You are viewing a plain text version of this content. The canonical link for it is here.
Posted to commits@pulsar.apache.org by "lhotari (via GitHub)" <gi...@apache.org> on 2023/01/31 12:56:50 UTC

[GitHub] [pulsar] lhotari commented on a diff in pull request #19015: [improve][admin] Improve partitioned-topic condition evaluation

lhotari commented on code in PR #19015:
URL: https://github.com/apache/pulsar/pull/19015#discussion_r1091886551


##########
pulsar-broker/src/main/java/org/apache/pulsar/broker/admin/impl/PersistentTopicsBase.java:
##########
@@ -2608,80 +2608,87 @@ protected void internalGetSubscriptionProperties(AsyncResponse asyncResponse, St
     protected void internalResetCursorOnPosition(AsyncResponse asyncResponse, String subName, boolean authoritative,
             MessageIdImpl messageId, boolean isExcluded, int batchIndex) {
         CompletableFuture<Void> ret;
-        if (topicName.isGlobal()) {
-            ret = validateGlobalNamespaceOwnershipAsync(namespaceName);
+        // If the topic name is a partition name, no need to get partition topic metadata again
+        if (!topicName.isPartitioned()) {
+            ret = getPartitionedTopicMetadataAsync(topicName, authoritative, false)
+                    .thenCompose(topicMetadata -> {
+                        if (topicMetadata.partitions > 0) {
+                            log.warn("[{}] Not supported operation on partitioned-topic {} {}",
+                                    clientAppId(), topicName, subName);
+                            asyncResponse.resume(new RestException(Status.METHOD_NOT_ALLOWED,
+                                    "Reset-cursor at position is not allowed for partitioned-topic"));
+                        }
+                        return CompletableFuture.completedFuture(null);
+                    });
         } else {
             ret = CompletableFuture.completedFuture(null);
         }
-        ret.thenAccept(__ -> {
+
+        CompletableFuture<Void> future;
+        if (topicName.isGlobal()) {
+            future = ret.thenCompose(__ -> validateGlobalNamespaceOwnershipAsync(namespaceName));
+        } else {
+            future = CompletableFuture.completedFuture(null);
+        }
+        future.thenAccept(__ -> {
             log.info("[{}][{}] received reset cursor on subscription {} to position {}", clientAppId(), topicName,
                     subName, messageId);
-            // If the topic name is a partition name, no need to get partition topic metadata again
-            if (!topicName.isPartitioned()
-                    && getPartitionedTopicMetadata(topicName, authoritative, false).partitions > 0) {

Review Comment:
   This fixes an issue where Zookeeper session dies because the EventThread is blocked
   Stack trace
   ```
   "main-EventThread" #17 daemon prio=5 os_prio=0 cpu=1788.73ms elapsed=5176.10s tid=0x00007fa97a581000 nid=0x54 waiting on condition  [0x00007fa8ffbfb000]
      java.lang.Thread.State: WAITING (parking)
   	at jdk.internal.misc.Unsafe.park(java.base@11.0.16/Native Method)
   	- parking to wait for  <0x00000007dbaff838> (a java.util.concurrent.CompletableFuture$Signaller)
   	at java.util.concurrent.locks.LockSupport.park(java.base@11.0.16/LockSupport.java:194)
   	at java.util.concurrent.CompletableFuture$Signaller.block(java.base@11.0.16/CompletableFuture.java:1796)
   	at java.util.concurrent.ForkJoinPool.managedBlock(java.base@11.0.16/ForkJoinPool.java:3128)
   	at java.util.concurrent.CompletableFuture.waitingGet(java.base@11.0.16/CompletableFuture.java:1823)
   	at java.util.concurrent.CompletableFuture.get(java.base@11.0.16/CompletableFuture.java:1998)
   	at org.apache.pulsar.broker.admin.AdminResource.fetchPartitionedTopicMetadata(AdminResource.java:492)
   	at org.apache.pulsar.broker.admin.AdminResource.getPartitionedTopicMetadata(AdminResource.java:480)
   	at org.apache.pulsar.broker.admin.impl.PersistentTopicsBase.lambda$internalResetCursorOnPosition$212(PersistentTopicsBase.java:2595)
   	at org.apache.pulsar.broker.admin.impl.PersistentTopicsBase$$Lambda$1475/0x0000000840b46840.accept(Unknown Source)
   	at java.util.concurrent.CompletableFuture$UniAccept.tryFire(java.base@11.0.16/CompletableFuture.java:714)
   	at java.util.concurrent.CompletableFuture.postComplete(java.base@11.0.16/CompletableFuture.java:506)
   	at java.util.concurrent.CompletableFuture.complete(java.base@11.0.16/CompletableFuture.java:2073)
   	at org.apache.pulsar.broker.web.PulsarWebResource.lambda$checkLocalOrGetPeerReplicationCluster$21(PulsarWebResource.java:859)
   	at org.apache.pulsar.broker.web.PulsarWebResource$$Lambda$1135/0x000000084096fc40.accept(Unknown Source)
   	at java.util.concurrent.CompletableFuture$UniAccept.tryFire(java.base@11.0.16/CompletableFuture.java:714)
   	at java.util.concurrent.CompletableFuture.postComplete(java.base@11.0.16/CompletableFuture.java:506)
   	at java.util.concurrent.CompletableFuture.complete(java.base@11.0.16/CompletableFuture.java:2073)
   	at org.apache.pulsar.metadata.impl.ZKMetadataStore.handleGetResult(ZKMetadataStore.java:244)
   	at org.apache.pulsar.metadata.impl.ZKMetadataStore.lambda$batchOperation$7(ZKMetadataStore.java:188)
   	at org.apache.pulsar.metadata.impl.ZKMetadataStore$$Lambda$177/0x00000008402da040.processResult(Unknown Source)
   	at org.apache.pulsar.metadata.impl.PulsarZooKeeperClient$3$1.processResult(PulsarZooKeeperClient.java:490)
   	at org.apache.zookeeper.ClientCnxn$EventThread.processEvent(ClientCnxn.java:712)
   	at org.apache.zookeeper.ClientCnxn$EventThread.run(ClientCnxn.java:553)
      Locked ownable synchronizers:
   	- None
   ```
   
   



-- 
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: commits-unsubscribe@pulsar.apache.org

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