YongGang commented on code in PR #14909:
URL: https://github.com/apache/druid/pull/14909#discussion_r1326675056
##########
extensions-contrib/kubernetes-overlord-extensions/src/main/java/org/apache/druid/k8s/overlord/KubernetesTaskRunner.java:
##########
@@ -309,23 +309,23 @@ public Optional<InputStream> streamTaskReports(String
taskid) throws IOException
@Override
public List<Pair<Task, ListenableFuture<TaskStatus>>> restore()
{
- List<Pair<Task, ListenableFuture<TaskStatus>>> restoredTasks = new
ArrayList<>();
+ return ImmutableList.of();
+ }
+
+ @Override
+ @LifecycleStart
+ public void start()
+ {
for (Job job : client.getPeonJobs()) {
try {
- Task task = adapter.toTask(job);
- restoredTasks.add(Pair.of(task, joinAsync(task)));
+ joinAsync(adapter.toTask(job));
Review Comment:
As George mentioned, this change has reduced the frequency of task failures
during rollovers. To comprehensively address the issue, we might consider
persisting the status of `SeekableStreamIndexTaskRunner` in the database. This
would allow for accurate restoration upon startup, so don't need to rely on
TaskRunner for the latest task statuses when start. However, this enhancement
will be tackled in a subsequent PR.
(we've also observed similar symptoms with Middle Manager streaming
ingestion)
--
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.
To unsubscribe, e-mail: [email protected]
For queries about this service, please contact Infrastructure at:
[email protected]
---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]
