Github user tillrohrmann commented on a diff in the pull request:
https://github.com/apache/flink/pull/1790#discussion_r56159993
--- Diff:
flink-runtime/src/main/scala/org/apache/flink/runtime/taskmanager/TaskManager.scala
---
@@ -781,6 +787,32 @@ class TaskManager(
}
}
+ private def handleRequestTaskManagerLog
+ (sender: ActorRef, requestType: LogTypeRequest, jobManager: ActorRef):
+ Unit = {
+ if (blobService.isDefined && jobManager.path.address.host.isDefined) {
+ //create new Thread to upload log, to not block the TM for too long
+ new Thread() {
+ override def run(): Unit = {
+ val jmHost = jobManager.path.address.host.get
+ val port = blobService.get.getPort
+ val logFilePath = System.getProperty("log.file");
+ val file: File = requestType match {
+ case LogFileRequest => new File(logFilePath);
+ case StdOutFileRequest =>
+ new File(logFilePath.substring(0, logFilePath.length - 4) +
".out");
+ }
+
+ val client: BlobClient = new BlobClient(new
InetSocketAddress(jmHost, port))
+
+ sender ! client.put(new FileInputStream(file))
+ }
+ }.run()
--- End diff --
This won't start a new thread but simply call the just overwritten `run`
method. Use `start` instead. Furthermore, it is not encouraged to extend the
Thread class. It is better to implement a `Runnable` or even better avoiding to
deal with raw threads at all. I guess it would be easier to simply spawn a
`Future` which is executed by a respective `ExecutionContext`.
---
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.
---
