spring-projects/spring-ai · error · RuntimeException

Error serializing documentMap to JSON.

Error message

Error serializing documentMap to JSON.

What it means

SimpleVectorStore.getVectorDbAsJson() wraps Jackson serialization failures of the in-memory document map into a RuntimeException 'Error serializing documentMap to JSON.' The internal store maps IDs to documents/embeddings and Jackson fails if those objects are not serializable (e.g. a custom Document or embedding type Jackson cannot handle).

Source

Thrown at spring-ai-vector-store/src/main/java/org/springframework/ai/vectorstore/SimpleVectorStore.java:247

	public void load(Resource resource) {
		TypeReference<HashMap<String, SimpleVectorStoreContent>> typeRef = new TypeReference<>() {

		};
		try {
			this.store = this.jsonMapper.readValue(resource.getInputStream(), typeRef);
		}
		catch (IOException ex) {
			throw new RuntimeException(ex);
		}
	}

	private String getVectorDbAsJson() {
		ObjectWriter objectWriter = this.jsonMapper.writerWithDefaultPrettyPrinter();
		try {
			return objectWriter.writeValueAsString(this.store);
		}
		catch (JacksonException ex) {
			throw new RuntimeException("Error serializing documentMap to JSON.", ex);
		}
	}

	private float[] getUserQueryEmbedding(String query) {
		return this.embeddingModel.embed(query);
	}

	@Override
	public VectorStoreObservationContext.Builder createObservationContextBuilder(String operationName) {

		return VectorStoreObservationContext.builder(VectorStoreProvider.SIMPLE.value(), operationName)
			.dimensions(this.embeddingModel.dimensions())
			.collectionName("in-memory-map")
			.similarityMetric(VectorStoreSimilarityMetric.COSINE.value());
	}

	public static final class EmbeddingMath {

View on GitHub (pinned to 98a7beda4f)

Solutions

1. Inspect the wrapped JacksonException cause to find the unserializable type.
2. Ensure stored Documents use serializable field types or add proper Jackson annotations/mixins.
3. Verify Jackson versions on the classpath are consistent with the Spring AI dependency requirements.

Example fix

// before
vectorStore.persist(myFile); // fails if a document has an unserializable field

// after
// remove/replace non-serializable fields, e.g.
public class MyDocument extends Document {
    private transient OutputStream debugStream; // excluded from serialization
}

// ensure all stored documents contain only Jackson-serializable fields
objects.forEach(d -> assertSerializable(d));

try {
    vectorStore.persist(file);
} catch (RuntimeException e) {
    if (e.getMessage() != null && e.getMessage().contains("serializing documentMap")) {
        logger.error("Unserializable document in store", e.getCause());
    }
    throw e;
}

When it happens

Trigger: Calling persist() (which calls json()) after inserting documents whose stored representation cannot be serialized by the ObjectMapper, or when the Jackson mapper encounters an incompatible type.

Common situations: Custom Document subclasses with non-serializable fields; mixing documents loaded from another store version; classpath Jackson version mismatch.

Understand the failure class

Background: json.Marshal / "failed to marshal" errors in Go: why "unsupported type" happens and how to fix it — this error's family across 22 libraries.

Related errors

• Invalid JSON format for the input request:
• Failed to serialize the Document metadata:
• Unknown message type:
• 'spring.ai.chat.client.tool-search-advisor.tool-index-type=v
• Failed to convert JsonValue to string

AI-assisted analysis of spring-projects/spring-ai@98a7beda4f (2026-09-11). Data as JSON: /api/errors/49d336cba5efb740. Report an issue: GitHub.