ElasticSearch避免100％的JVM堆

Question

如何避免ElasticSearch中100％的JVM堆并且垃圾收集器清理堆？

在我的情况下，当JVM堆达到99％时，该站点不再响应。一些提示？

这是Marvel的一个屏幕。

elasticsearch.yml config

cluster.name：xxx

node.name: xxxx
node.data: true
node.master: true
bootstrap.mlockall: true
transport.tcp.compress: true;
transport.tcp.port: 9300
http.port: 9200
http.max_content_length: 500mb

index.search.slowlog.threshold.query.warn: 10s
index.search.slowlog.threshold.query.info: 5s
index.search.slowlog.threshold.query.debug: 2s
index.search.slowlog.threshold.query.trace: 500ms

script.engine.groovy.inline.aggs: on
script.inline: on
script.indexed: on
index.max_result_window: 10000

9G索引的堆大小为24g。

编辑 - GC日志

: 553545K->8330K(613440K), 0.0557428 secs] 732889K->188315K(25097728K), 0.0558526 secs] [Times: user=0.40 sys=0.00, real=0.06 secs]
2017-06-05T10:55:25.416+0200: 88.445: Total time for which application threads were stopped: 0.0561829 seconds, Stopping threads took: 0.0000684 seconds
2017-06-05T10:55:26.416+0200: 89.446: Total time for which application threads were stopped: 0.0002122 seconds, Stopping threads took: 0.0000832 seconds
2017-06-05T10:55:26.833+0200: 89.863: [GC (Allocation Failure) 2017-06-05T10:55:26.834+0200: 89.863: [ParNew
Desired survivor size 34865152 bytes, new threshold 6 (max 6)
- age   1:     530096 bytes,     530096 total
- age   2:     920856 bytes,    1450952 total
- age   3:    1311064 bytes,    2762016 total
- age   4:     258584 bytes,    3020600 total
- age   5:     600504 bytes,    3621104 total
- age   6:     717384 bytes,    4338488 total
: 553674K->5918K(613440K), 0.0450758 secs] 733659K->186089K(25097728K), 0.0451704 secs] [Times: user=0.35 sys=0.00, real=0.04 secs]