Skip to content

Releases: taosdata/TDengine

3.3.4.3

22 Nov 05:44
@tomchon tomchon
ver-3.3.4.3
b917c64
Compare
Choose a tag to compare
3.3.4.3 Latest
Latest

Behavior change and compatibility

New Features

  • feat: the twa function in stream computing support time-driven result push mode.
  • feat: the interp function in stream computing support time-driver result push mode.
  • feat: optimized the error msg when compact + split vgroup operation is performed in sequence
  • feat: support Azure Blob

Optimizations

  • enh: improve the efficiency of pulling data between dnodes when performing large concurrent query.
  • enh: support use AVX2 and AVX512 to perform decoding optimization for double, timestamp and bigint types.
  • enh: adjust the result type of the case when statement.
  • enh: improve the data scanning speed of the query "select... from... where ts in (...)".
  • enh: add a compatiability guarantee mechanism for stream, to avoid new compatiblity issued caused by subsequent functions, but the previously created stream must be deleted and recreated.
  • enh: improve the data synchronization performance of taosX in interlace scenarios.
  • enh: support turn off encoding for integer and float data types.
  • enh: in a multi-replica database, snode must be used in stream computing.
  • enh: generate unique ID to identify each query task to avoid memory corruption.
  • enh: speed up the creation time of the database.
  • enh: modify the default value of s3MigrateEnabled to 0.
  • enh: support recording delete operations in the audit database.
  • enh: support create database in the specified dnode list [enterprise].
  • fix: adjust the error message when deleting data columns of the super table.

Fixed issues

  • fix: query performance of last_row has significantly decreased in version 3.3.3.0.
  • fix: taosd cannot start when the WAL entry is incomplete.
  • fix: incorrect query results when partitioning by a constant.
  • fix: when the scalar function include _wstart and the fill method is prev, the calculation result is wrong.
  • fix: the time zone setting problem on Windows.
  • fix: the translation can't finish if compacting on empty database [enterprise]
  • fix: logic error with transaction conflicts
  • fix: some errors of the mnode can cause transactions to be unable to stop.
  • fix: some errors of the mnode can cause transactions to be unable to stop.
  • fix: the problem of taosc driver retrying errors after clearing the dnode data.
  • fix: after Data Compact is abnormally terminated, intermediate files are not cleaned up.
  • fix: after adding a new column, the earliest mode of the Kafka connector cannot consume the data of the new column.
  • fix: the behavior of the interp function is incorrect when use fill(prev).
  • fix: the problem that TSMA abnormally stops during high-frequency metadata operations.
  • fix: the execution result of the "show create stable" statement is incorrect.
  • fix: the percentile function will crash when querying a large amount of data.
  • fix: syntax error when using partition by and having together.
  • fix: tbname is empty when interp function is used with partition by tbname, c1.
  • fix: taosd may crash while writing illegal boolean values through stmt.
  • fix: update table name description in opc-ua data-in
  • fix: library symbol 'version' confict with other lib which use the same symbol. #25920
  • fix: the number of JAVA process handles in JDBC driver keeps increasing on Windows environment.
  • fix: the occasional startup failure when upgrading from version 3.3.3.1 to 3.3.4.0.
  • fix: memory leak caused by repeated addition and deletion of tables on the Windows platform.
  • fix(stream): check the right return code for concurrent checkpoint trans.
  • fix: the "too many session" problem while perform large concurrent queries.
  • fix: the problem of taos shell crashing in slow query scenarios on the Windows platform.
  • fix: the encrypted database cannot be recovered when opening the dnode log.
  • fix: the problem that taosd cannot be started due to mnode synchronization timeout.
  • fix: the slow sorting of file group data during snapshot synchronization leads to the inability of Vnode to recover.
  • fix: when writing data with escape characters to a varchar field throug line protocol, taosd will crash.
  • fix: metadata file damage caused by incorrect logic processing of error code
  • fix: when a query statement contains multiple nested "not" conditional statements, not setting the scalar mode will lead to query errors.
  • fix: the problem of dnode going offline due to timeout of vnode stat report.
  • fix: taosd failed to start on servers that not support AVX instructions.
  • fix(taosX): handle 0x09xx error codes in migration

What's Changed

Read more

Contributors

yinheli, diorcety, and 44 other contributors
Assets 2
Loading

3.3.3.0

08 Oct 07:31
@gccgdb1234 gccgdb1234
ver-3.3.3.0
e6fe950
Compare
Choose a tag to compare
3.3.3.0

New Features/Improvements:

  1. New functions: pi、truncate/tunc、exp、ln、mod、rand、sign、degress、radians、char、ascii、position、trim、replace、repeat、substring/substr、substring_index、week、weekday、weekofyear、dayofweek、stddev_pop、var_pop
  2. Microsoft Azure Blob can be used with multi-level storage [Enterprise]
  3. Data transfer from MongoDB to TDengine [Enterprise]
  4. TDengine macOS client [Enterprise]
  5. By default, the logs of taosX are not recorded in syslog [Enterprise]
  6. TDengine server records all slow query information to the database storing monitoring data
  7. Display server version in the query result of show cluster machines
  8. Turned keywords LEVEL/ENCODE/COMPRESS into non-keywords, they can be used as column names/table names/database names
  9. Disable the method for dynamic modifying temporary directories

Optimizations:

  1. Performance enhancement of taosX replication [Enterprise]
  2. Unified storage policies for slow query and common logs
  3. Optimize the subscription feature where the offset file is too large and the loading is too slow
  4. Upgrade JDBC driver to 3.2.5
  5. Change commit to sync mode in Kafka connector
  6. Add new parameter: query from table whether an exception is reported when the table does not exist
  7. Expand the maximum window of the tsma interval to one year
  8. Create tables from CSV files

Fixed issues:

  1. MQTT data-in task does not restart automatically after server restart [Enterprise]
  2. Task termination issue caused by agent and PI connector connection errors on Windows [Enterprise]
  3. MQTT tasks fail to synchronize data after restarting taosadapter [Enterprise]
  4. MQTT tasks status are not correct [Enterprise]
  5. Tasks are not scheduled after restarting taosx [Enterprise]
  6. Taosx: Log optimization for MongoDB DataIn task [Enterprise]
  7. Backup failure issue when the local.toml file is empty [Enterprise]
  8. The issue of taosx replica when the network is interrupted [Enterprise]
  9. Fix the problem that meta data is not synchronized between version 3.1 and version 3.3 [Enterprise]
  10. Fix the issue where the MQTT task cannot automatically recover after the taoadapter restarts abnormally [Enterprise]
  11. Kafka Datain tasks stop consuming data [Enterprise]
  12. Crash issue when using last and interval in query
  13. Crash issue caused by updating stt_trigger value during database insertation
  14. Taoskeeper restart repeatedly after modifying the password of root user
  15. The stream computing can not continue working after the taosd instance in cloud service restart
  16. taosd deadlock caused by loading tsdb cache
  17. Unexpected error message when query with fill(0)
  18. Incorrect result of query with last_row
  19. taosd crash issue when the user specifies the information_schema database to log in the TDengine server
  20. The group by statement supports positional parameters and alias syntax
  21. Crash caused by an error of metadata clone function object
  22. taosd crash when filled data during cache updates
  23. Fix the random incorrect result of query with query with stddev
  24. Offline node can not be added into cluster for multi-level storage with database encryption scenario
  25. taos CLI can not enter password longer than 20 characters
  26. The ‘int data overflow’ error when inserting SQL string concatenation
  27. metadata consistency in a large number of concurrent queries
  28. MQTT task can not be stopped normally
  29. Can't export values contain comma or quote as CSV file
  30. Crash issue when multiple threads query to describe the same child table concurrently
  31. Results returned by data explorer do not follow the timezone configuration
  32. Memory leak issue of taosadapter when creating websocket connection
  33. Community version can not start properly on the redhat operating system
  34. Crash caused by invalid assert when two stt files are located in the data block timeline distribution void scenario
  35. S3 fails to pull data for a long time
  36. Use result set function to judge the order of timeline
  37. Performance issue of supertable projection query
  38. In multi-node environment, transaction accumulation is caused by inconsistent transaction return code and state
  39. Error appears after delete data when stt_trigger = 1
  40. taosd service OOM during during node recovery
  41. Modify the judgment error when limit is too small
  42. Issues in historian data source connectivity and retrieving sample data function
  43. taosd service cannot be stopped normally
  44. crash caused by last function query for the whole empty values of some column
  45. coredump caused by assert when the query cid is not in the schema of the current version
  46. OOM issue when writing large message, and add the syncLogBufferCacheAllowed parameter control
  47. Query error when timezone is set to UTC and data contains '1970'
  48. disable timeout when mnode sending sync message
  49. After the cluster node recovers abnormally, the authorization status of the cluster becomes invalid
  50. Query null pointer problem
  51. set vgroup as single replica temporarily when restoring vnode
  52. SQL Reserved Word could NOT be used as column name and tag name in TDengine database
  53. Continuous memory growth during highly concurrent queries
  54. Crash caused by the assertion failure during checkpoint processing occurred in stream computation
  55. Query issue in information_schema database
  56. taosd service OOM during during node recovery
  57. taosd crashes when importing data through taosx when schema changes.
  58. taosd Memory Leak
  59. taos client memory leak
  60. Open source users can not modify other database options after the stt_trigger value is modified
  61. Query result is incorrect with operator of NOT IN (NULL)
  62. taos shell and taosBenchmark can not connect to TDengine Cloud Instance

What's Changed

Read more

Contributors

cadem, zitsen, and 40 other contributors
Assets 2
Loading

3.3.2.0

01 Jul 04:33
@gccgdb1234 gccgdb1234
ver-3.3.2.0
959f95d
Compare
Choose a tag to compare
3.3.2.0

New Features/Improvements

  1. alter table add column can support encode and compress parameters
  2. Configuration parameter supportVnodes can be changed dynamically using alter command
  3. Minimize the impact on read/write by compact operation
  4. Adjust the default value of supportVnodes to 2*Cores+5
  5. lossyColumnn parameter is obsoleted
  6. Support specifying ENCODE and COMPRESS when adding columns to existing table
  7. Update SupportVnodes on the fly
  8. Support CentOS Stream

New Features/Improvements (Enterprise only)

  1. balance vgroup leader can be performed on a database (Enterprise only)
  2. Each mount point in multi-level storage can be configured as allowing new file or not (Enterprise only)
  3. Speed control can be set for cross-level data transfer
  4. IP white list can be adjusted dynamically
  5. New user doesn't have privilege to create database by default
  6. Optimized process of configuring key for database encryption
  7. Data compression for data transfer between TDengine 2.x and 3.x
  8. Data transfer from Oracle to TDengine
  9. Data transfer from SQL Server to TDengine
  10. OPC UA data points can be found and added in data in tasks dynamically
  11. Pi backfill task can be paused and resumed
  12. Pi backfill can work with data transformation
  13. Performance and stability improvement for Pi data in tasks
  14. VARBINARY and GEOMETRY data types can be used when creating tables in taos-explorer
  15. User can transfer user/password and privilege data between TDengine clusters using taos-explorer
  16. Pi data transfer supports newly added assets of an element
  17. taos-explorer can use native connection for the sink TDengine
  18. GSSAPI support for Kafka connector
  19. MQTT tasks can obtain sample data from the data source
  20. Object array support in MQTT data
  21. Data parsing with custom scripts
  22. Dynamic data filtering with plugins

Fixed Bugs

  1. metadata broken by TSMA
  2. count(*) result is wrong when filtering by supertable tags or where conditions
  3. The output precision is wrong of function to_iso8601()
  4. taosd exit abnormally because of querying first() and last() together with interval() and fill()
  5. Deleting a consumer group of one topic caused consumption failure of another consumer group of same name for another topic
  6. Column index of STMT binding over boundary caused taosd exited abnormally
  7. Only one parameter is changed successfully when altering multiple parameters for a table
  8. Union caused taosd hang when the leader is switched frequently
  9. Cast() caused taosd exited abnormally
  10. count window and sliding caused taosd exited abnormally
  11. Performance downgrade of group by tbname
  12. Other fixing for known bugs

What's Changed

Read more

Contributors

cadem, zitsen, and 27 other contributors
Assets 2
Loading

3.3.1.0

28 Jun 07:33
@gccgdb1234 gccgdb1234
ver-3.3.1.0
6fa1d7f
Compare
Choose a tag to compare
3.3.1.0

Features/Improvements

  1. Improve the performance of transferring metadata

Fixed Bugs

  1. create/drop stable is not logged in operation logs
  2. steam-task-status is abnormal after performing redistribute and restart dnode
  3. Some other bug fixes for known issues
Assets 2
Loading

3.3.0.0

10 May 08:26
@gccgdb1234 gccgdb1234
ver-3.3.0.0
eb6094f
Compare
Choose a tag to compare
3.3.0.0

New Features and Improvements

  1. Full functional Join: including LEFT JOIN, RIGHT JOIN, FULL JOIN, SEMI JOIN, ANTI-SEMI JOIN in traditional databases, as well as ASOF JOIN and WINDOW JOIN in time series databases
  2. Composite Primary Key: time-series data is organized and sorted according to both timestamp and an additioanl column, which is called primary key
  3. Double replicas (Enterprise only): high availability and high reliability can be achieved with 2 replicas, i.e. lower hardware cost compared to 3 replicas
  4. S3 storage (Enterprise only): cold data can be stored in S3 to achieve lower storage cost, without compromising query or writing behavior
  5. TSMA: time-range SMA to accelerate aggregate query
  6. Configurable compression: encoding method, compression algorithm and compression level can be configured per column when creating super table
  7. Count Window: A new type of window based on counting rows can be used in both stream and batch processing
  8. Active-Active (Enterprise only): two standalone TDengine deployments can form a single business system to provide high availability with transparency to the client programs. It can also be used in disaster recovery
  9. Database Encryption (Enterprise only): database storage files are encrypted with transparency for reading and writing
  10. Multi-level storage (Enterprise only): Up to 64 mount points can be configured for each storage level
  11. Data In (Enterprise only): time-series data stored in postgreSql and MySQL can be replicated to TDengine in real time
  12. Data In (Enterprise only): optimized configuration method for OPC data points
  13. Data Explorer: a simplified version of data explorer is public for the community

What's Changed

Read more

Contributors

cadem, tjuzyp, and 27 other contributors
Assets 2
Loading

3.2.3.0

29 Feb 13:44
@gccgdb1234 gccgdb1234
ver-3.2.3.0
e27fdcf
Compare
Choose a tag to compare
3.2.3.0

New Features

  1. Count Window for stream processing. The starting and ending condition of a stream processing window is determined by the number of rows in the window
  2. TDengine monitoring
    • Refactored the metrics collected by taosd service, and refined the Grafana dashboard for monitoring taosd services
    • Collected rich metrics for taosX service, and designed new Grafana dashboard for monitoring resource usage, taosX services, agents, and data in tasks of various data source types
  3. New License Mechanism
    • Advanced database features need to be granted, a single license key can grant the fundamental database and/or one or more advanced features
    • Data in features need to be granted, a single license key can grant one or more data source types
  4. Data In: Wonderware Historian data can be ingested into TDengine
  5. Data Compression between taosX-Agent and taosX to save bandwidth in case they are not in same network

Improvements

  1. Performance improvemnt
    • The cache for last is built when creating super tables and child tables to improve the performance of querying last() or last_row() the first time
    • The cache for last can be utilized when querying both last() and last_row() together in a single SQL statement
    • The performance of querying last_row() when there is a column full of NULL values is improved significantly
    • Performance improvement when transferring data from Pi/OPC to TDengine
    • Performance improvement when transferring data from TDengine 2.6 to TDengine 3.x in case there are millions of tables but each table only has few rows
  2. Query optimization
    • Query statement "select count(*)" can filter out empty tables
  3. System Operations
    • The impact on data writing when performing "redistribute vgroup" operation is minimized using RAFT learner mechanism
    • Command "restore vnode" can be performed on a single vgroup ID
    • An empty dnode, i.e. without any vnode, can be dropped by force, regardless of online or not
  4. Data In Optimizations
    • The initial timestamp can be ignored when writing into TDengine from OPC
    • Advanced options are consolidated and refined for some data source types
    • More options, such as skipRows, delimiter, quoteChar, commentPrefix, have been added for CSV data source
  5. Explorer
    • support HTTPS
    • Users can view task related logs

What's Changed

Read more

Contributors

cadem, flyingangel2013, and 28 other contributors
Assets 2
Loading

3.2.2.0

22 Dec 01:19
@gccgdb1234 gccgdb1234
ver-3.2.2.0
e38f1a2
Compare
Choose a tag to compare
3.2.2.0

New Features

  1. Data compacting operations can be queried and killed (Enterprise only)
  2. Simple Storage Service or Object Storage Service (aka S3) can be used at the last tier of multi-tier storage (Enterprise only)
  3. Stream processing supports Event Window
  4. TDenginer server supports Windows/x64 platform (Enterprise only)
  5. ODBC driver

Improvements

  1. Performance improvement for some query scenarios, like partition by and sorting for super tables
  2. Adaptive API for BI tool integration
  3. Performance improvement for snapshot transfer, which accelerates the speed of redistributing vgroup or altering replica
  4. Floating point numbers can be written into columns of integer type

What's Changed

Read more

Contributors

freemine, cadem, and 27 other contributors
Assets 2
Loading

3.2.1.0

17 Nov 06:55
@gccgdb1234 gccgdb1234
ver-3.2.1.0
234463f
Compare
Choose a tag to compare
3.2.1.0

New Fetures:

  • View
  • Optimized query for BI tool integration
  • Optimized data recovery when one disk is disrupted using multi-level storage

Improvements

  • Performance optimization when querying last()/last_row() with primary key or other functions together

What's Changed

Read more

Contributors

cadem, tjuzyp, and 24 other contributors
Assets 2
Loading

3.2.0.0

30 Oct 02:18
@gccgdb1234 gccgdb1234
ver-3.2.0.0
0d8276b
Compare
Choose a tag to compare
3.2.0.0

New Features

  1. IP White List
  2. Configurable floating point lossy compression
  3. balance vgroup leader for a specific vgroup

Improvements

  1. Performance optimization for partition by column
  2. Performance optimization for partition by column with limit/slimit
  3. Performance optimization for partition by column with interval
  4. Block leve lcache for retrieving data on S3
  5. Split vgroup and redistribue vnode with existing of data subscription

What's Changed

Read more

Contributors

sangshuduo, cadem, and 32 other contributors
Assets 2
Loading

3.1.1.4

20 Sep 03:56
@gccgdb1234 gccgdb1234
ver-3.1.1.4
b5fafb3
Compare
Choose a tag to compare
3.1.1.4

Fixed bugs:

  1. core dump when selecting ins_tables
  2. replica data size is much bigger than expected when executing split vgroup
  3. wrongly removing other components in install.sh and rmtaos

What's Changed

Full Changelog: ver-3.1.1.1...ver-3.1.1.4

Contributors

cadem, wangmm0220, and 15 other contributors
Assets 2
Loading