[ https://issues.jboss.org/browse/WFCORE-1075?page=com.atlassian.jira.plugi... ] Brian Stansberry updated WFCORE-1075: ------------------------------------- Description: Host Controllers frequently take 15 seconds to shutdown and reload, logging the following after a 15s pause: {code} WFLYCTL0019: Graceful shutdown of the handler used for native management requests did not complete within [15000] ms but shutdown of the underlying communication channel is proceeding {code} This is because MasterDomainControllerOperationHandlerImpl and ServerToHostProtocolHandler both handle a GET_FILE_REQUEST by registering an ActiveOperation with ActiveOperationSupport, but then the GetFileOperation handlers for the request never call "done" or "failed" on the ActiveOperation.ResultHandler. The activeRequests map in the ActiveOperationSupport therefore ends up with extraneous data resulting in the shutdown delay. This is also a minor memory leak. was: Host Controllers frequently take 15 seconds to shutdown and reload, logging the following after a 15s pause: {code} WFLYCTL0019: Graceful shutdown of the handler used for native management requests did not complete within [15000] ms but shutdown of the underlying communication channel is proceeding {code} This is because MasterDomainControllerOperationHandlerImpl and ServerToHostProtocolHandler both handle a GET_FILE_REQUEST by registering and ActiveOperation with ActiveOperationSupport, but then the GetFileOperation handlers for the request never call "done" or "failed" on the ActiveOperation.ResultHandler. The activeRequests map in the ActiveOperationSupport therefore ends up with extraneous data resulting in the shutdown delay. This is also a minor memory leak. -- This message was sent by Atlassian JIRA (v6.4.11#64026)