如何监视滚动API为每个滚动ID使用的资源？调用clearScroll Api时会发生什么 [英] How to monitor resources used by scroll API for each scroll Id? what happens when clearScroll Api is called

问题描述

如何针对每个滚动ID监视滚动API所使用的内存资源或搜索上下文？这些搜索上下文存储在哪里，有没有办法查看或监视它们？参数keepAlive对Java api公共SearchRequest scroll（TimeValue keepAlive）{...}

How to monitor memory resources or search contexts used by scroll API for each scroll Id? where does these search contexts are stored, is there a way to see or monitor them? what does the parameter keepAlive mean for the java api public SearchRequest scroll(TimeValue keepAlive) {...}

Ref： https ：//www.elastic.co/guide/zh-CN/elasticsearch/client/java-rest/6.5/java-rest-high-clear-scroll.html#java-rest-high-clear-scroll-async

推荐答案

您可以通过调用节点统计信息API ，但是（我知道）无法查看有多少内存由他们使用：

You can check how many search contexts are currently opened by calling the Node stats API but there's no way (that I know of) to see how much memory is used by them:

GET _nodes/stats/indices?filter_path=**.open_contexts

=>

{
  "nodes" : {
    "QjjWj45NQkGC9qcaylDDVQ" : {
      "indices" : {
        "search" : {
          "open_contexts" : 0
        }
      }
    },
    "eZbpkckXSxSxzD8c_QOkEA" : {
      "indices" : {
        "search" : {
          "open_contexts" : 3
        }
      }
    },
    "0VFJaC2uTraudPCMqODhiw" : {
      "indices" : {
        "search" : {
          "open_contexts" : 2
        }
      }
    }
  }
}

搜索上下文存储在堆中。

The search contexts are stored on the heap.

此外， keepAlive 参数告诉上下文在自动清除之前在内存中保持可用状态的时间。

Furthermore, the keepAlive parameter tells how long the context stays available in memory before being cleared automatically.

这篇关于如何监视滚动API为每个滚动ID使用的资源？调用clearScroll Api时会发生什么的文章就介绍到这了，希望我们推荐的答案对大家有所帮助，也希望大家多多支持IT屋！