apache/iceberg · error · UnsupportedOperationException

Cannot drop partition field in non-Iceberg table: $table

Error message

Cannot drop partition field in non-Iceberg table: $table

What it means

DropPartitionFieldExec throws this when ALTER TABLE ... DROP PARTITION FIELD targets a non-Iceberg table. Partition evolution (dropping partition fields via UpdateSpec) is an Iceberg-specific capability; other V2 catalogs do not implement it, so the executor throws UnsupportedOperationException when the table is not a SparkTable wrapping an Iceberg table.

Solutions

1. Qualify the table with the Iceberg catalog so the statement resolves to an Iceberg table
2. Confirm the partition field exists on the Iceberg table spec (Spark3Util.toIcebergTerm must translate the transform)
3. For non-Iceberg tables, use that engine's own partition-management commands or migrate to Iceberg

Example fix

// before
ALTER TABLE my_delta_table DROP PARTITION FIELD days(ts);
// after
ALTER TABLE iceberg_catalog.db.my_table DROP PARTITION FIELD days(ts);

val loaded = catalog.loadTable(ident)
require(loaded.isInstanceOf[org.apache.iceberg.spark.source.SparkTable], s"Partition evolution requires Iceberg: $ident")

def isIcebergTable(t: org.apache.spark.sql.connector.catalog.Table): Boolean = t.isInstanceOf[org.apache.iceberg.spark.source.SparkTable]

try { spark.sql(s"ALTER TABLE $ident DROP PARTITION FIELD days(ts)") } catch { case e: UnsupportedOperationException if e.getMessage.contains("non-Iceberg") => log.error(s"Cannot evolve partitions on non-Iceberg table $ident") }

When it happens

Trigger: Running `ALTER TABLE ... DROP PARTITION FIELD days(ts)` where the table resolves through a non-Iceberg V2 catalog, so the `case iceberg: SparkTable` branch is not taken.

Common situations: Iceberg partition-evolution scripts applied to Delta or Hive tables; wrong catalog binding; test environments where the same table name exists under different catalogs.

Understand the failure class

Background: UnsupportedOperationException and "is not supported" errors: when a library deliberately refuses a call — this error's family across 30 libraries.

Related errors

• Cannot replace partition field in non-Iceberg table: $table
• Cannot drop branch on non-Iceberg table: $table
• Cannot drop identifier fields in non-Iceberg table: $table
• Cannot drop tag on non-Iceberg table: $table
• Cannot set identifier fields in non-Iceberg table: $table

AI-assisted analysis of apache/iceberg@86d9c8fc54 (2026-09-12). Data as JSON: /api/errors/36436180c90bbf38. Report an issue: GitHub.

Appendix: source

Thrown at spark/v4.1/spark-extensions/src/main/scala/org/apache/spark/sql/execution/datasources/v2/DropPartitionFieldExec.scala:58

        val schema = iceberg.table.schema
        transform match {
          case IdentityTransform(FieldReference(parts))
              if parts.size == 1 && schema.findField(parts.head) == null =>
            // the name is not present in the Iceberg schema, so it must be a partition field name, not a column name
            iceberg.table
              .updateSpec()
              .removeField(parts.head)
              .commit()

          case _ =>
            iceberg.table
              .updateSpec()
              .removeField(Spark3Util.toIcebergTerm(transform))
              .commit()
        }

      case table =>
        throw new UnsupportedOperationException(
          s"Cannot drop partition field in non-Iceberg table: $table")
    }

    Nil
  }

  override def simpleString(maxFields: Int): String = {
    s"DropPartitionField ${catalog.name}.${ident.quoted} ${transform.describe}"
  }
}

View on GitHub (pinned to 86d9c8fc54)