SQL CLI

Serverless Stack

Elasticsearch ships with a script to run the SQL CLI in its bin directory:

 $ ./bin/elasticsearch-sql-cli

You can pass the URL of the Elasticsearch instance to connect to as the first parameter:

 $ ./bin/elasticsearch-sql-cli https://some.server:9200

If security is enabled on your cluster, you can pass the username and password in the form username:password@host_name:port to the SQL CLI:

 $ ./bin/elasticsearch-sql-cli https://sql_user:strongpassword@some.server:9200

Once the CLI is running you can use any query that Elasticsearch supports:

  sql> SELECT * FROM library WHERE page_count > 500 ORDER BY page_count DESC; author | name | page_count | release_date -----------------+--------------------+---------------+--------------- Peter F. Hamilton|Pandora's Star |768 |1078185600000 Vernor Vinge |A Fire Upon the Deep|613 |707356800000 Frank Herbert |Dune |604 |-144720000000 Alastair Reynolds|Revelation Space |585 |953078400000 James S.A. Corey |Leviathan Wakes |561 |1306972800000  

The jar containing the SQL CLI is a stand alone Java application and the scripts just launch it. You can move it around to other machines without having to install Elasticsearch on them. Without the already provided script files, you can use a command similar to the following to start the SQL CLI:

  $ ./java -jar [PATH_TO_CLI_JAR]/elasticsearch-sql-cli-[VERSION].jar https://some.server:9200  

  $ ./java -cp [PATH_TO_CLI_JAR]/elasticsearch-sql-cli-[VERSION].jar org.elasticsearch.xpack.sql.cli.Cli https://some.server:9200  

The jar name will be different for each Elasticsearch version (for example elasticsearch-sql-cli-7.3.2.jar), thus the generic VERSION specified in the example above. Furthermore, if not running the command from the folder where the SQL CLI jar resides, you’d have to provide the full path, as well.

CLI commands

Apart from SQL queries, CLI can also execute some specific commands:

allow_partial_search_results = <boolean> (default false): If true, returns partial results if there are shard request timeouts or shard failures. If false, returns an error with no partial results.

  sql> allow_partial_search_results = true; allow_partial_search_results set to true  

fetch_size = <number> (default 1000): Allows to change the size of fetches for query execution. Each fetch is delimited by fetch separator (if explicitly set).

 sql> fetch_size = 2000; fetch size set to 2000

fetch_separator = <string> (empty string by default): Allows to change the separator string between fetches.

  sql> fetch_separator = "---------------------"; fetch separator set to "---------------------"  

lenient = <boolean> (default false): If false, Elasticsearch SQL returns an error for fields containing array values. If true, Elasticsearch SQL returns the first value from the array with no guarantee of consistent results.

 sql> lenient = true; lenient set to true

info: Returns server information.

 sql> info; Node:mynode Cluster:elasticsearch Version:8.3

exit: Closes the CLI.

 sql> exit; Bye!

cls: Clears the screen.

 sql> cls;

logo: Prints Elastic logo.

  sql> logo; asticElasticE ElasticE sticEla sticEl ticEl Elast lasti Elasti tic cEl ast icE icE as cEl icE as cEl icEla las El sticElasticElast icElas las last ticElast El asti asti stic El asticEla Elas icE El Elas cElasticE ticEl cE Ela ticEl ticElasti cE las astic last icE sticElas asti stic icEl sticElasticElast icE sticE ticEla icE sti cEla icEl sti Ela cEl sti cEl Ela astic ticE asti ElasticElasti ticElasti lasticElas ElasticElast SQL 8.3.0  