记一次elasticsearch错误(SearchPhaseExecutionException: all shards failed)

一、问题

logstash突然无法对elasticsearch服务进行读写操作了,提示elasticsearch的地址有问题,检测elasticsearch发现端口存在。查看日志发现有报错。

二、问题原因

有一些索引的数据损坏了

三、解决

官网文档:https://www.elastic.co/guide/cn/elasticsearch/guide/cn/_cluster_health.html

  • 查看索引信息

    1
    2
    3
    4
    5
    6
    7
    8
    9
    10
    11
    12
    13
    14
    15
    16
    17
    18
    19
    curl -XGET 'http://localhost:9200/_cluster/health?pretty=true'

    {
    "cluster_name" : "test",
    "status" : "yellow",
    "timed_out" : false,
    "number_of_nodes" : 1,
    "number_of_data_nodes" : 1,
    "active_primary_shards" : 134,
    "active_shards" : 134,
    "relocating_shards" : 0,
    "initializing_shards" : 0,
    "unassigned_shards" : 30,
    "delayed_unassigned_shards" : 0,
    "number_of_pending_tasks" : 0,
    "number_of_in_flight_fetch" : 0,
    "task_max_waiting_in_queue_millis" : 0,
    "active_shards_percent_as_number" : 98.19711538461539 #数据的正常率,100表示一切ok
    }
  • 查看所有索引状态

    green:所有的主分片和副本分片都已分配。你的集群是 100% 可用的。
    yellow:所有的主分片已经分片了,但至少还有一个副本是缺失的。不会有数据丢失,所以搜索结果依然是完整的。不过,你的高可用性在某种程度上被弱化。如果 更多的 分片消失,你就会丢数据了。把 yellow 想象成一个需要及时调查的警告。
    red:至少一个主分片(以及它的全部副本)都在缺失中。这意味着你在缺少数据:搜索只能返回部分数据,而分配到这个分片上的写入请求会返回一个异常。

    1
    curl  'localhost:9200/_cat/indices?v'
  • 删除Red索引

    1
    curl -XDELETE http://localhost:9200/index_name