[GitHub] [kafka] hachikuji commented on a diff in pull request #13579: KAFKA-14904: Flaky Test kafka.api.TransactionsBounceTest.testWithGroupId()

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

hachikuji commented on code in PR #13579:
URL: https://github.com/apache/kafka/pull/13579#discussion_r1170680513


##########
core/src/main/scala/kafka/coordinator/transaction/TransactionCoordinator.scala:
##########
@@ -332,24 +331,45 @@ class TransactionCoordinator(txnConfig: TransactionConfig,
       debug(s"Returning ${Errors.INVALID_REQUEST} error code to client for $transactionalId's AddPartitions request for verification")
       responseCallback(AddPartitionsToTxnResponse.resultForTransaction(transactionalId, partitions.map(_ -> Errors.INVALID_REQUEST).toMap.asJava))
     } else {
-      val result: ApiResult[(Int, TransactionMetadata)] = getTransactionMetadata(transactionalId, producerId, producerEpoch, partitions)
-      
+      val result: ApiResult[Map[TopicPartition, Errors]] =
+        txnManager.getTransactionState(transactionalId).flatMap {
+          case None => Left(Errors.INVALID_PRODUCER_ID_MAPPING)
+
+          case Some(epochAndMetadata) =>
+            val txnMetadata = epochAndMetadata.transactionMetadata
+
+            // generate the new transaction metadata with added partitions
+            txnMetadata.inLock {
+              if (txnMetadata.producerId != producerId) {
+                Left(Errors.INVALID_PRODUCER_ID_MAPPING)
+              } else if (txnMetadata.producerEpoch != producerEpoch) {
+                Left(Errors.PRODUCER_FENCED)
+              } else if (txnMetadata.pendingTransitionInProgress && !(txnMetadata.pendingState == Some(Ongoing) && txnMetadata.state == Ongoing)) {
+                // return a retriable exception to let the client backoff and retry
+                Left(Errors.CONCURRENT_TRANSACTIONS)

Review Comment:
   Yeah, I think that's right. It works because we remove the partition after we have confirmed that the end transaction marker has been written. So if the partition is included, then it means markers are still to come. This assumes we fix https://issues.apache.org/jira/browse/KAFKA-14884 of course.



-- 
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: jira-unsubscribe@kafka.apache.org

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