apache/iceberg · error · IllegalArgumentException

Unsupported compression codec

Error message

Unsupported compression codec: %s

What it means

Parquet.Context.toCodec converts a user-supplied compression codec string to a Parquet CompressionCodecName; unknown names raise this IllegalArgumentException. It validates the write compression setting before any file is written.

Solutions

1. Use one of the valid codec names: NONE, UNCOMPRESSED, SNAPPY, GZIP, LZO, BROTLI, LZ4, ZSTD
2. Check for whitespace or casing issues in the configured value
3. Upgrade parquet-mr/Iceberg if the codec (e.g. ZSTD) isn't in the enum version you use

Example fix

// before
table.updateProperties().set("write.parquet.compression-codec", "gzip2").commit();
// after
table.updateProperties().set("write.parquet.compression-codec", "GZIP").commit();

String codec = props.getProperty("write.parquet.compression-codec");
if (codec != null) {
  Set<String> valid = Set.of("NONE","UNCOMPRESSED","SNAPPY","GZIP","LZO","BROTLI","LZ4","ZSTD");
  Preconditions.checkArgument(valid.contains(codec.toUpperCase(Locale.ROOT)), "Unsupported compression codec: %s", codec);
}

try {
  newContext();
} catch (IllegalArgumentException e) {
  if (e.getMessage().startsWith("Unsupported compression codec")) { /* correct the property */ }
  else throw e;
}

When it happens

Trigger: Passing TableProperties.WRITE_COMPRESSION_CODEC / Parquet.PARQUET_COMPRESSION with a value not matching a Parquet codec enum name (valueOf is case-insensitive via toUpperCase but the name must otherwise match exactly).

Common situations: Typos like 'snappy ' or 'lz4hadoop' or 'brotli' on unsupported versions; setting the property with engine-specific codec names not present in parquet-mr's enum.

Understand the failure class

Background: Invalid enum value errors: "Unknown type", "Invalid scope", "must be one of" — when a string is not on the library's allowed list — this error's family across 23 libraries.

Related errors

• Class was not found
• Fallback ID assignment in Parquet is UNSAFE and will be…
• Skipping bloom filter config for missing field
• Skipping column statistics config for missing field
• Skipping dictionary encoding config for missing field

AI-assisted analysis of apache/iceberg@86d9c8fc54 (2026-09-12). Data as JSON: /api/errors/7886e9dfa3da4ee3. Report an issue: GitHub.

Appendix: source

Thrown at parquet/src/main/java/org/apache/iceberg/parquet/Parquet.java:794

            compressionLevel,
            rowGroupCheckMinRecordCount,
            rowGroupCheckMaxRecordCount,
            PARQUET_BLOOM_FILTER_MAX_BYTES_DEFAULT,
            PARQUET_BLOOM_FILTER_ADAPTIVE_ENABLED_DEFAULT,
            ImmutableMap.of(),
            ImmutableMap.of(),
            ImmutableMap.of(),
            ImmutableMap.of(),
            dictionaryEnabled,
            ImmutableMap.of(),
            dataContext.trackUncompressedRowGroupSize());
      }

      private static CompressionCodecName toCodec(String codecAsString) {
        try {
          return CompressionCodecName.valueOf(codecAsString.toUpperCase(Locale.ROOT));
        } catch (IllegalArgumentException e) {
          throw new IllegalArgumentException("Unsupported compression codec: " + codecAsString);
        }
      }

      private static WriterVersion toWriterVersion(String pageVersion) {
        try {
          return WriterVersion.fromString(pageVersion);
        } catch (IllegalArgumentException e) {
          throw new IllegalArgumentException(
              "Unsupported Parquet page version: " + pageVersion + " (must be v1 or v2)");
        }
      }

      int rowGroupSize() {
        return rowGroupSize;
      }

      int pageSize() {
        return pageSize;

View on GitHub (pinned to 86d9c8fc54)