April 1, 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 | >
April 1, 2011 NOTICE: This channel is no longer actively logged.
[00:02:19] <Nihility> stuartdouglas: i have a few more mins to talk about it[00:02:30] <ALR> JAX-RS test case now locally passing on Sun JDK 6u24[00:02:34] <stuartdouglas> I think I have it figured out, I did not have the right service import[00:02:37] <Nihility> stuartdouglas: what issue are you seeing[00:02:58] <ALR> I'll mail the list and discuss further whether we should add a Woodstox dep or Maven Enforcer config to require a minimal JDK version[00:03:24] <Nihility> ALR: an alternative fix is to somehow have arq set the jaxp provider through modules[00:03:36] <Nihility> ALR: i must admit i dont know how that all works[00:03:42] *** jclingan has joined #jboss-as7[00:03:51] <ALR> :)[00:03:55] <ALR> Stopgap is Enforcer config[00:04:11] <ALR> But that won't help us w/ testing users on older (not THAT old) JDKs[00:04:11] *** jclingan has quit IRC[00:04:12] <Nihility> ALR if it did set the jaxp provider option it would be the most consistent with a real server[00:04:27] <ALR> Nihility: I do dislike that option though[00:04:31] <Nihility> thats what standalone.sh does[00:04:40] <Nihility> there is an api way to do it too[00:04:40] <ALR> Just because every time we add some command-line config, it makes things so much less usable[00:04:57] <ALR> An API/Service for MSC would be great if we could[00:05:19] <Nihility> it has to happen early in the bootstrap though before msc[00:05:29] <Nihility> as it stubs out the jdk jaxp provider[00:05:39] <Nihility> its part of module boot[00:05:42] *** jclingan has joined #jboss-as7[00:06:40] * Nihility goes to look at the source for the embedded container[00:07:22] * ALR onto reapplying his patches for OSGi[00:07:48] <Nihility> ALR: ah i think i can make it consitent here[00:08:11] *** jclingan has quit IRC[00:08:23] <ALR> Nihility: In the Embedded Container?[00:08:29] <Nihility> yeah[00:08:35] <ALR> Nihility: Point me?[00:08:41] <Nihility> k sec[00:09:18] *** jclingan has joined #jboss-as7[00:10:38] <Nihility> __JAXPRedirected.changeAll(jaxpModuleIdentifier, Module.getBootModuleLoader());[00:10:43] <Nihility> that should be added to[00:11:10] <Nihility> EmbeddedServerFactory.create()[00:11:14] <Nihility> before starting the server[00:11:23] <ALR> JIRA coming.[00:11:33] <ALR> I wonder if that'll address both failures I saw.[00:11:36] <Nihility> not sure if we want to hardocde that module[00:11:45] <Nihility> name[00:11:45] *** jclingan has joined #jboss-as7[00:11:46] <Nihility> that is[00:11:54] <Nihility> my guess is we do[00:12:27] <Nihility> ALR: do you have the oldd JDK still?[00:12:28] *** jclingan has quit IRC[00:12:40] <Nihility> i can push a patch to my repo for you to try[00:13:43] <ALR> Nihility: Yeah, but it's not my default[00:13:50] <ALR> Nihility: https://issues.jboss.org/browse/JBAS-9210[00:13:52] <jbossbot> jira [JBAS-9210] Test Failures in SUn JDK 1.6.0_u20 [Open (Unresolved) Bug, Major, Andrew Rubinger] https://issues.jboss.org/browse/JBAS-9210[00:14:03] <ALR> I wanna just jot this down and curb it for now while I do the more pressing OSGi stuff.[00:14:10] *** jclingan has joined #jboss-as7[00:14:14] <stuartdouglas> can someone merge my master?[00:14:17] <Nihility> ALR: ok yeah ill just leave a note on that[00:14:22] <ALR> Cool, thanks.[00:14:26] <stuartdouglas> IT has some resteasy fixes[00:14:32] <Nihility> stuartdouglas: can you recap that jaxb issue[00:14:33] *** jclingan has quit IRC[00:14:42] <stuartdouglas> It was a missing services="import"[00:14:50] <Nihility> ok so let me guess[00:15:04] <stuartdouglas> https://github.com/stuartwdouglas/jboss-as/commit/e6c999987eccd9a960bd5047520715b6a690f4c9#L2R31[00:15:04] <jbossbot> git [jboss-as] e6c9999.. Stuart Douglas Add Resteasy JAXB provider[00:15:10] <Nihility> ohhhhhh i got it[00:15:15] <Nihility> now your statement makes sense[00:15:39] * stuartdouglas really should have picked it up earlier[00:15:41] <Nihility> the api change did fix your problem but you needed the import[00:15:46] <stuartdouglas> yes[00:15:47] *** jclingan has joined #jboss-as7[00:16:07] <Nihility> the way it attempts to use the platform version wasnt working because we filter it[00:16:09] *** jclingan has quit IRC[00:16:14] <Nihility> i get it[00:16:18] <stuartdouglas> which resteasy providers are we going to provide in AS7 ?[00:16:34] <stuartdouglas> I just added JAXB, but there are a heap of others like yaml and JSON[00:16:45] <dmlloyd> probably JSON at least[00:16:51] <dmlloyd> what did we provide in AS6?[00:17:01] * stuartdouglas looks[00:17:01] <Nihility> yeah as long as they are tiny[00:17:18] <Nihility> ALR: dude you have to switch to idea[00:17:28] <Nihility> ALR: you will love it[00:17:30] <ALR> Nihility: I've tried a few times.[00:17:44] <ALR> Nihility: And for some stuff it made life easier. For instance the old AS tree.[00:17:45] <Nihility> anyone who has 80 million project[00:17:49] <Nihility> should love this stuff[00:18:13] <ALR> I use Eclipse, but have separate workspaces for each project group[00:18:19] <Nihility> (or just the AS which is 80 million projects haha[00:18:27] <ALR> For instance AS has its own Formatting rules to pass checkstyle.[00:18:37] <ALR> And in Eclipse I close all the modules I'm not working on to save RAM.[00:18:51] <ALR> So I have a flawed, yet finely tuned working process.[00:19:07] <ALR> You wanna know the REAL reason I don't use IDea?[00:19:11] <ALR> The fonts look weird.[00:19:14] <Nihility> not OSS?[00:19:15] <ALR> And I can't get used to it.[00:19:17] <stuartdouglas> It looks like we provided atom, jettison, jackson, multipart and yaml[00:19:19] <Nihility> oh you can change that[00:19:20] <ALR> Nope. Just LnF[00:19:23] <Nihility> i use Monaco[00:19:27] <stuartdouglas> I will create a jira to add then[00:19:30] <Nihility> i love Monaco[00:19:30] <ALR> I know. I couldn't get everything to look just right.[00:19:34] <Nihility> best font ever[00:19:39] <Nihility> yeah i agree with you there[00:19:42] <ALR> I use whatever that new Fedora monospaced thing is[00:19:45] <Nihility> swing L&F not as good as SWT[00:19:51] <Nihility> oh that font sucks[00:20:02] <Nihility> well actually its ok if you have the right version[00:20:08] <ALR> SWT *is* nice.[00:20:32] <Nihility> the best thing about idea is[00:20:36] <Nihility> i no longer sit and wait[00:20:41] *** jclingan has joined #jboss-as7[00:20:42] <Nihility> while it builds for 6 years[00:20:44] <Nihility> and crashes[00:21:04] <ALR> I'll move to Idea when Jesper...moves to any editor.[00:21:07] <stuartdouglas> hmm, jira is dead[00:21:08] *** jclingan has quit IRC[00:21:16] <Nihility> stuartdouglas: k ill merge your stuff[00:21:16] <ALR> I think he uses a steady hand and a magnet.[00:22:10] <stuartdouglas> With that latest merge the Seam 3 archetype works on AS7 out of the box, with JPA JAX-RS and CDI[00:22:17] *** jclingan has joined #jboss-as7[00:23:31] *** pferraro has quit IRC[00:23:32] *** jclingan has quit IRC[00:23:56] *** jclingan has joined #jboss-as7[00:23:58] <Nihility> ALR: i wonder what would happen to him if he used an IDE[00:24:24] *** irooskov has quit IRC[00:24:24] *** jclingan has quit IRC[00:24:36] <Nihility> he would have less to drink about i guess :)[00:24:37] <ALR> Nihility: He's l33t[00:25:39] * stuartdouglas can't image trying to write java without an IDE[00:25:57] *** jclingan has joined #jboss-as7[00:26:23] *** jclingan has quit IRC[00:26:33] * jamezp agrees with stuartdouglas[00:27:03] *** mbg has joined #jboss-as7[00:27:03] *** ChanServ sets mode: +v mbg[00:27:10] <jamezp> I'd lose my mind with out content assist and JavaDoc pop-ups.[00:27:45] *** jclingan has joined #jboss-as7[00:28:03] <stuartdouglas> jamezp: Guess we a not l33t enough :-)[00:28:21] <jamezp> Hahaha. Probably not :-)[00:28:22] *** jclingan has quit IRC[00:28:47] <Nihility> stuartdouglas: i still do it from time to time[00:28:55] <Nihility> eclipse made me use vim alot[00:29:14] <Nihility> do i feel like spending 20 mins creating a new workspace....[00:29:15] <Nihility> no i dont[00:29:16] <Nihility> :)[00:29:16] <stuartdouglas> that is different :-)[00:29:27] <stuartdouglas> The seam booking example nearly deploys[00:29:39] <stuartdouglas> it fails loading in the initial database[00:29:46] *** jclingan has joined #jboss-as7[00:30:06] *** jclingan has quit IRC[00:30:35] *** pferraro has joined #jboss-as7[00:30:36] *** ChanServ sets mode: +v pferraro[00:30:48] *** irooskov has joined #jboss-as7[00:31:35] *** jclingan has joined #jboss-as7[00:32:32] <baileyje> dmlloyd: https://github.com/baileyje/jboss-invocation/commits/master[00:33:06] *** opalka has quit IRC[00:33:07] *** jclingan has quit IRC[00:33:12] <Nihility> bstansberry: hey can you take a quick look at stuartdouglas' patch[00:33:30] <Nihility> oh wait[00:33:30] <Nihility> nm[00:33:31] <bstansberry> sure[00:33:37] <Nihility> i fast forwarded to his[00:33:38] <bstansberry> that was easy :)[00:33:42] <Nihility> which i think he rebased on yours[00:33:53] <Nihility> it lookedd like he was changing a bunch of managment stuff[00:33:53] <Nihility> haha[00:34:12] *** jclingan has joined #jboss-as7[00:34:46] <baileyje> dmlloyd: I gotta take the dog for a walk, Be back on in awhile.[00:34:47] *** jclingan has quit IRC[00:34:57] <dmlloyd> baileyje: are we actually using void setInterceptors(final List<Interceptor> interceptors, int nextIndex)?[00:35:02] *** asaldhan has left #jboss-as7[00:35:03] <dmlloyd> other than in the other setInterceptors method[00:35:43] <dmlloyd> also if we're going the clone route then we should do clone = (InterceptorContext) super.clone();[00:35:48] <dmlloyd> and not copy over the fields[00:36:37] *** jclingan has joined #jboss-as7[00:36:40] <dmlloyd> otherwise looks good[00:37:02] <dmlloyd> ah I see the ChainedInterceptor uses the index[00:37:03] *** jclingan has quit IRC[00:37:08] <dmlloyd> but the clone thing stands[00:38:17] <jbossbot> git [jboss-as] push master 7237588.. Stuart Douglas Fix CDI injection into JAX-RS resources[00:38:17] <jbossbot> git [jboss-as] push master 5f1c9de.. Stuart Douglas Expand JAXRS+CDI tests[00:38:17] <jbossbot> git [jboss-as] push master ed066f0.. Stuart Douglas fix CDI injection problem[00:38:18] <jbossbot> git [jboss-as] push master e6c9999.. Stuart Douglas Add Resteasy JAXB provider[00:38:18] <jbossbot> git [jboss-as] push master URL: http://github.com/jbossas/jboss-as/compare/17c9d0d...e6c9999[00:42:14] *** jpearlin has joined #jboss-as7[00:43:26] <stuartdouglas> hmm, weld needs a dep on all PU's in the application[00:49:14] *** alesj has quit IRC[00:52:06] * dmlloyd will just make that change for baileyje to speed things along when he gets back...[00:57:35] *** mbg is now known as mbg|away[01:01:30] <dmlloyd> http://github.com/dmlloyd/jboss-invocation/commit/00ebfd5[01:01:30] <jbossbot> git [jboss-invocation] 00ebfd5.. David M. Lloyd Minor cleanup[01:07:51] *** bstansberry has quit IRC[01:09:58] *** mbg|away is now known as mbg[01:11:38] <baileyje> dmlloyd: back..[01:11:59] <dmlloyd> check my minor cleanup, ended up being more minor than I thought :)[01:12:21] <dmlloyd> but you didn't have my earlier change so I needed to fix the test anyway[01:13:07] <baileyje> dmlloyd: Looks good to me.[01:13:29] <jbossbot> git [jboss-invocation] push master 5cefea9.. John E. Bailey Add cloning support to InterceptorContext[01:13:29] <jbossbot> git [jboss-invocation] push master 00ebfd5.. David M. Lloyd Minor cleanup[01:13:29] <jbossbot> git [jboss-invocation] push master URL: http://github.com/jbossas/jboss-invocation/compare/06c69b5...00ebfd5[01:13:57] <dmlloyd> found a nasty little issue in SecureClassLoader. It's using CodeSource as a hash key, which includes URL. Blargh.[01:14:21] <dmlloyd> might modify modules to skip SecureClassLoader and work out my own codeSource cache. Might give a teeny perf boost anyway[01:14:31] <stuartdouglas> can someone merge my master?[01:14:41] * dmlloyd can...[01:14:52] <stuartdouglas> It has a fix that sets up dependencies so web does not start util weld and JPA are ready[01:15:26] <stuartdouglas> The seam booking example nearly works, there is just only little weld + EJB issue I have found ...[01:17:12] *** slaboure has quit IRC[01:20:45] <jbossbot> git [jboss-as] push master a33c114.. Stuart Douglas Set up depedencies so web deployments do not start util JPA and weld are up[01:20:45] <jbossbot> git [jboss-as] push master URL: http://github.com/jbossas/jboss-as/compare/e6c9999...a33c114[01:20:52] <dmlloyd> tests are getting slower and slower[01:21:43] *** jwulf has joined #jboss-as7[01:21:44] <stuartdouglas> due to there being more of them or because the AS is slowing down?[01:22:19] *** frainone has quit IRC[01:27:13] <dmlloyd> I think the tests are slowing down[01:27:20] <dmlloyd> some take as long as 10s or more to run[01:32:08] *** jwulf_ has joined #jboss-as7[01:37:09] *** pferraro has quit IRC[01:44:18] *** jnoer has joined #jboss-as7[01:48:27] *** mbg has quit IRC[02:01:06] <ALR> Got quiet. :)[02:01:24] <Nihility> Not just the tests[02:01:32] <Nihility> Our boot time is slower too[02:01:52] <smarlow> nice, passed 6 of 6 tck tests. its a start :)[02:02:03] <Nihility> cool[02:02:14] <Nihility> is that the web profile jpa stuff?[02:02:37] <smarlow> yes, shelly was getting stuck and I took a look, had to fix a bug but got it going.[02:02:47] <Nihility> ALR: Yeah a lot of us do the family thing around 6-8[02:03:05] <ALR> Definitely.[02:03:16] <ALR> Nihility: She's 14 months now?[02:03:20] <ALR> ish?[02:03:27] <Nihility> Im partially back[02:03:41] <Nihility> Still have to get her to sleep probably[02:03:53] <Nihility> 14 yep[02:03:59] <Nihility> Oh wait[02:04:00] <Nihility> O[02:04:02] <Nihility> N[02:04:05] <Nihility> no[02:04:08] <Nihility> 15[02:04:10] <Nihility> Haha[02:04:25] <Nihility> iPad :)[02:04:43] *** jpearlin has quit IRC[02:04:52] <ALR> Hehe[02:05:36] <Nihility> dmlloyd: are you doing a 1 year party for Simon[02:05:55] <dmlloyd> I don't think so[02:06:54] <Nihility> what you don't want to eat the cardboard at chuk e cheese?[02:08:20] <dmlloyd> they never remember their first birthday anyway[02:08:26] <Nihility> new feature request for jbossbot anytime some types a version he says how many issues are left[02:08:54] <ALR> Girl, you got so many issues, you need JIRA.[02:09:55] <Nihility> Maybe one of these I'll write a pester extension[02:10:18] <Nihility> He tracks you down on irc and bitches that you still have x issues open[02:11:09] <ALR> Nihility: Trying to make yourself obsolete?[02:11:31] <Nihility> then I could write more code![02:12:15] <ALR> I love squashing now.[02:12:41] <ALR> Scared you're gonna lose stuff but don't wanna make it a commit on its own? Squash it.[02:13:03] <ALR> Also I have a new programming paradigm:[02:13:06] <smarlow> alright, here is mine[02:13:11] <ALR> Git Filter, Motherf*cker[02:13:17] *** bgeorges has quit IRC[02:14:26] <smarlow> when I launch the AS, it could notice an env variable setting "AS_RunThisClass". That launches a little gui that tells me that I have an AS instance running[02:14:58] <ALR> smarlow: How would it know?[02:15:20] <ALR> smarlow: We'd need to leave some sort of a non-configurable breadcrumb[02:15:41] <smarlow> the class is in an external jar (not shipped as well). its just a gadget[02:16:01] <ALR> Otherwise the config per server instance could put it off the map. For instance you can't just check port 8080. That could be another process, or maybe AS is bound to a different port.[02:16:08] <ALR> It's not about where it's packaged.[02:16:27] <ALR> More like: what's the registration point when AS comes up that this util will ping for running instances?[02:16:36] <smarlow> well, having a port checker is cool too[02:17:12] <ALR> There are port checkers.[02:17:29] <Nihility> Netstat -p is your friend[02:17:31] * ALR is just not that empathetic to users who don't know what processes are on their machine. :)[02:17:38] <dmlloyd> it's not just the ports - AS instances can't share runtime directories either[02:17:54] <ALR> Mhmm.[02:17:58] <smarlow> netstat is a big friend :) I scripted some port checkers with that[02:18:23] <Nihility> I would rather have tetris[02:18:41] *** jpearlin has joined #jboss-as7[02:18:58] <Nihility> jpearlin: I merged your patch btw[02:19:07] <jpearlin> the smoke test? thanks[02:19:14] <smarlow> its customizable, when the AS is launched, tetris is too. you would never start a build with the AS running again :)[02:20:19] <jpearlin> Nihility: I was thinking about the base64 issue some more...[02:20:28] <Nihility> I would have stuck with eclipse if it would have at least offered tetras to me when it hung for longamou ts of time[02:20:45] <jpearlin> shouldn't every decoded string's length be a multiple of 4, based on the algorithm?[02:21:00] <jpearlin> so isnt the issue only when the stream has underflow?[02:21:09] <Nihility> no that's the whole point of the padding[02:21:22] <jpearlin> that's my point...[02:21:33] <jpearlin> I thought you said you found the issue when decoding an encoded string[02:21:41] <Nihility> The padding is because the input isn't big enough[02:22:02] <Nihility> Right so the decoding needs to understand how to treat thenpadding[02:22:02] <jpearlin> right...but that guarantees the output is a multiple of 4 bytes per every 3[02:22:12] <jpearlin> right...so I guess I misunderstood...[02:22:23] <jpearlin> the issue is just with the end when it needs to determine if padding was added originally[02:22:28] <jpearlin> ?[02:22:32] <Nihility> when there is padding it means the decoded output should have less bytes[02:22:41] <dmlloyd> http://en.wikipedia.org/wiki/Base_64#Padding[02:22:44] <jpearlin> ok....that makes a lot more sense[02:22:54] <jpearlin> I thought something really weird was going on[02:23:04] <jpearlin> so its an issue with the logic that handles removing the padding on decode[02:23:25] <jpearlin> so the test is to feed in strings for encoding that are not multiples of 3 then testing the decode[02:23:48] <dmlloyd> the transcoding algorithm should not rely on data coming in in any multiple[02:23:58] <jpearlin> it doesn't[02:24:22] <jpearlin> it only adds the padding if we hit the end of stream and the current chunk is less than 3 bytes[02:24:26] <dmlloyd> the 8 examples in that wikipedia article should be sufficient to fully test the state machine in both directions[02:24:32] <jpearlin> ok[02:24:39] <jpearlin> I'll create unit tests based on those to start[02:24:57] <dmlloyd> though the choice of plaintext in that article is perhaps unfortunate[02:25:07] <jpearlin> ha[02:25:15] <jpearlin> I'll use the length as a guide, not the words themselves[02:25:28] <Nihility> it is ironic[02:25:54] <Nihility> ok bib[02:25:59] <Nihility> Bbiab[02:26:34] *** jamezp has quit IRC[02:43:55] *** jnoer has quit IRC[02:51:24] *** pferraro has joined #jboss-as7[02:51:24] *** ChanServ sets mode: +v pferraro[02:56:21] *** lgao has joined #jboss-as7[03:05:28] *** fnasser has quit IRC[03:17:20] <ALR> Nihility, dmlloyd: Twice today I've gotten a hang in the AS build:[03:17:35] <ALR> http://pastebin.com/iLkfnpTs[03:17:37] <ALR> Known issue?[03:18:02] *** maxandersen has joined #jboss-as7[03:18:02] *** ChanServ sets mode: +v maxandersen[03:18:14] <ALR> Ah, more context, part of Running org.jboss.as.controller.test.ExistingConnectionRemoteProxyControllerTestCase[03:19:39] *** pferraro has quit IRC[03:22:11] <baileyje> anyone have good luck debugging tests in the testsuite ?[03:22:26] <stuartdouglas> integration or smoke?[03:23:00] <baileyje> smok[03:23:01] <baileyje> e[03:23:16] <stuartdouglas> you can use -Dmaven.surefire.debug, but it does not work that well[03:23:53] <stuartdouglas> so you need to go through all the tests till you get to the one you want, or change to pom so only 1 is enabled[03:25:12] <baileyje> thanks, I will see if I can avoid it..[03:27:30] *** irooskov has quit IRC[03:29:10] <ALR> stuartdouglas: Getting the problem where -Dtest= doesn't set jboss.home ?[03:29:16] <ALR> (I saw that one earlier today)[03:29:46] <stuartdouglas> that could be what caused it, I am not sure[03:31:38] *** smcgowan has joined #jboss-as7[03:31:38] *** ChanServ sets mode: +v smcgowan[03:32:05] *** jwulf has quit IRC[03:32:51] *** bgeorges has joined #jboss-as7[03:33:50] *** jwulf has joined #jboss-as7[03:34:45] <maxandersen> im testing JBAS-9177 on windows - as starts up, jboss-admin.bat works but localhost:8080 shows nothing …is that expected ?[03:34:46] <jbossbot> jira [JBAS-9177] Launch script does not work on Windows [Open (Unresolved) Bug, Critical, Unassigned] https://issues.jboss.org/browse/JBAS-9177[03:35:20] <maxandersen> shouldn't beta1 show something on localhost:8080 ? (even just a white blank page)[03:37:34] <maxandersen> dmlloyd or Nihility ^ ?[03:42:56] <baileyje> dmlloyd: I think we are in pretty good shape. We need a release of jboss-invocation, and I have to route an executor into this process from the configuration.[03:44:23] <smcgowan> maxandersen: yes that's expected, there is no landing page currently[03:46:41] <maxandersen> smcgowan: actually i'm getting not even a white page when accessing lcoalhost:8080; 127.0.0.1:8080 works though weirdly enough - i'll blame the virtualization layer for that one ;)[03:47:07] *** bgeorges has quit IRC[03:48:37] <maxandersen> smcgowan: besides the space issue - is it expected that there is no .conf file for it to read ? it looks for .conf.bat not just .conf.[03:48:52] *** jclingan has joined #jboss-as7[03:50:32] *** jclingan1 has joined #jboss-as7[03:50:34] <smcgowan> maxandersen: not sure about that - i'd have to check on windows[03:51:01] <maxandersen> smcgowan: the bat file explicitly looks for that conf.bat ..but there is only a conf file in the distro.[03:51:11] <smcgowan> i see[03:53:18] *** jclingan has quit IRC[03:53:21] *** jpearlin has left #jboss-as7[03:54:06] <maxandersen> there I fixed the spacing issue...[03:54:16] <maxandersen> ill try and create a patch ;)[03:54:35] <maxandersen> jetlag makes you do crazy things[04:03:33] <smcgowan> stuartdouglas: u there?[04:03:40] <stuartdouglas> smcgowan: yes[04:04:03] <smcgowan> stuartdouglas: with one of smarlows latest fixes, we have some JPA tck tests passing[04:05:00] <smcgowan> stuartdouglas: but i still have some questions regarding the defining logical names for resources[04:05:20] *** bstansberry has joined #jboss-as7[04:05:20] *** ChanServ sets mode: +v bstansberry[04:05:23] <smcgowan> i can get the tess to pass if i use a jndi-name of java:/DB1 and add the jta-data-source as java:/DB1[04:05:52] <smcgowan> but can't seem to get the same results when defining jdbc/DB1 which is what the tests have defined[04:06:18] <smcgowan> as well as the entries of the resource-ref elements in the jboss*.xml files[04:06:33] <stuartdouglas> at the moment we only support java: AFAIK[04:07:13] <smcgowan> so that blocks much of the TCK that depends on that[04:07:23] <smcgowan> that's all i've got to work as well[04:07:54] <stuartdouglas> so the tck requires us to support non java: names?[04:07:56] <smcgowan> but have changed it to java: to try to move forward on testing such as in the JPA case[04:08:44] <smcgowan> description of https://issues.jboss.org/browse/JBCTS-1089 is per the set up instructions[04:09:24] *** ALR has quit IRC[04:10:12] <smcgowan> and then the lookup, obviously is: java:comp/env/jdbc/DB1[04:11:57] <stuartdouglas> how are you doing the java:comp/env/jdbc/DB1 binding ?[04:13:02] <maxandersen> anyone know if -jboss-home is equivalent to -Djboss.home.dir ?[04:13:20] <maxandersen> seems the scripts uses both…just not sure which one is the right one[04:14:12] <maxandersen> standalone uses jboss.home.dir domain -jboss-home...[04:14:19] <maxandersen> which is correct?[04:16:12] <smcgowan> stuartdouglas: there is a resource-ref element in the WEB-INF/web.xml[04:16:27] <bstansberry> maxandersen: -jboss-home is an arg understood by the process that controls all the process in domain mode[04:16:50] <stuartdouglas> resource-ref is only half implemented, but it should work if it is using lookup-name[04:16:54] <maxandersen> bstansberry: ok, so that is a expected difference then I assume ?[04:16:55] <bstansberry> I'm checking whether the main in standalone mode understands it[04:17:12] <maxandersen> bstansberry: k[04:18:49] <bstansberry> maxandersen: no, it doesn't[04:19:08] <maxandersen> bstansberry: ok so i wont touch that then ;)[04:20:01] <maxandersen> bstansberry: is there a fast way to have the build just assemble the build/target ? (just in case you know top of your head…tired of waiting on maven building everything ;)[04:21:18] <bstansberry> you mean cd build/; mvn install is slow?[04:21:22] * bstansberry agrees[04:21:42] <bstansberry> i'll take a quick look if the "build all the modules step" can be skipped[04:22:36] <dmlloyd> baileyje: okay, I can do a release of that stuff tomorrow.[04:22:40] <maxandersen> bstansberry: hmm..might be i'm actually doiing it in "root"...[04:22:59] <bstansberry> jumping into build/ will save a lot[04:23:06] <baileyje> Great. I am just cleaning up the commit. The async stuff should be done. I will clean it up and have you take a look at it.[04:23:09] <maxandersen> yup - sorry - jetlag brain.[04:23:21] <bstansberry> cool; glad that helps[04:23:47] <bstansberry> build/ still has a fairly lengthy process, but I don't see how to make it go away[04:24:23] <maxandersen> bstansberry: so if maven would finishing downloading the internet (seems there are some new dependencies update since this morning) then ill have a patch for windows .bat files.[04:24:41] <bstansberry> nic[04:24:42] <bstansberry> e[04:24:57] <maxandersen> bstansberry: it seems to just be 2 stupid missing quotes.[04:26:00] <smcgowan> didn't like that I changed to java:/DB1: Caused by: java.lang.IllegalArgumentException: Invalid context name 'java:/DB1' for binding[04:26:19] <smcgowan> changed s/b forced[04:27:16] <stuartdouglas> It should be possible to use resource-ref entries in web.xml, and use lookup-name to point them at java:/DB1[04:29:49] *** ccrouch has quit IRC[04:31:22] *** JimMa has joined #jboss-as7[04:32:00] <smcgowan> i'll give that a try[04:34:25] <maxandersen> bstansberry: mvn complains now about not finding org.jboss.ws:jbossws-common:jar:2.0.0.Alpha4 ….known issue ?[04:34:35] * maxandersen built AS 7 fine earlier today...[04:35:06] <bstansberry> I'll try[04:35:14] <bstansberry> I've built many times today as well[04:36:10] <maxandersen> hmmm…maybe I need to force the full build…and not just cd build…?[04:36:33] <maxandersen> well i guess org.jboss.ws is not part of the as build but just a pure dependency..so that sholuldnt matter...[04:38:41] <bstansberry> maxandersen: works fine for me and the jar is at https://repository.jboss.org/nexus/content/groups/public/org/jboss/ws/jbossws-common/2.0.0.Alpha4/[04:40:49] <maxandersen> bstansberry: mkay…ill give my mvn repo and nexus a hard shake then[04:49:59] <baileyje> dmlloyd: You never pushed anything for the async correct?[04:53:41] <smcgowan> stuartdouglas: hmm, no luck with that: http://pastebin.com/Y8j41bwJ[04:55:50] <smcgowan> that's with: http://pastebin.com/H6R2SGr9[04:58:28] <maxandersen> bstansberry: following http://community.jboss.org/wiki/HackingonAS7, shouldnt git push -f my_cool_ffeature in step 12 be git push -f origin my_cool_feature ?[04:59:57] <bstansberry> maxandersen: yep; that's wrong[05:01:12] <maxandersen> bstansberry: fixed[05:01:24] <maxandersen> (once jboss.org starts waiting ;)[05:01:27] <bstansberry> thanks![05:01:28] <maxandersen> stops i mean[05:01:33] <bstansberry> hehehe[05:02:15] <maxandersen> so … I have to signup for as pullrequests mailing list to submit pullrequests ? doing a github pullrequest on master is not enough ?[05:02:19] <maxandersen> bstansberry: ^[05:02:52] <bstansberry> you can ask here for pulls or use the list[05:03:13] <bstansberry> github, maybe it gets noticed maybe not[05:04:43] <maxandersen> bstansberry: https://github.com/jbossas/jboss-as/pull/19[05:06:01] <maxandersen> bstansberry: which fixes https://issues.jboss.org/browse/JBAS-9177[05:06:02] <jbossbot> jira [JBAS-9177] Launch script does not work on Windows [Open (Unresolved) Bug, Critical, Unassigned] https://issues.jboss.org/browse/JBAS-9177[05:09:46] <jbossbot> git [jboss-as] push master 254ba95.. Max Rydahl Andersen JBAS-9177 added missing quotes where JBOSS_HOME is used[05:09:47] <jbossbot> jira [JBAS-9177] Launch script does not work on Windows [Open (Unresolved) Bug, Critical, Unassigned] https://issues.jboss.org/browse/JBAS-9177[05:09:47] <jbossbot> git [jboss-as] push master URL: http://github.com/jbossas/jboss-as/compare/a33c114...254ba95[05:10:10] <bstansberry> maxandersen: pushed[05:10:18] <maxandersen> excellent ;)[05:10:41] <maxandersen> bstansberry: now where is that empty welcome page being generated ? :)[05:11:14] <Nihility> maxandersen: sort of a inner debate on that one[05:11:34] <Nihility> we need a root.war for that[05:11:44] <Nihility> but we dont want deployments in the main boot[05:12:11] <maxandersen> Nihility: ah yes now I recall…so you needed some default root.war when no root.war otherwise present kinda thing...[05:13:13] <Nihility> need to talk it out with remy[05:13:13] <maxandersen> Nihility: wanted to go and setup that google analytics phone home feature and at least have a "Congratulations - your JBoss AS 7 is running…" text ;)[05:13:37] <maxandersen> i guess that will have to wait for next night with jetlag then ;)[05:13:38] <Nihility> that would be cool[05:14:00] <maxandersen> so if I put a root.war in the deployment now it would actually get picked up ?[05:14:10] <maxandersen> I could create the root.war then..[05:14:23] <Nihility> maxandersen: go to localhost:9990/console[05:14:27] <maxandersen> then once you and remy have fought the battle you have something...[05:15:05] <Nihility> then once you go through the splash page[05:15:09] <Nihility> login is admin admin[05:15:16] <Nihility> it doesnt do much yet[05:15:22] <Nihility> but click on the debug link[05:15:31] <Nihility> you can browse the management model there[05:16:08] <maxandersen> hmm…anything I click on says "could not reval <name>"[05:16:29] <Nihility> yeah i get the same on the stuff on the left[05:17:11] <maxandersen> Nihility: ah datasources works …but that is about it ;)[05:17:30] <maxandersen> and jms[05:17:35] <maxandersen> so thats sweet ;)[05:17:37] <Nihility> yeah he hasnt implemented any of that stuff yet i think he was just playing with the UI elements[05:17:51] <Nihility> but anyway the debug link on the right side[05:17:55] <maxandersen> Nihility: would be great to have a default fallback ;)[05:17:58] <Nihility> will dump the works[05:18:06] <maxandersen> ah sorry - missed that part[05:18:52] <Nihility> ug reminds me i need to fix that property value junk in the jsson[05:18:53] <Nihility> json[05:19:01] *** rawbdor has quit IRC[05:19:21] <maxandersen> Nihility: I was using the updated console today and noticed that deployment node was behaving weird…[05:20:26] <maxandersen> Nihility: tab completion in there basically doesn't work…as if read-operation-names etc. aren't returning the available operations.[05:20:36] <maxandersen> just wondering what so special about deployments ;)[05:20:45] <maxandersen> since the other parts seem to work.[05:21:01] <Nihility> bstansberry might know something[05:21:22] <Nihility> bstansberry: does the deployment node not have operations?[05:21:38] <Nihility> most of the deployment operations i know are global[05:21:42] <bstansberry> sure, it has ops[05:22:07] <maxandersen> i can run operations in there….just the api that the tab completion uses doesn't seem to return anything[05:22:23] <maxandersen> let me show..2 sec.[05:23:19] *** smarlow has quit IRC[05:23:19] <maxandersen> ./jboss-admin.sh[05:23:21] <maxandersen> connect[05:23:25] <maxandersen> cd deployment[05:23:33] <maxandersen> ./:<tab>[05:23:38] <maxandersen> that should return "something"[05:24:10] <bstansberry> that's because there are no deployments[05:24:12] <maxandersen> ./:read-operation-names does return things[05:24:26] <maxandersen> sure, but it has operations...[05:24:29] <bstansberry> that's because there is a root[05:24:39] <bstansberry> yeah, i understand, not saying it's good[05:25:01] *** pgier has quit IRC[05:25:29] * bstansberry is pondering a bit[05:25:30] <maxandersen> ah …because its on the nodetype and no actual node child (i.e. deployment=foo.war)[05:26:05] <maxandersen> and doing deployment:read-operation-names actually falls back to /:read-operation-names (is my guess)[05:26:16] <maxandersen> since subsystem=ee works.[05:26:52] <maxandersen> Nihility: would it make sense I created a basic root.war showing the phone home setup I mentioned ?[05:27:20] <maxandersen> Nihility: that could then also be integrated into the console …since its just a basic href link...[05:30:38] <bstansberry> Nihility: woops[05:30:39] <bstansberry> [localhost:9999 /] :read-resource-description(recursive=1,operations=1)[05:30:40] <bstansberry> {[05:30:40] <bstansberry> "outcome" => "failed",[05:30:40] <bstansberry> "failure-description" => [("java.util.MissingResourceException" => "Can't find resource for bundle java.util.PropertyResourceBundle, key shutdown")][05:30:40] <bstansberry> }[05:31:10] <bstansberry> the smoke test must not have the "operations=true"[05:31:50] *** rawbdor has joined #jboss-as7[05:32:49] <Nihility> Doh[05:32:55] <Nihility> My cuckoo[05:32:59] <Nihility> Lol[05:33:08] <Nihility> fuckup[05:33:33] <Nihility> Rocking kid to sleep (again)[05:33:35] <bstansberry> i'm fixing, along with the smoke test[05:35:54] <Nihility> Ill have my hands free in a few mins hopefully[05:38:21] *** smarlow has joined #jboss-as7[05:38:21] *** ChanServ sets mode: +v smarlow[05:42:56] *** lgao has left #jboss-as7[05:43:45] <bstansberry> Nihility: https://github.com/bstansberry/jboss-as/commit/1c7f20a74390aee1be637e6975ff3ba64d3a3d5f[05:43:46] <jbossbot> git [jboss-as] 1c7f20a.. bstansberry at jboss dot com [JBAS-9104] Add description text...[05:43:47] <jbossbot> jira [JBAS-9104] Add missing standalone shutdown operation [Resolved (Done) Task, Blocker, Jason Greene] https://issues.jboss.org/browse/JBAS-9104[05:45:59] <Nihility> Thanks[05:46:14] <Nihility> push it![05:48:06] *** mbg has joined #jboss-as7[05:48:06] *** ChanServ sets mode: +v mbg[05:48:25] <maxandersen> Nihility: so …how does one deploy a root.war to as7 ? :) root.war in deployments doesn't work...[05:48:55] <maxandersen> bstansberry: btw. wasn't the default for AS 7 to be that exploded directories aren't deployed automatically ?[05:49:28] <bstansberry> hmm, is it not?[05:49:47] <bstansberry> ah, that reminds me of a possible bug i should check out[05:50:19] <maxandersen> bstansberry: well im checking but im pretty sure im seeing foo.war.deployed for foo.war directoreis witthout doing anything...[05:50:24] <maxandersen> let me try again[05:50:24] <bstansberry> i think when it persists standalone.xml its writing the wrong value or something, so if you restart the behavior is different[05:50:51] <maxandersen> md foo.war ...[05:51:00] <maxandersen> a few seconds later… foo.war.deployed[05:52:23] <bstansberry> yep, it's a cut-and-paste bug in the xml marshalling logic[05:53:11] <bstansberry> JBAS-9070[05:53:12] <jbossbot> jira [JBAS-9070] Support for auto-deployment by the deployment scanner [Resolved (Done) Task, Major, Brian Stansberry] https://issues.jboss.org/browse/JBAS-9070[05:53:25] * bstansberry remembered the right number :)[05:53:44] <maxandersen> bstansberry: should I open separte jira or will you just go with it ? :)[05:54:01] <bstansberry> no need for a new one[05:54:21] <maxandersen> bstansberry: okey ill let you run with it then ;)[05:54:24] <stuartdouglas> smcgowan: sorry, did not see your reply, that looks like you are trying to bind something with a null name, which is odd[05:55:02] <maxandersen> bstansberry: so do you know how to deploy what corresponds to a root.war to as 7 ? i can't seem to get it to pick it up ;)[05:55:16] <smcgowan> stuartdouglas: ya, i'm not able to find a working combination[05:55:33] <bstansberry> it won't deploy?[05:55:38] <smcgowan> well, except for java:/DB1 as we discussed but nothing with jdbc/DB1[05:56:03] <Nihility> maxandersen: try adding a jboss-web.xml with a context of /[05:56:24] <bstansberry> or try ROOT.war, not root.war[05:56:42] <Nihility> yeah[05:56:50] <Nihility> their might be some kind of safegaurd[05:56:54] <Nihility> in jboss web[05:57:15] <Nihility> i really could use a beer[05:57:19] <bstansberry> i think it's a simple x.equals(ROOT.war) not equalsIgnoreCase[05:57:20] <maxandersen> ROOT.war was that magic word[05:57:36] * bstansberry seconds Nihillity's suggestion![05:57:43] <bstansberry> back in a minute :)[05:57:47] * Nihility is out :([06:00:05] <Nihility> i found a hard cider in the fridge[06:00:09] <bstansberry> ok, i won't describe the enjoyment then[06:00:26] <Nihility> not exactly what i wanted[06:00:29] <Nihility> but it will do[06:00:32] <stuartdouglas> cider < beer[06:01:07] <Nihility> stuartdouglas: does anyone in australia drink fosters[06:01:12] <stuartdouglas> nope[06:01:17] <stuartdouglas> no where even sells it[06:01:17] <Nihility> thats what i thought[06:02:08] <stuartdouglas> except for a few boutique places[06:02:13] <stuartdouglas> It's crap anyway[06:03:46] <Nihility> yeah its water[06:04:01] <Nihility> although better than many of the domestics here[06:04:11] <stuartdouglas> wow, they must be really bad[06:04:14] <Nihility> id drink it over budweiser any day[06:04:53] <Nihility> there is good beer here, microbrews usually[06:05:58] <stuartdouglas> plenty of good beers here too, but none as good as Belgian beer :-)[06:06:04] <Nihility> but yeah i always wondered the way the advertise[06:06:31] <Nihility> they act like its the only beer there[06:06:33] <stuartdouglas> If it was not for the export market they would have stopped making it long ago[06:06:34] <maxandersen> Duval….[06:06:40] <Nihility> and i was thinking it would be very sad if it was[06:06:41] <Nihility> haha[06:06:51] <stuartdouglas> Duval < Chimay[06:07:00] <Nihility> yeah i like chimay alot[06:07:05] *** mbg is now known as mbg|away[06:07:35] <Nihility> bstansberry: is probably drinking a fosters right now[06:07:36] <Nihility> hahahaha[06:08:31] <bstansberry> i'm having patriotic beer[06:08:57] <Nihility> pats blue ribbon?[06:08:58] <Nihility> haha[06:11:58] <stuartdouglas> I have been looking at trying to get the weld interceptor bindings integrated with EJB, but I have not really had any luck[06:12:19] <stuartdouglas> I think it may have to wait till next release[06:12:34] <Nihility> ok thats fine[06:13:46] <stuartdouglas> I have all the prices mostly set up, but I need to figure out which ejb the interceptor is actually intercepting[06:14:11] <stuartdouglas> and at the moment there is not really any way to do that, (at least, not without adding some massive hacks into ee)[06:15:52] <maxandersen> hey anyone know if its possible to write a relative url that will change the port ? i..e. <a href=":9000/console:>Console</a> doesn't work ..have to specify localhost but that won't be good..[06:16:03] <maxandersen> i guess javascript is the only way ;([06:16:36] <jbossbot> git [jboss-as] push master 3de27df.. bstansberry at jboss dot com [JBAS-8204] Handle domain and server group deploy removes separately...[06:16:38] <jbossbot> jira [JBAS-8204] Deployment via a deployment plan [Open (Unresolved) Task, Major, Brian Stansberry] https://issues.jboss.org/browse/JBAS-8204[06:16:38] <jbossbot> git [jboss-as] push master 81524f8.. bstansberry at jboss dot com Don't return DISABLED as status of started but non-auto-start server[06:16:38] <jbossbot> git [jboss-as] push master 99feae3.. bstansberry at jboss dot com Restore ability to dump host model in xml format[06:16:38] <jbossbot> git [jboss-as] push master e559d19.. bstansberry at jboss dot com Clean up the domain.interactive demo runner[06:16:38] <jbossbot> git [jboss-as] push master 482a1d0.. bstansberry at jboss dot com Fix the domain.configs demo[06:16:39] <jbossbot> git [jboss-as] push master 97b21df.. bstansberry at jboss dot com [JBAS-9070] Fix cut and paste bugs...[06:16:39] <jbossbot> jira [JBAS-9070] Support for auto-deployment by the deployment scanner [Resolved (Done) Task, Major, Brian Stansberry] https://issues.jboss.org/browse/JBAS-9070[06:16:40] <jbossbot> git [jboss-as] push master URL: http://github.com/jbossas/jboss-as/compare/254ba95...97b21df[06:17:30] <bstansberry> maxandersen: the auto-deploy of exploded should be fixed[06:17:39] <maxandersen> bstansberry: coolio[06:21:42] *** odin_ has quit IRC[06:21:59] <smarlow> Nihility: so, my memory leak fix, loses the benefit of ConcurrentReferenceHashMap (due to strong references keeping the lists from compacting) but otherwise, avoids leaking the entity manager which kept a bunch of the deployment in memory previously..[06:21:59] <smarlow> Nihility: I have another fix on my branch that smcgowan would like to have for getting better testing results.[06:22:15] *** jclingan1 has quit IRC[06:23:17] *** jma has joined #jboss-as7[06:24:56] <jbossbot> git [jboss-as] push master f43d71f.. Scott Marlow add/remove some comments[06:24:56] <jbossbot> git [jboss-as] push master aea1073.. Scott Marlow JBAS-9194 fix JPA memory leak[06:24:57] <jbossbot> jira [JBAS-9194] Memory leak in SFSBPCMap [Coding In Progress (Unresolved) Sub-task, Major, Scott Marlow] https://issues.jboss.org/browse/JBAS-9194[06:24:57] <jbossbot> git [jboss-as] push master 18a295e.. Scott Marlow after irc discussion with Emmanuel, I need to change non-tx em to not use xpc[06:24:57] <jbossbot> git [jboss-as] push master 1b92a98.. Scott Marlow fix for pu/pc at class level[06:24:57] <jbossbot> git [jboss-as] push master URL: http://github.com/jbossas/jboss-as/compare/97b21df...1b92a98[06:25:02] <Nihility> smarlow: there you go[06:25:16] <smarlow> Nihility: thanks :)[06:26:05] <Nihility> smarlow: don't forget to upate jira![06:27:47] <smcgowan> the JIRA i was referring to was JBCTS-1096[06:27:59] <smarlow> Nihility: good thing you said that, I'll do that now :)[06:31:19] <maxandersen> Nihility: bstansberry: btw. i'm still confused on what compatibility level the java api will have ? when we talked at devoxx I understood as the java api would have same problems as older AS's (i.e. I would need to compile/run against the matching runtime jars to get it to work) …is that no longer true ?[06:31:50] <bstansberry> that is no longer true[06:32:16] <smcgowan> nihility: the tests are running locally but won't be able to run in Hudson till I can resolve lookups to jdbc/DB1[06:33:12] *** jclingan has joined #jboss-as7[06:34:36] <bstansberry> maxandersen: the library we use for the "detyped" request and response objects will remain compatible. the wire protocol will remain compatible. the simple API to make requests will remain compatible[06:35:30] <maxandersen> and this is something more than dmr, correct ?[06:35:33] * smarlow communication breakdown[06:35:33] * smarlow The call to the JIRA server did not complete within the timeout period. We are unsure of the result of this operation.[06:36:13] <maxandersen> bstansberry: any way to know which API will remain "highly" compatible ? some specific jar or java package ?[06:36:28] <bstansberry> the simple API? it's the controller-client module in the AS source tree[06:36:48] <bstansberry> that's really the API, then it's maven deps[06:37:30] <bstansberry> right now it's depending on another AS module jboss-as-protocol for the transport stuff, but that will be replaced in a later beta with something long-term[06:39:12] <maxandersen> bstansberry: thats *great* news...[06:40:11] <maxandersen> so this api currently seem to cover "raw" domain model operations and hten some specific api for deployment… correct?[06:40:27] <maxandersen> bstansberry: will that include more before final or is that the basic goal ?[06:40:57] <bstansberry> I don't want to go much beyond that[06:41:22] <maxandersen> bstansberry: but thats great then[06:41:34] <maxandersen> bstansberry: that simplifies things a lot for us then...[06:41:44] <maxandersen> bstansberry: at least I hope so ;)[06:41:46] <bstansberry> yes, that's the idea[06:42:14] <maxandersen> bstansberry: so this would give us "easy" remote deployment I would guess ;)[06:42:15] *** lgao has joined #jboss-as7[06:42:33] <maxandersen> bstansberry: does the remote api support or plan to support incremental updates or is that out of scope ?[06:43:02] <maxandersen> rawbdor: good news for you and Andre ;) ^[06:43:25] <bstansberry> not sure what you mean by incremental updates[06:43:45] <maxandersen> just update lets say foo.war/index.html and not the whole 20 MB war...[06:45:39] *** mbg|away is now known as mbg[06:46:40] <maxandersen> bstansberry: that would reduce the need for our file and ssh support for AS7 :)[06:47:04] <maxandersen> and users could just stay on one of the deployments mechanism instead of toggling between them too much.[06:47:34] <bstansberry> oh, yeah, incremental is out of scope[06:48:39] *** kkhan has quit IRC[06:48:58] *** smarlow has quit IRC[06:49:57] <bstansberry> battery's dying, time to call it a night[06:50:00] <bstansberry> g'night all[06:50:08] *** bstansberry is now known as bstans_zzz[07:00:21] *** mbg has quit IRC[07:11:12] *** smcgowan has quit IRC[07:22:50] *** jclingan1 has joined #jboss-as7[07:25:44] *** jclingan has quit IRC[07:49:34] *** lgao has quit IRC[07:56:15] *** lgao has joined #jboss-as7[08:07:04] <jbossbot> git [jboss-dmr] push master ac4a57f.. Jason T. Greene Encode properties as objects with JSON[08:07:04] <jbossbot> git [jboss-dmr] push master URL: http://github.com/jbossas/jboss-dmr/compare/09d2ea0...ac4a57f[08:10:02] *** emuckenhuber has joined #jboss-as7[08:10:03] *** emuckenhuber has joined #jboss-as7[08:10:03] *** ChanServ sets mode: +v emuckenhuber[08:16:24] *** maxandersen is now known as max_sunrise[08:23:29] *** jfclere has joined #jboss-as7[08:23:29] *** ChanServ sets mode: +v jfclere[08:35:11] *** kkhan has joined #jboss-as7[08:35:11] *** ChanServ sets mode: +v kkhan[08:36:21] *** kkhan has quit IRC[08:36:44] *** kkhan has joined #jboss-as7[08:36:44] *** ChanServ sets mode: +v kkhan[08:45:08] *** kkhan has quit IRC[08:54:40] *** torben has joined #jboss-as7[08:54:40] *** ChanServ sets mode: +v torben[08:59:14] *** kkhan has joined #jboss-as7[08:59:15] *** ChanServ sets mode: +v kkhan[09:29:54] *** asoldano has joined #jboss-as7[09:29:54] *** ChanServ sets mode: +v asoldano[09:36:59] *** Jaikiran has joined #jboss-as7[09:36:59] *** ChanServ sets mode: +v Jaikiran[09:42:53] *** max_sunrise is now known as maxandersen[09:45:04] *** opalka has joined #jboss-as7[09:45:04] *** opalka has joined #jboss-as7[09:45:04] *** ChanServ sets mode: +v opalka[09:45:55] <opalka> morning[09:47:42] *** adietisheim has joined #jboss-as7[09:59:13] *** slaboure has joined #jboss-as7[10:06:09] *** kkhan_ has joined #jboss-as7[10:06:09] *** kkhan has quit IRC[10:06:27] *** kkhan_ has quit IRC[10:06:49] *** kkhan has joined #jboss-as7[10:06:50] *** ChanServ sets mode: +v kkhan[10:12:03] *** kkhan_ has joined #jboss-as7[10:12:03] *** kkhan has quit IRC[10:14:10] *** bgeorges has joined #jboss-as7[10:14:42] *** darranl has joined #jboss-as7[10:14:43] *** darranl has joined #jboss-as7[10:14:43] *** ChanServ sets mode: +v darranl[10:15:18] *** kkhan_ has quit IRC[10:17:37] *** kkhan has joined #jboss-as7[10:17:37] *** ChanServ sets mode: +v kkhan[10:18:05] *** odin_ has joined #jboss-as7[10:25:05] *** alesj has joined #jboss-as7[10:38:59] *** maeste has joined #jboss-as7[10:38:59] *** ChanServ sets mode: +v maeste[10:42:34] *** jma has quit IRC[10:42:34] *** JimMa has quit IRC[10:45:12] *** jcosta has joined #jboss-as7[10:45:12] *** ChanServ sets mode: +v jcosta[10:56:23] *** maxandersen1 has joined #jboss-as7[10:56:23] *** maxandersen has quit IRC[10:56:36] *** maxandersen1 has quit IRC[10:56:54] *** maxandersen has joined #jboss-as7[10:56:54] *** ChanServ sets mode: +v maxandersen[11:11:15] *** jhalliday has joined #jboss-as7[11:24:22] <opalka> kkhan, ping[11:24:32] <kkhan> opalka: hey[11:24:51] *** Jaikiran is now known as Jaikiran|AFK[11:25:20] <opalka> kkhan, I'm wondering where the following launch 'script' is created - http://fpaste.org/kkIg/[11:26:16] <kkhan> http://community.jboss.org/wiki/JBossModulesSurefirePlugin i[11:27:49] <opalka> kkhan, thx[11:42:18] *** miclorb_ has quit IRC[11:42:50] *** stalep has joined #jboss-as7[11:44:25] *** emmanuel has joined #jboss-as7[11:44:25] *** ChanServ sets mode: +v emmanuel[11:44:39] *** emmanuel has quit IRC[11:57:02] *** wolfc has joined #jboss-as7[11:57:02] *** ChanServ sets mode: +v wolfc[11:58:40] *** miclorb_ has joined #jboss-as7[12:00:50] *** bgeorges has quit IRC[12:01:36] *** rmaucher has joined #jboss-as7[12:02:57] *** miclorb_ has quit IRC[12:03:59] *** jma has joined #jboss-as7[12:04:00] *** JimMa has joined #jboss-as7[12:14:48] *** sannegrinovero has joined #jboss-as7[12:14:48] *** sannegrinovero has quit IRC[12:14:48] *** sannegrinovero has joined #jboss-as7[12:14:48] *** ChanServ sets mode: +v sannegrinovero[12:17:45] *** lgao has quit IRC[12:19:25] *** miclorb has joined #jboss-as7[12:27:17] *** jclingan1 has quit IRC[12:37:09] *** Jaikiran|AFK is now known as Jaikiran[12:45:00] *** miclorb has quit IRC[12:48:40] *** torben has quit IRC[13:35:27] *** asoldano is now known as asoldano_lunch[13:40:04] *** dimitris_ has joined #jboss-as7[13:40:04] *** dimitris_ has joined #jboss-as7[13:40:04] *** ChanServ sets mode: +v dimitris_[13:44:01] *** torben has joined #jboss-as7[13:44:02] *** torben has joined #jboss-as7[13:44:02] *** ChanServ sets mode: +v torben[13:44:48] *** pil-afk is now known as pilhuhn[14:04:26] *** vtunka has joined #jboss-as7[14:04:26] *** ChanServ sets mode: +v vtunka[14:05:55] *** smarlow has joined #jboss-as7[14:05:55] *** ChanServ sets mode: +v smarlow[14:09:02] *** torben has quit IRC[14:23:24] *** slaboure has quit IRC[14:25:40] *** jma has quit IRC[14:25:40] *** JimMa has quit IRC[14:26:39] *** torben has joined #jboss-as7[14:26:39] *** ChanServ sets mode: +v torben[14:27:55] *** torben has quit IRC[14:34:37] *** jpederse has joined #jboss-as7[14:34:42] *** torben has joined #jboss-as7[14:34:42] *** ChanServ sets mode: +v torben[14:34:47] *** jpederse has quit IRC[14:34:47] *** jpederse has joined #jboss-as7[14:36:22] <Jaikiran> wolfc: dmlloyd: Nihility: http://community.jboss.org/wiki/EJB3supportinJBossAS700Beta2[14:36:54] <Jaikiran> i'll be updating that with more info later today (as and when i have some more tests to verify what's working)[14:37:41] <wolfc> @Remote is working except for the remoty bits :-)[14:38:11] <wolfc> Jaikiran, I think it's best to say: @Remote now functions as @Local[14:38:17] <Jaikiran> wolfc: yeah, let me add that bit to be more clear[14:38:22] <Jaikiran> both @Remote and business-remote[14:38:41] <Jaikiran> anything else that i'm missing?[14:39:12] <wolfc> EJBTHREE-2120[14:39:13] <jbossbot> jira [EJBTHREE-2120] Tx logic doesn't take method-intf type into acount [Open (Unresolved) Bug, Major, Carlo de Wolf] https://issues.jboss.org/browse/EJBTHREE-2120[14:39:29] * Jaikiran makes a note[14:39:56] <wolfc> I think everywhere Method-Intf can be used doesn't function yet[14:40:22] <wolfc> What happened to MDB parsing?[14:40:43] <Jaikiran> it's there. i haven't added it to that wiki, since i haven't tested it yet[14:40:49] <Jaikiran> will be doing it later today[14:41:05] <Jaikiran> basically, it has the same level of support as @MessageDriven[14:41:15] *** slaboure has joined #jboss-as7[14:42:29] *** asoldano_lunch is now known as asoldano[14:42:50] *** torben has quit IRC[14:42:53] <wolfc> nice[14:43:23] *** torben has joined #jboss-as7[14:43:23] *** ChanServ sets mode: +v torben[14:45:00] *** torben has quit IRC[14:50:02] *** mmoyses has joined #jboss-as7[14:50:02] *** ChanServ sets mode: +v mmoyses[14:50:14] *** emmanuel has joined #jboss-as7[14:50:14] *** ChanServ sets mode: +v emmanuel[14:54:50] *** smcgowan has joined #jboss-as7[14:54:50] *** ChanServ sets mode: +v smcgowan[14:57:09] *** balunasj has joined #jboss-as7[14:57:09] *** ChanServ sets mode: +v balunasj[14:59:36] *** frainone has joined #jboss-as7[14:59:36] *** ChanServ sets mode: +v frainone[15:03:48] *** ccrouch has joined #jboss-as7[15:03:50] *** ccrouch has joined #jboss-as7[15:03:50] *** ChanServ sets mode: +v ccrouch[15:04:52] *** pgier has joined #jboss-as7[15:04:52] *** ChanServ sets mode: +v pgier[15:06:28] *** vtunka has quit IRC[15:06:38] *** jfd has joined #jboss-as7[15:06:38] *** jfd has quit IRC[15:06:38] *** jfd has joined #jboss-as7[15:06:38] *** ChanServ sets mode: +v jfd[15:08:40] *** bstans_zzz is now known as bstansberry[15:08:56] <bstansberry> emuckenhuber: how are things?[15:09:42] <emuckenhuber> bstansberry: good[15:09:51] <bstansberry> moved?[15:10:02] <emuckenhuber> yeah, loads of boxes to unpack[15:10:12] <bstansberry> :)[15:10:16] <emuckenhuber> and they just told me that i won't have internet at home for the next two weeks :(([15:10:41] <bstansberry> well, you can sleep at the office ;)[15:11:38] <emuckenhuber> hehe, actually the place is close to the office so i hope it won't be necessary[15:11:48] <bstansberry> I'm going through domain demos, fixing stuff[15:14:15] <emuckenhuber> ok - are there a lot of issues with the demos?[15:15:05] <bstansberry> hopefully not :)[15:15:23] <bstansberry> domain.interactive has a problem with the JMS part[15:16:18] <bstansberry> domain.ds has what looks like a JBAS-9010 issue[15:16:20] <jbossbot> jira [JBAS-9010] Provide more comprehensive error messages for missing dependencies [Open (Unresolved) Task, Blocker, David Lloyd] https://issues.jboss.org/browse/JBAS-9010[15:16:49] <bstansberry> I doubt that's domain management related; i'll report that to someone else in a bit[15:17:02] <bstansberry> domain.configs and description are ok[15:17:07] *** bgeorges has joined #jboss-as7[15:17:07] <bstansberry> i'm fixing domain.host[15:17:21] * smarlow finished shoveling[15:17:26] <bstansberry> the others, I don't know[15:17:29] <bstansberry> smarlow: bummer[15:17:52] <smarlow> bstansberry: at least, I had one day to grill a few weeks ago :)[15:18:06] <bstansberry> I like shoveling in Dec/Jan/Feb. April, not so much[15:18:41] <bstansberry> keep that grill memory alive![15:19:23] <smarlow> haha, happy thoughts![15:20:23] *** vtunka has joined #jboss-as7[15:20:23] *** ChanServ sets mode: +v vtunka[15:24:12] *** fnasser has joined #jboss-as7[15:24:41] <bstansberry> emuckenhuber: can you have a look at the JMS part of domain.interactive?[15:25:10] <darranl> what are you guys shoveling? snow?[15:25:23] <smcgowan> smarlow: i'm getting great results on the persistence TCK tests, only issue found so far regarding inclusion of orm.xml in PU: JBCTS-1097[15:25:30] <bstansberry> darranl: yes[15:25:39] <bstansberry> well, smarlow is[15:25:52] <smarlow> smcgowan: that is great news![15:25:58] <smcgowan> smarlow: i'll try to get the jdbc look up stuff figured out and see if we can run these in Hudson[15:25:59] <darranl> I am just thinking if the grass will be dry enough to cut tonight ;-)[15:26:08] <bstansberry> me too[15:26:09] <dmlloyd> good morning[15:26:16] <bstansberry> morning[15:26:27] <smarlow> smcgowan: I'll take a look at JBCTS-1097[15:26:49] <smcgowan> smarlow: i've attached a .war to the JIRA[15:27:01] <smarlow> smcgowan: thanks![15:27:50] <emuckenhuber> bstansberry: well i was just running the domain.messaging demo and it works fine for me - that the demo you mentioned?[15:28:13] <bstansberry> no, domain.interactive, item 11 on the menu creates a queue[15:28:17] <emuckenhuber> ah ok[15:30:56] <wolfc> dmlloyd, Jaikiran, here is a nice scenario I got in a support case: https://github.com/wolfc/jboss-as/commit/9150e4abe5c05a8a17f82afe8ac61abe0acaf857#L3R30[15:30:57] <jbossbot> git [jboss-as] 9150e4a.. Carlo de Wolf Test exception throwing from an interceptor[15:30:59] <smarlow> smcgowan: it reproduced locally, thanks for the test jar. "org.xml.sax.EntityResolver from [Module "org.dom4j:main""[15:31:02] <wolfc> something to keep in mind[15:31:50] <smarlow> smcgowan: ncdfe ^ as you know.[15:32:12] <smcgowan> smarlow: ya[15:34:25] <wolfc> most preferably I would like to punish those people right away after coming out the interceptor, but you could argue that only top side the exception must never show.[15:35:12] *** starrgrass has joined #jboss-as7[15:35:39] *** aloubyansky has joined #jboss-as7[15:38:45] <wolfc> smarlow, ping[15:38:57] <smarlow> wolfc: pong[15:39:07] <wolfc> did you rollback to version 2 of http://community.jboss.org/wiki/EJB3supportinJBossAS700Beta2/version ?[15:39:37] <smarlow> yeah, I saved a minor revision (I fixed a typo and it trashed the document)[15:40:29] <smarlow> wolfc: so I rolled back my change[15:41:03] <wolfc> Okay, the version history doesn't show restores. Bummer.[15:41:56] <jbossbot> git [jboss-as] push master 7c2ead1.. bstansberry at jboss dot com [JBAS-9216] Remove incorrect cast[15:41:57] <jbossbot> jira [JBAS-9216] Invalid cast in SpecifiedInterfaceRemoveHandler [Open (Unresolved) Bug, Major, Brian Stansberry] https://issues.jboss.org/browse/JBAS-9216[15:41:58] <jbossbot> git [jboss-as] push master URL: http://github.com/jbossas/jboss-as/compare/1b92a98...7c2ead1[15:44:10] <opalka> wolfc, What are the plans for ws-ejb3 branch WRT AS7 Beta 2 ?[15:45:07] *** bstansberry is now known as bstans_afk[15:45:45] <wolfc> dmlloyd: ^ ? opalka, AFAIK there is nothing blocking for merge. It is junkware, so it is going to be tossed during the next refactoring round, but the test in there is valid.[15:47:35] <wolfc> latest rebase is in https://github.com/wolfc/jboss-as/commits/ejb3-ws[15:49:14] <dmlloyd> well, you know, someone has to say "please merge this" :)[15:49:24] <dmlloyd> my ESP has been weak of late[15:49:39] * dmlloyd has a look[15:53:04] <wolfc> we've discussed the branch a couple of times, but we never got to an 'yeah let's merge this'. :-)[15:53:30] * opalka nods[15:53:56] <wolfc> probably because the code I wrote isn't that cheerful :-P[15:54:38] <dmlloyd> thought I merged this stuff yesterday[15:55:05] <dmlloyd> guess not[15:56:16] <wolfc> hehe I had something similar yesterday until I saw 3 svn-commit tmp files, got distracted before svn acknowledged[15:56:31] *** vtunka has quit IRC[15:56:48] <opalka> dmlloyd, please merge, will fix many tests for us (we'll cleanup the code later)[15:57:20] <dmlloyd> it's testing now, opalka[15:57:40] <wolfc> Jaikiran also has some fixes. When do we close shop?[15:58:06] <Nihility> morning folks[15:58:11] <opalka> morning[15:58:14] <Nihility> just got back dropping off tax forms[15:58:21] <dmlloyd> there's 5 issues left[15:58:29] <Nihility> yep and one of them is mind[15:58:34] <dmlloyd> baileyje should have something on JBAS-8978[15:58:36] <jbossbot> jira [JBAS-8978] Complete async EJB invocation [Open (Unresolved) Task, Major, John Bailey] https://issues.jboss.org/browse/JBAS-8978[15:58:40] <Nihility> bstans_afk: hey you going to do that dep plan issue[15:58:43] <baileyje> dmlloyd: Yeah.[15:58:51] <dmlloyd> not sure what ALR is gonna do with JBAS-8946, I think it's probably gonna slide[15:58:52] <jbossbot> jira [JBAS-8946] Upgrade to latest Arquillian release [Open (Unresolved) Task, Major, Andrew Rubinger] https://issues.jboss.org/browse/JBAS-8946[15:59:13] <dmlloyd> that leaves JBAS-8959 and JBAS-8204[15:59:14] <jbossbot> jira [JBAS-8959] JPA 2.0 implementation [Open (Unresolved) Task, Major, Scott Marlow] https://issues.jboss.org/browse/JBAS-8959[15:59:15] <jbossbot> jira [JBAS-8204] Deployment via a deployment plan [Open (Unresolved) Task, Major, Brian Stansberry] https://issues.jboss.org/browse/JBAS-8204[15:59:18] <wolfc> last I heard he was trying to get it in[16:00:04] <wolfc> 8959 is a big one with subtasks already moved to CR1[16:00:05] <baileyje> I need two things for it. The jboss-invocation release and I have to track down exactly what needs to be carried over into the async thread. I know security and txn for sur[16:00:17] <wolfc> baileyje, not txn[16:00:41] <dmlloyd> not txn for void async methods perhaps, but I thought future ones were supposed to propagate txn context[16:00:49] <wolfc> no, no[16:00:59] <dmlloyd> it's been a while since I read that text though[16:01:20] <dmlloyd> ah guess not[16:01:34] <wolfc> you get a very nice situation if you try to run a tx on two threads with different outcomes[16:01:37] <baileyje> wolfc: Ok. So what needs to be carried along?[16:01:38] <dmlloyd> well in any case the async interceptor should probably go last[16:02:08] <dmlloyd> the txn interceptor would presumably know the disposition of txn for that method and not attach a context[16:02:09] *** starrgrass has quit IRC[16:02:26] <dmlloyd> anyway going last in the client chain would let all previous interceptors attach their contexts to the invocation[16:02:37] <dmlloyd> where the "server-side" chains can find them[16:02:49] <baileyje> that was my other question. Should the interceptors be called async or just the actual component invocation?[16:03:07] *** mbg has joined #jboss-as7[16:03:07] *** ChanServ sets mode: +v mbg[16:03:10] <dmlloyd> "Called" == "named" or "invoked"?[16:03:23] <baileyje> so first or last from a component interceptor.[16:03:28] <smarlow> I moved JBAS-8959 to beta3[16:03:28] <baileyje> invoked.[16:03:29] <jbossbot> jira [JBAS-8959] JPA 2.0 implementation [Open (Unresolved) Task, Major, Scott Marlow] https://issues.jboss.org/browse/JBAS-8959[16:03:31] *** jma has joined #jboss-as7[16:03:42] *** JimMa has joined #jboss-as7[16:03:58] <dmlloyd> the interceptors take care of calling the component asynchronously[16:03:58] <wolfc> the async request should call without a tx context. the tx interceptor will take care of setting one up (or not).[16:04:32] <dmlloyd> the async interceptor is last on the client chain. The component chain needs no specific knowledge of it.[16:04:44] *** opalka has quit IRC[16:05:15] <wolfc> in that case you need to split between sec and tx client side[16:05:47] <dmlloyd> nah, the client txn interceptor should be aware that there is no txn propagation for that method already[16:06:01] <dmlloyd> just putting the async interceptor last makes things a lot simpler[16:06:11] <dmlloyd> well, a little simpler[16:06:42] <wolfc> the view can declare @TX(required)[16:07:05] <dmlloyd> that would be an error, would it not?[16:07:12] <wolfc> no[16:07:20] <dmlloyd> hm, that's annoying[16:07:30] <jhalliday> dmlloyd, Nihility: can one of you spare a few minutes to discuss XTS packaging and deployment?[16:07:38] <dmlloyd> what's XTS?[16:07:50] <jhalliday> the web services transactions component of JBossTS[16:07:54] <wolfc> 4.5.3: This means, for example, that the semantics of the REQUIRED transaction attribute on an asynchronous method are exactly the same as REQUIRES_NEW.[16:07:59] <dmlloyd> ah[16:08:23] <dmlloyd> wolfc: okay but that can still be done by the client tx interceptor right?[16:08:35] <dmlloyd> well, I guess that's just an impl detail[16:08:47] <dmlloyd> okay, you win, sec then async then txn[16:08:51] *** vtunka has joined #jboss-as7[16:08:55] *** ChanServ sets mode: +v vtunka[16:09:00] <dmlloyd> actually either way txn will get run[16:09:05] <baileyje> I think I need coffee. dmlloyd where is the client interceptor actually being created/uesed?[16:09:13] <dmlloyd> so the client txn int has to be aware that the method is async no matter what[16:09:17] <wolfc> async will be in a new thread[16:09:21] <dmlloyd> baileyje: I think it's created in AbstractComponent[16:09:25] <wolfc> so the client txn won't find a running tx[16:09:47] <dmlloyd> the client txn shouldn't look for a running tx if the method is async[16:09:56] <baileyje> dmlloyd: That is what I am struggling with. I see "createClient" but it is not called anywhere[16:10:00] <dmlloyd> whether it's run before or after[16:10:19] <dmlloyd> baileyje: okay, look for whatever craetes the managed references[16:10:25] <wolfc> in that case async client should skip the rest of the client chain[16:10:25] <Nihility> i have a call[16:10:31] <Nihility> bbiab[16:10:54] <dmlloyd> wolfc, why? if the client txn int knows that method is async then it can just act accordingly[16:11:16] <baileyje> dmlloyd: That would be the views[16:11:21] <wolfc> doable, but why should it bother[16:11:51] *** bstans_afk is now known as bstansberry[16:12:51] <wolfc> baileyje, look for createClientInterceptor(Class<?>) on the EJB components[16:13:52] *** pferraro has joined #jboss-as7[16:13:52] *** ChanServ sets mode: +v pferraro[16:14:38] <baileyje> wolfc: Ok. So if it is asnc, we return the async interceptor[16:14:53] <emuckenhuber> bstansberry: this would fix the jms part of the interactive demo: https://github.com/emuckenhuber/jboss-as/commit/6a615c836cd8f0941a6a660d86a75a3fbcbb45d1[16:14:54] <jbossbot> git [jboss-as] 6a615c8.. Emanuel Muckenhuber fix querying for the socket-binding info[16:15:21] <wolfc> baileyje, the entire view needs not be async[16:15:41] <bstansberry> great. I have a couple fixes you can review too -- 1 sec[16:16:06] <baileyje> wolfc: ok.. in actual terms what does that mean. I am not awake yet[16:16:55] *** asaldhan has joined #jboss-as7[16:16:55] *** ChanServ sets mode: +v asaldhan[16:17:06] <dmlloyd> baileyje, the client interceptor here is a generalized entry point for the component[16:17:22] <dmlloyd> this code is, as always, split into three different duplicate entry poitns[16:17:22] <wolfc> in fact, async is an attribute of a component method, not a view method[16:17:22] <baileyje> I would assume returning an asyn interceptor from the createClientInterceptor would make it async since the proxy uses it as its first interceptor[16:17:47] <dmlloyd> wolfc: it does however have to operate in the view, since it's on the client side, regardless of where it is specified.[16:18:09] <wolfc> true. so some methods (or all) of the view are async[16:18:11] <dmlloyd> baileyje: yes and no. The async client interceptor has to run after the others though. It should just be one in a chain[16:18:42] <dmlloyd> I don't think we currently have just one single entry point.[16:19:04] <baileyje> that is what confuses me. I hear that, and then wolfc says the whole thing has to be.[16:20:06] *** mbg is now known as mbg|away[16:20:07] <wolfc> has to be waht?[16:20:11] <bstansberry> emuckenhuber: last 2 commits on https://github.com/bstansberry/jboss-as/commits/demos -- if those look ok to you, i'll push them[16:20:16] <baileyje> ASYNC[16:20:37] <dmlloyd> the method does, not the view, but we presently only use one interceptor as an entry point for all methods[16:20:56] *** mbg|away is now known as mbg[16:20:58] <dmlloyd> in fact I don't think we even *have* per-method interceptors on the client side outside of the new ComponentEntry interface which nobody is using[16:21:11] <baileyje> dmlloyd: Right. We are creating the client interceptor from a request from a view.[16:21:19] <emuckenhuber> bstansberry: looks good[16:21:34] <wolfc> but that is something we need to decide and refactor upon. We got too many different interceptor chains.[16:22:00] <dmlloyd> I"d say we don't have enough![16:22:12] <dmlloyd> I went through all of our interceptors the other day[16:22:14] <baileyje> ok. Right now I care about beta2. We can refactor after monday's meeting.[16:22:24] <bstansberry> emuckenhuber: so, sounds like domain.rar, domain.servers, domain.subsystem are left. may all work :) please pick one to try![16:22:33] <dmlloyd> I identified possibly five different interceptor types[16:23:00] <bstansberry> emuckenhuber: hmm, let me push this stuff first, in case it fixes anything ;)[16:23:11] <dmlloyd> client per-method, view per-method, view per-method+per-instance, component per-method, component instance per-method+per-instance[16:23:13] <emuckenhuber> k[16:24:21] *** jamezp has joined #jboss-as7[16:26:01] <dmlloyd> anyway none of that helps.[16:26:48] <baileyje> What I have now is the async interceptor is the very first component interceptor for any method chain that is async.[16:27:02] <baileyje> It sounds like I need to move it back a step into the client interceptor[16:27:16] <baileyje> which has no concept of method.[16:27:43] <jbossbot> git [jboss-as] push master 921c6db.. bstansberry at jboss dot com [JBAS-8204] Fixes to rollback plan creation[16:27:45] <jbossbot> jira [JBAS-8204] Deployment via a deployment plan [Open (Unresolved) Task, Major, Brian Stansberry] https://issues.jboss.org/browse/JBAS-8204[16:27:45] <jbossbot> git [jboss-as] push master ead3272.. bstansberry at jboss dot com Improve the domain.host demo a bit[16:27:45] <jbossbot> git [jboss-as] push master c1719e8.. Emanuel Muckenhuber fix querying for the socket-binding info[16:27:45] <jbossbot> git [jboss-as] push master URL: http://github.com/jbossas/jboss-as/compare/7c2ead1...c1719e8[16:28:19] <wolfc> the client interceptor knows the method that is invoked[16:28:39] <baileyje> Right the interceptor does when it is being invoked, but not when it is being setup.[16:28:48] <baileyje> I can deal with that.[16:28:53] <bstansberry> emuckenhuber: pushed. i'm going to see what's up with domain.ds. can you test the others? if any are broken, please pick one an d let me know which[16:30:00] *** maxandersen is now known as max_away[16:30:06] <dmlloyd> baileyje: I guess the issue is right now we aren't even creating proper client interceptors at all.[16:30:21] <wolfc> depends on what the definition is :-)[16:31:11] <baileyje> They are being created, they just don't do a lot more than attache the component to the private data[16:31:56] <dmlloyd> yeah which is something that should be done by the view interceptor, not the client[16:32:08] <dmlloyd> the client interceptor chain is a completely different animal - it has to be serializable for one[16:32:40] <dmlloyd> it seems we don't even have sec, txn client interceptors[16:32:53] <wolfc> we don't have remote, so there is no need yet[16:33:55] <dmlloyd> don't we need those even for local?[16:34:02] <dmlloyd> esp. once async comes into play[16:34:13] <wolfc> no, not really[16:34:15] <dmlloyd> the only diff between local & remote, in vm anyway, is pass-by-value afaict[16:34:35] <wolfc> yes[16:34:40] <dmlloyd> you can still be calling into a different invocation context completely, and with async, a different thread even[16:35:24] <wolfc> as long as an in-vm remote call stays on one thread no propagation is needed[16:36:01] <wolfc> async jumps threads which break security propagation[16:36:04] <dmlloyd> so as an optimization we could prune the txn/sec client and component interceptors from non-async method invocations if we wanted to[16:36:05] <dmlloyd> ?[16:36:38] <wolfc> just the txn/sec client/server interceptors, not the component interceptors.[16:37:05] <dmlloyd> right now the component interceptors are taking the context from the current thread context right?[16:37:14] <dmlloyd> not from the invocation private data?[16:37:18] <wolfc> yes[16:37:25] <wolfc> that's how a TM works[16:38:48] <dmlloyd> so really we'll have three txn interceptors when @Remote is done - one to get the current context and attach it to the invocation, one to get the context attached to the invocation and make it "current" in the thread, and the txn interceptor which actually does something with it[16:39:03] <wolfc> yes[16:39:17] <dmlloyd> first would be client, second would be early on the component, third would be late on the component[16:39:43] <wolfc> I call the second one, server side interceptor[16:43:08] <emuckenhuber> bstansberry: domain.rar works - the other 2 seem to have some issues[16:43:29] *** asoldano is now known as asoldano_away[16:43:33] <bstansberry> ok, take your pick :-)[16:45:22] <emuckenhuber> bstansberry: ok, starting to look at domain.subsystem - it might have affected the other one[16:46:12] * wolfc is afk for bit[16:48:41] <bstansberry> baileyje: the ds-mbean.sar used in the ds and domain.ds demos is not deploying correctly with a JBAS-9010-ish issue[16:48:43] <jbossbot> jira [JBAS-9010] Provide more comprehensive error messages for missing dependencies [Open (Unresolved) Task, Blocker, David Lloyd] https://issues.jboss.org/browse/JBAS-9010[16:48:56] <bstansberry> if I restart the server so it tries to deploy at boot, I get this:[16:48:57] <bstansberry> 09:46:34,079 ERROR [org.jboss.as] (MSC service thread 1-4) JBoss AS 7.0.0.Beta2 "(TBD)" started (with errors) in 3116ms - Started 99 of 126 services (1 services failed or missing dependencies, 26 services are passive or on-demand)[16:48:58] <bstansberry> Services missing dependencies:[16:48:58] <bstansberry> service jboss.deployment.unit."ds-mbean.sar".POST_MODULE[16:49:19] <baileyje> bstansberry: hmmmm[16:49:39] <bstansberry> ring any bells? the sar does nothing fancy, just a jboss-service.xml that names the class[16:49:48] *** aloubyansky has quit IRC[16:50:10] <bstansberry> ah, I lied, MANIFEST.MF has some stuff[16:50:10] <baileyje> I can't remember anything..[16:50:32] <bstansberry> ok, i'll check the MANIFEST stuff; probably something wrong there[16:50:57] <baileyje> I guess I don't often redeploy the demo SARs[16:51:35] <bstansberry> I bet there's an equivalent smoke test and the sar's just got out of sync![16:51:41] * bstansberry hopes for easy fix ;)[16:55:29] <emuckenhuber> bstansberry: easy one: https://github.com/emuckenhuber/jboss-as/commit/b392eca4387bf48b6279fde51596a2756330acfb[16:55:30] <jbossbot> git [jboss-as] b392eca.. Emanuel Muckenhuber use correct web connector protocol[16:56:06] <emuckenhuber> and the domain.server demo actually works[16:56:18] <bstansberry> :-)[16:58:12] <rmaucher> nice :)[16:58:49] <jbossbot> git [jboss-as] push master b392eca.. Emanuel Muckenhuber use correct web connector protocol[16:58:49] *** jpederse has quit IRC[16:58:49] <jbossbot> git [jboss-as] push master URL: http://github.com/jbossas/jboss-as/compare/c1719e8...b392eca[16:59:31] *** emmanuel has quit IRC[17:01:21] *** max_away has quit IRC[17:04:58] *** balunasj has quit IRC[17:06:29] *** jpederse has joined #jboss-as7[17:06:36] <bstansberry> baileyje: https://github.com/bstansberry/jboss-as/commit/daa9dea9389f2acef1804461a3534ef4e1c8fb51 fixes the ds demos. And the need for the fix is a nice sign of progress![17:06:37] <jbossbot> git [jboss-as] daa9dea.. bstansberry at jboss dot com Fix the ds demo sar's dependencies[17:06:49] *** jpederse has quit IRC[17:06:49] *** jpederse has joined #jboss-as7[17:11:56] *** mbg is now known as mbg|away[17:13:51] <baileyje> bstansberry: For sure it is[17:14:02] <bstansberry> i'll push that[17:14:31] <jbossbot> git [jboss-as] push master daa9dea.. bstansberry at jboss dot com Fix the ds demo sar's dependencies[17:14:31] <jbossbot> git [jboss-as] push master URL: http://github.com/jbossas/jboss-as/compare/b392eca...daa9dea[17:14:38] <smarlow> smcgowan: I have a possible orm.xml fix https://github.com/scottmarlow/jboss-as/commit/c05293acc5b68b87175cc4f0b3abc3400275474d[17:14:39] <jbossbot> git [jboss-as] c05293a.. Scott Marlow JBCTS-1097[17:14:53] *** emuckenhuber has quit IRC[17:18:30] *** mbg|away is now known as mbg[17:18:40] *** jclingan has joined #jboss-as7[17:21:49] *** alesj has left #jboss-as7[17:21:53] *** alesj has quit IRC[17:26:33] <rmaucher> still tagging later today ?[17:27:44] <rmaucher> the TCK does not appear to run right now (not so good), but the checkout that is used seems a bit old[17:28:16] *** jamezp has quit IRC[17:30:54] <dmlloyd> anyone: http://github.com/dmlloyd/jboss-invocation/commit/8125ab9[17:30:55] <jbossbot> git [jboss-invocation] 8125ab9.. David M. Lloyd Code cleanup - SVUIDs, unused imports, unused methods, unused classes, do not require non-static for method interceptor[17:31:17] <dmlloyd> smcgowan: see rmaucher's comment[17:33:11] *** mmoyses is now known as mmoyses_[17:33:14] <baileyje> dmlloyd: I will review that if it gets the release faster?[17:33:17] <baileyje> :)[17:33:24] <dmlloyd> :D[17:33:26] <Nihility> i can review stuff[17:33:35] <Nihility> just got off a LONNNG call[17:33:50] <dmlloyd> I like to release only once we know everything is working[17:33:56] <dmlloyd> granted -invocation is comparatively stable[17:34:11] <baileyje> dmlloyd: It looks good to me.[17:34:31] <jbossbot> git [jboss-invocation] push master 8125ab9.. David M. Lloyd Code cleanup - SVUIDs, unused imports, unused methods, unused classes, do not require non-static for method interceptor[17:34:31] <jbossbot> git [jboss-invocation] push master URL: http://github.com/jbossas/jboss-invocation/compare/00ebfd5...8125ab9[17:35:15] <smcgowan> dmlloyd, rmaucher: the tck is not running due to an issue with svn co[17:35:42] <rmaucher> ok sandy[17:35:51] <dmlloyd> sandy?[17:35:59] <rmaucher> shelly lol[17:36:09] <dmlloyd> ah :)[17:36:20] <rmaucher> :D[17:36:28] <Nihility> ROFL[17:36:42] <smcgowan> thats' kinda of funny, many people cal me Sandy because maiden name is Sanborn[17:36:50] <Nihility> i almost called her sally on the phone call, although that was tounge tied[17:36:51] <smcgowan> well, those that knew me then[17:37:19] <rmaucher> well, ok, you guys have the logs to make fun of me now ;)[17:37:30] <bstansberry> that's why we turned on logging![17:37:48] <smcgowan> there's an eng ticket filed to get the issue resolved[17:37:53] <smcgowan> eng-ops[17:38:10] <bstansberry> you filed a ticket to get rmaucher to use your correct name[17:38:12] <bstansberry> ?[17:38:15] <bstansberry> seems excessive[17:38:17] *** maeste has quit IRC[17:39:22] <rmaucher> meh ;)[17:39:29] <smcgowan> one issue i'm seeing with the test runs is OOM java heap[17:39:43] <dmlloyd> yeah you have to beef up maven these days[17:40:10] <rmaucher> so we would have an undeploy leak that was not there before ?[17:40:12] <smcgowan> the TCK test runs, that is - no maven[17:40:16] *** jclingan1 has joined #jboss-as7[17:40:17] <smarlow> smcgowan: should we try and get the orm.xml fix into beta2? Its on my branch currently[17:40:20] *** asoldano_away is now known as asoldano[17:40:26] <dmlloyd> oh I see[17:40:51] <dmlloyd> rmaucher, that is quite possible. I think deployment is going to remain a bit shaky until closer to Beta3[17:41:28] <smcgowan> i don't want to cause alarm but it is something i've just started seeing[17:41:38] <smcgowan> need to investigate more[17:42:01] <rmaucher> I look a bit at memory use, and this is going up from where it was two week ago (about +10%), probably due to the introduction of more subsystems[17:42:36] <dmlloyd> yeah and we haven't exactly been proactive at trimming mem usage lately[17:42:41] *** jclingan has quit IRC[17:42:52] <smcgowan> rmaucher: see the JSF runs, on mar 24 able to run the entire suite, yesterday OOM[17:43:07] <baileyje> dmlloyd: Do you want to pre-review some of this async stuff?[17:43:16] *** jamezp has joined #jboss-as7[17:43:38] <dmlloyd> baileyje: sure[17:43:42] <rmaucher> mem use is still not too bad, but this is for their cloud stuff where the reqs are really tight[17:44:09] <smcgowan> smarlow: your fix looks trivial[17:44:14] <baileyje> dmlloyd: There were a few changes on your branch I based this off of. They don't seem to have been merged.[17:44:15] <Nihility> yeah we could disable subsystems as a workaround for the cloud use[17:44:27] <Nihility> they only want a small number of things[17:44:28] <dmlloyd> baileyje: yeah I didn't want to merge in unused code.[17:44:37] <smarlow> smcgowan: well yeah, its best when they are :)[17:44:54] <Nihility> it does suck having a leak though[17:45:12] <Nihility> smcgowan: do we do a heap dump on OOM[17:45:13] <Nihility> ?[17:45:16] <baileyje> dmlloyd:[17:45:20] <baileyje> https://github.com/baileyje/jboss-as/compare/daa9dea9389f2acef180...df8fe74f1d6d96109fad[17:45:39] <smcgowan> nihility: let me see if i captured one[17:46:29] <baileyje> dmlloyd: If it is going to be awhile, I may move all the async stuff back into the EJB subsystem.[17:46:38] <smarlow> too bad we cannot generate java heap dumps by default but that could be a security issue (leaving copies of Java memory on disk).[17:47:16] <Nihility> smcgowan: i wonder if this jpa[17:47:31] <baileyje> dmlloyd: Crap. I have more cleanup to do.[17:47:33] <Nihility> smarlow added XPC support recently[17:47:42] <Nihility> smarlow: and then he worked around an ejb issue[17:47:51] <Nihility> smcgowan: which i merged last night[17:47:58] <Nihility> bad tab complete[17:48:37] <smcgowan> nihility: i did know about that, the one I saw locally was with JPA but did not have scott's fix in my loal[17:49:10] *** dimitris_ has quit IRC[17:50:26] <smcgowan> but, yes, we used to take the heap dump on OOM in TCK - let me confirm we still do it with the AS 7 set up[17:50:47] <smarlow> Nihility: I have one more fix for JBCTS-1097 (a orm.xml bug). https://github.com/scottmarlow/jboss-as/commit/c05293acc5b68b87175cc4f0b3abc3400275474d[17:50:49] <jbossbot> git [jboss-as] c05293a.. Scott Marlow JBCTS-1097[17:51:04] <smarlow> but that wouldn't help with the OOM :)[17:52:12] <smarlow> I did notice some classloaders were sticking around, when I looked at memory the other day.[17:52:16] <smcgowan> jpederse: what is jboss.jdbc.spy ?[17:52:26] <dmlloyd> that'd probably affect permgen more than heap.[17:52:33] <jpederse> smcgowan: logging of SQL statements[17:52:39] <dmlloyd> anyway we expect to have many small class loaders.[17:53:46] <smcgowan> jpederse: https://issues.jboss.org/browse/JBCTS-1098[17:54:52] <dmlloyd> those should definitely be TRACE[17:55:31] *** rawbdor has quit IRC[17:55:42] <jpederse> smcgowan: spy="false" or disable the logging category if you don't want it[17:56:07] <jpederse> dmlloyd: diabled by default - and documented - specialized logging cateogy[17:56:23] <smcgowan> well just trying to figure out if the CNFE is significant - only a WARN[17:56:34] <dmlloyd> still should be TRACE[17:56:46] <jpederse> smcgowan: I would ask Steve[17:59:57] <smarlow> jpederse: why Steve? is that more of a classloading question? Why can't we load org.apache.derby.tools.sysinfo?[18:00:51] *** jwulf has quit IRC[18:01:03] <jpederse> smarlow: well, what is the code in question ?[18:05:18] <smcgowan> http://pastebin.com/B4fYy6vx[18:05:24] <smarlow> jpederse: your right, looks like a TCCL[18:05:40] <jpederse> smarlow: well, there you go :)[18:05:52] <Nihility> whats loading sysinfo[18:05:56] <Nihility> a test debloyment[18:05:56] <Nihility> ?[18:06:03] <smarlow> code is here: http://pastie.org/1744160 http://pastie.org/1744167[18:06:11] *** ALR has joined #jboss-as7[18:06:12] *** ChanServ sets mode: +v ALR[18:06:21] <smarlow> systeinfo is referenced from the hibernate dialect for derby[18:06:21] <Nihility> thats nice but it doesnt answer my question[18:06:22] <Nihility> :)[18:06:29] <Nihility> oh[18:07:07] <smarlow> sysinfo is in the apache derby client jar but hibernate is using the TCCL[18:07:08] *** jclingan1 has quit IRC[18:07:14] <Nihility> well thats just wrong[18:07:33] *** jma has quit IRC[18:07:34] <dmlloyd> the problem is that hibernate needs all these deps for dialects[18:07:39] <dmlloyd> we'll have to figure out some way to deal with that[18:07:47] *** JimMa has quit IRC[18:07:50] <Nihility> we probably have to patch hibernate[18:07:57] <Nihility> which sucks[18:08:20] <Nihility> only solution atm is to put the driver in the users deployment[18:08:27] <jpederse> smarlow: we did a classloading spi for the JDBC layer - you'll need something similar[18:08:53] <jpederse> smarlow: in order to remove all dependencies on TCCL[18:08:57] <dmlloyd> or package the dialect with the driver[18:08:58] <Nihility> it would have to take the JCA resource[18:09:03] <Nihility> and then it would have to unwrap it[18:09:16] <Nihility> and then call getClass.getClassLoader[18:09:41] *** jcosta has quit IRC[18:09:49] <jpederse> yeah, that is another idea[18:10:22] <jpederse> but the JCA layer only adds the JDBC driver jar as a dependecy[18:10:39] <Nihility> yeah its make a really bad assumption here[18:10:44] <jpederse> so if anything is outside of that Hibernate will need additional info[18:10:49] <Nihility> its assuming that the Resource JCA is giving it[18:11:05] <Nihility> happens to refer to the driver on the classpath of the application[18:11:46] <Nihility> right there is no contract with JCA to get access to the Driver is there?[18:11:53] *** jfd has quit IRC[18:11:54] <Nihility> other than unwrapping a JDBC object[18:11:55] *** adietisheim has quit IRC[18:11:59] *** adietisheim has joined #jboss-as7[18:12:30] <jpederse> that is an internal detail of the resource adapter - there is no contract[18:12:34] <bstansberry> rmaucher: i'm going through all the demos, and the web.connector one fails[18:12:49] <dmlloyd> it could just as well use the driver's DataSource impl, if it had one[18:13:20] <Nihility> honestly i dont see why the dialects need to do this[18:13:52] <Nihility> just so it can log a version?[18:14:35] <bstansberry> early lunch; bbiab[18:14:41] *** bstansberry is now known as bstans_afk[18:15:15] <rmaucher> bstansberry, emanuel supposedly has fixed it[18:15:22] <smarlow> Nihility: they are using the version to determine jdbc driver capabilities (if driver version is greater than 10, then limit is supported)[18:16:14] <bstans_afk> rmaucher: hmm, fails on master. the fix he gave me earlier was for domain.subsystem[18:16:39] <bstans_afk> i gotta go for an hour or so[18:17:38] <rmaucher> bstansberry, ok, change "protocol" value to "HTTP/1.1" from "http"[18:18:26] <Nihility> smarlow: smcgowan ok until we get this sorted, i think for the tck we need to use derby as a module, and make all deployments import it[18:19:09] <smcgowan> nihility: this problem does not impact testing, derby is currently configured as a module[18:19:12] <rmaucher> https://github.com/rmaucher/jboss-as/commit/80b96619ec12e5645ddab0dc622958d967d0a8c4[18:19:13] <jbossbot> git [jboss-as] 80b9661.. Rémy Maucherat Fix "protocol" value for the connector demo.[18:19:32] <Nihility> rmaucher: thanks ill merge that[18:19:37] <rmaucher> ok, thanks[18:19:42] <rmaucher> afk too now ;)[18:19:53] <Nihility> ok thanks for your help[18:20:05] <jpederse> Nihility: there is a nice juicy one from maeste too ;)[18:20:33] <rmaucher> I'll improve the behavior for the next build when there's a bad value for protocol ;)[18:20:52] <Nihility> hahahaha[18:21:24] <Nihility> WTF GET / http doesnt work[18:22:37] <Nihility> baileyje:how is async going?[18:23:15] <smarlow> Nihility, smcgowan: the org.hibernate module doesn't reference the derby module currently. Could we inject the jdbc driver into hibernate perhaps? Are we injecting it into user deployments that use the jdbc driver directly?[18:23:47] <baileyje> Nihility: Pretty good. I am doing some cleanup while we get jboss-invocation updates[18:23:52] <Nihility> cool[18:24:35] <Nihility> smarlow: if a user uses jdbc directly then they just add it to their class space via a module import, or including it in the deployment, or a class-path ref[18:24:36] *** pilhuhn has quit IRC[18:24:43] *** vtunka has quit IRC[18:24:45] *** pilhuhn has joined #jboss-as7[18:24:45] *** pilhuhn has joined #jboss-as7[18:24:45] *** ChanServ sets mode: +v pilhuhn[18:25:06] <Nihility> smarlow: we should also have a way to bind the Driver instances to deployments though[18:25:20] <Nihility> perhaps by supporting @Resource on Driver[18:26:23] <Nihility> @DataSourceDefinition currently requires the class space visibility[18:26:35] *** darranl is now known as darranl_afk[18:26:40] <smarlow> I assume it still needs to get into the persistence provider (org.hibernate module in this case). Should we hack a bunch of jdbc drivers as optional dependencies into the hibernate module.[18:26:52] <Nihility> hell no[18:26:54] <Nihility> :)[18:27:06] <smarlow> :) good[18:27:33] <Nihility> hibernate should have a way to get it from the server relative to the deployment[18:27:53] <Nihility> based on the actual resource that is being used[18:28:33] <Nihility> to me i think the simplest solution would be a way to go from a JCA resource to the Driver instance[18:28:40] <Nihility> some kind of spi[18:29:14] <jbossbot> git [jboss-as] push master 80b9661.. Rémy Maucherat Fix "protocol" value for the connector demo.[18:29:14] <jbossbot> git [jboss-as] push master URL: http://github.com/jbossas/jboss-as/compare/daa9dea...80b9661[18:29:46] <Nihility> ALR: hey you going to do JBAS-8946 today?[18:29:47] <jbossbot> jira [JBAS-8946] Upgrade to latest Arquillian release [Open (Unresolved) Task, Major, Andrew Rubinger] https://issues.jboss.org/browse/JBAS-8946[18:30:11] <jpederse> Nihility: that is hacky[18:30:43] <jpederse> Nihility: we have the services necessary for the Hibernate integration to fix this[18:30:55] <Nihility> well the problem i more worry about is correlation[18:31:03] <jpederse> Nihility: correct, they need to hook into it[18:31:19] <Nihility> if we require that the user specify hibernate metadata that specifies a driver[18:31:24] <Nihility> it might not match what jca is using[18:31:29] <jpederse> true[18:31:46] <jpederse> but the dependencies can be found in AS 7[18:31:59] <Nihility> really the problem to me is that this should be done via properties[18:32:02] <Nihility> on DataSource[18:32:04] <jpederse> since the datasource deployment depends on the JDBC service[18:32:15] <Nihility> like it should be DataSource.getProperty("Version")[18:32:19] <Nihility> and then it all just works[18:32:24] <jpederse> :)[18:33:29] *** mbg is now known as mbg|away[18:33:47] <Nihility> yeah basically the user provides a jndi name right[18:33:57] <jpederse> persistence.xml -> jndi name -> lookup datasource -> get reference to driver reference -> add to deployment ?[18:33:59] <Nihility> so we just need a way to go from that jndi name to the driver[18:34:28] <Nihility> right except instead of add to deployment[18:34:36] <Nihility> it would be "make info available to hibernate"[18:34:49] <jpederse> I don't see why that can't be handled automatically in the JPA layer[18:34:51] <asoldano> Nihility, dmlloyd: all clear till now from ws pov for the Beta2 release http://jbossws.jboss.org:8180/hudson/job/CXF-CORE-AS-7.0.0-SUN-JDK-6/ ; should you want to re-run the tests just before tagging, let me know and I can send you the hudson user/pwd (I'm leaving in half an hour from now)[18:34:58] <jpederse> what ever you call it ;)[18:35:02] <Nihility> true yeah[18:35:10] <Nihility> the code to do this can be anywhere[18:35:26] <Nihility> but it will have to know how to go from jndi name to driver value in the ds config[18:35:32] <jpederse> sure, the entry point is the jndi name[18:36:11] <Nihility> asoldano: thats great![18:36:27] <jpederse> well, it is just a matter of lookups - if the information isn't exposed on the existing services we can just add it - we use it today internally[18:36:51] <baileyje> dmlloyd: Once the jboss-invocation is ready, I can update as to use it..[18:37:09] <baileyje> since I need it anyway, and have already fixed the tests that needed updating for it[18:37:39] <jpederse> smarlow: so try that[18:37:43] <dmlloyd> any reason you want to do it right now rather than when you're ready to merge, baileyje?[18:38:04] <baileyje> dmlloyd: Nope. Either way..[18:38:27] <baileyje> I just cleaned things up a bit. I will have your review in a little bit (if you have time)[18:39:44] <dmlloyd> sure[18:41:57] <ALR> Nihility: I've been doing it the past 2 days.[18:42:10] <ALR> JBAS-8946[18:42:11] <jbossbot> jira [JBAS-8946] Upgrade to latest Arquillian release [Open (Unresolved) Task, Major, Andrew Rubinger] https://issues.jboss.org/browse/JBAS-8946[18:42:17] <smarlow> Nihility, jpederse: what does the datasource service have to expose and do we have an instance of the hibernate module per deployment?[18:42:33] <ALR> What I have now is moved all the stuff that can't be in ARQ under the AS7 tree.[18:42:47] <ALR> Except there's also some changes that were made to existing ARQ stuff.[18:43:01] <ALR> And sorting through those is tougher than just copying modules/classes/packages[18:43:26] <jpederse> smarlow: the reference specified in the <driver> section[18:43:40] <jpederse> s/section/element[18:43:42] <Nihility> ALR: right im just thinking perhaps JBAS-8946 should be moved to beta3[18:43:43] <jbossbot> jira [JBAS-8946] Upgrade to latest Arquillian release [Open (Unresolved) Task, Major, Andrew Rubinger] https://issues.jboss.org/browse/JBAS-8946[18:43:46] <Nihility> since we are releasing today[18:43:53] <ALR> Nihility: Ah in that case.[18:43:59] <ALR> Yeah I'd wanted to have it knocked out.[18:44:05] <ALR> I thought it was just extra modules that were made[18:44:25] <smarlow> jpederse: driver is the module name?[18:44:31] <ALR> Nihility: There are more changes to existing ARQ stuff than I was expecting. I'll bump it.[18:44:32] <Nihility> right id be worried about throwing a big change in at the last minute before you had chance to let it settle[18:44:38] <jpederse> smarlow: yes, and version[18:44:54] <ALR> Nihility: Bumped.[18:44:55] <jpederse> smarlow: well, the driver name[18:45:04] <Nihility> cool[18:45:15] <smarlow> jpederse: don't we need the module name?[18:45:17] <Nihility> so now its just baileyje and i holding up the release[18:45:31] <Nihility> smarlow: take a look at this wiki:[18:45:57] <Nihility> http://community.jboss.org/wiki/DataSourceConfigurationinAS7[18:46:43] *** mmoyses_ is now known as mmoyses[18:47:49] <jpederse> smarlow: you will have to hook into the JDBC service based on the <driver> entry[18:51:12] <baileyje> dmlloyd: https://github.com/baileyje/jboss-as/compare/daa9dea9389f2acef180...dab1e381e3e010b1bf0a[18:52:31] <smarlow> jpederse, Nihility: If I get the ModuleSpecification for Hibernate and inject into it, am I changing a global module or a per deployment module?[18:53:17] <Nihility> no you dont want to do that[18:53:32] <Nihility> you dont want hibernate to have all jdbc drivers on its classpath[18:53:43] <Nihility> we can have muliple drivers with different versions[18:53:48] <Nihility> so it will eventually break[18:54:15] <dmlloyd> yeah what we need is to split out the dialect impls and attach them to the drivers somehow[18:54:21] *** pferraro has quit IRC[18:54:36] <Nihility> hibernate needs to go from the jndi entry in persistence.xml to the Driver instance being used[18:54:37] *** pferraro has joined #jboss-as7[18:54:37] *** ChanServ sets mode: +v pferraro[18:54:42] <dmlloyd> or else have them rely solely on reflection somehow[18:55:02] <Nihility> if it does that it can do getClass().getClassLoader() on Driver[18:55:13] <Nihility> and load all the classes in the driver it wants[18:55:19] <smarlow> Nihility: oh, you mean like a new version of hibernate, like 4.0?[18:55:51] <Nihility> i mean our jpa code has to know how to do that, cooperating with the datasource subsystem somehow[18:56:04] <Nihility> and then hibernate needs to access that information[18:56:17] <dmlloyd> baileyje: https://github.com/baileyje/jboss-as/commit/dab1e381e3e010b1bf0afc292c18693d09a2da73#L16R162 - just use Arrays.equals() here - the second if statement is incorrect anyway[18:56:18] <jbossbot> git [jboss-as] dab1e38.. John E. Bailey [JBAS-8978] - More Async impl.[18:56:19] <jbossbot> jira [JBAS-8978] Complete async EJB invocation [Open (Unresolved) Task, Major, John Bailey] https://issues.jboss.org/browse/JBAS-8978[18:56:48] <smarlow> if our jpa layer sets the TCCL to include the jdbc driver, that would hack around this bug[18:56:52] *** slaboure has quit IRC[18:57:29] <Nihility> the only reasonable workaround is for the deployment to reference the driver[18:57:44] <Nihility> extra configuration[18:58:15] <Nihility> it could do a Class-Path on the driver for example[18:58:19] <Nihility> or it could use a module import[18:58:59] <smarlow> looking at the error message: java.lang.ClassNotFoundException: org.apache.derby.tools.sysinfo from [Module "org.hibernate:main" from local module loader @fee4648[18:59:02] *** pferraro has quit IRC[18:59:02] <Nihility> the jpa subsystem could dynamicly add that information sure[18:59:08] <Nihility> but it would be throw away code[18:59:12] <smcgowan> nihility: one other issue (kind of a blocker) is to be able to do a lookup of java:/comp/env of the DB resource. i was speaking with stuartdouglas last night and it may be a case of resource-ref not being fully implemented but I'd like to discuss with you quickly to be sure this is true[18:59:32] <Nihility> supposedly resource-ref was implemented[18:59:42] <Nihility> in web.xml[18:59:45] <smcgowan> nihility: once I get this resolved i should be able to run the jpa tests in hudson[18:59:56] <Nihility> stuartdouglas was the one that added it[18:59:56] <baileyje> dmlloyd: good catch. I should not write core at 12:30 AM, or should make sure I review it in the morning.[18:59:58] *** pferraro has joined #jboss-as7[18:59:58] *** ChanServ sets mode: +v pferraro[18:59:59] <smcgowan> i have this derby config: https://svn.devel.redhat.com/repos/jboss-tck/tck6/trunk/jee6tck-as7-mods/derby/derby-config.xml[19:00:05] <Nihility> i know it works in ejb deployments[19:00:16] <smcgowan> this web.xml and jboss-web.xml: http://pastebin.com/H6R2SGr9[19:00:41] <smcgowan> but I get this error: http://pastebin.com/Y8j41bwJ ; e.g,, #[19:00:41] <smcgowan> Caused by: java.lang.IllegalArgumentException: key is null[19:00:57] <dmlloyd> that jndi-name of java:/DB1 isn't valid[19:01:03] <dmlloyd> it's gotta be in a namespace[19:01:20] <smcgowan> you mean jdbc/DB1 ?[19:01:21] <dmlloyd> java:global, :module, :app, or :comp[19:01:31] <dmlloyd> no the jndi-name[19:01:52] <dmlloyd> the res-ref-name is fine, assuming that there really is something in java:comp/jdbc/DB1[19:01:57] <dmlloyd> or comp/enf[19:02:00] <dmlloyd> or whatever it defaults to[19:02:13] <Nihility> dmlloyd: right now ds' are binding in java:[19:02:17] *** jhalliday has quit IRC[19:02:49] <smcgowan> dmlloyd: i've tried several combinations but none are working for me so any hints are welcome[19:03:01] <Nihility> dmlloyd: take a look at the default standalone config[19:03:07] <dmlloyd> well Nihility that's not going to work at all with injection[19:03:11] <dmlloyd> you'll get goofy exceptions[19:03:13] <smcgowan> for the JPA tests to run, i had to defined jta-data-source as java:/DB1[19:03:17] <dmlloyd> because there's no mapping to a source service[19:04:11] <baileyje> dmlloyd: updated. https://github.com/baileyje/jboss-as/compare/daa9dea9389f2acef180...216b249fbcb8820161e7[19:04:16] <dmlloyd> if we want to add support for namespaceless things, then we need to enhance stuff[19:04:40] *** pferraro has quit IRC[19:04:49] <Nihility> the DS code isnt using a component service to bind this stuff[19:05:01] <dmlloyd> then it will never work for two reasons :)[19:05:26] <dmlloyd> the res-ref stuff uses MSC services[19:05:37] <dmlloyd> everything bound has to use BinderService or it won't be injectable[19:05:49] <dmlloyd> and will ultimately stop working once JNDI is read-only[19:05:55] <dmlloyd> like HQ[19:07:04] <dmlloyd> https://github.com/baileyje/jboss-as/compare/jbossas:master...baileyje:JBAS-8978#L21R172 <- this could just be == void.class because nothing but void is assignable to void[19:07:06] <jbossbot> jira [JBAS-8978] Complete async EJB invocation [Open (Unresolved) Task, Major, John Bailey] https://issues.jboss.org/browse/JBAS-8978[19:07:09] <dmlloyd> not really super important though[19:07:32] <dmlloyd> same here; https://github.com/baileyje/jboss-as/compare/jbossas:master...baileyje:JBAS-8978#L23R335[19:07:33] <jbossbot> jira [JBAS-8978] Complete async EJB invocation [Open (Unresolved) Task, Major, John Bailey] https://issues.jboss.org/browse/JBAS-8978[19:08:13] *** pferraro has joined #jboss-as7[19:08:13] *** ChanServ sets mode: +v pferraro[19:08:16] <dmlloyd> otherwise I say, good 'nuff[19:08:38] *** asoldano has quit IRC[19:10:10] <Nihility> right so basically we wont be able to progress with tck until the jndi strategy uses BinderService[19:11:47] <wolfc> the async test might be a bit beefier[19:12:01] <smarlow> Nihility: I'm not following how changing the deployment to reference the driver helps. Its the Hibenate module that needs access to the driver[19:12:39] <Nihility> smarlow: no the hibernate module doesnt need to see the driver, just the dialect code does[19:12:43] *** pferraro has quit IRC[19:12:53] <Nihility> and the dialect code can get a classloader from a driver[19:13:00] <smcgowan> thanks, Nihility/dmlloyd for feedback on that issue[19:13:00] <wolfc> dmlloyd, baileyje, ultimately we need an executor per component for in-vm asyncs and a local executor for client asyncs[19:13:20] <Nihility> so we should create a high priority item on that[19:13:34] <dmlloyd> no, we should be able to share a single one wolfc[19:13:35] <Nihility> (jca using binder service)[19:13:38] <smcgowan> nihility: how does amiller's app look up the DS[19:13:58] <dmlloyd> ultimately it should be an on-demand thread pool service which is started when one or more async EJB is deployed[19:14:01] <Nihility> smcgowan: it works just fine since it use a persistence.xml that does not point to a resource-ref[19:14:01] *** rawbdor has joined #jboss-as7[19:14:01] *** ChanServ sets mode: +v rawbdor[19:14:02] <wolfc> dmlloyd, that means a single component can consume all server resources[19:14:17] <Nihility> so bypasses the problem[19:14:21] <dmlloyd> the async invocation pool should be a queueless pool with no upper bound[19:14:30] <smcgowan> nihility: ah, ok[19:14:53] <Nihility> so that means in a nutshell[19:14:58] <wolfc> dmlloyd, that can lead to resource starvation[19:15:09] <Nihility> resource-ref pointing to ds' or jms destionations is completely broken[19:15:29] <baileyje> dmlloyd: pushed a fix for it.[19:15:31] <dmlloyd> anything can lead to resource starvation, the question is, how many hoops do we make normal users jump through to accomodate pathological cases[19:15:38] <wolfc> dmlloyd, the whole idea behind EJB is to make resources manageable from an admin perspective[19:15:40] <Nihility> it only is working for ejbs, managed beans, and env entries[19:16:03] <jpederse> Nihility: jndi stuff for connector is under connector.jndi - if you want to have a look[19:16:10] <dmlloyd> well it's a discussion for another day anyway.[19:16:14] <wolfc> dmlloyd, if we have an unbounded queue per component with a set amount of threads, then we should be good[19:16:28] *** pferraro has joined #jboss-as7[19:16:28] *** ChanServ sets mode: +v pferraro[19:16:42] <smarlow> Nihility: not currently in Hibernate 3.6. For Hibernate 4.0, they redesigned the dialects so that this could be fixed in Hibernate.[19:16:43] <Nihility> smcgowan: i had it in my head that we had a workaround but dmlloyd says we dont which explains the failure[19:16:50] <baileyje> wolfc: That sounds like a good beta3 task :)[19:17:04] <Nihility> smarlow: yeah thats why i mentioned we would have to patch hibernate[19:17:10] <smcgowan> nihility:i can file that JIRA if one does not yet exist[19:17:11] <Nihility> which sucks[19:17:19] <wolfc> yes, right now if it works, it works. the test doesn't prove it though.[19:17:24] <Nihility> smcgowan: yes pease, feel free to make it a blocker[19:17:39] <smcgowan> nihility: Beta3 blocker, right?[19:18:07] <smcgowan> :)[19:18:35] <Nihility> correct[19:19:40] <Nihility> dmlloyd: btw the workaround i had in my head was that if the jndi name was not in an EE namespace, we just looked it up and hoped it was there, which would likely be true[19:20:20] <smarlow> smcgowan: I didn't create the jira yet[19:20:23] *** darranl_afk is now known as darranl[19:20:53] <smarlow> smcgowan: never mind, I thought you were in the same conversation[19:21:20] * smarlow creating a Hibernate jira to stop using TCCL in Dialects[19:21:46] <dmlloyd> Nihility: that doesn't really help us in the long term though, unless we decide to make JNDI user-writable after all[19:22:20] <Nihility> dmlloyd: certainly[19:22:50] <Nihility> dmlloyd: fixing jca can be done quickly, jms on the other hand...[19:23:13] <Nihility> well i guess we could make it work[19:23:25] <Nihility> by forking temporarily[19:23:34] <Nihility> im hoping to avoid that though[19:25:27] <dmlloyd> we should be able to fix JCA though[19:25:58] *** jfd has joined #jboss-as7[19:25:58] *** jfd has quit IRC[19:25:58] *** jfd has joined #jboss-as7[19:25:58] *** ChanServ sets mode: +v jfd[19:26:11] <Nihility> yeah thats no big deal[19:26:21] <Nihility> we need to create that jboss namespace as part of that[19:26:25] <Nihility> ill put that on my plate[19:26:44] *** ccrouch has quit IRC[19:26:46] <Nihility> java:jboss/datasources/blah[19:27:18] <dmlloyd> we can possibly trick hornetq as well - hand it a fake Context and trap the ObjectFactory they bind or whatever[19:27:20] *** ccrouch has joined #jboss-as7[19:27:20] *** ChanServ sets mode: +v ccrouch[19:27:34] <Nihility> oh thats a good idea actually[19:27:44] <Nihility> we can create a EmulatedBinderService[19:27:51] <Nihility> that uses that context[19:27:56] <dmlloyd> depending on how long AndyTaylor and clebert hold out on us :)[19:28:06] <Nihility> worse comes to worse[19:28:09] <Nihility> ill send them a patch[19:28:31] <Nihility> and fork until they merge it[19:28:37] <Nihility> i dont want to wait another release cycle[19:29:01] <dmlloyd> yeah let's put getting JNDI & binding completely fixed on the beta3 roadmap[19:29:13] <dmlloyd> fits with the stuff we need to do to EE anyway[19:29:32] <baileyje> dmlloyd: Gotta go grab lunch. Be back in a little bit[19:29:35] <Nihility> yeah i really want smcgowan to be able to make progress with the TCK[19:29:42] <Nihility> and this seems essentially[19:29:47] <Nihility> the biggest blocker[19:30:04] <dmlloyd> yeah there are many connected problems[19:30:07] <smcgowan> that would be great,[19:30:08] <ALR> My AS build hangs have become more and more frequent[19:30:20] <ALR> http://pastebin.com/zNV0iBZv[19:30:36] <dmlloyd> I've never seen this, ALR[19:30:48] <dmlloyd> and I do full clean rebuilds on every patch I merge[19:30:49] <ALR> dmlloyd: I have a magic machine :)[19:31:06] <ALR> dmlloyd: It's transient, but lately I've been hitting it just right.[19:31:08] <smarlow> smcgowan, Nihility: HHH-6073 is for the Hibernate TCCL issue[19:31:08] <jbossbot> jira [HHH-6073] Dialects cannot use the Thread Context ClassLoader with AS7, please change to use the [Open (Unresolved) Task, Blocker, Unassigned] http://opensource.atlassian.com/projects/hibernate/browse/HHH-6073[19:31:40] <smcgowan> and i'll continue to look at the failures related to this that are exposed differently[19:32:36] <Nihility> smarlow: we can't use java.sql.Connection, that is a wrapper from JCA, it has to either unwrap it (kind of a hack) or use a new spi thing we define[19:35:01] <Nihility> smarlow: an example would be like what they do for scanning a callback class[19:35:16] <Nihility> if we had a callback we could figure it out[19:35:20] <Nihility> and return the right class to them[19:39:07] *** jfclere has quit IRC[19:41:42] *** ccrouch has quit IRC[19:42:16] <smarlow> Nihility: There is a ClassLoaderService in Hibernate 4.0, I'm talking to steve about using that or maybe just fixing the dialects to not use the TCCL[19:42:47] <smarlow> for the ClassLoaderService, like you said ^, I would have to pass the callback class[19:44:06] *** ALR has quit IRC[19:44:57] <jbossbot> git [jboss-invocation] push master d8d152d.. David M. Lloyd Prep 1.0.0.Beta4[19:44:57] <jbossbot> git [jboss-invocation] push master 415706f.. David M. Lloyd Next is Beta5[19:44:57] <jbossbot> git [jboss-invocation] push master URL: http://github.com/jbossas/jboss-invocation/compare/8125ab9...415706f[19:46:25] *** smcgowan is now known as smcgowan_lunch[19:46:46] <jbossbot> git [jboss-dmr] push master db87bb9.. Jason T. Greene Revert Base64 impl for the time being[19:46:46] <jbossbot> git [jboss-dmr] push master URL: http://github.com/jbossas/jboss-dmr/compare/ac4a57f...db87bb9[19:47:02] *** fnasser has quit IRC[19:51:48] *** frainone has quit IRC[19:52:22] *** mbg|away is now known as mbg[19:54:08] <jbossbot> git [jboss-as] push master 0d6db33.. David M. Lloyd Fix varargs warning[19:54:08] <jbossbot> git [jboss-as] push master 98ef9fa.. David M. Lloyd Minor javadoc fix[19:54:09] <jbossbot> git [jboss-as] push master 91c9feb.. David M. Lloyd [JBAS-8978] Asynchronous invocation support[19:54:09] <jbossbot> jira [JBAS-8978] Complete async EJB invocation [Open (Unresolved) Task, Major, John Bailey] https://issues.jboss.org/browse/JBAS-8978[19:54:10] <jbossbot> git [jboss-as] push master 5d9d234.. John E. Bailey [JBAS-8978] - Implement the initial @Asynchronous parsing[19:54:10] <jbossbot> git [jboss-as] push master 216b249.. John E. Bailey [JBAS-8978] - More Async impl.[19:54:10] <jbossbot> git [jboss-as] push master b332558.. David M. Lloyd Use Beta4 of jboss-invocation[19:54:10] <jbossbot> git [jboss-as] push master URL: http://github.com/jbossas/jboss-as/compare/80b9661...b332558[19:59:56] <jbossbot> git [jboss-dmr] push master ce60046.. Jason T. Greene Release 1.0.0.Beta5[19:59:56] <jbossbot> git [jboss-dmr] push master 3edfa44.. Jason T. Greene Next is Beta6[19:59:56] <jbossbot> git [jboss-dmr] push master URL: http://github.com/jbossas/jboss-dmr/compare/ff822a8...3edfa44[20:00:10] <jbossbot> git [jboss-as] push master 0c312ff.. Jason T. Greene Update to DMR 1.0.0.Beta5[20:00:10] <jbossbot> git [jboss-as] push master URL: http://github.com/jbossas/jboss-as/compare/b332558...0c312ff[20:01:18] <Nihility> baileyje: so that means i can close it right[20:01:19] <Nihility> ?[20:01:35] <dmlloyd> he's eating lunch but yeah I say close it out[20:02:04] <dmlloyd> let the Last Issue Closing ceremony commence[20:03:08] <Nihility> cool[20:03:13] <Nihility> anyone in here have a windows box[20:03:37] <jamezp> I do, but only XP.[20:03:44] <Nihility> thats perfect[20:04:16] <Nihility> jamezp: when i tag can you build and start it ?[20:04:51] <jamezp> Sure, I can handle that.[20:05:12] <jbossbot> git [jboss-as] push master afc3e3c.. Jason T. Greene Code name![20:05:12] <jbossbot> git [jboss-as] push master URL: http://github.com/jbossas/jboss-as/compare/0c312ff...afc3e3c[20:05:26] <Nihility> jamezp: thanks![20:06:04] <jamezp> No problem. Getting ready to clone now :-)[20:06:43] <Nihility> 0 issues remaining[20:07:50] <Nihility> baileyje wins hte last issue prize[20:07:57] <Nihility> 123 issues in 2 weeks[20:07:58] <Nihility> not bad[20:07:59] <Nihility> :)[20:08:05] * jamezp thinks git on Windows sucks.[20:08:50] *** torben has joined #jboss-as7[20:08:54] *** torben has joined #jboss-as7[20:08:54] *** ChanServ sets mode: +v torben[20:13:30] *** torben has quit IRC[20:14:21] <baileyje> dmlloyd Thanks[20:14:38] <dmlloyd> no prob[20:14:46] <dmlloyd> 54 issues before beta3![20:14:50] <dmlloyd> let's get to work![20:18:44] <wolfc> dmlloyd, Nihility, I don't see https://github.com/wolfc/jboss-as/commits/ejb3-ws[20:19:17] <dmlloyd> ah the push was rejected. I'll have to rebase it[20:19:36] *** bstans_afk is now known as bstansberry[20:19:37] <dmlloyd> which of course now fails due to conflicts.[20:19:52] <dmlloyd> wolfc: can you fix it and I'll do it again?[20:20:17] <wolfc> dmlloyd, on it[20:23:46] <jamezp> Nihility: Deployment Scanner failure on Windows.[20:25:16] <Nihility> jamezp: doh whats the error[20:25:44] <Nihility> stuartdouglas: hey is resteasy integrated with ejb3 or is that still pending[20:25:45] <Nihility> ?[20:25:50] <jamezp> Nihility: http://fpaste.org/ShYw/[20:26:35] <jamezp> I wonder if I should try on a path with no space.[20:26:47] <Nihility> thats probably whats doing it[20:27:16] <jamezp> I'll try on a path with no space and see if that works.[20:29:37] <Nihility> ok thats a bug in AutoDeployTestSupport[20:29:43] <Nihility> it's use getPath vs getRealPath[20:30:21] <dmlloyd> we did have an issue with spaces before but it should have been fixed already[20:30:23] <Nihility> ill fix that[20:30:35] <Nihility> this is outside of vfs[20:30:41] <dmlloyd> could be a shrinkwarp issue[20:30:54] <dmlloyd> ah I see[20:30:57] <dmlloyd> it's using vfs[20:31:01] <dmlloyd> when it shouldn't, I mean[20:31:20] *** ccrouch has joined #jboss-as7[20:31:20] *** ChanServ sets mode: +v ccrouch[20:33:24] <Nihility> nah[20:33:30] <Nihility> its just bad url usage[20:33:51] <Nihility> its using a URL from classloader[20:33:58] <Nihility> and then calling getPath on it[20:34:01] <Nihility> and shoving it into a file[20:34:12] <dmlloyd> I'm thinking we may want to make codeSource URLs reflect real jar:file: and file: URLs[20:34:19] <dmlloyd> seems like a lot of things depend on that[20:34:48] <Nihility> this is our code so no excuses :)[20:34:48] <dmlloyd> it'd just suck having a bunch of temp dir crap in there[20:34:58] <dmlloyd> maybe configurable[20:35:23] <Nihility> jamezp: i am going to push a commit if you you could try it[20:35:38] <jamezp> Sure thing.[20:37:10] *** ALR has joined #jboss-as7[20:37:10] *** ChanServ sets mode: +v ALR[20:42:09] <jbossbot> git [jboss-as] push master e914bbb.. Jason T. Greene Fix URL handling of paths with spaces[20:42:09] <jbossbot> git [jboss-as] push master URL: http://github.com/jbossas/jboss-as/compare/afc3e3c...e914bbb[20:42:12] <Nihility> feel free to retro review[20:42:13] <Nihility> :)[20:42:20] <Nihility> jamezp: can you try that[20:42:25] <jamezp> Sure thing.[20:43:45] <dmlloyd> nice[20:46:28] <bstansberry> that code was cut-and-pasted from something in the integration testsuite, so I won't be surprised if it fails there as well[20:46:40] <Nihility> ah[20:47:40] *** smarlow has quit IRC[20:47:41] <bstansberry> http://community.jboss.org/wiki/WritingaAS7TestCaseintestsuitemodule[20:47:54] * bstansberry goes to look for more of same[20:48:11] <Nihility> ah yes its asaldhan's stuff[20:48:12] <Nihility> thats it[20:48:35] <Nihility> i think i need to make a global class in AS[20:48:41] <Nihility> that does URL to file conversion[20:48:47] *** darranl has quit IRC[20:48:48] <Nihility> everyone screws it up one way or the other[20:49:10] *** smarlow has joined #jboss-as7[20:49:11] *** ChanServ sets mode: +v smarlow[20:49:37] <Nihility> if that works for you jamezp ill fix the other areas[20:50:13] <jamezp> Sounds good. Testing now.[20:53:05] *** torben has joined #jboss-as7[20:53:06] *** torben has joined #jboss-as7[20:53:06] *** ChanServ sets mode: +v torben[20:53:26] * jamezp has a crappy laptop and keeps running out of memory.[20:55:08] <Nihility> actually im going to change this to not use a file at all[20:55:51] *** frainone has joined #jboss-as7[20:55:51] *** ChanServ sets mode: +v frainone[20:57:15] <jamezp> Nihility: Still getting the same error. Let me double check that my pull request worked.[20:58:28] <jamezp> Pull worked, but same failure.[20:58:33] <Nihility> same class?[20:58:46] <jamezp> Yes, exact same error message.[20:59:01] <Nihility> maybe maven cached the class or something[20:59:33] <jamezp> I did an mvn clean first to be safe.[20:59:39] <Nihility> actually i have a better fix anyway[20:59:48] <jamezp> Sweet[20:59:59] <Nihility> it doesnt use files from urls[21:00:03] <Nihility> just testing it here[21:00:25] <Nihility> it should be impossible to get a %20 in the path[21:00:31] <Nihility> with the patch i did though[21:01:00] <jamezp> Good point...[21:01:13] <jamezp> Let me try again then to make sure the clean worked.[21:01:14] <Nihility> anyway one sec[21:03:03] <bstansberry> Nihility: yeah, the Shrinkwrap stuff can take a URL[21:03:27] <Nihility> actually im a moron on that older patch[21:04:08] <Nihility> URI.getPath[21:04:13] <Nihility> it would it should have done[21:04:14] <Nihility> duh[21:04:16] <Nihility> well no matter[21:04:18] <Nihility> better patch comng[21:08:35] *** pilhuhn has quit IRC[21:16:07] <dmlloyd> wolfc: progress?[21:16:30] <wolfc> dmlloyd, it's running. I'm being hit by a kernel bug.[21:16:42] <dmlloyd> that doesn't sound good.[21:17:04] <wolfc> it's not supposed to take 10 minutes to compile AS 7[21:18:15] <wolfc> I'll push after I'm done with rebasing, then we can both run integration-tests. You'll be there much faster.[21:18:32] <dmlloyd> ok[21:20:13] *** smcgowan_lunch is now known as smcgowan[21:23:39] <wolfc> hmm, supposedly fixed in kernel 2.6.36.2. might as well go to 2.6.38 then. It has been a long since I needed a custom kernel.[21:24:23] <dmlloyd> what issue[21:24:29] <Nihility> bstansberry: https://github.com/n1hility/jboss-as/commit/8cf1052469df97c9a5efb171201db2d22753b582[21:24:30] <jbossbot> git [jboss-as] 8cf1052.. Jason T. Greene Use URLs instead of files with tests that lookup resources on TCCL[21:24:36] <Nihility> does that match what you came up with[21:24:36] <Nihility> ?[21:24:46] <Nihility> oh no[21:24:47] <Nihility> thats wrong[21:24:48] <Nihility> haha[21:24:49] <Nihility> dman[21:25:38] <bstansberry> hehe[21:25:40] <Nihility> https://github.com/n1hility/jboss-as/commit/0d8dbd3c15970952d4ded1c1d994f1b033febecc[21:25:41] <jbossbot> git [jboss-as] 0d8dbd3.. Jason T. Greene Use URLs instead of files with tests that lookup resources on TCCL[21:25:55] <Nihility> or i can just take yours[21:26:18] <bstansberry> Nihility: approved[21:26:27] <jbossbot> git [jboss-as] push master 0d8dbd3.. Jason T. Greene Use URLs instead of files with tests that lookup resources on TCCL[21:26:27] <jbossbot> git [jboss-as] push master URL: http://github.com/jbossas/jboss-as/compare/e914bbb...0d8dbd3[21:26:38] <Nihility> jamezp: finally[21:26:51] <jamezp> Cool, pulling now[21:26:53] <Nihility> i hate it when things that simple are harder than they need to be[21:26:57] <bstansberry> i'll go fix that wiki[21:27:35] <wolfc> dmlloyd, finally after 12 minutes of compiling. https://github.com/wolfc/jboss-as/commits/ejb3-ws I haven't tested the final commit.[21:27:40] <jamezp> Nihility: Isn't that how it usually goes? :-)[21:28:58] <Nihility> so i was thinking of posting an april fools joke about AS being rewritten in scala[21:30:25] <Nihility> "I was hanging out on ##java, and someone brought up an interesting point. Java is a dead language, and functional languages are superior. .... So we have now decided to switch to Scala, finally we can use functions in our software properly as the software gods intended"[21:30:47] <bstansberry> be sure to file a JIRA for Beta3[21:33:29] <jamezp> Hmmm... ...getting a weird missing dependency error in Web Subsystem.[21:34:15] <rmaucher> like what ?[21:34:28] <wolfc> I'm going to reboot, brb[21:34:30] <jamezp> Missing com.sun.faces:jsf-api:jar:2.0.4-b09[21:35:20] *** wolfc has quit IRC[21:36:08] <rmaucher> rebuilding ... ;)[21:37:01] <jamezp> Seems to be working on my Mac, but not my WinXP machine.[21:37:04] <rmaucher> jsf-impl is in the dependency list, but jsf-api ?[21:37:51] <jamezp> I'm actually seeing both missing. It could be a setting I have somewhere though.[21:38:00] <rmaucher> maybe a manifest dependency from your war ?[21:38:20] *** wolfc has joined #jboss-as7[21:38:20] *** ChanServ sets mode: +v wolfc[21:38:59] <jamezp> I'm just executing the build.bat to see if it builds. Maybe I'll just manually install them.[21:39:29] <rmaucher> there's a com.sun.jsf-impl module, but no jsf-api[21:40:14] <jamezp> Yeah, works fine on my Mac. It seems more of a maven issue than a AS build issue.[21:42:52] <Nihility> hmm[21:43:09] <Nihility> dmlloyd: you merging ejb3-ws?[21:47:57] <baileyje> dmlloyd: For the JBAS-8990, are you looking to make them all configurable from the model, or just use thread factories with names?[21:47:59] <jbossbot> jira [JBAS-8990] Eliminate usage of Executors.defaultThreadFactory() [Open (Unresolved) Task, Major, John Bailey] https://issues.jboss.org/browse/JBAS-8990[21:48:57] <jamezp> Think I found it. I was using https://repository.jboss.org/nexus/content/groups/public-jboss/ instead of https://repository.jboss.org/nexus/content/groups/public/ in my settings.[21:49:30] <smcgowan> jamezp: yes, i was just going to say that, there was a problem getting the jsf stuff[21:49:38] <smcgowan> it was on the jboss-dev mail list a while ago[21:49:53] <jamezp> Shows how much I pay attention :-)[21:50:22] <smcgowan> it was a while ago,[21:50:42] <jamezp> I'll blame it on my timing of joining the list then ;-)[21:50:44] <Nihility> wooo i got a windows build working[21:50:53] <Nihility> well[21:50:56] <Nihility> a windows install[21:51:00] <jamezp> I'm trying now, so far so good.[21:51:01] <Nihility> not the as build[21:51:17] <Nihility> i remembered i had an image on my wifes mac[21:51:56] <Nihility> jamezp: neat tip you can get an RDC client for windows[21:52:08] <Nihility> so i can control the shitty windows instance from my mac[21:52:16] <wolfc> ApplicationPathOverrideIntegrationTestCase has stalled[21:52:20] <dmlloyd> Nihility: yeah I'm gonna merge that[21:52:35] <dmlloyd> baileyje: just thread factories with names[21:52:54] <jamezp> Yes, i use it from time to time. It's funny because it actually looks better on my mack than on my ThinkPad.[21:52:55] <baileyje> dmlloyd: Great, I will knock these all out then[21:53:17] <asaldhan> bstansberry: anything wrong with the shrinkwrap usage?[21:53:46] <bstansberry> the URL to File conversion was causing issues on Windows[21:55:37] <bstansberry> so we fixed those 2 tests and another I wrote that was using the same pattern[21:56:01] <bstansberry> we == Nihility[21:56:33] <Jaikiran> dmlloyd: Nihility: this branch was probably never merged upstream yesterday https://github.com/jaikiran/jboss-as/commits/ejb3-dd[21:57:01] <dmlloyd> I believe that is included in wolfc's branch?[21:57:09] <wolfc> it looks like Shrinkwrap blew a fuse http://fpaste.org/MON1/[21:57:21] <wolfc> dmlloyd, it is not included in ejb3-ws[21:57:31] <dmlloyd> wolfc: are you sure you aren't just running out of RAM or something[21:57:44] <asaldhan> bstansberry: ok. I thought Java took care of file paths on windows.[21:57:51] <wolfc> got 3GB left[22:00:03] <jamezp> Windows XP build = success.[22:00:55] <wolfc> I just dumped it into SHRINKWRAP-269 and I'm going to retry the build[22:00:56] <jbossbot> jira [SHRINKWRAP-269] Shrinkwrap is hanging [Open (Unresolved) Bug, Major, Unassigned] https://issues.jboss.org/browse/SHRINKWRAP-269[22:01:23] <Nihility> jamezp: awesome thanks[22:01:38] <jamezp> NP, now let's see if it starts.[22:02:02] <asaldhan> bstansberry: that URL handling in the test case is not right. Ideally Shrinkwrap should just use the path from tccl. But it was not working on eclipse junit.[22:02:02] <Nihility> asaldhan: URL does not decode very well[22:02:37] <asaldhan> Nihility: that test setup was the only way I could get eclipse junit to work in the test. I put it in the email for ALR to figure and fix.[22:02:37] <jamezp> Nihility: Started with no issues.[22:02:58] <Nihility> jamezp: awesome thanks[22:03:05] <jamezp> No problem.[22:03:43] <Nihility> asaldhan: all shrinkwrap does is open the inputstream on the URL[22:04:29] <ALR> asaldhan: Yeah, we have a series of Asset impls[22:04:44] <ALR> So we'll take in any URL w/ a known handler[22:04:46] <asaldhan> ALR: if u can beef your examples in the user guide, it would be better.[22:04:57] <ALR> asaldhan: After API goes locky[22:05:17] <ALR> For now just JavaDocy[22:05:33] <ALR> How do I force git merge to take in the new thing?[22:07:08] <jbossbot> git [jboss-as] push master 5fa4b50.. jaikiran Initial support for interceptors in ejb-jar.xml[22:07:09] <jbossbot> git [jboss-as] push master 52f13cb.. jaikiran Fix processing of interceptors in ejb-jar.xml. Also added more tests around dd based interceptors[22:07:09] <jbossbot> git [jboss-as] push master 330d43f.. jaikiran Add a test for post-construct lifecyle of interceptors configured in ejb-jar.xml[22:07:09] <jbossbot> git [jboss-as] push master d781354.. jaikiran Fix processing of singleton beans in ejb-jar.xml and add test coverage for the same[22:07:09] <jbossbot> git [jboss-as] push master URL: http://github.com/jbossas/jboss-as/compare/0d8dbd3...d781354[22:07:53] *** Jaikiran has quit IRC[22:08:02] *** sannegrinovero has quit IRC[22:08:03] <dmlloyd> (yay at *yet another* unique implementation of lifecycle callbacks)[22:11:04] <wolfc> hanging again in ApplicationPathOverrideIntegrationTestCase[22:12:09] <dmlloyd> wolfc: can you just rebase and push it out? I'll test it locally[22:12:45] <wolfc> (09:27:58 PM) wolfc: dmlloyd, finally after 12 minutes of compiling. https://github.com/wolfc/jboss-as/commits/ejb3-ws I haven't tested the final commit.[22:13:25] <dmlloyd> conflicts with Jaikiran's branch[22:13:31] <dmlloyd> can you re-rebase it?[22:13:47] <wolfc> sure, give me an hour or so :-)[22:15:37] <dmlloyd> what is wrong with your computer[22:16:00] <Nihility> ROFL[22:16:10] * dmlloyd has other things he would like to get to today[22:16:30] <wolfc> Jaikiran changed the whitespaces, great :-)[22:16:52] <ALR> Do we really care about checkstyle?[22:17:08] <ALR> Trailing whitespace in comments, go f off. :)[22:18:10] <Nihility> haha enable trailing WS filtering in eclipse[22:18:22] <Nihility> ALR: we care because it lessens conflictss[22:18:56] <wolfc> apparently not :-)[22:19:00] <ALR> Hehe.[22:19:11] <ALR> I think in practice the formatters aren't 100%[22:19:20] <Nihility> i was just going to say, just like wolfc is seeing right now![22:19:27] <wolfc> dmlloyd, I pushed a new one https://github.com/wolfc/jboss-as/commits/ejb3-ws[22:19:45] <ALR> My mantra this week is: Complain, Motherf*cker![22:19:57] <wolfc> Luckily the week is almost done :-)[22:20:13] <ALR> Nah, I'm on through the weekend. Monday and most of Tues were blown.[22:20:16] <dmlloyd> not for me[22:20:30] <dmlloyd> I need to figure out wtf we're doing before the monday meeting[22:20:49] <Nihility> oh btw i might not be able to make that[22:20:52] <Nihility> its at 7am right[22:20:55] <ALR> We should write simple wrappers above Spring and call it a product.[22:20:56] <dmlloyd> yeah[22:21:06] <ALR> April Fools to you guys, we stole your stuff.[22:21:09] <Nihility> but you dont need me[22:21:33] <dmlloyd> we can get in to plenty of trouble on our own[22:21:36] <smarlow> wolfc: which kernel bug are you hitting?[22:21:48] <smarlow> I hit a xorg talking all the cpu last year bug[22:21:56] <Nihility> oh crap[22:22:02] <Nihility> did we ever merge thomas' change[22:22:07] * dmlloyd shruf[22:22:09] <dmlloyd> g[22:22:12] *** jpederse has quit IRC[22:22:29] <wolfc> smarlow, it manifests itself as https://bugzilla.kernel.org/show_bug.cgi?id=14426[22:23:54] <wolfc> but I think that's just a symptom. Although the hpet code might truly be broken.[22:24:32] <smarlow> wolfc: yup, different than I hit. I ran with a custom kernel for a while but not long enough to test the management console bug that should be fixed now (used to barf with long kernel names)[22:24:39] <mmoyses> bstansberry: do you have a few minutes to help me with the descriptions?[22:24:56] <bstansberry> sure[22:25:03] <mmoyses> pm[22:25:35] *** jclingan has joined #jboss-as7[22:28:51] <jbossbot> git [jboss-as] push master 8d74b56.. Carlo de Wolf Be more informative when an interceptor can not be found[22:28:52] <jbossbot> git [jboss-as] push master 5aebfc6.. Carlo de Wolf Expose componentInterceptor[22:28:52] <jbossbot> git [jboss-as] push master 3ee7bd2.. Carlo de Wolf Create interceptor chain for component methods as well[22:28:52] <jbossbot> git [jboss-as] push master ae0e8e5.. Carlo de Wolf Added SessionBeanComponent.invoke[22:28:52] <jbossbot> git [jboss-as] push master a9541a5.. Carlo de Wolf Use SessionBeanComponent.invoke[22:28:52] <jbossbot> git [jboss-as] push master 305ac9e.. Carlo de Wolf Test Singleton EJB WS[22:28:53] <jbossbot> git [jboss-as] push master 0aa7122.. Carlo de Wolf Exposing component methods hack to make WS test pass[22:28:53] <jbossbot> git [jboss-as] push master 340cde9.. Carlo de Wolf Remove iteration over super class public methods[22:28:53] <jbossbot> git [jboss-as] push master URL: http://github.com/jbossas/jboss-as/compare/d781354...340cde9[22:32:34] *** adietisheim has quit IRC[22:33:03] <jbossbot> git [jboss-as] push master 35e17b6.. Thomas Diesler [JBAS-9003] Add org.osgi.util.tracker to framework sources[22:33:04] <jbossbot> jira [JBAS-9003] Two modules include "org.osgi:org.osgi.compendium" maven artifact [Resolved (Done) Bug, Major, Thomas Diesler] https://issues.jboss.org/browse/JBAS-9003[22:33:04] <jbossbot> git [jboss-as] push master ac4a77f.. Thomas Diesler Upgrade to jbosgi-framework-1.0.0.Alpha23[22:33:04] <jbossbot> git [jboss-as] push master URL: http://github.com/jbossas/jboss-as/compare/340cde9...ac4a77f[22:33:23] <Nihility> and hopefully thats it[22:33:30] <Nihility> jamezp: would you mind rebuilding one last time[22:33:45] <jamezp> Nihility: Not at all.[22:33:54] <Nihility> great[22:34:07] <Nihility> bstansberry: on my windows box the deploy api test fails due to a timeout[22:34:30] <Nihility> bstansberry: i didnt debug it but i suspect because its a slow system[22:34:43] <Nihility> testDeploymentFileApi[22:34:49] <bstansberry> hmm[22:35:06] <bstansberry> it waits i think 20 secs[22:35:11] <Nihility> junit.framework.AssertionFailedError: Timed out waiting for registration/unregistration[22:35:17] <bstansberry> so it would have to be really really slow[22:41:01] <bstansberry> Nihility: there used to be a race in that test, that I thought I fixed a few weeks back. maybe not[22:41:16] <Nihility> i changed it to 40 sseconds[22:41:20] <Nihility> i dont think its working[22:41:21] <Nihility> :)[22:41:50] *** mmoyses has quit IRC[22:41:54] <rmaucher> hum, the amount of last minute commits is staggering :D[22:42:18] <bstansberry> race == test puts down marker files for step 2 be fore step 1 is complete, so step one deletes them, and then step 2 doesn't happen[22:42:28] <bstansberry> i'll look again[22:42:36] <Nihility> well this is the Api case[22:42:39] <Nihility> not the FS[22:42:43] <bstansberry> oh[22:42:59] <Nihility> it doesnt seem to affect jamezp[22:43:12] <bstansberry> time to fire up my windoze laptop[22:43:16] <jamezp> Not yet, but I do have a test failure.[22:43:39] <Nihility> rmaucher: don't worry we didnt touch web :) [yet cough][22:45:07] * jamezp is going to try a clean, then build.[22:45:44] <rmaucher> right ;)[22:46:13] <Nihility> actually max was asking for a splash page[22:46:21] <Nihility> and im like that will have to wait[22:46:52] *** alesj has joined #jboss-as7[22:47:25] <Nihility> alesj: are you on indows?[22:47:48] <Nihility> i think he switched to mac[22:47:55] <Nihility> its kind of funny because back in the day[22:48:01] <bstansberry> Nihility: you have a test report?[22:48:08] <Nihility> it was impossible to see anyone not using doz[22:48:23] *** jfd has quit IRC[22:48:24] <Nihility> ill pastebin it[22:48:39] <bstansberry> the server log too please[22:49:04] <bstansberry> i.e. the ZZZZ-output.txt[22:50:06] <wolfc> Yes! it has passed integration tests.[22:50:22] <Nihility> http://fpaste.org/Ozop/[22:50:32] <Nihility> log acts like it worked[22:50:50] <wolfc> I'm going to do check my bios and do a hardware check. This is silly.[22:50:57] <Nihility> http://fpaste.org/3Frd/[22:51:13] <Nihility> the error isnt so useful[22:51:17] <Nihility> it just waits[22:51:20] <Nihility> so i might need to debug[22:51:28] <Nihility> if someone else gets a clean build on windows[22:51:35] <Nihility> ill sideline this for later[22:51:40] <Nihility> this is an old old box[22:51:46] *** wolfc has quit IRC[22:51:46] <jamezp> Mine is working now.[22:52:13] <Nihility> jamezp: so smoke tests all pass[22:52:15] <Nihility> ?[22:52:22] <jamezp> It's processing them now.[22:54:03] <Nihility> you know[22:54:07] <Nihility> it could just be a port thing[22:55:35] * smarlow starting to look at JBAS-9040[22:55:36] <jbossbot> jira [JBAS-9040] Support for pluggable JPA providers [Open (Unresolved) Sub-task, Major, Scott Marlow] https://issues.jboss.org/browse/JBAS-9040[22:56:26] <Nihility> hmm i should see other failures though[22:56:33] <jamezp> Nihility: Build successfully for me.[22:56:38] <Nihility> ok[22:56:45] <Nihility> awesome[22:56:45] <jamezp> or built even.[22:56:51] <bstansberry> Nihility: do you have the xml report for that case?[22:56:59] <Nihility> yes[22:57:05] <bstansberry> my assumption is the server log is showing the success of the other two tests[22:58:52] <ALR> Aaaaaand I ditched hours of work by resetting --hard[22:59:07] *** wolfc has joined #jboss-as7[22:59:07] *** wolfc has quit IRC[22:59:51] <jamezp> Starts on Windows too. Looks like there is a new info message for this special occasion :-)[23:01:35] *** torben has quit IRC[23:01:50] <Nihility> http://fpaste.org/JcWd/[23:02:19] <Nihility> jamezp: hhaha you see the code name then huh[23:02:27] <Nihility> ALR: git stash is your friend[23:02:33] <jamezp> Ha, I do indeed.[23:02:39] <ALR> Nihility: I know.[23:02:44] <ALR> Been stashing and popping a lot[23:02:48] <ALR> But working with like 7 repos.[23:02:51] <ALR> And lost track.[23:02:54] <ALR> Too many terminals[23:03:10] <ALR> What *should* have been my friend:[23:03:13] <ALR> Frequent commits.[23:03:21] <Nihility> yeah you can get the commits back[23:03:23] <Nihility> if you did any[23:03:25] <ALR> Then interactive rebase squash later[23:03:33] <jamezp> IDEA has the best git support IMO. Makes it easier to not get confused.[23:03:53] <Nihility> jamezp: havew you tried the git prompt yet?[23:04:10] <ALR> Actually I think the hours was in figuring *what* to do. Re-doing it is much faster[23:04:33] <jamezp> Nihility: Not yet. I've just been amazed I hardly ever have to go to the command line.[23:05:32] <Nihility> ALR: it always is[23:08:14] <dmlloyd> alr, even when you do a hard reset you can still restore using the lost-and-found feature[23:08:28] <dmlloyd> unless you never committed that is :)[23:08:33] <ALR> dmlloyd: Never committed[23:08:38] <ALR> And when I do that, I do:[23:08:41] <ALR> git reflog[23:08:45] <ALR> and then something like[23:08:55] <ALR> git reset --hard HEAD{2} or whatever[23:09:04] <ALR> lost-and-found easier?[23:09:22] <dmlloyd> never had to use it because I use the prompt feature, so I usually know what revision I just abandoned[23:09:28] <dmlloyd> making it easy to go back to it[23:09:45] <ALR> Ah. Yeah it's quite a toolset. But nothing can save uncommitted stuff.[23:10:47] <dmlloyd> that's where IDEA comes in :)[23:10:52] <dmlloyd> local history![23:13:48] <ALR> Eclipse has it too. But I had too many modules to sort through[23:13:50] <ALR> Easier to redo.[23:14:25] <ALR> You guys act like IDEA is the World's First IDE. :)[23:14:43] <ALR> Just to spite you I'm going back to CafeBuilder.[23:14:48] <stuartdouglas> morning[23:14:49] <ALR> Or whatever it was called.[23:14:58] <ALR> stuartdouglas: Morning.[23:15:12] <jamezp> I just started using it last week. It's way better than eclipse IMO.[23:16:03] <jamezp> stuartdouglas: Morning.[23:17:58] <ALR> Maven Support is better for sure. Though m2e has improved so much over the years.[23:18:17] <ALR> And as I told Jason yesterday, I can't get used to the LnF. I'm so superficial.[23:18:45] <jamezp> I personally think NetBeans has the best Maven support, but the worst git support.[23:20:10] <jamezp> I can't stand m2e. It constantly freezes eclipse for me and turns my MacBook into a heater.[23:20:38] <stuartdouglas> m2e has improved a lot, but it is still terrible :-)[23:20:51] <Nihility> stuartdouglas: hey does RF work with ejb?[23:20:59] <Nihility> stuartdouglas: just wondering after the one million patchess[23:21:35] <stuartdouglas> RF?[23:21:45] <Nihility> oops[23:21:51] <Nihility> RE[23:21:52] <Nihility> resteasy[23:21:59] <Nihility> jax-rs[23:22:03] *** wolfc has joined #jboss-as7[23:22:03] *** ChanServ sets mode: +v wolfc[23:22:09] <stuartdouglas> possibly :-)[23:22:16] <bstansberry> :)[23:22:17] <stuartdouglas> It has not been tested yet[23:22:55] <stuartdouglas> but I did port that bit of the integration from AS6, so it is possible in theory :-)[23:23:11] <Nihility> hahaha[23:23:16] <Nihility> ok just wondering[23:23:29] <Nihility> ok guys[23:23:36] <Nihility> im going to do a pre release tag[23:23:49] <Nihility> until brian's build comes back successful[23:24:03] <bstansberry> 1/2 the internet is downloaded[23:24:11] <Nihility> haha[23:24:29] *** alesj has quit IRC[23:24:59] <stuartdouglas> JAX-RS does work with CDI though[23:25:14] <stuartdouglas> and the EJB support is only needed for non-cdi deployments[23:25:26] <Nihility> ah ok[23:25:29] <stuartdouglas> otherwise the CDI integration takes over and gets the EJB's from the BM[23:25:49] <stuartdouglas> the CDI stuff even has an integration test :-)[23:30:23] <ALR> dmlloyd: I'm not making up this hang. I'll look into it later, but there's definitely an issue w/ the controller test.[23:32:49] <stuartdouglas> There is a known issue with the current version of MSC that is fixed in the next (unreleased) version that can occasionally cause a hang[23:33:20] <stuartdouglas> but I have only seen it once, so it is probably unlikely that it is causing your problem[23:35:55] * ALR is just well-hung.[23:36:43] <stuartdouglas> If you see a situation where services claim to have missing deps even though the deps are installed, then you have hit the msc bug[23:39:00] *** wolfc has quit IRC[23:40:27] <ALR> stuartdouglas: OK, cool. I think this is lower-level.[23:40:39] <ALR> stuartdouglas: Though I did see a misssing deps thing yesterday.[23:40:57] <ALR> Which went away alongside another XML parsing problem which was fixed by upgrading the JDK[23:41:20] <ALR> JBAS-9010[23:41:22] <jbossbot> jira [JBAS-9010] Provide more comprehensive error messages for missing dependencies [Open (Unresolved) Task, Blocker, David Lloyd] https://issues.jboss.org/browse/JBAS-9010[23:41:27] <ALR> Oops[23:41:31] <ALR> JBAS-9210[23:41:32] <jbossbot> jira [JBAS-9210] Test Failures in SUn JDK 1.6.0_u20 [Open (Unresolved) Bug, Major, Andrew Rubinger] https://issues.jboss.org/browse/JBAS-9210[23:41:34] <ALR> That one.[23:52:12] *** jclingan1 has joined #jboss-as7[23:53:47] *** jpearlin has joined #jboss-as7[23:55:26] *** jclingan has quit IRC[23:58:37] *** kkhan has quit IRC[23:59:22] *** jclingan1 has quit IRC