Elasticsearch TOO_MANY_REQUESTS/

原文链接 : https://blog.csdn.net/piaoxue820/article/details/123638653

ES插入大量的数据时报错：TOO_MANY_REQUESTS/12/disk usage exceeded flood-stage watermark, index has read-only-allow-delete block 的解决方法
原因：
是因为一次请求中批量插入的数据条数巨多，以及短时间内的请求次数巨多引起ES节点服务器内存超过限制，ES主动给索引上锁。
解决方法:

PUT _all/_settings
{
  "index.blocks.read_only_allow_delete": null
}

解决
紧急给Elasticsearch的硬盘扩容，扩容完毕后执行以下语句关闭索引的只读状态：

PUT _all/_settings
{
  "index.blocks.read_only_allow_delete": null
}

如果来不及给Elasticsearch硬盘扩容，可以先关闭磁盘分配保护，让最后仅有的5%的磁盘空间缓冲一点时间，然后再给硬盘扩容。

关闭磁盘分配保护

PUT _cluster/settings
{
  "transient": {
    "cluster.routing.allocation.disk.threshold_enabled": false
  }
}

关闭索引的只读状态

PUT _all/_settings
{
  "index.blocks.read_only_allow_delete": null
}

#! this request accesses system indices: [.apm-agent-configuration, .apm-custom-link, .async-search, .kibana_1, .kibana_7.12.0_001, .kibana_task_manager_1, .kibana_task_manager_7.12.0_001, .reporting-2019.12.01, .reporting-2020.11.08, .reporting-2020.12.06, .reporting-2021-11-07, .security-7, .tasks], but in a future major version, direct access to system indices will be prevented by default
{
  "acknowledged" : true
}

磁盘扩容完毕后，启用磁盘分配保护

PUT _cluster/settings
{
  "transient": {
    "cluster.routing.allocation.disk.threshold_enabled": true
  }
}