Skip to content

Commit

Permalink
[DOCS] SQL: Document async SQL search (#74975) (#75037)
Browse files Browse the repository at this point in the history
Documents async SQL search functionality.

I plan to add formal API documentation for the async APIs with a later PR.

Relates to #73991 and #74845.
# Conflicts:
#	docs/reference/release-notes/highlights.asciidoc
  • Loading branch information
jrodewig committed Jul 7, 2021
1 parent 04308b5 commit 0f5a0e7
Showing 1 changed file with 191 additions and 0 deletions.
191 changes: 191 additions & 0 deletions docs/reference/sql/endpoints/rest.asciidoc
Expand Up @@ -11,6 +11,7 @@
* <<sql-rest-params>>
* <<sql-runtime-fields>>
* <<sql-rest-fields>>
* <<sql-async>>

[[sql-rest-overview]]
=== Overview
Expand Down Expand Up @@ -561,6 +562,170 @@ Frank Herbert |Dune |604 |1965-06-01T00:00:00.000Z|TUESDAY
----
// TESTRESPONSE[non_json]

[[sql-async]]
=== Run an async SQL search

By default, SQL searches are synchronous. They wait for complete results before
returning a response. However, results can take longer for searches across large
data sets or <<data-tiers,frozen data>>.

To avoid long waits, run an async SQL search. Set `wait_for_completion_timeout`
to a duration you’d like to wait for synchronous results.

[source,console]
----
POST _sql?format=json
{
"wait_for_completion_timeout": "2s",
"query": "SELECT * FROM library ORDER BY page_count DESC",
"fetch_size": 5
}
----
// TEST[setup:library]
// TEST[s/"wait_for_completion_timeout": "2s"/"wait_for_completion_timeout": "0"/]

If the search doesn’t finish within this period, the search becomes async. The
API returns:

* An `id` for the search.
* An `is_partial` value of `true`, indicating the search results are incomplete.
* An `is_running` value of `true`, indicating the search is still running in the
background.

For CSV, TSV, and TXT responses, the API returns these values in the respective
`Async-ID`, `Async-partial`, and `Async-running` HTTP headers instead.

[source,console-result]
----
{
"id": "FnR0TDhyWUVmUmVtWXRWZER4MXZiNFEad2F5UDk2ZVdTVHV1S0xDUy00SklUdzozMTU=",
"is_partial": true,
"is_running": true,
"rows": [ ]
}
----
// TESTRESPONSE[s/FnR0TDhyWUVmUmVtWXRWZER4MXZiNFEad2F5UDk2ZVdTVHV1S0xDUy00SklUdzozMTU=/$body.id/]
// TESTRESPONSE[s/"is_partial": true/"is_partial": $body.is_partial/]
// TESTRESPONSE[s/"is_running": true/"is_running": $body.is_running/]

To check the progress of an async search, use the search ID with the get async
SQL search status API.

[source,console]
----
GET _sql/async/status/FnR0TDhyWUVmUmVtWXRWZER4MXZiNFEad2F5UDk2ZVdTVHV1S0xDUy00SklUdzozMTU=
----
// TEST[skip: no access to search ID]

If `is_running` and `is_partial` are `false`, the async search has finished with
complete results.

[source,console-result]
----
{
"id": "FnR0TDhyWUVmUmVtWXRWZER4MXZiNFEad2F5UDk2ZVdTVHV1S0xDUy00SklUdzozMTU=",
"is_running": false,
"is_partial": false,
"expiration_time_in_millis": 1611690295000,
"completion_status": 200
}
----
// TESTRESPONSE[s/FnR0TDhyWUVmUmVtWXRWZER4MXZiNFEad2F5UDk2ZVdTVHV1S0xDUy00SklUdzozMTU=/$body.id/]
// TESTRESPONSE[s/"expiration_time_in_millis": 1611690295000/"expiration_time_in_millis": $body.expiration_time_in_millis/]

To get the results, use the search ID with the get async SQL search API. If the
search is still running, specify how long you’d like to wait using
`wait_for_completion_timeout`. You can also specify the response `format`.

[source,console]
----
GET _sql/async/FnR0TDhyWUVmUmVtWXRWZER4MXZiNFEad2F5UDk2ZVdTVHV1S0xDUy00SklUdzozMTU=?wait_for_completion_timeout=2s&format=json
----
// TEST[skip: no access to search ID]

[discrete]
[[sql-async-retention]]
==== Change the search retention period

By default, {es} stores async SQL searches for five days. After this period,
{es} deletes the search and its results, even if the search is still running. To
change this retention period, use the `keep_alive` parameter.

[source,console]
----
POST _sql?format=json
{
"keep_alive": "2d",
"wait_for_completion_timeout": "2s",
"query": "SELECT * FROM library ORDER BY page_count DESC",
"fetch_size": 5
}
----
// TEST[setup:library]

You can use the get async SQL search API's `keep_alive` parameter to later
change the retention period. The new period starts after the request runs.

[source,console]
----
GET _sql/async/FmdMX2pIang3UWhLRU5QS0lqdlppYncaMUpYQ05oSkpTc3kwZ21EdC1tbFJXQToxOTI=?keep_alive=5d&wait_for_completion_timeout=2s&format=json
----
// TEST[skip: no access to search ID]

Use the delete async SQL search API to delete an async search before the
`keep_alive` period ends. If the search is still running, {es} cancels it.

[source,console]
----
DELETE _sql/async/delete/FmdMX2pIang3UWhLRU5QS0lqdlppYncaMUpYQ05oSkpTc3kwZ21EdC1tbFJXQToxOTI=
----
// TEST[skip: no access to search ID]

[discrete]
[[sql-store-searches]]
==== Store synchronous SQL searches

By default, {es} only stores async SQL searches. To save a synchronous search,
specify `wait_for_completion_timeout` and set `keep_on_completion` to `true`.

[source,console]
----
POST _sql?format=json
{
"keep_on_completion": true,
"wait_for_completion_timeout": "2s",
"query": "SELECT * FROM library ORDER BY page_count DESC",
"fetch_size": 5
}
----
// TEST[setup:library]

If `is_partial` and `is_running` are `false`, the search was synchronous and
returned complete results.

[source,console-result]
----
{
"id": "Fnc5UllQdUVWU0NxRFNMbWxNYXplaFEaMUpYQ05oSkpTc3kwZ21EdC1tbFJXQTo0NzA=",
"is_partial": false,
"is_running": false,
"rows": ...,
"columns": ...,
"cursor": ...
}
----
// TESTRESPONSE[s/Fnc5UllQdUVWU0NxRFNMbWxNYXplaFEaMUpYQ05oSkpTc3kwZ21EdC1tbFJXQTo0NzA=/$body.id/]
// TESTRESPONSE[s/"rows": \.\.\./"rows": $body.rows/]
// TESTRESPONSE[s/"columns": \.\.\./"columns": $body.columns/]
// TESTRESPONSE[s/"cursor": \.\.\./"cursor": $body.cursor/]

You can get the same results later using the search ID with the get async SQL
search API.

Saved synchronous searches are still subject to the `keep_alive` retention
period. When this period ends, {es} deletes the search results. You can also
delete saved searches using the delete async SQL search API.

[[sql-rest-fields]]
=== Supported REST parameters

Expand Down Expand Up @@ -623,6 +788,32 @@ More information available https://docs.oracle.com/javase/8/docs/api/java/time/Z
|Defines one or more <<runtime-search-request,runtime fields>> in the search
request. These fields take precedence over mapped fields with the same name.

|keep_alive
|`5d`
a|Period {es} stores the search and its results. Must be greater than `1m` (one
minute).

When this period end, {es} deletes the search and its results, even if the
search is still running.

If `keep_on_completion` is `false`, {es} only stores searches that don't finish
within the `wait_for_completion_timeout`, regardless of this value.

|keep_on_completion
|`false`
|If `true`, {es} stores the search and its results. If `false`, {es} stores the
search and its results only if it doesn't finish before the
`wait_for_completion_timeout`.

|wait_for_completion_timeout
|None
a|Duration to wait for the search to finish. Defaults to no timeout, meaning the
request waits for complete results.

If you specify this parameter and the search finishes during the timeout, the
request returns complete results. If the search doesn't finish during this
period, the search becomes async. See <<sql-async>>.

|===

Do note that most parameters (outside the timeout and `columnar` ones) make sense only during the initial query - any follow-up pagination request only requires the `cursor` parameter as explained in the <<sql-pagination, pagination>> chapter.
Expand Down

0 comments on commit 0f5a0e7

Please sign in to comment.