Github user tgravescs commented on a diff in the pull request:
https://github.com/apache/spark/pull/10794#discussion_r57732500
--- Diff:
yarn/src/main/scala/org/apache/spark/deploy/yarn/YarnAllocator.scala ---
@@ -81,8 +82,20 @@ private[yarn] class YarnAllocator(
new ConcurrentHashMap[ContainerId, java.lang.Boolean])
@volatile private var numExecutorsRunning = 0
- // Used to generate a unique ID per executor
- private var executorIdCounter = 0
+
+ /**
+ * Used to generate a unique ID per executor
+ *
+ * Init `executorIdCounter`. when AM restart, `executorIdCounter` will
reset to 0. Then
+ * the id of new executor will start from 1, this will conflict with the
executor has
+ * already created before. So, we should initialize the
`executorIdCounter` by getting
+ * the max executorId from driver.
+ *
+ * @see SPARK-12864
+ */
+ private var executorIdCounter: Int = {
--- End diff --
So I think there is a race condition here. For instance, AM dies right
after it sends launch command for a new executor, if the AM comes up and get
the currentExecutorId right before the executor registers, this number is going
to be off by one (or possibly more then one if it sent launch commands for many)
---
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]
