Issues

• Trino connector: centralize Trino and Hudi type mapping in one bidirectional converter

  area:schemacomponent:trinotype:refactor
  apache/hudi#20210 · Filter by author voonhous opened on Oct 6, 2026

• Compose record keys with : (colon) causes commit failure

  type:bug
  apache/hudi#20209 · Filter by author chandrashekar-kyamanna opened on Oct 6, 2026

• Record index updates fail forever after a first commit that carries no schema

  area:indexarea:metadata-tablepriority:hightype:bug
  apache/hudi#20204 · Filter by author rahil-c opened on Oct 5, 2026
  1

• Complex keygen encoding deduction can read a record another writer keyed

  area:writerpriority:mediumtype:bug
  apache/hudi#20202 · Filter by author ad1happy2go opened on Oct 5, 2026
  1

• Single-field ComplexKeyGenerator tables duplicate records on branch-0.x across the 0.14.1 key encoding change

  area:writerpriority:hightype:bug
  apache/hudi#20199 · Filter by author ad1happy2go opened on Oct 5, 2026
  1

• read_blob() fails on OUT_OF_LINE blobs under Kryo: broadcast storage configuration arrives with a null Hadoop configuration

  area:readerpriority:mediumtype:bug
  apache/hudi#20191 · Filter by author rahil-c opened on Oct 2, 2026

• Triage/Review/land PRs related to 19802

  type:devtask
  apache/hudi#20185 · Filter by author vinothchandar opened on Oct 1, 2026
  

• [BUG] Spark file-group reader picks the vectorized Parquet reader from stale supportBatch state and mutates spark.sql.parquet.enableVectorizedReader, causing ClassCastException ColumnarBatchRow -> UnsafeRo

  type:bug
  apache/hudi#20182 · Filter by author sumitsingh-in opened on Oct 1, 2026

• Make the Hudi scan native in the Comet bundle through hudi-rs

  area:buildarea:query-enginepriority:medium
  Task

  apache/hudi#20176 · Filter by author yihua opened on Oct 1, 2026

• Add an update as delete and insert operation mode for merge-on-read tables

  area:engine-interoparea:writerpriority:high
  apache/hudi#20175 · Filter by author yihua opened on Oct 1, 2026

• A failed lock release leaves a half-released provider, and the timeline archivers leak or misuse their lock

  apache/hudi#20162 · Filter by author linliu-code opened on Sep 30, 2026
  1

• Rename of a VARIANT column: failures to unblock under schema-on-read and schema-on-write

  area:schemaengine:sparkengine:spark4.1type:improvement
  apache/hudi#20141 · Filter by author voonhous opened on Sep 29, 2026
  

• [BUG] HoodieFileGroupReaderBasedFileFormat writes the per-query batch decision into the session conf spark.sql.parquet.enableVectorizedReader

  area:readerengine:sparktype:bug
  apache/hudi#20140 · Filter by author voonhous opened on Sep 29, 2026

• [BUG] count(*) under schema-on-read fails on a VARIANT table once an internal schema is committed

  area:readerarea:schemaengine:sparkengine:spark4.1type:bug
  apache/hudi#20139 · Filter by author voonhous opened on Sep 29, 2026

• Dangling lock with StorageBasedLockProvider when the lock release write fails with an indeterminate storage error

  apache/hudi#20128 · Filter by author linliu-code opened on Sep 28, 2026
  1

• A DataFrame write does not invalidate a CACHE TABLE'd Hudi table, so queries silently return stale rows

  type:bug
  apache/hudi#20113 · Filter by author linliu-code opened on Sep 28, 2026

• Add functional guards against executor .hoodie access and task payload growth on the Spark read path

  area:performancearea:testspriority:mediumtype:improvement
  apache/hudi#20111 · Filter by author yihua opened on Sep 27, 2026

• Spark row-based parquet reads regenerate the same row projection for every file

  area:performancearea:readerpriority:mediumtype:improvement
  apache/hudi#20109 · Filter by author yihua opened on Sep 27, 2026

• Schema-on-read batch reads return wrong values after some column type changes

  area:readerarea:schemapriority:hightype:bug
  apache/hudi#20108 · Filter by author yihua opened on Sep 27, 2026
  1

  

• [Vector Index] Integrate partitioned RLI detection with vector finalist arbitration

  apache/hudi#20106 · Filter by author chrevanthreddy opened on Sep 27, 2026

• CACHE TABLE on a Hudi table strands the cached dataset on every write — fileStatusCache participates in HoodieFileIndex equality

  type:bug
  apache/hudi#20104 · Filter by author linliu-code opened on Sep 27, 2026
  1

• Spark parquet base file reads convert and compare the full footer schema for every file

  area:performancearea:readerpriority:mediumtype:improvement
  apache/hudi#20101 · Filter by author yihua opened on Sep 27, 2026

• Spark parquet base file reads repeat conf copies and schema work that is constant for the scan

  area:performancearea:readerpriority:mediumtype:improvement
  apache/hudi#20100 · Filter by author yihua opened on Sep 27, 2026

• Table service tasks ship the table, and bootstrap listing ignores the job's Hadoop configuration

  area:performancearea:table-servicepriority:mediumtype:improvement
  apache/hudi#20094 · Filter by author yihua opened on Sep 27, 2026

• Metadata index record generation tasks carry table state and repeat metadata table I/O

  area:metadata-tablearea:performancearea:writerpriority:mediumtype:improvement
  apache/hudi#20093 · Filter by author yihua opened on Sep 27, 2026