apache/iceberg · error · java.lang.UnsupportedOperationException

Cannot convert unknown type to Flink: %s

Error message

Cannot convert unknown type to Flink: %s

What it means

TypeToFlinkType.primitive maps Iceberg primitive Types to Flink DataTypes; an unrecognized primitive hits the default branch and throws this message with the type's toString. It means an Iceberg type reached the converter that this Iceberg-Flink version does not know how to map (often a newer type than the runtime supports).

Source

Thrown at flink/v2.1/flink/src/main/java/org/apache/iceberg/flink/TypeToFlinkType.java:148

        } else {
          // NANOS
          return new TimestampType(9);
        }
      case STRING:
        return new VarCharType(VarCharType.MAX_LENGTH);
      case UUID:
        // UUID length is 16
        return new BinaryType(16);
      case FIXED:
        Types.FixedType fixedType = (Types.FixedType) primitive;
        return new BinaryType(fixedType.length());
      case BINARY:
        return new VarBinaryType(VarBinaryType.MAX_LENGTH);
      case DECIMAL:
        Types.DecimalType decimal = (Types.DecimalType) primitive;
        return new DecimalType(decimal.precision(), decimal.scale());
      default:
        throw new UnsupportedOperationException(
            "Cannot convert unknown type to Flink: " + primitive);
    }
  }
}

View on GitHub (pinned to 86d9c8fc54)

Solutions

1. Upgrade the iceberg-flink runtime to a version that maps the offending primitive type
2. Migrate the column to a supported type (e.g. timestamp -> micros timestamp) with add/update-schema rewrite
3. Check for mixed Iceberg jar versions on the classpath and align them

Example fix

// before
Table column: timestamp_ns with old iceberg-flink runtime -> throws
// after
Upgrade iceberg-flink runtime, or rewrite column to Types.TimestampType (micros)

for (Types.NestedField f : schema.columns()) {
  if (f.type().isPrimitiveType()) {
    Type.PrimitiveType p = f.type().asPrimitiveType();
    // e.g. reject TimestampNanoType / unknown types before conversion
  }
}

boolean isSupportedIcebergPrimitive(Type t) {
  return switch (t.typeId()) {
    case BOOLEAN, INT, LONG, FLOAT, DOUBLE, DATE, TIME, TIMESTAMP,
         STRING, UUID, FIXED, BINARY, DECIMAL -> true;
    default -> false;
  };
}

try {
  DataType dt = TypeToFlinkType.toFlinkType(primitive);
} catch (UnsupportedOperationException e) {
  throw new IllegalStateException("Upgrade iceberg-flink or migrate column: " + e.getMessage(), e);
}

When it happens

Trigger: Converting an Iceberg schema to Flink types where a column's primitive type is missing from the switch (e.g. Types.TimestampNanoType/unknown Variant on older runtimes) — via FlinkSchemaUtil.convert(schema) or table schema-to-Flink conversion at scan/sink setup.

Common situations: Iceberg tables written by newer engines with nano-timestamp or unknown future types being read by an older iceberg-flink runtime; mixed Iceberg versions on the classpath.

Understand the failure class

Background: "is not a compatible type" / "cannot merge" errors: when a value's type doesn't match what the library requires — this error's family across 65 libraries.

Related errors

• Unsupported YearMonthIntervalType.
• Unsupported DayTimeIntervalType.
• Unsupported DistinctType.
• Unsupported StructuredType.
• Unsupported type: %s

AI-assisted analysis of apache/iceberg@86d9c8fc54 (2026-09-12). Data as JSON: /api/errors/9ecd903bf3060399. Report an issue: GitHub.