Github user pwendell commented on a diff in the pull request:
https://github.com/apache/spark/pull/42#discussion_r10597200
--- Diff: core/src/main/scala/org/apache/spark/ui/SparkUI.scala ---
@@ -68,19 +105,53 @@ private[spark] class SparkUI(sc: SparkContext) extends
Logging {
/** Initialize all components of the server */
def start() {
- // NOTE: This is decoupled from bind() because of the following
dependency cycle:
- // DAGScheduler() requires that the port of this server is known
- // This server must register all handlers, including JobProgressUI,
before binding
- // JobProgressUI registers a listener with SparkContext, which
requires sc to initialize
+ storage.start()
jobs.start()
+ env.start()
exec.start()
+
+ // Listen for events from the SparkContext if it exists, otherwise
from persisted storage
+ val eventBus = if (live) {
--- End diff --
I think it would be a bit nicer to have the `SparkUI` take an `EventBus` as
a constructor argument rather than reasoning about it inside of the UI. This
way you decoupled the UI functionality from whether or not the bus is running
live (it seems like there are some places where you still have to do that, but
as much as possible it might be nice to separate this)
---
If your project is set up for it, you can reply to this email and have your
reply appear on GitHub as well. If your project does not have this feature
enabled and wishes so, or if the feature is enabled but not working, please
contact infrastructure at infrastruct...@apache.org or file a JIRA ticket
with INFRA.
---
