[ES|QL] When requesting GET /_query/async/{id}?keep_alive={time} while the query is running keep_alive is ignored

[kertal]

Elasticsearch Version

main

Installed Plugins

No response

Java Version

bundled

OS Version

MacOs 15.6.1

Problem Description

When requesting async ESQL query results using

GET /_query/async/{id}?keep_alive={time}

While there query is still running, the new value of keep_alive doesn't prevent the query to 404 after the initial keep_alive time that was set

Steps to Reproduce

Start a query with a given DELAY to make sure it's running while you use the GET API

POST /_query/async?drop_null_columns
{
   "wait_for_completion_timeout":"200ms",
   "keep_on_completion":true,
   "keep_alive":"10000ms",
   "query":"FROM kibana_sample_data_ecommerce | EVAL delay = DELAY(20000ms) | LIMIT 1",
   "locale":"en",
   "include_ccs_metadata":true
}

Use the id of the query to increase the intended keep_alive value

GET /_query/async/{id}?keep_alive=1500000ms

Then use the following request to get the results of a query, request it after roughly 30s or a little bit more, after the query returned results and the keep_alive period is over

GET /_query/async/{id}

then the request will return a 404, because the resource with the actual id as been cleaned up.

Expected result should be that the keep_alive can be extended also in the running state, like it works in DSL

Logs (if relevant)

No response

[kertal]

very similar issue was resolved in #130619 ... note that it would be very helpful if the GET endpoint would return the actual keep alive value, or there would be another way to gain insights about async queries currently running ,and their keep alive stats (and more). But this is a feature request

[elasticsearchmachine]

Pinging @elastic/es-analytical-engine (Team:Analytics)

[dnhatn]

@kertal I opened a fix in #135209