[[update-desired-nodes]] === Create or update desired nodes API ++++ Create or update desired nodes ++++ NOTE: {cloud-only} ..New API reference [sidebar] -- For the most up-to-date API details, refer to {api-es}/group/endpoint-cluster[Cluster APIs]. -- Creates or updates the desired nodes. [[update-desired-nodes-request]] ==== {api-request-title} [source,console] -------------------------------------------------- PUT /_internal/desired_nodes// { "nodes" : [ { "settings" : { "node.name" : "instance-000187", "node.external_id": "instance-000187", "node.roles" : ["data_hot", "master"], "node.attr.data" : "hot", "node.attr.logical_availability_zone" : "zone-0" }, "processors" : 8.0, "memory" : "58gb", "storage" : "2tb" } ] } -------------------------------------------------- // TEST[s//test/] // TEST[s//1/] ////////////////////////// [source,console] -------------------------------------------------- DELETE /_internal/desired_nodes -------------------------------------------------- // TEST[continued] ////////////////////////// [[update-desired-nodes-query-params]] ==== {api-query-parms-title} include::{es-ref-dir}/rest-api/common-parms.asciidoc[tag=master-timeout] `dry_run`:: (Optional, Boolean) If `true`, then the request simulates the update and returns a response with `dry_run` field set to `true`. [[update-desired-nodes-desc]] ==== {api-description-title} This API creates or update the desired nodes. External orchestrators can use this API to let Elasticsearch know about the cluster topology, including future changes such as adding or removing nodes. Using this information, the system is able to take better decisions. It's possible to run the update in "dry run" mode by adding the `?dry_run` query parameter. This will validate the request result, but will not actually perform the update. [[update-desired-nodes-examples]] ==== {api-examples-title} In this example, a new version for the desired nodes with history `Ywkh3INLQcuPT49f6kcppA` is created. This API only accepts monotonically increasing versions. [source,console] -------------------------------------------------- PUT /_internal/desired_nodes/Ywkh3INLQcuPT49f6kcppA/100 { "nodes" : [ { "settings" : { "node.name" : "instance-000187", "node.external_id": "instance-000187", "node.roles" : ["data_hot", "master"], "node.attr.data" : "hot", "node.attr.logical_availability_zone" : "zone-0" }, "processors" : 8.0, "memory" : "58gb", "storage" : "2tb" } ] } -------------------------------------------------- // TEST The API returns the following result: [source,console-result] -------------------------------------------------- { "replaced_existing_history_id": false, "dry_run": false } -------------------------------------------------- Additionally, it is possible to specify a processors range. This is helpful in environments where Elasticsearch nodes can be deployed in hosts where the number of processors that the Elasticsearch process can use is guaranteed to be at least the lower range and up to the upper range. This is a common scenario in Linux deployments where cgroups is used. [source,console] -------------------------------------------------- PUT /_internal/desired_nodes/Ywkh3INLQcuPT49f6kcppA/101 { "nodes" : [ { "settings" : { "node.name" : "instance-000187", "node.external_id": "instance-000187", "node.roles" : ["data_hot", "master"], "node.attr.data" : "hot", "node.attr.logical_availability_zone" : "zone-0" }, "processors_range" : {"min": 8.0, "max": 10.0}, "memory" : "58gb", "storage" : "2tb" } ] } -------------------------------------------------- ////////////////////////// [source,console] -------------------------------------------------- DELETE /_internal/desired_nodes -------------------------------------------------- // TEST[continued] //////////////////////////