[
https://issues.apache.org/jira/browse/FLINK-5482?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=15823989#comment-15823989
]
ASF GitHub Bot commented on FLINK-5482:
---------------------------------------
Github user uce commented on a diff in the pull request:
https://github.com/apache/flink/pull/3120#discussion_r96232469
--- Diff:
flink-runtime/src/main/java/org/apache/flink/runtime/query/QueryableStateClient.java
---
@@ -341,7 +341,34 @@ public void shutDown() {
return previous;
}
} else {
- return cachedFuture;
+ // do not retain futures which failed as they
will remain in
+ // the cache even if the error cause is not
present any more
+ // and a new lookup may succeed
+ boolean isFailedFuture = false;
+ if (cachedFuture.isCompleted()) {
+ // find out if the future failed
+ try {
+
cachedFuture.value().get().get();
--- End diff --
I think you can do `cachedFuture.value().get().isFailure()` instead of
catching the Exception here.
> QueryableStateClient does not recover from a failed lookup due to a
> non-running job
> -----------------------------------------------------------------------------------
>
> Key: FLINK-5482
> URL: https://issues.apache.org/jira/browse/FLINK-5482
> Project: Flink
> Issue Type: Bug
> Affects Versions: 1.2.0
> Reporter: Nico Kruber
> Assignee: Nico Kruber
>
> When the QueryableStateClient is used to issue a query but the job is not
> running yet, its internal lookup result is cached with an
> IllegalStateException that the job was not found. It does, however, never
> recover from that.
--
This message was sent by Atlassian JIRA
(v6.3.4#6332)
