Object

  • Fixed bug on convert of nested collections of objects to document structure - Issue #7548

Server

  • Improved shutdown process with early signal to all the execution threads

Remote

HA

  • Fixed issue with backup of old database on server startup - Internal Issue #8934
  • Locks were kept for too much time as the background thread that checks the timeout was not working correctly. Now fixed - Issue #5012
  • Node-id is now used as suffix of the temporary files used for the distributed backup. This avoids that the files are overlapped in case the nodes are running all on the same machine - partial fix for internal Issue #8923
  • Recreating a database after dropping it was failing in embedded distributed mode. Now fixed - Issue #7571

Other

Neo4j to OrientDB Importer

  • Added more tests in the automated test suite
  • Migration / compatibility tests with Neo4j 3.2.2

2.2.23 - July 11, 2017

Functionality Added or Changed

Packaging

  • The Teleporter plugin is now integrated into OrientDB Community Edition (previously it was available only in the Enterprise Edition). As a result the package name of OrientDB Community has now been changed in "orientdb-teleporter-community-major.minor.hotfix.format", e.g. "orientdb-teleporter-community-2.2.23.zip"

Teleporter

  • Using Teleporter's user interface included in Studio, it is now possible to map in a visual way the source schema of the RDBMS with the graph model in OrientDB.

Bugs Fixed

Core

  • Attempting to remove a non existing field from a class instead of throwing an OSchemaException exception like in 2.1 was dumping an entire stack trace to the logger as an ERROR. As a result in certain conditions logs were filled with unnecessary stack traces, and an ERROR event instead of a WARNING was produced. Now fixed - Internal Issue #8841
  • A NPE was fixed - Internal Issue #8835
  • A NPE during database upgrade was fixed - Issue #7512
  • Under rare conditions, when -1 was being sent as record version, uniqueness constraint was violated and duplicates in the database could occur. Now fixed - Issue #7518

SQL

  • Fixed square bracket filtering on maps
  • Fixed usage of Lucene index in SQL, when multiple indexes are defined on a property and the query uses LUCENE operator - Issue #7513
  • Fixed parsing problem in INSERT FROM SELECT where the SELECT has no target - Issue #7517
  • Removed misleading log message on long running queries - Issue #6889
  • CONFLICT-STRATEGY field was empty after setting a new CONFLICTSTRATEGY using the ALTER CLUSTER command - Internal Issue #8879
  • Fixed NPE that can happen when defining custom record conflict strategies
  • Fixed SQL transactions with a mix of binary and non-binary filtering strategies - Issue #7534

Server

  • Fixed issue on error management on sessions without token - Issue #7528

Studio

  • Fixed bug with LinkList/LinkSet when editing/creating Vertex/Edge - Issue #478
  • Fixed bug on document create in browse section that didn't show the red bar in case of error - Issue #468

Distributed

  • A ClassCastException was being logged each time a distributed transaction was rolled back. Now fixed - Internal Issue #8858
  • Refreshed active session with updated storage in case of HA SYNC - Internal Issue #8804
  • Fixed bug during resurrection of a record - Internal Issue #8791

Security

  • Fixed vulnerability in SQL for queries on OUser/ORole without the privileges. An independent security researcher has reported this vulnerability to Beyond Security’s SecuriTeam Secure Disclosure program.

2.2.22 - June 19, 2017

Functionality Added or Changed

High Availability (HA)

  • Supported hot / rolling upgrade between hotfix versions when running in HA - Issue #7480
  • New SQL Command HA SET to change at runtime server role (MASTER or REPLICA), cluster's owner and database status - Issue #7471

Bugs Fixed

Core

  • Fixed bug on toLowerCase/toUpperCase that didn't take care of Locale - Issue #7464

Server

  • Fixed bug on server startup when automatic backup plugin is disabled - Issue #7473

SQL

  • Use of indexes for range queries when >= and <= operators are used has been optimized - Issue #7501

High Availability (HA)

  • Fixed bug that causes a STARTING server to never go ONLINE

Studio

  • Fixed compilation problem with node > 5 Issue #472

Lucene

  • Fixed bug in OLuceneResultSet#fetchMore with Filters Issue #7532

2.2.21 - May 31, 2017

Functionality Added or Changed

High Availability (HA)

  • Improved performance of messages exchange

Bugs Fixed

Core

  • Fixed bug on file management after restoring of 2.1.x backup on 2.2.x

Remote

  • Fixed bug on client blocked in case of error on message reading

Graph

  • Fixed integration with gremlin for result set of collections, Issue #7308
  • Heartbeat message now is Gossip message transferring the server coordinator to check consistency of distributed configuration

High Availability (HA)

  • Fixed issue with a recursive call causing a stack overflow exception. The user receives this problem as huge delay on distributed responses or random errors
  • Unable to create new users in distributed setup #Issue #7446
  • Fixed issue in StressTest tool with HA configuration about round-robin-connect strategy that caused the schema to being not updated on concurrent threads
  • Fixed locking problem with distributed transactions
  • Fixed issue when exception ODistributedRecordLockedException was raised instead of ORecordDuplicatedException
  • Fixed issue with recycling record position
  • Fixed OSGi dependency of distributed module from graphdb


2.2.20 - May 10, 2017

Functionality Added or Changed

Bugs Fixed

Core

  • Fixed retain of storage instance in the profile after close/delete - Issue #7320

Studio

Lucene

2.2.19 - April 26, 2017

Functionality Added or Changed

Core

  • Improved statistic of byte buffer exposed through JMX

SQL

  • Supported wildcards in exclude() and include() functions - Issue #4773
  • New -force and -full options in HA SYNC DATABASE command - Issue #7355

HA

  • Improved distributed logs
    • Information on who is the coordinator server are now printed
  • Added timeout in every lock request

Bugs Fixed

Core

  • Storage name now are registered in case insensitive way - Issues #7358 and #7327

Studio

  • Removed in/out properties when adding/editing new edges from the GraphEditor - Issues #7339 and #464
  • Fixed Graph in memory cache on vertices/edges add/remove - Issue #466
  • Fixed issue with the dropdown menu in New Edge Popup - Issue #465

HA

  • Fixed bug with re-election of coordinator
  • Fixed deadlock in case multiple servers went down
  • Fixed problem when nodes start at the same time
  • Fixed synchronization issue between nodes in case of high concurrency (many threads running on each available nodes and trying to modify the same vertex) - Internal Issue #8371
  • Offline nodes were automatically removed from quorum despite DISTRIBUTED_AUTO_REMOVE_OFFLINE_SERVERS was -1 - Internal Issue 8561

SQL

  • Fix index estimation on MATCH statement execution planning - Internal Issue #8562
  • Fix parsing of UPDATE statement with multiple SET/ADD/REMOVE... clauses - Issue #7362

JDBC

  • Some null values were not included in the result set: field's names mapping from documents to reusultSET has now been improved - Internal Issue #8555
  • Improved data type mapping and Spark compatibility of JDBC driver - Internal Issue #8594

2.2.18 - April 06, 2017

Functionality Added or Changed

Studio

  • Implemented notification hide on route change - Issue #437
  • Implemented History navigation with Alt + Up/Down - Issue #5425
  • Implemented a better rendering with multiple edges - Issue #428
  • Implemented same color generation in Schema and GraphEditor - Issue #453
  • Implemented refresh of metadata on Vertex/Edge creation (GraphEditor) - Issue #454
  • Implemented Vertex/Edge label template - Issue #374

Core

  • Add access of active Intent on database

HA

Console

Bugs Fixed

Core

  • Fixed minor issue in ORecordNotFoundException id retention - Issue #7172
  • Fix same content detection for document binary fields - Issue #7146
  • Fixed live query with record level security - Issue #7213
  • Fixed internal clusters detection for security and jdbc - Issue #7182
  • Fixed ReadOnly and Mandatory handling with default values for document deserialized from JSON - Issue #7211
  • Fixed asBoolean method - Issue #7284
  • Exception in data restore procedure was fixed
  • Under some circumstances an index was being marked healthy even if the rebuild process didn't complete - Issue #7274 and Internal Issue #8446
  • A unclean shutdown was forcing index rebuilds. Now fixed - Internal Issue #8495

Graph

  • Fix connection strategy management with pool factory- Issue #7110
  • Delete of edges that involves supernodes has been optimized and as a result it is now more fast - Internal Issue #8517

Server

Object

  • Fixed issue on type detection in case of domain with one to one relationship of a class that implement Serializable - Issue #7172

HA

  • Rebooted node does not fetch changes from cluster that occurred while the node was down - Internal Issue #8208
  • Fixed bug on distributed lock release

Lucene

Studio

  • Fixed visualization problem with Match - Issue #452
  • Fixed color of edges in schema panel - Issue #457
  • Fixed visibility issue in Backup Management with Firefox/IE - Issue #7195
  • Fixed bug in Schema Manager with multiple inheritance - Issue #460

ETL

  • Fixed wrong mapping of column's names when aliases are used - Issue #7307

2.2.17 - February 16, 2017

Important Notes

  • A critical issue which in very rare cases may lead to corruption of a record during update was fixed (Issue #7151). All users and Customers running OrientDB 2.2 are invited to upgrade to v. 2.2.17 or higher as soon as possible

Functionality Added or Changed

Scripts

  • New 'delay' option in stresstester script
  • Increase the default heap size on console.sh/.bat to 1GB and let configure via variable ORIENTDB_OPTS_MEMORY Issue #7169

Bugs Fixed

Core

  • Fixed regression on document constructors - Issue #7150
  • Fixed minor issue on class description management in remote - Issue #7065
  • Fixed http double response generation bug - Issue #6474
  • In very rare cases, during a record update, if the difference between the free space available in the page that will be used to contain the record and the size of the record itself is less than three (freeSpace - recordSize < 3) a record could be truncated. Now fixed - Issue #7151
  • Fixed lost of index's metadata when reopened after a complete rebuild - Issue #7125

SQL

  • Fixed a bug in the MATCH statement when 'optional' attribute was being used on empty vertex classes - Internal Issue #8289

Graph

Object

  • Fixed wrong record dirty mark on field read

JDBC

  • Improve interaction with managemnt tools about database metadata - Issue #7126

HA

  • Fixed ConcurrentModificationException when requesting the database size - Issue #7149

Neo4j to OrientDB Importer

  • Under Windows, the startup script was filing due to a code comment in a wrong position

Studio

2.2.16 - February 02, 2017

Functionality Added or Changed

Studio

  • A new OrientDB logo that includes the text "Community" or "Enterprise" is now shown according to the OrientDB Edition in use - Issue #407

Scripts

HA

  • More verbose logs when the download of a database is skipped because autoDeploy=false
  • Removed, by default, the ping to Hazelcast servers

Bugs Fixed

Core

  • Fixed Issue on binary compare of dates - Internal Issue #8227
  • Changed create callback parameter of async save for receive the valid new rid instead of invalid presave rid - Issue #7105
  • Fixed bug on dirty management during delete using record api

Graph

  • Fixed regression introduced in 2.2.15 during vertex delete that caused stale pointers - Issue #7118
  • Fixed graph query of properties with kebab-case issue - Issue #7128

SQL

  • SQL UUID function was returnng a Java Object instead of a String - Issue #7116
  • Fixed a NPE on gremlin() function execution with invalid input - Issue #5714
  • Fixed a NPE on iteration of a mix of regular and lightweight edges - Issue #7018
  • SQL A* function on was returning wrong results if the same query was being executed multiple times - Issue #7103

ETL

  • Fixed NPE when database type is graph and "dbAutoDropIfExists": true is configured - Issue #7109 and Issue #7119

Lucene

  • Add to indexes new metadata fields _CLASS and _CLUSTER to enable partitioned queries - Issue #6594

Studio

  • Fulltext Lucene Index was being created as SBTREE - Issue #442
  • 'SEVER' was added as log type in the Studio Servers Management's Logs tab - Internal Issue #8176
  • Edge properties were ignored when creating a new edge by clicking on a vertex from the Graph Editor - Issue #438

High Availability

  • Under certain conditions, synchronization of a node was failing - Issues #6999 and #7073
  • An "ODistributedStartupException: Error on registering local node on cluster" was happening when starting a cluster node from a system where Java 7 (instead of 8) was installed. Compatibility with Java 7 has now been restored - Issue #7059
  • Fixed stall in case the current database is newer than the one in the cluster

Scripts

  • On console.sh split memory option and ssl options - Issue #7084

2.2.15 - January 19, 2017

Functionality Added or Changed

Studio

  • Internal: Release process has been improved: a pom.xml file has now been added and Studio snapshots are now being published - Issue #426
  • Internal: A migration to WebPack and Babel has been done and Angular 2 has been added as dependency - Issue #419
  • The Schema panel has been improved: now there are two main categories of classes: "User" Classes and "System" Classes. In the "User" Classes there is an additional separation between "Vertex", "Edge" and "Generic" Classes - Issue #21
  • New forms to create Vertex, Edge and Generic Classes, along with their buttons, have been added in the Schema panel. It is now possible to specify number of clusters during class creation - Issue #431

ETL

  • Logging in case of errors is improved with input data, executed command and exception message - Issue #6872

Neo4j to OrientDB Importer

  • A new command line option -i, default to false, has been added. When -i is true the Neo4j to OrientDB Importer will create a unique index on the property Neo4jRelID, for all migrated edge classes. This will allow users to query relationships by original Neo4j relationship ids - Issue #9
  • The value of the parameter MaxDirectMemorySize has been rised from 4g to 512g in the startup scripts to avoid possible OOM errors during the import - Issue #14
  • Internal: The index MultipleLabelNeo4jConversion.Neo4jLabelList is rebuilt before starting schema migration. Affects only migrations from databases that include Neo4j nodes with multiple labels - Issue #12

Bugs Fixed

Storage

  • A NullPointerException was happening when calling ODatabaseDocumentTx.open() after all indexes were rebuilt - Issue #6542
  • A NullPointerException was happening when calling index.getVersion() - Issue #6651
  • Package sun.nio.ch in artifact orientdb-core was mistakenly marked as non-optional. Marked as optional now - Issue #7042
  • Inside a transaction, the parameter fromInclusive was ignored by the iterateEntriesMajor function - Issue #7052
  • A NullPointerException was happening in OSBTree.findBucket - Issue #7053
  • A NullPointerException was happening in OEngineLocalPaginated.changeCacheSize - Issue #7088

SQL

  • MATCH statement with conditions on $matched was returning incorrect results - Issue #6931
  • SET was not working correctly with double numbers in scientific notation - Issue #7013

Studio

  • When editing a function, the "Save" command was not enabled when changing the 'idempotent' checkbox - Issue #423
  • When creating a Role, the "Inherited Role" was not filled, even if saved correctly - Issue #421
  • The "In Edges" and "Out Edges" frames were mistakenly enabled during new Vertex creation before the Vertex was created - Issue #429
  • A refresh issue has been fixed when editing a Vertex: created edges were not shown even if correctly created - Issue #325
  • Backticks are now used in all Studio-generated queries - Issue #366
  • "Source" and "Target" vertex frames are now shown when editing an Edge, instead of the "In Edges" and "Out Edges" frames - Issue #435
  • When moving from the "Graph Editor" tab to another tab, and then returning to the Graph Editor, the previous query was lost. Now it appears again, like the graph. When logging into another database, the graph of the previous database was loaded. Now it is not loaded anymore - Issue #6923
  • Edges were not visualized in the Graph Editor if queried directly through a query like SELECT * FROM E - Issue #7039

Supportability

  • Fixed an issue that caused "java.lang.Class" to be written in the log instead of "OMemoryAndLocalPaginatedEnginesInitializer" - Issue #6962

Other

  • An export and import operation where lucene indexes were involved was resulting in empty indexes - Issue #6977
  • The script gremlin.sh was missing logic to figure out symlinks - Issue #7057