Index API- Index Management for Elasticsearch

In Elasticsearch, an index can be thought of as an optimized collection of documents, and each document is a collection of fields, which are key-value pairs containing your data.

That is: index → document → field → data.

An Elasticsearch index is just a logical group of one or more physical shards, each of which is actually a separate index. By distributing documents in the index over multiple shards and those shards over multiple nodes.

That is: node → sharding → document (index).

Shards are divided into master shards and copies, with each document belonging to a master shard, and then a deputy shard is a copy of the master shard.

Core concept analogy:

Starting from 6.0.0, a single index can have only one type. Starting from 7.0.0, it is not recommended. Not supported at all after 8.0.0.

The index set

Index Settings are divided into:

• Static index Settings: set when the index is created, or can be modified after the index is closed.
• Dynamic index Settings: This can be changed directly using the update index API.

Related parameters are shown in the following table:

With so many Settings, it is not normally configured. Number_of_shards (the number of primary shards) and number_of_replicas (the number of replicas of the primary shards) are commonly used.

Create a sample

1. Create an index and configure number_of_SHards and number_of_replicas. The master shard is set to 3, and the master shard’s copy is set to 2. The index name is displayed in the path and the following JSON is the request body content. The index name must comply with the following rules:

The index name must comply with the following conventions:

• The value can only be lowercase characters

• Cannot contain the characters \, /, *,? , “, <, >, |, (space), and #

• Indexes can contain colons (:) prior to 7.0, but not recommended after.

• Cannot start with -, _, or +

• Cannot be. Or..

• The length cannot exceed 255 bytes (note bytes, so multi-byte characters will reach the 255 limit faster)

• Names starting with. Are not recommended, except for hidden and internal indexes managed by plug-ins

PUT /my-index-000001

{

 "settings": {

  "index": {

   "number_of_shards": 3."number_of_replicas": 2}}}Copy the code

Further, instead of displaying the specified index section:

PUT /my-index-000001
{
  "settings": {
    "number_of_shards": 3."number_of_replicas": 2}}Copy the code

2. Plus the mapping configuration

PUT /test
{
  "settings": {
    "number_of_shards": 1}."mappings": {
    "properties": {
      "field1": { "type": "text"}}}}Copy the code

3. Configuring index Aliases

{
  "aliases": {
    "alias_1": {},
    "alias_2": {
      "filter": {
        "term": { "user.id": "kimchy"}},"routing": "shard-1"}}}Copy the code

Sample response:

{
  "acknowledged": true."shards_acknowledged": true."index": "test"
}
Copy the code

Trees indicates whether the index was successfully created in the cluster, and shards_trees indicates whether the required number of shard copies was started for each shard before timeout.

It is important to note that both values can be false, but the index can also be created successfully, indicating only the state of the operation before the timeout.

Index Management API

Further, the API related to index management is as follows:

Based on the data in the above table, we can further summarize as follows.

1. The index ofCRUDOperating as follows:

• Add: Creates an indexPUT /{index}
• Delete: Deletes an indexDELETE /{index}
• Search: get the indexGET /{index}
• Modified: there is no API to modify index directly, there is API to modify index attributes, and need to see whether the attributes are static or dynamic attributes, etc.

2. Index existence operation:

• The index isHEAD /{index}

Elasticsearch responds 200 if it has the index, 404 if it doesn’t.

3. Index on and off operations:

• Shut downPOST /{index}/_close
• openPOST /{index}/_open

After an index is closed, it cannot be read or written. Switch status can be found at GET /_cat/indices/ Customer? V, the value is open or close

4. Index freezing and unfreezing operations:

• freezePOST /{index}/_freeze
• thawPOST /{index}/_unfreeze

After the index is frozen, it cannot write. The frozen state can be viewed in GET /{index}, in {index}-> Settings ->index->frozen, with a value of false or true.

5. Index shrink and expand operations:

• shrinkagePOST /{index}/_shrink/{target_index}
• Break upPOST /{index}/_split/{target_index}

Generate a new index and reduce or increase the number of shards for the index.

6. Other operations:

• copyPOST /{index}/_clone/{target_index}

Copy to generate a new index.

• flipPOST / {rollover - target} / _rollover or / {rollover - target} / _rollover / {target - index}

When the specified condition is met, the write data of the index alias points to the new index to achieve the effect similar to data archiving.

• parsingGET /_resolve/index/{name}

Gets the name, alias, and data flow of the related index with the specified name.

From the above summary, we can see that the API for index management can only operate directly on the index itself, and does not involve much more detailed things. Once again, Elasticsearch’s action keyword starts with an underscore _, such as _clone.

In the next article we will learn about index mapping management.

Welcome to my blog: The Border city of A-woo

Welcome to pay attention to my public number: A woo programming