DataStax Enterprise 4.6 Release notes

DataStax Enterprise release notes cover components, changes and enhancements, issues, and resolved issues for DataStax Enterprise 4.6.x releases.

Before you upgrade 

The latest version of DataStax Enterprise 4.6 is 4.6.13.

DataStax Enterprise 4.6.13 release notes 

18 August 2016

4.6.13 Component changes
  • Apache Cassandra™ 2.0.17.1420
4.6.13 Resolved issue
  • TTL collector shouldn't read whole rows from Cassandra. (DSP-9446)
  • Spark Worker error logs during shutdown. (DSP-10128)

4.6.13 Apache Cassandra™ changes

Familiarize yourself with the changes and features in the new Apache Cassandra™ version. A complete list of changes is available in CHANGES.txt. The following changes identify only the DataStax Enterprise production certified changes that are in addition to Apache Cassandra™ 2.0.17:
  • Avoid writing range tombstones after END_OF_ROW marker (CASSANDRA-10791)

DataStax Enterprise 4.6.12 release notes 

21 April 2016

4.6.12 Component changes
  • Apache Cassandra™ 2.0.17.903
  • Apache Solr 4.6.0.3.413
  • Apache Tomcat 6.0.45
  • Netty 4.0.19.Final
4.6.12 Changes and enhancements
  • Add a warning after dsetool create_core if reindex=true is not specified. (DSP-7631)
  • Provide user credentials in a configuration file for use with dse and dsetool. (DSP-8128)
  • TomcatSolrRunner should get the cipher suites for SSL connections from cassandra.yaml. (DSP-8370)
  • Make EverywhereStrategy return the correct primary endpoint from calculateNaturalEndpoints. (DSP-8507)
4.6.12 Resolved issues
  • Throw error if Solr schema unique key has a tokenized field. (DSP-5431)
  • Solr slow sub-query request modification consumes POST content. (DSP-6609)
  • Removed byoh-env.sh: PIG_PARTITIONER='dsetool partitioner' call to speed up BYOH start. (DSP-6962)
  • After creating Thrift table for data import handler, unable to create valid Solr core. (DSP-6982)
  • Conversion of unordered docids into bitsets for a Solr join query. (DSP-7021)
  • User latency tracking no longer performs a reverse DNS lookup. (DSP-7252)
  • Exclude hadoop-client lib from Spark. (DSP-7253)
  • Solr and demo apps should be served from CATALINA_HOME. (DSP-7414)
  • Bootstrapping a new data center reverts dse_perf keyspace to SimpleStrategy. (DSP-7845)
  • Multiple CQL Solr query fq clauses should work. (DSP-7975)
  • EverywhereStrategy should calculate natural endpoints in sorted order. (DSP-8023)
  • Fix example JMX remote password file location in cassandra-env.sh. (DSP-8245)
  • Support all inter-node (node-to-node) encryption options in DSE Search legacy netty transport. Use server_encryption_options properties in cassandra.yaml. (DSP-8374)
  • Stop optimizing in the commit that occurs at the end of reindexing a core. (DSP-8613)
  • Node should exit when plugin fails to start (deadlock in PluginManager.activate/deactivate). (DSP-8648)
  • Avoid reading from Cassandra during first phase of CQL Solr Query without lazy field loading. (DSP-8833)
  • Solr rounds the timestamp key to second if used as primary key when distributed. (DSP-8879)
  • Document deletion fails due to incorrect escaping of Solr special characters in unique key. (DSP-8884)
  • Fix duplicate results from Solr query. (DSP-8936)
  • CFS keyspace warning about phantom datacenter DC1 or UNKNOWN-DC. (DSP-9246)

4.6.12 Apache Cassandra changes

Familiarize yourself with the changes and features in the new Apache Cassandra™ version. A complete list of changes is available in CHANGES.txt. The following changes identify only the DataStax Enterprise production certified changes that are in addition to the Apache Cassandra™ 2.0.17 version:
  • Fix streaming_socket_timeout_in_ms not enforced (CASSANDRA-11286)
  • Make paging logic consistent between searcher impls (CASSANDRA-10683)
  • Merge range tombstones during compaction (CASSANDRA-7953)
  • Fix pending range calculation during moves (CASSANDRA-10887)
  • Gossiper#isEnabled is not thread safe (CASSANDRA-11116)
  • CrcCheckChance should adjust based on live CFMetadata not sstable metadata (CASSANDRA-7978)
  • Reduce garbage on codec flag deserialization (CASSANDRA-7244)
  • Avoid early loading of non-system keyspaces before compaction-leftovers cleanup at startup (CASSANDRA-6913)
  • Throw IRE if a prepared statement has more markers than supported (CASSANDRA-5598)
  • Change snapshot response message verb to INTERNAL to avoid dropping it (CASSANDRA-6415)
  • Fix cache persistence when both row and key cache are enabled (CASSANDRA-6413)
  • Don't list CQL3 table in CLI describe even if named explicitly (CASSANDRA-5750)
  • (Hadoop) fix quoting in CqlPagingRecordReader and CqlRecordWriter (CASSANDRA-5824)
  • (Hadoop) fix support for Thrift tables in CqlPagingRecordReader (CASSANDRA-5752)
  • Leveled compaction performs size-tiered compactions in L0 (CASSANDRA-5371, 5439)
  • drop unnecessary keyspace parameter from user-defined compaction API (CASSANDRA-5139)
  • Fix tracing when operation completes before all responses arrive (CASSANDRA-5668)
  • Have BulkLoader ignore snapshots directories (CASSANDRA-5587)
  • Don't throw away initial causes exceptions for internode encryption issues (CASSANDRA-5644)
  • Connection thrashing during multi-region ec2 during upgrade, due to messaging version (CASSANDRA-5669)
  • Avoid duplicate index entries ind PrecompactedRow and ParallelCompactionIterable (CASSANDRA-5395)
  • Remove the index entry on oldColumn when new column is a tombstone (CASSANDRA-5395)
  • Invalid streamId in cql binary protocol when using invalid CL (CASSANDRA-5164)
  • Copy resultSet on count query to avoid ConcurrentModificationException (CASSANDRA-5382)
  • Fix cf name extraction from manifest in Directories.migrateFile() (CASSANDRA-5242)
  • Avoid allocating SSTableBoundedScanner during repair when the range does not intersect the sstable (CASSANDRA-5249)
  • Add threadpool size adjustment methods to JMXEnabledThreadPoolExecutor and CompactionManagerMBean (CASSANDRA-5044)
  • Apply disk_failure_policy to bad disks on initial directory creation (CASSANDRA-4847)
  • Add debug logging to list filenames processed by Directories.migrateFile method (CASSANDRA-4939)
  • Adjust blockFor calculation to account for pending ranges due to node movement (CASSANDRA-833)
  • (CQL3) Make prepared statement global instead of per connection (CASSANDRA-4449)
  • (CQL3) Fix validation when using counter and regular columns in the same table (CASSANDRA-4706)
  • (cql3) Force provided columns in clustering key order in 'CLUSTERING ORDER BY' (CASSANDRA-4881)
  • Fix assumption error in CLI when updating/describing keyspace (CASSANDRA-4322)
  • Fix re-created keyspace disappearing after 1.1.5 upgrade (CASSANDRA-4698, 4752)
  • Fix cassandra-env.sh on RHEL and other non-dash-based systems (CASSANDRA-4494)
  • Flush based on data size, not throughput; overwritten columns no longer artificially inflate liveRatio (CASSANDRA-4399)
  • Identify and blacklist corrupted SSTables from future compactions (CASSANDRA-2261)
  • Kick off background compaction when min/max thresholds change (CASSANDRA-4279)
  • Respect 1.0 cache settings as much as possible when upgrading (CASSANDRA-4088)
  • Relax path length requirement for SSTable files when upgrading on non-Windows platforms (CASSANDRA-4110)
  • Display byte data instead of erroring out on decode failure (CASSANDRA-3874)
  • Recycle commitlog segments for improved performance (CASSANDRA-3411, 3543, 3557, 3615)
  • Only snapshot CF being compacted for snapshot_before_compaction (CASSANDRA-3803)
  • (CQL) Proper ColumnFamily metadata validation on CREATE COLUMNFAMILY (CASSANDRA-3565)
  • Fix compression "chunk_length_kb" option to set correct kb value for thrift/avro (CASSANDRA-3558)
  • Validate compression parameters on add/update of the ColumnFamily (CASSANDRA-3573)
  • Avoid rescheduling compactions when no compaction was executed (CASSANDRA-3484)
  • Fix bug that caused first column in per row indexes to be ignored (CASSANDRA-3441)
  • Optimize UUIDGen to avoid lock contention on InetAddress.getLocalHost (CASSANDRA-3387)
  • Post-task execution, enabling more consistent leveled compaction (CASSANDRA-3391)
  • Remove procrun (bin\daemon) from Cassandra source tree and artifacts (CASSANDRA-3331)
  • Prevent nodes that failed to join from hanging around forever (CASSANDRA-3351)
  • Evict gossip state immediately when a token is taken over by a new IP (CASSANDRA-3259)
  • Remove dynamic_snitch boolean from example configuration (defaulting to true) and set default badness threshold to 0.1 (CASSANDRA-3229)
  • Fix deadlock in commit log during flush (CASSANDRA-3253)
  • Replace CollatingIterator, ReducingIterator with MergeIterator (CASSANDRA-2062)
  • Fixed the ability to set compaction strategy in cli using create column family command (CASSANDRA-2778)
  • Store hints in the coordinator node instead of in the closest replica (CASSANDRA-2914)
  • Allow topology sort to work with non-unique rack names between datacenters (CASSANDRA-3152)
  • Fix of numeric-only and string column names handling in CLI "drop index" (CASSANDRA-3054)
  • Avoid retaining references to dropped CFS objects in CompactionManager.estimatedCompactions (CASSANDRA-2708)
  • Change TokenRing.endpoints to be a list of rpc addresses instead of listen/broadcast addresses (CASSANDRA-1777)
  • Keep gossipped version in sync with actual on migration coordinator (CASSANDRA-2946)
  • Flush memtables on shutdown when durable writes are disabled (CASSANDRA-2958)
  • Fix bug where dirty commitlog segments were removed (and avoid keeping segments with no post-flush activity permanently dirty) (CASSANDRA-2829)
  • Add asynchronous and half-sync/half-async (hsha) thrift servers (CASSANDRA-1405)
  • Fix potential use of free'd native memory in SerializingCache (CASSANDRA-2951)
  • Adjust hinted handoff page size to avoid OOM with large columns (CASSANDRA-2652)
  • Faster flushes and compaction from fixing excessively pessimistic rebuffering in BRAF (CASSANDRA-2581)
  • Disable compaction throttling during bootstrap (CASSANDRA-2612)
  • Add support for presenting row key as a column in CQL result sets (CASSANDRA-2622)
  • Remove checking all column families on startup for compaction candidates (CASSANDRA-2444)
  • Refuse to apply migrations with older timestamps than the current schema (CASSANDRA-2536)
  • Support for columns that act as incr/decr counters (CASSANDRA-1072, 1937, 1944, 1936, 2101, 2093, 2288, 2105, 2384, 2236, 2342, 2454)
  • CQL (CASSANDRA-1703, 1704, 1705, 1706, 1707, 1708, 1710, 1711, 1940, 2124, 2302, 2277, 2493)
  • Count a row deletion as one operation towards memtable threshold (CASSANDRA-2519)
  • Avoid exporting an un-requested row in sstable2json, when exporting a key that does not exist (CASSANDRA-2168)
  • Refactor stress.py to have only one copy of the format string used for creating row keys (CASSANDRA-2108)
  • Fix Cassandra cli to respect timeout if schema does not settle (CASSANDRA-2187)
  • Fix for compaction and cleanup writing old-format data into new-version sstable (CASSANDRA-2211, -2216)
  • Improve detection and cleanup of partially-written sstables (CASSANDRA-2206)
  • Fix CFMetaData.apply to only compare objects of the same class (CASSANDRA-1962)
  • Ignore messages from newer versions, keep track of nodes in gossip regardless of version (CASSANDRA-1970
  • Add cli support for setting username/password at 'connect' command (CASSANDRA-2111)

DataStax Enterprise 4.6.11 release notes 

16 November 2015

4.6.11 Component changes
  • Apache Cassandra™ 2.0.17.858
  • Thrift 0.9.3
4.6.11 Resolved issues
  • Error loading class 'solr.FastLRUCache' in search. (DSP-4702)
  • Disallow setting Solr concurrency level at 1 via JMX. (DSP-6868)
  • SSL + Kerberos qop=auth-int or auth-conf not reported as warning. (DSP-6939)
  • Better dsetool core name error handling. (DSP-6953)
  • Possible out of memory error from malformed Thrift protocol messages. (DSP-7019)

4.6.11 Apache Cassandra™ changes

Familiarize yourself with the changes and features in the new Apache Cassandra™ version. A complete list of changes is available in CHANGES.txt. The following changes identify only the DataStax Enterprise production certified changes that are in addition to the Apache Cassandra™ 2.0.17 version.
  • Fix sorting (ORDER BY) for queries with an IN condition on partition key columns (CASSANDRA-10363)
  • Fix rare race where older gossip states can be shadowed (CASSANDRA-10366)
  • Fix consolidating racks violating the RF contract (CASSANDRA-10238)
  • Disallow decommission when node is in drained state (CASSANDRA-8741)

DataStax Enterprise 4.6.10 release notes 

8 October 2015

4.6.10 Component changes
  • Apache Cassandra™ 2.0.16.762
  • Apache Solr 4.6.0.3.199
4.6.10 Changes and enhancements
  • Use BIO Tomcat connector to improve DSE Search throughput. (DSP-6302)
  • Allow for configuring the NIO connector via Tomcat server.xml. (DSP-6645)
    The BIO connector is more performant. However, you can revert back to NIO by setting the protocol to org.apache.coyote.http11.Http11NioProtocol in the Tomcat server.xml file under ~/dse/resources/tomcat/conf:
    <Connector port="${http.port}" protocol="org.apache.coyote.http11.Http11NioProtocol"
    connectionTimeout="20000" 
    redirectPort="8443" URIEncoding="UTF-8"/>
    After you start DataStax Enterprise, you can verify the NIO connector with these lines in the Tomcat Catalina log file:
    Sep 14, 2015 11:22:39 PM org.apache.coyote.http11.Http11NioProtocol init
    INFO: Initializing Coyote HTTP/1.1 on http-8983
    Sep 14, 2015 11:22:41 PM org.apache.coyote.http11.Http11NioProtocol start
    INFO: Starting Coyote HTTP/1.1 on http-8983
4.6.10 Resolved issues
  • The default value of load_max_time_per_core is changed to 5 minutes. (DSP-6260)
  • Deleted doc could still show up in DSE Search query when using facet tagging and excluding filters or query result cache is on. (DSP-6529)
  • 'dse beeline' failed with password auth. (DSP-6598)
  • Top deletion may cause unwanted deletes from the index. (DSP-6654)
  • FuzzySet integer min value causes assertion bug. (DSP-6658)
  • Solr slow sub-query log doesn't report prepare and process times for CQL Solr queries. (DSP-6724)
  • Unable to use multiple fq in a CQL Solr query. (DSP-6763)
  • Solr CQL queries fail with non-compound timestamp keys. (DSP-6810)
4.6.10 Apache Cassandra™ changes
  • Backport CASSANDRA-8013 to 2.0 (CASSANDRA-10144)
  • Make getFullyExpiredSSTables less expensive (CASSANDRA-9882)
  • Add tool to find why expired sstables are not getting dropped (CASSANDRA-10015)
  • Remove erroneous pending HH tasks from tpstats/jmx (CASSANDRA-9129)
  • Don't cast expected bf size to an int (CASSANDRA-9959)

DataStax Enterprise 4.6.9 release notes 

4.6.9 Resolved issues
  • Range tombstones might cause unwanted deletes from the index. (DSP-6301)

DataStax Enterprise 4.6.8 release notes 

4.6.8 Component changes
  • Apache Cassandra™ 2.0.16.678
  • Apache Solr 4.6.0.3.113
  • Apache Tomcat 6.0.44
  • Shark 1.1.2
  • Spark Cassandra Connector 1.1.2
  • Apache Spark 1.1.1.1
4.6.8 Changes and enhancements
  • Provide a way to disable DSE from adding _partitionKey field to Solr documents. (DSP-6021)
  • Distribution of requests is more even while using shard.shuffling.seed. (DSP-6127)
4.6.8 Resolved issues
  • Compaction of files stored on the cfs-archive layer is not disabled, and files are compacted automatically. (DSP-4081)
  • Spark Workers inherit JAVA_HOME from Spark Submitter Application. (DSP-4832)
  • Disable SSLv3 in the TomcatSolrRunner. (DSP-5070)
  • Spark Connector Java API the spark.cassandra.input.split.size does not change the number of splits. (DSP-5238)
  • Hive queries fail with Thrift size error. (DSP-5247)
  • New dsetool command to automatically generate schema at stdout. (DSP-5373)
  • DSE spark-submit --deploy-mode cluster does not pass credentials. (DSP-5451)
  • Sparksql query fails with HiveContext against secure cluster. (DSP-5493)
  • StreamingContext job submission fails against a secure cluster while using textFileStream. (DSP-5536)
  • RETRIEVE phase of named CQL Solr queries not recorded properly. (DSP-5643)
  • DSE hangs on restart with empty XML config files after unclean shutdown. (DSP-5645)
  • Demos: Test for dse in path not correct. (DSP-5671)
  • Solr CQL index may handle RangeTombstones incorrectly. (DSP-5940)
  • Bootstrapping TDE nodes try to read from themselves. (DSP-5701)
  • HIVE - INNER JOIN, replacing nulls with other rows data. (DSP-5994)
  • Netty client starvation with more nodes than the configured max connections. (DSP-6061)
  • DSE unique key bloom filter is not being used when the unique key is composite. (DSP-6090)
  • DSE Search should shutdown the related threads after dropping cores. (DSP-6110)
4.6.8 Apache Cassandra™ changes
  • Log when messages are dropped due to cross_node_timeout (CASSANDRA-9793)
  • checkForEndpointCollision fails for legitimate collisions (CASSANDRA-9765)
  • Complete CASSANDRA-8448 fix (CASSANDRA-9519)
  • Don't include auth credentials in debug log (CASSANDRA-9682)
  • Can't transition from write survey to normal mode (CASSANDRA-9740)
  • Avoid NPE in AuthSuccess#decode (CASSANDRA-9727)
  • Add listen_address to system.local (CASSANDRA-9603)
  • Bug fixes to resultset metadata construction (CASSANDRA-9636)
  • Fix setting 'durable_writes' in ALTER KEYSPACE (CASSANDRA-9560)
  • Avoid ballot clash in Paxos (CASSANDRA-9649)
  • Improve trace messages for RR (CASSANDRA-9479)
  • Fix suboptimal secondary index selection when restricted clustering column is also indexed (CASSANDRA-9631)
  • (cqlsh) Add min_threshold to DTCS option autocomplete (CASSANDRA-9385)
  • Fix error message when attempting to create an index on a column in a COMPACT STORAGE table with clustering columns (CASSANDRA-9527)
  • 'WITH WITH' in alter keyspace statements causes NPE (CASSANDRA-9565)
  • Display min timestamp in sstablemetadata viewer (CASSANDRA-6767)
  • Scrub (recover) sstables even when -Index.db is missing, (CASSANDRA-9591)
  • Fix growing pending background compaction (CASSANDRA-9662)
  • Don't track hotness when opening from snapshot for validation (CASSANDRA-9382)
  • Expose some internals of SelectStatement for inspection (CASSANDRA-9532)
  • ArrivalWindow should use primitives (CASSANDRA-9496)
  • Periodically submit background compaction tasks (CASSANDRA-9592)
  • Set HAS_MORE_PAGES flag to false when PagingState is null (CASSANDRA-9571)
  • Backport indexed value validation fix from CASSANDRA-9057 (CASSANDRA-9564)
  • Don't accumulate more range than necessary in RangeTombstone.Tracker (CASSANDRA-9486)
  • Add broadcast and rpc addresses to system.local (CASSANDRA-9436)
  • Always mark sstable suspect when corrupted (CASSANDRA-9478)
  • Add database users and permissions to CQL3 documentation (CASSANDRA-7558)
  • Allow JVM_OPTS to be passed to standalone tools (CASSANDRA-5969)
  • Fix bad condition in RangeTombstoneList (CASSANDRA-9485)
  • Fix potential StackOverflow when setting CrcCheckChance over JMX (CASSANDRA-9488)
  • Fix null static columns in pages after the first, paged reversed queries (CASSANDRA-8502)
  • Fix failing bound statement after adding a collection (CASSANDRA-9411)
  • Fix counting cache serialization in request metrics (CASSANDRA-9466)
  • (cqlsh) Add LOGIN command to switch users (CASSANDRA-7212)
  • Clone SliceQueryFilter in AbstractReadCommand implementations (CASSANDRA-8940)
  • Push correct protocol notification for DROP INDEX (CASSANDRA-9310)
  • token-generator - generated tokens too long (CASSANDRA-9300)
  • Add option not to validate atoms during scrub (CASSANDRA-9406)
  • Backport CASSANDRA-3569 (CASSANDRA-9455)

DataStax Enterprise 4.6.7 release notes 

4.6.7 Component changes
  • Apache Cassandra™ 2.0.14.459
  • Apache Solr 4.6.0.3.5

4.6.7 Resolved issues

  • Fix Ec2MultiRegionSnitch not reconnecting on private IP during upgrade. (DSP-4748)
  • Do not store spatial types in Lucene index by default. (DSP-4694)
  • CQL Solr queries do not error out when additional WHERE clauses are present. (DSP-5109)
  • DocTransformers do not work for DSE Search. (DSP-5201)
  • CQL Solr queries are not registered in the QueryMetrics RETRIEVE phase. (DSP-5223)
  • Bootstrap of Cassandra-only nodes delayed by unnecessary Solr index build. (DSP-5273)
  • Remove Spark run-example script from DSE build. (DSP-5328)
  • java.lang.ArrayIndexOutOfBoundsException when no keyspace.cf provided in dsetool. (DSP-5384)
  • Sparksql query fails with HiveContext against secure cluster. (DSP-5480)
  • Audit log ttl isn't being applied to entries. (DSP-5571)
  • Extend request handling and processing to support custom searchers. (DSP-5075)
  • Password might appear in audit log when creating or altering a user. (DSP-5593)
4.6.7 Apache Cassandra™ changes
  • Fix counting of tombstones for TombstoneOverwhelmingException (CASSANDRA-9299)
  • Fix ReconnectableSnitch reconnecting to peers during upgrade (CASSANDRA-6702)
  • Include keyspace and table name in error log for collections over the size limit (CASSANDRA-9286)
  • Avoid potential overlap in LCS with single-partition sstables (CASSANDRA-9322)
  • Log warning message when a table is queried before the schema has fully propagated (CASSANDRA-9136)
  • Overload SecondaryIndex#indexes to accept the column definition (CASSANDRA-9314)
  • (cqlsh) Add SERIAL and LOCAL_SERIAL consistency levels (CASSANDRA-8051)
  • Fix streaming hang when retrying (CASSANDRA-9132)

DataStax Enterprise 4.6.6 release notes 

4.6.6 Component changes
  • Cassandra 2.0.14.425
  • Apache Spark 1.1.0.4
4.6.6 Changes and enhancements
  • Allow passing JMX username and password to DSE commands (DSP-5217)

4.6.6 Resolved issues

  • Bad token to string in log message (DSP-2970)
  • Index updates fail due to Assertion error in Cassandra (DSP-4663)
  • Audit log recorded table_name as null when deleting index (DSP-4721)
  • Audit log recorded table_name as null when deleting index (DSP-5049)
  • Node slow log should use UTC timezone instead of local timezone (DSP-5125)
  • Enable FieldInputTransformer for multivalue fields (DSP-5138)
  • Addressing corner cases related to existing solr_query column before Solr index is created on table (DSP-5186)
  • dsetool reload_core does not update schema (DSP-5204)
  • Bootstrap of Cassandra-only nodes is delayed by unnecessary Solr index build (DSP-5273)
  • Lock when reading from Cassandra using Spark (Remote ClassLoading Issue) (DSP-5389)
4.6.6 Patches
  • Fix index selection during rebuild with certain table layouts (CASSANDRA-9281)
  • Fix partition-level-delete-only workload accounting (CASSANDRA-9194)
  • Allow scrub to handle corrupted compressed chunks (CASSANDRA-9140)
  • Fix assertion error when resetlocalschema is run during repair (CASSANDRA-9249)
  • Disable single sstable tombstone compactions for DTCS by default (CASSANDRA-9234)
  • Do more agressive ttl expiration checks to be able to drop more sstables (CASSANDRA-8243)
  • Incoming TcpConnection thread is not named (CASSANDRA-9262)
  • Close incoming connections when MessagingService is stopped (CASSANDRA-9238)
  • Avoid overflow when calculating max SSTable size in LCS (CASSANDRA-9235)
  • Make SSTable blacklisting work with compression (CASSANDRA-9138)
  • Do not attempt to rebuild indexes if no index accepts any column (CASSANDRA-9196)
  • Do not initiate snitch reconnection for dead states (CASSANDRA-7292)
  • Fix ArrayIndexOutOfBoundsException in CQLSSTableWriter (CASSANDRA-8978)
  • Add shutdown gossip state to prevent timeouts during rolling restarts (CASSANDRA-8336)
  • Fix running with java.net.preferIPv6Addresses=true (CASSANDRA-9137)
  • Fix failed bootstrap/replace attempts being persisted in system.peers (CASSANDRA-9180)
  • Flush system.IndexInfo after marking index built (CASSANDRA-9128)
  • Fix updates to min/max_compaction_threshold through cassandra-cli (CASSANDRA-8102)
  • Backport CASSANDRA-6863 without digest updates (CASSANDRA-8989)
  • Do not include tmp files when doing offline relevel (CASSANDRA-9088)
  • Use the proper CAS WriteType when finishing a previous round during Paxos preparation (CASSANDRA-8672)

DataStax Enterprise 4.6.5 release notes 

4.6.5 Components
  • Cassandra 2.0.14.352
4.6.5 Changes and enhancements
  • Support for Java 8

4.6.5 Resolved issues

  • Revert CASSANDRA-7816 gossip changes (DSP-5205)

DataStax Enterprise 4.6.4 release notes 

4.6.4 Components
  • Cassandra 2.0.14.348
4.6.4 Resolved issues
  • Cannot resize MOS table if OldSize+NewSize > memtable_total_space_in_mb (DSP-4240)
  • Spark SQL "drop table if exists" gives error if table does not exist (DSP-4387)
  • sstableloader incorrectly pulls directory information from cassandra.yaml (DSP-4630)
  • Restarting a node after an abrupt shutdown might cause index data loss (DSP-4802)
  • Issuing "dsetool jobtracker" on non-Hadoop workload nodes causes an exception (DSP-4835)
  • Temporary directories for Spark are created for commands other than during startup (DSP-4939)
  • A limitation of the current Sqoop import requires at least one value column (non-primary key) to do an import (DSP-5043)
  • Make updates replayed by the Solr commit log on startup immediately searchable (DSP-5095)
  • Creating Solr_query column before using dsetool causes Solr queries in CQL to fail (DSP-5105)
  • slow_sub_query_log with multi-variable parameters prints an object reference instead of the actual parameters that are passed in (DSP-5106)
  • DSE fails to start when config_encryption_active = True; LDAP options are now commented out by default in dse.yaml (DSP-5112)
  • dsetool should default to localhost, not the broadcast address for JMX. (DSP-5122)
  • A system key created with AES/CBC/PKCS5Padding encryption cannot be initialized (DSP-5159)
  • Solr auto soft commits are erroneously distributed (DSP-5165)

4.6.4 Patches

  • Avoid race in cancelling compactions (CASSANDRA-9070)
  • More aggressive check for expired sstables in DTCS (CASSANDRA-8359)
  • Don't set clientMode to true when bulk-loading sstables to avoid a NullPointerException (CASSANDRA-8238)
  • Fix ignored index_interval change in ALTER TABLE statements (CASSANDRA-7976)
  • Do more aggressive compaction in old time windows in DTCS (CASSANDRA-8360)
  • java.lang.AssertionError when reading saved cache (CASSANDRA-8740)
  • "disk full" when running cleanup (CASSANDRA-9036)
  • Make PasswordAuthenticator number of hashing rounds configurable (CASSANDRA-8085)
  • Lower logging level from ERROR to DEBUG when a scheduled schema pull cannot be completed due to a node being down (CASSANDRA-9032)
  • Fix MOVED_NODE client event (CASSANDRA-8516)
  • Allow overriding MAX_OUTSTANDING_REPLAY_COUNT (CASSANDRA-7533)
  • Fix malformed JMX ObjectName containing IPv6 addresses (CASSANDRA-9027)
  • Fix potential data loss in CompressedSequentialWriter (CASSANDRA-8949)
  • (cqlsh) Allow increasing CSV field size limit through cqlshrc config option (CASSANDRA-8934)
  • Stop logging range tombstones when exceeding the threshold (CASSANDRA-8559)
  • Fix NullPointerException when nodetool getendpoints is run against invalid keyspaces or tables (CASSANDRA-8950)
  • Allow specifying the tmp dir (CASSANDRA-7712)
  • Improve compaction estimated tasks estimation (CASSANDRA-8904)
  • Fix duplicate up/down messages sent to native clients (CASSANDRA-7816)

DataStax Enterprise 4.6.3 release notes 

4.6.3 Components
  • Cassandra 2.0.12.275
4.6.3 Resolved issues
  • JMX defaults to binding only localhost to mitigate "CVE-2015-0225: Apache Cassandra remote execution of arbitrary code." After you upgrade to DataStax Enterprise 4.6.3, get the latest default values from cassandra-env.sh and ensure that your local cassandra-env.sh file has the latest default values. If you require remote access of JMX, you can safely enable remote JMX access by turning on JMX security. The latest default values from cassandra-env.sh are:
    LOCAL_JMX=yes
    if [ "$LOCAL_JMX" = "yes" ]; then  
        JVM_OPTS="$JVM_OPTS -Dcassandra.jmx.local.port=$JMX_PORT -XX:+DisableExplicitGC"
    else  
        JVM_OPTS="$JVM_OPTS -Dcom.sun.management.jmxremote.port=$JMX_PORT"  
        JVM_OPTS="$JVM_OPTS -Dcom.sun.management.jmxremote.rmi.port=$JMX_PORT"  
        JVM_OPTS="$JVM_OPTS -Dcom.sun.management.jmxremote.ssl=false"  
        JVM_OPTS="$JVM_OPTS -Dcom.sun.management.jmxremote.authenticate=true"  
        JVM_OPTS="$JVM_OPTS -Dcom.sun.management.jmxremote.password.file=/etc/dse/cassandra/jmxremote.password"
    fi

DataStax Enterprise 4.6.2 release notes 

DataStax Enterprise 4.6.2 release notes cover components, changes and enhancements, issues, and resolved issues.

4.6.2 Components
  • Cassandra 2.0.12.274
4.6.2 Known issues
  • DataStax Enterprise 4.6.2 and earlier releases are affected by "CVE-2015-0225: Apache Cassandra remote execution of arbitrary code." Upgrade to the latest release or enable JMX security immediately if your JMX port is possibly exposed to malicious users.
4.6.2 Resolved issues
  • Added validation error message if Solr schema version is 1.0 (DSP-4421)
  • Join in Spark with case class causes exception (DSP-4645)
  • Lucene library version issue causes degraded query performance compared with Apache Solr 4.6.x (DSP-4811)
  • Mahout demo fails to run when DSE_HOME is set for package installs [DSP-4837]
  • CQL parameterized Solr query causes error (DSP-4847)
  • CompoundFileDirectory exceptions (DSP-4848)
  • Set concurrent_compactors to 2 by default in cassandra.yaml (DSP-4849)
  • Default consistency level is LOCAL_ONE, not QUORUM for Spark (DSP-4860)
  • Set internode compression to "dc" in default cassandra.yaml (DSP-4861)
  • Fix Solr document expiration when row was physically deleted from Cassandra (DSP-4869)
  • NullPointerException when using Solr data import handler (DSP-4888)
  • Examples of using Pig to access Cassandra table with SSL enabled (DSP-4949)
4.6.2 Patches
  • Expose commit log archive status via JMX (CASSANDRA-8734)
  • Provide better exceptions for invalid replication strategy parameters (CASSANDRA-8909)
  • Fix regression in mixed single and multi-column relation support for SELECT statements (CASSANDRA-8613)
  • Add ability to limit number of native connections (CASSANDRA-8086)
  • Fix CQLSSTableWriter throwing exception and spawning threads (CASSANDRA-8808)
  • Add offline tool to relevel sstables (CASSANDRA-8301)
  • Preserve stream ID for more protocol errors (CASSANDRA-8848)
  • Fix combining token() function with multi-column relations on clustering columns (CASSANDRA-8797)
  • Make CFS.markReferenced() resistant to bad refcounting (CASSANDRA-8829)
  • Fix StreamTransferTask abort/complete bad refcounting (CASSANDRA-8815)
  • Fix AssertionError when querying a DESC clustering ordered table with ASC ordering and paging (CASSANDRA-8767)
  • AssertionError: "Memory was freed" when running cleanup (CASSANDRA-8716)
  • Make it possible to set max_sstable_age to fractional days (CASSANDRA-8406)
  • Fix memory leak in SSTableSimple*Writer and SSTableReader.validate() (CASSANDRA-8748)
  • Fix some multi-column relations with indexes on some clustering columns (CASSANDRA-8275)
  • 'nodetool info' prints exception against older node (CASSANDRA-8796)
  • Ensure SSTableSimpleUnsortedWriter.close() terminates if disk writer has crashed (CASSANDRA-8807)

DataStax Enterprise 4.6.1 release notes 

DataStax Enterprise 4.6.1 release notes cover components, changes and enhancements, issues, and resolved issues.

4.6.1 Components
  • Cassandra 2.0.12.200
  • Shark 1.1.1
  • Solr 4.6.0.3.4
  • Sqoop 1.4.5.15.1
4.6.1 Known issues
  • DataStax Enterprise 4.6.1 and earlier releases are affected by "CVE-2015-0225: Apache Cassandra remote execution of arbitrary code." Upgrade to the latest DataStax Enterprise release or enable JMX security immediately if your JMX port is possibly exposed to malicious users.
4.6.1 Resolved issues
  • Unify permissions to the cassandra user (DSP-3585)
  • Include steps and values for non-vnode setups in Single data center deployment and Multi data center deployment (DSP-3705)
  • Implement faceting inside CQL Solr queries, see Using CQL Solr queries (DSP-3935)
  • Support for runtime configuration of Solr performance objects in dsetool (DSP-4000)
  • Document Spark Java API (DSP-4042)
  • Dropping a Solr table leaves data behind in solr_admin (DSP-4068)
  • Passing JAR or script file to the Spark shell is broken (DSP-4116)
  • Sqoop cql-import fails because the DataStax Java Driver is using the wrong data center (DSP-4324)
  • cassandra.connection.host is no longer used to find remote Cassandra for Hive tables (DSP-4366)
  • Speed up dsetool (DSP-4396)
  • Remove "dse on path" dependency for PySpark, see PySpark prerequisites (DSP-4403)
  • Internal authentication support for PySpark (DSP-4436)
  • Can't start analytics node when rest of cluster is stopped (DSP-4464)
  • Authentication failure events are not logged with the audit logger by Kerberos authenticator (DSP-4482)
  • Change strategy for saving version-specific custom Hive meta data (DSP-4501)
  • Thrift interface and Solr warning in DSE 4.6 (DSP-4510)
  • Batch updates that contain static columns prevent indexing of all rows in the batch (DSP-4511)
  • Implement sparse offheap filter (DSP-4518)
  • Kerberos + Pig cannot load table from Cassandra (DSP-4521)
  • Add option to automatically configure doc values in the generated schema (DSP-4524)
  • Stress write can crash nodes (DSP-4533)
  • DSE Pyspark not Python 2.6 compatible (DSP-4570)
  • Automatic Solr schema generation fails when descending clustering order is specified (DSP-4576)
  • DSE Sqoop from Oracle seems to send incorrect SQL syntax (DSP-4579)
  • Upgrade to commons-codec 1.9 (DSP-4590)
  • Indexing is disabled after race condition causing double trigger registration (DSP-4594)
  • Upgrade from DSE 4.0.5 and earlier loses metadata (DSP-4615)
  • Missing events in audit log when using CassandraAuditWriter (DSP-4628)
  • Logs showing vnodes not supported for Solr in DSE 4.6 (DSP-4635)
  • Search fails to start up with "Cannot find core" (DSP-4662)
  • Upgrade Sqoop to 1.4.5 (DSP-4665)
  • Delete by query failing on mixed workload clusters (DSP-4675)
  • Allow custom filter cache implementations (DSP-4686)
  • LDAP authenticator allows empty password to be authenticated (DSP-4687)
  • Race condition while loading secondary indexes upon upgrade (DSP-4689)
  • Running Spark commands against a remote cluster (DSP-4693)
  • Include LUCENE-5401 (DSP-4700)
  • Tracing doesn't work for CQL Solr queries (DSP-4751)
  • Fixed log error/warning "SparkMaster recovery failure" at startup (DSP-4806)
4.6.1 Patches
  • Throw correct exception when trying to bind a keyspace or table name (CASSANDRA-6952)
  • Fix race condition in StreamTransferTask that could lead to infinite loops and premature sstable deletion (CASSANDRA-7704)
  • Ignore Paxos commits for truncated tables (CASSANDRA-7538)
  • JBOD threshold to prevent unbalanced disk utilization (CASSANDRA-7386)
  • Expose off-heap memory usage stats (CASSANDRA-7897)
  • Fix error message on read repair timeouts (CASSANDRA-7947)
  • Fix non-distinct results in DISTNCT queries on static columns when paging is enabled (CASSANDRA-8087)
  • Fix validation of indexes in COMPACT tables (CASSANDRA-8156)
  • don't block SocketThread for MessagingService (CASSANDRA-8188)
  • Introduce background cache refreshing to permissions cache (CASSANDRA-8194)
  • Move all hints related tasks to hints internal executor (CASSANDRA-8285)
  • Increase quarantine delay on replacement (CASSANDRA-8260)
  • Fix some failing queries that use multi-column relations on COMPACT STORAGE tables (CASSANDRA-8264)
  • Regression in ORDER BY (CASSANDRA-8286)
  • Validate size of indexed column values (CASSANDRA-8280)
  • Fix InvalidRequestException with ORDER BY (CASSANDRA-8286)
  • LeveledCompactionStrategy should split large files across data directories when compacting (CASSANDRA-8329)
  • Fix NPE after dropping a keyspace (CASSANDRA-8332)
  • Dropping a column family doesn't remove the latency-sampling task (CASSANDRA-8401)
  • Refuse Paxos operation with more than one pending endpoint (CASSANDRA-8346, 8640)
  • Fix MOVED_NODE topology event never being emitted when a node moves its token (CASSANDRA-8373)
  • Add batch remove iterator to ABSC (CASSANDRA-8414, 8666)
  • cancel latency-sampling task when CF is dropped (CASSANDRA-8401)
  • Fix paging for multi-partition IN queries (CASSANDRA-8408)
  • Avoid StackOverflowError when a large list of IN values is used for a clustering column (CASSANDRA-8410)
  • Make HHOM.compact synchronized (CASSANDRA-8416)
  • Default DTCS base_time_seconds changed to 60 (CASSANDRA-8417)
  • Fix IllegalArgumentException in dynamic snitch (CASSANDRA-8448)
  • Add an extra version check to MigrationTask (CASSANDRA-8462)
  • Fix NPE when writetime() or ttl() calls are wrapped by another function call (CASSANDRA-8451)
  • Move MeteredFlusher to its own thread (CASSANDRA-8485)
  • Fix DISTINCT queries with LIMITs or paging when some partitions contain only tombstones (CASSANDRA-8490)
  • Ensure SSTableWriter cleans up properly after failure (CASSANDRA-8499)
  • Increase bf true positive count on key cache hit (CASSANDRA-8525)
  • Use more efficient slice size for querying internal secondary index tables (CASSANDRA-8550)
  • Fix potentially returning deleted rows with range tombstone (CASSANDRA-8558)
  • Out of memory caused by large tombstone warning (CASSANDRA-8559)
  • Make sure we unmark compacting after scrub/cleanup etc (CASSANDRA-8548)
  • Check for available disk space before starting a compaction (CASSANDRA-8562)
  • Using CQLSSTableWriter gives ConcurrentModificationException (CASSANDRA-8619)
  • Add support for UPDATE ... IF EXISTS (CASSANDRA-8610)
  • Fix SSTableSimpleUnsortedWriter ConcurrentModificationException (CASSANDRA-8619)
  • Paxos requires all nodes for Cassandra (CASSANDRA-8640)
  • Round up time deltas lower than 1ms in BulkLoader (CASSANDRA-8645)
  • Prevent non-zero default_time_to_live on tables with counters (CASSANDRA-8678)
  • Fix isClientMode check in Keyspace (CASSANDRA-8687)
  • Fix reversal of list prepends (CASSANDRA-8733)

DataStax Enterprise 4.6 release notes 

DataStax Enterprise 4.6 release notes cover components, changes and enhancements, resolved issues, and issues in DataStax Enterprise 4.6.

4.6 Components
  • Apache Cassandra 2.0.11.83
  • Apache Hadoop 1.0.4.13
  • Apache Hive 0.12.0.5
  • Apache Mahout 0.8
  • Apache Pig 0.10.1
  • Apache Shark 1.1.0
  • Apache Solr 4.6.0.3.3
  • Apache Spark 1.1.0
  • Spark Cassandra Connector 1.1.0
  • Apache Sqoop 1.4.4.14.2
  • Apache Tomcat 6.0.39
  • Apache Thrift 0.7.0
  • Apache Commons (several subpackages)
  • JBCrypt 0.3m
  • SLF4J 1.7.2
  • Guava 15.0
  • JournalIO 1.4.2
  • Netty 4.0.13.Final
  • Faster XML 3.1.3
  • HdrHistogram 1.2.1.1
  • Snappy 1.0.5
  • DataStax Java Driver for Apache Cassandra 2.1.2
4.6 Known issues
  • DataStax Enterprise 4.6 is affected by "CVE-2015-0225: Apache Cassandra remote execution of arbitrary code." Upgrade to the latest DataStax Enterprise release or enable JMX security immediately if your JMX port is possibly exposed to malicious users.
  • The DataStax Enterprise demos in package installations are in usr/share/dse-demos, which is different than the documented locations.

4.6 Changes and enhancements

DataStax Enterprise 4.6 introduces the following changes and enhancements in addition to the new major features:
  • The endpoint snitch is set in cassandra.yaml, not dse.yaml. The com.datastax.bdp.snitch.DseDelegateSnitch is replaced by com.datastax.bdp.snitch.DseSimpleSnitch in cassandra.yaml and the endpoint_snitch option has been removed from dse.yaml. Follow the upgrading to DataStax Enterprise 4.6 instructions to set the endpoint snitch.
  • Solr
    • DataStax Enterprise 4.6 introduces for experts only the capability for tuning index size relative to query range speed by adjusting the precision step of DataStax internal field types.
    • The DSENRTCachingDirectoryFactory is deprecated. Modify the solrconfig to use the StandardDirectoryFactory instead of the DSENRTCachingDirectoryFactory.
    • The stored=true DSE Search/Solr copy field directive is deprecated and will not be supported in the next release. DataStax recommends upgrading your schema.
    • In the event of an IOException during any index update operations, DSE Search now activates the configured Cassandra disk_failure_policy if enabled. To enable set enable_index_disk_failure_policy to true in the dse.yaml.
    • DSE Search now logs a warning when users do not map dynamic fields correctly.
    • Using the new allowPartialDeletes parameter prevents deletes if a node is down.
  • Spark
    • DataStax Enterprise sets the spark.cassandra.connection.host property automatically and this setting is now applied only if the user does not provide a setting.
  • Other
    • The default Kerberos authenticator now supports the Digest authentication.
4.6 Resolved Issues
  • Fixed the issue causing DataStax Enterprise to treat a CQL-based core as a Thift-backed core if the schema consists of only primary key fields. (DSP-3822)
  • Resolved the CQL TRUNCATE command problem related to memory-only tables. (DSP-4135)
  • Problems using S3 as an external file system have been resolved. (DSP-4082)
  • Resolved the problem running a Hive query against an external Cassandra cluster. (DSP-4094)
  • Fixed a problem causing an out-of-memory condition when running nodetool repair. (DSP-4104, CASSANDRA-7983)
  • Fixed an issue that prevented Hive from working with Kerberos if the user is not initialized on the node. (DSP-4127)
  • Fixed an issue causing Shark to silently drop inserts. (DSP-4148)
  • The DataStax Enterprise integration of Spark respects the SPARK_DRIVER_HOST environment variable on start up. (DSP-4166)
  • Fixed bug causing core creation to fail if the Cassandra timeuuid type is used inside a list, set, or map collection. (DSP-4288)
  • Querying partitions directly now works. (DSP-4293)
  • Fixed the issue preventing Hive from accessing uppercase keyspaces created using the Cassandra CLI. (DSP-4361)

4.6 Known issues

  • A bug exists in the Shark OSS project that causes indexes not to work. DataStax does not know when/if this bug will be fixed in the OSS project. (DSP-4152)
  • If you upgrade to DataStax Enterprise 4.6.0 and use an HSHA rpc server, change rpc_max_threads in the cassandra.yaml file from the default, which is Integer.MAX_VALUE to a reasonable value to prevent failure. (DSP-4205)
  • When you create a table that includes TBLPROPERTIES ("shark.cache" = "true"), the table continues to exist after the session ends. Shark should drop the table when the session ends. Do not design applications to depend on this persistance between sessions in the future. (DSP-4354)
  • The Shark JavaAPI does not work with Spark 1.1. Use the Kryo serializer as a workround.
  • The new Kerberos authenticator does not include login failure events in the log. (DSP-4482)
  • In DataStax Enterprise 4.6, SQL insertion is not supported for Hive operations in Spark due to a Spark 1.1.0 issue (SPARK-3816).

4.6 Patches

  • Disable SSLv3 for POODLE (CASSANDRA-8265)
  • Fix millisecond timestamps in Tracing (CASSANDRA-8297)
  • Include keyspace name in error message when there are insufficient live nodes to stream from (CASSANDRA-8221)
  • Avoid overlap in L1 when L0 contains many nonoverlapping sstables (CASSANDRA-8211)
  • Improve PropertyFileSnitch logging (CASSANDRA-8183)
  • Abort liveRatio calculation if the memtable is flushed (CASSANDRA-8164)
  • Correctly handle non-text column names in cql3 (CASSANDRA-8178)
  • Fix deletion for indexes on primary key columns (CASSANDRA-8206)
  • Add 'nodetool statusgossip' (CASSANDRA-8125)
  • Improve client notification that nodes are ready for requests (CASSANDRA-7510)
  • Handle negative timestamp in writetime method (CASSANDRA-8139)
  • Pig: Remove errant LIMIT clause in CqlNativeStorage (CASSANDRA-8166)
  • Throw ConfigurationException when hsha is used with the default rpc_max_threads setting of 'unlimited' (CASSANDRA-8116)
  • Allow concurrent writing of the same table in the same JVM using CQLSSTableWriter (CASSANDRA-7463)
  • Fix totalDiskSpaceUsed calculation (CASSANDRA-8205)
  • Reject conditions on DELETE unless full PK is given (CASSANDRA-6430)
  • Properly reject the token function DELETE (CASSANDRA-7747)
  • Force batchlog replay before decommissioning a node (CASSANDRA-7446)
  • Fix hint replay with many accumulated expired hints (CASSANDRA-6998)
  • Fix duplicate results in DISTINCT queries on static columns with query paging (CASSANDRA-8108)
  • Add DateTieredCompactionStrategy (CASSANDRA-6602)
  • Properly validate ascii and utf8 string literals in CQL queries (CASSANDRA-8101)
  • (cqlsh) Fix autocompletion for alter keyspace (CASSANDRA-8021)
  • Create backup directories for commitlog archiving during startup (CASSANDRA-8111)
  • Reduce totalBlockFor() for LOCAL_* consistency levels (CASSANDRA-8058)
  • Fix merging schemas with re-dropped keyspaces (CASSANDRA-7256)
  • Fix counters in supercolumns during live upgrades from 1.2 (CASSANDRA-7188)
  • Notify DT subscribers when a column family is truncated (CASSANDRA-8088)
  • Add sanity check of $JAVA on startup (CASSANDRA-7676)
  • Schedule fat client schema pull on join (CASSANDRA-7993)
  • Don't reset nodes' versions when closing IncomingTcpConnections (CASSANDRA-7734)
  • Record the real messaging version in all cases in OutboundTcpConnection (CASSANDRA-8057)
  • SSL does not work in cassandra-cli (CASSANDRA-7899)
  • Fix potential exception when using ReversedType in DynamicCompositeType (CASSANDRA-7898)
  • Better validation of collection values (CASSANDRA-7833)
  • Track min/max timestamps correctly (CASSANDRA-7969)
  • Fix possible overflow while sorting CL segments for replay (CASSANDRA-7992)
  • Increase nodetool Xmx (CASSANDRA-7956)
  • Archive any commitlog segments present at startup (CASSANDRA-6904)
  • Ignore fat clients when checking for endpoint collision (CASSANDRA-7939)
  • CrcCheckChance should adjust based on live CFMetadata not sstable metadata (CASSANDRA-7978)
  • Token() should only accept columns in the partitioning key order (CASSANDRA-6075)
  • Add method to invalidate permission cache via JMX (CASSANDRA-7977)
  • Allow propagating multiple gossip states atomically (CASSANDRA-6125)
  • Log exceptions related to unclean native protocol client disconnects at DEBUG or INFO (CASSANDRA-7849)
  • Allow permissions cache to be set via JMX (CASSANDRA-7698)
  • Include schema_triggers CF in readable system resources (CASSANDRA-7967)
  • Fix RowIndexEntry to report correct serializedSize (CASSANDRA-7948)
  • Make CQLSSTableWriter sync within partitions (CASSANDRA-7360)
  • Potentially use non-local replicas in CqlConfigHelper (CASSANDRA-7906)
  • Explicitly disallowing mixing multi-column and single-column relations on clustering columns (CASSANDRA-7711)
  • Better error message when condition is set on PK column (CASSANDRA-7804)
  • Forbid re-adding dropped counter columns (CASSANDRA-7831)
  • Fix CFMetaData#isThriftCompatible() for PK-only tables (CASSANDRA-7832)
  • Always reject inequality on the partition key without token() (CASSANDRA-7722)
  • Always send Paxos commit to all replicas (CASSANDRA-7479)
  • Make disruptor_thrift_server invocation pool configurable (CASSANDRA-7594)
  • Make repair no-op when RF=1 (CASSANDRA-7864)
  • Fix NPE when table dropped during streaming (CASSANDRA-7946)
  • Fix wrong progress when streaming uncompressed (CASSANDRA-7878)
  • Fix possible infinite loop in creating repair range (CASSANDRA-7983)
  • Fix unit in nodetool for streaming throughput (CASSANDRA-7375)
  • Do not exit nodetool repair when receiving JMX NOTIF_LOST (CASSANDRA-7909)
  • Stream to private IP when available (CASSANDRA-8084)
  • Don't index tombstones (CASSANDRA-7828)
  • Don't send schema change responses and events for no-op DDL statements (CASSANDRA-7600)
  • (Hadoop) fix cluster initialisation for a split fetching (CASSANDRA-7774)
  • Configure system.paxos with LeveledCompactionStrategy (CASSANDRA-7753)
  • Fix ALTER clustering column type from DateType to TimestampType when using DESC clustering order (CASSANRDA-7797)
  • Stop inheriting liveRatio and liveRatioComputedAt from previous memtables (CASSANDRA-7796)
  • Throw EOFException if we run out of chunks in compressed datafile (CASSANDRA-7664)
  • Throw InvalidRequestException when queries contain relations on entire collection columns (CASSANDRA-7506)
  • Fix PRSI handling of CQL3 row markers for row cleanup (CASSANDRA-7787)
  • (cqlsh) enable CTRL-R history search with libedit (CASSANDRA-7577)
  • Fix dropping collection when it's the last regular column (CASSANDRA-7744)
  • Properly reject operations on list index with conditions (CASSANDRA-7499)
  • (Hadoop) allow ACFRW to limit nodes to local DC (CASSANDRA-7252)
  • (cqlsh) Wait up to 10 sec for a tracing session (CASSANDRA-7222)
  • Fix NPE in FileCacheService.sizeInBytes (CASSANDRA-7756)
  • (cqlsh) cqlsh should automatically disable tracing when selecting from system_traces (CASSANDRA-7641)
  • (Hadoop) Add CqlOutputFormat (CASSANDRA-6927)
  • Don't depend on cassandra config for nodetool ring (CASSANDRA-7508)
  • (cqlsh) Fix failing cqlsh formatting tests (CASSANDRA-7703)
  • Fix MS expiring map timeout for Paxos messages (CASSANDRA-7752)
  • Do not flush on truncate if durable_writes is false (CASSANDRA-7750)
  • Give CRR a default input_cql Statement (CASSANDRA-7226)
  • Better error message when adding a collection with the same name than a previously dropped one (CASSANDRA-6276)
  • Fix validation when adding static columns (CASSANDRA-7730)
  • (Thrift) fix range deletion of supercolumns (CASSANDRA-7733)
  • Fix potential AssertionError in RangeTombstoneList (CASSANDRA-7700)
  • Validate arguments of blobAs* functions (CASSANDRA-7707)
  • Fix potential AssertionError with 2ndary indexes (CASSANDRA-6612)
  • Avoid logging CompactionInterrupted at ERROR (CASSANDRA-7694)
  • Minor leak in sstable2jon (CASSANDRA-7709)
  • Add cassandra.auto_bootstrap system property (CASSANDRA-7650)
  • Remove CqlPagingRecordReader/CqlPagingInputFormat (CASSANDRA-7570)
  • Fix IncompatibleClassChangeError from hadoop2 (CASSANDRA-7229)
  • Add 'nodetool sethintedhandoffthrottlekb' (CASSANDRA-7635)
  • Update java driver (for hadoop) (CASSANDRA-7618)
  • Fix truncate to always flush (CASSANDRA-7511)
  • Remove shuffle and taketoken (CASSANDRA-7601)
  • Switch liveRatio-related log messages to DEBUG (CASSANDRA-7467)
  • (cqlsh) Add tab-completion for CREATE/DROP USER IF [NOT] EXISTS (CASSANDRA-7611)
  • Always merge ranges owned by a single node (CASSANDRA-6930)
  • Pig support for hadoop CqlInputFormat (CASSANDRA-6454)
  • Fix ReversedType(DateType) mapping to native protocol (CASSANDRA-7576)
  • (Windows) force range-based repair to non-sequential mode (CASSANDRA-7541)
  • Fix range merging when DES scores are zero (CASSANDRA-7535)
  • Warn when SSL certificates have expired (CASSANDRA-7528)
  • Workaround JVM NPE on JMX bind failure (CASSANDRA-7254)
  • Fix race in FileCacheService RemovalListener (CASSANDRA-7278)
  • Fix inconsistent use of consistencyForCommit that allowed LOCAL_QUORUM operations to incorrect become full QUORUM (CASSANDRA-7345)
  • Properly handle unrecognized opcodes and flags (CASSANDRA-7440)
  • (Hadoop) close CqlRecordWriter clients when finished (CASSANDRA-7459)
  • Make sure high level sstables get compacted (CASSANDRA-7414)
  • Fix AssertionError when using empty clustering columns and static columns(CASSANDRA-7455)
  • Add inter_dc_stream_throughput_outbound_megabits_per_sec (CASSANDRA-6596)
  • Add option to disable STCS in L0 (CASSANDRA-6621)
  • Fix error when doing reversed queries with static columns (CASSANDRA-7490)
  • Backport CASSNADRA-3569/CASSANDRA-6747 (CASSANDRA-7560)
  • Track max/min timestamps for range tombstones (CASSANDRA-7647)
  • Fix NPE when listing saved caches dir (CASSANDRA-7632)
  • Fix sstableloader unable to connect encrypted node (CASSANDRA-7585)
  • Make StreamReceiveTask thread safe and gc friendly (CASSANDRA-7795)
  • Don't allow compacted sstables to be marked as compacting (CASSANDRA-7145)
  • Track expired tombstones (CASSANDRA-7810)
  • Validate empty cell names from counter updates (CASSANDRA-7798)
  • Improve PasswordAuthenticator default super user setup (CASSANDRA-7788)
  • Remove duplicates from StorageService.getJoiningNodes (CASSANDRA-7478)
  • Clone token map outside of hot gossip loops (CASSANDRA-7758)
  • Add stop method to EmbeddedCassandraService (CASSANDRA-7595)
  • Support connecting to ipv6 jmx with nodetool (CASSANDRA-7669)
  • Set gc_grace_seconds to seven days for system schema tables (CASSANDRA-7668)
  • SimpleSeedProvider no longer caches seeds forever (CASSANDRA-7663)
  • Set correct stream ID on responses when non-Exception Throwables are thrown while handling native protocol messages (CASSANDRA-7470)
  • Fix row size miscalculation in LazilyCompactedRow (CASSANDRA-7543)
  • Fix race in background compaction check (CASSANDRA-7745)
  • Don't clear out range tombstones during compaction (CASSANDRA-7808)
  • Fix CC#collectTimeOrderedData() tombstone optimisations (CASSANDRA-7394)