[
https://issues.apache.org/jira/browse/GOBBLIN-1099?focusedWorklogId=411512&page=com.atlassian.jira.plugin.system.issuetabpanels:worklog-tabpanel#worklog-411512
]
ASF GitHub Bot logged work on GOBBLIN-1099:
-------------------------------------------
Author: ASF GitHub Bot
Created on: 27/Mar/20 23:41
Start Date: 27/Mar/20 23:41
Worklog Time Spent: 10m
Work Description: autumnust commented on pull request #2940:
GOBBLIN-1099: Handle orphaned Yarn containers in Gobblin-on-Yarn clus…
URL: https://github.com/apache/incubator-gobblin/pull/2940#discussion_r399578539
##########
File path:
gobblin-cluster/src/main/java/org/apache/gobblin/cluster/GobblinTaskRunner.java
##########
@@ -366,20 +377,74 @@ boolean isStopped() {
}
@VisibleForTesting
- void connectHelixManager() {
- try {
- this.jobHelixManager.connect();
- this.jobHelixManager.getMessagingService()
-
.registerMessageHandlerFactory(GobblinHelixConstants.SHUTDOWN_MESSAGE_TYPE,
- new ParticipantShutdownMessageHandlerFactory());
- this.jobHelixManager.getMessagingService()
-
.registerMessageHandlerFactory(Message.MessageType.USER_DEFINE_MSG.toString(),
- getUserDefinedMessageHandlerFactory());
- if (this.taskDriverHelixManager.isPresent()) {
- this.taskDriverHelixManager.get().connect();
+ void connectHelixManager() throws Exception {
+ this.jobHelixManager.connect();
+ //Ensure the instance is enabled.
+ this.jobHelixManager.getClusterManagmentTool().enableInstance(clusterName,
helixInstanceName, true);
+ this.jobHelixManager.getMessagingService()
+
.registerMessageHandlerFactory(GobblinHelixConstants.SHUTDOWN_MESSAGE_TYPE,
+ new ParticipantShutdownMessageHandlerFactory());
+ this.jobHelixManager.getMessagingService()
+
.registerMessageHandlerFactory(Message.MessageType.USER_DEFINE_MSG.toString(),
+ getUserDefinedMessageHandlerFactory());
+ if (this.taskDriverHelixManager.isPresent()) {
+ this.taskDriverHelixManager.get().connect();
+ //Ensure the instance is enabled.
+
this.taskDriverHelixManager.get().getClusterManagmentTool().enableInstance(this.taskDriverHelixManager.get().getClusterName(),
helixInstanceName, true);
+ }
+ }
+
+ /**
+ * A method to handle failures joining Helix cluster. The method will
perform the following steps before attempting
+ * to re-join the cluster:
+ * <li>
+ * <ul>Disconnect from Helix cluster, which would close any open
clients</ul>
+ * <ul>Drop instance from Helix cluster, to remove any partial instance
structure from Helix</ul>
+ * <ul>Re-construct helix manager instances, used to re-join the
cluster</ul>
+ * </li>
+ */
+ private void onClusterJoinFailure() {
+ logger.warn("Disconnecting Helix manager..");
+ disconnectHelixManager();
+
+ HelixAdmin admin = new
ZKHelixAdmin(clusterConfig.getString(GobblinClusterConfigurationKeys.ZK_CONNECTION_STRING_KEY));
Review comment:
I couldn't fully get the purpose of disconnect from existing helixManager,
creating a HelixAdmin and use the newly-created admin to conduct
instance-dropping. Isn't `helixManager.getClusterManagmentTool()` essentially a
same HelixAdmin that we can use to drop instance ? What's the gain of
disconnecting helixManager?
----------------------------------------------------------------
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.
For queries about this service, please contact Infrastructure at:
us...@infra.apache.org
Issue Time Tracking
-------------------
Worklog Id: (was: 411512)
Time Spent: 1h (was: 50m)
> Handle orphaned Yarn containers in Gobblin-on-Yarn clusters
> -----------------------------------------------------------
>
> Key: GOBBLIN-1099
> URL: https://issues.apache.org/jira/browse/GOBBLIN-1099
> Project: Apache Gobblin
> Issue Type: Improvement
> Components: gobblin-yarn
> Affects Versions: 0.15.0
> Reporter: Sudarshan Vasudevan
> Assignee: Abhishek Tiwari
> Priority: Major
> Fix For: 0.15.0
>
> Time Spent: 1h
> Remaining Estimate: 0h
>
> A Yarn application may leave behind orphaned containers, which can happen due
> to lost node managers. The orphaned containers however can continue to run
> (potentially forever) as participants in the Helix cluster. This can cause
> the following problems for a Gobblin-on-Yarn application:
> # Double publish of data and commit of state
> # Task failures and partition starvation during application restarts, as
> Helix may assign tasks to the orphaned containers which have a stale state
> and configuration
> # Container failures on application restarts due to Helix instance name
> collisions with orphaned containers
>
--
This message was sent by Atlassian Jira
(v8.3.4#803005)
