Shrink an index into a new index with fewer primary shards.
Before you can shrink an index:
- The index must be read-only.
- A copy of every shard in the index must reside on the same node.
- The index must have a green health status.
To make shard allocation easier, we recommend you also remove the index's replica shards. You can later re-add replica shards as part of the shrink operation.
The requested number of primary shards in the target index must be a factor of the number of shards in the source index. For example an index with 8 primary shards can be shrunk into 4, 2 or 1 primary shards or an index with 15 primary shards can be shrunk into 5, 3 or 1. If the number of shards in the index is a prime number it can only be shrunk into a single primary shard Before shrinking, a (primary or replica) copy of every shard in the index must be present on the same node.
IMPORTANT: If the source index already has one primary shard, configuring the shrink operation with 'index.number_of_shards: 1' will cause the request to fail. An index with one primary shard cannot be shrunk further.
The current write index on a data stream cannot be shrunk. In order to shrink the current write index, the data stream must first be rolled over so that a new write index is created and then the previous write index can be shrunk.
A shrink operation:
- Creates a new target index with the same definition as the source index, but with a smaller number of primary shards.
- Hard-links segments from the source index into the target index. If the file system does not support hard-linking, then all segments are copied into the new index, which is a much more time consuming process. Also if using multiple data paths, shards on different data paths require a full copy of segment files if they are not on the same disk since hardlinks do not work across disks.
- Recovers the target index as though it were a closed index which had just been re-opened. Recovers shards to the
.routing.allocation.initial_recovery._idindex setting.
IMPORTANT: Indices can only be shrunk if they satisfy the following requirements:
- The target index must not exist.
- The source index must have more primary shards than the target index.
- The number of primary shards in the target index must be a factor of the number of primary shards in the source index. The source index must have more primary shards than the target index.
- The index must not contain more than 2,147,483,519 documents in total across all shards that will be shrunk into a single shard on the target index as this is the maximum number of docs that can fit into a single shard.
- The node handling the shrink process must have sufficient free disk space to accommodate a second copy of the existing index.
Required authorization
- Index privileges:
manage
Query parameters
-
Period to wait for a connection to the master node. If no response is received before the timeout expires, the request fails and returns an error.
Values are
-1or0.External documentation -
Period to wait for a response. If no response is received before the timeout expires, the request fails and returns an error.
Values are
-1or0.External documentation -
The number of shard copies that must be active before proceeding with the operation. Set to
allor any positive integer up to the total number of shards in the index (number_of_replicas+1).Values are
allorindex-setting.
curl \
--request PUT 'http://api.example.com/{index}/_shrink/{target}' \
--header "Content-Type: application/json" \
--data '"{\n \"settings\": {\n \"index.routing.allocation.require._name\": null,\n \"index.blocks.write\": null\n }\n}"'{
"settings": {
"index.routing.allocation.require._name": null,
"index.blocks.write": null
}
}