Skip to content

ESQL: Get and list APIs should use async execution id format #127187

@nik9000

Description

@nik9000

Description

The get and list APIs for ESQL should use the same async execution id format that the async API uses. You should be able to do a _query/async which returns an id and use that id immediately in the get running query API.

The API also shouldn't return node, coordinating_node, or data_node or anything. That's a concept that's not universally applicable. So:

{ "queries" : { "abc:1234": { <---- use the encoding returned by async search "node": "abc", <---- zap this "id": 1234, <---- zap this "start_time_millis": 14585858875292, "running_time_nanos": 762794, "query": "FROM logs* | STATS BY hostname" }, "abd:1729": { <---- use the encoding returned by async search "node": "abd", <---- zap this "id":1729, <---- zap this "start_time_millis": 14585858823573, "running_time_nanos": 90231, "query": "FROM orders | LOOKUP country_code ON country" } } } 
GET /_query/queries/abc:1234 <---- use the encoding returned by async search { "id": 1234, <---- zap this "node": "abc", <---- zap this "start_time_millis": 14585858875292, "running_time_nanos": 762794, "query": "FROM logs* | STATS BY hostname" "coordinating_node": "oTUltX4IQMOUUVeiohTt8A" <---- zap this "data_nodes" : [ "DwrYwfytxthse49X4", "i5msnbUyWlpe86e7"] <---- zap this } 

Metadata

Metadata

Assignees

Type

No type

Projects

No projects

Milestone

No milestone

Relationships

None yet

Development

No branches or pull requests

Issue actions