GridGain Ultimate Edition 8.1.2 Release Notes

Features and Improvements

  • IgniteCacheP2pUnmarshallingTxErrorTest#testResponseMessageOnUnmarshallingFailed fails [#IGNITE-5680]
  • JDBC thin driver: add example [#IGNITE-5679]
  • Web Console: Need to printout time taken evenif result set is empty [#IGNITE-5639]
  • Add complex DDL+DML test [#IGNITE-5449]
  • JDBC thin: support statement close on completion [#IGNITE-5435]
  • JDBC thin: support readOnly [#IGNITE-5426]
  • Investigate hangs in JDBC driver testIndexState() [#IGNITE-5401]
  • Do not perform cache key validation when BinaryMarshaller is used [#IGNITE-5383]
  • JDBC thin: support JdbcPreparedStatement.getParameterMetaData method [#IGNITE-5381]
  • DDL: Improve test coverage for objects with different cases (upper/lower) [#IGNITE-5279]
  • Create DDL Example [#IGNITE-5159]
  • Add ability to change end month to period [#IGNITE-4801]
  • Web Console: Refactor CSS [#IGNITE-4467]
  • ODBC: Implement DML operations with parameters in batch [#IGNITE-4370]
  • K-means clustering [#IGNITE-5113]
  • WAL queue flusher for background mode can be replaced with Timeout processor [#IGNITE-5591]
  • Archive WAL segment after significant period of grid inactivity [#IGNITE-5589]
  • Generate File WAL Segment Archive Completed Event [#IGNITE-5587]
  • Improve TcpCommunicationSpi protocol for node stopping [#IGNITE-5566]
  • Add ability to read WAL outside of an Ignite node [#IGNITE-5558]
  • ServiceProcessor recalculates all service assignments even if there is a pending topology change [#IGNITE-5552]
  • PersistentStoreConfiguration improvements [#IGNITE-5479]
  • Set 2.1.1 version as incompatible to 2.1.2 [#GG-12465]
  • Save user message in snapshot metadata [#GG-12451]
  • Add onActivate/onDeactivate methods for GridPluginProcessor [#GG-12450]
  • Add SnapshotsExampleNodeStartup in GridGainExamplesTestSuite [#GG-12378]
  • Save binary metadata to disk [#GG-12347]
  • Update license generator on the web site [#GG-12345]
  • SnapshotUtilityTest.testCommandRestore failed with NPE [#GG-12327]
  • vert.x-ignite: Update Ignite version to 2.0 [#GG-12307]
  • Snapshots utility: implement CLI for snapshot operations scheduling [#GG-12258]
  • Check should fail if some nodes don't have locally partitions which will belongs to them on restore [#GG-12233]
  • Need to return metadata cache to persistence disk storage [#GG-12212]
  • Snapshot creation may attempt to copy page which is not checkpointed [#GG-12052]
  • Visor GUI: Implement new tab with memory pools and caches metrics [#GG-11683]
  • Optimize snapshot progress tracking [#GG-12390]
  • GridDhtAtomicSingleUpdateRequest misses topologyVersion() method override [#GG-12350]
  • Snapshot creation took too much time [#GG-12270]
  • LIST command works very slowly [#GG-12236]
  • Can't invoke snapshot operation from client [#GG-12222]
  • Start using new ongoingSnapshotOperation to realize STATUS command in SnapshotUtility [#GG-12096]

Fixed

  • Skip deploy lgpl modules into repository [#IGNITE-5762]
  • Ignore non user caches when automatically choosing a queryable cache inside JDBC driver [#IGNITE-5744]
  • NPE during populated data (CacheQueryDdlExample) [#IGNITE-5735]
  • Duplicate dependencies in POM [#IGNITE-5726]
  • SQL: It should be possible to execute SQL queries over cache with no indexed types [#IGNITE-5711]
  • Web console: H2 version in pom file depends from Ignite version [#IGNITE-5710]
  • Some nodes have partitionUpdateCounter == 0 after rebalancing [#IGNITE-5701]
  • IgniteAtomicLongChangingTopologySelfTest.testClientQueueCreateCloseFailover() fails sometimes [#IGNITE-5694]
  • IgnitePartitionedSemaphoreSelfTest.testSemaphore() sometimes fails [#IGNITE-5693]
  • IgnitePdsCacheRebalancingAbstractTest suite hangs on TC [#IGNITE-5692]
  • Fix SparseDistributedMatrixTest::testCacheBehaviour crash [#IGNITE-5688]
  • Endless partition eviction during node shutdown [#IGNITE-5686]
  • JDBC: Prepared statement shouldn't clear params after query execution [#IGNITE-5685]
  • IgniteClientCacheStartFailoverTest.testClientStartLastServerFails* fails [#IGNITE-5684]
  • Web Console: missing fully qualified name for generated indexed types on Models Screen [#IGNITE-5683]
  • IgniteAtomicLongChangingTopologySelfTest.testClientQueueCreateCloseFailover fails [#IGNITE-5681]
  • ODBC: Few consecutive inserts lead to exception [#IGNITE-5663]
  • Primary index name should contain type ID or name [#IGNITE-5662]
  • Web Console: some checkbox are misaligned [#IGNITE-5656]
  • AtomicSequence usage inside transactions may cause deadlock [#IGNITE-5613]
  • Activate fails in case node was stopped with cancel=true and big objects in cache [#IGNITE-5604]
  • Web console: Agent should check that node URL is valid [#IGNITE-5599]
  • WAL Iteration stopped with exception after writting log with background mode [#IGNITE-5598]
  • Actualize VisorMemoryPolicyConfiguration, check fields [#IGNITE-5577]
  • ServiceProcessor may process failed reassignments in timeout thread [#IGNITE-5554]
  • Deadlock Detection uses IgniteCheckedException instead of TransactionTimeoutException [#IGNITE-5548]
  • NPE (PersistentStoreExample) [#IGNITE-5538]
  • CREATE INDEX failed if table has been re-created [#IGNITE-5533]
  • Assertion Error in GridDhtPartitionsExchangeFuture : msg == null [#IGNITE-5529]
  • IS_EVICT_DISABLED flag is not cleared when cache store throws an exception [#IGNITE-5528]
  • GridDeferredAckMessageSender can cause starvation in stripe pool on unstable topology [#IGNITE-5527]
  • Large near caches lead to cluster instability with metrics enabled [#IGNITE-5521]
  • IgniteChangeGlobalStateFailOverTest hangs activate on join node [#IGNITE-5520]
  • IgnitePdsCacheRebalancingAbstractTest hangs [#IGNITE-5514]
  • AssertionError: GridDhtPartitionsExchangeFuture [#IGNITE-5512]
  • SqlQuery fails with WHERE clause on LocalDateTime field [#IGNITE-5483]
  • Visor shows wrong statistics for off heap memory [#IGNITE-5461]
  • Tree is being concurrently destroyed error [#IGNITE-5454]
  • SQL: Incorrect partition is derived from query when argument type differs from column type [#IGNITE-5437]
  • GridServiceProxy does not unwraps exception message from InvocationTargetException [#IGNITE-5424]
  • Add possibility to generate alias for primary key in WebConsole [#IGNITE-5369]
  • AssertionError in index name check [#IGNITE-5340]
  • SqlFieldsQuery result set should expose fields metadata [#IGNITE-5252]
  • GridDhtPartitionDemander.requestPartitions invokes sendMessages consequently, which lead to significant increase of node start time on large clusters with ssl [#IGNITE-5232]
  • Web Console: Add support for Ignite 2.0 cluster on Queries screen [#IGNITE-5231]
  • The Unicode character in the value of a field which are included in an un-unique index will cause "stack overhead" exception [#IGNITE-5204]
  • NPE at CacheDataStoreImpl.canUpdateOldRow during load test [#IGNITE-5192]
  • ArrayIndexOutOfBoundsException in GridMergeIndexSorted [#IGNITE-5190]
  • Web console: Implement configuration of cache memory pagination [#IGNITE-5185]
  • Dependency to outdated Lucene 3.5.0 [#IGNITE-3562]
  • MOVE lost a part of the snapshot data if nodes count while moving is less then in the snapshot [#GG-11936]
  • Visor in ignite-db-x branch shows incorrect information for keys count in Heap and Off-heap [#GG-11217]
  • Closing NIO session because of unhandled exception [#GG-10949]
  • Need to create gc log for each client separately [ yardstick-ignite ] [#IGNITE-3176]
  • Snapshot message types are conflicting with SQL message types [#GG-12523]
  • Snapshot ongoing operation is requested from a wrong node [#GG-12519]
  • LIST snapshots is not working if it launches from client node [#GG-12501]
  • Restoration caches state after failed snapshot RESTORE is broken [#GG-12499]
  • Enabled logs verbose mode by default [#GG-12498]
  • Snapshot Security tests fail [#GG-12494]
  • RESTORE hangs on PHASE2 [#GG-12481]
  • SnapshotOperationType is private API, but is accessible through snapshot operation future [#GG-12466]
  • DR: DrSender send sends whole batch buffer regardless free space [#GG-12454]
  • QueueFlusher hangs on any error in FileWriteHandle.flush(WALRecord) [#GG-12418]
  • Server node does not start caches when joins during activation [#GG-12389]
  • NPE when deleting snapshots after destroying caches [#GG-12371]
  • GridSnapshotOperation.clusterNodes() is always null [#GG-12368]
  • Fix counting pages in snapshots during snapshot operations [#GG-12358]
  • Review IgniteDbSnapshotDuringTopologyChangeTest tests [#GG-12351]
  • NPE during completeSnapshotMoving operation [#GG-12348]
  • Activate cluster in all snapshot tests [#GG-12343]
  • Restore operation in snapshot-utility.sh is not restoring the data [#GG-12340]
  • Visor node throws NPE on exit [#GG-12336]
  • Connecting to grid with Visor emits GridSecurityProcessor exceptions [#GG-12333]
  • Tracking page update fails after restart on new topology [#GG-12330]
  • Cluster transition to an inactive state fails with assertion [#GG-12329]
  • Caches start after restore hangs if concurrent transactions run [#GG-12324]
  • Cancelling snapshot operation is not working [#GG-12323]
  • Restart future is not being removed after restore on client side [#GG-12322]
  • Java fatal error on trying to Deactivate a cluster [#GG-12313]
  • Snapshot restore is not working in through snaphot-utility.sh [#GG-12304]
  • SQL doesn't work [#GG-12298]
  • SNAPSHOT command failed in 8.1.1 [#GG-12287]
  • snapshot-utility is not working [#GG-12282]
  • Snapshot utility: improve error handling [#GG-12275]
  • LFS cache writer thread may die due to concurrent compaction in SYNC_BUFFER and ASYNC_BUFFERED modes [#GG-12274]
  • Snapshot creation hangs on segmentation of participating node [#GG-12264]
  • Get fails after restore from snapshot [#GG-12261]
  • Activating huge grid via visor throws (timeout) exception though activation itself finally succeeds [#GG-12247]
  • Unable to activate cluster with 1 node [#GG-12245]
  • In some reason LRS locked in live lock [#GG-12242]
  • NPE in DrSenderMetadataManager [#GG-12187]
  • improper escaping in index name [#GG-12179]
  • Assertion error in OffheapReadWriteLock leads to grid hang-up [#GG-12170]
  • ClassCastException in GridCacheContext.topology during visor-datastreamer load test [#GG-12168]
  • CANCEL of RESTORE operation in progress crashes grid node [#GG-12158]
  • CANCEL of SNAPSHOT operation in progress hangs at utility side and grid node crashes [#GG-12153]
  • CANCEL operation incorrectly reports itself as STATUS [#GG-12152]
  • DR Active-Passive test fails [#GG-12151]
  • We will lose data if we cancel snapshot restore [#GG-12140]
  • Example db-server-ext.xml delivered with leaked private IPs [#GG-12122]
  • INFO shows incorrect information aftre snapshot was moved [#GG-12106]
  • A misconfiguration of rollingUpdatesEnabled flag at client and server can lead to packet storm, resulting in OOM/NPE [#GG-12085]

.NET: Features and Improvements

  • .NET: IgniteConfiguration.ActiveOnStart [#IGNITE-5522]
  • .NET: Persistent Store tests [#IGNITE-5502]
  • .NET: PersistentStoreMetrics [#IGNITE-5491]
  • .NET: Support local collection joins in LINQ [#IGNITE-4636]
  • .NET: Build failure [#GG-12280]
  • .NET: Create Doxygen documentation [#GG-12075]
  • .NET: IGridGain.GetSnapshot() [#GG-12310]
  • .NET: NuGet version should not contain build number [#GG-12346]

.NET: Fixed

  • .NET: Negative field offset values can be observed on 2-byte offsets [#IGNITE-5716]
  • .NET: incorrect jvm.dll lookup paths for JRE [#IGNITE-5628]
  • .NET: CacheConfiguration copy constructor [#IGNITE-5629]
  • .NET: Inject resources into event listeners and filters [#IGNITE-5605]
  • .NET: InstanceResourceAttribute does not work in scan queries [#IGNITE-5588]
  • .NET: LINQ Conditional statement will fail for some types [#IGNITE-5442]
  • .NET: Stack trace is lost for exceptions in ICacheStore [#IGNITE-5441]
  • .NET: Join fails in LINQ when subquery is using Select clause [#IGNITE-4926]
  • .Net Nuget packages couldn't be created for SB releases [#GG-11578]

C++: Features and Improvements