按原文讲透Elasticsearch生命周期策略ILM_elasticsearch ilm-优快云博客

LM moves indices through the lifecycle according to their age. To control the timing of these transitions, you set a minimum age for each phase. For an index to move to the next phase, all actions in the current phase must be complete and the index must be older than the minimum age of the next phase.（必须完成当前阶段的所有操作，且年龄大于下个阶段的最小年龄限制）

The minimum age defaults to zero, which causes ILM to move indices to the next phase as soon as all actions in the current phase complete.(如果年龄限制为0，则完成所有操作后立即进入下个阶段)

If an index has unallocated shards and the cluster health status is yellow, the index can still transition to the next phase according to its index lifecycle management policy. However, because Elasticsearch can only perform certain clean up tasks on a green cluster, there might be unexpected side effects.

To avoid increased disk usage and reliability issues, address any cluster health problems in a timely fashion.

所以我们可以看到一个阶段的典型定义如下

{
	"phases": {
#阶段
		"hot": {
#年龄限制
            "min_age":"0ms",
#操作目录
			"actions": {
#操作1
				"rollover": {
#操作1的option
					"max_docs": "5"
				}
			}
		}
	}
}

1.2阶段的概念

阶段按顺序分为hot,warm,cold,delete。

索引经历4个阶段，最后消失，每个索引某一时刻只能处于一种阶段。

HOT：索引被激活，更新，和查询。原文:The index is actively being updated and queried

WARM：索引不能被更新(不支持增删改)，可以查询。原文：The index is no longer being updated but is still being queried

COLD:索引不能被更新，很少被查询。查询速度变慢。原文：The index is no longer being updated and is seldom queried. The information still needs to be searchable, but it’s okay if those queries are slower.

DELETE：索引不再被需要了~。原文：The index is no longer needed and can safely be removed.

1.3操作的概念

action目录下，目前有12种操作，每个阶段只能执行其中的部分操作，比如在HOT阶段不能执行delete，对应关系如下表。

	HOT	WARM	COLD	DELETE
Allocate		√	√
Delete				√
Force merge	√	√
Freeze			√
Migrate		√	√
Read only		√
Rollover	√
Searchable snapshot	√	√	√	√
Set priority	√	√	√
Shrink		√
Unfollow	√	√	√
Wait for snapshot				√

1.3.1.Allocate分配

更新索引设置以更改代理分片的节点，并更改副本的数量。

1.3.2.Delete删除

删除索引而非数据

1.3.3.Force merge强制合并

强制将索引合并为指定的segment数量，并且索引变为Read only。

1.3.4.Freeze冻结

冻结索引使内存占用达到最小。

1.3.5.Migrate迁移

将索引移动到与当前阶段相对应的数据层。如果在allocate操作中没有指定分配选项，ILM会自动在warm和cold阶段注入迁移操作。

1.3.6.Read only只读

将索引变为只读。

1.3.7.Rollover滚动

根据模板创建新的索引，将当前写入的索引变为旧索引。

1.3.8.Searchable snapshot可搜索的快照

在配置的存储库中获取被管理的索引的快照，并将其挂载为可搜索的快照。如果索引是数据流的一部分，则将挂载的索引将替换流中的原始索引。这个比较多逻辑，建议看原文。

原文:Takes a snapshot of the managed index in the configured repository and mounts it as a searchable snapshot. If the index is part of a data stream, the mounted index replaces the original index in the stream.

1.3.9.Set priority设施优先级

设置阶段的优先级，在节点断电重启后，将按优先级高低启动阶段。所以需要从高到到低设置。Hot最高，以此类推。

1.3.10.Shrink收缩

将索引设置为只读，并将其缩小为具有较少初始分片的新索引。（创建索引时依赖template参数，可能会有较多分片）

1.3.11.Unfollow反跟随

将跨群集复制的跟随索引转换为常规索引。这使收缩、滚动和可搜索的快照操作能够在跟随索引上安全执行。

1.3.12.Wait for snapshot等待快照

删除前对索引做快照。不会完全丢失。

1.4Template模板的概念

模板是rollover操作创建索引时的依据。它阐述了1、索引自身的属性 2、索引关联的生命周期策略。

如下，

PUT _index_template/my_template
{
#索引的名称样式均为text-xxx
  "index_patterns": ["test-*"], 
  "template": {
    "settings": {
#索引的分区分片数量
      "number_of_shards": 1,
      "number_of_replicas": 1,
#索引绑定的生命周期策略
      "index.lifecycle.name": "my_policy", 
#索引指向的别名
      "index.lifecycle.rollover_alias": "test-alias-all" 
    }
  }
}

2.制作生命周期策略

以上概念了解后，就可以开始写一个生命周期策略了。

1、创建生命周期策略

PUT /_ilm/policy/test-policy
{
	"policy": {
		"phases": {
			"hot": {
				"min_age": "0ms",
				"actions": {
					"set_priority": {
						"priority": 100
					},
					"rollover": {
						"max_age": "30d"
					}
				}
			},
			"warm": {
				"min_age": "31d",
				"actions": {
					"forcemerge": {
						"max_num_segments": 1
					},
					"set_priority": {
						"priority": 50
					},
					"allocate": {
						"number_of_replicas": 1
					}
				}
			},
			"cold": {
				"min_age": "60d",
				"actions": {
					"set_priority": {
						"priority": 0
					}
				}
			},
			"delete": {
				"min_age": "360d",
				"actions": {
					"delete": {
						"delete_searchable_snapshot": true
					}
				}
			}
		}
	}
}

2、创建模板

见1.4定义

3、创建别名下得第一个索引

PUT /test-index-000001
{
	"aliases": {
		"test-alias-all": {
			"is_write_index": true
		}
	}
}

4、设置刷新时间

PUT /_cluster/settings
{
	"transient": {
		"indices.lifecycle.poll_interval": "5m"
	}
}

每次刷新会检查当前索引是否可以进入下一阶段