Searchable snapshotedit

Phases allowed: hot, cold, frozen.

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.

The searchable_snapshot action requires data tiers. The action uses the index.routing.allocation.include._tier_preference setting to mount the index directly to the phase’s corresponding data tier. In the frozen phase, the action mounts a partially mounted index to the frozen tier. In other phases, the action mounts a fully mounted index to the corresponding data tier.

If the searchable_snapshot action is used in the hot phase the subsequent phases cannot include the shrink, forcemerge, or freeze actions.

This action cannot be performed on a data stream’s write index. Attempts to do so will fail. To convert the index to a searchable snapshot, first manually roll over the data stream. This creates a new write index. Because the index is no longer the stream’s write index, the action can then convert it to a searchable snapshot. Using a policy that makes use of the rollover action in the hot phase will avoid this situation and the need for a manual rollover for future managed indices.

Mounting and relocating the shards of searchable snapshot indices involves copying the shard contents from the snapshot repository. This may incur different costs from the copying between nodes that happens with regular indices. These costs are typically lower but, in some environments, may be higher. See Reduce costs with searchable snapshots for more details.

By default, this snapshot is deleted by the delete action in the delete phase. To keep the snapshot, set delete_searchable_snapshot to false in the delete action.


(Required, string) Repository used to store the snapshot.
(Optional, Boolean) Force merges the managed index to one segment. Defaults to true. If the managed index was already force merged using the force merge action in a previous action the searchable snapshot action force merge step will be a no-op.

The forcemerge action is best effort. It might happen that some of the shards are relocating, in which case they will not be merged. The searchable_snapshot action will continue executing even if not all shards are force merged.


PUT _ilm/policy/my_policy
  "policy": {
    "phases": {
      "cold": {
        "actions": {
          "searchable_snapshot" : {
            "snapshot_repository" : "backing_repo"