elasticsearch线程池解析

每个Elasticsearch节点内部都维护着多个线程池，如index、search、get、bulk等，用户可以修改线程池的类型和大小，线程池默认大小跟CPU逻辑一致。

一、查看当前线程组状态

[html]view plaincopy

curl-XGET'https://localhost:9200/_nodes/stats?pretty'[html]view plaincopy

"thread_pool":{

"bulk":{

"threads":32,

"queue":0,

"active":0,

"rejected":0,

"largest":32,

"completed":659997

},

"index":{

"threads":2,

"queue":0,

"active":0,

"rejected":0,

"largest":2,

"completed":2

}

[html]view plaincopy

threadpool:

generic:

keep_alive:2m[html]view plaincopy

threadpool:

index:

size:30

queue_size:1000

3、scaling

可变大小的pool，大小根据负载在1到size间，同样keep_alive参数指定了闲置线程被回收的时间。

[html]view plaincopy

threadpool:

warmer:

size:8

keep_alive:2m

四、修改线程池配置

1、elasticsearch.yml

[html]view plaincopy

threadpool.index.type:fixed

threadpool.index.size:100

threadpool.index.queue_size:500[html]view plaincopy

curl-XPUT'localhost:9200/_cluster/settings'-d'{

"transient":{

"threadpool.index.type":"fixed",

"threadpool.index.size":100,

"threadpool.index.queue_size":500

}

}'[html]view plaincopy

EsRejectedExcutionException[rejectedexecution(queuecapacity50)on.......] 这个错误明显是默认大小为50的队列(queue)处理不过来了，解决方法是增大bulk队列的长度[html]view plaincopy

threadpool.bulk.queue_size:1000