[ https://issues.apache.org/jira/browse/ARTEMIS-4352?focusedWorklogId=869861&page=com.atlassian.jira.plugin.system.issuetabpanels:worklog-tabpanel#worklog-869861 ]
ASF GitHub Bot logged work on ARTEMIS-4352: ------------------------------------------- Author: ASF GitHub Bot Created on: 07/Jul/23 19:44 Start Date: 07/Jul/23 19:44 Worklog Time Spent: 10m Work Description: clebertsuconic opened a new pull request, #4543: URL: https://github.com/apache/activemq-artemis/pull/4543 If the Security Manager is using Netty, and in particular the same Netty connection, you could run into a deadlock / starvation. This is particularly true in the Wildfly case where they reuse the same connection for everything via XNIO. Issue Time Tracking ------------------- Worklog Id: (was: 869861) Remaining Estimate: 0h Time Spent: 10m > Deadlock can happen during JMS authentication with custom realm which > requests the same server > ---------------------------------------------------------------------------------------------- > > Key: ARTEMIS-4352 > URL: https://issues.apache.org/jira/browse/ARTEMIS-4352 > Project: ActiveMQ Artemis > Issue Type: Bug > Components: Broker > Affects Versions: 2.29.0 > Reporter: Emmanuel Hugonnet > Priority: Major > Time Spent: 10m > Remaining Estimate: 0h > > On session creation the IO thread is blocked during the authentication, this > may lead to deadlock or starvation. > The issue is in > org.apache.activemq.artemis.core.protocol.core.impl.ActiveMQPacketHandler#handlePacket > which should pass the process to an 'actor' -- This message was sent by Atlassian Jira (v8.20.10#820010)