Shrink

Phases allowed: hot, warm.

Blocks writes on a source index and shrinks it into a new index with fewer primary shards. The name of the resulting index is shrink-<random-uuid>-<original-index-name>. This action corresponds to the shrink API.

After the shrink action, any aliases that pointed to the source index point to the new shrunken index. If ILM performs the shrink action on a backing index for a data stream, the shrunken index replaces the source index in the stream. You cannot perform the shrink action on a write index.

To use the shrink action in the hot phase, the rollover action must be present. If no rollover action is configured, ILM will reject the policy.

The shrink action will unset the index’s index.routing.allocation.total_shards_per_node setting, meaning that there will be no limit. This is to ensure that all shards of the index can be copied to a single node. This setting change will persist on the index even after the step completes.

response = client.ilm.put_lifecycle(
  policy: 'my_policy',
  body: {
    policy: {
      phases: {
        warm: {
          actions: {
            shrink: {
              number_of_shards: 1
            }
          }
        }
      }
    }
  }
)
puts response

PUT _ilm/policy/my_policy
{
  "policy": {
    "phases": {
      "warm": {
        "actions": {
          "shrink" : {
            "number_of_shards": 1
          }
        }
      }
    }
  }
}

response = client.ilm.put_lifecycle(
  policy: 'my_policy',
  body: {
    policy: {
      phases: {
        warm: {
          actions: {
            shrink: {
              max_primary_shard_size: '50gb'
            }
          }
        }
      }
    }
  }
)
puts response

PUT _ilm/policy/my_policy
{
  "policy": {
    "phases": {
      "warm": {
        "actions": {
          "shrink" : {
            "max_primary_shard_size": "50gb"
          }
        }
      }
    }
  }
}