apache/flink · error · IllegalArgumentException

no codec for codecByte: {}

Error message

no codec for codecByte: {}

What it means

AvroOutputFormat's embedded Codec enum maps a byte code to a compression codec; forCodecByte(byte) throws IllegalArgumentException when the byte does not match any known codec (null, deflate, snappy, bzip2, xz, zstd per the enum). It is used when parsing/validating codec settings, so an unknown byte means an unsupported or corrupt codec identifier.

Source

Thrown at flink-formats/flink-avro/src/main/java/org/apache/flink/formats/avro/AvroOutputFormat.java:77

            this.codecByte = codecByte;
            this.codecFactory = codecFactory;
        }

        private byte getCodecByte() {
            return codecByte;
        }

        private CodecFactory getCodecFactory() {
            return codecFactory;
        }

        private static Codec forCodecByte(byte codecByte) {
            for (final Codec codec : Codec.values()) {
                if (codec.getCodecByte() == codecByte) {
                    return codec;
                }
            }
            throw new IllegalArgumentException("no codec for codecByte: " + codecByte);
        }
    }

    private static final long serialVersionUID = 1L;

    private final Class<E> avroValueType;

    private transient Schema userDefinedSchema = null;

    private transient Codec codec = null;

    private transient DataFileWriter<E> dataFileWriter;

    public AvroOutputFormat(Path filePath, Class<E> type) {
        super(filePath);
        this.avroValueType = type;
    }

View on GitHub (pinned to 2f3c205e92)

Solutions

1. Use the public codec setters with well-known names (e.g. setCodecAsString("snappy")) instead of raw bytes.
2. Keep Flink/Avro versions aligned across client and cluster so both sides know the same codec set.
3. Validate the byte against the known codec values before calling forCodecByte, and fail with a descriptive message.
4. If a newer Avro wrote an unknown codec byte, upgrade flink-avro on the reading side to a version that supports it.

Example fix

// before
Codec c = Codec.forCodecByte(rawByte); // throws on unknown

// after
Codec c = Arrays.stream(Codec.values())
        .filter(k -> k.getCodecByte() == rawByte)
        .findFirst()
        .orElseThrow(() -> new IllegalArgumentException(
                "unsupported codec byte " + rawByte + "; upgrade flink-avro"));

boolean known = Arrays.stream(Codec.values())
        .anyMatch(c -> c.getCodecByte() == rawByte);
if (!known) {
    throw new IllegalArgumentException("unsupported codec byte: " + rawByte);
}

When it happens

Trigger: Constructing or deserializing an AvroOutputFormat whose codec byte was set from an unvalidated string/name that maps to no enum constant — e.g. serializing an AvroOutputFormat instance across a cluster where the codec field carried an out-of-range byte, or calling Codec.forCodecByte directly with an arbitrary value.

Common situations: Custom code mutating the codec field via reflection or partial deserialization; format class serialized with a different Flink/Avro plugin version that added new codecs the reader's enum lacks; validation code probing codec bytes without bounds checks.

Related errors

• Encountered exception when reading from avro format file
• Interrupted while waiting for the previous batch to be consu
• Option %s.%s is required for serialization
• Schema provided for '%s' format does not match the table sch
• {e.getMessage()}

AI-assisted analysis of apache/flink@2f3c205e92 (2026-08-14). Data as JSON: /api/errors/a897ce3bdfbeb221. Report an issue: GitHub.