apache/hadoop · error · InvalidRecordException

Null endpoint

Error message

Null endpoint

What it means

RegistryTypeUtils.retrieveAddressURLs converts a URI-typed endpoint's addresses into java.net.URL objects. Unlike retrieveAddressesUriType — which returns null for a null endpoint — this method deliberately rejects null and throws InvalidRecordException ('Null endpoint', with an empty path).

Source

Thrown at hadoop-common-project/hadoop-registry/src/main/java/org/apache/hadoop/registry/client/binding/RegistryTypeUtils.java:233

    String val = address.get(field);
    if (val == null) {
      throw new InvalidRecordException("", "Missing address field: " + field);
    }
    return val;
  }

  /**
   * Get the address URLs. Guranteed to return at least one address.
   * @param epr endpoint
   * @return the address as a URL
   * @throws InvalidRecordException if the type is wrong, there are no addresses
   * or the payload ill-formatted
   * @throws MalformedURLException address can't be turned into a URL
   */
  public static List<URL> retrieveAddressURLs(Endpoint epr)
      throws InvalidRecordException, MalformedURLException {
    if (epr == null) {
      throw new InvalidRecordException("", "Null endpoint");
    }
    List<String> addresses = retrieveAddressesUriType(epr);
    List<URL> results = new ArrayList<URL>(addresses.size());
    for (String address : addresses) {
      results.add(new URL(address));
    }
    return results;
  }

  /**
   * Validate the record by checking for null fields and other invalid
   * conditions
   * @param path path for exceptions
   * @param record record to validate. May be null
   * @throws InvalidRecordException on invalid entries
   */
  public static void validateServiceRecord(String path, ServiceRecord record)
      throws InvalidRecordException {

View on GitHub (pinned to 2add963021)

Solutions

1. Null-check the endpoint before calling: if (epr == null) skip or handle the absent case
2. Filter null endpoints out of record.external/internal lists before processing
3. Use retrieveAddressesUriType when a null-tolerant read (returns null) fits the control flow better

Example fix

// before
List<URL> urls = RegistryTypeUtils.retrieveAddressURLs(firstExternal(record)); // may be null

// after
Endpoint epr = firstExternal(record);
if (epr != null && AddressTypes.ADDRESS_URI.equals(epr.addressType)
    && epr.addresses != null && !epr.addresses.isEmpty()) {
  List<URL> urls = RegistryTypeUtils.retrieveAddressURLs(epr);
}

if (epr == null) {
  return Collections.emptyList(); // no endpoint -> no URLs
}
return RegistryTypeUtils.retrieveAddressURLs(epr);

static boolean isRetrievableEndpoint(Endpoint epr) {
  return epr != null
      && AddressTypes.ADDRESS_URI.equals(epr.addressType)
      && epr.addresses != null && !epr.addresses.isEmpty();
}

try {
  urls = RegistryTypeUtils.retrieveAddressURLs(epr);
} catch (InvalidRecordException e) {
  if ("Null endpoint".equals(e.getMessage())) {
    return Collections.emptyList();
  }
  throw e;
}

When it happens

Trigger: Passing a null Endpoint: taken from an empty external/internal list lookup, record.external.get(i) where the list contains nulls, or code assuming the null-tolerant behavior of retrieveAddressesUriType.

Common situations: Iterating ServiceRecord.external/internal without per-element null checks; endpoints cleared after record mutation; chaining accessors without intermediate validation.

Related errors

• Address type of %s does not match required type of %s
• No addresses in endpoint
• Null record
• null component type not allowed
• null value not allowed

AI-assisted analysis of apache/hadoop@2add963021 (2026-08-22). Data as JSON: /api/errors/6fb093658a18567c. Report an issue: GitHub.