zaynt4606 commented on code in PR #2989:
URL: https://github.com/apache/celeborn/pull/2989#discussion_r1886247312
##########
worker/src/main/scala/org/apache/celeborn/service/deploy/worker/PushDataHandler.scala:
##########
@@ -273,26 +273,26 @@ class PushDataHandler(val workerSource: WorkerSource)
extends BaseMessageHandler
val writePromise = Promise[Array[StatusCode]]()
// for primary, send data to replica
if (doReplicate) {
+ val peer = location.getPeer
+ val peerWorker = new WorkerInfo(
+ peer.getHost,
+ peer.getRpcPort,
+ peer.getPushPort,
+ peer.getFetchPort,
+ peer.getReplicatePort)
+ if (unavailablePeers.containsKey(peerWorker)) {
+ fileWriter.decrementPendingWrites()
+
workerSource.incCounter(WorkerSource.REPLICATE_DATA_CREATE_CONNECTION_FAIL_COUNT)
+ logError(
+ s"PushData replication failed caused by unavailable peer for
partitionLocation: $location")
+ callbackWithTimer.onFailure(
+ new
CelebornIOException(StatusCode.PUSH_DATA_CREATE_CONNECTION_FAIL_REPLICA))
+ return
+ }
+
pushData.body().retain()
replicateThreadPool.submit(new Runnable {
override def run(): Unit = {
- val peer = location.getPeer
Review Comment:
Done~
--
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]
