May 24, 2011 
< | 1 | 2 | 3 | 4 | 5 | 6 | 7 | 8 | 9 | 10 | 11 | 12 | 13 | 14 | 15 | 16 | 17 | 18 | 19 | 20 | 21 | 22 | 23 | 24 | 25 | 26 | 27 | 28 | 29 | 30 | 31 | >
May 24, 2011 NOTICE: This channel is no longer actively logged.
[00:02:17] *** emuckenhuber has quit IRC[00:06:01] <dmlloyd> if it's blocking... a certain compliance test whose identity I will not reveal..., it's a priority[00:07:00] *** kevinpollet has quit IRC[00:07:41] *** kevinpollet has joined #jboss-as7[00:07:51] *** kevinpollet has quit IRC[00:10:55] *** jwulf has joined #jboss-as7[00:11:49] *** opalka has quit IRC[00:37:41] *** tcrawley has quit IRC[00:38:57] *** tcrawley has joined #jboss-as7[00:38:57] *** ChanServ sets mode: +v tcrawley[00:42:27] <dmlloyd> anyone else getting checksum errors downloading BCEL during AS build?[00:42:41] <dmlloyd> I've cleared it from my repos a couple of times[00:48:38] *** clebert has quit IRC[00:50:24] <dmlloyd> pgier: ping..[00:51:30] *** slaboure has quit IRC[00:52:11] *** pferraro has quit IRC[00:53:50] *** jpearlin has joined #jboss-as7[00:55:10] *** sannegrinovero has quit IRC[01:14:08] *** asaldhan has left #jboss-as7[01:15:45] *** smarlow has joined #jboss-as7[01:15:45] *** ChanServ sets mode: +v smarlow[01:19:27] *** miclorb_ has joined #jboss-as7[01:25:16] *** jhalliday has joined #jboss-as7[01:28:50] *** bgeorges has quit IRC[01:29:31] *** pferraro has joined #jboss-as7[01:29:31] *** ChanServ sets mode: +v pferraro[01:29:49] *** alexsmirnov has quit IRC[01:31:51] <jhalliday> is there a client .jar for AS7?[01:33:48] *** aslak has quit IRC[01:34:56] *** ccrouch has quit IRC[01:35:13] *** jamezp is now known as jamezp_afk[01:44:25] *** bstansberry has joined #jboss-as7[01:44:25] *** ChanServ sets mode: +v bstansberry[01:44:57] <dmlloyd> not yet that I know of[01:48:23] <jhalliday> does 'not yet' imply there is a plan to have one eventually?[01:50:07] <dmlloyd> I think there was talk about having a single client JAR[01:50:45] <jhalliday> any talk of what may be in it?[01:56:54] *** frainone has quit IRC[02:05:59] <dmlloyd> not sure, seems like it'd have to contain a lot of EE APIs though...[02:06:12] <dmlloyd> not sure what the client end of most of our services look like[02:20:07] <dmlloyd> stuartdouglas: http://github.com/dmlloyd/jboss-as/commit/6045821[02:20:08] <jbossbot> git [jboss-as] 6045821.. David M. Lloyd Exclusions for Xalan[02:22:06] <stuartdouglas> I love maven's pom.xml format[02:24:30] <stuartdouglas> 264 lines just to exclude two artifacts[02:26:18] *** jamezp_afk is now known as jamezp[02:26:40] <dmlloyd> no kidding[02:27:02] <dmlloyd> it will inevitably collapse under its own bulk eventually[02:27:14] * dmlloyd hopes pgier has us switched over to gradle by then :)[02:27:58] <stuartdouglas> I have played around with gradle a bit, it seemed much better, but still a bit immature[02:28:07] <stuartdouglas> or maybe I was just doing it wrong[02:28:14] <misty> hibernate has switched to gradle[02:28:18] <dmlloyd> yeah it has a bit of growing up to do[02:28:34] <dmlloyd> but the architecture looks good at my first readthrough[02:28:56] <stuartdouglas> it is heaps more flexible than maven, and a lot less verbose[02:29:22] <stuartdouglas> you can do things that are a bit unusual without having to write a custom plugin for one thing :-)[02:31:55] *** adinn has joined #jboss-as7[02:31:56] *** ChanServ sets mode: +v adinn[02:32:45] <jamezp> dmlloyd: Do we want to convert the messages in this class? https://github.com/jbossas/jboss-as/blob/master/controller/src/main/java/org/jboss/as/controller/registry/FastCopyHashMap.java[02:32:50] *** adinn has left #jboss-as7[02:33:30] *** adinn has joined #jboss-as7[02:33:32] *** ChanServ sets mode: +v adinn[02:36:34] <dmlloyd> no, jamezp[02:37:05] <jamezp> Sweet, the controller message bundle is rather large :-)[02:38:11] *** magesh has joined #jboss-as7[02:39:12] *** magesh1 has quit IRC[02:53:56] *** Binbin has joined #jboss-as7[03:12:50] <dmlloyd> test failures now stuartdouglas[03:13:03] <stuartdouglas> which ones?[03:13:05] *** sgilda has quit IRC[03:13:18] <dmlloyd> org.jboss.as.arquillian.container.managed.IntegrationTestCase[03:13:28] <dmlloyd> (arq2)[03:13:32] <stuartdouglas> do you have a jbas process running in the background?[03:13:39] <dmlloyd> I don't think so...[03:13:48] <dmlloyd> no[03:14:43] <stuartdouglas> hmm, I have only ever seen that one fail when I left an AS instance running[03:15:06] <dmlloyd> I can build again I guess[03:15:28] <stuartdouglas> is this my master rebased against upstream? Or just my master?[03:15:29] <dmlloyd> might it have failed by doing a different project build at the same time?[03:15:31] <dmlloyd> like some socket[03:15:40] <dmlloyd> it's your master rebased on upstream plus the exclusions patch[03:16:05] <stuartdouglas> can you push it out? I will see if I get the same failure[03:16:25] <dmlloyd> okay it's in mid build but if it fails again I will push it out[03:17:26] <dmlloyd> failed again, same spot[03:17:49] <dmlloyd> it's pushed to my master[03:18:55] <stuartdouglas> ok, I'll give it a go[03:24:03] *** magesh1 has joined #jboss-as7[03:25:07] *** magesh has quit IRC[03:25:07] <stuartdouglas> built fine for me[03:25:25] <dmlloyd> wtf.[03:25:36] <stuartdouglas> what is the output of the test failure?[03:25:41] <dmlloyd> if I have a messed up m2 repos I'm going to be pissed off[03:26:02] <dmlloyd> http://fpaste.org/J4kg/[03:26:25] <dmlloyd> I love fpaste. In fedora you just do "fpaste arquillian2/container-managed/target/surefire-reports/org.jboss.as.arquillian.container.managed.IntegrationTestCase.txt" and it gives you the URL[03:26:26] <stuartdouglas> what about the -output.txt file?[03:26:35] <dmlloyd> http://fpaste.org/i2xc/[03:27:05] <dmlloyd> that looks kinda serious tbh[03:27:08] <stuartdouglas> thats interesting[03:27:27] <stuartdouglas> does not look related to my changes, does upstream master build fine for you atm?[03:27:43] <stuartdouglas> it could be something from carlo's changes maybe[03:27:46] * stuartdouglas looks[03:27:46] <dmlloyd> last time I built upstream was friday I think[03:28:08] <dmlloyd> oh shit[03:28:10] <dmlloyd> check this out[03:28:10] <dmlloyd> haha[03:28:15] <dmlloyd> 20:19:09,208 INFO [org.jboss.as.arquillian.container.managed.JBossAsManagedContainer] (main) Starting container with: [/home/david/local/jdk/ibm-jdk-1.6.0-x86_64/bin/java, -Xmx512m, -XX:MaxPermSize=128m, -Djboss.home.dir=target/jbossas, -Dorg.jboss.boot.log.file=target/jbossas/standalone/log/boot.log, -Dlogging.configuration=file:target/jbossas/standalone/configuration/logging.properties, -jar, /home/david/src/java/as7/merge/arquillian2/contain[03:28:15] <dmlloyd> er-managed/target/jbossas/jboss-modules.jar, -mp, target/jbossas/modules, -logmodule, org.jboss.logmanager, -jaxpmodule, javax.xml.jaxp-provider, org.jboss.as.standalone][03:28:19] <dmlloyd> read that line carefully[03:28:31] <stuartdouglas> IBM JDK?[03:28:40] <dmlloyd> yeah, I was testing AS startup on different JDKs[03:28:45] <dmlloyd> forgot I left IBM on ;|[03:28:53] <dmlloyd> still it's good to know, something is fishy with IBM![03:29:10] <dmlloyd> no wonder maven was suddenly hitting OOME[03:29:15] <stuartdouglas> 1.6.0 ? is that the latest one ?[03:29:24] <dmlloyd> I dunno, I downloaded that ages ago[03:30:06] <dmlloyd> man no wonder stuff has been acting weird[03:30:08] <dmlloyd> sheesh[03:35:19] <jamezp> Knowing IBM that probably is the latest.[03:36:09] <jbossbot> git [jboss-as] push master ab81002.. Carlo de Wolf JBAS-9181: an implementation of EJB 3.1 Embeddable[03:36:10] <jbossbot> jira [JBAS-9181] Redirected to: https://issues.jboss.org/si/jira.issueviews:issue-xml/AS7-434/AS7-434.xml[03:36:11] <jbossbot> jira [AS7-434] Implement EJB 3.1 FR 22 Embeddable Usage [Open (Unresolved) Task, Critical, Carlo de Wolf] https://issues.jboss.org/browse/AS7-434[03:36:11] <jbossbot> git [jboss-as] push master 7bdc128.. Carlo de Wolf JBAS-9181: fix NPE on non-existent path elements[03:36:12] <jbossbot> git [jboss-as] push master 5cd8ec3.. Carlo de Wolf JBAS-9181: ignore empty class path entries[03:36:12] <jbossbot> git [jboss-as] push master ecd56d8.. Carlo de Wolf Increase startup timeout and always destroy the process if timed out[03:36:12] <jbossbot> git [jboss-as] push master 34826d4.. Carlo de Wolf AS7-434: work-around a jboss-modules bug[03:36:12] <jbossbot> git [jboss-as] push master b922e07.. Carlo de Wolf AS7-434: bring up EJB3 Embeddable with an existing ModuleLoader[03:36:12] <jbossbot> git [jboss-as] push master 4ec30db.. Carlo de Wolf AS7-434: properly set the dependencies of jboss-as-embedded[03:36:13] <jbossbot> git [jboss-as] push master 0976a25.. Carlo de Wolf AS7-434: add classpath to deployment if running within classpath modular enviroment (requires exp-ejbemb)[03:36:13] <jbossbot> git [jboss-as] push master 61a01a0.. Carlo de Wolf AS7-485: make sure jsfunit-aquillian does not end up on the test class path and disable JSFTestCase[03:36:14] <jbossbot> jira [AS7-485] Fix the testsuite/integration testcase failures [Open (Unresolved) Task, Major, Stan Silvert] https://issues.jboss.org/browse/AS7-485[03:36:14] <jbossbot> git [jboss-as] push master 1123506.. Carlo de Wolf AS7-835: refactored EjbAnnotationProcessor into EJBComponentDescriptionFactory[03:36:15] <jbossbot> jira [AS7-835] ejb-jar.xml doesn't need to define a session-type [Open (Unresolved) Bug, Critical, Carlo de Wolf] https://issues.jboss.org/browse/AS7-835[03:36:15] <jbossbot> git [jboss-as] push master c4edbc1.. Stuart Douglas Initial implementation of default interceptors[03:36:16] <jbossbot> git [jboss-as] push master 3036f95.. Stuart Douglas WIP interceptor ordering[03:36:16] <jbossbot> git [jboss-as] push master d91d07f.. Stuart Douglas WIP interceptors[03:36:17] <jbossbot> git [jboss-as] push master a3a3ec2.. Stuart Douglas WIP intercptors[03:36:17] <jbossbot> git [jboss-as] push master e78bb11.. Stuart Douglas Fix CDI/EJB integration[03:36:18] <jbossbot> git [jboss-as] push master 6785d17.. Stuart Douglas Update to jboss-invocation snapshot[03:36:18] <jbossbot> git [jboss-as] push master 0cf8a4f.. Stuart Douglas Change the way duplicate bindings are detected...[03:36:19] <jbossbot> git [jboss-as] push master 19a745e.. Stuart Douglas Change the way interceptor methods declared in the deployment descriptor are handled[03:36:19] <jbossbot> git [jboss-as] push master 55d5f94.. Stuart Douglas Fix equals / hashCode for EJB views, and also implemented SessionContext.getBusinessObject[03:36:20] <jbossbot> git [jboss-as] push master 11c54c1.. Stuart Douglas Throw IllegalStateException as required by the spec[03:36:20] <jbossbot> git [jboss-as] push master 3437506.. Stuart Douglas Change the order of parsing so that @ApplicationException annotations are overriden by XML[03:36:21] <jbossbot> git [jboss-as] push master 549c9f9.. Stuart Douglas Add support for injecting the SessionContext through SessionBean.setSessionContext()[03:36:21] <jbossbot> git [jboss-as] push master 56b724b.. Stuart Douglas Add singleton BMT interceptor[03:36:22] <jbossbot> git [jboss-as] push master 043aba5.. Stuart Douglas Fix env-entry problems[03:36:22] <jbossbot> git [jboss-as] push master 39a64e9.. Stuart Douglas Change the web NamingListener to a valve[03:36:23] <jbossbot> git [jboss-as] push master 7e1b660.. Stuart Douglas Improve resource-env-ref injection from DD's[03:36:23] <jbossbot> git [jboss-as] push master d5937d1.. Stuart Douglas jboss-invocation version[03:36:24] <jbossbot> git [jboss-as] push master 792e3cc.. Carlo de Wolf Using jboss-metadata-ejb-7.0.0.Beta2-6-g0f72d3d[03:36:24] <jbossbot> git [jboss-as] push master 6941745.. Carlo de Wolf AS7-835: move lookup to common.Naming[03:36:25] <jbossbot> git [jboss-as] push master e492d6a.. Carlo de Wolf AS7-859: find back the callback method using the correct ClassReflectionIndex[03:36:25] <jbossbot> jira [AS7-859] On same named private lifecycle callbacks the super class callback is not called [Open (Unresolved) Bug, Critical, Carlo de Wolf] https://issues.jboss.org/browse/AS7-859[03:36:26] <jbossbot> git [jboss-as] push master f8bfa8f.. Carlo de Wolf AS7-835: work-around JBMETA-332[03:36:26] <jbossbot> jira [JBMETA-332] EjbJarMetaData.getEnterpriseBean(String) throws NPE when enterpriseBeans is null [Open (Unresolved) Bug, Major, Unassigned] https://issues.jboss.org/browse/JBMETA-332[03:36:27] <jbossbot> git [jboss-as] push master b442171.. Carlo de Wolf Re-enabled StatefulSessionSynchronizationInterceptor[03:36:27] <jbossbot> git [jboss-as] push master 2f61cf9.. Carlo de Wolf Make sure the StatefulSessionSynchronizationInterceptor gets the proper config[03:36:28] <jbossbot> git [jboss-as] push master 782a860.. Stuart Douglas fix @Resource injection for managed beans[03:36:28] <jbossbot> git [jboss-as] push master 6045821.. David M. Lloyd Exclusions for Xalan[03:36:29] <jbossbot> git [jboss-as] push master URL: http://github.com/jbossas/jboss-as/compare/2892f86...6045821[03:37:48] <stuartdouglas> yay[03:38:26] <dmlloyd> I wonder why IBM triggered all those xalan things[03:38:31] <dmlloyd> actually I probably don't want to know[03:38:54] <jamezp> No you don't, FWIW though GlassFish has xalan issues with the J9 JVM.[03:39:21] <jamezp> I've never been real successful at getting anything but WebSphere to run on the IBM JDK.[03:39:39] <dmlloyd> AS7 seems to run on it well enough[03:39:52] <stuartdouglas> used to[03:39:56] <dmlloyd> they actually don't have the stupid classloader lock which gives us so much grief[03:40:07] <dmlloyd> it does, stuartdouglas, I don't know why this test fails[03:40:28] <stuartdouglas> it's probably something screwing the way the managed container starts the AS[03:40:30] <jamezp> That's actually good to know.[03:40:55] <dmlloyd> might be tccl-related[03:42:08] <dmlloyd> we start up in jrockit too, though you need the latest modules release else it crashes hard immediately :)[03:42:27] <dmlloyd> er latest snapshot[03:43:51] <stuartdouglas> I'm going to be up in the brisbane office thursday/friday this week, so I may not be around that much[03:46:06] *** jamezp is now known as jamezp_afk[03:48:29] *** bgeorges has joined #jboss-as7[03:48:29] *** ChanServ sets mode: +v bgeorges[03:51:04] *** lgao has joined #jboss-as7[03:55:08] *** smcgowan has joined #jboss-as7[03:55:08] *** ChanServ sets mode: +v smcgowan[03:57:33] *** bbrowning has quit IRC[04:11:19] *** pferraro has quit IRC[04:11:48] *** pferraro has joined #jboss-as7[04:11:48] *** ChanServ sets mode: +v pferraro[04:26:19] *** JimMa has joined #jboss-as7[04:28:39] *** stliu has joined #jboss-as7[04:28:39] *** ChanServ sets mode: +v stliu[04:34:26] *** JimMa has quit IRC[04:46:12] *** pgier has quit IRC[04:49:18] *** jpearlin has left #jboss-as7[04:54:49] *** stliu has quit IRC[05:27:50] *** magesh1 has quit IRC[05:30:05] *** smarlow has quit IRC[05:30:45] *** magesh has joined #jboss-as7[05:31:22] *** smcgowan has quit IRC[05:33:55] *** jc3 has quit IRC[05:46:36] *** jwulf has quit IRC[06:13:12] <stuartdouglas> dmlloyd: Still awake?[06:13:18] <dmlloyd> yep[06:13:37] <stuartdouglas> I have done an impl of interceptor ordering https://github.com/stuartwdouglas/jboss-as/compare/master[06:15:22] *** JimMa has joined #jboss-as7[06:16:03] <stuartdouglas> I am not sure if we need to do the same thing for configurators as well, but I would think not, as all most of them do is add interceptors[06:16:11] <stuartdouglas> so there order does not really matter now[06:16:16] *** pferraro has quit IRC[06:17:00] *** sguilhen has quit IRC[06:18:09] *** sguilhen has joined #jboss-as7[06:20:23] *** JimMa has quit IRC[06:46:33] *** asoldano has joined #jboss-as7[06:46:33] *** ChanServ sets mode: +v asoldano[06:59:31] <dmlloyd> yeah without interceptor ordering, configurator ordering doesn't matter[06:59:36] <dmlloyd> might as well be a list[07:03:18] <stuartdouglas> CDI TCK should be passing in web profile now[07:03:40] <stuartdouglas> It turns out that the 1 test that was failing is bogus, so I just excluded it[07:04:03] <stuartdouglas> it does seem like quite a few non-web profile tests get excluded by the web profile selector though[07:04:51] <dmlloyd> reviewed the patch, looks ok[07:04:54] <dmlloyd> still building though :([07:05:32] <dmlloyd> be sure to let jaikiran know that this is all done[07:05:37] <stuartdouglas> I just pushed another patch[07:05:51] <stuartdouglas> that should fix some CDI issues that aren't tested by the web profile for some reason[07:06:01] <stuartdouglas> yea, I will let him know[07:06:15] <jbossbot> git [jboss-as] push master 26d23aa.. Stuart Douglas AS7-834 Change component interceptors to use a well defined ordering[07:06:16] <jbossbot> jira [AS7-834] Implement a deterministic way of adding interceptors and configurators to a view and component [Open (Unresolved) Task, Major, Stuart Douglas] https://issues.jboss.org/browse/AS7-834[07:06:16] <jbossbot> git [jboss-as] push master URL: http://github.com/jbossas/jboss-as/compare/6045821...26d23aa[07:07:28] <dmlloyd> okay[07:13:40] *** mlinhard has joined #jboss-as7[07:13:40] *** ChanServ sets mode: +v mlinhard[07:26:05] <dmlloyd> okay I'm out for the night, have a good one all[07:26:16] <stuartdouglas> night[07:46:02] *** bstansberry has quit IRC[07:49:57] <nickarls> stuartdouglas: mornings. any pointers on where to start looking? http://pastebin.com/3yUHSQUw[07:50:04] <nickarls> seam, cdi, logging or something else?[07:50:10] <nickarls> upstream master 20 min ago[07:50:33] *** misty has left #jboss-as7[07:50:41] <stuartdouglas> looks like the static initalizer for LoggingProviders is failing[07:51:40] <stuartdouglas> stick a breakpoint of LoggerProvider.findProvider and see what the exception is[07:53:00] <nickarls> could it be the attachmentkey-stuff that is warned about?[07:53:39] <stuartdouglas> doubt it[07:53:52] <stuartdouglas> actually it probably is[07:54:15] <stuartdouglas> are you bundling jboss logging in your app?[07:54:39] *** lazarotti has quit IRC[07:55:55] <nickarls> not be but my buddy maven apparently is ;-) signature change apparently...[07:56:27] <stuartdouglas> we are talking about doing something to fix that, so the AS ignores bundled stuff that will break things[07:56:54] <nickarls> envers pulled in 3.0.0.Beta5[08:11:33] *** jfclere has joined #jboss-as7[08:11:33] *** ChanServ sets mode: +v jfclere[08:12:31] <nickarls> stuartdouglas: do you know of any specialization bugs on AS7: http://pastebin.com/PVkWTb6W[08:12:40] *** opalka has joined #jboss-as7[08:12:40] *** ChanServ sets mode: +v opalka[08:13:00] <nickarls> same thing worked on AS6. Strange that there would be a low-level Weld difference like that[08:13:21] <stuartdouglas> AS6 did a dodgey bean archive merging thing[08:13:25] <stuartdouglas> that could explain it[08:13:34] <stuartdouglas> I would need to look at the classes and how they are packaged[08:13:41] *** opalka has quit IRC[08:13:45] *** opalka has joined #jboss-as7[08:13:45] *** ChanServ sets mode: +v opalka[08:14:49] <nickarls> the base bean is ApplicationBundles from seam international and my class extends it directly (WAR packed archive)[08:15:16] *** jfclere has quit IRC[08:15:31] <stuartdouglas> does seam-int have a beans.xml?[08:16:32] *** jfclere has joined #jboss-as7[08:16:32] *** ChanServ sets mode: +v jfclere[08:17:05] *** stliu has joined #jboss-as7[08:17:15] *** ChanServ sets mode: +v stliu[08:17:46] *** JimMa has joined #jboss-as7[08:18:16] <nickarls> must have, it works ;-)[08:20:10] <stuartdouglas> hmm it could be some kind of bean visibility problem[08:20:23] <stuartdouglas> can you turn it into an ARQ test case?[08:20:28] <stuartdouglas> please :-)[08:20:49] <stuartdouglas> if not just file a JIRA and assign it to me[08:21:23] *** AndyTaylor has quit IRC[08:21:26] <nickarls> I can start with distilling it down to a one class war first and see how far I get after that ;-)[08:22:30] <stuartdouglas> you will probably need a bean archive in war/lib that is then specialised by a beans in war/classes[08:30:55] <nickarls> stuartdouglas: https://issues.jboss.org/browse/AS7-880[08:30:56] <jbossbot> jira [AS7-880] Specializing bean in bundled archive doesn't work [Open (Unresolved) Bug, Major, Stuart Douglas] https://issues.jboss.org/browse/AS7-880[08:31:59] *** torben has joined #jboss-as7[08:31:59] *** ChanServ sets mode: +v torben[08:32:48] *** mgoldmann has joined #jboss-as7[08:32:48] *** ChanServ sets mode: +v mgoldmann[08:38:52] *** tdiesler has joined #jboss-as7[08:38:52] *** ChanServ sets mode: +v tdiesler[08:40:56] *** adietisheim has joined #jboss-as7[08:43:32] *** emuckenhuber has joined #jboss-as7[08:43:32] *** ChanServ sets mode: +v emuckenhuber[08:51:45] *** kevinpollet has joined #jboss-as7[08:59:03] *** davidbos has joined #jboss-as7[09:06:35] *** emuckenhuber has quit IRC[09:08:49] *** mgoldmann has quit IRC[09:12:19] *** rmaucher has joined #jboss-as7[09:12:39] *** tdiesler has quit IRC[09:14:24] *** emuckenhuber has joined #jboss-as7[09:14:24] *** emuckenhuber has joined #jboss-as7[09:14:24] *** ChanServ sets mode: +v emuckenhuber[09:15:21] *** hardy_ has joined #jboss-as7[09:21:55] *** galderz has joined #jboss-as7[09:21:55] *** ChanServ sets mode: +v galderz[09:27:35] *** hardy_ has quit IRC[09:29:38] *** hardy_ has joined #jboss-as7[09:36:33] *** tdiesler has joined #jboss-as7[09:36:33] *** ChanServ sets mode: +v tdiesler[09:37:01] *** AndyTaylor has joined #jboss-as7[09:37:01] *** ChanServ sets mode: +v AndyTaylor[09:37:18] *** miclorb_ has quit IRC[09:42:58] *** kevinpollet has quit IRC[09:43:06] *** kevinpollet has joined #jboss-as7[09:43:16] *** jhalliday has quit IRC[09:43:18] *** slaboure has joined #jboss-as7[09:44:10] *** adinn has left #jboss-as7[09:53:45] *** wolfc has joined #jboss-as7[09:53:45] *** ChanServ sets mode: +v wolfc[10:02:19] *** ALR has quit IRC[10:10:01] <wolfc> jfclere, good morning[10:12:38] <jfclere> wolfc: good morning[10:13:35] <wolfc> jfclere: right now we have a dependency on a temp release of metadata, so I want to get a Beta3 a.s.a.p. stuartdouglas, anything that needs to go in there? jfclere, when can we get a Beta3?[10:14:08] <stuartdouglas> I have a minor one : https://github.com/stuartwdouglas/metadata[10:15:04] <jfclere> ok looking and releasing in the next minutes :D[10:16:16] <jfclere> new features ;-)[10:16:32] *** aslak has joined #jboss-as7[10:16:32] *** ChanServ sets mode: +v aslak[10:19:59] *** alesj has joined #jboss-as7[10:26:17] *** alesj has quit IRC[10:27:35] *** hbraun has joined #jboss-as7[10:27:46] *** alesj has joined #jboss-as7[10:30:13] *** bgeorges has quit IRC[10:33:21] *** lgao has quit IRC[10:37:25] *** magesh1 has joined #jboss-as7[10:37:42] <jfclere> stuartdouglas: hm you have more changes than just this one. Do you need all changes?[10:38:00] <jfclere> this one: https://github.com/stuartwdouglas/metadata/commit/eaa329ac5e8324f73b718bfbe07a9c156b05d793[10:38:01] <jbossbot> git [metadata] eaa329a.. Stuart Douglas Parse transaction synchronization methods[10:38:11] *** alesj has quit IRC[10:38:32] *** alesj has joined #jboss-as7[10:38:35] <stuartdouglas> that should be the only one, the rest should already be merged[10:38:55] *** magesh has quit IRC[10:39:02] <stuartdouglas> hmm, I'm not sure what happened there[10:39:07] <stuartdouglas> looks like a wonkey rebase[10:40:05] <jfclere> no pb I will just merge than one then :D[10:40:45] *** miclorb_ has joined #jboss-as7[10:52:29] *** kkhan has joined #jboss-as7[10:52:29] *** ChanServ sets mode: +v kkhan[10:59:23] *** opalka has quit IRC[11:11:46] <jfclere> wolfc: stuartdouglas: staging ready do you want to test?[11:12:00] <jfclere> or should I release now[11:12:05] <wolfc> huh, clean your mouth :-)[11:13:12] <wolfc> I've never figured out how we can pickup stuff from stages into Hudson jobs[11:23:34] <jfclere> wolfc: you have to use the staging repos instead http://repository.jboss.org/nexus/content/groups/public/[11:23:50] <jfclere> I have tried it once... It worked[11:24:38] <jfclere> but no pb I have released :D[11:24:43] <wolfc> yeah, but it would be a bitch to setup for jobs[11:25:11] <jfclere> releasing is one click: that is more easy :D[11:25:17] <jfclere> released :D[11:25:29] *** magesh1 has left #jboss-as7[11:33:14] <jbossbot> git [jboss-as] push master 23b4870.. Shelly McGowan update javax.mail.mail to version 1.4.4[11:33:14] <jbossbot> git [jboss-as] push master URL: http://github.com/jbossas/jboss-as/compare/26d23aa...23b4870[11:42:37] *** sannegrinovero has joined #jboss-as7[11:42:37] *** ChanServ sets mode: +v sannegrinovero[11:43:43] *** vtunka has joined #jboss-as7[11:43:43] *** ChanServ sets mode: +v vtunka[11:49:04] *** opalka has joined #jboss-as7[11:49:04] *** opalka has joined #jboss-as7[11:49:04] *** ChanServ sets mode: +v opalka[11:51:14] *** darranl has joined #jboss-as7[11:51:15] *** darranl has joined #jboss-as7[11:51:15] *** ChanServ sets mode: +v darranl[11:55:37] *** Jaikiran has joined #jboss-as7[11:55:37] *** ChanServ sets mode: +v Jaikiran[11:55:54] <Jaikiran> opalka: got a minute?[12:00:47] <wolfc> does anybody know how I can execute an operation without persisting it in the model?[12:01:52] <emuckenhuber> wolfc: just don't implement ModelUpdateOperationHandler ?[12:02:29] <wolfc> emuckenhuber: it's an existing operation[12:02:58] <wolfc> I want to add extensions and deployments, but only temporarily[12:10:22] *** kevinpollet has quit IRC[12:14:51] <opalka> Jaikiran, TTYL, doing some paper work :([12:15:00] <Jaikiran> opalka: sure, np[12:17:30] *** Binbin is now known as Binbingone[12:18:45] *** opalka is now known as opalka_lunch[12:19:40] <kkhan> Jaikiran: Does your https://github.com/jbossas/jboss-as/pull/40 fix the tests you mentioned, or introduce a new failure?[12:20:08] <Jaikiran> kkhan: it shows up a new issue, which looks related to the testcase than the impl. asoldano ^[12:20:30] <Jaikiran> kkhan: the functionality/fix for AS7-852 has been implemented, it's just a matter of fixing that testcase i believe[12:20:32] <jbossbot> jira [AS7-852] No component found for type '...' [Pull Request Sent (Unresolved) Bug, Major, jaikiran pai] https://issues.jboss.org/browse/AS7-852[12:20:51] <kkhan> I think then rather than pulling it in to master, it is better if you and asoldano sync up and then commit that?[12:21:00] <kkhan> i.e. once it is working[12:21:18] <Jaikiran> kkhan: sure, that should be fine[12:21:25] <kkhan> Cool :-)[12:21:28] <Jaikiran> opalka_lunch: asoldano: this is the error i'm seeing now http://pastie.org/1965489[12:21:56] <Jaikiran> for some reason, the entire exception stacktrace isn't being shown by the testcase output but a quick look suggests something to do with the testcase itself[12:22:12] <asoldano> Jaikiran, yes, the error seems unrelated...[12:29:17] <asoldano> Jaikiran, I have no special idea of the reason for that exception... Richard might know. I'll try pulling your changes locally and run the jbossws testsuite (that does not rely on arquillian, etc.) and see how that go[12:29:29] <asoldano> Jaikiran, I'll let you know after lunch, thanks for now[12:29:30] <Jaikiran> ok, thanks[12:29:36] <Jaikiran> sure[12:31:02] <stuartdouglas> Jaikiran: with that patch, I don't think the optional dependency will necessarily work, because if the service tries to start before the dependency is installed it may just start[12:31:10] <stuartdouglas> even though the dependency is going to be installed later[12:32:17] <stuartdouglas> I have actually been working on something similar in my master, but it does this horrible thing where is saves potential lazy injections for later in the process, and checks them when the DD is processed[12:34:05] <stuartdouglas> I have remove methods from DD and an interceptor that prevents invocation of non business methods in my master[12:34:51] <Jaikiran> stuartdouglas: ah yes, i had missed that case[12:36:09] <Jaikiran> btw, i did not get what you meant by "I have remove methods from DD and an interceptor that prevents invocation of non business methods in my master". is that related to something else?[12:36:28] <stuartdouglas> there should have been an 'also' in there[12:36:36] <Jaikiran> oh ok :)[12:36:49] <stuartdouglas> just letting you know so we don't double up :-)[12:36:56] <Jaikiran> sure :)[12:37:11] <Jaikiran> i'm going to take a look at the latest results and see what needs to be fixed next[12:37:15] <stuartdouglas> are you happy with the interceptor ordering changes?[12:37:25] <Jaikiran> they have been done?[12:37:28] <Jaikiran> i haven't seen yet[12:37:31] <Jaikiran> which branch is it?[12:37:35] <stuartdouglas> upstream[12:37:44] *** stliu has quit IRC[12:37:46] * Jaikiran checks[12:37:46] <stuartdouglas> It all has been merged[12:38:05] <Jaikiran> you mean the Phase.java type ordering right?[12:38:27] <stuartdouglas> yes, I was not 100% sure what the correct interceptor order was[12:38:39] <stuartdouglas> so I just took a guess, feel free to change it if it is wrong[12:38:44] <Jaikiran> ok[12:39:26] <stuartdouglas> https://github.com/jbossas/jboss-as/blob/master/ee/src/main/java/org/jboss/as/ee/component/interceptors/InterceptorOrder.java[12:41:25] <Jaikiran> yep, looks similar to what i was thinking[12:42:18] <Jaikiran> although i hadn't thought of how user interceptors would be added. this approach looks good[12:42:53] <stuartdouglas> because it uses a stable sort they just go in the order they are added to the configuration[12:43:18] <stuartdouglas> I did not add an ordering thing to the configurators as the order does not really matter now[12:44:01] <Jaikiran> yeah, it won't matter when those configurators run, anymore[12:45:21] *** JimMa has quit IRC[12:45:53] <stuartdouglas> the @Resource injections that only exist if there is an appropriate binding are kinda problematic with comp / module aliasing[12:46:14] <Jaikiran> it's only a special case with env-entry[12:46:21] <Jaikiran> rest of it is mandatory injection[12:46:24] *** miclorb_ has quit IRC[12:46:35] <Jaikiran> what's comp/module aliasing?[12:46:39] <stuartdouglas> because if you have @Resource("blagh"), depending on if that class is used in a EJB or managed bean changes the location that the env entry is bound to[12:46:54] <stuartdouglas> in a war the java:comp is aliased to java:module[12:47:06] <stuartdouglas> and all EJB's in the war have a shared java:comp[12:47:11] <stuartdouglas> that is really java:module[12:47:16] <Jaikiran> right, so ultimately it's java:module[12:47:17] <Jaikiran> yep[12:48:01] <Jaikiran> well, in case of MBs the only way the env-entry can be specified is via web.xml[12:48:09] <Jaikiran> (i.e ultimately java:module)[12:48:16] <Jaikiran> since MBs don't have descriptor of their own[12:48:22] <stuartdouglas> but if you use the same class (say an interceptor) in both a web module and in a ejb-jar there are actually two different bindings[12:48:32] <stuartdouglas> a java:comp one and a java:module one[12:48:56] <stuartdouglas> so it may be that you only want the injection to run on the ejb, as there is a comp binding but no module binding[12:49:16] <stuartdouglas> what we have at the moment does not support that very well as it is all per class, rather than per component[12:50:10] <Jaikiran> yep, that's correct. ultimately the @Resource(name="blah") should translate to a absolute name and then whether or not an injection is required will be decided on whether there's an entry for that absolute name[12:51:45] <stuartdouglas> yea, but that can be per component to, as not all components have the same java:comp bindings[12:52:06] <stuartdouglas> so it's a bit of a pain and does not really fit well with the rest of the injection stuff :-([12:52:12] <Jaikiran> agreed[12:52:55] <Jaikiran> to be honest, i think this corner case is only going to used/tested in the TCK. but yeah, we'll somehow have to accomodate this one[12:53:20] <hbraun> darranl: fyi https://issues.jboss.org/browse/AS7-881[12:53:21] <jbossbot> jira [AS7-881] DigestAuthenticator calculates wrong hash [Open (Unresolved) Bug, Major, Heiko Braun] https://issues.jboss.org/browse/AS7-881[12:53:28] <hbraun> darranl: just send a pull request[12:53:44] <hbraun> darranl: https://github.com/heiko-braun/jboss-as/commit/2f1d1cd6cfc5d407730e4df66ee358b77ffda071[12:53:45] <jbossbot> git [jboss-as] 2f1d1cd.. Heiko Braun Fix AS7-881[12:53:46] <jbossbot> jira [AS7-881] DigestAuthenticator calculates wrong hash [Open (Unresolved) Bug, Major, Heiko Braun] https://issues.jboss.org/browse/AS7-881[12:54:00] <darranl> hbraun, excellent, thanks for continuing to debug[12:54:17] <hbraun> it didn't consider the actual request method[12:54:27] <hbraun> always did assume GET[12:54:32] <hbraun> hence my problems[12:59:27] *** sgilda has joined #jboss-as7[13:00:34] <Jaikiran> stuartdouglas: this might run into CCE isn't it? https://github.com/stuartwdouglas/jboss-as/commit/8d044ff09cfe972839b0c0076cbb7aa5949d3100#L2R61[13:00:35] <jbossbot> git [jboss-as] 8d044ff.. Stuart Douglas Remove methods from deployment descriptor[13:00:35] *** kevinpollet has joined #jboss-as7[13:02:09] <stuartdouglas> it could i suppose if the xml descriptor has a remove method for something that is not a SFSB[13:02:26] <darranl> hbraun, have you submitted a pull for that? or do you want me to pull it into my branch and get it included in the changes I send at the end of today?[13:02:27] * stuartdouglas should probably check for that[13:02:42] <hbraun> yes, i've send a pull already[13:02:48] <darranl> hbraun, cool[13:20:45] *** jwulf has joined #jboss-as7[13:22:38] *** ccrouch has joined #jboss-as7[13:22:38] *** ChanServ sets mode: +v ccrouch[13:24:30] *** AndyTaylor has quit IRC[13:29:19] *** asoldano is now known as asoldano_lunch[13:30:32] *** bbrowning has joined #jboss-as7[13:30:32] *** ChanServ sets mode: +v bbrowning[13:33:36] *** opalka_lunch is now known as opalka_away[13:34:20] *** balunasj has joined #jboss-as7[13:34:20] *** balunasj has joined #jboss-as7[13:34:20] *** ChanServ sets mode: +v balunasj[13:38:30] *** jc3 has joined #jboss-as7[13:38:30] *** ChanServ sets mode: +v jc3[13:39:01] *** balunasj has quit IRC[13:40:44] *** opalka_away has quit IRC[13:43:16] *** maeste has joined #jboss-as7[13:43:17] *** ChanServ sets mode: +v maeste[13:52:33] <alesj> with Ispan being integrated, do we also have support for http session replication, etc?[13:53:30] <rmaucher> the clustered session manager hasn't been ported yet (assuming the plan is to port the AS 6 session manager)[13:54:33] <alesj> ah, ok, makes sense then why i don't see any discovery :-)[13:56:00] <alesj> but i guess cache should be shared … let me try that[13:56:26] *** torben is now known as torben|afk[13:56:39] <hbraun> darranl: ping[13:56:44] <darranl> hi hbraun[13:57:12] <hbraun> darranl: when auth fails with digest (i.e. wrong pw) i never get to see the login prompt again[13:57:35] <hbraun> darranl: unless I quit the browser[13:57:46] <hbraun> problem with the impl. or the spec?[13:57:47] <darranl> hbraun, ok let me check[13:58:23] <hbraun> darranl: I would expect it to bring up the prompt when reloading the page[13:59:55] <darranl> hbraun, https://issues.jboss.org/browse/AS7-882[13:59:57] <jbossbot> jira [AS7-882] Verify invalid password behaviout for Digest auth (Domain Management) [Open (Unresolved) Task, Major, Darran Lofthouse] https://issues.jboss.org/browse/AS7-882[14:00:31] <hbraun> did you create it just now?[14:00:35] *** bgeorges has joined #jboss-as7[14:00:59] <darranl> yes just created[14:01:06] <hbraun> thanks[14:02:40] <darranl> my current task is a little internal refactoring and then I will debug the failed password checks[14:04:21] *** pmuir has joined #jboss-as7[14:04:21] *** ChanServ sets mode: +v pmuir[14:06:05] <hbraun> darranl: I don't really understand why failed password returns 403[14:06:23] <hbraun> guess if it would return 401 the prombpt would re-appear[14:09:08] <darranl> yes it probably should be 401[14:09:51] <hbraun> 403 is not mentioned in RFC 2617 at all[14:11:28] <hbraun> dai can only see 401 and 401 w. "stale=true" in response to an "Authorization" header[14:11:48] <hbraun> darranl: i can take care of it if you like[14:12:04] <hbraun> darranl: but I don't want to mess up your commit[14:12:26] *** jpederse has joined #jboss-as7[14:12:26] *** ChanServ sets mode: +v jpederse[14:14:01] <darranl> hbraun, this should be what you need - https://github.com/darranl/jboss-as/commit/e83ae5f08ea7c575e5f09b752461ae3a6b8f0a78[14:14:02] <jbossbot> git [jboss-as] e83ae5f.. Darran Lofthouse [AS7-882] Re-send a 401 challenge after failure to verify the users credentials.[14:14:03] <jbossbot> jira [AS7-882] Verify invalid password behaviour for Digest auth (Domain Management) [Open (Unresolved) Task, Major, Darran Lofthouse] https://issues.jboss.org/browse/AS7-882[14:14:56] <hbraun> darranl: i'll give it a try[14:15:53] <darranl> hbraun, will just be away for 15min - need to grab some food[14:16:12] <hbraun> darranl: bon appetit[14:17:54] *** darranl is now known as darranl_afk[14:26:40] *** magesh has joined #jboss-as7[14:31:11] *** sgilda_ has joined #jboss-as7[14:31:49] <davidbos> hbraun: ping[14:32:07] <hbraun> davidbos: pong[14:32:08] <davidbos> hbraun: is there a subsystem that would be a good starting point to look at?[14:32:18] <davidbos> hbraun: one that works in the console...[14:32:45] <hbraun> you can look at the datasource subsystem[14:32:48] <hbraun> in domain mode[14:33:05] <hbraun> davidbos: or jms is even better[14:33:13] <hbraun> it#s a little more simply[14:33:15] <hbraun> simple[14:33:23] <davidbos> hbraun: ok - will take a look there thanks :)[14:33:27] *** sgilda has quit IRC[14:33:42] <hbraun> davidbos: let me know how you get along[14:34:17] <davidbos> hbraun: at this point all is good - it starts up and mvn install runs to completion without problems...[14:34:59] *** AndyTaylor has joined #jboss-as7[14:34:59] *** ChanServ sets mode: +v AndyTaylor[14:37:25] <bobmcw> hey, so we have integ tests that start the AS, deploy an app, undeploy[14:37:38] <bobmcw> sometimes the deploying starts services that take a while (15-40s) to start[14:37:46] <bobmcw> we're seeing the app get undeployed before it's fully deployed[14:37:55] <bobmcw> and not sure our long-starting services get shutdown completely[14:37:57] <bobmcw> expected?[14:38:05] <bobmcw> we're using the .asynchronous stuff etc to start'em[14:38:33] <bobmcw> but just seems weird that the app can be undeployed before it's fully deployed[14:38:41] <jbossbot> git [jboss-as] push master ae6c41e.. Heiko Braun Fix AS7-881[14:38:42] <jbossbot> jira [AS7-881] DigestAuthenticator calculates wrong hash [Resolved (Done) Bug, Major, Heiko Braun] https://issues.jboss.org/browse/AS7-881[14:38:42] <jbossbot> git [jboss-as] push master URL: http://github.com/jbossas/jboss-as/compare/23b4870...ae6c41e[14:38:44] <bobmcw> and after deployment, services related to the app are still churning[14:38:49] <bobmcw> *undeployment[14:40:00] *** magesh has quit IRC[14:46:35] *** mmoyses has joined #jboss-as7[14:46:36] *** ChanServ sets mode: +v mmoyses[14:48:03] <jc3> i don't think there's anything wrong with the shutdown api, btw.[14:49:05] <hbraun> darranl_afk: pig me when you get back[14:50:53] *** asoldano_lunch is now known as asoldano[14:52:51] *** smarlow has joined #jboss-as7[14:52:51] *** ChanServ sets mode: +v smarlow[14:54:46] *** smcgowan has joined #jboss-as7[14:54:46] *** ChanServ sets mode: +v smcgowan[14:55:01] <smcgowan> kkahn: is there action from me on my pull request[14:56:16] *** magesh has joined #jboss-as7[14:56:36] <smcgowan> kkhan: specifically, this: http://lists.jboss.org/pipermail/jbossas-pull-requests/2011-May/000770.html[14:59:16] *** darranl_afk is now known as darranl[14:59:24] <darranl> hi hbraun[14:59:39] <hbraun> darranl: i'll get back to you in a inute[14:59:43] <hbraun> minute[15:00:35] *** aloubyansky has joined #jboss-as7[15:03:37] *** sgilda_ has quit IRC[15:03:59] *** sgilda has joined #jboss-as7[15:06:34] *** opalka has joined #jboss-as7[15:06:34] *** opalka has joined #jboss-as7[15:06:34] *** ChanServ sets mode: +v opalka[15:10:07] *** pferraro has joined #jboss-as7[15:10:08] *** ChanServ sets mode: +v pferraro[15:10:22] <kkhan> smcgowan: i see you have a merge commit in your branch, I'm not 100% sure but I believe they screw things up down the line[15:11:39] <smcgowan> kkhan: ah, ok, what should I do?[15:11:41] <hbraun> darranl: doesn't work at all[15:11:56] <hbraun> but I have no idea why[15:12:04] <darranl> ok well I will debug properly once I have finished this task[15:12:20] <hbraun> darranl: I cannot even debug it in standlone, because it esacpes into a runtime task[15:12:59] <hbraun> AFAIK it should prompt the challenge with stale=false to force a new login prompt[15:13:28] <hbraun> dathanks[15:13:32] <hbraun> darranl: thanks[15:13:45] <kkhan> smcgowan: Keep https://github.com/smcgowan/jboss-as/commit/aee8a9bad1fa7221e5983b831eb32070e13d63b5 safe somewhere, for example 'git checkout -b backup'[15:13:46] <jbossbot> git [jboss-as] aee8a9b.. Shelly McGowan JBCTS-1101, JBEE-81 - Fix for JSTL SQL Actions, Updates to Final API versions[15:13:47] <jbossbot> jira [JBCTS-1101] Redirected to: https://issues.jboss.org/login.jsp?permissionViolation=true&os_destination=%2Fsi%2Fjira.issueviews%3Aissue-xml%2FJBCTS-1101%2FJBCTS-1101.xml[15:13:48] <jbossbot> jira [JBEE-81] Release Final versions of APIs for inclusion in AS 7.0.0.Final [Resolved (Done) Release, Major, Shelly McGowan] https://issues.jboss.org/browse/JBEE-81[15:14:04] <kkhan> Then checkout master[15:14:11] <kkhan> Do a git fetch upstream[15:14:44] <kkhan> Then git reset --hard upstream/master[15:15:01] <kkhan> Then git cherry-pick aee8a9bad1fa7221e5983b831eb32070e13d63b5[15:15:38] *** bobmcw has quit IRC[15:16:05] <smcgowan> kkhan: thanks for the reference. i'm heading into the office, will fix my enviroment and re-request the pull[15:16:07] <kkhan> i don't remember exactly what the conflict was but I think you had upgraded to a lower version of something than had been done in master[15:16:13] <kkhan> smcgowan: On final thing[15:16:15] <kkhan> One[15:16:32] <kkhan> Normally to get your branch in fit state for a commit[15:16:42] <kkhan> You git fetch upstream[15:16:48] <kkhan> git rebase upstream/master[15:16:57] <kkhan> that avoids the merge commits[15:16:59] <smcgowan> i did do the rebase for sure[15:17:51] <kkhan> ok, I'm not sure where that merge comes from then[15:18:30] <smcgowan> kkhan: last week i did run into an issue, which I thought I resolved but apparently not[15:18:32] <smcgowan> i'll clean it up[15:18:41] <kkhan> I think if you have your master and accidentally do a git merge upstream/master it can happen[15:18:44] <kkhan> but not sure[15:19:09] <kkhan> thanks[15:19:21] *** mmoyses has quit IRC[15:21:11] *** smcgowan has quit IRC[15:21:59] *** mmoyses has joined #jboss-as7[15:21:59] *** ChanServ sets mode: +v mmoyses[15:22:33] *** bobmcw has joined #jboss-as7[15:22:33] *** ChanServ sets mode: +v bobmcw[15:22:50] <asoldano> opalka, ping[15:23:14] <asoldano> Jaikiran, I've tried your fix with the jbossws testuite and it works fine[15:23:45] *** JimMa has joined #jboss-as7[15:23:53] <Jaikiran> asoldano: thanks for testing that. we'll have to wait for opalka then to make sure that testsuite/integration test is fixed to make it pass[15:26:12] <asoldano> Jaikiran, in any case, the sooner that is merged, the better, as we have many jaxws TCK regressions due to recent ejb3/ee changes and removing those because of AS7-852 would really simplify the work (also to Shelly)[15:26:13] <jbossbot> jira [AS7-852] No component found for type '...' [Pull Request Sent (Unresolved) Bug, Major, jaikiran pai] https://issues.jboss.org/browse/AS7-852[15:27:18] *** opalka has quit IRC[15:28:03] *** mlinhard has quit IRC[15:32:00] *** magesh has quit IRC[15:32:03] *** magesh1 has joined #jboss-as7[15:33:30] *** frainone has joined #jboss-as7[15:33:31] *** ChanServ sets mode: +v frainone[15:35:12] <dmlloyd> good morning all[15:36:27] <wolfc> good morning[15:36:49] <baileyje> goo morning[15:36:53] <wolfc> I'm trying to get it working on modules CR3, but I keep running into issues[15:37:01] *** mmoyses has quit IRC[15:40:01] *** mmoyses has joined #jboss-as7[15:40:01] *** ChanServ sets mode: +v mmoyses[15:40:58] *** fnasser has joined #jboss-as7[15:47:04] <alesj> pferraro: i'm abusing "web" cache container, to create new cache … on two diff nodes[15:47:19] <alesj> pferraro: but it seems it is not shared among the two nodes[15:47:26] <alesj> pferraro: is this expected by-default?[15:49:20] <alesj> galderz: ^ … any idea?[15:50:34] <galderz> alesj: maybe your cluster is not forming?[15:50:54] <galderz> also, what cache mode does it have? invalidation/repl/dist?[15:51:31] *** bstansberry has joined #jboss-as7[15:51:31] *** ChanServ sets mode: +v bstansberry[15:51:34] <alesj> galderz: i can see the cluster formed: Received new cluster view: [skywalker-6037|1] [skywalker-6037, solaris-18654][15:52:09] <alesj> but then I create the new cache on the fly: EmbCM::getCache(<name>, true)[15:52:28] <alesj> hence no idea which mode gets used[15:52:36] <alesj> i would styill expect it to be shared[15:52:49] <alesj> but i might (and probably am) be wrong :-)[15:52:57] <alesj> galderz: ^[15:54:49] <galderz> well, depends on what that cache name is[15:55:17] <galderz> if it does not match any of the caches configured in that "stolen" cache manager, the mode is defined by the defaultCache configuration[15:55:29] <galderz> you can find the cache mode by checking[15:56:04] *** pgier has joined #jboss-as7[15:56:04] *** ChanServ sets mode: +v pgier[15:56:06] <galderz> cache.getConfiguration.getCacheMode[15:56:17] <alesj> let me try an abuse the cache container some more, with repl-async cache usage[15:56:58] <smarlow> ..[15:59:15] <bstansberry> dmlloyd, baileyje: WDYT about breaking ModelNodeRegistration on new-controllers (by using NewStepHandler) ?[15:59:49] <bstansberry> the options are that or a bunch of New*** in that package (which I have but don't particularly want to push)[16:01:19] <wolfc> dmlloyd: I have https://github.com/wolfc/jboss-as/commit/4a5acf3c7ca0bf8a4eabfe5fdd0e0fe8a6559d07 with which I can poke any admin operation into the embedded model controller[16:01:20] <jbossbot> git [jboss-as] 4a5acf3.. Carlo de Wolf AS7-434: do not store model and expose a ModelControllerClient[16:01:21] <jbossbot> jira [AS7-434] Implement EJB 3.1 FR 22 Embeddable Usage [Open (Unresolved) Task, Critical, Carlo de Wolf] https://issues.jboss.org/browse/AS7-434[16:01:39] <alesj> galderz: yup, abusing repl-async helped :-)[16:01:41] <wolfc> from what I recall the plan was to add an extension which registers a processor which adds a dep[16:02:02] <wolfc> problem is that registering a dup is only allowed for boot operations[16:02:14] <galderz> alesj: use cache.getConfiguration.getCacheMode to see what the mode is[16:04:37] <alesj> nah, that's fine, i'm just hacking some simple demo for today's JUG[16:09:05] <dmlloyd> wolfc: why not just implement ConfigurationPersister? what's the point of using XmlConfigurationPersister[16:09:29] <wolfc> dmlloyd: I need to load the standalone.xml else nothing happens[16:10:04] <dmlloyd> for embedded you should be generating config programmatically[16:10:28] <dmlloyd> if you rely on standalone.xml then it'll only work if you have a standalone.xml[16:10:39] <bobmcw> dmlloyd: so the torquebox core module would like to bring in jruby.jar from outside of $JBOSS_HOME/modules/...[16:10:58] <bobmcw> user brings his own jruby installation, we want to use the jars where they sit in $JRUBY_HOME in our subsystem[16:10:58] <dmlloyd> bobmcw: you guys like making things difficult don't you :)[16:11:05] <bobmcw> java's dead, man.[16:11:25] <dmlloyd> okay so by default you read JRUBY_HOME, always?[16:11:28] <bobmcw> and I mentioned this like 2 weeks ago on skype! it's not done yet?![16:11:49] <bobmcw> well, we have a strategy... we respect $JRUBY_HOME env, or -Djruby.home sysprop, or if not set, we look for $JBOSS_HOME/../jruby[16:11:59] <bobmcw> and from there can then find our jars relative to the effective jrubyHome[16:12:04] <dmlloyd> why not ship it?[16:12:11] <bobmcw> well, we do currently[16:12:14] <bobmcw> but users like to have their own[16:12:16] <bobmcw> it's a ruby thing[16:12:27] <bobmcw> they manage their gemsets etc external to the AS[16:12:32] <bbrowning> that's like asking why AS doesn't shit a jdk ;)[16:12:37] <dmlloyd> heh[16:12:37] <bobmcw> and if torquebox goes into jboss-as-dist, do you want to ship jruby too?[16:12:37] <wolfc> dmlloyd: you need a full distribution anyway[16:12:37] <bbrowning> ship*[16:12:46] <dmlloyd> why not![16:12:53] <dmlloyd> okay, well I was just curious[16:12:56] <dmlloyd> I don't know much about ruby[16:13:02] <bobmcw> it's a feature to use an external JRUBY_HOME[16:13:06] <bobmcw> we'd like to support it[16:13:20] <dmlloyd> okay so here's what I'd do[16:13:27] <bobmcw> our boot operation can do some seeking and searching, it doesn't need jruby itself[16:13:31] *** clebert has joined #jboss-as7[16:13:31] *** ChanServ sets mode: +v clebert[16:13:34] <dmlloyd> in your subsystem add op, locate the ruby you want to use[16:13:38] <baileyje> bstansberry: I am ok with breaking it if it isn't for a long time. How long are you thinking?[16:13:58] <bobmcw> dmlloyd: 'k[16:14:10] <bobmcw> can we mutate our subsystem's Module to add external non-module deps?[16:14:17] <dmlloyd> then record that info... then in your deployer, create or reuse a new module in the ServiceModuleLoader[16:14:24] <bstansberry> baileyje: well, forever[16:14:42] <bstansberry> baileyje: it uses OperationHandler in its API; needs to become NewStepHandler[16:14:53] <baileyje> Oh, so you mean fixing it..[16:14:56] *** bgeorges has quit IRC[16:14:58] <bobmcw> our deployers do need to touch org.jruby.Ruby classes'n'such. They can mix in that external jruby Module we create?[16:15:16] <dmlloyd> hm, okay, that makes it a bit trickier...[16:15:18] <bstansberry> baileyje: right, but that will break everything that does registration[16:15:29] <bobmcw> well, they touch services, which touch pools, which touch org.jruby.Ruby[16:15:33] <bobmcw> so, sorta linky, but not directly[16:15:47] <dmlloyd> services and pools all being defined in static modules today?[16:15:47] <bstansberry> but that's probably fine; everytime you touch a handler you are breaking 1 line of registration anyway[16:15:53] <baileyje> bstansberry: It is all broken in my version anyway since half the Op Handlers are gone[16:16:11] <bobmcw> yah, services and pools are our code[16:16:17] <baileyje> bstansberry: So really you are trying to fix it for me :)[16:16:18] <bobmcw> Service<T> and Pool<Ruby>[16:16:28] <dmlloyd> bstansberry: yeah let's start breaking[16:17:04] <bstansberry> dmlloyd, baileyje: ok, will do.[16:17:14] <bobmcw> dmlloyd: so ServiceModuleLoader is what creates new modules, and I can programatically assemble one that doesn't have a module.xml or live under modules/ ?[16:17:38] <dmlloyd> bobmcw: ServiceModuleLoader is what creates modules for deployments. Deployments can see it, our code can't (other than via reflection)[16:17:56] <bstansberry> baileyje: i should merge your stuff today. i'm concerned about not rebasing this branch onto master and losing op handler changes[16:18:27] <bstansberry> onto?? from?? whatever the correct preposition is[16:18:28] <dmlloyd> bobmcw: would it be enough to support jruby.home? maybe have standalone.sh set it from JRUBY_HOME if it's there?[16:18:40] <baileyje> bstansberry: Yeah. I have all the adds migrated, working on the removes now[16:19:19] <asoldano> kkhan, Jaikiran: please use this as a testcase for AS7-852: https://github.com/asoldano/jboss-as/commit/f5fc276c70675901f35b8f80b61f2db19b3a3679[16:19:21] <jbossbot> jira [AS7-852] No component found for type '...' [Pull Request Sent (Unresolved) Bug, Major, jaikiran pai] https://issues.jboss.org/browse/AS7-852[16:19:22] <jbossbot> git [jboss-as] f5fc276.. asoldano [AS7-852] Adding non-ejb3 test[16:19:24] <dmlloyd> baileyje: did you see my base class for add handlers last night?[16:19:33] <baileyje> bstansberry/dmlloyd: No sure if I should be using context.writeModel more than I am.[16:19:34] <bobmcw> dmlloyd: if we somehow know it before boot-time, we still have to mix in the jars, right?[16:19:39] <bobmcw> some machinations?[16:19:54] <baileyje> dmlloyd: I saw it. I was afraid to look at what it meant too all my changes :)[16:20:00] *** mgoldmann has joined #jboss-as7[16:20:00] *** ChanServ sets mode: +v mgoldmann[16:20:03] <bstansberry> baileyje: see what I pushed last night; dmlloyd added a readModelForUpdate[16:20:06] <Jaikiran> asoldano: ok, let me check[16:20:11] <dmlloyd> bobmcw: well I'm thinking we can still define an org.jruby module... but for the resource root we can have ${jruby.home} and I can add prop interpolation[16:20:14] <baileyje> bstansberry: That is what I figured.[16:20:32] <baileyje> Since write model is more related to addressable children.[16:20:39] <baileyje> Most adds do very little of that[16:20:52] <bobmcw> dmlloyd: ah... hm[16:21:04] <bobmcw> could be useful[16:21:18] <asoldano> Jaikiran, that test is passing with your fork, while it fails with the current master[16:21:23] <bobmcw> I'll poke around some and see what kinda classloaders I piss off[16:21:25] <dmlloyd> bobmcw, it's really the only way I can think of to make it a static module, else we're getting into some more esoteric hacking[16:21:34] <asoldano> Jaikiran, basically, the test Richard provided is also affected by other ejb3 issues[16:21:51] *** pmuir has quit IRC[16:22:04] <bobmcw> dmlloyd: could my subsystem add do some work to figure out jruby.home and set it as a domain model <path> or whatnot[16:22:10] <bobmcw> so we still have our searchiness[16:22:11] <baileyje> bstansberry: dmlloyd's changes are in your push, correct?[16:22:46] <bstansberry> baileyje: yes[16:22:55] <baileyje> great.[16:23:04] <dmlloyd> bobmcw: yeah I think we have some system-defined paths today (right bstansberry?)[16:23:30] <baileyje> dmlloyd: What was your base add class[16:23:38] <bstansberry> dmlloyd: yes[16:23:45] * bstansberry goes to look[16:23:51] <bstansberry> to see which[16:24:04] <dmlloyd> baileyje: AbstractDeploymentChainStep[16:24:09] <bobmcw> but not currently interpolated in module.xml, I'm guessing?[16:24:16] <dmlloyd> not currently, no[16:24:30] <dmlloyd> that's an enhancement I'd have to do[16:25:00] <baileyje> Oh. For the boot ops[16:25:03] <baileyje> cool..[16:25:09] <tdiesler> aslak, hi alsak. shall we sync up again on AS7/ARQ this week?[16:25:38] <baileyje> dmlloyd: Do we need to break the processor registration out from the existing ops?[16:25:40] <bstansberry> bobmcw: https://github.com/jbossas/jboss-as/blob/master/server/src/main/java/org/jboss/as/server/ApplicationServerService.java#L154[16:26:07] <bstansberry> those are the paths that are always there[16:26:13] <dmlloyd> baileyje: well they should query the controller type at least... actually I was going to break that out in to a "isBooting()" flag...[16:26:30] <dmlloyd> right now there's a special type for "server boot" but that's kinda weak[16:26:51] <dmlloyd> instead it should be more like: "if (controllerType == SERVER && context.isBooting()) { add step; }"[16:27:00] <baileyje> dmlloyd: Yeah. I have a lot of ops that have to check for both[16:27:03] <bobmcw> bstansberry: 'k, I'll poke, and probably whinge later[16:27:13] <dmlloyd> also we don't have "restart required" figured out yet do we[16:27:21] <baileyje> Since they could run in either runtime or server boot.[16:27:40] <dmlloyd> all our runtime stuff should be checking for SERVER at least.[16:27:53] <dmlloyd> okay any problem with be adding a booting flag?[16:27:59] <dmlloyd> and getting rid of SERVER_BOOT?[16:28:10] <aslak> tdiesler, hey[16:28:53] <aslak> tdiesler, sure.. from my knowledge right now, i believe alr is working on the beta-1 upgrade and removal of arq1.[16:29:08] <aslak> tdiesler, have you looked at creating a bundle for beta-1 ?[16:29:41] *** aloubyansky has quit IRC[16:30:13] <baileyje> dmlloyd: I have them all checking for SERVER at a minimum but the booting flag would be much cleaner[16:30:19] <tdiesler> aslak, what do you mean? We have a beta-1 bundle in container-osgi, no?[16:30:30] <dmlloyd> okay[16:31:23] <aslak> tdiesler, no i mean the 'static' subsystem deployment. but the upgrade alr is doing might take care of that build?[16:32:23] *** pmuir has joined #jboss-as7[16:32:23] *** pmuir has joined #jboss-as7[16:32:23] *** ChanServ sets mode: +v pmuir[16:33:54] <tdiesler> aslak, ok lets talk about that on Thu. Perhaps I can help porting https://github.com/tdiesler/jboss-as/tree/as773 to beta-1[16:34:05] <tdiesler> aslak, It should not be too difficult[16:34:33] <dmlloyd> bstansberry, baileyje: boot flag change is pushed[16:34:52] <baileyje> dmlloyd: thanks.[16:35:07] <baileyje> bstansberry: Let me know when you have some time integrate the change[16:35:14] <aslak> tdiesler, should be fairly simple.. most of it could be 'copied' from the osgi branch[16:35:19] <dmlloyd> bstansberry: we may want to rebase soon before we really start breaking things[16:35:21] <aslak> branch/container[16:35:44] <bstansberry> baileyje: i'll do that now then the ModelNodeRegistration bit[16:35:56] <baileyje> bstansberry: Sounds good[16:36:39] <tdiesler> aslak, yes agreed, but lets hear ALR first before I jump in and duplicate what he is working on. Perhaps he is happy to delegate this thing away[16:37:09] <aslak> tdiesler, :)[16:38:35] *** mbg has joined #jboss-as7[16:38:35] *** ChanServ sets mode: +v mbg[16:39:41] <tdiesler> Nihility, could you please hava a look at https://github.com/jbossas/jboss-as/pull/35[16:39:45] <bstansberry> dmlloyd: noodling a bit about validation[16:40:01] <bstansberry> mostly a path forward w/o doing anything right now :-)[16:40:08] <tdiesler> Nihility, this would complete OSGi for AS7 CR1[16:40:09] *** magesh has joined #jboss-as7[16:40:27] <Nihility> tdiesler: this looks like it contains merge commits[16:41:00] <bstansberry> one thought is StepHandler exposes isAutoValidationAllowed() -- false only in cases where we think it won't work[16:41:01] <tdiesler> Nihility, yes. I did an experiment with rebase and merge. This final result was the same[16:41:07] *** magesh1 has quit IRC[16:41:15] <tdiesler> Nihility, whats wrong with merge commits?[16:41:35] <dmlloyd> they make history hard to follow[16:41:47] *** smcgowan has joined #jboss-as7[16:41:47] *** ChanServ sets mode: +v smcgowan[16:41:53] <bstansberry> then the context exposes isOperationAutoValidated() to tell a handler if the op has already been checked (so it doesn't have to)[16:42:07] <Nihility> tdiesler: merge commits make the history hard to follow, and they can cover up conflicts in some cases[16:42:14] <dmlloyd> bstansberry: hmmm.[16:42:59] <bstansberry> basically i'm thinking about how to add hooks to support *future* validation by the controller without actually doing it now[16:43:01] <tdiesler> dmlloyd, Nihility, I have folks branching off my work. When I rebase it screws them. So what do we do. Git recommends to never rebase stuff that you have published publically.[16:43:16] <dmlloyd> tdiesler: it really doesn't screw them, if they rebase as well[16:43:46] <dmlloyd> thus far the no-rebasing rule has only applied to upstream[16:44:12] <dmlloyd> we've done multi-level branches with rebasing in the past and it has worked out okay[16:44:37] <Nihility> the unfortunate things is it might be a PITA to rebase this branch[16:44:52] <Nihility> since it has it looks like 3 merge commits interleaved[16:44:58] <bstansberry> baileyje: pushed isBooting()[16:45:09] <Nihility> hmmm[16:45:35] <Nihility> ill see what it does[16:45:46] <tdiesler> dmlloyd, ok. where can we go from here? I could cheery pick my commits onto a new branch based of current master[16:47:17] <dmlloyd> looks like Nihility is going to give it a try, but otherwise yeah that'd accomplish the same thing[16:48:53] <Nihility> wow[16:48:56] <Nihility> i dont believe it[16:48:59] <Nihility> it merges clean[16:49:19] <Nihility> tdiesler: im not sure how you pulled that off :)[16:50:09] <bobmcw> I had cherry-pick freak me out the other day, but spookily following complex file movements[16:50:16] <bobmcw> "holy shit, it worked?"[16:51:18] *** balunasj has joined #jboss-as7[16:51:18] *** balunasj has joined #jboss-as7[16:51:18] *** ChanServ sets mode: +v balunasj[16:51:42] <tdiesler> Nihility, merci[16:52:25] *** alesj has quit IRC[16:53:49] <tdiesler> Nihility, you can run mvn -Dtest=osgi/**/*TestCase test in integration. That should give you the full level of supported osgi functionality[16:54:31] *** balunasj is now known as balunasj_mtg[16:54:32] <Nihility> tdiesler: btw two ways you can avoid merges, one is to hold off rebasing until everyone has checked in their work. Another is to rebase your branch, and have them do a rebase on their patches before you merge[16:55:00] *** pferraro has quit IRC[16:55:14] <Nihility> the latter works pretty well[16:56:28] <tdiesler> Nihillity, i.e. don't rebase before you have consumed your dependees, right?[16:56:30] <Nihility> baileyje: hey if you are looking for tiny tasks to do for filler , adding Service and ServiceMBean interfaces into our sar stuff[16:57:09] <bstansberry> dmlloyd: do you see any reason to allow ModelType.UNDEFINED as the newData param passed to writeModel ?[16:57:42] <dmlloyd> I don't see any reason to disallow it, but the last op handler I wrote was months ago so I'm probably not an authority[16:57:44] *** mbg has quit IRC[16:57:47] <bstansberry> think that would be a bug, as a resource would never be represented as undefined[16:57:54] *** torben|afk has quit IRC[16:57:55] <bstansberry> it would either be defined or removed[16:58:01] <Nihility> tdiesler: thats the former approach, which is fine too, although a bit more restrictive to the branch maintainer. The latter approach is where you rebase as often as you like, but you tell everyone who is a dependant that you have rebased and they need to as well[16:59:40] <Nihility> tdiesler: that branch has a snapshot dep on osgi[16:59:47] <Nihility> tdiesler: are you doing a release of that?[16:59:51] <tdiesler> should be CR3[16:59:57] <tdiesler> fixing it now[17:01:00] <Nihility> btw guys i ran into a major productivity problem[17:01:08] <Nihility> my beer fridge is currently out of beer![17:01:19] *** jamezp_afk is now known as jamezp[17:01:25] <Jaikiran> asoldano: do you want me to remove those ejb3ws test that Richard introduced?[17:01:34] <Jaikiran> i don't see any point in letting them stay[17:01:45] <Nihility> need help![17:01:48] <Nihility> :)[17:01:50] <Jaikiran> i'll be introducing the tests that you provided[17:02:21] <asoldano> Jaikiran, no, I'd prefer to keep them, as they were put there for other reasons and just added the @Resource annotation later to serve as tests for AS7-852 too[17:02:23] <jbossbot> jira [AS7-852] No component found for type '...' [Pull Request Sent (Unresolved) Bug, Major, jaikiran pai] https://issues.jboss.org/browse/AS7-852[17:02:31] <Nihility> darranl: hey when you have a chance can you give me an update wrt to domain security[17:02:34] *** mbg has joined #jboss-as7[17:02:34] *** ChanServ sets mode: +v mbg[17:02:46] <Nihility> Jaikiran: do you expect to have ejb security in CR1?[17:02:54] <asoldano> Jaikiran, basically Richard needs them for his ejb3-ws integration work[17:03:16] <Jaikiran> Nihility: i think we should be able to have that, now that the EJB3 tck looks in far better shape[17:03:20] *** asaldhan has joined #jboss-as7[17:03:20] *** ChanServ sets mode: +v asaldhan[17:03:22] <Jaikiran> wolfc: wdyt? ^[17:03:32] <Jaikiran> i mean considering the june 15 timeline[17:03:36] <darranl> Nihility, sure just adding some additional features to the core of it at the moment and will then be pulling the latest from kkhan to integrate on the native side[17:04:02] <wolfc> We don't have any good security interceptors yet[17:04:43] <tdiesler> Nihility, https://github.com/tdiesler/jboss-as/commit/6af3ca5907fbb0bbcc02de27cf18627f7619bcda[17:04:44] <jbossbot> git [jboss-as] 6af3ca5.. Thomas Diesler Upgrade to jbosgi-framework-1.0.0.CR3[17:07:11] <wolfc> Let's finish up the TCK first and then bring in more. In the mean time I thought darranl would push this forward?[17:07:14] *** magesh has quit IRC[17:08:17] <Nihility> wolfc: agreed, i was just asking (actually from a TCK perspective)[17:08:35] *** alexsmirnov has joined #jboss-as7[17:09:15] <Nihility> asaldhan: hey do you have a resource available to work on the ejb3 security interceptor?[17:09:40] <asaldhan> Nihility: wasnt it decided that DL will be doing it?[17:09:54] <Nihility> asaldhan: ah didnt know :)[17:09:59] <asaldhan> Nihility: I sent email 2 weeks ago[17:10:00] <Nihility> darranl: are you doing that?[17:10:09] <asaldhan> Nihility: check the internal jboss-as ML[17:10:12] <Nihility> asaldhan: sorry my email buffer was overflowed[17:10:14] <asaldhan> Nihility: response from Brino[17:10:17] <Nihility> oh that email list[17:10:18] <asaldhan> Nihility: np. :)[17:10:19] <darranl> Nihility, it is what I am going to be looking into[17:10:43] <Nihility> im not a fan of internal private mailing lists for dicussing OSS software[17:10:45] <Nihility> so i dont read it[17:10:48] <dmlloyd> darranl: when you get to that let me know, we need to discuss how association works for remote inv[17:10:49] <Nihility> :)[17:11:03] <Jaikiran> asoldano: quick question, with that latest test i see this error:[17:11:04] <Jaikiran> 20:36:50,150 ERROR [org.apache.catalina.core.ContainerBase.[jboss.web].[localhost].[/ws-endpoint-example].[TestService]] (http-localhost.localdomain-127.0.0.1-8080-3) Allocate exception for servlet TestService: java.lang.ClassCastException: org.jboss.as.testsuite.integration.ws.SimpleWebserviceEndpointImpl cannot be cast to javax.servlet.Servlet[17:11:21] <darranl> dmlloyd, which one - EJB3? Native Domain Management or both?[17:11:24] <Nihility> in case you guys are wondering[17:11:24] <asaldhan> Nihility: resourcing can be an internal debatable matter. :)[17:11:31] <Nihility> today im going through our "loose ends"[17:11:35] <dmlloyd> darranl: EJB security interceptors[17:11:39] <asoldano> Jaikiran, mmh... I don't see that locally...[17:11:42] <darranl> dmlloyd, ok[17:11:52] <Jaikiran> asoldano: that test passes for you locally?[17:12:02] <Jaikiran> i rebased against upstream by the way[17:12:10] <Nihility> asaldhan: word brother[17:12:11] <asoldano> Jaikiran, with a clone of your 852 branch, yes[17:12:11] <Nihility> :)[17:12:29] <asoldano> Jaikiran, with master, I have the injection error[17:12:35] <Nihility> asaldhan: hey does chicago office have a big meeting room[17:12:46] <asaldhan> Nihility: how many people?[17:13:01] <asaldhan> Nihility: pm[17:13:15] <Jaikiran> asoldano: the test adds that impl as a servlet class, so shouldn't it implement that interface?[17:13:35] <asoldano> Jaikiran, no, the WS subsystem rewrites the web.xml metadata at deploy time[17:14:59] <asoldano> Jaikiran, let me do another test, I actually updated the ws stack before running the test locally... it should make any difference, but...[17:15:05] <asoldano> s/should/shouldn't[17:15:21] <Jaikiran> asoldano: i just pushed my AS7-852 branch. give it a try if you have some time[17:15:22] <jbossbot> jira [AS7-852] No component found for type '...' [Pull Request Sent (Unresolved) Bug, Major, jaikiran pai] https://issues.jboss.org/browse/AS7-852[17:15:38] <Jaikiran> i mean the new test that you added[17:17:14] <asaldhan> darranl: the ejb3 interceptors will probably be simple to implement as long as you use the sec api. :)[17:17:28] *** Jaikiran is now known as Jaikiran|bbl[17:17:45] <darranl> ok[17:17:55] * asaldhan 's isp is having a fit[17:21:37] *** frainone has quit IRC[17:21:37] *** sguilhen has quit IRC[17:26:35] <asoldano> Jaikiran|bbl, I confirm, test passing locally[17:33:46] *** mmoyses is now known as mmoyses_[17:34:16] <smcgowan> kkhan: pull this please: https://github.com/smcgowan/jboss-as/commit/d087f3293fd88c7bbdb72727dfc51b470eb53f73[17:34:17] <jbossbot> git [jboss-as] d087f32.. Shelly McGowan JBCTS-1101, JBEE-81 - Fix for JSTL SQL Actions, Updates to Final API versions[17:34:17] <jbossbot> jira [JBCTS-1101] Redirected to: https://issues.jboss.org/login.jsp?permissionViolation=true&os_destination=%2Fsi%2Fjira.issueviews%3Aissue-xml%2FJBCTS-1101%2FJBCTS-1101.xml[17:34:18] <jbossbot> jira [JBEE-81] Release Final versions of APIs for inclusion in AS 7.0.0.Final [Resolved (Done) Release, Major, Shelly McGowan] https://issues.jboss.org/browse/JBEE-81[17:34:53] <Nihility> baileyje: https://issues.jboss.org/browse/AS7-887 is a small task for filler like you were asking for[17:34:54] <Nihility> :)[17:34:55] <jbossbot> jira [AS7-887] Support legacy ServiceMBeans in SAR deployments [Open (Unresolved) Enhancement, Major, Unassigned] https://issues.jboss.org/browse/AS7-887[17:35:00] <kkhan> smcgowan: Looking[17:37:38] *** sguilhen has joined #jboss-as7[17:38:03] <kkhan> smcgowan: history looks better, just doing a build and will merge once that is done[17:39:14] <smcgowan> kkhan: thank you, apologize for the inconvenience[17:39:21] <kkhan> np[17:39:56] *** frainone has joined #jboss-as7[17:39:56] *** ChanServ sets mode: +v frainone[17:43:13] <jbossbot> git [jboss-as] push master d087f32.. Shelly McGowan JBCTS-1101, JBEE-81 - Fix for JSTL SQL Actions, Updates to Final API versions[17:43:14] <jbossbot> jira [JBCTS-1101] Redirected to: https://issues.jboss.org/login.jsp?permissionViolation=true&os_destination=%2Fsi%2Fjira.issueviews%3Aissue-xml%2FJBCTS-1101%2FJBCTS-1101.xml[17:43:15] <jbossbot> jira [JBEE-81] Release Final versions of APIs for inclusion in AS 7.0.0.Final [Resolved (Done) Release, Major, Shelly McGowan] https://issues.jboss.org/browse/JBEE-81[17:43:15] <jbossbot> git [jboss-as] push master URL: http://github.com/jbossas/jboss-as/compare/ae6c41e...d087f32[17:43:37] <dmlloyd> sorry, I meant to push those but I got sidetracked[17:48:32] *** pferraro has joined #jboss-as7[17:48:32] *** ChanServ sets mode: +v pferraro[17:51:52] *** AndyTaylor has quit IRC[17:57:29] *** kevinpollet has quit IRC[17:58:00] *** sebersole has joined #jboss-as7[17:58:00] *** ChanServ sets mode: +v sebersole[17:58:25] *** asoldano is now known as asoldano_away[18:07:24] *** stansilvert has joined #jboss-as7[18:09:48] *** tdiesler has quit IRC[18:10:03] <mgoldmann> guys (and girls), another heads-up:[18:10:04] <mgoldmann> https://fedoraproject.org/wiki/JBossAS7[18:10:08] <mgoldmann> https://fedoraproject.org/wiki/JBossAS7DependencyTree[18:10:17] <mgoldmann> there is some progress now! :[18:10:19] <mgoldmann> :)[18:11:03] *** kcbabo has joined #jboss-as7[18:11:03] *** ChanServ sets mode: +v kcbabo[18:19:31] *** emuckenhuber has quit IRC[18:19:42] *** pferraro has quit IRC[18:20:11] <darranl> has there been a classloader change in standalone that would remove visibility or com.sun.jndi.ldap.LdapCtxFactory ?[18:20:15] *** pferraro has joined #jboss-as7[18:20:15] *** ChanServ sets mode: +v pferraro[18:21:25] *** jfclere has quit IRC[18:23:51] <dmlloyd> change? no[18:24:02] <dmlloyd> that class was never visible[18:25:04] <darranl> it definitely was loading it before[18:25:22] <darranl> weird[18:25:36] <dmlloyd> chances are, what changed was the TCCL[18:25:50] <dmlloyd> null TCCL = able to load classes from the system class path[18:27:32] *** pilhuhn is now known as pil-dinner-bbl[18:28:17] <darranl> could be two issues then, the domain HTTP server is not setting a TCCL but somehow from the thread from the executor one has been set[18:28:38] <dmlloyd> how are you loading the class?[18:29:48] <darranl> the class is being loaded by instantiating a javax.naming.directory.InitialDirContext[18:30:00] *** clebert has quit IRC[18:30:12] *** pferraro has quit IRC[18:30:44] <bobmcw> how do the cool kids write mbeans these days?[18:30:51] <bobmcw> there's like a brazillion options[18:31:04] *** kevinpollet has joined #jboss-as7[18:31:10] <bobmcw> as7 isn't liking my FooMBean interfaces, so I guess I'm doing it wrongly[18:31:24] <bobmcw> (trying MBeanRegistrationService)[18:32:14] <dmlloyd> darranl: would you mind pastebinning the stack trace?[18:32:35] *** clebert has joined #jboss-as7[18:32:35] *** ChanServ sets mode: +v clebert[18:32:38] <dmlloyd> bobmcw: if you want them to deploy automatically, you need a jboss-service.xml currently[18:33:03] <darranl> dmlloyd, here it is http://pastebin.com/mxdNtDhA[18:33:25] <bobmcw> dmlloyd: I don't know what that means? I'm currently standing up an MBeanRegistrationService, and catching this:[18:33:36] <bobmcw> https://gist.github.com/989076[18:33:42] <bstansberry> baileyje, dmlloyd: new-controllers now has ModelNodeRegistration using NewStepHandler[18:33:50] <bstansberry> buh-bye compilation[18:33:51] <bobmcw> I have a SharedRubyRuntimePoolMBean interface[18:34:17] <bobmcw> so it seems like it's trying to work[18:34:54] <bobmcw> oh, wait, I had my impls wrong[18:34:59] <bobmcw> PBKAC[18:35:07] <dmlloyd> darranl: where's the rest of it?[18:35:14] <dmlloyd> there should be at least one Caused By:[18:35:24] <dmlloyd> bstansberry: awesome![18:35:47] <darranl> dmlloyd, sorry here is the full trace http://pastebin.com/hBFF0hib[18:36:10] <dmlloyd> bobmcw: oh ok, you're just doing it a different way. If you do an AS4-style jboss-service.xml you can deploy MBeans as services, but using a reg service is fine too[18:36:16] <bstansberry> baileyje, dmlloyd: woops, maybe not; give me a minute[18:36:34] <bobmcw> yah, I originally had Foo impls BarMBean, but it didn't like that[18:36:40] <bobmcw> making Foo impls FooMBean seemed happy now[18:36:52] <bobmcw> AS6 didn't mind the mismatch[18:37:01] <dmlloyd> darranl: there's no more after that?[18:37:27] <darranl> dmlloyd, no InitialContextFactoryBuilder drops it !! - let me log the original exception as well[18:37:32] <bstansberry> baileyje, dmlloyd: k, now it's pushed[18:38:45] <dmlloyd> that's annoying.[18:39:13] *** frainone has quit IRC[18:39:15] <dmlloyd> darranl: well, try wrapping the code which instantiates that with a setContextClassLoader(null) for now[18:39:19] <dmlloyd> and we'll figure something out[18:39:25] <bobmcw> dmlloyd: got it working: https://skitch.com/bobmcwhirter/fbd1i/sun.tools.jconsole.jconsole[18:39:35] <bobmcw> weird classloader problem, but I can work through that[18:39:48] <dmlloyd> that's a jconsole issue[18:40:09] <bobmcw> ah, 'k[18:40:10] *** mmoyses_ is now known as mmoyses[18:40:16] <dmlloyd> the problem is you're sending ruby classes back[18:40:30] <bobmcw> yeah, should just toString() the sucker[18:40:39] <darranl> ok thanks dmlloyd will do that - FYI the missing exception is http://pastebin.com/jPDUmffV[18:40:39] <bobmcw> or otherwise make regular java[18:41:11] <dmlloyd> bobmcw: or use openmbean[18:41:32] <bobmcw> that's the cool way to do mbeans now?[18:42:03] <dmlloyd> bobmcw: if a class is a java bean, with the appropriate you can send them back even without class loaders[18:42:11] <dmlloyd> you might need mxbean magic for that[18:42:18] *** pferraro has joined #jboss-as7[18:42:18] *** ChanServ sets mode: +v pferraro[18:42:23] <dmlloyd> which is a magic way to transform POJO-y things to openmbean-y things[18:42:27] <bobmcw> maybe I'll revisit that later[18:42:39] *** hbraun has quit IRC[18:43:57] *** adietisheim has quit IRC[18:45:00] <dmlloyd> ah![18:45:10] <dmlloyd> darranl: that's kind of our fault[18:46:41] <darranl> dmlloyd, must be something else changed, setting the Classloader to null and I get a NPE at org.jboss.as.naming.InitialContextFactoryBuilder.createInitialContextFactory(InitialContextFactoryBuilder.java:55)[18:47:25] <dmlloyd> I think our InitialContextFactoryBuilder is not doing the right thing[18:47:34] <dmlloyd> granted, I'm not entirely certain what the right thing is[18:47:40] <dmlloyd> the NPE though is definitely a bug[18:47:44] <darranl> dmlloyd, I will double check something, maybe some of my own properties have not been set properly causing the InitialContextFactoryBuilder to do it's things[18:47:51] <dmlloyd> it should be using Class.forName, not cl.loadClass[18:49:13] <darranl> you mean it should be doing Class.forName when the cl is null or all the time?[18:49:54] <dmlloyd> all the time[18:53:34] *** kcbabo has quit IRC[18:54:10] *** hardy_ has quit IRC[18:55:01] <bobmcw> dmlloyd: is there some way, at runtime, for us to stop/start/stop/start an MSC Service<T>?[18:55:54] <dmlloyd> yeah lots of ways[18:56:05] *** hardy_ has joined #jboss-as7[18:56:05] <dmlloyd> it depends on what you're trying to do[18:56:10] <bbrowning> any way that I can do remotely or from a web app running inside AS?[18:56:14] <bobmcw> we want to be able to (temporarily) stop() a JMSTopicService, let's say[18:56:23] <dmlloyd> for testing purposes?[18:56:27] <bbrowning> yes[18:56:37] <dmlloyd> just open jconsole and go into the org.jboss.msc mbean[18:56:45] <bbrowning> automated test :)[18:56:47] <bobmcw> howabout non-testing purposes[18:56:59] <dmlloyd> the preferred mechanism is to use dependencies to control services[18:57:12] <bobmcw> this is like a user saying "yo, stop that service, please"[18:57:19] <bbrowning> aye[18:57:23] <bobmcw> and then deciding to turn it back on[18:57:31] <bobmcw> not something we can solve, per se[18:57:32] <dmlloyd> okay so if the service is something managed then the best way to do that is via an operation[18:57:42] <bobmcw> nope, our services, part of a user app deployment[18:57:48] <dmlloyd> if this is a service you've created, then just change the mode[18:57:57] <dmlloyd> Mode.NEVER = stop[18:57:59] <bobmcw> ServiceController.setMode(..)[18:58:01] <dmlloyd> yes[18:58:06] <bobmcw> that'll stop it, and dependencies?[18:58:14] <dmlloyd> yes BUT it's non-blocking[18:58:15] <bobmcw> setMode( ACTIVE ) and it goes again?[18:58:36] <dmlloyd> so setting the mode to NEVER gets everything to stop, dependent-order-first[18:58:42] *** jfclere has joined #jboss-as7[18:59:02] <dmlloyd> correct[18:59:13] <dmlloyd> if you set mode ACTIVE before everything is stopped, then everything will start again[18:59:26] <dmlloyd> there's a little bit of inertia in there though[18:59:35] <dmlloyd> takes a while to turn around a big ship, so to speak[18:59:40] <bobmcw> this is mostly leafs[18:59:50] <dmlloyd> okay, then it should be fine[18:59:59] <dmlloyd> just don't make the mistake of assuming the service is stopped after setMode returns[19:00:08] <bbrowning> how do I verify it's stopped?[19:00:13] <dmlloyd> service listener[19:00:56] <dmlloyd> as always, ensure that you're not performing blocking operations in the listener[19:02:05] <bobmcw> ffs, you and your non-blocking async lifestyle![19:02:09] *** ALR has joined #jboss-as7[19:02:12] *** ALR has quit IRC[19:02:12] *** ALR has joined #jboss-as7[19:02:12] *** ChanServ sets mode: +v ALR[19:02:39] *** hardy_ has quit IRC[19:04:10] <darranl> dmlloyd, none of that works, the class is always not found, do we have any other options just to get it visible for now?[19:05:27] <dmlloyd> wow, that's odd[19:05:58] <dmlloyd> baileyje: I think this is leaving my area of expertise...[19:06:26] <dmlloyd> darranl, setting TCCL to null was my option to just get it visible for now :)[19:07:05] <darranl> setting it to null - and using Class.forName has not effect ;-([19:07:20] <dmlloyd> same stack trace?[19:07:44] <baileyje> I am missing some context. What is the issue?[19:07:49] <dmlloyd> http://pastebin.com/jPDUmffV[19:07:52] <dmlloyd> baileyje: ^^[19:08:04] <dmlloyd> this happens when trying to create an LDAP initial context[19:08:20] <darranl> yes still going through the same org.jboss.modules.ConcurrentClassloader[19:09:26] <baileyje> It is using the context classloader[19:10:20] <baileyje> Which in the case does not have access to the Ldap factory[19:13:36] <darranl> being a com.sun class is this one we should not be bringing in from system ?[19:14:12] <dmlloyd> that's what I thought[19:14:27] <dmlloyd> but we set TCCL to null (and changed cl.loadClass to Class.forName()) and it still didn't work[19:14:37] <dmlloyd> darranl: you did use the 3-arg form of Class.forName() right?[19:15:06] <darranl> dmlloyd, that could be why it is not working ;-)[19:15:31] <dmlloyd> haha yeah that'd do it[19:15:43] <dmlloyd> I was thinking something really crazy was going on[19:17:07] <darranl> that does fix it ;-)[19:23:10] <darranl> dmlloyd, is this something we want to switch to or just use as a local workaround for the moment?[19:24:26] <dmlloyd> we definitely want to change that code to Class.forName()[19:24:54] <darranl> in the format I just used? if so I can add this in my next pull request[19:26:24] <darranl> bstansberry, may have found the issue you had with the LDAP based auth - just weird it was not triggered for me sooner[19:27:33] <bstansberry> darranl: good :)[19:28:32] *** bstansberry is now known as bstans_afk[19:28:49] <dmlloyd> darranl: yeah. And TCCL->null is a decent enough workaroudn for now[19:29:07] <dmlloyd> there will be an issue with user deployments that want to use LDAP though, which we need to resolve[19:30:00] <darranl> do you mean it should be passing the context classloader into Class.forName so still use that if set and for my case set it to null just before my call?[19:30:57] <dmlloyd> yeah transform classLoader.loadClass(x) to Class.forName(x, false, classLoader); adn then set TCCL to null around your call[19:31:17] <darranl> thanks understood - I will do that[19:32:53] <dmlloyd> no[19:32:55] <dmlloyd> wait![19:33:00] <dmlloyd> I didn't mean always use a null class loader[19:33:14] <dmlloyd> I meant use TCCL always, but use Class.forName() so that if it IS null it doesn't explode[19:33:22] <darranl> yeah I know, from your last comment I know what you mean now ;-)[19:33:25] <dmlloyd> oh okay[19:33:36] <dmlloyd> whew, just saw the JIRA descr and panicked :)[19:33:39] <darranl> just updated the Jira ;-)[19:33:45] <dmlloyd> ah I see it :)[19:34:09] <darranl> thanks for your help, I will get back to testing it[19:37:54] <darranl> mmoyses, let us know if you end up having any problems with the LDAP login modules, just had an issue with ClassLoading for the InitialDirContext that we need to work around, not sure if you will be affected as well[19:39:05] <mmoyses> darranl: thanks for the heads up. i will test those login modules[19:40:05] <darranl> mmoyses, I will be getting this fix in tomorrow and then the workaround I need to use is to set the context ClassLoader to null before creating the initial context, just need to perform some additional testing with that change[19:40:23] <darranl> https://issues.jboss.org/browse/AS7-892[19:40:25] <jbossbot> jira [AS7-892] InitialContextFactoryBuilder should use Class.forName passing in the ClassLoader instead of using it directly [Open (Unresolved) Task, Major, Darran Lofthouse] https://issues.jboss.org/browse/AS7-892[19:43:24] *** darranl is now known as darranl_afk[19:45:18] *** frainone has joined #jboss-as7[19:45:18] *** ChanServ sets mode: +v frainone[19:45:43] *** Jaikiran|bbl is now known as Jaikiran[19:48:00] <jamezp> dmlloyd: Can you double check this bit for me real quick? http://fpaste.org/ENsy/[19:48:55] <dmlloyd> looks good... *except*... you should not do the controller or controller-client modules for now[19:49:05] <dmlloyd> we're refactoring those pretty heavily[19:50:17] <jamezp> Ah, crap :-([19:50:28] *** emuckenhuber has joined #jboss-as7[19:50:28] *** ChanServ sets mode: +v emuckenhuber[19:51:03] <jamezp> I have 1,255 line message bundle already.[19:52:17] <dmlloyd> ah...[19:52:24] <dmlloyd> well hang on to it, you can rebase it later[19:52:29] <dmlloyd> it'll probably be... mostly correct :)[19:52:32] *** bstans_afk is now known as bstansberry[19:53:32] <jamezp> dmlloyd: I will for sure. I'm actually almost done. It's rather large :-)[19:54:23] <jamezp> BTW, fpaste command for Fedora awesome. Thanks for sharing that yesterday.[19:55:29] <dmlloyd> better living through technology[19:56:21] <jamezp> Indeed.[19:58:11] *** tcrawley has quit IRC[20:03:36] *** jwulf has quit IRC[20:04:36] <jbossbot> git [jboss-as] push master 0aaead0.. jaikiran AS7-852 Fix env-entry and its corresponding @Resource handling[20:04:38] <jbossbot> jira [AS7-852] No component found for type '...' [Pull Request Sent (Unresolved) Bug, Major, jaikiran pai] https://issues.jboss.org/browse/AS7-852[20:04:38] <jbossbot> git [jboss-as] push master bd4c849.. jaikiran AS7-852 Implement @Resource processing for WebServiceContext type[20:04:38] <jbossbot> git [jboss-as] push master a2c2eeb.. asoldano [AS7-852] Adding non-ejb3 test[20:04:38] <jbossbot> git [jboss-as] push master URL: http://github.com/jbossas/jboss-as/compare/d087f32...a2c2eeb[20:06:16] *** pmuir has quit IRC[20:09:21] *** pmuir has joined #jboss-as7[20:09:21] *** ChanServ sets mode: +v pmuir[20:10:55] *** Jaikiran has quit IRC[20:14:54] <jpederse> mgoldmann: kick ass ! congrats ![20:15:09] <mgoldmann> jpederse: thanks! :)[20:19:04] <dmlloyd> mgoldmann: cool.. be sure to get me in the loop for packaging jboss as with fedora though - there's some special stuff with modules we should be doing[20:19:37] <mgoldmann> dmlloyd: can you say more?[20:20:06] <dmlloyd> well it'd be nice to have a better correlation between modules and RPMs for one thing[20:20:26] <mgoldmann> ha[20:20:27] <dmlloyd> and we should try to do it in such a way that it dovetails into our EAP productization process[20:20:48] *** davidbos has quit IRC[20:20:53] <mgoldmann> we were talking that it would be nice to have moles related to rpms so we can package some stuff earlier and some later[20:20:57] <mgoldmann> but still having AS working[20:21:27] <mgoldmann> dmlloyd: so, I'm open for advices :)[20:22:33] <dmlloyd> do the fedora people like to build from source always?[20:22:40] <mgoldmann> yes[20:22:44] <mgoldmann> only from sourse[20:22:47] <mgoldmann> source*[20:22:58] <mgoldmann> we cannot bundle anything[20:23:58] <mgoldmann> so, if you add a new dependency - I'll need to package it into Fedora[20:24:13] <mgoldmann> (or I'm lucky and it;s alredy there)[20:24:23] <mgoldmann> crap, I cannot write today[20:25:05] <dmlloyd> the existing deps are tricky because we certify AS with a specific set of JARs and I think if we start upgrading them we might not be certified anymore...[20:25:27] <mgoldmann> dmlloyd: so, with FEdora the goal is to use always newest versions[20:25:28] <dmlloyd> also it's a bit fuzzy if I have AS 7.0.2 and I update, say, apache-commons-io, is it still AS 7.0.2[20:25:28] <dmlloyd> ?[20:25:33] <dmlloyd> right[20:26:04] <mgoldmann> it's not always possible, but you cannot be sure you'll have always exact version[20:26:49] <dmlloyd> so we have to know what that means for certification. In Fedora then, AS 7.0.2 == the core AS modules are 7.0.2, but the dependencies could be ???. Differentiates from the AS ZIP distribution which contains fixed dependencies.[20:27:07] <mgoldmann> yes, that's true[20:27:18] <dmlloyd> but I guess it's fairly similar to something like apache httpd, which has its own version but links against common .so libs which could be ???[20:27:34] <mgoldmann> good analogy I think[20:28:03] <mgoldmann> we can make also -compat packages, but Fedora is not happy to review or include them[20:28:38] <dmlloyd> the thing I'd really like to accomplish though is to have it set up so that users can just run a single command like "modular-java" and be able to boot java via jboss-modules.jar, with all the installed distro modules available to them[20:29:20] <dmlloyd> including everything AS provides[20:30:30] <dmlloyd> or make a wrapper script for the openjdk/icedtea "java" command[20:30:30] *** lazarotti has joined #jboss-as7[20:30:33] <dmlloyd> one of the two[20:30:43] <dmlloyd> whatever we can get away with :)[20:31:06] <mgoldmann> creating wrapper is simple[20:32:06] <mgoldmann> wrapper as a wrapper, I'm not talking about implementing your idea now :)[20:32:29] *** kcbabo has joined #jboss-as7[20:32:29] *** ChanServ sets mode: +v kcbabo[20:32:45] *** mlinhard has joined #jboss-as7[20:32:45] *** ChanServ sets mode: +v mlinhard[20:33:06] <mgoldmann> dmlloyd: feel free to join #fedora-java[20:33:30] <mgoldmann> there are some guys really interested in JBoss projects and are hapilly to tell you what is possible, what not[20:33:38] <mgoldmann> I'm really fresh to this topic[20:33:38] <dmlloyd> cool[20:33:46] <mgoldmann> (previously packaged rubygems)[20:33:55] <mgoldmann> (which is different story...)[20:34:11] <dmlloyd> unlike previous ASes, if AS7 was in Fedora I'd probably always use the Fedora version for my personal stuff![20:47:11] *** pil-dinner-bbl is now known as pilhuhn[20:54:06] *** kcbabo has quit IRC[20:54:24] *** kevinpollet has quit IRC[20:57:11] *** galderz has quit IRC[20:57:56] *** sannegrinovero has quit IRC[21:00:28] *** jamezp is now known as jamezp_afk[21:03:12] <jbossbot> git [jboss-as] push master bbb38b9.. Paul Gier Remove source plugin configuration because it is duplicated in jboss-parent pom.[21:03:12] <jbossbot> git [jboss-as] push master URL: http://github.com/jbossas/jboss-as/compare/a2c2eeb...bbb38b9[21:07:59] *** opalka has joined #jboss-as7[21:07:59] *** opalka has joined #jboss-as7[21:07:59] *** ChanServ sets mode: +v opalka[21:11:23] *** kcbabo has joined #jboss-as7[21:11:23] *** ChanServ sets mode: +v kcbabo[21:15:35] <pilhuhn> Hey[21:16:45] <pilhuhn> Would it make sense to unite all file appenders under one file-appender node in teh management tree and have a switch to determine which specific version to use?[21:17:09] <dmlloyd> can you rephrase that?[21:17:55] *** torben has joined #jboss-as7[21:17:55] *** torben has quit IRC[21:17:55] *** torben has joined #jboss-as7[21:17:55] *** ChanServ sets mode: +v torben[21:18:27] <pilhuhn> If I do[21:18:28] <pilhuhn> [standalone@localhost:9999 subsystem=logging] ls[21:18:28] <pilhuhn> periodic-rotating-file-handler size-rotating-file-handler logger file-handler[21:18:28] <pilhuhn> async-handler console-handler[21:18:39] <pilhuhn> I see 3 *file-handler subtrees[21:18:54] <pilhuhn> where each does almost exactly the same as the others.[21:19:35] <pilhuhn> I think it could make sense to only have the node 'file-handler' and then an attribute on it "plain flavor", "periodic", "size"[21:19:45] <dmlloyd> hmm that is odd[21:19:58] <dmlloyd> well it makes sense though, to an extent[21:20:19] <dmlloyd> within the <logging> tag there's a different handler element for each type[21:20:56] <pilhuhn> They are not exactly the same. But config is the same to 90%. Only difference is "nothing provided", "size provided" and "time given"[21:21:00] <dmlloyd> I'm not sure the entities are similar enough[21:21:08] <dmlloyd> yeah that could work[21:21:12] <dmlloyd> can you work up a patch for that?[21:21:15] <dmlloyd> actually[21:21:21] <dmlloyd> wait a week, *then* work up a patch :)[21:21:30] <pilhuhn> lol[21:22:05] <pilhuhn> I can try - I would like to finally have contributed a piece of code to AS7 and not only bug reports[21:26:59] *** kevinpollet has joined #jboss-as7[21:27:42] <jbossbot> git [jboss-as] push master d9ce8dd.. Scott M Stark AS7-893, use hasDefined(ALIAS) to avoid IllegalArgumentException error when...[21:27:43] <jbossbot> jira [AS7-893] web subsystem virtual-server is requiring an alias to succeed [Open (Unresolved) Bug, Major, Scott Stark] https://issues.jboss.org/browse/AS7-893[21:27:43] <jbossbot> git [jboss-as] push master URL: http://github.com/jbossas/jboss-as/compare/bbb38b9...d9ce8dd[21:28:19] <rmaucher> thanks scott[21:33:23] *** kkhan has quit IRC[21:37:13] <jbossbot> git [jboss-as] push master ae8c859.. unknown Add missing *.conf.bat files so that the server starts with the same options on linux and windows[21:37:13] <jbossbot> git [jboss-as] push master URL: http://github.com/jbossas/jboss-as/compare/d9ce8dd...ae8c859[21:51:09] *** adietisheim has joined #jboss-as7[21:56:11] *** kcbabo has quit IRC[21:58:27] *** mgoldmann has quit IRC[22:01:06] <jbossbot> git [jboss-as] push master e4e80c3.. bstansberry at jboss dot com Fix javadoc links[22:01:06] <jbossbot> git [jboss-as] push master 43ce9a4.. bstansberry at jboss dot com Refactor some infrastructure stuff out of domain mode test classes[22:01:06] <jbossbot> git [jboss-as] push master 3b6b04d.. bstansberry at jboss dot com [AS7-853] Fix deployment handling issues, add domain mode deployment tests...[22:01:07] <jbossbot> jira [AS7-853] :add deploment on server group with enabled=true fails [Open (Unresolved) Bug, Major, Brian Stansberry] https://issues.jboss.org/browse/AS7-853[22:01:07] <jbossbot> git [jboss-as] push master URL: http://github.com/jbossas/jboss-as/compare/ae8c859...3b6b04d[22:01:16] *** jamezp_afk is now known as jamezp[22:10:46] *** emuckenhuber has quit IRC[22:11:04] *** opalka has quit IRC[22:11:31] *** sebersole has quit IRC[22:16:20] *** galderz has joined #jboss-as7[22:16:20] *** ChanServ sets mode: +v galderz[22:16:31] *** galderz has quit IRC[22:17:53] *** pilhuhn has quit IRC[22:18:44] <jbossbot> git [jboss-as] push master 7eec7a6.. Alexey AS7-786 fix completion of file paths starting with disc root on windows, disable escaping for command argument values on windows[22:18:45] <jbossbot> jira [AS7-786] CLI upload doesn't like Windows paths [Resolved (Done) Bug, Major, Alexey Loubyansky] https://issues.jboss.org/browse/AS7-786[22:18:45] <jbossbot> git [jboss-as] push master URL: http://github.com/jbossas/jboss-as/compare/3b6b04d...7eec7a6[22:20:30] *** asoldano_away is now known as asoldano[22:25:45] *** adietisheim has quit IRC[22:29:48] *** kcbabo has joined #jboss-as7[22:29:48] *** ChanServ sets mode: +v kcbabo[22:31:57] *** torben has quit IRC[22:33:04] *** adietisheim has joined #jboss-as7[22:34:11] *** smcgowan has quit IRC[22:34:16] *** pmuir has quit IRC[22:36:01] *** pferraro has left #jboss-as7[22:37:26] *** pferraro has joined #jboss-as7[22:37:26] *** ChanServ sets mode: +v pferraro[22:37:42] *** pferraro has left #jboss-as7[22:38:17] *** rmaucher has left #jboss-as7[22:41:12] *** smarlow has quit IRC[22:41:47] *** asoldano has quit IRC[22:42:00] *** balunasj_mtg has quit IRC[22:42:09] *** jpederse has quit IRC[22:43:41] *** maeste has quit IRC[22:44:37] *** mlinhard has quit IRC[22:53:10] *** mbg has quit IRC[23:00:27] *** jwulf has joined #jboss-as7[23:04:26] *** misty has joined #jboss-as7[23:07:52] *** mmoyses has quit IRC[23:10:49] <jamezp> Anyone else using IDEA 10.5 on Fedora?[23:14:29] <dmlloyd> yes[23:15:03] <jamezp> Does it seem slower to you? Seems a lot slower to me.[23:15:22] <asaldhan> jamezp: idea by nature is slow. try eclipse.[23:15:37] <jamezp> Especially in bigger projects like jboss-as.[23:16:11] *** jwulf has quit IRC[23:16:23] <jamezp> asaldhan: Last time I used eclipse I wanted to throw my laptop out the window :-)[23:16:42] <jamezp> m2eclipse kept locking up my MacBook.[23:17:12] <asaldhan> jamezp: last time I used idea, it could not port my license key (mac to win). I threw it out the window.[23:17:22] <dmlloyd> jamezp: he's trolling you.[23:17:23] *** mbg has joined #jboss-as7[23:17:23] *** ChanServ sets mode: +v mbg[23:17:41] <dmlloyd> jamezp: yeah 10.5 uses a really slow completion algorithm[23:17:45] <dmlloyd> it's brand new[23:17:54] <dmlloyd> tbh I liked the old one better[23:18:17] <jamezp> asaldhan: That's why I use the Community Edition ;-)[23:19:01] <jamezp> dmlloyd: Me too. I keep reading on Twitter how others find it much faster.[23:19:14] <dmlloyd> it is faster sometimes[23:19:22] *** sgilda has quit IRC[23:19:23] <dmlloyd> I think it has like a caching algorithm or something[23:19:31] *** wolfc has quit IRC[23:20:00] <jamezp> Could be, seems to be after a build it starts slowing down.[23:22:46] *** kcbabo has quit IRC[23:23:56] *** kevinpollet has quit IRC[23:25:15] *** adietisheim has quit IRC[23:30:10] *** kcbabo has joined #jboss-as7[23:30:10] *** ChanServ sets mode: +v kcbabo[23:30:16] *** jc3` has joined #jboss-as7[23:30:39] *** jc3 has quit IRC[23:38:25] *** sgilda has joined #jboss-as7[23:38:56] *** clebert has quit IRC[23:40:29] <dmlloyd> you guys may be interested to know that I'm presently running maven through jboss modules[23:40:44] <dmlloyd> mainly as an experiment to see if anything shakes loose[23:40:54] <dmlloyd> seems to work fine though[23:40:55] <bobmcw> replacing classworlds?[23:41:02] <dmlloyd> no, just encapsulating it[23:41:04] <bobmcw> wtf wrote classworlds?[23:41:14] <bobmcw> dmlloyd: Nihility tell you about that?[23:41:22] <dmlloyd> yeah I know! there hasn't been a release since like 1985![23:41:31] <dmlloyd> yeah I forgot to tell him that was yours :)[23:41:33] <bobmcw> "We don't know much about classworlds, or who wrote it"[23:41:36] <dmlloyd> haha[23:41:37] <bobmcw> "Uh, I wrote it, Jason..."[23:42:07] <bobmcw> it was ahead of its time![23:42:30] <bobmcw> screw classloader hierarchies! directed graphs ftw![23:42:39] <dmlloyd> hip hip![23:42:51] <jamezp> hooray![23:45:03] <dmlloyd> I'm not sure that it'd be quite so easy to replace classwords in maven, or desirable even[23:45:25] <dmlloyd> but by running maven inside of modules it opens another option for embedded testing of modular stuff[23:45:33] <dmlloyd> makes runtime linking much easier[23:50:25] <jbossbot> git [jboss-as] push master 49bd34c.. jfclere fix multicast parsing[23:50:25] <jbossbot> git [jboss-as] push master 6d11b01.. jfclere initial mod_cluster integration work[23:50:25] <jbossbot> git [jboss-as] push master fe3c53d.. jfclere Arrange parser logic.[23:50:25] <jbossbot> git [jboss-as] push master 0ebb87a.. jfclere Arrange mod_cluster schema and logic.[23:50:25] <jbossbot> git [jboss-as] push master 2d0ee07.. jfclere Arrange code after a rebase.[23:50:25] <jbossbot> git [jboss-as] push master b1c2bd5.. jfclere Arrange modcluster part.[23:50:26] <jbossbot> git [jboss-as] push master 0a5236a.. jfclere Complet the modcluster logic.[23:50:26] <jbossbot> git [jboss-as] push master bead92b.. jfclere Arrange the advertise-socket (that is just a reference).[23:50:26] <jbossbot> git [jboss-as] push master 6d8b028.. jfclere Remove empty entries.[23:50:27] <jbossbot> git [jboss-as] push master a3a5477.. jfclere Arrange version.[23:50:27] <jbossbot> git [jboss-as] push master 90e5b71.. jfclere Add missing metrics.[23:50:28] <jbossbot> git [jboss-as] push master 9aaef29.. jfclere Add addCustomLoadMetrics().[23:50:28] <jbossbot> git [jboss-as] push master 1417856.. jfclere Remove the jmx logic for the web part.[23:50:29] <jbossbot> git [jboss-as] push master 223d260.. jfclere Remove the MBeanServer.[23:50:29] <jbossbot> git [jboss-as] push master e209d07.. jfclere Arrange version.[23:50:30] <jbossbot> git [jboss-as] push master 4538dfb.. jfclere Add list-proxies command.[23:50:30] <jbossbot> git [jboss-as] push master 7629439.. jfclere Use a released version of mod_cluster.[23:50:31] <jbossbot> git [jboss-as] push master 437f9c2.. jfclere Make the schema simpler: now it fits to the actual mod_cluster doc page.[23:50:31] <jbossbot> git [jboss-as] push master e0de662.. jfclere Add test for mod_cluster configuration.[23:50:32] <jbossbot> git [jboss-as] push master 8de5115.. jfclere move all the parsing logic in separate file.[23:50:32] <jbossbot> git [jboss-as] push master 63a784a.. jfclere oops mod_cluster was there twice...[23:50:33] <jbossbot> git [jboss-as] push master 5b484a7.. jfclere Arrange version.[23:50:33] <jbossbot> git [jboss-as] push master c70e0ac.. jfclere Port to next version.[23:50:34] <jbossbot> git [jboss-as] push master 406f42d.. jfclere Don't use catalina we removed it in jbossweb :D[23:50:34] <jbossbot> git [jboss-as] push master 927c987.. jfclere Arrange build.[23:50:35] <jbossbot> git [jboss-as] push master 327ffef.. jfclere remove traces.[23:50:35] <jbossbot> git [jboss-as] push master 87e9108.. jfclere typo[23:50:36] <jbossbot> git [jboss-as] push master 0975ba8.. jfclere Add missing description.[23:50:36] <jbossbot> git [jboss-as] push master 2cd9d94.. bstansberry at jboss dot com Include mod-cluster subsys in domain mode testsuite config[23:50:37] <jbossbot> git [jboss-as] push master URL: http://github.com/jbossas/jboss-as/compare/7eec7a6...2cd9d94[23:51:44] <misty> that was impressive[23:53:41] <misty> dmlloyd: who do I talk to about what bits of Transactions will be exposed in EAP6?[23:53:44] *** jc3` is now known as jc3[23:53:58] <misty> I've talked to Jonathan a bit but it seems like the details are not clear yet[23:54:08] *** bstansberry is now known as bstans_afk[23:54:30] <misty> I am trying to get a feel for what bits about JTS need to be documented for EAP[23:54:45] <dmlloyd> hmm it's hard to say[23:54:51] <dmlloyd> Jonathan is your guy overall though[23:55:14] <misty> well he keeps talking about a "they"[23:55:30] <misty> "They are not exposing that in EAP6, I don't know what They want to do with that in EAP6"[23:55:45] <misty> so he is a good resource but I really need to talk to They too[23:55:50] <dmlloyd> well sounds like we need to get him in the same room as They[23:55:56] <misty> yeah :D[23:56:08] * dmlloyd wonders if he is one of They[23:56:16] <misty> if you have to ask ...[23:58:06] <misty> so one of the biggest things is to get a handle on what users will have to / be able to configure in EAP[23:58:42] <misty> so far it seems like they do not need to worry about ObjectStore, recovery, or ORB (but apparently YOU guys need to worry about ORB quite a lot!)