dongjoon-hyun commented on a change in pull request #30046:
URL: https://github.com/apache/spark/pull/30046#discussion_r505165348



##########
File path: 
core/src/main/scala/org/apache/spark/storage/BlockManagerDecommissioner.scala
##########
@@ -82,23 +83,35 @@ private[storage] class BlockManagerDecommissioner(
               Thread.sleep(SLEEP_TIME_SECS * 1000L)
             case Some((shuffleBlockInfo, retryCount)) =>
               if (retryCount < maxReplicationFailuresForDecommission) {
-                logInfo(s"Trying to migrate shuffle ${shuffleBlockInfo} to 
${peer}")
-                val blocks =
-                  bm.migratableResolver.getMigrationBlocks(shuffleBlockInfo)
+                logDebug(s"Trying to migrate shuffle ${shuffleBlockInfo} to 
${peer}")
+                val blocks = 
bm.migratableResolver.getMigrationBlocks(shuffleBlockInfo)
                 logDebug(s"Got migration sub-blocks ${blocks}")
-                blocks.foreach { case (blockId, buffer) =>
-                  logDebug(s"Migrating sub-block ${blockId}")
-                  bm.blockTransferService.uploadBlockSync(
-                    peer.host,
-                    peer.port,
-                    peer.executorId,
-                    blockId,
-                    buffer,
-                    StorageLevel.DISK_ONLY,
-                    null)// class tag, we don't need for shuffle
-                  logDebug(s"Migrated sub block ${blockId}")
+
+                // Migrate the components of the blocks.
+                try {
+                  blocks.foreach { case (blockId, buffer) =>
+                    logDebug(s"Migrating sub-block ${blockId}")
+                    bm.blockTransferService.uploadBlockSync(
+                      peer.host,
+                      peer.port,
+                      peer.executorId,
+                      blockId,
+                      buffer,
+                      StorageLevel.DISK_ONLY,
+                      null)// class tag, we don't need for shuffle
+                    logDebug(s"Migrated sub block ${blockId}")
+                  }
+                  logDebug(s"Migrated ${shuffleBlockInfo} to ${peer}")
+                } catch {
+                  case e: IOException =>
+                    // If a block got deleted before netty opened the file 
handle, then trying to

Review comment:
       Could you describe when this happens logically?




----------------------------------------------------------------
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.

For queries about this service, please contact Infrastructure at:
[email protected]



---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]

Reply via email to