Github user tdas commented on a diff in the pull request:
https://github.com/apache/spark/pull/2065#discussion_r16683031
--- Diff:
external/flume-sink/src/main/scala/org/apache/spark/streaming/flume/sink/SparkAvroCallbackHandler.scala
---
@@ -124,6 +139,8 @@ private[flume] class SparkAvroCallbackHandler(val
threads: Int, val channel: Cha
*/
def shutdown() {
logInfo("Shutting down Spark Avro Callback Handler")
+ stopped = true
+ processorsToShutdown.foreach(_.shutdown())
--- End diff --
There maybe a race condition here. A thread that might have called
`getEventBatch()` and created a processor, before the shutting down thread
(i.e., the one that called `shutdown()`) marks `stopped` as true. But that
thread may not have added the processor to `processorsToShutdown` yet, and the
shutting down thread may have already completed the shutdown function. In that
case, there is a zombie processor that is never shutdown.
---
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.
---
---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]
