You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
Copy file name to clipboardExpand all lines: docs/modules/spark-k8s/pages/usage-guide/listenerclass.adoc
+8-2Lines changed: 8 additions & 2 deletions
Display the source diff
Display the rich diff
Original file line number
Diff line number
Diff line change
@@ -1,5 +1,7 @@
1
-
= History service exposition with listener classes
2
-
:description: Configure the Spark history service exposure with listener classes: cluster-internal, external-unstable, or external-stable.
1
+
= Service exposition with listener classes
2
+
:description: Configure the Spark connect and history services exposure with listener classes: cluster-internal, external-unstable, or external-stable.
3
+
4
+
== History services
3
5
4
6
The operator deploys a xref:listener-operator:listener.adoc[Listener] for each spark history pod.
5
7
The default is to only being accessible from within the Kubernetes cluster, but this can be changed by setting `.spec.clusterConfig.listenerClass`:
@@ -17,3 +19,7 @@ spec:
17
19
<1> Specify one of `external-stable`, `external-unstable`, `cluster-internal` (the default setting is `cluster-internal`).
18
20
19
21
For the example above, the listener operator creates a service named `spark-history-node-default-0-listener` where `spark-history` is the name of the SparkHistoryServer, `node` is the service role (the only service role available for history servers) and `default` is the role group.
22
+
23
+
== Connect services
24
+
25
+
Connect pods can be exposed using listener classes in exactly tha same fashion as history servers.
0 commit comments