目录

  • ElasticSearch 设置备份文件地址
  • 注册快照存储库
  • 查看快照存储库保存结果
  • 创建快照
  • 异步创建
  • 指定索引进行快照
  • 在服务器查看备份的数据
  • 恢复数据
  • 本机恢复
  • 其他服务器恢复
  • 常见问题
  • 报错 doesn't match any of the locations specified by path.repo because this setting is empty

ElasticSearch 设置备份文件地址

  • 在elasticsearch.yml 新增
path.repo: ["/home/es/snapshot"]

注册快照存储库

PUT _snapshot/ecopherebbs
{
    "type": "fs",
    "settings": {
        "location": "/home/es/snapshot"
    }
}

保存成功

基于hfds的es备份恢复_基于hfds的es备份恢复

查看快照存储库保存结果

GET _snapshot

返回结果

{
  "ecopherebbs" : {
    "type" : "fs",
    "settings" : {
      "location" : "/home/es/snapshot"
    }
  }
}

基于hfds的es备份恢复_集群环境_02

创建快照

异步创建

  • 创建
PUT _snapshot/ecopherebbs/snapshot_20210631
{
  "ignore_unavailable": true,
   "include_global_state": true
}
  • 查看状态
GET _snapshot/ecopherebbs/snapshot_20210631

处理中

{
  "snapshots" : [
    {
      "snapshot" : "snapshot_20210631",
      "uuid" : "PTBkWV-pQ8CQ7l4RNxISfA",
      "version_id" : 7090399,
      "version" : "7.9.3",
      "indices" : [
        "t_e_s_t___2_0_2_1_0_4_2_1_0_9_2_9_index",
        "test0627",
        "test110",
        ".kibana-event-log-7.9.3-000001",
        "fang_fa_index",
        "filebeat-7.9.3-2021.06.29-000001",
        "gateway_requests",
        "my_index",
        "ilm-history-2-000003",
        ".kibana-event-log-7.9.3-000002",
        ".kibana_1"
      ],
      "data_streams" : [ ],
      "include_global_state" : true,
      "state" : "IN_PROGRESS",
      "start_time" : "2021-06-30T05:26:10.530Z",
      "start_time_in_millis" : 1625030770530,
      "end_time" : "1970-01-01T00:00:00.000Z",
      "end_time_in_millis" : 0,
      "duration_in_millis" : 0,
      "failures" : [ ],
      "shards" : {
        "total" : 0,
        "failed" : 0,
        "successful" : 0
      }
    }
  ]
}

处理完成

{
  "snapshots" : [
    {
      "snapshot" : "snapshot_20210631",
      "uuid" : "PTBkWV-pQ8CQ7l4RNxISfA",
      "version_id" : 7090399,
      "version" : "7.9.3",
      "indices" : [
        "t_e_s_t___2_0_2_1_0_4_2_1_0_9_2_9_index",
        "test0627",
        "test110",
        ".kibana-event-log-7.9.3-000001",
        "fang_fa_index",
        "filebeat-7.9.3-2021.06.29-000001",
        "gateway_requests",
        "my_index",
        "ilm-history-2-000003",
        ".kibana-event-log-7.9.3-000002",
        ".kibana_1"
      ],
      "data_streams" : [ ],
      "include_global_state" : true,
      "state" : "SUCCESS",
      "start_time" : "2021-06-30T05:26:10.530Z",
      "start_time_in_millis" : 1625030770530,
      "end_time" : "2021-06-30T05:26:16.937Z",
      "end_time_in_millis" : 1625030776937,
      "duration_in_millis" : 6407,
      "failures" : [ ],
      "shards" : {
        "total" : 200,
        "failed" : 0,
        "successful" : 200
      }
    }
  ]
}

指定索引进行快照

PUT _snapshot/ecopherebbs/snapshot_20210630?wait_for_completion=true
{
  "indices": "index1,index2",
  "ignore_unavailable": true,
   "include_global_state": true
}
  1. wait_for_completion=true时会一直阻塞直到快照完成，kibana执行时超时为30秒，超时后快照依然会在后台执行
  2. snapshot_20200729 就是指定的备份快照命中
  3. 请求体重indices指定需要备份的索引（多个用逗号隔开），也可以不指定，这样会备份所有的es索引库数据

在服务器查看备份的数据

每次执行快照都会生成一个 meta和一个snap文件

[es@test12 snapshot]$ ls -lrt
总用量 1404
drwxrwxr-x. 62 es es   4096 6月  30 13:23 indices
-rw-rw-r--.  1 es es 460329 6月  30 13:24 meta--WCM58RpQNm5bEywCjO_cg.dat
-rw-rw-r--.  1 es es   1596 6月  30 13:24 snap--WCM58RpQNm5bEywCjO_cg.dat
-rw-rw-r--.  1 es es   1596 6月  30 13:26 snap-PTBkWV-pQ8CQ7l4RNxISfA.dat
-rw-rw-r--.  1 es es 460329 6月  30 13:26 meta-PTBkWV-pQ8CQ7l4RNxISfA.dat
-rw-rw-r--.  1 es es    302 6月  30 13:52 snap-N3zRMDnDTEqoiShEwjKosA.dat
-rw-rw-r--.  1 es es 460329 6月  30 13:52 meta-N3zRMDnDTEqoiShEwjKosA.dat
-rw-rw-r--.  1 es es  24990 6月  30 13:52 index-2
-rw-rw-r--.  1 es es      8 6月  30 13:52 index.latest

恢复数据

本机恢复

POST _snapshot/ecopherebbs/snapshot_20210631_02/_restore

执行恢复要将之前的索引删除，否则会报错

POST /_snapshot/my_backup/snapshot_1/_restore
{
    "indices": "index_1", 
    "rename_pattern": "index_(.+)", 
    "rename_replacement": "restored_index_$1" 
}

其他服务器恢复

  1. 配置第一步设置的目录
    path.repo: ["/home/es/snapshot"]
  2. 创建仓库
PUT _snapshot/ecopherebbs
{
    "type": "fs",
    "settings": {
        "location": "/home/es/snapshot-test"
    }
}
  1. 执行恢复索引
POST _snapshot/ecopherebbs/snapshot_20210631_02/_restore

常见问题

报错 doesn't match any of the locations specified by path.repo because this setting is empty

{
  "error" : {
    "root_cause" : [
      {
        "type" : "repository_exception",
        "reason" : "[ecopherebbs] location [/home/es/snapshot] doesn't match any of the locations specified by path.repo because this setting is empty"
      }
    ],
    "type" : "repository_exception",
    "reason" : "[ecopherebbs] failed to create repository",
    "caused_by" : {
      "type" : "repository_exception",
      "reason" : "[ecopherebbs] location [/home/es/snapshot] doesn't match any of the locations specified by path.repo because this setting is empty"
    }
  },
  "status" : 500
}
  • 解决方法：

在为es集群环境做灾备和恢复时候，首先需要创建创建一个仓库，并往仓库中存放快照（每个快照中会区分不同的索引）。但是在创建仓库的时候，要求仓库的地址必须在每个集群环境中的elasticsearch.yml中进行配置