Re: CPM test case hang
Thilo Goetz wrote: Works fine with UIMA 2.2.2. I guess it's time to nuke the maven repo, do a clean extract from svn and see what happens. And do a reboot in the bargain :-) Did the above except nuke the maven repo, and it still hangs. Let me know if you want me to do anything to investigate. --Thilo --Thilo Thilo Goetz wrote: Jaroslaw Cwiklik wrote: I've applied a patch to VinciCasProcessorDeployer.java. Build with a version before the patch was applied and see if this problem goes away. No luck. I went back as far as I could so it would still compile, and still had the same problem. I guess the next thing to do would be to try our last release on that machine with that jvm and see if I have the same issue. I'll do that when I have time. --Thilo JC On Fri, Sep 4, 2009 at 11:44 AM, Thilo Goetz twgo...@gmx.de wrote: Jaroslaw Cwiklik wrote: Ran on the command line multiple times with IBM 1.5 and Sun's 1.5 and no hang. Yes, I can't reproduce it either on my laptop. Same code level, same jvm, both dual core processors, same maven version even. Still hangs reliably on my workstation, though. Some subtle synchronization issue, undoubtedly. Let me know if you want me to go back to an earlier code level and try there, or if there's anything else I can do to help. --Thilo JC On Fri, Sep 4, 2009 at 10:26 AM, Thilo Goetz twgo...@gmx.de wrote: Jaroslaw Cwiklik wrote: Thilo, I've tried IBM's 1.5.8 and Sun's 1.5 and I cant recreate the hang. I run the test cases in eclipse. Is anyone else having this problem? Please try it on the command line, that's where I had the problem. --Thilo Jerry On Thu, Sep 3, 2009 at 9:19 AM, Jaroslaw Cwiklik uim...@gmail.com wrote: Thilo, recently I applied a patch attached to JIRA UIMA-1304. I only tested this with JDK1.6. I will look into this today. Jerry On Thu, Sep 3, 2009 at 5:00 AM, Thilo Goetz twgo...@gmx.de wrote: Hi all, has anything changed in the CPM recently? I get a reliable test case hang here (no cpu activity): --- T E S T S --- Running org.apache.uima.collection.impl.cpm.CpmProcessingTest Tests run: 4, Failures: 0, Errors: 0, Skipped: 0, Time elapsed: 1.344 sec Running org.apache.uima.collection.impl.cpm.CasPoolTest Tests run: 1, Failures: 0, Errors: 0, Skipped: 0, Time elapsed: 0.125 sec Running org.apache.uima.collection.impl.cpm.CpmInitTest Initialize was called: true Sep 3, 2009 10:50:48 AM org.apache.uima.collection.impl.cpm.BaseCPMImpl process WARNING: Processing the request to stop the CPM. (Thread Name: main) Sep 3, 2009 10:50:48 AM org.apache.uima.collection.impl.cpm.engine.CPMEngine process INFO: The collection reader thread state is: 2007 (Thread Name: main) I have to kill the process, so I don't have more info than this. It only happens with this jvm (on windows): java version 1.5.0 Java(TM) 2 Runtime Environment, Standard Edition (build pwi32dev-20090707 (SR10 )) IBM J9 VM (build 2.3, J2RE 1.5.0 IBM J9 2.3 Windows XP x86-32 j9vmwi3223-20090707 (JIT enabled) J9VM - 20090706_38445_lHdSMr JIT - 20090623_1334_r8 GC - 200906_09) JCL - 20090705 I tried 3 others, and it goes through with all of them: java version 1.6.0 Java(TM) SE Runtime Environment (build pwi3260sr5-20090529_04(SR5)) IBM J9 VM (build 2.4, J2RE 1.6.0 IBM J9 2.4 Windows XP x86-32 jvmwi3260sr5-20090519_35743 (JIT enabled, AOT enabled) J9VM - 20090519_035743_lHdSMr JIT - r9_20090518_2017 GC - 20090417_AA) JCL - 20090529_01 java version 1.5.0_11 Java(TM) 2 Runtime Environment, Standard Edition (build 1.5.0_11-b03) Java HotSpot(TM) Client VM (build 1.5.0_11-b03, mixed mode) java version 1.6.0_13 Java(TM) SE Runtime Environment (build 1.6.0_13-b03) Java HotSpot(TM) Client VM (build 11.3-b02, mixed mode) Let me know if you need more information. --Thilo
Re: CPM test case hang
Thilo Goetz wrote: Thilo Goetz wrote: Works fine with UIMA 2.2.2. I guess it's time to nuke the maven repo, do a clean extract from svn and see what happens. And do a reboot in the bargain :-) Did the above except nuke the maven repo, and it still hangs. Let me know if you want me to do anything to investigate. Please do the following when you get the hang to produce some diagnositic info: Press ctrl-break to generate a Java Dump Look in the Java Dump for the section labeled: LOCKS subcomponent dump routine See if you see anything like Deadlock detected !!! - If so, please post the relevant info :-) . -Marshall --Thilo --Thilo Thilo Goetz wrote: Jaroslaw Cwiklik wrote: I've applied a patch to VinciCasProcessorDeployer.java. Build with a version before the patch was applied and see if this problem goes away. No luck. I went back as far as I could so it would still compile, and still had the same problem. I guess the next thing to do would be to try our last release on that machine with that jvm and see if I have the same issue. I'll do that when I have time. --Thilo JC On Fri, Sep 4, 2009 at 11:44 AM, Thilo Goetz twgo...@gmx.de wrote: Jaroslaw Cwiklik wrote: Ran on the command line multiple times with IBM 1.5 and Sun's 1.5 and no hang. Yes, I can't reproduce it either on my laptop. Same code level, same jvm, both dual core processors, same maven version even. Still hangs reliably on my workstation, though. Some subtle synchronization issue, undoubtedly. Let me know if you want me to go back to an earlier code level and try there, or if there's anything else I can do to help. --Thilo JC On Fri, Sep 4, 2009 at 10:26 AM, Thilo Goetz twgo...@gmx.de wrote: Jaroslaw Cwiklik wrote: Thilo, I've tried IBM's 1.5.8 and Sun's 1.5 and I cant recreate the hang. I run the test cases in eclipse. Is anyone else having this problem? Please try it on the command line, that's where I had the problem. --Thilo Jerry On Thu, Sep 3, 2009 at 9:19 AM, Jaroslaw Cwiklik uim...@gmail.com wrote: Thilo, recently I applied a patch attached to JIRA UIMA-1304. I only tested this with JDK1.6. I will look into this today. Jerry On Thu, Sep 3, 2009 at 5:00 AM, Thilo Goetz twgo...@gmx.de wrote: Hi all, has anything changed in the CPM recently? I get a reliable test case hang here (no cpu activity): --- T E S T S --- Running org.apache.uima.collection.impl.cpm.CpmProcessingTest Tests run: 4, Failures: 0, Errors: 0, Skipped: 0, Time elapsed: 1.344 sec Running org.apache.uima.collection.impl.cpm.CasPoolTest Tests run: 1, Failures: 0, Errors: 0, Skipped: 0, Time elapsed: 0.125 sec Running org.apache.uima.collection.impl.cpm.CpmInitTest Initialize was called: true Sep 3, 2009 10:50:48 AM org.apache.uima.collection.impl.cpm.BaseCPMImpl process WARNING: Processing the request to stop the CPM. (Thread Name: main) Sep 3, 2009 10:50:48 AM org.apache.uima.collection.impl.cpm.engine.CPMEngine process INFO: The collection reader thread state is: 2007 (Thread Name: main) I have to kill the process, so I don't have more info than this. It only happens with this jvm (on windows): java version 1.5.0 Java(TM) 2 Runtime Environment, Standard Edition (build pwi32dev-20090707 (SR10 )) IBM J9 VM (build 2.3, J2RE 1.5.0 IBM J9 2.3 Windows XP x86-32 j9vmwi3223-20090707 (JIT enabled) J9VM - 20090706_38445_lHdSMr JIT - 20090623_1334_r8 GC - 200906_09) JCL - 20090705 I tried 3 others, and it goes through with all of them: java version 1.6.0 Java(TM) SE Runtime Environment (build pwi3260sr5-20090529_04(SR5)) IBM J9 VM (build 2.4, J2RE 1.6.0 IBM J9 2.4 Windows XP x86-32 jvmwi3260sr5-20090519_35743 (JIT enabled, AOT enabled) J9VM - 20090519_035743_lHdSMr JIT - r9_20090518_2017 GC - 20090417_AA) JCL - 20090529_01 java version 1.5.0_11 Java(TM) 2 Runtime Environment, Standard Edition (build 1.5.0_11-b03) Java HotSpot(TM) Client VM (build 1.5.0_11-b03, mixed mode) java version 1.6.0_13 Java(TM) SE Runtime Environment (build 1.6.0_13-b03) Java HotSpot(TM) Client VM (build 11.3-b02, mixed mode) Let me know if you need more information. --Thilo
Re: [jira] Reopened: (UIMA-1462) SimpleUimaAsService has checked in SimpleServer libraries as binaries
Hi Marshall, Lookin in the SandboxDistr POM I see SimpleUimaAsService is commented because SimpleUimaAsService needs uima-as for building and uima-as is not in the lifecycle of Sandbox, so finally SimpleUimaAsService was left out to avoid the Sandbox building failure. For this reason I did not put the parent SandboxDistr POM inside SimpleUimaAsService. Tommaso 2009/10/2 Marshall Schor m...@schor.com Tommaso Teofili (JIRA) wrote: [ https://issues.apache.org/jira/browse/UIMA-1462?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel] Tommaso Teofili reopened UIMA-1462: --- some plugins useful for licensing purposes were wrongly removed from pom Hi Tommaso - This POM appears to be missing the parent link. The POM parents add a lot of common build / dependency behavior; can you add that link, and then the much simpler POM will maybe work :-) -Marshall SimpleUimaAsService has checked in SimpleServer libraries as binaries - Key: UIMA-1462 URL: https://issues.apache.org/jira/browse/UIMA-1462 Project: UIMA Issue Type: Bug Components: Sandbox-SimpleUimaAsService Affects Versions: 2.2.2S, 2.3S Reporter: Thilo Goetz Assignee: Tommaso Teofili Fix For: 2.3S Attachments: patch1462.txt, patch1462b1.txt, patch1462b2.txt, patch1462c1.txt, patch1462c2.txt, simpleServerDistributionPatch.txt, simpleUimaAsServicePatch.txt For the upcoming release, this should be changed so it depends on the SimpleServer project directly.
[jira] Updated: (UIMA-1581) RegularExpressionAnnotator tests fail
[ https://issues.apache.org/jira/browse/UIMA-1581?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ] Marshall Schor updated UIMA-1581: - Fix Version/s: (was: 2.3) RegularExpressionAnnotator tests fail - Key: UIMA-1581 URL: https://issues.apache.org/jira/browse/UIMA-1581 Project: UIMA Issue Type: Bug Components: Sandbox-RegexAnnotator Affects Versions: 2.3S Environment: MacOS X Reporter: Tommaso Teofili Assignee: Tommaso Teofili the following TestSetFeatureValues tests fail: * testSetFeatureValues1 * testSetFeatureValues2 * testSetFeatureValues3 -- This message is automatically generated by JIRA. - You can reply to this email to add a comment to the issue online.
[jira] Closed: (UIMA-1555) repackage UIMA-AS to work as an add-on to base UIMA
[ https://issues.apache.org/jira/browse/UIMA-1555?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ] Marshall Schor closed UIMA-1555. Resolution: Fixed repackage UIMA-AS to work as an add-on to base UIMA --- Key: UIMA-1555 URL: https://issues.apache.org/jira/browse/UIMA-1555 Project: UIMA Issue Type: Task Components: Async Scaleout, Build, Packaging and Test Affects Versions: 2.3AS Reporter: Marshall Schor Fix For: 2.3AS Change the UIMA-AS binary packaging to be an add-on to base UIMA. # remove uima base jars from uima-as bin packaging # change instructions for installing uima-as to unzipping/taring over base uima level # update docs to reflect this packaging. # verify that javadocs generate, and that docs say there are 2 javadocs sets - one for base UIMA and one for uima-as addon # verify that uima-as startup checks for compatible level of uima base (today this means equal level) # verify that the binary examples (in the UIMA_HOME/examples directory) is the proper merge of base and UIMA-AS # verify that the scripts work -- This message is automatically generated by JIRA. - You can reply to this email to add a comment to the issue online.
[jira] Updated: (UIMA-1532) Generics in FSIndex maybe incorrect
[ https://issues.apache.org/jira/browse/UIMA-1532?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ] Marshall Schor updated UIMA-1532: - Affects Version/s: 2.3 Fix Version/s: (was: 2.3) Defer past 2.3.0 Generics in FSIndex maybe incorrect --- Key: UIMA-1532 URL: https://issues.apache.org/jira/browse/UIMA-1532 Project: UIMA Issue Type: Bug Components: Core Java Framework Affects Versions: 2.3 Reporter: Jörn Kottmann The generics in FSIndex still need to be checked for correctness, there is already a thread about that on the mailing list. This issue should make sure we do not forget it. Here is the thread form the dev list: http://markmail.org/message/padxg7yhw5rnsjjo -- This message is automatically generated by JIRA. - You can reply to this email to add a comment to the issue online.
[jira] Updated: (UIMA-1481) add Javadocs to UIMA-AS assembly
[ https://issues.apache.org/jira/browse/UIMA-1481?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ] Marshall Schor updated UIMA-1481: - Fix Version/s: 2.3AS add Javadocs to UIMA-AS assembly Key: UIMA-1481 URL: https://issues.apache.org/jira/browse/UIMA-1481 Project: UIMA Issue Type: Improvement Components: Async Scaleout Reporter: Marshall Schor Assignee: Marshall Schor Priority: Minor Fix For: 2.3AS -- This message is automatically generated by JIRA. - You can reply to this email to add a comment to the issue online.
[jira] Closed: (UIMA-1481) add Javadocs to UIMA-AS assembly
[ https://issues.apache.org/jira/browse/UIMA-1481?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ] Marshall Schor closed UIMA-1481. Resolution: Fixed add Javadocs to UIMA-AS assembly Key: UIMA-1481 URL: https://issues.apache.org/jira/browse/UIMA-1481 Project: UIMA Issue Type: Improvement Components: Async Scaleout Reporter: Marshall Schor Assignee: Marshall Schor Priority: Minor -- This message is automatically generated by JIRA. - You can reply to this email to add a comment to the issue online.
Re: CPM test case hang
Marshall Schor wrote: Thilo Goetz wrote: Thilo Goetz wrote: Works fine with UIMA 2.2.2. I guess it's time to nuke the maven repo, do a clean extract from svn and see what happens. And do a reboot in the bargain :-) Did the above except nuke the maven repo, and it still hangs. Let me know if you want me to do anything to investigate. Please do the following when you get the hang to produce some diagnositic info: Press ctrl-break to generate a Java Dump Look in the Java Dump for the section labeled: LOCKS subcomponent dump routine See if you see anything like Deadlock detected !!! - Well, no deadlock detected, I'm afraid. I include the locks and threads section below. This is starting to look like a maven problem to me. I see all these threads waiting, and they all seem to be maven threads, not from any CPM code afaict. The JSR 166 backport classes seem to come out of the maven jar. Unfortunately I can't try a different maven version as our build appears to require 2.2 now. Great hint about Ctrl-Break, I didn't know you could do that. --Thilo 0SECTION LOCKS subcomponent dump routine NULL === NULL 1LKPOOLINFOMonitor pool info: 2LKPOOLTOTAL Current total number of monitors: 21 NULL 1LKMONPOOLDUMP Monitor Pool Dump (flat inflated object-monitors): 2LKMONINUSE sys_mon_t:0x62AB2998 infl_mon_t: 0x62AB29D8: 3LKMONOBJECT hidden/edu/emory/mathcs/backport/java/util/concurrent/linkedblockingqueue$serializablel...@00be1de8/00BE1DF4: unowned 3LKNOTIFYQWaiting to be notified: 3LKWAITNOTIFYpool-1-thread-1 (0x62695900) 3LKWAITNOTIFYpool-1-thread-2 (0x62EC1500) 3LKWAITNOTIFYpool-1-thread-3 (0x62EC1900) 3LKWAITNOTIFYpool-1-thread-4 (0x62EC1D00) 3LKWAITNOTIFYpool-1-thread-5 (0x62C3FD00) NULL 0SECTION THREADS subcomponent dump routine NULL = NULL 1XMCURTHDINFO Current Thread Details NULL -- NULL 1XMTHDINFO All Thread Details NULL -- NULL 2XMFULLTHDDUMP Full thread dump J9 VM (J2RE 5.0 IBM J9 2.3 Windows XP x86-32 build 20090706_38445_lHdSMr, native threads): 3XMTHREADINFO main (TID:0x0009F300, sys_thread_t:0x00036E28, state:R, native ID:0x084C) prio=5 4XESTACKTRACE at java/lang/ProcessImpl.waitFor(Native Method) 4XESTACKTRACE at org/apache/maven/surefire/booter/shade/org/codehaus/plexus/util/cli/CommandLineUtils.executeCommandLine(CommandLineUtils.java:146) 4XESTACKTRACE at org/apache/maven/surefire/booter/shade/org/codehaus/plexus/util/cli/CommandLineUtils.executeCommandLine(CommandLineUtils.java:98) 4XESTACKTRACE at org/apache/maven/surefire/booter/SurefireBooter.fork(SurefireBooter.java:673) 4XESTACKTRACE at org/apache/maven/surefire/booter/SurefireBooter.forkSuites(SurefireBooter.java:479) 4XESTACKTRACE at org/apache/maven/surefire/booter/SurefireBooter.runSuitesForkOnce(SurefireBooter.java:379) 4XESTACKTRACE at org/apache/maven/surefire/booter/SurefireBooter.run(SurefireBooter.java:245) 4XESTACKTRACE at org/apache/maven/plugin/surefire/SurefirePlugin.execute(SurefirePlugin.java:537) 4XESTACKTRACE at org/apache/maven/plugin/DefaultPluginManager.executeMojo(DefaultPluginManager.java:490) 4XESTACKTRACE at org/apache/maven/lifecycle/DefaultLifecycleExecutor.executeGoals(DefaultLifecycleExecutor.java:694) 4XESTACKTRACE at org/apache/maven/lifecycle/DefaultLifecycleExecutor.executeGoalWithLifecycle(DefaultLifecycleExecutor.java:556) 4XESTACKTRACE at org/apache/maven/lifecycle/DefaultLifecycleExecutor.executeGoal(DefaultLifecycleExecutor.java:535) 4XESTACKTRACE at org/apache/maven/lifecycle/DefaultLifecycleExecutor.executeGoalAndHandleFailures(DefaultLifecycleExecutor.java:387) 4XESTACKTRACE at org/apache/maven/lifecycle/DefaultLifecycleExecutor.executeTaskSegments(DefaultLifecycleExecutor.java:348) 4XESTACKTRACE at org/apache/maven/lifecycle/DefaultLifecycleExecutor.execute(DefaultLifecycleExecutor.java:180) 4XESTACKTRACE at org/apache/maven/DefaultMaven.doExecute(DefaultMaven.java:328) 4XESTACKTRACE at org/apache/maven/DefaultMaven.execute(DefaultMaven.java:138) 4XESTACKTRACE at org/apache/maven/cli/MavenCli.main(MavenCli.java:362(Compiled Code)) 4XESTACKTRACE at org/apache/maven/cli/compat/CompatibleMain.main(CompatibleMain.java:60) 4XESTACKTRACE at sun/reflect/NativeMethodAccessorImpl.invoke0(Native Method) 4XESTACKTRACE at sun/reflect/NativeMethodAccessorImpl.invoke(NativeMethodAccessorImpl.java:79) 4XESTACKTRACE at sun/reflect/DelegatingMethodAccessorImpl.invoke(DelegatingMethodAccessorImpl.java:43) 4XESTACKTRACE at java/lang/reflect/Method.invoke(Method.java:618) 4XESTACKTRACE at
[jira] Closed: (UIMA-1590) fix extractAndBuild scripts
[ https://issues.apache.org/jira/browse/UIMA-1590?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ] Marshall Schor closed UIMA-1590. Resolution: Fixed works except the .sh versions don't build the PDF versions due to font missing. fix extractAndBuild scripts --- Key: UIMA-1590 URL: https://issues.apache.org/jira/browse/UIMA-1590 Project: UIMA Issue Type: Improvement Components: Build, Packaging and Test Reporter: Marshall Schor Fix For: 2.3S, 2.3, 2.3AS The extractAndBuild scripts need the following improvements: # since the builds for uima-as and the sandbox require that these projects be in the same directory as the base uima, fix the script to extract into a subdir of the base uima extraction, and then copy into the same base uima dir. # The scripts have been failing when it comes to doing the doc building, because they are missing saxon, jai, and the docbook 4.5 xml files; the script tries to get these from the internet (with permission) but fail on the 2nd request due to a problem of getting Ant I/O to the console to work more than once when Ant is invoked from maven. Also, it's a nuisance. Change this to attempt to copy over any of this tooling that might be in the same projects that contain the extractAndBuild scripts in the first place. So - if you've fixed those up once, these will be reused. # regularize the arguments - so they all take 2 args for the release and release-candidate, so all extractAndBuild args are the same. # for the sandbox build - fix the assembly to include the -N parameter - no need to build twice. -- This message is automatically generated by JIRA. - You can reply to this email to add a comment to the issue online.
[jira] Updated: (UIMA-1099) The Cas Editor leaks memory with each opened file.
[ https://issues.apache.org/jira/browse/UIMA-1099?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ] Marshall Schor updated UIMA-1099: - Fix Version/s: (was: 2.3AS) 2.3 The Cas Editor leaks memory with each opened file. -- Key: UIMA-1099 URL: https://issues.apache.org/jira/browse/UIMA-1099 Project: UIMA Issue Type: Bug Components: CasEditor Affects Versions: 2.2.2CE Reporter: Jörn Kottmann Assignee: Jörn Kottmann Fix For: 2.3 For each file which was openend in the Cas Editor a certain amount of memory is leaking. -- This message is automatically generated by JIRA. - You can reply to this email to add a comment to the issue online.
[jira] Updated: (UIMA-1506) update Bean Scripting Framework Annotator with info about licenses and documentation
[ https://issues.apache.org/jira/browse/UIMA-1506?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ] Marshall Schor updated UIMA-1506: - Fix Version/s: (was: 2.3AS) 2.3S update Bean Scripting Framework Annotator with info about licenses and documentation Key: UIMA-1506 URL: https://issues.apache.org/jira/browse/UIMA-1506 Project: UIMA Issue Type: Improvement Components: Sandbox-BSFAnnotator Reporter: Marshall Schor Assignee: Marshall Schor Priority: Minor Fix For: 2.3S Attachments: BSFAnnotator.zip update with info received in email: http://markmail.org/thread/e3q5gk6h42jggg5c -- This message is automatically generated by JIRA. - You can reply to this email to add a comment to the issue online.
Re: CPM test case hang
Thilo Goetz wrote: Marshall Schor wrote: Thilo Goetz wrote: Thilo Goetz wrote: Works fine with UIMA 2.2.2. I guess it's time to nuke the maven repo, do a clean extract from svn and see what happens. And do a reboot in the bargain :-) Did the above except nuke the maven repo, and it still hangs. Let me know if you want me to do anything to investigate. Please do the following when you get the hang to produce some diagnositic info: Press ctrl-break to generate a Java Dump Look in the Java Dump for the section labeled: LOCKS subcomponent dump routine See if you see anything like Deadlock detected !!! - Well, no deadlock detected, I'm afraid. I include the locks and threads section below. This is starting to look like a maven problem to me. I see all these threads waiting, and they all seem to be maven threads, not from any CPM code afaict. The JSR 166 backport classes seem to come out of the maven jar. Unfortunately I can't try a different maven version as our build appears to require 2.2 now. Great hint about Ctrl-Break, I didn't know you could do that. Can you run the tests from Eclipse? Do they hang if you do that? Thanks... -Marshall --Thilo 0SECTION LOCKS subcomponent dump routine NULL === NULL 1LKPOOLINFOMonitor pool info: 2LKPOOLTOTAL Current total number of monitors: 21 NULL 1LKMONPOOLDUMP Monitor Pool Dump (flat inflated object-monitors): 2LKMONINUSE sys_mon_t:0x62AB2998 infl_mon_t: 0x62AB29D8: 3LKMONOBJECT hidden/edu/emory/mathcs/backport/java/util/concurrent/linkedblockingqueue$serializablel...@00be1de8/00BE1DF4: unowned 3LKNOTIFYQWaiting to be notified: 3LKWAITNOTIFYpool-1-thread-1 (0x62695900) 3LKWAITNOTIFYpool-1-thread-2 (0x62EC1500) 3LKWAITNOTIFYpool-1-thread-3 (0x62EC1900) 3LKWAITNOTIFYpool-1-thread-4 (0x62EC1D00) 3LKWAITNOTIFYpool-1-thread-5 (0x62C3FD00) NULL 0SECTION THREADS subcomponent dump routine NULL = NULL 1XMCURTHDINFO Current Thread Details NULL -- NULL 1XMTHDINFO All Thread Details NULL -- NULL 2XMFULLTHDDUMP Full thread dump J9 VM (J2RE 5.0 IBM J9 2.3 Windows XP x86-32 build 20090706_38445_lHdSMr, native threads): 3XMTHREADINFO main (TID:0x0009F300, sys_thread_t:0x00036E28, state:R, native ID:0x084C) prio=5 4XESTACKTRACE at java/lang/ProcessImpl.waitFor(Native Method) 4XESTACKTRACE at org/apache/maven/surefire/booter/shade/org/codehaus/plexus/util/cli/CommandLineUtils.executeCommandLine(CommandLineUtils.java:146) 4XESTACKTRACE at org/apache/maven/surefire/booter/shade/org/codehaus/plexus/util/cli/CommandLineUtils.executeCommandLine(CommandLineUtils.java:98) 4XESTACKTRACE at org/apache/maven/surefire/booter/SurefireBooter.fork(SurefireBooter.java:673) 4XESTACKTRACE at org/apache/maven/surefire/booter/SurefireBooter.forkSuites(SurefireBooter.java:479) 4XESTACKTRACE at org/apache/maven/surefire/booter/SurefireBooter.runSuitesForkOnce(SurefireBooter.java:379) 4XESTACKTRACE at org/apache/maven/surefire/booter/SurefireBooter.run(SurefireBooter.java:245) 4XESTACKTRACE at org/apache/maven/plugin/surefire/SurefirePlugin.execute(SurefirePlugin.java:537) 4XESTACKTRACE at org/apache/maven/plugin/DefaultPluginManager.executeMojo(DefaultPluginManager.java:490) 4XESTACKTRACE at org/apache/maven/lifecycle/DefaultLifecycleExecutor.executeGoals(DefaultLifecycleExecutor.java:694) 4XESTACKTRACE at org/apache/maven/lifecycle/DefaultLifecycleExecutor.executeGoalWithLifecycle(DefaultLifecycleExecutor.java:556) 4XESTACKTRACE at org/apache/maven/lifecycle/DefaultLifecycleExecutor.executeGoal(DefaultLifecycleExecutor.java:535) 4XESTACKTRACE at org/apache/maven/lifecycle/DefaultLifecycleExecutor.executeGoalAndHandleFailures(DefaultLifecycleExecutor.java:387) 4XESTACKTRACE at org/apache/maven/lifecycle/DefaultLifecycleExecutor.executeTaskSegments(DefaultLifecycleExecutor.java:348) 4XESTACKTRACE at org/apache/maven/lifecycle/DefaultLifecycleExecutor.execute(DefaultLifecycleExecutor.java:180) 4XESTACKTRACE at org/apache/maven/DefaultMaven.doExecute(DefaultMaven.java:328) 4XESTACKTRACE at org/apache/maven/DefaultMaven.execute(DefaultMaven.java:138) 4XESTACKTRACE at org/apache/maven/cli/MavenCli.main(MavenCli.java:362(Compiled Code)) 4XESTACKTRACE at org/apache/maven/cli/compat/CompatibleMain.main(CompatibleMain.java:60) 4XESTACKTRACE at sun/reflect/NativeMethodAccessorImpl.invoke0(Native Method) 4XESTACKTRACE at
Re: [jira] Reopened: (UIMA-1462) SimpleUimaAsService has checked in SimpleServer libraries as binaries
Tommaso Teofili wrote: Hi Marshall, Lookin in the SandboxDistr POM I see SimpleUimaAsService is commented because SimpleUimaAsService needs uima-as for building and uima-as is not in the lifecycle of Sandbox, so finally SimpleUimaAsService was left out to avoid the Sandbox building failure. For this reason I did not put the parent SandboxDistr POM inside SimpleUimaAsService. hmmm. I had left out the SimpleUimaAsService from the modules section originally in the mistaken belief that I needed to do that to avoid including it in the sandbox release. (That's not required, because the sandbox release content is set just by the assembly descriptors). But, in any case, you can set the parent of this to be the SandboxDistr, independently of whether or not the SimpleUimaAsService is in the module aggregation. Then, you can delete all of the plugins in the build section :-) , and it will still build, nicely. There's some issue we'll need to dig through, though, because it pulls in a large number of jars related to activemq (I think), not all of which would need to be distributed. Note: this is not being released with our 2.3.0 release per previous discussions which indicated it needed to be documented first. -Marshall Tommaso 2009/10/2 Marshall Schor m...@schor.com Tommaso Teofili (JIRA) wrote: [ https://issues.apache.org/jira/browse/UIMA-1462?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel] Tommaso Teofili reopened UIMA-1462: --- some plugins useful for licensing purposes were wrongly removed from pom Hi Tommaso - This POM appears to be missing the parent link. The POM parents add a lot of common build / dependency behavior; can you add that link, and then the much simpler POM will maybe work :-) -Marshall SimpleUimaAsService has checked in SimpleServer libraries as binaries - Key: UIMA-1462 URL: https://issues.apache.org/jira/browse/UIMA-1462 Project: UIMA Issue Type: Bug Components: Sandbox-SimpleUimaAsService Affects Versions: 2.2.2S, 2.3S Reporter: Thilo Goetz Assignee: Tommaso Teofili Fix For: 2.3S Attachments: patch1462.txt, patch1462b1.txt, patch1462b2.txt, patch1462c1.txt, patch1462c2.txt, simpleServerDistributionPatch.txt, simpleUimaAsServicePatch.txt For the upcoming release, this should be changed so it depends on the SimpleServer project directly.
[jira] Commented: (UIMA-1593) AEs that declare multipleDeploymentAllowed=false can't be deployed as services and called from multiple clients in the same JVM
[ https://issues.apache.org/jira/browse/UIMA-1593?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanelfocusedCommentId=12761657#action_12761657 ] Marshall Schor commented on UIMA-1593: -- If we override this value in the service wrapper in all cases, does that break anything? If not, I would be +1 for doing this in the service wrapper. AEs that declare multipleDeploymentAllowed=false can't be deployed as services and called from multiple clients in the same JVM --- Key: UIMA-1593 URL: https://issues.apache.org/jira/browse/UIMA-1593 Project: UIMA Issue Type: Bug Components: Async Scaleout, Core Java Framework Affects Versions: 2.2.2 Reporter: Adam Lally If an AnalysisEngine descriptor declares multipleDeploymentAllowed=false, the intention is that there cannot be multiple instances of it in the same JVM. UIMA reports an error if you try to embed such an AE in an aggregate that declares multipleDeploymentAllowed=true. But this only makes sense for co-located components, not services. Currently, the service wrapper returns the multipleDeploymentAllowed=false as part of the getMetaData() call. The aggregate AE checks this metadata for consistency, and at that point it isn't aware that it came from a service, so the exception is thrown. Perhaps the service wrapper should be modified to set mutlipleDeploymentAllowed to true when responding to all getMetaData() calls? Alternatively it could be done on the client side, but to me it makes more sense for the service to do it - why expose this setting if clients have no use for it? -- This message is automatically generated by JIRA. - You can reply to this email to add a comment to the issue online.
Cas Editor Eclipse naming
I was testing the Cas Editor plugin, as a new naive user. There is nothing in the documentation that says to create one or more new projects, of a special type, for use by the Cas Editor - so I was trying to use a plain Java project, and of course, nothing was working. Part of the problem is that the label for the Cas Editor things in Eclipse is NLP. I did not recognize or associate that with either UIMA or the Cas Editor. (So, for instance, if you say in Eclipse, menu - file - new - other you get a pick list with lots of things, including one labeled NLP. Same with menu - window - show view - other same NLP entry.) I think we should rename this categorization tag to something that uses the words Uima, and CasEditor - maybe it should be a 2 level categorization, the top one being UIMA (which already exists, and is used by the Component Descriptor Editor), and a then CasEditor. WDYT? (What do you think)? -Marshall
[jira] Closed: (UIMA-1422) Improve/fix build process for Eclipse Plugins
[ https://issues.apache.org/jira/browse/UIMA-1422?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ] Marshall Schor closed UIMA-1422. Resolution: Fixed Assignee: Marshall Schor Cas Editor plugin working now. Improve/fix build process for Eclipse Plugins - Key: UIMA-1422 URL: https://issues.apache.org/jira/browse/UIMA-1422 Project: UIMA Issue Type: Bug Components: Eclipse plugins Reporter: Marshall Schor Assignee: Marshall Schor Priority: Minor Fix For: 2.3 Eclipse update site build not working, parts seem broken. See http://markmail.org/thread/5m6kzwczxlqdyknh -- This message is automatically generated by JIRA. - You can reply to this email to add a comment to the issue online.
[jira] Updated: (UIMA-1333) JCasFlow_ImplBase.setJCas() should be deprecated, and this action always done by framework
[ https://issues.apache.org/jira/browse/UIMA-1333?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ] Marshall Schor updated UIMA-1333: - Summary: JCasFlow_ImplBase.setJCas() should be deprecated, and this action always done by framework (was: JCasFlow_ImpleBase.setJCas() should be deprecated, and this action always done by framework) JCasFlow_ImplBase.setJCas() should be deprecated, and this action always done by framework -- Key: UIMA-1333 URL: https://issues.apache.org/jira/browse/UIMA-1333 Project: UIMA Issue Type: Improvement Components: Core Java Framework Reporter: Eddie Epstein Priority: Minor Everytime I create a new FC to use this.getJCas() in the next() method the program crashes because the JCas had not been previously set. Can anyone explain why it is not right to have the framework automatically make the association between flow object and specific CAS whenever a flow object is returned from computeFlow and from newCasProduced methods? -- This message is automatically generated by JIRA. - You can reply to this email to add a comment to the issue online.
[jira] Closed: (UIMA-1333) JCasFlow_ImplBase.setJCas() should be deprecated, and this action always done by framework
[ https://issues.apache.org/jira/browse/UIMA-1333?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ] Marshall Schor closed UIMA-1333. Resolution: Fixed Fix Version/s: 2.3 JCasFlow_ImplBase.setJCas() should be deprecated, and this action always done by framework -- Key: UIMA-1333 URL: https://issues.apache.org/jira/browse/UIMA-1333 Project: UIMA Issue Type: Improvement Components: Core Java Framework Reporter: Eddie Epstein Priority: Minor Fix For: 2.3 Everytime I create a new FC to use this.getJCas() in the next() method the program crashes because the JCas had not been previously set. Can anyone explain why it is not right to have the framework automatically make the association between flow object and specific CAS whenever a flow object is returned from computeFlow and from newCasProduced methods? -- This message is automatically generated by JIRA. - You can reply to this email to add a comment to the issue online.
[jira] Closed: (UIMA-1390) UIMA should include a User Library containing all the necessary jars for Eclipse projects
[ https://issues.apache.org/jira/browse/UIMA-1390?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ] Marshall Schor closed UIMA-1390. Resolution: Won't Fix an alternative was done in UIMA-857 UIMA should include a User Library containing all the necessary jars for Eclipse projects -- Key: UIMA-1390 URL: https://issues.apache.org/jira/browse/UIMA-1390 Project: UIMA Issue Type: Improvement Reporter: Burn Lewis Priority: Minor Attachments: UIMA-1390_uima-as-distr.patch, UIMA-1390_uimaj-distr.patch We could include a uima.userlibraries file containing all the jars needed by projects using UIMA ... would avoid the need to add the jars one by one when setting up the classpath ... simply import this file and then add the library to the project's build path. It could contain separate sets for core UIMA and UIMA-AS. The adjustExamplePaths script could correct its absolute paths. -- This message is automatically generated by JIRA. - You can reply to this email to add a comment to the issue online.
[jira] Created: (UIMA-1598) make CasManager_impl thread-safe
make CasManager_impl thread-safe Key: UIMA-1598 URL: https://issues.apache.org/jira/browse/UIMA-1598 Project: UIMA Issue Type: Improvement Components: Core Java Framework Affects Versions: 2.3 Reporter: Marshall Schor Priority: Minor Because it is not thread-safe, users have to remember to serialize on using its services. See UIMA-1400 for instance. Defer past 2.3.0 release. -- This message is automatically generated by JIRA. - You can reply to this email to add a comment to the issue online.
[jira] Updated: (UIMA-1249) CasManager_impl logic for lazy merge of the type system could lead to excessive work or missed errors
[
https://issues.apache.org/jira/browse/UIMA-1249?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
]
Marshall Schor updated UIMA-1249:
-
Affects Version/s: (was: 2.2.2)
2.3
defer past 2.3.0
CasManager_impl logic for lazy merge of the type system could lead to
excessive work or missed errors
-
Key: UIMA-1249
URL: https://issues.apache.org/jira/browse/UIMA-1249
Project: UIMA
Issue Type: Improvement
Affects Versions: 2.3
Reporter: Marshall Schor
Priority: Minor
The CasManager_impl class is sometimes (mostly?) (but not always) used when
assembling a pipeline of UIMA components.
There is one instance of this associated with each Resource Manager instance.
(PearWrapper versions of the ResourceManager share this component).
It has 2 phases. At initialization, its method {{addMetaData}} is
repeatedly called as a part of the initialization phase of components being
assembled to run under one ResourceManager instance, to collect all the
metadata from the components (e.g., their individual type systems, type
priorities, and index definitions).
At the first call that requires the merged result, e.g.
{{getCasDefinition()}}, the class merges all the collected metadata and uses
it to produce the CAS's type system, indexes, etc.
After this first call, additional calls to {{addMetaData}} which attempt to
add new things not already in the merge, should result in an error.
In the current implementation, the call to {{addMetaData}} in this case not
only won't result in any error, but it will reset the class instance, so that
a subsequent call to get the CAS definition will result in merging being
again called, and a new, non-identical merged result will be returned. This
could result in CASes in a pool, for instance, having different type systems.
Normally, this sequence will never happen; however, in the multi-threaded
case, where initialization and processing could occur at the same time across
multiple instances, it could happen that {{addMetaData}} could be called by a
thread still initializing, while another thread has already obtained the
final merged CAS definition. In these cases, the {{addMetaData}} call
could be ignored, but in the general case, one would need to check to see
if the metaData being added would change the existing type system, and throw
an error if it did.
--
This message is automatically generated by JIRA.
-
You can reply to this email to add a comment to the issue online.
[jira] Updated: (UIMA-1113) DocumentAnalyzer falls down when running pear specifier and html result view
[ https://issues.apache.org/jira/browse/UIMA-1113?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ] Marshall Schor updated UIMA-1113: - Affects Version/s: (was: 2.2.2) 2.3 Fix Version/s: (was: 2.3) defer past 2.3.0 DocumentAnalyzer falls down when running pear specifier and html result view Key: UIMA-1113 URL: https://issues.apache.org/jira/browse/UIMA-1113 Project: UIMA Issue Type: Bug Components: Tools Affects Versions: 2.3 Reporter: Thilo Goetz Priority: Minor Throws an invalid XML exception. -- This message is automatically generated by JIRA. - You can reply to this email to add a comment to the issue online.
[jira] Closed: (UIMA-985) Javdocs for getSofaData calls do not say null may be returned
[ https://issues.apache.org/jira/browse/UIMA-985?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ] Marshall Schor closed UIMA-985. --- Resolution: Fixed Fix Version/s: 2.3 Assignee: Marshall Schor did the update by hand, because patch was no longer valid Javdocs for getSofaData calls do not say null may be returned - Key: UIMA-985 URL: https://issues.apache.org/jira/browse/UIMA-985 Project: UIMA Issue Type: Bug Components: Core Java Framework Reporter: Burn Lewis Assignee: Marshall Schor Priority: Minor Fix For: 2.3 Attachments: UIMA-985.patch getDocumentText() getSofaDataString() getSofaDataArray() getSofaURI() getSofaDataStream() should say null is returned if sofa data not set or set to a different type. Should we throw an exception if no sofa data not set? -- This message is automatically generated by JIRA. - You can reply to this email to add a comment to the issue online.
[jira] Updated: (UIMA-999) Should have a way to ignore or omit the imports in an aggregate that will be overridden by the deployment descriptor
[ https://issues.apache.org/jira/browse/UIMA-999?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ] Marshall Schor updated UIMA-999: Affects Version/s: 2.3AS defer past 2.3.0 Should have a way to ignore or omit the imports in an aggregate that will be overridden by the deployment descriptor Key: UIMA-999 URL: https://issues.apache.org/jira/browse/UIMA-999 Project: UIMA Issue Type: Bug Components: Async Scaleout Affects Versions: 2.3AS Reporter: Burn Lewis Priority: Minor Entries in the deployment descriptor for a delegate override the import specified in the aggregate descriptor for that delegate ... but the import must still be present and valid. It would be nice to have a way to indicate that such imports will be overriden and should be ignored. -- This message is automatically generated by JIRA. - You can reply to this email to add a comment to the issue online.
[jira] Commented: (UIMA-1149) [UIMA-AS] fix potential thread safety issue with flow controller code
[ https://issues.apache.org/jira/browse/UIMA-1149?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanelfocusedCommentId=12761731#action_12761731 ] Burn Lewis commented on UIMA-1149: -- Is synchronization also needed around the FlowContainer next() call in executeFlowStep in AggregateAnalysisEngineController_impl.java ? [UIMA-AS] fix potential thread safety issue with flow controller code - Key: UIMA-1149 URL: https://issues.apache.org/jira/browse/UIMA-1149 Project: UIMA Issue Type: Bug Components: Async Scaleout Affects Versions: 2.2.2AS Reporter: Marshall Schor Priority: Minor Fix For: 2.3AS Attachments: uimaj-as-core-UIMA-1149-patch.txt The general contract regarding threading for user-written UIMA components is that components do not have to be written to be thread-safe, or, more precisely, the framework guarantees it will not call an instance of a user-written component class on multiple threads at the same time. See http://incubator.apache.org/uima/downloads/releaseDocs/2.2.2-incubating/docs/html/tutorials_and_users_guides/tutorials_and_users_guides.html#ugr.tug.aae.contract_for_annotator_methods This is intended to make writing UIMA components by our user community easier, and to eliminate many hard-to-diagnose issues that could occur otherwise. The FlowController main object class, the one that is specified in the custom flow control specification for an aggregate, in UIMA-AS, can be (we think, but to be investigated :-) ) be called on multiple threads at the same time, in the current implementation. This Jira is to fix this, by synchronizing on the flow controller object so that users don't have to write thread safe implementations of this class. Also, the docs in the above referenced manual should be updated to include the flow controller object in the set of objects where the framework guarantees that instances of this object will not be called on multiple threads at the same time. -- This message is automatically generated by JIRA. - You can reply to this email to add a comment to the issue online.
[jira] Updated: (UIMA-1065) CFE - configurable feature extrator for UIMA annotation comparison, evaluation, testing, generation of machine learning features
[ https://issues.apache.org/jira/browse/UIMA-1065?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ] Igor Sominsky updated UIMA-1065: Attachment: CFE-patch-20091002.txt Code fixes after running through FindBug tool CFE - configurable feature extrator for UIMA annotation comparison, evaluation, testing, generation of machine learning features Key: UIMA-1065 URL: https://issues.apache.org/jira/browse/UIMA-1065 Project: UIMA Issue Type: New Feature Components: Tools Reporter: Igor Sominsky Assignee: Marshall Schor Priority: Minor Attachments: CFE-20080908.zip, CFE-20080908.zip.md5, CFE-patch-20090827.txt, CFE-patch-20090828.txt, CFE-patch-20090930.txt, CFE-patch-20091002.txt, CFE_sominsky-A4.pdf, CFE_UG-images.zip, CFE_UG_v9.pdf, images.zip CFE - configurable feature extrator for UIMA annotation comparison, evaluation of performance metrics such as precision/recall/f-score, regression testing, generation of machine learning features. It provides Feature Extraction Specification language (FESL), that specifies rules for feature extraction. The criteria for the extraction could be customized by multiple conditions written in FESL in normal disjunctive form. Extracted features can be output into a file or CAS. Extractor can be implemented as CAS consumer of TAE -- This message is automatically generated by JIRA. - You can reply to this email to add a comment to the issue online.
[jira] Commented: (UIMA-1153) thread safety issue with sample flow controller AdvancedFixedFlowController
[ https://issues.apache.org/jira/browse/UIMA-1153?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanelfocusedCommentId=12761732#action_12761732 ] Burn Lewis commented on UIMA-1153: -- A nice solution ... we should change the FC example ... but may be hard to test .. can FindBugs help us? thread safety issue with sample flow controller AdvancedFixedFlowController --- Key: UIMA-1153 URL: https://issues.apache.org/jira/browse/UIMA-1153 Project: UIMA Issue Type: Bug Components: Async Scaleout, Examples Affects Versions: 2.2.2 Reporter: Marshall Schor Assignee: Burn Lewis Priority: Minor The AdvancedFixedFlowController implements its Flow Object class as an inner class of its Flow Controller class. In UIMA-AS, it is possible that more than one Flow Object can be accessing the Flow Controller fields on different threads, and (for Cas Multipliers) it is possible for an individual Flow Object to be called on multiple threads at the same time (via the method (via the newCasProduced method). In the example, the newCasProduced method is marked *synchronized*, so I think the same Flow Object will not be running multiple threads at the same time, because the other methods to this are all calls done on 1 thread on behalf of the main CAS associated with this Flow Object. However, multiple flow objects running on different threads could have their newCasProduced method called. In this case, the references in that method to common mutable fields in the shared Flow Controller object are not synchronized. This code should be fixed so that those references are synchronized on the same monitor used to synchronize other field access in the Flow Controller. UIMA-1149 I hope will use the class instance of the flow controller for this purpose, and, assuming that it does, this code should explicitly synchronize on that same object, for references to mutable fields in that class. There is one field, mSequence, which is in the associated Flow Controller class, which is mutable (in theory). References to this field should be synchronized with the same monitor used by the flow controller class when manipulating it. The other field references in the flow object should be checked, and synchronized if they refer to mutable Flow Controller fields. -- This message is automatically generated by JIRA. - You can reply to this email to add a comment to the issue online.
Re: [jira] Reopened: (UIMA-1462) SimpleUimaAsService has checked in SimpleServer libraries as binaries
Il giorno 02/ott/2009, alle ore 19.21, Marshall Schor ha scritto: hmmm. I had left out the SimpleUimaAsService from the modules section originally in the mistaken belief that I needed to do that to avoid including it in the sandbox release. (That's not required, because the sandbox release content is set just by the assembly descriptors). But, in any case, you can set the parent of this to be the SandboxDistr, independently of whether or not the SimpleUimaAsService is in the module aggregation. Ok, it's right. I was in doubt so I did not add the parent. Then, you can delete all of the plugins in the build section :-) , and it will still build, nicely. There's some issue we'll need to dig through, though, because it pulls in a large number of jars related to activemq (I think), not all of which would need to be distributed. Yes, the large amount of jars is related to activemq; to avoid this we could set activemq scope to 'provided' or use the exclusion tag to import only wanted transitive dependencies. Note: this is not being released with our 2.3.0 release per previous discussions which indicated it needed to be documented first. -Marshall Tommaso
