This is an automated email from the ASF dual-hosted git repository. HTHou pushed a commit to branch rc/2.0.12 in repository https://gitbox.apache.org/repos/asf/iotdb.git
commit ae2a8597489fb0d2fdecedfa7fe6f029aa01ff95 Author: HTHou <[email protected]> AuthorDate: Tue Nov 17 10:51:38 2026 +0800 Update release notes and versions --- RELEASE_NOTES.md | 99 ++++++++++++++++++++++++++ distribution/pom.xml | 2 +- iotdb-client/client-cpp/pom.xml | 4 +- iotdb-client/jdbc/src/main/feature/feature.xml | 2 +- iotdb-core/ainode/pom.xml | 12 ++-- iotdb-core/ainode/pyproject.toml | 2 +- 6 files changed, 110 insertions(+), 11 deletions(-) diff --git a/RELEASE_NOTES.md b/RELEASE_NOTES.md index 96b941b107d..6608f03db57 100644 --- a/RELEASE_NOTES.md +++ b/RELEASE_NOTES.md @@ -19,6 +19,105 @@ --> +# Apache IoTDB 2.0.12 + +## Features & Improvements + +**SQL Injection Prevention:** Refactored the table model's PreparedStatement to prevent SQL injection +**Cluster Management:** Support IPv6 cluster deployment +**REST:** REST API supports setting the client timezone +**Data Sync:** Optimized filtering performance for non-first TAG columns in the table model +**Data Import/Export Tools:** Degradation strategy for Load files +**Database Management:** `SHOW CREATE DATABASE/TOPIC/PIPE` +**Computation - Query Execution - SQL Clauses:** Optimize the error message when a `select *` query has too many columns +**Functions:** Added time-frequency analysis function (FFT) +**Functions:** Added filtering functions (HighPass, LowPass) +**Latest Point Cache:** The latest point cache supports being enabled/disabled at table/database granularity +**Functions:** Built-in rate/increase/irate/delta aggregate functions for the table model +**Functions:** Added sequence correlation function (XCorr) +**Functions:** Built-in M4 table-valued function +**Functions:** Time-series trend matching table-valued function +**Functions:** Added the built-in table-valued function `read_tsfile`, supporting direct querying from a specified tsfile using SQL +**Tree Model Metadata:** When creating a new time series under an existing device, ignore the aligned/unaligned attributes specified in the create statement and use the current device's aligned/unaligned attributes instead of reporting an error +**Measurement:** Optimized the validation result of Measurement during cache writes +**Device-Granularity Deletion:** Fixed an issue where, after deleting a table with a large amount of data and then re-creating the table and writing data, querying that table had high latency +**Memory Optimization:** Memory optimization when copying the memtable TVList during queries +**Pipe Data Sync:** Optimized the matching algorithm for `**` +**Tree Model Metadata:** Added a boolean flag to device nodes indicating whether there are other devices below, and tuned the query `select measurement from xxx.**` +**SQL Clauses:** Performance optimization for full-table `order by time + limit` in the table model +**SQL Clauses:** Memory optimization for full-table queries in the table model +**Pipe Data Sync:** Optimized the Pipe memory block expansion/contraction algorithm + +## Bugs + +**Online File Merge:** Fixed an issue where `show disk_usage` caused the merge to hang +**Tree Model Metadata:** Fixed the "no matched database" issue +**Data Sync:** Support storing TsFile parsing checkpoint progress, so that on retry after failure there is no need to re-parse the entire file +**Basic Write:** Fixed an issue where, for the same aligned device, first writing a request with fewer columns and more rows and then writing a request with more columns and one row generated a WAL that, upon restarting the DataNode, caused severe memtable null-value expansion and ultimately caused the DataNode restart to hang +**TsFile Sync:** When loading a tsfile, modify the replica set comparison logic: when comparing replica sets, no longer care about list order, only compare whether the elements within the sets are exactly identical; change the list comparison to a set comparison +**Pipe Data Sync:** Fixed an issue where Pipe threw an NPE for null measurement points when converting a partial insert write into a tablet +**Pipe Data Sync:** Fixed an issue where Pipe stopped after a tsfile parsing timeout +**Online File Merge:** Fixed an issue where the merge did not fall back to copying when creating hard links failed +**Pipe Data Sync:** Added per-Pipe independent memory quota limits or fair scheduling +**Pipe Data Sync:** Implemented exponential backoff and circuit breaker protection for scenarios where the disk is full or the receiver is unavailable +**Pipe Data Sync:** Added global and per-Pipe limits on in-flight TsFiles +**Pipe Data Sync:** Fixed an issue where `drop pipe` did not wait for the sink's return +**Data Import/Export Tools:** Fixed an error when executing `tsfile load` while a database named with a `root.` prefix already exists +**Table Model Metadata:** Optimized the handling of special characters escaped with backticks in table views +**Pipe Data Sync:** When deleting a Pipe there are asynchronous processes (e.g., recover); the asynchronous execution result needs to be communicated to the user +**Pipe Data Sync:** `Drop pipe` no longer waits for the sink's return, ensuring complete removal of the Pipe +**Pipe Data Sync:** Fixed the lag in the assigner portion when executing Pipe create/start/stop/delete commands +**Pipe Data Sync:** Fixed the excessive logging issue by reducing the amount of logs on the Pipe receiver side +**Pipe Data Sync:** Fixed an issue where splitting a tsfile in 1.3.x to sync partial data to 2.x failed + +# Apache IoTDB 2.0.11 + +## Break change +- This version raises the minimum required JDK version to 17. + +## Features & Improvements + +- Metadata: SHOW TIMESERIES supports sorting by measurement name. +- Query: Support SQL statements that contain only a SELECT clause. +- Region: Support manually repairing partitioned tables using SQL. +- Views: Support logical views. +- Metadata: Add the COUNT DATABASE statement for the table model. +- IoTConsensus: Support multiple directories when IoTConsensus receives snapshots. +- Region: The EXTEND REGION and REMOVE REGION syntaxes support a regionId list, allowing multiple regions to be processed at once. +- Query: EXPLAIN ANALYZE supports JSON output format. +- Mods: Strengthen memory control for queries involving mods. Memory usage must also be controllable for a single large, GB-level mods2 file. +- Data Ingestion: When deleting data, send requests only to the region group to which the current table belongs. +- Protocol Layer: Add memory control at the protocol layer. +- UDF: Add an InternalSession interface to the UDF API. +- Pipe: Modify the real-time data definition logic. +- Functions: Support subsequent-value filling. +- SQL Clauses: GROUP BY and ORDER BY support aliases defined with AS in the SELECT clause. +- Parameter Configuration: Support hot reloading of cluster runtime configurations. +- Communication Encryption: Support mutual authentication for encrypted client communications. +- Query: Support writing query results to a TsFile at a specified path using SQL. +- High Availability: Support high availability for table management, device management, writable-view management, TTL management, database management, and user privileges. +- Client: Support a Node.js client. Only version 2.x is supported. +- Edge Edition: To facilitate use on edge devices with limited memory, introduce an Edge Edition that limits the combined memory usage of the ConfigNode and DataNode to 512 MB. The target workload is approximately 10,000 measurements operating at 1 Hz for read and write workloads. The first version will primarily adjust certain JVM and default configuration parameters; future versions may continue evolving toward a more integrated DataNode–ConfigNode architecture. + +## Bugs +- Query: In the tree and table models, comparisons between integer and floating-point values in query filter conditions behave inconsistently. +- Scaling: During scale-in validation, IoTConsensusQueue under IoTConsensus Used Memory on the monitoring dashboard displays a negative value. +- Load TsFile: When the receiving end has write privileges, LOAD TSFILE fails with No permissions for this operation, please add privilege SYSTEM, and the data is not synchronized. +- SSL: Fix NullPointerException: Cannot invoke "String.contains(java.lang.CharSequence)" because the return value of "java.lang.Throwable.getMessage()" is null. +- SSL: Improve the error message displayed when SSL is enabled in the security package but the CLI connects to the cluster without specifying SSL, making the message more explicit. +- Cluster Management: During validation of Issue 0228, a new problem was encountered: after restarting the cluster, one node remained in the Unknown state and produced no enjoy logs, indicating that the main thread was blocked. However, the node could still perform compaction normally. +- Scaling: During scale-in, the CompressionRatio field for a DataRegion in the Adding state displays a negative value in SHOW REGIONS. +- Region: Preserve two decimal places for the CompressionRatio field in SHOW REGIONS results. +- Scaling: During scale-in, after restarting the cluster, the CompressionRatio field for most DataRegions in SHOW REGIONS is NaN, even though it had a value before the restart. +- ConfigNode: Fix a ConfigNode thread leak. In an IoTV2 3C3D deployment with three replicas, running SQL covering tree-model test cases caused the number of ConfigNodeRPC-Processor threads to reach 3,000. The cluster could no longer be connected to, and the ConfigNode/DataNode reported NullPointerExceptions. +- Query: Fix an error in CAST(1.1 AS FLOAT): java.lang.UnsupportedOperationException: org.apache.tsfile.read.common.block.column.DoubleColumn. The operation is expected to succeed. +- DataNode: PipeTsFileEpochProgressIndexKeeper retains more than 50,000 TsFileResources, causing the DataNode to run out of memory. In an IoTV2 3C3D deployment with two replicas, concurrent reads, writes, and deletes were performed while one DataNode was stopped; after 54 hours of testing, one DataNode ran out of memory and generated a 23-GB OOM dump with ON_HEAP_MEMORY="20G". +- User Management: Concurrent execution of LIST USER, CREATE USER, and DROP USER may result in 301: Ratis request failed Unknown. The ConfigNode log shows Caused by: java.util.ConcurrentModificationException: null. +- IoTConsensus: In a 3C3D deployment with two replicas, concurrent reads, writes, and deletes were performed while a DataNode was stopped. One DataNode process exited and generated a 25-GB dump file. +- Metadata: Prohibit creating devices on template measurements and their subpaths. +- Region: A frame size error, Frame size (69670459) larger than protect max size (67108864), may cause a query to hang. The system should return an appropriate error message instead. + + # Apache IoTDB 2.0.10 ## Features & Improvements diff --git a/distribution/pom.xml b/distribution/pom.xml index 4a8b1c33c94..824af36cf93 100644 --- a/distribution/pom.xml +++ b/distribution/pom.xml @@ -253,7 +253,7 @@ <dependency> <groupId>org.apache.iotdb</groupId> <artifactId>iotdb-ainode</artifactId> - <version>2.0.11-SNAPSHOT</version> + <version>2.0.12</version> </dependency> </dependencies> <build> diff --git a/iotdb-client/client-cpp/pom.xml b/iotdb-client/client-cpp/pom.xml index 656f0349974..209d7cda43a 100644 --- a/iotdb-client/client-cpp/pom.xml +++ b/iotdb-client/client-cpp/pom.xml @@ -24,7 +24,7 @@ <parent> <groupId>org.apache.iotdb</groupId> <artifactId>iotdb-client</artifactId> - <version>2.0.11-SNAPSHOT</version> + <version>2.0.12</version> </parent> <artifactId>client-cpp</artifactId> <packaging>pom</packaging> @@ -69,7 +69,7 @@ <dependency> <groupId>org.apache.iotdb</groupId> <artifactId>iotdb-thrift-commons</artifactId> - <version>2.0.11-SNAPSHOT</version> + <version>2.0.12</version> <scope>provided</scope> </dependency> </dependencies> diff --git a/iotdb-client/jdbc/src/main/feature/feature.xml b/iotdb-client/jdbc/src/main/feature/feature.xml index 06d0e1c8375..02e1e93b64e 100644 --- a/iotdb-client/jdbc/src/main/feature/feature.xml +++ b/iotdb-client/jdbc/src/main/feature/feature.xml @@ -18,7 +18,7 @@ --> <features xmlns="http://karaf.apache.org/xmlns/features/v1.5.0" name="driver-s7-feature"> - <feature name="iotdb-feature" description="iotdb-feature" version="1.3.3"> + <feature name="iotdb-feature" description="iotdb-feature" version="2.0.12"> <details>Feature to install required Bundle to use IoTDB inside Karaf container</details> <feature prerequisite="true">wrap</feature> <feature>scr</feature> diff --git a/iotdb-core/ainode/pom.xml b/iotdb-core/ainode/pom.xml index 80cab2a052b..1fce5a0b585 100644 --- a/iotdb-core/ainode/pom.xml +++ b/iotdb-core/ainode/pom.xml @@ -24,7 +24,7 @@ <parent> <groupId>org.apache.iotdb</groupId> <artifactId>iotdb-core</artifactId> - <version>2.0.11-SNAPSHOT</version> + <version>2.0.12</version> </parent> <artifactId>iotdb-ainode</artifactId> <name>IoTDB: Core: AINode</name> @@ -33,31 +33,31 @@ <dependency> <groupId>org.apache.iotdb</groupId> <artifactId>iotdb-thrift</artifactId> - <version>2.0.11-SNAPSHOT</version> + <version>2.0.12</version> <scope>provided</scope> </dependency> <dependency> <groupId>org.apache.iotdb</groupId> <artifactId>iotdb-thrift-confignode</artifactId> - <version>2.0.11-SNAPSHOT</version> + <version>2.0.12</version> <scope>provided</scope> </dependency> <dependency> <groupId>org.apache.iotdb</groupId> <artifactId>iotdb-thrift-commons</artifactId> - <version>2.0.11-SNAPSHOT</version> + <version>2.0.12</version> <scope>provided</scope> </dependency> <dependency> <groupId>org.apache.iotdb</groupId> <artifactId>iotdb-thrift-ainode</artifactId> - <version>2.0.11-SNAPSHOT</version> + <version>2.0.12</version> <scope>provided</scope> </dependency> <dependency> <groupId>org.apache.iotdb</groupId> <artifactId>iotdb-python-api</artifactId> - <version>2.0.11-SNAPSHOT</version> + <version>2.0.12</version> <scope>provided</scope> </dependency> </dependencies> diff --git a/iotdb-core/ainode/pyproject.toml b/iotdb-core/ainode/pyproject.toml index 0dc630fb0ff..198a124a551 100644 --- a/iotdb-core/ainode/pyproject.toml +++ b/iotdb-core/ainode/pyproject.toml @@ -21,7 +21,7 @@ build-backend = "poetry.core.masonry.api" [tool.poetry] name = "apache-iotdb-ainode" -version = "2.0.7.dev" +version = "2.0.12" description = "Apache IoTDB AINode" readme = "README.md" authors = ["Apache Software Foundation <[email protected]>"]
