Documenting master_is_stable health API settings (#87901)

docs/reference/settings/health-diagnostic-settings.asciidoc

@@ -0,0 +1,29 @@
+[[health-diagnostic-settings]]
+=== Health diagnostic settings in {es}
+[subs="attributes"]
+++++
+<titleabbrev>Health Diagnostic settings</titleabbrev>
+++++
+
+The following are the _expert-level_ settings available for configuring an internal diagnostics service.
+The output of this service is currently exposed through the Health API <<health-api, Health API>>. It
+is not recommended to change any of these from their default values.
+
+==== Cluster level settings
+
+`health.master_history.has_master_lookup_timeframe`::
+(<<static-cluster-setting,Static>>) The amount of time a node looks back to see if it has observed
+a master at all, before moving on with other checks. Defaults to `30s` (30 seconds).
+
+`master_history.max_age`::
+(<<static-cluster-setting,Static>>) The timeframe we record the master history 
+to be used for diagnosing the cluster health. Master node changes older than this time will not be considered when
+diagnosing the cluster health. Defaults to `30m` (30 minutes).
+
+`health.master_history.identity_changes_threshold`::
+(<<static-cluster-setting,Static>>) The number of master identity changes witnessed by a node that indicates the cluster is not healthy.
+Defaults to `4`.
+
+`health.master_history.no_master_transitions_threshold`::
+(<<static-cluster-setting,Static>>) The number of transitions to no master witnessed by a node that indicates the cluster is not healthy.
+Defaults to `4`.

docs/reference/setup.asciidoc

@@ -64,6 +64,8 @@ include::modules/discovery/discovery-settings.asciidoc[]
 
 include::modules/indices/fielddata.asciidoc[]
 
+include::settings/health-diagnostic-settings.asciidoc[]
+
 include::settings/ilm-settings.asciidoc[]
 
 include::modules/indices/index_management.asciidoc[]