Github user kayousterhout commented on a diff in the pull request:
https://github.com/apache/spark/pull/1309#discussion_r14695912
--- Diff:
core/src/main/scala/org/apache/spark/ui/jobs/JobProgressListener.scala ---
@@ -147,6 +155,11 @@ class JobProgressListener(conf: SparkConf) extends
SparkListener {
val info = taskEnd.taskInfo
if (info != null) {
+ val accumulables = stageIdToAccumulables.getOrElseUpdate(sid,
HashMap[String, String]())
+ for ((name, value) <- info.accumulableValues) {
+ accumulables(name) = value
--- End diff --
Is this right? Isn't this setting the value for the whole stage to be the
partial update for the task (which could explain the discrepancy in your
screenshot)?
---
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 [email protected] or file a JIRA ticket
with INFRA.
---
