The Apache IoTDB team is pleased to announce the release of Apache IoTDB

Apache IoTDB (Database for Internet of Things) is an IoT native database
with high performance for data management and analysis, deployable on the
edge and the cloud.

This is a bug-fixed version of 0.12.0, which includes a number of improvements:

## Bug Fixes
* [GITHUB-3373] Remove the broken cached leader connection & optimize
the insertRecords method in session
* [IOTDB-1433] Fix bug in getMetadataAndEndOffset when querying non-exist device
* [IOTDB-1432] fix level compaction loss data
* [IOTDB-1427] Fix compaction lock with query
* [IOTDB-1420] Fix compaction ttl bug
* [IOTDB-1419] Remove redundant clearCompactionStatus, fix continuous
compaction doesn't take effect when enablePartition
* [IOTDB-1415] Fix OOM caused by ChunkCache
* [IOTDB-1414] NPE occurred when call getStorageGroupNodeByPath()
method using not exist path
* [IOTDB-1412] Unclear exception message thrown when executing empty
* [IOTDB-1411] Fix thriftMaxFrameSize and thriftDefaultBufferSize does
not in effect
* [IOTDB-1398] Do not select unseq files when there are uncompacted
old unseq files
* [IOTDB-1390] Fix unseq compaction loss data bug
* [IOTDB-1384] Fix group by bug
* [ISSUE-3378] Fix NPE when clear upgrade folder; Fix some upgraded
pageHeader missing statistics
* [GITHUB-3339] Try to fix sg dead lock
* [GITHUB-3329] Fix upgrade NPE and DeadLock
* [GITHUB-3319] Fix upgrade tool cannot close file reader
* [IOTDB-1212] Fix The given error message is not right when executing
select sin(non_existence) from root.sg1.d1
* [IOTDB-1219] Fix a potential NPE issue in UDF module
* [IOTDB-1286] Fix 4 C++ mem-leak points
* [IOTDB-1294] Fix delete operation become invalid after compaction
* [IOTDB-1313] Fix lossing time precision when import csv with
unsupported timestamp format
* [IOTDB-1316] The importCsv tool should continue inserting if a part
of insertion failed
* [IOTDB-1317] Fix log CatchUp always failed due to not check the
follower's match index
* [IOTDB-1323] Fix return a success message when encounter
RuntimeException during the insertion process
* [IOTDB-1325] Fix StackOverflow Exception in group by natural month query
* [IOTDB-1330] Fix the load tsfile bug when the cross multi
partition's tsfile only have one page
* [IOTDB-1348] Fix Last plan not work in cluster mode
* [IOTDB-1376] Fix BatchProcessException was not correctly handled in
* [ISSUE-3277] Fix TotalSeriesNumber in MManager counted twice when recovering
* [ISSUE-3116] Fix bug when using natural month unit in time interval
in group by query
* [ISSUE-3309] Fix InsertRecordsOfOneDevice runs too slow
* Fix the plan index is always zero when using insertRecords interface
to run the cluster
* Add authority check for users create timeseries using executeBatch
interface without the privilege
* Fix versionInfo NPE when query upgrading 0.11 tsfile
* Fix upgrade tool cannot load old tsfile if time partition enabled in 0.11
* Fix import csv throw ArrayOutOfIndexError when the last value in a
line is null
* Fix upgrade tool cannot close file reader

## Improvements
* [GITHUB-3399] Change the default primitive array size to 32
* [IOTDB-1387] Support Without Null ALL in align by device clause,
Filter RowRecord automatically if any column in it is null or all
columns are null
* [IOTDB-1385] Extract the super user to the configuration
* [IOTDB-1315] ExportCsvTool should support timestamp
* [IOTDB-1339] optimize TimeoutChangeableTSnappyFramedTransport
* [IOTDB-1356] Separate unseq_file_num_in_each_level from selecting
candidate file in unseq compaction
* [IOTDB-1357] Compaction use append chunk merge strategy when chunk
is already large enough
* [IOTDB-1380] Automatically close the dataset while there is no more data
* Optimize sync leader for meta

## New Features
* [GITHUB-3389] TTL can be set to any path
* [GITHUB-3387] Add parameter compaction_interval=10000ms
* [IOTDB-1190] Fully support HTTP URL char set in timeseries path
* [IOTDB-1321][IOTDB-1322] Filter RowRecord automatically if any
column in it is null or all columns are null
* [IOTDB-1357] Compaction use append chunk merge strategy when chunk
is already large
* [ISSUE-3089] Make it possible for storage groups to have name with hyphen

## Miscellaneous changes
* [GITHUB-3346] upgrade netty and claim exclusion for enforcer check
* [IOTDB-1259] upgrade libthrift from 0.12.0/0.13.0 to 0.14.1
* Uncomment the less used configurations
* Enable the configration `concurrent_writing_time_partition`

The full release note is available at:

The release is available for download at:

Maven artifacts for JDBC driver, session SDK, TsFile SDK, Spark-connector,
Hadoop-connector, Hive-connector and Flink-connector can be found at:

Docker image of IoTDB server can be found at:

Python API package can be found at:

The Apache IoTDB team

Reply via email to