March 16, 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 | >
March 16, 2011 NOTICE: This channel is no longer actively logged.
[00:00:06] <frainone> maybe the problem involves my commit, the one I wanted to add to AS[00:03:12] <frainone> the actual cause of the NPE is at line 64[00:05:14] <frainone> at JavaNamespaceSetup, line 51, namespaceSelector.getValue() is null[00:06:27] <frainone> dmlloyd: I don't think this has to do with my commit, but I think that maybe it is better to skip it for this release[00:06:42] <frainone> and, then, check calmly, with time, to see what is going wrong here[00:07:10] <frainone> the commit is simple and not worth the extra work at this point[00:10:06] <frainone> dmlloyd: you can check yourself if you want: https://github.com/fl4via/jboss-as/commit/57f91581689870256cd2ffa11ae2639c2c0e6827[00:10:07] <jbossbot> git [jboss-as] 57f9158.. Flavia Rainone Add support to @WebServiceRef annotation on classes, and improve @WebServiceRef demo test appliction[00:10:22] <frainone> there is some work that is ongoing and not ready yet, so this commit is pretty simple[00:11:38] *** alesj has quit IRC[00:11:48] <dmlloyd> okay[00:12:17] <dmlloyd> this fixes the test?[00:13:20] <frainone> if you're talking about the NPE, nope, it doesn't. The NPE is caused by JavaNamespaceSetup, line 51, and I would need to investigate further to discover why is namespaceSelector.getValue()[00:14:25] <frainone> this is the commit that I wanted you to add, but now, with the NPE, and given we are all tired, I think we'd better skip it than spend time investigating it now[00:15:35] <frainone> that commit is not rebased against the lated version,as after I rebased, I got the NPE... that commit was done before rebasing (I got no test errors at that time)[00:15:48] <dmlloyd> where are you getting the NPE? I can't seem to duplicate it[00:15:49] <frainone> s/lated/latest[00:15:58] <dmlloyd> oh I see[00:16:01] <frainone> dmlloyd: I"ll paste it[00:16:03] <dmlloyd> so this commit introduces the NPE apparently?[00:16:57] <frainone> http://fpaste.org/xr9P/[00:17:14] <frainone> if added to an updated version of jboss-as, apparently yes[00:24:09] <dmlloyd> #L1L196 I guess?[00:24:42] <dmlloyd> well anyway, I'll just go with what we have now I guess[00:24:42] <frainone> dmlloyd: ?[00:24:48] <frainone> yes, this is the best to do[00:25:17] <dmlloyd> I'll probably have to tag after dinner though, it's almost done cooking...[00:25:20] <frainone> the commit is too insignificant to justify any effort on fixing the NPE now[00:26:23] *** fnasser has quit IRC[00:28:50] <frainone> dmlloyd: ok, I'm off now... I got a fever and I'd better rest so I can be better tomorrow[00:30:25] *** frainone is now known as frainone_afk[00:32:10] *** stuartdouglas has quit IRC[00:32:29] *** asaldhan has quit IRC[00:36:14] *** maxandersen has quit IRC[00:48:19] *** asaldhan has joined #jboss-as7[00:49:56] *** jwulf has joined #jboss-as7[00:55:19] *** emmanuel has quit IRC[00:55:49] *** stuartdouglas has joined #jboss-as7[01:02:13] <dmlloyd> okay no problem frainone_afk[01:06:16] *** smcgowan has joined #jboss-as7[01:06:16] *** ChanServ sets mode: +v smcgowan[01:09:22] <maeste> bstans_afk, dmlloyd , baileyje : could someone have a look at https://github.com/maeste/jboss-as/commit/a5449e843d5046af53d4e16cc24e10ec0bb2c171 I've added description to datasources and resource-adapters subsystem for admin console ans kkhan asked me today.[01:09:23] <jbossbot> git [jboss-as] a5449e8.. Stefano Maestri read-resource-description support for datasources and resource-adapters subsystem[01:09:27] *** aslak has quit IRC[01:13:21] *** starksm64 has joined #jboss-as7[01:15:41] *** jpearlin has quit IRC[01:21:13] *** jpearlin has joined #jboss-as7[01:21:28] *** starksm64 has quit IRC[01:22:14] <maeste> well, it's late night here. head off to bed.[01:23:03] *** maeste is now known as maeste_zzzz[01:28:13] <dmlloyd> okay. Since brian/kabir/etc aren't here to OK that patch I think I'll hold off on it until after the tag.[01:32:18] *** stuartdouglas has quit IRC[01:54:43] <dmlloyd> you always find interesting issues when you build under a new version name for the first time[01:54:52] <dmlloyd> [ERROR] Failed to execute goal on project jboss-as-domain-controller: Could not resolve dependencies for project org.jboss.as:jboss-as-domain-controller:jar:7.0.0.Beta1: Could not find artifact org.jboss.as:jboss-as-domain-client:jar:7.0.0.Beta1 in jboss (https://repository.jboss.org/nexus/content/repositories/public) -> [Help 1][01:54:54] <dmlloyd> \o/[01:56:01] *** mbg is now known as mbg|away[01:56:04] *** mbg|away is now known as mbg[02:11:30] <baileyje> how are we ding[02:11:37] <jbossbot> git [jboss-as] push master 3dac07f.. David M. Lloyd POM cleanups[02:11:38] <jbossbot> git [jboss-as] push master f336625.. David M. Lloyd Prep 7.0.0.Beta1[02:11:38] <jbossbot> git [jboss-as] push master 6866906.. David M. Lloyd Next version will be 7.0.0.Beta2[02:11:38] <jbossbot> git [jboss-as] push master URL: http://github.com/jbossas/jboss-as/compare/ea971e1...6866906[02:12:17] <dmlloyd> hopefully this turned out right...[02:13:08] *** asaldhan has left #jboss-as7[02:17:04] <dmlloyd> and that's that, I suppose[02:17:21] <dmlloyd> have a good night everyone[02:17:25] <dmlloyd> and thanks![02:17:39] <dmlloyd> tomorrow it's back to work on Beta2 - the deadline is coming up fast![02:18:58] <baileyje> dmlloyd: Yeah. My first task is to actually use model operations to deploy datasources.[02:19:09] <baileyje> Anyway. good night dmlloyd thanks for the hard work[02:19:32] <dmlloyd> thank you for fixing that DS thing.[02:20:16] <dmlloyd> (oh, and all the other stuff too!) :)[02:20:31] <baileyje> ha[02:21:46] <jpearlin> wow[02:21:55] <jpearlin> oops...sorry...wrong window ;)[02:25:26] <smcgowan> baileyje: is it too late for a question? i saw your changes to the datasource, but when configuring derby, get: a sealing violation[02:26:01] <baileyje> sealing violation eh. Not sure what would cause that. Is it added as a module?[02:26:23] <smcgowan> yes, i have it added as a module and did not add it as a user deployment - is that still needed?[02:26:25] <smcgowan> http://pastebin.com/4vwVj935[02:26:37] <smcgowan> adding the .jars as a user deployment[02:26:57] <baileyje> With the change you should not need to add as a deployment. just list as a jdbc driver in the config[02:28:06] <smcgowan> i wonder if i'm now conflicting with derby in the JDK?[02:28:14] <smcgowan> here's my config: http://pastebin.com/LdaruZ01[02:29:17] <baileyje> and the jar is not a deployment in your server?[02:29:50] <baileyje> I think the sealing issue would be if there were multiple jars being loaded with the sealed package. But I am not really clear.[02:30:38] <smcgowan> the derby jars are not a deployment in my server[02:31:40] <baileyje> Can I see the module config for derby?[02:32:28] <smcgowan> baileyje: sure[02:33:08] <smcgowan> http://pastebin.com/v3M5H0Si[02:34:16] <smcgowan> i've been in meetings all day just wanted to look at this update and also the JPA changes[02:36:37] <smcgowan> so i see that derbyclient.jar and derby.jar are in usr/jdk1.6.0_21/db/lib[02:40:40] *** stuartdouglas has joined #jboss-as7[02:40:40] *** ChanServ sets mode: +v stuartdouglas[02:49:09] *** jpearlin has left #jboss-as7[02:58:51] <smcgowan> baileyje: i still have a conflict but am able to get rid of the sealing exception once i remove derbyclient.jar from the module[02:59:30] <baileyje> Do the two jars have duplicate classes?[03:03:58] <smcgowan> once I removed the derbyclient.jar i was getting an DuplicateException version 10.4 was already installed[03:04:07] <smcgowan> my jars are 10.5 though[03:04:29] <smcgowan> so i'm just cleaning and going to reconfigure the module[03:04:50] <baileyje> weird. It gets the version from the driver class itself[03:06:47] <smcgowan> maybe i should get back to this in the morning :)[03:15:25] <baileyje> Yeah. Not a bad idea[03:17:01] <smcgowan> baileyje: i got rid of the exceptions but the datasources have missing dependencies - so ya need to think about this[03:24:06] *** JimMa has joined #jboss-as7[03:43:03] *** bgeorges has joined #jboss-as7[03:43:03] *** ChanServ sets mode: +v bgeorges[04:06:21] *** smcgowan has quit IRC[04:18:36] *** bgeorges has quit IRC[04:27:39] *** frainone_afk has quit IRC[04:30:47] *** bgeorges has joined #jboss-as7[04:30:47] *** ChanServ sets mode: +v bgeorges[04:37:38] *** bobmcw has quit IRC[04:58:42] *** Nihility has joined #jboss-as7[04:58:43] *** Nihility has joined #jboss-as7[04:58:43] *** ChanServ sets mode: +v Nihility[04:58:48] *** Nihility has quit IRC[04:59:14] *** Nihility has joined #jboss-as7[04:59:15] *** Nihility has joined #jboss-as7[04:59:15] *** ChanServ sets mode: +v Nihility[04:59:21] <Nihility> saw the good news[05:00:39] <baileyje> Nihility: We have Driver deployment from modules now[05:00:48] <Nihility> yeah i saw that[05:01:01] <Nihility> i reread the irc log[05:01:06] <Nihility> that took awhile :)[05:01:17] <Nihility> btw the classloader spi[05:01:26] <Nihility> has nothing to do wit the parser[05:01:37] <baileyje> I had to get a crash course in getting all the descriptions and such lined up for new model ops[05:01:42] <baileyje> Yeah. I tried to tell him that..[05:01:48] <baileyje> totally unrelated.[05:02:03] <Nihility> the agreement was just about how to "share" the jdbc code you pulled in with IJ[05:02:18] <Nihility> making it be a separate jar[05:02:24] <baileyje> They DS elements are all parsed in IJ and they still reference "module" in the DS description[05:02:35] <Nihility> has zilch to do with xml parsing[05:02:52] <baileyje> anyway it worked well.[05:02:57] <baileyje> I still need to javadoc it all[05:03:15] <baileyje> I also plan to fix the way DSs are deployed. Actually make them operations.[05:03:28] <baileyje> will do that tomorrow[05:05:11] <Nihility> stuartdouglas: ping[05:05:49] <stuartdouglas> Nihility: pong[05:08:08] *** rawblem has quit IRC[05:09:20] <Nihility> but yeah drivers from modules is kick ass![05:13:38] <stuartdouglas> Nihility: you pinged?[05:18:43] <Nihility> stuartdouglas: yeah i just wanted to get a summary of what weld stuff is in the tag[05:19:08] <Nihility> i am closer to your timezone btw[05:19:16] <Nihility> (in Vegas for the wedding)[05:19:16] <stuartdouglas> were are you?[05:19:18] <stuartdouglas> ah[05:19:38] <Nihility> well closer from the hour perspective[05:19:44] <stuartdouglas> very basic EJB integration, so CDI can create EJB's and inject them[05:19:44] <Nihility> your still a day off[05:19:45] <Nihility> haha[05:20:03] <Nihility> i also saw JPA in irc logs[05:20:07] <stuartdouglas> however interceptors are not working (as in CDI interceptors are not being applied to EJB's)[05:20:12] <stuartdouglas> JPA works[05:20:13] <Nihility> does that work?[05:20:17] <Nihility> cool[05:20:22] <stuartdouglas> 100% until proven otherwise :-)[05:20:32] <Nihility> also do either of you know if ejbs in web work[05:20:41] <stuartdouglas> not sure[05:20:49] <stuartdouglas> CDI injection into web components work[05:21:06] <stuartdouglas> The ejb integration is the only real iffy bit[05:21:16] <stuartdouglas> the TCK is currently at 90%[05:21:31] <stuartdouglas> but when I sort out some JSF 1.2 issues to will probably go to around 95[05:21:34] <Nihility> baileyje: said something like "if a web thing is a component it gets @EJB"[05:21:44] <stuartdouglas> web things are components[05:21:53] <baileyje> stuartdouglas: Is that merged in?[05:21:54] <stuartdouglas> and they have @Resource injection[05:22:03] <stuartdouglas> yea, that has been in for ages[05:22:14] <baileyje> are they being added as EE components?[05:22:14] <stuartdouglas> as in, 2 weeks[05:22:34] <baileyje> I must have missed that when I was out[05:22:39] <stuartdouglas> which is ages compared to how long everything else has been in :-)[05:22:42] <Nihility> btw release codename david came up with[05:22:44] <stuartdouglas> yea, you were on holidays[05:22:46] <Nihility> "Ides of March"[05:22:52] <Nihility> which is bad ass imo[05:23:09] <Nihility> and fits alphas calendar theme[05:23:17] <baileyje> Nihility: I am into that.[05:23:20] <baileyje> Good name for sure.[05:23:24] <baileyje> or just "Ides"[05:24:11] <Nihility> yeah maybe just Ides[05:24:15] <baileyje> Or.. "The Ides"[05:25:19] <Nihility> does anyone in here have windows?[05:27:12] <Nihility> i have a windows nt vm on my notebook[05:27:17] <Nihility> i wonder if that will work[05:28:54] <Nihility> Caused by: java.lang.NullPointerException[05:28:54] <Nihility> at org.jboss.as.ee.structure.EjbJarDeploymentProcessor.deploy(EjbJarDeploymentProcessor.java:65)[05:29:09] <Nihility> ear demo fails :([05:34:22] <stuartdouglas> crap, my fault, fixed in my master[05:35:57] <Nihility> which commit[05:37:10] <stuartdouglas> hang on, my push was rejected[05:38:13] <stuartdouglas> should be there now[05:41:32] <stuartdouglas> something I have been thinking about recently is a module auditing tool[05:41:44] <stuartdouglas> that scans the const pools of all classes in modules[05:41:53] <stuartdouglas> and tells you which ones are not accessible[05:42:07] <stuartdouglas> and then tells you which modules those classes are in[05:42:20] <stuartdouglas> does this sound like something worth while?[05:42:22] <Nihility> yeah we need tools[05:43:21] *** irooskov has quit IRC[05:43:35] <stuartdouglas> I can't help but think that there are heaps on NCDFE's just waiting to be discovered when something takes a code path we have not seen in testing[05:47:42] <Nihility> cool looking good[05:48:17] <Nihility> rar demo also fails[05:49:40] <Nihility> hmm now it works[05:58:17] *** stuartdouglas_ has joined #jboss-as7[05:59:44] *** stuartdouglas has quit IRC[06:00:54] <Nihility> java 6 doesnt run on win[06:00:57] <Nihility> nt[06:05:25] *** bobmcw has joined #jboss-as7[06:05:25] *** ChanServ sets mode: +v bobmcw[06:35:31] *** Nihility has quit IRC[06:35:41] *** Nihility has joined #jboss-as7[06:35:41] *** ChanServ sets mode: +v Nihility[06:54:55] *** stuartdouglas_ has quit IRC[07:10:05] *** Nihility has quit IRC[07:25:24] *** mbg has joined #jboss-as7[07:25:24] *** ChanServ sets mode: +v mbg[07:29:18] *** pferraro has joined #jboss-as7[07:29:18] *** ChanServ sets mode: +v pferraro[07:38:27] *** pferraro has quit IRC[07:49:30] *** mbg is now known as mbg|away[07:50:03] *** mbg|away is now known as mbg[08:05:51] *** Jaikiran has joined #jboss-as7[08:05:51] *** ChanServ sets mode: +v Jaikiran[08:07:07] *** jfclere has joined #jboss-as7[08:07:07] *** ChanServ sets mode: +v jfclere[08:22:04] *** mbg is now known as mbg|away[08:22:32] *** miclorb_ has quit IRC[08:22:47] *** mbg|away is now known as mbg[08:38:17] *** mbg is now known as mbg|away[08:43:51] *** mbg|away is now known as mbg[08:45:20] *** mbg has quit IRC[08:46:35] *** miclorb has joined #jboss-as7[08:52:50] *** rmaucher has joined #jboss-as7[08:53:18] *** aslak has joined #jboss-as7[08:53:18] *** ChanServ sets mode: +v aslak[08:55:30] *** opalka has joined #jboss-as7[08:55:30] *** ChanServ sets mode: +v opalka[09:01:14] *** bgeorges has quit IRC[09:07:43] *** pil-afk-dinner has joined #jboss-as7[09:11:58] *** epbernard has joined #jboss-as7[09:11:58] *** epbernard is now known as emmanuel[09:11:58] *** ChanServ sets mode: +v emmanuel[09:14:37] *** maxandersen has joined #jboss-as7[09:14:37] *** ChanServ sets mode: +v maxandersen[09:17:03] *** slaboure has joined #jboss-as7[09:20:35] *** emuckenhuber has quit IRC[09:26:33] *** maxandersen has quit IRC[09:26:45] *** nickboldt1 has quit IRC[09:28:34] *** Jaikiran has quit IRC[09:28:57] *** jwulf has quit IRC[09:30:33] *** jwulf has joined #jboss-as7[09:31:13] *** maxandersen has joined #jboss-as7[09:31:14] *** ChanServ sets mode: +v maxandersen[09:32:19] *** nickboldt has joined #jboss-as7[09:32:20] *** ChanServ sets mode: +v nickboldt[09:42:53] *** miclorb has quit IRC[09:42:54] *** Jaikiran has joined #jboss-as7[09:42:54] *** ChanServ sets mode: +v Jaikiran[09:43:04] *** tdiesler has joined #jboss-as7[09:43:05] *** ChanServ sets mode: +v tdiesler[09:43:46] *** emuckenhuber has joined #jboss-as7[09:43:46] *** emuckenhuber has joined #jboss-as7[09:43:46] *** ChanServ sets mode: +v emuckenhuber[09:50:49] *** maeste_zzzz is now known as maeste[09:53:17] <maeste> emuckenhuber: when you have time would you have a look at https://github.com/maeste/jboss-as/commit/a5449e843d5046af53d4e16cc24e10ec0bb2c171[09:53:18] <jbossbot> git [jboss-as] a5449e8.. Stefano Maestri read-resource-description support for datasources and resource-adapters subsystem[10:10:32] *** torben has joined #jboss-as7[10:10:32] *** torben has joined #jboss-as7[10:10:32] *** ChanServ sets mode: +v torben[10:10:56] *** slaboure has quit IRC[10:11:19] *** wolfc has joined #jboss-as7[10:11:19] *** ChanServ sets mode: +v wolfc[10:14:32] *** alesj has joined #jboss-as7[10:21:38] *** adietisheim has joined #jboss-as7[10:22:35] <opalka> AS7 tag is broken, at least from JBossWS PoV :( http://fpaste.org/RMOd/[10:30:14] <emuckenhuber> maeste: sure will do[10:32:32] <emuckenhuber> opalka: hmm, are the tests part of the AS build/testsuite somewhere - then i can take a look at the issues with the deployment mgr?[10:33:43] <opalka> emuckenhuber, they're not :([10:33:55] <opalka> emuckenhuber, I'm having a closer look ...[10:34:13] <rmaucher> the actual bug that would cause that is unclear though, looking at the error[10:39:45] <emuckenhuber> opalka: well Brian moved the DeplomentManager thing to as-controller[10:40:09] <emuckenhuber> sorry controller-client[10:40:25] <emuckenhuber> https://github.com/emuckenhuber/jboss-as/tree/master/controller-client/src/main/java/org/jboss/as/controller/client/helpers/standalone[10:44:42] <opalka> emuckenhuber, this should be OK, he update AS7 tests integration module accordingly[10:44:52] <opalka> s/update/updated/[11:03:03] <emuckenhuber> so then it's just an issue of updating the pom of your tests to use the Beta1 artifact? :)[11:07:37] <opalka> emuckenhuber, yes[11:08:09] <emuckenhuber> ok, cool[11:09:00] *** jcosta has joined #jboss-as7[11:09:00] *** ChanServ sets mode: +v jcosta[11:20:36] <wolfc> stuartdouglas, yt? I see a branch phaseDeps on top of my ejb3-sessionid. It is probably old stuff, not?[11:24:30] *** stalep has joined #jboss-as7[11:25:08] *** pil-afk-dinner is now known as pilhuhn[11:25:09] *** pilhuhn has joined #jboss-as7[11:26:01] *** miclorb has joined #jboss-as7[11:48:52] *** jhalliday has joined #jboss-as7[11:54:05] *** jwulf has quit IRC[11:58:35] *** darranl has joined #jboss-as7[11:58:36] *** darranl has joined #jboss-as7[11:58:36] *** ChanServ sets mode: +v darranl[11:59:16] *** emmanuel has quit IRC[12:06:35] *** jwulf has joined #jboss-as7[12:08:40] *** epbernard has joined #jboss-as7[12:08:40] *** epbernard is now known as emmanuel[12:08:40] *** ChanServ sets mode: +v emmanuel[12:21:49] *** aslak has quit IRC[12:22:25] *** aslak has joined #jboss-as7[12:22:25] *** ChanServ sets mode: +v aslak[12:35:55] *** balunasj_away has quit IRC[12:36:37] *** JimMa has quit IRC[12:46:07] *** Jaikiran has quit IRC[12:46:20] *** Jaikiran has joined #jboss-as7[12:46:30] *** ChanServ sets mode: +v Jaikiran[12:52:25] *** miclorb has quit IRC[13:05:19] *** jwulf has quit IRC[13:32:06] *** darranl has quit IRC[13:32:38] *** darranl has joined #jboss-as7[13:32:38] *** darranl has joined #jboss-as7[13:32:38] *** ChanServ sets mode: +v darranl[13:32:55] *** nickboldt has quit IRC[13:35:17] *** bgeorges has joined #jboss-as7[13:39:06] *** tdiesler has quit IRC[13:39:24] *** pgier has joined #jboss-as7[13:39:25] *** ChanServ sets mode: +v pgier[13:46:22] *** tdiesler has joined #jboss-as7[13:46:30] *** ChanServ sets mode: +v tdiesler[13:50:48] *** tdiesler has quit IRC[13:54:26] *** jpederse has joined #jboss-as7[13:54:30] *** jpederse has joined #jboss-as7[13:59:56] *** pferraro has joined #jboss-as7[13:59:56] *** ChanServ sets mode: +v pferraro[14:01:04] *** mbg has joined #jboss-as7[14:01:04] *** ChanServ sets mode: +v mbg[14:02:31] *** mmoyses has joined #jboss-as7[14:02:31] *** ChanServ sets mode: +v mmoyses[14:10:59] *** jfd has joined #jboss-as7[14:10:59] *** jfd has quit IRC[14:10:59] *** jfd has joined #jboss-as7[14:10:59] *** ChanServ sets mode: +v jfd[14:11:16] *** nickboldt has joined #jboss-as7[14:11:16] *** ChanServ sets mode: +v nickboldt[14:12:47] *** fnasser has joined #jboss-as7[14:12:47] *** ChanServ sets mode: +v fnasser[14:13:55] *** smarlow has joined #jboss-as7[14:13:55] *** ChanServ sets mode: +v smarlow[14:17:07] *** hbraun has joined #jboss-as7[14:17:24] <hbraun> bstans_afk: ping[14:18:46] <hbraun> emuckenhuber: ping[14:18:57] <emuckenhuber> hbraun: pong[14:19:02] <hbraun> emuckenhuber: hi[14:19:27] <hbraun> emuckenhuber: how do I adress top level elements? i.e. /subsystem[14:19:46] <hbraun> i.e. in order to fetch the child names (all subsystems in this case)[14:20:06] <hbraun> is there something like 'base/subsystem' ?[14:20:31] <emuckenhuber> hmm there is an operation to fetch child names[14:21:27] <emuckenhuber> hbraun: http://pastebin.com/HCbtX8Pf[14:21:48] <emuckenhuber> well basically it's called "read-children-names" with the parameter "child-type"[14:22:50] <hbraun> weird, why not : {"operation" => "read-children-names","address" => ["subsystem"],"operations" => true,"recursive" => true}[14:23:46] <hbraun> maybe more like this one: {"operation" => "read-children-names","address" => ["subsystem"]}[14:24:02] <hbraun> why is there a special address for this op?[14:24:12] <emuckenhuber> because an address is always e.g. "subsystem" => "web"[14:24:38] <hbraun> haha, that's exactly my question in the beginning[14:24:47] <hbraun> it only come sin tuples[14:25:10] <emuckenhuber> frankly i don't like all the parameters we have atm.[14:25:36] <hbraun> but it also means i cannot do: read-children-names, address=path[14:25:46] <hbraun> to figure out what path names are set[14:25:59] <hbraun> simlar to the subsystem request[14:26:01] <emuckenhuber> yes[14:26:11] <hbraun> that feels wrong to me[14:26:43] <hbraun> it mean for every leaf we need a dedicated type?[14:27:23] <emuckenhuber> hmm, what do you mean with dedicated type?[14:29:41] <hbraun> well, in the example you send ( http://pastebin.com/HCbtX8Pf) it refers to a type[14:29:51] <hbraun> operation.get(CHILD_TYPE).set(SUBSYSTEM);[14:30:08] <emuckenhuber> oh, well that's operation.get("child-type").set("subsystem")[14:30:09] <hbraun> which I guess means for the second example I posted it would be operation.get(CHILD_TYPE).set(PATH);[14:30:11] <hbraun> and so on[14:30:24] <emuckenhuber> it would be operation.get("child-type").set("path")[14:30:29] <hbraun> i see[14:30:32] <hbraun> ok[14:30:36] <hbraun> i try that[14:30:47] <emuckenhuber> did you give the CLI a try?[14:31:05] <hbraun> is it the element name?[14:31:24] <hbraun> nope. I am building a tree browser to debug the web interface[14:31:40] <emuckenhuber> yes, in most cases - although usually we have <paths><path name=""></paths >[14:31:56] <emuckenhuber> which then maps to an address like "path" => "my-path-name"[14:34:47] *** frainone has joined #jboss-as7[14:34:47] *** ChanServ sets mode: +v frainone[14:36:35] <hbraun> ok, thanks[14:40:06] <stalep> hi, i tried to deploy the specj benchmark on as7 beta1 and got this exception: "<message-driven> element parsing hasn't yet been implemented." do anyone know when its scheduled to be implemented?[14:42:44] <emuckenhuber> stalep: wolfc should know[14:42:46] *** smcgowan has joined #jboss-as7[14:42:46] *** ChanServ sets mode: +v smcgowan[14:43:16] <wolfc> stalep: it is still some time off. Nobody is assigned to it.[14:43:43] <wolfc> stalep, right now annotated session beans are working[14:43:54] <stalep> wolfc, ok. thanks.[14:45:05] *** pferraro has quit IRC[14:45:26] <Jaikiran> out of curiosity, does anyone have a list of features (Java EE and domain) which end users can try/test against 7.0 Beta1?[14:45:33] <Jaikiran> end users == me too :)[14:45:46] <wolfc> dmlloyd, ping[14:46:03] <dmlloyd> pong wolfc[14:46:06] <stalep> Jaikiran, im hanging on this too. this would be very valuable.[14:46:09] <wolfc> do you like https://github.com/wolfc/jboss-as/commit/eddb9241003a7846e10a5fd231278167574b578f ?[14:46:10] <jbossbot> git [jboss-as] eddb924.. Carlo de Wolf Store ComponentView directly into Component[14:46:25] <Jaikiran> stalep: indeed[14:46:38] <dmlloyd> wolfc: I'm not opposed to it, if necessary[14:47:02] <wolfc> https://github.com/wolfc/jboss-as/commit/1e1296393efffb732482b7557b70d036fd4785fa#L0R71[14:47:02] <jbossbot> git [jboss-as] 1e12963.. Carlo de Wolf Implement SessionContext.getBusinessObject[14:47:35] <wolfc> although I don't like that sessionId going through EE either[14:47:36] <dmlloyd> should probably use an IdentityHashMap[14:47:47] <dmlloyd> yeah we need to get that out of there[14:47:55] <wolfc> the classes are unique[14:48:15] <dmlloyd> IHM is more memory-efficient and usually more performant than HashMap[14:48:34] <dmlloyd> if you're storing things which use identity anyway, it's usually the better choice[14:49:25] *** asaldhan has joined #jboss-as7[14:49:25] *** ChanServ sets mode: +v asaldhan[14:50:53] <smcgowan> smarlow: i saw the Beta1 list included JPA and PC injection - do you have list of functionality associated with taht bullet[14:51:33] <wolfc> hmm, it really shouldn't matter.[14:53:31] *** pferraro has joined #jboss-as7[14:53:31] *** ChanServ sets mode: +v pferraro[14:57:26] <bgeorges> wolfc: dmlloyd - what's the situation with MDBs ?[14:57:50] <dmlloyd> no MDB support presently[14:58:03] <bgeorges> dmlloyd: ?[14:58:05] <dmlloyd> it's on my beta2/3 list[14:58:23] <dmlloyd> just web profile stuff, which I believe does not include MDBs, is in beta1[14:58:31] <smarlow> smcgowan: there are things that could work but aren't tested yet (we are at the works at all stage). We can do PC injection into an EntityManager field and use of that EM under a CMT. PU injection is coded but not tested yet. The PersistenceUnit scoped search (PC referencing a persistence.xml) seems to work but will likely have some broken edge cases yet to be hit.[14:58:35] *** hbraun has quit IRC[14:58:38] <dmlloyd> the infrastructure is all there though so it should be a fairly minimal effort[14:58:42] <wolfc> Close to all the pieces are in place[14:58:44] <dmlloyd> barring anything crazy with JCA[14:59:17] <wolfc> Basically a 'stateless' bean with a rar 'injected'[14:59:26] <smarlow> smcgowan: the demo doesn't push very hard, just creates a new entity, persists it and finds it under the same transaction. Pretty lightweight.[14:59:33] <dmlloyd> yeah in fact that would be a great task for beta2[14:59:42] <dmlloyd> MDB component type + @MessageDriven support[14:59:49] <smcgowan> smarlow: my PU deploys, but then does nothing[15:00:07] <smarlow> smcgowan: lets pm for a sec...[15:00:07] <wolfc> I asked Stuart to look into message inflow stuff.[15:00:09] <smcgowan> smarlow: i'm working out datasource configuration issues too so need to resolve those first,[15:00:23] *** pferraro has quit IRC[15:00:35] <bgeorges> dmlloyd: ok, will you be doing it ?[15:01:01] <smcgowan> but the containers seem to not find the javax.persistence APIS, see the .jtr file here: http://hudson.qa.jboss.com/hudson/view/TCK6-AS7-WEB/job/tck6-as7-signaturetest_web-profile/[15:01:09] <dmlloyd> probably not personally, maybe just parts[15:01:09] <smarlow> smcgowan: I'm very interested in fixing any bugs you find/hit :)[15:01:42] <wolfc> the important bit is too understand RARs properly.[15:07:38] <baileyje> smcgowan: What dep issues are you having with the datasources?[15:09:34] <smcgowan> baileyje: this is all I see: http://pastebin.com/C61jdj9n[15:10:00] <baileyje> smcgowan: Have you looked at the service dump from the MCS mbean?[15:10:16] <baileyje> If you open jconsole you can get tons of info[15:10:41] <smcgowan> baileyje: i'll do that, and my derby module only contains the one jar: derby.jar[15:10:57] <smcgowan> which should be OK[15:11:21] <baileyje> can you paste your latest config and your server boot log?[15:11:38] *** maxandersen has quit IRC[15:11:48] <smcgowan> baileyje: sure, give me one se[15:11:58] <smarlow> I assume we still need to disabled OSGI as well (via something like: Disable osgi in standalone.xml (so that JDBC datasources works). Find <subsystem xmlns="urn:jboss:domain:osgi" and delete up to the matching "</subsystem>")[15:13:48] <baileyje> smarlow: You don't need that with the change to load Drivers from the subsystem.[15:13:57] <smcgowan> baileyje: i do get an ERROR loading the module: http://pastebin.com/W9f3tHFw[15:14:13] <smarlow> baileyje: nice :)[15:14:20] <smcgowan> baileyje: so give me a few minutes to work on my config[15:14:24] <baileyje> smcgowan: Is that the exact name of the module?[15:14:38] <smcgowan> yes[15:14:41] <wolfc> dmlloyd, what do you want first SessionContext.getBO or sessionId out of EE?[15:15:06] <dmlloyd> whatever is easiest to deliver[15:15:36] <wolfc> https://github.com/wolfc/jboss-as/commits/ejb3-ee :-)[15:15:51] <wolfc> There is also a bug in WeldEjbServices.resolveEjb[15:16:00] <wolfc> and SessionObjectReferenceImpl[15:16:09] <wolfc> resolveEjb should initiate a new session[15:16:15] <wolfc> not SessionObjectReferenceImpl[15:20:44] <dmlloyd> do what needs to be done. I think that apart from the comparatively significant issue of lifecycle interceptors, our architecture should be "complete enough" to get the job done, so at this point we can make some compromises to get specific pieces working[15:21:10] <dmlloyd> there will be time to clean things up later, as long as we don't go too far off the reservation[15:24:39] <smarlow> wolfc: In JPADependencyProcessor, I'm currently injected javax.persistence.api and other JPA modules into deployments that have a persistence.xml. But, you mentioned that is wrong I think. You said an application without a persistence.xml, should be able to reference another (with fully scoped PU name reference) applications persistence.xml. Is this a spec requirement or something we historically have supported? Shelly is getting a test fai[15:24:39] <smarlow> lure that would be fixed, if I better supported this concept.[15:25:34] <wolfc> smarlow, do you also mark every deployment that has a @PersistenceContext or @PersistenceUnit?[15:26:09] <smarlow> wolfc: no but that would better handle this concept. :)[15:26:30] <smarlow> wolfc, smcgowan: for shelly's test error, I think adding javax.persistence.api to the EE module, would help.[15:26:55] <wolfc> smarlow, no that's the wrong way around[15:27:18] <smarlow> wolfc, smcgowan: the test failure is about looking for javax.persistence.api in a web app, without any JPA stuff...[15:27:46] <wolfc> hmm, dmlloyd, maybe we should add javaee.api to every deployment?[15:27:47] <smarlow> smcgowan: is that true? or do they have JPA annotations in the web app?[15:27:55] <Jaikiran> dmlloyd: can you please pull this fix upstream https://github.com/jaikiran/jboss-as/commit/0bebad6b522342d232c644a1cc2d20eb57719956[15:27:56] <jbossbot> git [jboss-as] 0bebad6.. jaikiran Fix NPE in ComponentInterceptorAnnotationParsingProcessor.processInterceptorConfigs[15:28:13] *** Nihility has joined #jboss-as7[15:28:13] *** ChanServ sets mode: +v Nihility[15:28:14] <wolfc> Oi, I had that one as well :-)[15:28:16] <smcgowan> smarlow: they are signature tests[15:28:29] <Jaikiran> too late :)[15:28:45] <wolfc> I asked a merge first :-)[15:29:00] <Jaikiran> then it's me who's too late ;)[15:29:16] <Jaikiran> btw, the JIRA for that one is JBAS-8964[15:29:17] <smarlow> wolfc, dmlloyd, smcgowan: so, signature tests for the JPA api methods.[15:29:19] <jbossbot> jira [JBAS-8964] NullPointerException in ComponentInterceptorAnnotationParsingProcessor.processInterceptorConfigs [Open (Unresolved) Bug, Major, jaikiran pai] https://issues.jboss.org/browse/JBAS-8964[15:29:42] <wolfc> Ah cool, we're raising issues again.[15:29:59] *** AndyTaylor has quit IRC[15:30:32] * smarlow I need to walk away for 30 minutes...[15:31:14] <dmlloyd> okay well I'll go with wolfc's branch if that's OK[15:31:58] <wolfc> we got the exact same tree has :-D[15:32:00] <wolfc> hash[15:32:20] *** Nihility has quit IRC[15:32:29] <dmlloyd> well, then I'll merge them both... simultaneously![15:32:40] * dmlloyd has mad skillz[15:33:01] <smcgowan> smarlow: JavaEESpec EE.6.1[15:33:14] *** Nihility has joined #jboss-as7[15:33:14] *** ChanServ sets mode: +v Nihility[15:35:29] <wolfc> alesj, do you have a Weld test for AS 7 that uses SessionObjectReference?[15:36:58] <alesj> wolfc: TCK?[15:37:25] <wolfc> No, just some test[15:37:57] <wolfc> I see a problem with WeldEjbServices and SessionObjectReferenceImpl. The creation of the session is mangled, but I want to be sure.[15:37:59] <alesj> dunno[15:38:14] <alesj> hack one yourself?[15:38:23] <alesj> as you know what you're looking for[15:39:31] *** Jaikiran is now known as Jaikiran|brb[15:44:29] *** Nihility has quit IRC[15:49:40] *** bgeorges has quit IRC[16:01:17] *** ALR has joined #jboss-as7[16:01:17] *** ChanServ sets mode: +v ALR[16:03:22] <jbossbot> git [jboss-as] push master 37dd809.. Carlo de Wolf Fixed @ExcludeClassInterceptors annotation parsing[16:03:23] <jbossbot> git [jboss-as] push master eddb924.. Carlo de Wolf Store ComponentView directly into Component[16:03:23] <jbossbot> git [jboss-as] push master 1e12963.. Carlo de Wolf Implement SessionContext.getBusinessObject[16:03:23] <jbossbot> git [jboss-as] push master 1fd20c8.. Carlo de Wolf Demo SessionContext.getBusinessObject[16:03:23] <jbossbot> git [jboss-as] push master URL: http://github.com/jbossas/jboss-as/compare/6866906...1fd20c8[16:06:26] *** bstans_afk is now known as bstansberry[16:07:06] <smarlow> smcgowan: so, its a test that checks if javax.persistence classes are available to all WEB deployments, even ones that don't have any direct use of the javax.persistence api.[16:08:09] <smcgowan> smarlow: web container[16:11:51] <smarlow> smcgowan, rmaucher: I don't get why they would test the web container directly but we don't currently include javax.persistence in the web container module.xml[16:13:43] <smarlow> smcgowan, rmaucher: is there any reason why that test should pass? Maybe we should challenge that one. I would expect a WEB deployment to have access to javax.persistence but don't see why the WEB container should.[16:14:00] <smcgowan> smarlow: the spec requires the APIs are available from the web container[16:14:31] <rmaucher> don't we add all the EE spec APIs to the web deployments ?[16:14:51] <smarlow> rmaucher: we currently don't include javax.persistence in the EE spec module.[16:15:02] <smarlow> we could and that would address this I think[16:15:15] <smarlow> but I was just questioning if we want to do that?[16:15:44] <wolfc> ah crock, the integration-tests are not compiling anymore[16:16:08] <wolfc> dmlloyd, I think we want them compiled by default.[16:16:38] <smarlow> rmaucher, smcgowan: seems like org.jboss.as.ee should include jpa also[16:17:03] <rmaucher> nice, problem solved :)[16:17:17] *** Jaikiran|brb is now known as Jaikiran[16:17:49] <smarlow> dmlloyd: do you have any concern about putting javax.persistence into the "org.jboss.as.ee" module?[16:17:55] <dmlloyd> no, it makes sense[16:18:07] * smarlow doing it...[16:18:08] <wolfc> ee doesn't use jpa[16:18:18] *** mmoyses has quit IRC[16:18:35] *** jfd has quit IRC[16:18:40] <wolfc> we should add javaee.api to all deployments via a processor[16:21:39] <dmlloyd> smarlow: ah wrong module[16:21:44] <dmlloyd> it should be added to javax.ee.api[16:21:51] <dmlloyd> er javaee.api rather[16:22:14] *** frainone has quit IRC[16:24:26] <smarlow> dmlloyd, wolfc: I gotta step away for a few hours. I am open to either direction on this. Challenging the test or including it in javaee.api[16:26:48] <wolfc> smarlow, dmlloyd is right it's missing from javaee.api. javaee.api is added to all deployments that have a Component.[16:27:20] <wolfc> there is no challenging the test. The test is valid.[16:32:00] <alesj> rmaucher: any plans to optimize jbw some more?[16:33:16] <rmaucher> mostly no, I can't shrink it much more[16:33:48] <alesj> rmaucher: what're final numbers?[16:33:56] <alesj> or, what about that native connector?[16:34:04] <rmaucher> maybe longer term items like removing dead code, etc[16:34:35] <rmaucher> I only have my quick test numbers that I sent by email[16:35:19] <rmaucher> which are ok[16:35:47] *** jfd has joined #jboss-as7[16:35:47] *** ChanServ sets mode: +v jfd[16:36:06] *** jamezp has joined #jboss-as7[16:36:14] <rmaucher> it would be interesting to get the native numbers, since that connector should retain decent performance[16:36:28] <rmaucher> (compared to the default config)[16:37:04] <dmlloyd> does using the native connector reduce the # of classes loaded?[16:37:25] <dmlloyd> .sos are much more easily shared than classes...[16:45:42] <alesj> smcgowan: ping?[16:48:45] *** emuckenhuber has quit IRC[16:50:50] *** opalka has quit IRC[16:50:53] <rmaucher> dmlloyd, well, if you're using SSL mostly ;)[16:51:01] <rmaucher> and it's faster for that too[16:51:34] <rmaucher> (historically, SSL support has more bugs though - supposedly it is ok now)[16:52:41] <dmlloyd> smarlow: yeah put it in javaee.api[16:53:05] <wolfc> dmlloyd, I just did that. Now it can't find javax.sql.DataSource[16:53:12] <wolfc> which is sort of weird[16:53:20] <dmlloyd> that's in javax.api iirc[16:53:27] <wolfc> yup[16:53:30] *** jfclere has quit IRC[16:54:56] <wolfc> dmlloyd, for immediate consumption https://github.com/wolfc/jboss-as/commit/b0f3ad3513334b13646db96afc3d841c886f26f5[16:54:57] <jbossbot> git [jboss-as] b0f3ad3.. Carlo de Wolf Removed workAroundBug[16:55:06] <wolfc> https://github.com/wolfc/jboss-as/commit/fd5696596545d9d9c2e4eb4c499bf3f1b7be62f1 actually broke the compilation[16:55:07] <jbossbot> git [jboss-as] fd56965.. Stuart Douglas Fix integration tests[17:00:34] <dmlloyd> hm[17:00:39] <dmlloyd> it built OK for me[17:01:25] <wolfc> with -Pintegration-tests ?[17:02:00] <dmlloyd> ah[17:02:01] <dmlloyd> no[17:02:21] <dmlloyd> cherry-picked, now testing[17:03:01] <wolfc> I think we should compile always and make the running of the tests configurable.[17:03:24] <bstansberry> +1[17:03:31] <dmlloyd> yeah agreed[17:03:39] <bstansberry> it was that way before, but it got refactored[17:03:41] <dmlloyd> so... who knows how to do that :)[17:03:41] *** emmanuel has quit IRC[17:06:11] *** epbernard has joined #jboss-as7[17:06:11] *** epbernard is now known as emmanuel[17:06:11] *** ChanServ sets mode: +v emmanuel[17:07:27] <dmlloyd> hm the integration tests actually fail[17:08:17] <wolfc> not on my end, what's the failure?[17:08:22] *** emmanuel has quit IRC[17:08:38] *** epbernard has joined #jboss-as7[17:08:38] *** epbernard is now known as emmanuel[17:08:38] *** ChanServ sets mode: +v emmanuel[17:09:20] <dmlloyd> wolfc: http://fpaste.org/arwL/[17:09:21] *** emuckenhuber has joined #jboss-as7[17:09:21] *** emuckenhuber has joined #jboss-as7[17:09:21] *** ChanServ sets mode: +v emuckenhuber[17:10:17] <wolfc> dmlloyd, that test is from a dirty branch[17:17:01] <wolfc> dmlloyd, how is java/ exposed through modules? javax.sql.DataSource needs java.sql[17:17:28] <dmlloyd> it's always present[17:17:37] <dmlloyd> legal requirement :)[17:18:36] <wolfc> :-)[17:23:26] *** bgeorges has joined #jboss-as7[17:26:54] *** emuckenhuber has quit IRC[17:29:03] <dmlloyd> what do you mean about the dirty branch[17:29:17] <wolfc> the test can't pass it is broken[17:29:19] <dmlloyd> this is a clean build of upstream plus b0f3ad3513334b13646db96afc3d841c886f26f5 cherry-picked[17:29:31] <dmlloyd> oh ok, so it's expected to fail at the moment?[17:29:37] <dmlloyd> it does compile at least[17:29:39] <dmlloyd> :)[17:29:52] *** Nihility_ has joined #jboss-as7[17:29:52] *** ChanServ sets mode: +v Nihility_[17:29:52] <wolfc> Hmm, it did not compile for me[17:29:57] <jbossbot> git [jboss-as] push master 2895bbe.. Carlo de Wolf Removed workAroundBug[17:29:58] <jbossbot> git [jboss-as] push master URL: http://github.com/jbossas/jboss-as/compare/1fd20c8...2895bbe[17:30:03] <dmlloyd> ah well it's in there ;)[17:30:29] <wolfc> smcgowan, https://github.com/wolfc/jboss-as/commit/7a93ed4622c59e906953eef1a3a3a8b8eca6e342 makes the signature test work[17:30:30] <jbossbot> git [jboss-as] 7a93ed4.. Carlo de Wolf Added the proper dependencies on and for javax.persistence.api[17:31:00] <dmlloyd> can I merge that?[17:31:04] <smcgowan> wolfc: thanks[17:31:07] <wolfc> yup[17:31:18] <wolfc> smcgowan, it doesn't make them pass though :-P[17:31:28] <jbossbot> git [jboss-as] push master 4fa477a.. Carlo de Wolf Added the proper dependencies on and for javax.persistence.api[17:31:28] <jbossbot> git [jboss-as] push master URL: http://github.com/jbossas/jboss-as/compare/2895bbe...4fa477a[17:31:32] <wolfc> Hibernate is not up to spec[17:31:35] <smcgowan> are you seeing the jaxrs issue still?[17:31:39] <wolfc> yes[17:31:51] <smcgowan> hibernate is the same API set certified in AS 6[17:31:56] <dmlloyd> I think we have more than one API JAR for some things...[17:33:28] <smcgowan> dmlloyd: the jaxr reflection error is not seen consistently - i'll look into that too[17:33:38] <dmlloyd> great :([17:34:40] *** Nihility_ has quit IRC[17:35:43] <wolfc> smcgowan, we have both org.hibernate.javax.persistence:hibernate-jpa-2.0-api and org.hibernate.java-persistence:jpa-api in there[17:35:51] <wolfc> the second one I think is the broken one[17:35:58] * wolfc is running off for dinner[17:38:38] <dmlloyd> <version.org.hibernate.javax.persistence.hibernate-jpa-2.0-api>1.0.0.Final</version.org.hibernate.javax.persistence.hibernate-jpa-2.0-api>[17:38:38] <dmlloyd> <version.org.hibernate.jpa-api>2.0-cr-1</version.org.hibernate.jpa-api>[17:38:45] *** emuckenhuber has joined #jboss-as7[17:38:46] *** emuckenhuber has joined #jboss-as7[17:38:46] *** ChanServ sets mode: +v emuckenhuber[17:40:27] <dmlloyd> argh that one's define twice, too[17:40:44] <dmlloyd> I think we need to have a discussion about what "alphabetical" means ;|[17:41:36] <dmlloyd> so the question is, who here knows which JPA API artifact is the right one?[17:42:03] <dmlloyd> if nobody knows I'm going with wolfc's guess[17:42:13] <rmaucher> good question ;)[17:42:34] <dmlloyd> emmanuel: do you happen to know, offhand?[17:42:51] <rmaucher> he may be busy moving[17:43:06] <dmlloyd> ah. Maybe I can track steve down..[17:43:06] <rmaucher> ah no it's the other emmanuel ;)[17:43:43] *** emmanuel has quit IRC[17:44:24] *** epbernard has joined #jboss-as7[17:44:24] *** epbernard is now known as emmanuel[17:44:24] *** ChanServ sets mode: +v emmanuel[17:45:09] <rmaucher> I also think org.hibernate.javax.persistence.hibernate-jpa-2.0-api is the right one[17:48:08] <dmlloyd> okay I'm going with that[17:48:25] <dmlloyd> yeah the build was using the other one, that explains the issues[17:53:35] *** bgeorges has quit IRC[17:56:00] <dmlloyd> wolfc: can I get an OK on http://github.com/dmlloyd/jboss-as/commit/ae0b5f0 when you're back[17:56:01] <jbossbot> git [jboss-as] ae0b5f0.. David M. Lloyd Remove and ban bad JPA API dependency; fix JPA pom[17:59:54] *** pilhuhn is now known as pil-dinner[18:03:28] *** jcosta has quit IRC[18:03:54] <smcgowan> dmlloyd: jdcasey said to sign him up for javadoc plugin - file JIRA with requirements[18:04:39] <dmlloyd> great.[18:10:17] <Jaikiran> who's the next jandex guy, now that jason isn't around this week[18:10:18] <Jaikiran> ?[18:10:41] <dmlloyd> I think a few people have at least poked around with it[18:10:56] <dmlloyd> I don't think there's a next guy per se[18:10:58] <Jaikiran> i have a API question[18:11:02] <Jaikiran> let me pastebin the code[18:12:00] <Jaikiran> https://gist.github.com/872854[18:12:42] <Jaikiran> briefly, i have a class hierarchy Child extends Parent and Parent extends GrandParent[18:13:04] <Jaikiran> assume Child and GrandParent have some annotations on them[18:13:13] <Jaikiran> where as Parent has 0 annotations[18:13:36] <dmlloyd> hmm[18:13:40] <dmlloyd> that's supposed to work, it might be a bug[18:13:58] <dmlloyd> for sure, if all the classes are in the same JAR[18:14:04] <dmlloyd> even annotation-less classes are supposed to be indexed[18:14:15] <Jaikiran> you mean if Parent doesn't have any annotations, index.getClassByName is supposed to return a ClassInfo for it?[18:14:20] <dmlloyd> yes[18:14:25] <Jaikiran> hmm interesting[18:14:28] <dmlloyd> as far as I understand[18:14:32] <Jaikiran> let me try a unit test in jandex then[18:14:36] <dmlloyd> okay[18:14:47] *** alesj has quit IRC[18:18:51] * wolfc is checking http://github.com/dmlloyd/jboss-as/commit/ae0b5f0[18:18:52] <jbossbot> git [jboss-as] ae0b5f0.. David M. Lloyd Remove and ban bad JPA API dependency; fix JPA pom[18:19:35] *** emmanuel has quit IRC[18:20:46] *** frainone has joined #jboss-as7[18:20:46] *** ChanServ sets mode: +v frainone[18:20:59] *** epbernard has joined #jboss-as7[18:20:59] *** epbernard is now known as emmanuel[18:20:59] *** ChanServ sets mode: +v emmanuel[18:22:27] *** torben has quit IRC[18:24:33] <dmlloyd> smcgowan: what project should I file that maven JIRA under?[18:30:38] *** torben has joined #jboss-as7[18:30:39] *** torben has joined #jboss-as7[18:30:39] *** ChanServ sets mode: +v torben[18:30:49] *** torben has quit IRC[18:31:12] <dmlloyd> wolfc: problems?[18:31:45] <wolfc> I run TCK on my slowest box :-)[18:32:08] <wolfc> AS 7 needs 10 minutes to build[18:32:13] <dmlloyd> ah[18:32:34] <dmlloyd> well if the commit itself looks sane I'd like to go ahead and push it out[18:32:41] <wolfc> test is running now[18:33:17] <wolfc> smcgowan, instead of sleep you can always look for .isdeployed :-)[18:33:41] <dmlloyd> the deployment API should now be stable enough to use directly afaik[18:34:33] <wolfc> dmlloyd, I say push it[18:34:58] <jbossbot> git [jboss-as] push master ae0b5f0.. David M. Lloyd Remove and ban bad JPA API dependency; fix JPA pom[18:34:59] <jbossbot> git [jboss-as] push master URL: http://github.com/jbossas/jboss-as/compare/4fa477a...ae0b5f0[18:38:39] <Jaikiran> dmlloyd: fails as expected https://github.com/jaikiran/jandex/commit/4d796e53912ef5f7ecad21dc35e646dbc0ddfe01[18:38:40] <jbossbot> git [jandex] 4d796e5.. jaikiran Add a class hierarchy test case[18:38:47] <Jaikiran> with https://gist.github.com/872893[18:39:19] <Jaikiran> so with the current API, I don't see a way to get hold of the annotated grand parent[18:39:25] <wolfc> Jaikiran, you didn't index all classes[18:39:34] <Jaikiran> oops[18:39:37] <Jaikiran> one sec let me do that[18:41:34] <Jaikiran> wolfc: that was it! :)[18:42:01] * Jaikiran now goes back to the tck test[18:42:56] <dmlloyd> maybe AS is somehow not indexing everything[18:44:09] <wolfc> I suspect indexing doesn't travel down Class-Path and Extensions[18:44:24] <dmlloyd> it does and it doesn't[18:44:33] <dmlloyd> if the JAR is internal yes it will get indexed[18:44:54] <wolfc> how about WEB-INF/lib stuff? will that become one index?[18:44:57] <dmlloyd> if it's external to the deployment, it may get indexed (depending on where it is) but odds are that index isn't included in the deployment index[18:45:02] <dmlloyd> yeah it should be wolfc[18:45:17] <wolfc> in that case I'll stop speculating :-)[18:46:32] *** pferraro has joined #jboss-as7[18:46:33] *** ChanServ sets mode: +v pferraro[18:49:21] <smcgowan> dmlloyd: about the JIRA for jdcasey, i was thinking it should be a JBAS JIRA for providing javadoc for public APIs[18:49:40] <dmlloyd> hm okay[18:49:46] <dmlloyd> I'll make a parent issue with subtasks then[18:49:46] <smcgowan> do you agree?[18:50:14] <dmlloyd> yeah that's reasonable. My initial impression was more from a maven perspective, with the AS docs being a pleasant side-effect, but it amoutns to the same thing really[18:52:43] *** hbraun has joined #jboss-as7[18:52:51] <hbraun> bstansberry: ping[18:53:00] <bstansberry> pong[18:53:10] <hbraun> bstansberry: how are you?[18:53:23] <hbraun> bstansberry: got a minute?[18:53:38] <hbraun> bstansberry: is this addressing correct? {"address" => [("server-group" => "main-server-group")],"operation" => "read-resource-description"}[18:54:14] <hbraun> because I always end up with the root resource[18:54:15] <bstansberry> yes, that looks correct[18:54:15] *** ALR has quit IRC[18:54:35] <hbraun> hm[18:54:37] <hbraun> weird[18:55:15] <bstansberry> new ModelNode().get("address").add("server-group" => "main-server-group");[18:55:26] <hbraun> what about this: addr.add("base", "domain"); ?[18:55:31] <hbraun> is it mandatory?[18:55:35] <bstansberry> no[18:55:44] <bstansberry> everything is rooted in the domain[18:55:49] <hbraun> taken from here: http://community.jboss.org/wiki/FormatofaDetypedOperationRequest[18:57:05] <bstansberry> yeah, your text output looked right, and the code I typed ^^^ is what generated the stuff in that wiki[18:57:30] *** stalep has quit IRC[18:58:30] <darranl> hbraun, I keep getting the root resource as well when testing with the CLI[18:58:37] *** frainone has quit IRC[18:58:44] <hbraun> darranl:[18:58:54] <hbraun> darranl: interesting[18:59:01] <hbraun> darranl: i'll get back to you later[18:59:41] <bstansberry> are you testing a DC or standalone?[19:00:31] *** Heiko has joined #jboss-as7[19:00:54] <Heiko> darranl: Did you figure it out?[19:01:21] <bstansberry> [/] /server-group=main-server-group:read-resource-description[19:01:21] <bstansberry> {[19:01:22] <bstansberry> "outcome" => "success",[19:01:22] <bstansberry> "result" => undefined,[19:01:22] <bstansberry> "compensating-operation" => undefined[19:01:22] <bstansberry> }[19:01:28] <bstansberry> which is a :([19:01:42] <bstansberry> meaning the metadata needs to be added[19:02:07] <Heiko> bstansberry: What do you mean?[19:02:19] <darranl> Heiko, no at this stage I didn't dig too far - at the time I was actually trying to find an operation that would be passed around a domain and had issues with crashes every time I attached the debugger[19:02:24] <Heiko> Ah[19:02:31] *** emmanuel has quit IRC[19:02:53] <Heiko> bstansberry: Why do i retrieve the root resource?[19:03:02] *** jfclere has joined #jboss-as7[19:03:10] <Heiko> Maybe my codebase is out of sync[19:03:32] <bstansberry> Heiko: is that the command you are using?[19:03:54] <Heiko> I am going over the http api[19:04:11] <Heiko> But it looks similar[19:04:45] <Heiko> I am building a tree browser for the console[19:06:14] <Heiko> bstansberry: I'll check later on. Maybe i need to sync with upstream[19:06:20] <Heiko> bstansberry: Tnx[19:16:30] *** pferraro has quit IRC[19:16:53] *** pferraro has joined #jboss-as7[19:16:53] *** ChanServ sets mode: +v pferraro[19:19:45] *** jfclere has quit IRC[19:25:31] *** adietisheim has quit IRC[19:25:32] *** adietisheim1 has joined #jboss-as7[19:25:33] *** Jaikiran has quit IRC[19:25:56] *** hbraun has quit IRC[19:28:18] *** maxandersen has joined #jboss-as7[19:28:18] *** ChanServ sets mode: +v maxandersen[19:30:58] <smcgowan> dmlloyd (or baileyje): if i wanted to use derby that is provided with the JDK how would I define the driver for that in the datasource config?[19:31:14] <dmlloyd> good question[19:31:20] <dmlloyd> we don't currently have any way to access that class[19:31:23] <smcgowan> i'm just thinking that there is no need to create the module if the JDK already supplies it[19:31:32] *** jhalliday has quit IRC[19:31:38] <dmlloyd> yeah it gets tricky though because of course not all JDKs would supply it[19:31:53] <dmlloyd> we don't really have a good way to automatically expose modules for stuff the JDK supplies[19:32:02] <dmlloyd> for the HTTP server for example, we always bundle our own[19:32:09] <smcgowan> ok, then i'll proceed with adding my own[19:32:09] *** maxandersen has quit IRC[19:33:58] *** alesj has joined #jboss-as7[19:36:06] *** Heiko has quit IRC[19:45:25] *** frainone has joined #jboss-as7[19:46:52] <asaldhan> bstansberry: is there some wiki etc on the DMR. I found the api to be pretty confusing.[19:47:10] <asaldhan> bstansberry: well I only looked at it today. so newbie issues, of course. :)[19:48:42] <bstansberry> asaldhan: I don't think so. dmlloyd, did you write any kind of DMR wiki?[19:49:07] <bstansberry> the most confusing thing is get(...) actualy mutates[19:49:13] <asaldhan> bstansberry: I need to figure out modelNode.asList() and modelNode.asPropertyList() diffs[19:49:26] <asaldhan> bstansberry: something for me to do. :)[19:49:34] *** mmoyses has joined #jboss-as7[19:49:34] *** ChanServ sets mode: +v mmoyses[19:50:02] <mmoyses> dmlloyd: bstansberry: can you please pull https://github.com/mmoyses/jboss-as/commit/b4e04f52046cad2687d60348bad3c8678ccd2963 ?[19:50:03] <jbossbot> git [jboss-as] b4e04f5.. Marcus Moyses replace jaas with security-domains in the model[19:51:07] <bstansberry> mmoyses: is that new? or did I dream I pulled it yesterday?[19:51:23] <mmoyses> bstansberry: it's new[19:51:30] <bstansberry> phew :-)[19:51:54] *** frainone has quit IRC[19:54:30] <bstansberry> asaldhan: think of modelNode.asPropertyList() as being similar to Map.entrySet()[19:54:55] <asaldhan> bstansberry: got it.[19:55:35] <bstansberry> more specifically, Map<String, ModelNode>.entrySet[19:57:47] <asaldhan> bstansberry: by the way, how do I pull in marcus's commit with above id?[19:57:59] <asaldhan> bstansberry: I did "git fetch and then git merge"[19:58:28] <bstansberry> that works[19:58:37] *** frainone has joined #jboss-as7[19:58:59] <bstansberry> you can also "git fetch mmoyses" and then "git cherry-pick b4e04f5204"[19:59:08] <bstansberry> if you don't want other stuff on his branch[19:59:43] *** frainone has quit IRC[19:59:52] *** frainone has joined #jboss-as7[19:59:53] <jbossbot> git [jboss-as] push master d8e774b.. David M. Lloyd Remove duplicate module def; be more emphatic about alphabetization[19:59:53] <jbossbot> git [jboss-as] push master URL: http://github.com/jbossas/jboss-as/compare/ae0b5f0...d8e774b[20:00:02] <bstansberry> grrr[20:00:57] <asaldhan> bstansberry: thx. I was just reading the doc on cherry-picj[20:01:05] <asaldhan> bstansberry: it worked[20:04:53] *** alesj has quit IRC[20:05:09] <mmoyses> bstansberry: want me to rebase?[20:05:34] <bstansberry> mmoyses, no[20:05:40] <bstansberry> just had to rebuild :)[20:09:28] *** darranl has quit IRC[20:10:37] *** hbraun has joined #jboss-as7[20:12:30] <jbossbot> git [jboss-as] push master b3e90b9.. Marcus Moyses replace jaas with security-domains in the model[20:12:30] <jbossbot> git [jboss-as] push master URL: http://github.com/jbossas/jboss-as/compare/d8e774b...b3e90b9[20:12:44] <bstansberry> mmoyses: ^^^[20:12:50] *** bstansberry is now known as bstans_afk[20:12:54] <mmoyses> bstans_afk: thanks[20:13:33] <asaldhan> mmoyses: if I ask for pull request now, should I be doing 3 commits?[20:13:41] <wolfc> and thus his cognitive dreams come true :-)[20:13:55] <asaldhan> mmoyses: I have pushed it to my master.[20:14:01] *** pil-dinner is now known as pilhuhn[20:16:01] <mmoyses> asaldhan: only your commits need to pulled now. those 2[20:16:41] <asaldhan> mmoyses: the first 2 and now the latest (fixing the subsystem.xml). so 3.[20:16:56] <mmoyses> asaldhan: right[20:17:42] *** Heiko has joined #jboss-as7[20:20:58] <dmlloyd> in case you guys haven't noticed, I'm populating JIRA with tasks for beta2[20:21:05] <dmlloyd> we've got just about 2 weeks until we tag[20:21:36] <dmlloyd> the filter you want is: project = JBAS AND fixVersion in ("7.0.0.Beta2") AND resolution = Unresolved AND issuetype not in ("Sub-task", "Component Upgrade")[20:21:48] <dmlloyd> that'll give you the top-level issues[20:25:15] <jamezp> dmlloyd: If you've got anything I could work on let me know.[20:25:32] <dmlloyd> grab any beta2 task :)[20:25:56] * jamezp goes to JIRA[20:26:14] <dmlloyd> you'll want to make sure you've read the http://community.jboss.org/wiki/HackingonAS7 document.[20:26:24] <dmlloyd> particularly the appendixes[20:26:31] <dmlloyd> well A&B anyway[20:26:39] <dmlloyd> most of the rest is about using git[20:27:12] <jamezp> Will do.[20:28:34] *** frainone has quit IRC[20:32:04] *** jfd has quit IRC[20:33:00] *** mbg is now known as mbg|away[20:33:26] *** Heiko has quit IRC[20:33:41] *** Heiko has joined #jboss-as7[20:34:58] *** irooskov has joined #jboss-as7[20:39:05] *** Heiko_ has joined #jboss-as7[20:39:16] *** hbraun has quit IRC[20:40:15] *** Heiko has quit IRC[20:44:56] *** hbraun has joined #jboss-as7[20:45:49] *** jfd has joined #jboss-as7[20:45:49] *** ChanServ sets mode: +v jfd[20:47:57] *** mbg|away is now known as mbg[20:48:19] <dmlloyd> we need to identify the deployment descriptors we do not know how to deal with and add them to JBAS-9016[20:48:20] <jbossbot> jira [JBAS-9016] Support for deployment descriptor parsing (remaining) [Open (Unresolved) Task, Major, Unassigned] https://issues.jboss.org/browse/JBAS-9016[20:50:50] <dmlloyd> if I've missed any[20:50:56] <dmlloyd> which I probably have[20:56:36] *** bstans_afk is now known as bstansberry[20:57:02] <mmoyses> bstans_afk: when you come back I have 3 more commits if you would be so kind :) http://pastebin.com/ypbXixpV[20:57:17] <dmlloyd> JBAS-9020[20:57:19] <jbossbot> jira [JBAS-9020] Deployment content management enhancement [Open (Unresolved) Task, Major, Unassigned] https://issues.jboss.org/browse/JBAS-9020[20:57:22] <dmlloyd> be advised[20:58:23] *** Heiko has joined #jboss-as7[20:59:13] *** Heiko_ has quit IRC[21:03:29] <smcgowan> baileyje: good news: http://pastebin.com/d3Zh4M2c bad news: http://pastebin.com/ErWMbjZ0 - heading out to a celtics game but wanted to give you an update before i left[21:04:23] <baileyje> smcgowan: Ok. Well I am working on changing the DS deployments a bit. But I still think we need to dig into what deps are missing for those service.[21:04:40] <smcgowan> baileyje: i'll continue to dig[21:04:55] <baileyje> jconsole will give you exactly what is missing[21:10:14] <jbossbot> git [jboss-as] push master ff07b4a.. Anil Saldhana add test case[21:10:15] <jbossbot> git [jboss-as] push master 23478c3.. Anil Saldhana test[21:10:15] <jbossbot> git [jboss-as] push master 328151e.. Anil Saldhana changed jaas to security-domains[21:10:15] <jbossbot> git [jboss-as] push master URL: http://github.com/jbossas/jboss-as/compare/b3e90b9...328151e[21:10:27] <bstansberry> mmoyses: ^^^[21:10:57] <smcgowan> exactly? says has missing dependency[21:11:00] <mmoyses> bstansberry: ty[21:12:48] *** frainone has joined #jboss-as7[21:12:48] *** ChanServ sets mode: +v frainone[21:13:43] *** smcgowan has quit IRC[21:13:45] <asaldhan> bstansberry: I bungled up my git master. :) mmoyses came to the rescue. I need to replace my master. :)[21:14:40] *** nickboldt has quit IRC[21:14:49] *** nickboldt has joined #jboss-as7[21:14:49] *** ChanServ sets mode: +v nickboldt[21:14:59] <bstansberry> you can just delete it with git branch -D then git fetch upstre am then git checkout upstream/master then git checkout -b master[21:15:15] <bstansberry> first one is "git branch -D master"[21:15:19] <asaldhan> bstansberry: right. :)[21:17:37] <dmlloyd> folks: I'm a bit concerned that we have too much piled up against CR1 and not enough on Beta2/Beta3[21:17:55] <dmlloyd> so if you have an issue on CR1 that you think could/should be addressed earlier then please - address it :)[21:18:16] <dmlloyd> 94 top-level JIRAs exist for Beta2, Beta3, and CR1[21:18:31] <dmlloyd> so we need to close about 3 a day at a minimum[21:24:28] <asaldhan> bstansberry: Anil's geriatric sheet: git fetch marcus; git merge marcus/master --ff-only[21:24:40] <asaldhan> bstansberry: sorry: http://community.jboss.org/docs/DOC-16629[21:25:22] <dmlloyd> I'd say your #3 is much more safely accomplished with a single command: git reset --hard upstream/master[21:27:57] *** pilhuhn is now known as pil-afk[21:28:05] <asaldhan> bstansberry: see, we always learn. thanks, dmlloyd[21:36:33] <asaldhan> dmlloyd: if I reset my master, I do not have to do a push?[21:36:54] <dmlloyd> it depends on whether you want to also reset your remote to upstream/master[21:37:07] <asaldhan> dmlloyd: yeah I want to reset basically.[21:37:21] <dmlloyd> well if your remote was already reset to upstream then you don't need to[21:37:29] <dmlloyd> usually a "git status" will tell you if you've diverted[21:37:30] <asaldhan> dmlloyd: case when I have totally messed up my master.[21:37:33] <dmlloyd> diverged*[21:37:46] <dmlloyd> it's not like svn[21:37:52] <dmlloyd> your local copy isn't a slave of the remote copy[21:38:01] <dmlloyd> they're two isolated git repositories[21:38:23] <dmlloyd> so if you haven't pushed anything to your origin/master in a while then it won't have changed without your knowledge[21:38:27] <asaldhan> dmlloyd: right. but if I have checked in commits to my master and they are bogus.[21:38:43] <dmlloyd> when you commit, it commits into the repository local to your machine[21:38:52] <dmlloyd> your local checkout is a complete repository[21:38:54] <asaldhan> dmlloyd: I meant pushed[21:39:12] <asaldhan> dmlloyd: if I have pushed some changes to my master and they are screwed up in relation to upstream[21:39:15] <dmlloyd> if you want to change your remote origin/master then yes pushing is what you want to do[21:40:06] <asaldhan> dmlloyd: when I reset git reset --hard upstream/master[21:40:11] <asaldhan> dmlloyd: what happened?[21:40:20] <asaldhan> dmlloyd: local branch was replaced with upstream/master?[21:40:27] <dmlloyd> you're changing your local master to point to the same commit as upstream/master[21:40:48] <dmlloyd> where upstream/master is as of the last time you did "git fetch upstream"[21:40:49] <asaldhan> dmlloyd: now I want my master to get synched with upstream/master.[21:41:26] <dmlloyd> if you have local commits and you want to combine your local commits with upstream commits, use "git rebase upstream/master", which will put the new upstream commits chronologically before your local commits.[21:41:39] <dmlloyd> well, sequentially before. whatever.[21:42:03] <asaldhan> Interactive rebase already started[21:42:08] <asaldhan> dmlloyd: ^ whats that?[21:42:16] <dmlloyd> that means you're in the middle of a rebase already.[21:42:20] <dmlloyd> "git rebase --cancel"[21:42:29] <dmlloyd> or maybe "--abort" I forget[21:42:47] <dmlloyd> use "git log upstream/master..master" to see what you've got that upstream hasn't, and "git log master..upstream/master" to see what upstream's got that you haven't[21:44:26] *** epbernard has joined #jboss-as7[21:44:26] *** epbernard is now known as emmanuel[21:44:26] *** ChanServ sets mode: +v emmanuel[21:52:10] *** pferraro has quit IRC[21:54:41] *** jfd has quit IRC[22:00:41] *** jpederse has quit IRC[22:02:12] *** stuartdouglas has joined #jboss-as7[22:02:12] *** ChanServ sets mode: +v stuartdouglas[22:09:06] *** frainone has quit IRC[22:13:45] *** fnasser has quit IRC[22:16:14] *** nickboldt has quit IRC[22:32:10] *** nickboldt has joined #jboss-as7[22:32:11] *** ChanServ sets mode: +v nickboldt[22:35:23] *** mmoyses has quit IRC[22:36:57] *** mbg is now known as mbg|away[22:49:15] *** jwulf has joined #jboss-as7[23:03:11] <baileyje> bstansberry: Found an issue. If an address has a path element with a ':', or a '/' it does not like it.[23:03:36] <bstansberry> the CLI?[23:04:11] <baileyje> hyeah[23:04:26] <baileyje> the CLI. Can i just quote it up?[23:04:37] <bstansberry> alexey said he was going to support quoting it[23:06:29] <baileyje> Yeah. I have tired several combinations with no luck[23:07:41] <bstansberry> I think there is a JIRA; i'm looking[23:08:17] <bstansberry> I really hate JIRA!!![23:09:56] <bstansberry> JBAS-8929[23:09:57] <jbossbot> jira [JBAS-8929] CLI address parsing is overly restrictive [Resolved (Done) Bug, Major, Alexey Loubyansky] https://issues.jboss.org/browse/JBAS-8929[23:10:15] <bstansberry> baileyje: can you comment on the JIRA, since it says it's fixed[23:10:37] <baileyje> when was it fixed?[23:10:50] <bstansberry> Monday[23:16:17] *** miclorb has joined #jboss-as7[23:21:58] <nickarls> hmm, I dropped a war into the deployments folder of a running standalone beta1-server but didn't get any deployment marker file...[23:22:56] <bstansberry> nickarls: see the README.txt in the deployments folder[23:23:04] <stuartdouglas> dmlloyd: I have a question about exactly when the dependencyUninstalled() ServiceListener method is called[23:23:27] <stuartdouglas> It looks like it gets called when a service is installed and a dependency is not added, is that correct?[23:23:48] <bstansberry> baileyje: quick approval please: http://fpaste.org/gUpu/[23:23:49] <stuartdouglas> I mean, when a dependency is missing[23:24:39] <dmlloyd> stuartdouglas: yeah. It may be a transient condition[23:24:42] <baileyje> bstansberry: We may conflict.[23:25:02] <bstansberry> maeste also has a big patch[23:25:20] <bstansberry> which i was about to push[23:25:23] <baileyje> bstansberry: I have update that file.[23:25:25] <nickarls> oops, misread, the dodeploy should be user-generated...[23:25:32] <baileyje> to add several of the bits you have there[23:25:52] <bstansberry> do you have them all?[23:25:55] <stuartdouglas> In that case in AbstractDeploymentUnitService.DeploymentServiceListener I think the dependencyUninstalled part of the listener may be casing the transient test failures[23:26:21] <stuartdouglas> as if a service is added before it's deps are present, the the listener will be removed and count decremented[23:26:34] <baileyje> bstansberry: I have a big DS chnage.[23:26:40] <bstansberry> I can skip committing that; i was reviewing another description patch and the description demo fails because those were missing, so i added them[23:26:47] <stuartdouglas> and the service could then have it's deps added and start up[23:28:07] <stuartdouglas> but because lister could report the deployment complete before these services are started[23:28:19] <bstansberry> baileyje: this is maeste's commit[23:28:21] <bstansberry> https://github.com/maeste/jboss-as/commit/a5449e843d5046af53d4e16cc24e10ec0bb2c171[23:28:22] <jbossbot> git [jboss-as] a5449e8.. Stefano Maestri read-resource-description support for datasources and resource-adapters subsystem[23:29:10] <baileyje> yeah. That will be fun..[23:29:32] <baileyje> I have converted all the DS instances to be added as services.[23:29:42] <baileyje> sorry.. Operation[23:29:43] <baileyje> s[23:29:52] <baileyje> Instead of all from the subsystem add[23:30:01] <bstansberry> :-)[23:30:28] <baileyje> So there will likely be a major conflict between me and maeste[23:30:36] <nickarls> is there any list of what is supposed to be semi-working and what not? For example, I get a "Unable to parse document" for a faces-config.xml but I don't know at what stage JSF is in AS 7 so I don't know if to JIRA or not.[23:30:39] <bstansberry> ah sure[23:31:13] <baileyje> just with the Datasource subsystem though[23:31:35] <baileyje> his touches RA as well[23:31:53] *** emuckenhuber has quit IRC[23:32:14] <baileyje> go ahead and push his. I will sort out the DS descriptions.[23:32:22] <bstansberry> nickarls: there's no master list, no. there is this on JSF though: http://community.jboss.org/wiki/JSFonAS7[23:33:09] <bstansberry> I know Nihility plans to write up something, but he's not in at the moment[23:33:13] <stuartdouglas> bstansberry: If you get the chance would you be able to try https://github.com/stuartwdouglas/jboss-as/tree/transient and see if you still get the transient test failures ?[23:33:52] <bstansberry> stuartdouglas: sure; 1 sec[23:37:02] <jbossbot> git [jboss-as] push master 2a13331.. Stefano Maestri read-resource-description support for datasources and resource-adapters subsystem[23:37:03] <jbossbot> git [jboss-as] push master bdfe4e5.. bstansberry at jboss dot com Use POOL_NAME not POOLNAME[23:37:03] <jbossbot> git [jboss-as] push master d5f2e6e.. bstansberry at jboss dot com Datasource description fix, and tests of the connector extension descriptions[23:37:03] <jbossbot> git [jboss-as] push master URL: http://github.com/jbossas/jboss-as/compare/328151e...d5f2e6e[23:37:29] <bstansberry> maeste: ^^^[23:38:38] <maeste> bstansberry: thanks[23:39:36] *** maeste has quit IRC[23:42:16] *** pgier has quit IRC[23:42:50] <bstansberry> stuartdouglas: smoke tests passed twice, 3rd run going....[23:43:09] <wolfc> dmlloyd, what do you mean with JBAS-8971 ?[23:43:11] <jbossbot> jira [JBAS-8971] Implement JMS MDB integration [Open (Unresolved) Sub-task, Major, Unassigned] https://issues.jboss.org/browse/JBAS-8971[23:43:32] <stuartdouglas> It is always nice that something that would explain the bug perfectly actually turns out to be the source of the bug :-)[23:43:34] <wolfc> a MDB should be able to do JMS inflow without any specifics[23:44:17] <dmlloyd> well MDBs are not restricted to JMS, so one would assume that it's a separate integration step[23:44:28] <stuartdouglas> If this is the cause the EJB tests would be much more prone to failure, as they set up deps between services that do not exist yet[23:44:49] <dmlloyd> stuartdouglas, they are, apparently, especially for certain people's systems[23:44:54] <dmlloyd> just not mine I guess :)[23:44:59] <wolfc> nah, just something that needs a test[23:45:29] <stuartdouglas> or for mine either[23:45:31] <wolfc> but we may need a rar that doesn't do JMS to test things properly[23:45:47] <dmlloyd> yeah perhaps[23:46:03] *** emuckenhuber has joined #jboss-as7[23:46:03] *** ChanServ sets mode: +v emuckenhuber[23:46:08] <wolfc> and if somebody says quartz then he'll be the designated volunteer[23:46:31] * dmlloyd says nothing of the sort :)[23:46:40] <bstansberry> quarks[23:46:44] <bstansberry> oh, so close!!![23:46:49] <bstansberry> yet so far[23:46:54] <dmlloyd> perhaps remoting would be a fun way to test inflow...[23:47:05] <dmlloyd> remoting.rawr[23:47:27] <bstansberry> stuartdouglas: it's passed 6 or 7 times now[23:47:31] * dmlloyd hopes we don't need to deploy a rar to use JMS inflow[23:47:51] <stuartdouglas> bstansberry: how often was it failing before?[23:47:55] <wolfc> ? you need the hornetq-ra.rar[23:48:18] <bstansberry> i doubt i could go two runs without a failure[23:48:22] * dmlloyd is thusly disappointed[23:48:37] <bstansberry> i'd say the odds of failure on at least 1 test were 75%[23:48:57] <bstansberry> my patch last night reduced it to 25%[23:49:09] * bstansberry is glad his patch was wrong[23:49:10] <wolfc> we could do a very simplistic message inflow though[23:49:22] <wolfc> IRC ? :-)[23:49:34] <bstansberry> jboss-bot.rar[23:49:38] <dmlloyd> heh[23:49:44] <wolfc> +1[23:49:45] <baileyje> bstansberry: What is the demo to run to test the descriptions?[23:50:00] <bstansberry> description[23:50:35] <baileyje> cool..[23:50:41] *** mbg|away is now known as mbg[23:50:57] <bstansberry> aha! there IS a unit test of that[23:51:22] <bstansberry> well, there was meant to be one, but it's missing the critical piece :-)[23:51:28] <stuartdouglas> bstansberry: unfortunately that is only half the fix, the problem now becomes that if a service is missing a dep deployment blocks forever[23:52:35] <stuartdouglas> I will do up a proper fix...[23:53:14] <bstansberry> k. thanks much for digging into it[23:54:18] <bstansberry> it appears I broke the build; better go fix that[23:55:51] <wolfc> dmlloyd, for ejb-jar parsing do you want to do merging and augmentation in one go?[23:56:41] <dmlloyd> I thought maybe that descriptors could go first, then the annotation processor would merge/augment[23:56:52] <dmlloyd> since descriptors can disable annotation processing[23:57:26] *** jpearlin has joined #jboss-as7[23:58:03] <wolfc> ah yes, bummer.[23:58:12] <jbossbot> git [jboss-as] push master 04f816a.. bstansberry at jboss dot com Don't commit test since I'm not committing the description text[23:58:13] <jbossbot> git [jboss-as] push master URL: http://github.com/jbossas/jboss-as/compare/d5f2e6e...04f816a[23:58:32] <dmlloyd> we'll almost certainly have to add an index or two to description to help with the merging