java - Elastic Search java (spring) 索引问题

Question

我已经将elastic-search 1.7.1与spring应用程序集成。我有一个cron作业，它在每次运行时更新elasticsearch的索引。 我遵循了 github 上提供的各种示例代码来使其工作。首先，我有 Autowired ElasticSearchOperations 用于索引目的:

@Autowired
private ElasticsearchOperations elasticsearchOperations;

然后按以下方式执行内部循环索引

for(int i=0;i<list.size();i++)
{
    CategoryProductSearch search = new CategoryProductSearch();
    // set data to fields
    System.out.println("BEFORE SAVING DATA");
    IndexQuery indexQuery =new      

    IndexQueryBuilder().withId(search.getId()).withObject(search).build();
    //indexQuery.setId(search.getId());
    //indexQuery.setObject(search);
    //elasticsearchOperations.createIndex(CategoryProductSearch.class);
    elasticsearchOperations.putMapping(CategoryProductSearch.class);
    elasticsearchOperations.index(indexQuery);
    elasticsearchOperations.refresh(CategoryProductSearch.class,true);

    System.out.println("SAVING DATA");
 }

当我第一次运行它时，它按预期工作。我已在 config 文件夹内的 elasticsearch.yml 中将集群重命名为“mycluster”。第一次运行后我可以看到创建的文件夹。索引和搜索(在另一个文件中实现)工作得很好。但有时代码会卡在下面提到的行并显示连续警告 [Chase Stein] 节点 null 不是集群的一部分 Cluster [elasticsearch]，忽略...

 elasticsearchOperations.putMapping(CategoryProductSearch.class);

然后一段时间后它抛出 NoNodeAvailableException。我已经阅读过有关此问题的信息，它说可能没有足够的磁盘空间供 Elasticsearch 索引数据。 我是 Spring 新手，第一次尝试 Elasticsearch 。这是磁盘空间问题还是我索引数据的方式有问题？ 另外，如果我从/data 目录中手动删除“mycluster”文件夹并重新启动应用程序，它会再次正常工作。!

我已在本地 PC 上设置好所有内容。每当我重新启动elasticsearch服务时，就会出现这个问题。

异常的堆栈跟踪是:

  org.elasticsearch.action.UnavailableShardsException: [mycluster][0]       

  Primary shard is not active or isn't assigned to a known node. Timeout:    

  [1m], request: index {[mycluster][categoryproductsearch][1], 

  source[{// Source string }]
at 
  org.elasticsearch.action.support.replication
  .TransportShardReplicationOperationAction$PrimaryPhase
  .retryBecauseUnavailable
  (TransportShardReplicationOperationAction.java:655)
at 
   org.elasticsearch.action.support.replication
  .TransportShardReplicationOperationAction$PrimaryPhase.doRun
  (TransportShardReplicationOperationAction.java:362)
at 

  org.elasticsearch.common.util.concurrent.AbstractRunnable.run
  (AbstractRunnable.java:36)
at 
   org.elasticsearch.action.support.replication.
   TransportShardReplicationOperationAction$PrimaryPhase$3.onTimeout
   (TransportShardReplicationOperationAction.java:515)
at  

 org.elasticsearch.cluster.ClusterStateObserver$ObserverClusterStateListener
 .onTimeout
 (ClusterStateObserver.java:231)
at    
     org.elasticsearch.cluster.service.
    InternalClusterService$NotifyTimeout.run
    (InternalClusterService.java:560)
at 
     java.util.concurrent.ThreadPoolExecutor.runWorker
     (ThreadPoolExecutor.java:1145)
at 
     java.util.concurrent.ThreadPoolExecutor$Worker.run
     (ThreadPoolExecutor.java:615)
at 
      java.lang.Thread.run(Thread.java:745)

Answer 1

嗨，我从 generator-jhipster-elasticsearch-reindexer 获取了此代码

@Transactional(readOnly = true)
@SuppressWarnings("unchecked")
private <T> void reindexForClass(Class<T> entityClass, JpaRepository<T, Long> jpaRepository,
                                                      ElasticsearchRepository<T, Long> elasticsearchRepository) {
    elasticsearchTemplate.deleteIndex(entityClass);
    try {
        elasticsearchTemplate.createIndex(entityClass);
    } catch (IndexAlreadyExistsException e) {
        // Do nothing. Index was already concurrently recreated by some other service.
    }
    elasticsearchTemplate.putMapping(entityClass);
    if (jpaRepository.count() > 0) {
        try {
            Method m = jpaRepository.getClass().getMethod("findAllWithEagerRelationships");
            elasticsearchRepository.save((List<T>) m.invoke(jpaRepository));
        } catch (Exception e) {
            elasticsearchRepository.save(jpaRepository.findAll());
        }
    }
    log.info("Elasticsearch: Indexed all rows for " + entityClass.getSimpleName());
}

正如您所看到的，索引首先被删除，然后再次创建，并放置了映射，我认为您的顺序是错误的，并且会导致一些损坏的分片。您可以在 localhost:9200 上访问 Elastic Rest API 并尝试获取请求/_cat/indices 以查看您的索引。