AbstractElasticsearchRepository : Cannot create index: Connection refused; nested exception is java.lang.RuntimeException: Connection refused

Question

I am implementing a microservices project that integrates Kafka, Elasticsearch and Kibana. I have configured network for communication between containers, but I can't get the data collection microservice to communicate with the container running an ElasticSearch oss image in its version 7.6.2.

I can see the following log when launching the Spring Boot microservice:

INFO 1 --- [ main] o.s.d.elasticsearch.support.VersionInfo : Version Spring Data Elasticsearch: 4.0.2.RELEASE
INFO 1 --- [ main] o.s.d.elasticsearch.support.VersionInfo : Version Elasticsearch Client in build: 7.6.2
INFO 1 --- [ main] o.s.d.elasticsearch.support.VersionInfo : Version Elasticsearch Client used: 7.6.2
WARN 1 --- [ main] .d.e.r.s.AbstractElasticsearchRepository : Cannot create index: Connection refused; nested exception is java.lang.RuntimeException: Connection refused

Below is the ElasticSearch container configuration:

elasticsearch:
    image: docker.elastic.co/elasticsearch/elasticsearch-oss:7.6.2
    environment:
      - discovery.type=single-node
      - cluster.name=covid-tweets-es-cluster
      - bootstrap.memory_lock=true
      - "ES_JAVA_OPTS=-Xms512m -Xmx512m"
    ulimits:
      memlock:
        soft: -1
        hard: -1
    volumes:
      - elasticsearch-data:/usr/share/elasticsearch/data
    ports:
      - 9300:9300
      - 9200:9200
    networks:
      - covid_processor_network
    

This is the Spring Data ElasticSearch configuration of the microservice:

## Kafka Binder Props
spring.cloud.stream.kafka.binder.brokers: kafka:9092
spring.cloud.stream.kafka.binder.auto-create-topics: false
spring.cloud.stream.kafka.binder.configuration.auto.offset.reset: latest
spring.cloud.stream.bindings.processed-tweets.group: tweets-collector
## Persistence
spring.data.elasticsearch.repositories.enabled: true
spring.data.elasticsearch.cluster-nodes: elasticsearch:9300
spring.data.elasticsearch.cluster-name: covid-tweets-es-cluster
spring.data.elasticsearch.rest.uris: elasticsearch:9200

From the microservice container I can connect to the elasticsearch container

This is the client code, it is quite simple, it simply uses a TweetReopistory repository that implements ElasticsearchRepository

@StreamListener(AppStreamsConfig.PROCESSED_TWEETS_CHANNEL)
    public void onNewProcessedTweet(
            @Payload final TweetDTO newProcessedTweet,
            @Header(KafkaHeaders.RECEIVED_TOPIC) String topic,
            @Header(KafkaHeaders.RECEIVED_PARTITION_ID) Integer partition,
            @Header(KafkaHeaders.OFFSET) Long offset,
            @Header(IntegrationMessageHeaderAccessor.DELIVERY_ATTEMPT) Integer deliveryAttempt) {

        log.info("NewsProcessedTweet with id '{}' and text '{}' received from bus. topic: {}, partition: {}, offset: {}, deliveryAttempt: {}",
                newProcessedTweet.getId(), newProcessedTweet.getText(), topic, partition, offset, deliveryAttempt);

        try {
            tweetService.save(newProcessedTweet);
        } catch (final Exception ex) {
            ex.printStackTrace();
            log.error("Collect Tweet Exception -> " + ex.getMessage());
        }
    }

@Service("tweetsService")
@RequiredArgsConstructor
public class TweetsServiceImpl implements ITweetsService {

    /**
     * Tweets Repository
     */
    private final TweetsRepository tweetsRepository;
    private final TweetDtoMapper tweetDtoMapper;

    @Override
    public void save(TweetDTO tweetDto) {
        Assert.notNull(tweetDto, "Tweet can not be null");

        final TweetEntity tweetToSave = tweetDtoMapper.dtoToEntity(tweetDto);
        tweetsRepository.save(tweetToSave);
    }

Anyone know what I am doing wrong? Thank you

Answer 1

Which client are you using? RestHighLevelClient or TransportClient? Try with 9200 port if using RestHighLevelClient.

Share the client usage code as well to understand better.