Re: Issues not closed for 2.3.0 release
Marshall Schor wrote: [...] Both of these approaches put uimaj-core classes at the root level of the application (i.e., System) class-loader chain. I can't think of any cases where that would not be OK; can anyone else? I think that may not work because the UIMA framework classes need to access the annotator classes. If the core is loaded by the system classloader, it will not have access to classes loaded by the bs loader. --Thilo
UIMA-AS error handling in case an exception is thrown from CM.next method
Hi everyone, I might found a problem in the error handling when an exception is thrown from the CasMultiplier.next method before getEmptyCas was called. UIMA-AS prints out the stack trace of the exception from the next method and afterwards it prints out the following exception stack trace, is that expected behavior ? Jörn Here is the exception stack trace: CASAdminException: Can't flush CAS, flushing is disabled. at org.apache.uima.cas.impl.CASImpl.reset(CASImpl.java:869) at org.apache.uima.util.CasPool.releaseCas(CasPool.java:224) at org.apache.uima.resource.impl.CasManager_impl.releaseCas(CasManager_impl.java:141) at org.apache.uima.cas.AbstractCas_ImplBase.release(AbstractCas_ImplBase.java:35) at org.apache.uima.cas.impl.CASImpl.release(CASImpl.java:3591) at org.apache.uima.cas.impl.CASImpl.release(CASImpl.java:3589) at org.apache.uima.aae.controller.BaseAnalysisEngineController.dropCAS(BaseAnalysisEngineController.java:1040) at org.apache.uima.aae.controller.BaseAnalysisEngineController.dropCAS(BaseAnalysisEngineController.java:1273) at org.apache.uima.aae.controller.AggregateAnalysisEngineController_impl.dropCAS(AggregateAnalysisEngineController_impl.java:302) at org.apache.uima.aae.controller.AggregateAnalysisEngineController_impl.sendReplyToRemoteClient(AggregateAnalysisEngineController_impl.java:1832) at org.apache.uima.aae.controller.AggregateAnalysisEngineController_impl.replyToClient(AggregateAnalysisEngineController_impl.java:1943) at org.apache.uima.aae.controller.AggregateAnalysisEngineController_impl.finalStep(AggregateAnalysisEngineController_impl.java:1567) at org.apache.uima.aae.controller.AggregateAnalysisEngineController_impl.abortProcessingCas(AggregateAnalysisEngineController_impl.java:971) at org.apache.uima.aae.controller.AggregateAnalysisEngineController_impl.process(AggregateAnalysisEngineController_impl.java:1010) at org.apache.uima.aae.error.handler.ProcessCasErrorHandler.handleError(ProcessCasErrorHandler.java:521) at org.apache.uima.aae.error.ErrorHandlerChain.handle(ErrorHandlerChain.java:57) at org.apache.uima.aae.handler.input.ProcessResponseHandler.handleProcessResponseWithException(ProcessResponseHandler.java:535) at org.apache.uima.aae.handler.input.ProcessResponseHandler.handle(ProcessResponseHandler.java:642) at org.apache.uima.aae.handler.HandlerBase.delegate(HandlerBase.java:149) at org.apache.uima.aae.handler.input.ProcessRequestHandler_impl.handle(ProcessRequestHandler_impl.java:951) at org.apache.uima.aae.spi.transport.vm.UimaVmMessageListener.onMessage(UimaVmMessageListener.java:107) at org.apache.uima.aae.spi.transport.vm.UimaVmMessageDispatcher$1.run(UimaVmMessageDispatcher.java:70) at java.util.concurrent.ThreadPoolExecutor$Worker.runTask(ThreadPoolExecutor.java:886) at java.util.concurrent.ThreadPoolExecutor$Worker.run(ThreadPoolExecutor.java:908) at java.lang.Thread.run(Thread.java:619) CASAdminException: Can't flush CAS, flushing is disabled. at org.apache.uima.cas.impl.CASImpl.reset(CASImpl.java:869) at org.apache.uima.util.CasPool.releaseCas(CasPool.java:224) at org.apache.uima.resource.impl.CasManager_impl.releaseCas(CasManager_impl.java:141) at org.apache.uima.cas.AbstractCas_ImplBase.release(AbstractCas_ImplBase.java:35) at org.apache.uima.cas.impl.CASImpl.release(CASImpl.java:3591) at org.apache.uima.cas.impl.CASImpl.release(CASImpl.java:3589) at org.apache.uima.aae.controller.BaseAnalysisEngineController.dropCAS(BaseAnalysisEngineController.java:1040) at org.apache.uima.aae.controller.BaseAnalysisEngineController.dropCAS(BaseAnalysisEngineController.java:1273) at org.apache.uima.aae.controller.AggregateAnalysisEngineController_impl.dropCAS(AggregateAnalysisEngineController_impl.java:302) at org.apache.uima.aae.controller.AggregateAnalysisEngineController_impl.sendReplyToRemoteClient(AggregateAnalysisEngineController_impl.java:1832) at org.apache.uima.aae.controller.AggregateAnalysisEngineController_impl.replyToClient(AggregateAnalysisEngineController_impl.java:1943) at org.apache.uima.aae.controller.AggregateAnalysisEngineController_impl.finalStep(AggregateAnalysisEngineController_impl.java:1567) at org.apache.uima.aae.controller.AggregateAnalysisEngineController_impl.abortProcessingCas(AggregateAnalysisEngineController_impl.java:971) at org.apache.uima.aae.controller.AggregateAnalysisEngineController_impl.process(AggregateAnalysisEngineController_impl.java:1010) at org.apache.uima.aae.error.handler.ProcessCasErrorHandler.handleError(ProcessCasErrorHandler.java:521) at org.apache.uima.aae.error.ErrorHandlerChain.handle(ErrorHandlerChain.java:57) at org.apache.uima.aae.handler.input.ProcessResponseHandler.handleProcessResponseWithException(ProcessResponseHandler.java:535) at org.apache.uima.aae.handler.input.ProcessResponseHandler.handle(ProcessResponseHandler.java:642) at
[jira] Closed: (UIMA-1624) Uima AS service should start listeners on its input queue on succesfull initialization only
[ https://issues.apache.org/jira/browse/UIMA-1624?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ] Jerry Cwiklik closed UIMA-1624. --- Resolution: Fixed Changed uima-as code to delay startup of listeners on input queues until service initializes. Uima AS service should start listeners on its input queue on succesfull initialization only --- Key: UIMA-1624 URL: https://issues.apache.org/jira/browse/UIMA-1624 Project: UIMA Issue Type: Bug Components: Async Scaleout Reporter: Jerry Cwiklik Fix For: 2.3AS Currently UIMA AS service JMS listener starts as soon as it is initialized. There are two listeners started on an input queue each with a different selector. One listener is for processing GetMeta while the other for Process messages. The listeners start in separate threads and become active despite the fact that the rest of service components may not have finished initializing. This creates a problem if the service fails in annotator code while initializing. Since the GetMeta listener is active, it may fetch GetMeta request from a queue and try to process it. The current code prevents the GetMeta from being processed since the service has not finished initializing. If the AE initialization fails, the getMeta is never returned to the client. Fix input queue listeners so that they become active only after successful initialization of AEs. In case of aggregates, allow listeners on temp reply queues to start immediately so that GetMeta replies from delegates can be processed. -- This message is automatically generated by JIRA. - You can reply to this email to add a comment to the issue online.
BaseAnalysisEngineController line 1742 IndexOutOfBoundsException
Hi, I got an IndexOutOfBoundsException, I think there is a concurrency problem at the mentioned place. Jörn Here is the stack trace: Caused by: java.lang.IndexOutOfBoundsException: Index: 2, Size: 0 at java.util.ArrayList.RangeCheck(ArrayList.java:547) at java.util.ArrayList.get(ArrayList.java:322) at org.apache.uima.aae.controller.BaseAnalysisEngineController.stop(BaseAnalysisEngineController.java:1741) at org.apache.uima.aae.controller.BaseAnalysisEngineController.stop(BaseAnalysisEngineController.java:1707) at org.apache.uima.aae.controller.AggregateAnalysisEngineController_impl.stop(AggregateAnalysisEngineController_impl.java:2753) at org.apache.uima.aae.controller.BaseAnalysisEngineController.terminate(BaseAnalysisEngineController.java:1967) at org.apache.uima.aae.controller.BaseAnalysisEngineController.terminate(BaseAnalysisEngineController.java:1940) at org.apache.uima.adapter.jms.activemq.SpringContainerDeployer.initializeContainer(SpringContainerDeployer.java:313) at org.apache.uima.adapter.jms.activemq.SpringContainerDeployer.deploy(SpringContainerDeployer.java:390) ... 7 more
[jira] Reopened: (UIMA-1433) UIMA AS service creates too many JMS connections
[ https://issues.apache.org/jira/browse/UIMA-1433?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ] Jerry Cwiklik reopened UIMA-1433: - UIMA AS service is not recovering properly when the broker is stopped and restarted. When the broker is bounced the service should invalidate its connection cache and create a new one. It should also invalidate all sessions and producers associated with a stale connection. The aggregate service with remote delegates will appear hung if the broker is bounced. UIMA AS service creates too many JMS connections - Key: UIMA-1433 URL: https://issues.apache.org/jira/browse/UIMA-1433 Project: UIMA Issue Type: Bug Components: Async Scaleout Reporter: Jerry Cwiklik Assignee: Jerry Cwiklik Fix For: 2.3AS UIMA AS service maintains connections to client's reply queue. These connections are cached and reused. When the connection becomes idle for too long, it is closed. Current default connection timeout is set to 30 minutes. This value can be changed via System property -DSessionTimeoutOverride=n. JMS Connections are expensive, and if too many are created the broker may become unstable. As optimization, the service should create a single JMS connection (per broker) and use it to create jms sessions and message producers for each client. This change will reduce number of threads the broker needs to manage in deployments where clients and services use a single broker for messaging. -- This message is automatically generated by JIRA. - You can reply to this email to add a comment to the issue online.
[jira] Created: (UIMA-1627) Application link not working on gcc 4.2
Application link not working on gcc 4.2 --- Key: UIMA-1627 URL: https://issues.apache.org/jira/browse/UIMA-1627 Project: UIMA Issue Type: Bug Components: C++ Framework Reporter: Eddie Epstein Assignee: Eddie Epstein Need to explicitly specify -lapr-1 -- This message is automatically generated by JIRA. - You can reply to this email to add a comment to the issue online.
Re: Issues not closed for 2.3.0 release
Thilo Goetz wrote: Marshall Schor wrote: [...] Both of these approaches put uimaj-core classes at the root level of the application (i.e., System) class-loader chain. I can't think of any cases where that would not be OK; can anyone else? I think that may not work because the UIMA framework classes need to access the annotator classes. If the core is loaded by the system classloader, it will not have access to classes loaded by the bs loader. Yes, this sounds right. How about this fix: put a copy of the UIMALogFormatter into the bootstrap loader Jar. This class does not reference any uima classes, only Java JRE classes. -Marshall --Thilo
Re: Issues not closed for 2.3.0 release
Marshall Schor wrote: Thilo Goetz wrote: Marshall Schor wrote: [...] Both of these approaches put uimaj-core classes at the root level of the application (i.e., System) class-loader chain. I can't think of any cases where that would not be OK; can anyone else? I think that may not work because the UIMA framework classes need to access the annotator classes. If the core is loaded by the system classloader, it will not have access to classes loaded by the bs loader. Yes, this sounds right. How about this fix: put a copy of the UIMALogFormatter into the bootstrap loader Jar. This class does not reference any uima classes, only Java JRE classes. I think this issue is general, not restricted to the Formatter class. The Javadocs for the LogManager say that the class objects named in the configuration are loaded by using the System class loader. It actually says the classes are first searched on the system class path before any user class path. But it seems that only the application initial class path that is used to launch things, is searched. So maybe the only approach will be to inject all the urls into the system class loader, as Jerry was trying to do (Jars and directories, not individual classes). Any other ideas? -Marshall -Marshall --Thilo
