[ 
https://issues.apache.org/jira/browse/GOBBLIN-1988?focusedWorklogId=900269&page=com.atlassian.jira.plugin.system.issuetabpanels:worklog-tabpanel#worklog-900269
 ]

ASF GitHub Bot logged work on GOBBLIN-1988:
-------------------------------------------

                Author: ASF GitHub Bot
            Created on: 18/Jan/24 01:37
            Start Date: 18/Jan/24 01:37
    Worklog Time Spent: 10m 
      Work Description: AndyJiang99 commented on code in PR #3862:
URL: https://github.com/apache/gobblin/pull/3862#discussion_r1456712626


##########
gobblin-utility/src/main/java/org/apache/gobblin/util/LoggingUncaughtExceptionHandler.java:
##########
@@ -42,6 +45,7 @@ public LoggingUncaughtExceptionHandler(Optional<Logger> 
logger) {
 
   @Override
   public void uncaughtException(Thread t, Throwable e) {
+    exception = e;

Review Comment:
   The exception variable is set but never used, maybe missing something?



##########
gobblin-runtime/src/main/java/org/apache/gobblin/runtime/StreamModelTaskRunner.java:
##########
@@ -155,13 +156,17 @@ protected void run() throws Exception {
     Thread thread = new Thread(() -> connectableStream.connect());
     thread.setName(this.getClass().getSimpleName());
     //Log uncaught exceptions (e.g.OOMEs) to prevent threads from dying 
silently
-    thread.setUncaughtExceptionHandler(new 
LoggingUncaughtExceptionHandler(Optional.absent()));
+    LoggingUncaughtExceptionHandler exceptionHandler = new 
LoggingUncaughtExceptionHandler(Optional.absent());
+    thread.setUncaughtExceptionHandler(exceptionHandler);
     thread.start();
-
-    if (!ExponentialBackoff.awaitCondition().callable(() -> 
this.forks.keySet().stream().map(Optional::get).allMatch(Fork::isDone)).
+    if (!ExponentialBackoff.awaitCondition().callable(() ->
+        this.forks.keySet().stream().map(Optional::get).allMatch(Fork::isDone) 
|| exceptionHandler.getException() != null).
         
initialDelay(initialDelay).maxDelay(initialDelay).maxWait(TimeUnit.MINUTES.toMillis(maxWaitInMinute)).await())
 {
       throw new TimeoutException("Forks did not finish withing specified 
timeout.");
     }
+    if (exceptionHandler.getException() != null) {
+      throw Throwables.propagate(exceptionHandler.getException());
+    }

Review Comment:
   You mentioned a mock to test this functionality to fail the container and 
retry the job again, but I don't see the rests associated with that





Issue Time Tracking
-------------------

    Worklog Id:     (was: 900269)
    Time Spent: 0.5h  (was: 20m)

> Fail the streaming container when OOM issue happens
> ---------------------------------------------------
>
>                 Key: GOBBLIN-1988
>                 URL: https://issues.apache.org/jira/browse/GOBBLIN-1988
>             Project: Apache Gobblin
>          Issue Type: Bug
>            Reporter: Zihan Li
>            Priority: Major
>          Time Spent: 0.5h
>  Remaining Estimate: 0h
>
> Now when the OOM error happens, we just log out the error and keep the 
> container running there without doing anything. In this case we should fail 
> the container and retry the task. 



--
This message was sent by Atlassian Jira
(v8.20.10#820010)

Reply via email to