IMPORTANT: No additional bug fixes or documentation updates will be released for this version. For the latest information, see the current release documentation.
« Start Snapshot Lifecycle Management Multiple deployments writing to the same snapshot repository »
Elastic Docs ›Elasticsearch Guide [8.12] ›Troubleshooting

Restore from snapshot

edit

Restore from snapshot

edit

Elasticsearch is using snapshots to store a copy of your data outside a cluster. You can restore a snapshot to recover indices and data streams for which there are no copies of the shards in the cluster. This can happen if the data (indices or data streams) was deleted or if the cluster membership changed and the current nodes in the system do not contain a copy of the data anymore.

Restoring the missing data requires you to have a backup of the affected indices and data streams that is up-to-date enough for your use case. Please do not proceed without confirming this.

In order to restore the indices and data streams that are missing data:

Use Kibana

  1. Log in to the Elastic Cloud console.

  2. On the Elasticsearch Service panel, click the name of your deployment.

    If the name of your deployment is disabled your Kibana instances might be unhealthy, in which case please contact Elastic Support. If your deployment doesn’t include Kibana, all you need to do is enable it first.

  3. Open your deployment’s side navigation menu (placed under the Elastic logo in the upper left corner) and go to Dev Tools > Console.

    Kibana Console

  4. To view the affected indices using the cat indices API.

    response = client.cat.indices(
  v: true,
  health: 'red',
  h: 'index,status,health'
)
puts response
    GET _cat/indices?v&health=red&h=index,status,health

    The response will look like this:

    index                                status health
.ds-my-data-stream-2022.06.17-000001 open   red
kibana_sample_data_flights           open   red

    The red health of the indices above indicates that these indices are missing primary shards, meaning they are missing data.

  5. In order to restore the data we need to find a snapshot that contains these two indices. To find such a snapshot use the get snapshot API.

    response = client.snapshot.get(
  repository: 'my_repository',
  snapshot: '*',
  verbose: false
)
puts response
    GET _snapshot/my_repository/*?verbose=false

    The response will look like this:

    {
  "snapshots" : [
    {
      "snapshot" : "snapshot-20200617",                                     
      "uuid" : "dZyPs1HyTwS-cnKdH08EPg",
      "repository" : "my_repository",                                       
      "indices" : [                                                         
        ".apm-agent-configuration",
        ".apm-custom-link",
        ".ds-ilm-history-5-2022.06.17-000001",
        ".ds-my-data-stream-2022.06.17-000001",
        ".geoip_databases",
        ".kibana-event-log-8.2.2-000001",
        ".kibana_8.2.2_001",
        ".kibana_task_manager_8.2.2_001",
        "kibana_sample_data_ecommerce",
        "kibana_sample_data_flights",
        "kibana_sample_data_logs"
      ],
      "data_streams" : [ ],
      "state" : "SUCCESS"                                                     
    }
  ],
  "total" : 1,
  "remaining" : 0
}

    The name of the snapshot.

    The repository of the snapshot.

    The indices backed up in the snapshot.

    If the snapshot was successful.
  6. The snapshot snapshot-20200617 contains the two indices we want to restore. You might have multiple snapshots from which you could restore the target indices. Choose the latest snapshot.

  7. Now that we found a snapshot, we will proceed with the data stream preparation for restoring the lost data. We will check the index metadata to see if any index is part of a data stream:

    response = client.indices.get(
  index: 'kibana_sample_data_flights,.ds-my-data-stream-2022.06.17-000001',
  features: 'settings',
  flat_settings: true
)
puts response
    GET kibana_sample_data_flights,.ds-my-data-stream-2022.06.17-000001?features=settings&flat_settings

    The response will look like this:

    {
  ".ds-my-data-stream-2022.06.17-000001" : {                                
    "aliases" : { },
    "mappings" : { },
    "settings" : {                                                          
      "index.creation_date" : "1658406121699",
      "index.hidden" : "true",
      "index.lifecycle.name" : "my-lifecycle-policy",
      "index.number_of_replicas" : "1",
      "index.number_of_shards" : "1",
      "index.provided_name" : ".ds-my-data-stream-2022.06.17-000001",
      "index.routing.allocation.include._tier_preference" : "data_hot",
      "index.uuid" : "HmlFXp6VSu2XbQ-O3hVrwQ",
      "index.version.created" : "8020299"
    },
    "data_stream" : "my-data-stream"                                        
  },
  "kibana_sample_data_flights" : {                                          
    "aliases" : { },
    "mappings" : { },
    "settings" : {
      "index.creation_date" : "1655121541454",
      "index.number_of_replicas" : "0",
      "index.number_of_shards" : "1",
      "index.provided_name" : "kibana_sample_data_flights",
      "index.routing.allocation.include._tier_preference" : "data_content",
      "index.uuid" : "jMOlwKPPSzSraeeBWyuoDA",
      "index.version.created" : "8020299"
    }
  }
}

    The name of an index.

    The settings of this index that contains the metadata we are looking for.

    This indicates that this index is part of a data stream and displays the data stream name.

    The name of the other index we requested.

    The response above shows that kibana_sample_data_flights is not part of a data stream because it doesn’t have a field called data_stream in the settings.

    On the contrary, .ds-my-data-stream-2022.06.17-000001 is part of the data stream called my-data-stream. When you find an index like this, which belongs to a data stream, you need to check if data are still being indexed. You can see that by checking the settings, if you can find this property: "index.lifecycle.indexing_complete" : "true", it means that indexing is completed in this index and you can continue to the next step.

    If index.lifecycle.indexing_complete is not there or is configured to false you need to rollover the data stream so you can restore the missing data without blocking the ingestion of new data. The following command will achieve that.

    response = client.indices.rollover(
  alias: 'my-data-stream'
)
puts response
    POST my-data-stream/_rollover

  8. Now that the data stream preparation is done, we will close the target indices by using the close indices API.

    response = client.indices.close(
  index: 'kibana_sample_data_flights,.ds-my-data-stream-2022.06.17-000001'
)
puts response
    POST kibana_sample_data_flights,.ds-my-data-stream-2022.06.17-000001/_close

    You can confirm that they are closed with the cat indices API.

    response = client.cat.indices(
  v: true,
  health: 'red',
  h: 'index,status,health'
)
puts response
    GET _cat/indices?v&health=red&h=index,status,health

    The response will look like this:

    index                                status health
.ds-my-data-stream-2022.06.17-000001 close   red
kibana_sample_data_flights           close   red

  9. The indices are closed, now we can restore them from snapshots without causing any complications using the restore snapshot API:

    response = client.snapshot.restore(
  repository: 'my_repository',
  snapshot: 'snapshot-20200617',
  body: {
    indices: 'kibana_sample_data_flights,.ds-my-data-stream-2022.06.17-000001',
    include_aliases: true
  }
)
puts response
    POST _snapshot/my_repository/snapshot-20200617/_restore
{
  "indices": "kibana_sample_data_flights,.ds-my-data-stream-2022.06.17-000001", 
  "include_aliases": true                                                       
}

    The indices to restore.

    We also want to restore the aliases.

    If any feature states need to be restored we’ll need to specify them using the feature_states field and the indices that belong to the feature states we restore must not be specified under indices. The Health API returns both the indices and feature_states that need to be restored for the restore from snapshot diagnosis. e.g.:

    response = client.snapshot.restore(
  repository: 'my_repository',
  snapshot: 'snapshot-20200617',
  body: {
    feature_states: [
      'geoip'
    ],
    indices: 'kibana_sample_data_flights,.ds-my-data-stream-2022.06.17-000001',
    include_aliases: true
  }
)
puts response
    POST _snapshot/my_repository/snapshot-20200617/_restore
{
  "feature_states": [ "geoip" ],
  "indices": "kibana_sample_data_flights,.ds-my-data-stream-2022.06.17-000001",
  "include_aliases": true
}

  10. Finally we can verify that the indices health is now green via the cat indices API.

    response = client.cat.indices(
  v: true,
  index: '.ds-my-data-stream-2022.06.17-000001,kibana_sample_data_flightsh=index,status,health'
)
puts response
    GET _cat/indices?v&index=.ds-my-data-stream-2022.06.17-000001,kibana_sample_data_flightsh=index,status,health

    The response will look like this:

    index                                status health
.ds-my-data-stream-2022.06.17-000001 open   green
kibana_sample_data_flights           open   green

    As we can see above the indices are green and open. The issue is resolved.

For more guidance on creating and restoring snapshots see this guide.

« Start Snapshot Lifecycle Management Multiple deployments writing to the same snapshot repository »