Issues

• Flink: Upserts and deletes after a partition spec change leave duplicate and deleted rows in the table

  bug
  apache/iceberg#18748 · Filter by author jiwen624 opened on Nov 2, 2026
  1

• AWS: S3FileIO with the default client factory needs the kms and sts SDK modules at runtime; aws.md lists Glue/S3/DynamoDB/KMS/STS

  apache/iceberg#18744 · Filter by author msabiransari opened on Nov 1, 2026

• JDBC catalog: a dropped connection after the commit UPDATE was applied is replayed by JdbcClientPool, reported as CommitFailedException, and cleanup deletes the committed snapshot's files

  apache/iceberg#18743 · Filter by author msabiransari opened on Nov 1, 2026

• Core: expireSnapshots() inside a Transaction deletes files before commitTransaction() since 1.10.0

  apache/iceberg#18737 · Filter by author namest504 opened on Nov 1, 2026

• API: Predicate evaluation and file pruning treat -0.0 and 0.0 as different values

  apache/iceberg#18723 · Filter by author Rishavbhattarai opened on Oct 31, 2026

• Spark: WAP snapshot isolation broken for format-version 3 tables when deletion vectors are used (spark.wap.id)

  bug
  apache/iceberg#18713 · Filter by author abjain123 opened on Oct 30, 2026
  3

• Spark: Expose the effective split target size as a driver metric

  apache/iceberg#18708 · Filter by author gerashegalov opened on Oct 30, 2026

• Core: Support spherical geography bounding box metrics

  apache/iceberg#18705 · Filter by author KaiqiJinWow opened on Oct 29, 2026
  1

• API: Evaluator matches null values for notEq and notIn

  improvement
  apache/iceberg#18701 · Filter by author Revanth14 opened on Oct 29, 2026

• API: Binding folds float comparisons with out-of-range literals incorrectly for NaN, infinity and null

  improvement
  apache/iceberg#18700 · Filter by author Revanth14 opened on Oct 29, 2026

• Parquet files written with the "default" CRS are not compatible with Iceberg Java EPSG:4326 table

  bug
  apache/iceberg#18680 · Filter by author paleolimbot opened on Oct 28, 2026
  3

• Core: Expiring unused schemas and adding a column in one transaction leaves REST table metadata unloadable

  apache/iceberg#18677 · Filter by author LuciferYang opened on Oct 28, 2026

• Core: Metadata DELETE can incorrectly remove NULL date partitions

  bug
  apache/iceberg#18672 · Filter by author rajendrag opened on Oct 28, 2026

• Tracking: Vectorized Spark reads of shredded VARIANT columns

  improvement
  apache/iceberg#18669 · Filter by author nssalian opened on Oct 27, 2026
  2

  

• Core: replaceTransaction probes viewExists causing 403 Forbidden under table-only permissions

  bug
  apache/iceberg#18662 · Filter by author lsherwin10 opened on Oct 26, 2026

• Core: TableMetadata.Builder doesn't enforce format version field requirements

  bug
  apache/iceberg#18646 · Filter by author kevinjqliu opened on Oct 25, 2026

• Core: TableMetadata.Builder allows adding encryption keys to v1/v2 tables

  bug
  apache/iceberg#18632 · Filter by author kevinjqliu opened on Oct 24, 2026
  1

• FileMetadata.Builder.copy(DeleteFile) drops equalityFieldIds

  apache/iceberg#18630 · Filter by author mikepapadim opened on Oct 24, 2026

• Flink: DynamicIcebergSink writer and write-result aggregator keep using a dropped table's location after the table is re-created with the same name

  apache/iceberg#18623 · Filter by author kuni-92 opened on Oct 23, 2026

• Spark: equality delete on a nested struct field deletes every row

  apache/iceberg#18620 · Filter by author Neuw84 opened on Oct 23, 2026
  2

• Equality delete with a composite key containing a uuid is never attached to FileScanTask.deletes()

  apache/iceberg#18619 · Filter by author Neuw84 opened on Oct 23, 2026
  1

• Equality delete on a fixed column deletes almost nothing: StructLikeSet collapses to one key

  apache/iceberg#18618 · Filter by author Neuw84 opened on Oct 23, 2026
  2

• Clarify that write.*.isolation-level properties do not apply to append/INSERT

  apache/iceberg#18608 · Filter by author SergeyTimofeev1989 opened on Oct 22, 2026
  3

• Spark: metadata-only DELETE can remove NULL rows for <>, single-value NOT IN and NOT LIKE

  apache/iceberg#18605 · Filter by author zhengxiang378928908-code opened on Oct 22, 2026

• Spark: vectorized read fails on projected shredded VARIANT without variant bounds

  bug
  apache/iceberg#18601 · Filter by author kevinjqliu opened on Oct 21, 2026
  1