Alibaba Cloud Elasticsearch provides some commands for you to create manual snapshots for the index data stored on your Elasticsearch cluster, store the snapshots in a shared repository, or restore data from the snapshots. This topic describes how to create manual snapshots and restore data from the snapshots.

Background information

The data backup and restoration of Alibaba Cloud Elasticsearch clusters depend on the elasticsearch-repository-oss plug-in. By default, the plug-in is installed on Alibaba Cloud Elasticsearch clusters and cannot be removed. For more information about this plug-in, see elasticsearch-repository-oss.

Prerequisites

  • Object Storage Service (OSS) is activated, and an OSS bucket is created. For more information, see Activate OSS and Create buckets.
    Important The storage class of the OSS bucket must be Standard. Elasticsearch does not support the Archive storage class. In addition, the OSS bucket must reside in the same region as your Elasticsearch cluster.
  • Your Elasticsearch cluster is in a normal state.

Precautions

  • Snapshots store only index data. The following information of an Elasticsearch cluster is not stored in snapshots: monitoring data (such as indexes whose names start with .monitoring or .security_audit), metadata, translogs, configurations, software packages, built-in and custom plug-ins, and logs.
  • You can run all the code provided in this topic in the Kibana console of your Elasticsearch cluster. For more information, see Log on to the Kibana console.

Create a repository

Create a repository named my_backup. 

PUT _snapshot/my_backup/
{
    "type": "oss",
    "settings": {
        "endpoint": "http://oss-cn-hangzhou-internal.aliyuncs.com",
        "access_key_id": "xxxx",
        "secret_access_key": "xxxxxx",
        "bucket": "xxxxxx",
        "compress": true,
        "chunk_size": "500mb",
        "base_path": "snapshot/"
    }
}
ParameterDescription
endpointThe internal endpoint of the OSS bucket. For more information about how to obtain the endpoint, see Regions and endpoints.
access_key_idThe AccessKey ID of your account. For more information about how to obtain the AccessKey ID, see Obtain an AccessKey pair.
secret_access_keyThe AccessKey secret of your account. For more information about how to obtain the AccessKey secret, see Obtain an AccessKey pair.
bucketThe name of the OSS bucket. For more information about how to obtain the name, see Create buckets.
compressSpecifies whether the data compression feature is enabled for snapshots. Valid values:
  • true: The data compression feature is enabled for snapshots. This feature applies only to index metadata, including the mappings and settings of indexes.
  • false: The data compression feature is disabled for snapshots. The default value is false.
chunk_sizeIf you want to upload large volumes of data to the OSS bucket, you can upload the data in multiple parts. In this case, you can use this parameter to set the size of each part. If the size of a part reaches the value of this parameter, the excess data is distributed to another part.
base_pathThe start location of the repository. The default value is the root directory. You can specify the directory where specific snapshots are stored. Example: snapshot/myindex/.

Query repository information

Query the information of all repositories

GET _snapshot

Query the information of a specific repository

GET _snapshot/my_backup

Create a snapshot

Create a snapshot for all enabled indexes

PUT _snapshot/my_backup/snapshot_1
The preceding command creates the snapshot_1 snapshot for all the enabled indexes and stores the snapshot in the my_backup repository. After you run the command, the system immediately returns a response and creates the snapshot. If you want the system to return a response after the snapshot is created, specify the wait_for_completion parameter in the command. This parameter blocks all API calls until the snapshot is created. If the total size of the indexes is large, the response is returned after a long period of time. 
PUT _snapshot/my_backup/snapshot_1?wait_for_completion=true
Note
  • A repository stores multiple snapshots. Each snapshot is a copy of all indexes, specific indexes, or a single index in a cluster.
  • The first snapshot is a full copy of the data in a cluster. Subsequent snapshots store only incremental data. If you create a subsequent snapshot, the system only adds data to or removes data from the previous snapshot. Therefore, less time is required to create a subsequent snapshot than the first snapshot.

Create a snapshot for specific indexes

By default, a snapshot contains all enabled indexes. If Kibana is used when you create a snapshot, you may want to ignore all diagnostic indexes (the .kibana indexes) because of limited disk space. In this case, you can run the following command to create a snapshot only for specific indexes: 
PUT _snapshot/my_backup/snapshot_2
{
    "indices": "index_1,index_2"
}

The preceding command creates a snapshot only for the index_1 and index_2 indexes.

Query snapshot information

Query the information of all snapshots

GET _snapshot/my_backup/_all
If the command is successfully run, the following result is returned:
{
  "snapshots": [
    {
      "snapshot": "snapshot_1",
      "uuid": "vIdSCkthTeGa0nSj4D****",
      "version_id": 5050399,
      "version": "5.5.3",
      "indices": [
        ".kibana"
      ],
      "state": "SUCCESS",
      "start_time": "2018-06-28T01:22:39.609Z",
      "start_time_in_millis": 1530148959609,
      "end_time": "2018-06-28T01:22:39.923Z",
      "end_time_in_millis": 1530148959923,
      "duration_in_millis": 314,
      "failures": [],
      "shards": {
        "total": 1,
        "failed": 0,
        "successful": 1
      }
    },
    {
      "snapshot": "snapshot_3",
      "uuid": "XKO_Uwz_Qu6mZrU3Am****",
      "version_id": 5050399,
      "version": "5.5.3",
      "indices": [
        ".kibana"
      ],
      "state": "SUCCESS",
      "start_time": "2018-06-28T01:25:00.764Z",
      "start_time_in_millis": 1530149100764,
      "end_time": "2018-06-28T01:25:01.482Z",
      "end_time_in_millis": 1530149101482,
      "duration_in_millis": 718,
      "failures": [],
      "shards": {
        "total": 1,
        "failed": 0,
        "successful": 1
      }
    }
  ]
}

Query the information of a specific snapshot based on the snapshot name

GET _snapshot/my_backup/snapshot_3
If the command is successfully run, the following result is returned:
{
  "snapshots": [
    {
      "snapshot": "snapshot_3",
      "uuid": "vIdSCkthTeGa0nSj4D****",
      "version_id": 5050399,
      "version": "5.5.3",
      "indices": [
        ".kibana"
      ],
      "state": "SUCCESS",
      "start_time": "2018-06-28T01:22:39.609Z",
      "start_time_in_millis": 1530148959609,
      "end_time": "2018-06-28T01:22:39.923Z",
      "end_time_in_millis": 1530148959923,
      "duration_in_millis": 314,
      "failures": [],
      "shards": {
        "total": 1,
        "failed": 0,
        "successful": 1
      }
    }
  ]
}

Call the _status API to query the information of a specific snapshot

GET _snapshot/my_backup/snapshot_3/_status
The _status API allows you to query the detailed information of a snapshot. The information includes both the status of the snapshot and the statistics about each index and shard. If the command is successfully run, the following result is returned: 
{
"snapshots": [
   {
      "snapshot": "snapshot_3",
      "repository": "my_backup",
      "state": "IN_PROGRESS", 
      "shards_stats": {
         "initializing": 0,
         "started": 1, 
         "finalizing": 0,
         "done": 4,
         "failed": 0,
         "total": 5
      },
      "stats": {
         "number_of_files": 5,
         "processed_files": 5,
         "total_size_in_bytes": 1792,
         "processed_size_in_bytes": 1792,
         "start_time_in_millis": 1409663054859,
         "time_in_millis": 64
      },
      "indices": {
         "index_3": {
            "shards_stats": {
               "initializing": 0,
               "started": 0,
               "finalizing": 0,
               "done": 5,
               "failed": 0,
               "total": 5
            },
            "stats": {
               "number_of_files": 5,
               "processed_files": 5,
               "total_size_in_bytes": 1792,
               "processed_size_in_bytes": 1792,
               "start_time_in_millis": 1409663054859,
               "time_in_millis": 64
            },
            "shards": {
               "0": {
                  "stage": "DONE",
                  "stats": {
                     "number_of_files": 1,
                     "processed_files": 1,
                     "total_size_in_bytes": 514,
                     "processed_size_in_bytes": 514,
                     "start_time_in_millis": 1409663054862,
                     "time_in_millis": 22
                }
             }
          }
        }
      }
    }
  ]
}

Delete a snapshot

You can run the following command to delete a specific snapshot. If the snapshot is being created, the system stops the creation and deletes the snapshot from the repository. 

DELETE _snapshot/my_backup/snapshot_3
Warning You can delete snapshots only by calling the DELETE API. You cannot manually delete a snapshot because the snapshot may be associated with the data in other snapshots. If you manually delete the snapshot, other snapshots may be damaged. If some data in the snapshot that you want to delete is associated with other snapshots, the DELETE API finds the data and deletes only the data that is not associated with other snapshots.

Restore indexes from a snapshot

Important
  • We recommend that you do not restore indexes whose names start with a period (.). If you restore these indexes, you may fail to access the Kibana console.
  • If an index with the same name as the index to be restored exists in your cluster, you must delete or disable the index in the cluster first. Otherwise, the restoration fails.
  • If you want to restore data in a snapshot across regions, you must migrate the data to an OSS bucket in the destination region and restore the data to the desired Elasticsearch cluster in the destination region. For more information, see Migrate data between Alibaba Cloud OSS buckets.

Create a shared OSS repository in the destination cluster

Before you restore data from a snapshot to a cluster, you must create a shared OSS repository in the cluster and map it to the same OSS endpoint as the snapshot. For more information, see Create a repository. 

PUT _snapshot/my_backup_restore/
{
    "type": "oss",
    "settings": {
        "endpoint": "http://oss-cn-hangzhou-internal.aliyuncs.com",
        "access_key_id": "xxxx",
        "secret_access_key": "xxxxxx",
        "bucket": "xxxxxx",
        "compress": true,
        "chunk_size": "500mb",
        "base_path": "snapshot/"
    }
}

Restore a specific index

If you only want to verify or process the data in an index and do not want to overwrite the data in the Elasticsearch cluster, use this method to restore the index. 
POST /_snapshot/my_backup/snapshot_1/_restore
{
 "indices": "index_1", 
 "rename_pattern": "index_(.+)", 
 "rename_replacement": "restored_index_$1" 
}
ParameterDescription
indicesThe name of the index that you want to restore. In this example, the system restores only the index_1 index from the specified snapshot.
rename_patternThe format of the index name. The system searches for the index that you want to restore based on the index name. The name must match the format you specified.
rename_replacementThe regular expression that is used to rename the index.

Restore all indexes other than indexes whose names start with a period (.)

POST _snapshot/my_backup/snapshot_1/_restore 
{"indices":"*,-.monitoring*,-.security*,-.kibana*","ignore_unavailable":"true"}

Restore all indexes, including indexes whose names start with a period (.)

POST _snapshot/my_backup/snapshot_1/_restore
  • For example, if the snapshot_1 snapshot contains five indexes, the preceding command restores all these indexes to the Elasticsearch cluster.
  • After you call the _restore API, the system immediately returns a response and restores the indexes. If you want to block all API calls until the restoration is complete, you can specify the wait_for_completion parameter in the command. 
    POST _snapshot/my_backup/snapshot_1/_restore?wait_for_completion=true

Query restoration information

You can call the _recovery API to query the information of an index restoration task, such as the status and progress of the task.

Query the restoration information of a specific index

GET restored_index_3/_recovery

Query the restoration information of all indexes

Note The information may include the information of shards that are not involved in the restoration process. 
GET /_recovery/
If the command is successfully run, the following result is returned: 
{
   "restored_index_3" : {
     "shards" : [ {
       "id" : 0,
       "type" : "snapshot",
       "stage" : "index",
       "primary" : true,
       "start_time" : "2014-02-24T12:15:59.716",
       "stop_time" : 0,
       "total_time_in_millis" : 175576,
       "source" : {
         "repository" : "my_backup",
         "snapshot" : "snapshot_3",
         "index" : "restored_index_3"
       },
       "target" : {
         "id" : "ryqJ5lO5S4-lSFbGnt****",
         "hostname" : "my.fqdn",
         "ip" : "10.0.**.**",
         "name" : "my_es_node"
       },
       "index" : {
         "files" : {
           "total" : 73,
           "reused" : 0,
           "recovered" : 69,
           "percent" : "94.5%"
         },
         "bytes" : {
           "total" : 79063092,
           "reused" : 0,
           "recovered" : 68891939,
           "percent" : "87.1%"
         },
         "total_time_in_millis" : 0
       },
       "translog" : {
         "recovered" : 0,
         "total_time_in_millis" : 0
       },
       "start" : {
         "check_index_time" : 0,
         "total_time_in_millis" : 0
       }
     } ]
   }
}
The returned result lists all indexes that are being restored and all shards of these indexes. The result contains the following information for each shard: restoration start time, restoration end time, restoration duration, restoration progress, and transmitted bytes. The following table describes some parameters in the preceding result.
ParameterDescription
typeThe type of the restoration. The value snapshot indicates that the shard is being restored from a snapshot.
sourceThe snapshot and repository to which the shard belongs.
percentThe progress of the restoration. The value 94.5% indicates that 94.5% of the data in the shard is restored.

Delete an index that is being restored from a snapshot

You can use the DELETE API to delete an index that is being restored to cancel the restoration of the index. 
DELETE /restored_index_3

If the restored_index_3 index is being restored, the preceding command stops the restoration and deletes the data that is restored to the Elasticsearch cluster.

References