[GitHub] [incubator-celeborn] pan3793 commented on a diff in pull request #1581: [CELEBORN-669] Avoid commit files on excluded worker list

["pan3793 (via GitHub)" <gi...@apache.org>]

pan3793 commented on code in PR #1581:
URL: https://github.com/apache/incubator-celeborn/pull/1581#discussion_r1228147074


##########
client/src/main/scala/org/apache/celeborn/client/commit/CommitHandler.scala:
##########
@@ -265,14 +266,29 @@ abstract class CommitHandler(
           slaveIds,
           getMapperAttempts(shuffleId),
           commitEpoch.incrementAndGet())
-        val res = requestCommitFilesWithRetry(worker.endpoint, commitFiles)
+        val res =
+          if (conf.clientCommitFilesIgnoreExcludedWorkers && workerStatusTracker.blacklist.containsKey(
+              worker)) {
+            CommitFilesResponse(
+              StatusCode.WORKER_IN_BLACKLIST,
+              List.empty.asJava,
+              List.empty.asJava,
+              masterIds,
+              slaveIds)
+          } else {
+            requestCommitFilesWithRetry(worker.endpoint, commitFiles)
+          }
 
         res.status match {
           case StatusCode.SUCCESS => // do nothing
           case StatusCode.PARTIAL_SUCCESS | StatusCode.SHUFFLE_NOT_REGISTERED | StatusCode.REQUEST_FAILED =>
             logDebug(s"Request $commitFiles return ${res.status} for " +
               s"${Utils.makeShuffleKey(applicationId, shuffleId)}")
             commitFilesFailedWorkers.put(worker, (res.status, System.currentTimeMillis()))
+          case StatusCode.WORKER_IN_BLACKLIST =>
+            // Do not record worker failure since it's from local excluded worker list
+            logDebug(s"Request $commitFiles return ${res.status} for " +

Review Comment:
   is the message too noisy? I think the abnormal message should be info at least.



-- 
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: issues-unsubscribe@celeborn.apache.org

For queries about this service, please contact Infrastructure at:
users@infra.apache.org