lhotari commented on pull request #11884: URL: https://github.com/apache/pulsar/pull/11884#issuecomment-914076442
Several comments have been about using a concurrent queue implementation. I don't think that is necessary as long as the access of the queue is properly synchronized. The solution in this PR demonstrates in a unit test how the ConcurrentModificationException can happen in a single thread. When using an ordinary ArrayDeque the test would fail. The root cause of #11783 is that while the current thread is executing the action in the forEach block, the callback code might try to add a new OpSendMsg entry in the calling thread. Some locations: https://github.com/apache/pulsar/blob/d86db3f4ec4fb6bd04216a123cde2fee5c43f9d9/pulsar- client/src/main/java/org/apache/pulsar/client/impl/ProducerImpl.java#L886-L891 https://github.com/apache/pulsar/blob/d86db3f4ec4fb6bd04216a123cde2fee5c43f9d9/pulsar-client/src/main/java/org/apache/pulsar/client/impl/ProducerImpl.java#L1638-L1657 I guess it's a valid case that after failing all pending messages the client code is allowed to access the producer in the callback to send more messages. (there are more details in #11783) @merlimat @codelipenghui Does the approach in this PR make sense, to postpone elements added while the forEach loop is in progress? -- 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-unsubscr...@pulsar.apache.org For queries about this service, please contact Infrastructure at: us...@infra.apache.org
