[ https://issues.apache.org/jira/browse/KAFKA-2046?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=14378944#comment-14378944 ]
Onur Karaman commented on KAFKA-2046: ------------------------------------- I just tried deleting a topic with 128 partitions and noticed that the delete topic node and the topic node were still in zk, and all but one replica on the brokers had not been deleted. grep "handling stop replica (delete=false)" kafka-state-change.log produced output for all of the partitions. So the controller was able to send a StopReplicaRequest to the brokers to transition from OnlineReplica to OfflineReplica. However, grep "handling stop replica (delete=true)" kafka-state-change.log only revealed only one replica. This was the replica that I noticed had actually been deleted from the filesystem. The other replicas never received the delete=true StopReplicaRequest. So the transition from OfflineReplica to ReplicaDeletionStarted for all the other replicas hang. A thread dump on the controller indicates that it's getting stuck because of a LinkedBlockingQueue in ControllerChannelManager: {code} "delete-topics-thread-xyz"... java.lang.Thread.State: WAITING (parking) ... at java.util.concurrent.locks.LockSupport.park(LockSupport.java:175) at java.util.concurrent.locks.AbstractQueuedSynchronizer$ConditionObject.await(AbstractQueuedSynchronizer.java:2039) at java.util.concurrent.LinkedBlockingQueue.put(LinkedBlockingQueue.java:350) at kafka.controller.ControllerChannelManager.sendRequest(ControllerChannelManager.scala:57) ... at kafka.controller.KafkaController.sendRequest(KafkaController.scala:670) at kafka.controller.ControllerBrokerRequestBatch$$anonfun$sendRequestsToBrokers$3$$anonfun$apply$10.apply(ControllerChannelManager.scala:320) at kafka.controller.ControllerBrokerRequestBatch$$anonfun$sendRequestsToBrokers$3$$anonfun$apply$10.apply(ControllerChannelManager.scala:317) at scala.collection.immutable.List.foreach(List.scala:318) at kafka.controller.ControllerBrokerRequestBatch$$anonfun$sendRequestsToBrokers$3.apply(ControllerChannelManager.scala:317) at kafka.controller.ControllerBrokerRequestBatch$$anonfun$sendRequestsToBrokers$3.apply(ControllerChannelManager.scala:310) at scala.collection.mutable.HashMap$$anonfun$foreach$1.apply(HashMap.scala:98) at scala.collection.mutable.HashMap$$anonfun$foreach$1.apply(HashMap.scala:98) at scala.collection.mutable.HashTable$class.foreachEntry(HashTable.scala:226) at scala.collection.mutable.HashMap.foreachEntry(HashMap.scala:39) at scala.collection.mutable.HashMap.foreach(HashMap.scala:98) at kafka.controller.ControllerBrokerRequestBatch.sendRequestsToBrokers(ControllerChannelManager.scala:310) at kafka.controller.ReplicaStateMachine.handleStateChanges(ReplicaStateMachine.scala:115) at kafka.controller.TopicDeletionManager$$anonfun$startReplicaDeletion$2.apply(TopicDeletionManager.scala:337) at kafka.controller.TopicDeletionManager$$anonfun$startReplicaDeletion$2.apply(TopicDeletionManager.scala:327) at scala.collection.immutable.Map$Map1.foreach(Map.scala:109) at kafka.controller.TopicDeletionManager.startReplicaDeletion(TopicDeletionManager.scala:327) {code} controller.message.queue.size property is currently set to be very small. I'll try bumping this up and see if this addresses the issue. > Delete topic still doesn't work > ------------------------------- > > Key: KAFKA-2046 > URL: https://issues.apache.org/jira/browse/KAFKA-2046 > Project: Kafka > Issue Type: Bug > Reporter: Clark Haskins > Assignee: Onur Karaman > > I just attempted to delete at 128 partition topic with all inbound producers > stopped. > The result was as follows: > The /admin/delete_topics znode was empty > the topic under /brokers/topics was removed > The Kafka topics command showed that the topic was removed > However, the data on disk on each broker was not deleted and the topic has > not yet been re-created by starting up the inbound mirror maker. > Let me know what additional information is needed -- This message was sent by Atlassian JIRA (v6.3.4#6332)