[00:17:54] <Tashtego> damn the embedded glassfish version i am using has eclipse link v2.0.0 which doesnt yet support typed named queries... ^^
[00:18:56] *** jganoff has joined #jbosstesting
[00:28:06] *** pmuir has quit IRC
[00:49:34] *** jeand has quit IRC
[00:56:03] <Tashtego> good night everyone...
[00:56:09] *** Tashtego has quit IRC
[00:56:14] *** ldimaggi_ has joined #jbosstesting
[01:00:03] *** ALR has quit IRC
[01:17:02] *** ldimaggi_ is now known as ldimaggi_home
[01:17:35] *** lightguard_jp has quit IRC
[01:39:21] *** bobmcw has quit IRC
[01:39:36] *** bobmcw has joined #jbosstesting
[01:40:44] *** bobmcw_ has joined #jbosstesting
[01:40:52] *** bobmcw_ has quit IRC
[01:41:00] *** bobmcw has quit IRC
[01:41:10] *** bobmcw has joined #jbosstesting
[01:41:42] *** mgoldmann has quit IRC
[01:41:48] *** aslak has quit IRC
[01:45:41] *** michaelschuetz has joined #jbosstesting
[02:39:24] *** rruss has joined #jbosstesting
[02:42:19] *** rruss has quit IRC
[02:49:40] *** jganoff has quit IRC
[03:11:32] *** michaelschuetz has quit IRC
[05:08:19] *** ldimaggi_home has quit IRC
[07:22:37] *** jdlee has quit IRC
[07:23:58] *** jdlee has joined #jbosstesting
[07:23:58] *** jdlee has joined #jbosstesting
[07:37:55] *** oskutka has joined #jbosstesting
[07:51:34] *** ge0ffrey has joined #jbosstesting
[07:52:55] *** aslak has joined #jbosstesting
[07:53:23] <hatchetman82> hey aslak
[07:53:40] <aslak> hatchetman82, good morning
[07:54:35] <hatchetman82> gm. short question - do you split *.war builds into 2 (code and war wrapper similar to jca builds) or do you just have a module with packaging = war
[07:54:36] <hatchetman82> ?
[07:55:49] <aslak> hatchetman82, normally just war packaged
[08:04:56] <hatchetman82> aslak - ever have a module that depends on code thats in such a war module ? the maven compiler plugin isnt smart enough to put the *.war on the compilation class path for me :-(
[08:06:15] <aslak> hatchetman82, no never.. ;)
[08:06:41] <hatchetman82> yeah, i never had the need either
[08:06:55] <hatchetman82> but its amazing what a team of people doing random stuff will find :-)
[08:07:05] <aslak> i can see EE6 with ejbs packaged in wars create the need tho
[08:07:47] <aslak> or well.. the interfaces for remote ejbs should then be packaged seperatly i guess, which is what you should depend on..
[08:08:00] <hatchetman82> in our case the arq tests module needs to depend on classes from the *.war (because in arq we're trying to create our own *.war for @Deployment, so we just need the classes)
[08:08:07] <hatchetman82> and currently it doesnt compile
[08:08:30] <aslak> .war is not understood by javaSE
[08:08:39] <aslak> if it is on classpath or not
[08:08:45] <hatchetman82> tru
[08:08:48] <hatchetman82> true*
[08:09:05] <hatchetman82> ill split it into 2 - a code module and a wrapper module that packages the code module as a .war - like JCA builds
[08:09:54] <aslak> hatchetman82, any reason why the tests are not part of the war?
[08:11:05] <hatchetman82> we have arq tests in a single separate module
[08:11:22] <hatchetman82> at first it was because everyone but me were using maven 2
[08:11:46] <hatchetman82> and now its because having jboss deps in a module slows maven down by 30-40 seconds
[08:11:50] <hatchetman82> per module
[08:11:53] <hatchetman82> we have ~20
[08:12:03] <nickarls> aslak: any theories why all my datasources go http://pastebin.com/WCq0ky9T when started as embedded as 6?
[08:12:54] <aslak> Caused by: java.lang.NoSuchFieldError: urlDelimiter
[08:12:54] <aslak>         at org.jboss.resource.adapter.jdbc.local.LocalManagedConnectionFactory.setConnectionURL(LocalManagedConnectionFactory.java:114) [jboss-as-connector-6.0.0.Final.jar:6.0.0.Final]
[08:13:40] <hatchetman82> either you didnt specify jboss_home properly or youre lacking several extra classpath items ?
[08:14:45] <nickarls> I have http://pastebin.com/euBpeHzA in arquillian.xml in the test resources
[08:15:18] <nickarls> wait, wrong namespace?
[08:15:37] <aslak> yea
[08:16:35] <aslak> urn:arq:org.jboss.arquillian.container.jbossas.embedded_6
[08:16:44] <nickarls> the internet is nice, you can copypaste stuff from it without reading
[08:16:49] <aslak> or are you on managed?
[08:17:02] <nickarls> urn:arq:org.jboss.arquillian.container.jbossas.managed_6
[08:17:09] <aslak> :)
[08:17:35] <nickarls> I needed the port set to 80 so I googled around for the first match
[08:17:56] <nickarls> I didn't find any sample arqullian.xml in the docs
[08:18:33] <aslak> http://docs.jboss.org/arquillian/reference/latest/en-US/html_single/#containers.configuration
[08:18:37] <nickarls> hmm. same result though
[08:20:00] <aslak> looking at the code, it looks like your getting a different v. of the Base class BaseWrapperManagedConnectionFactory
[08:20:47] <hatchetman82> the internet is nice, you can copypaste stuff from it without reading   <---- if i had a dollar for every time someone here did this ....
[08:21:47] <hatchetman82> its easy to tell in my case though. they usually commit canned maven projects and then i suddenly have a dep on junit 3.8.1 (thats what the maven archetype generates)
[08:28:12] <nickarls> strange, the errors are the same even without arquillian.xml
[08:28:26] <aslak> nickarls, i think it's a dependency issue
[08:29:04] <aslak> nickarls, do you have jboss-as-connector-6.0.0.Final in there?
[08:29:47] <nickarls> no
[08:30:02] <aslak> and dependency:tree in maven 3+ doesn't work so.. so.. try to print the surefire classpath somehow
[08:32:59] <hatchetman82> mvn -X is the only way in maven 3
[08:33:16] <hatchetman82> not that tree doesnt work, its just that it uses the old (naven 2) reactor
[08:35:07] <nickarls> manged works fine but not embedded
[08:56:18] <nickarls> aslak: -X gave 670k of log, where should I start looking? ;-)
[08:56:29] <aslak> grep? .)
[08:56:50] *** lfryc has joined #jbosstesting
[08:58:19] <hatchetman82> working*
[08:58:22] <hatchetman82> nickarls: maven prints out the module name its owrking on
[08:58:25] <hatchetman82> grep over to the arq. module
[08:58:42] <hatchetman82> there's a build order table near the top that contains all the names
[09:00:00] *** jharting has joined #jbosstesting
[09:01:00] <nickarls> is "Test Classpath" of interest?
[09:16:10] <nickarls> BTW, is @DataSourceDefinition useable in tests if I add an EJB with it?
[09:20:22] <nickarls> BTW2, what code coverage plugins do you use? emma?
[09:26:28] <hatchetman82> nickarls: another thing to consider when running embedded
[09:26:50] <hatchetman82> you have the junit runner, running in a JVM started by either maven or your IDE or whatever.
[09:26:54] <hatchetman82> it has a classloader
[09:27:16] <hatchetman82> then it boots jboss
[09:27:24] <hatchetman82> which loads its own classloader tree
[09:27:51] <hatchetman82> so sometimes you'll have classloader issues with classes found in the original (junit) classloader that should have been packed inyour deployment
[09:28:18] <nickarls> so "managed" is recommended you get a clean classloader?
[09:34:20] *** mgoldmann has joined #jbosstesting
[09:34:54] *** maeste_afk is now known as maeste
[09:36:46] *** mgoldmann is now known as mgoldmann|away
[09:41:13] <nickarls> there was no maven integration in ShrinkWrap yet, right?
[09:41:20] <aslak> nickarls, not released
[09:45:08] <nickarls> surprisingly, I get CNFE for emma-enchanched  bytecode
[09:51:16] *** wolfc has joined #jbosstesting
[09:58:47] *** Jaikiran has joined #jbosstesting
[09:58:52] *** Jaikiran has joined #jbosstesting
[10:00:44] <hatchetman82> nickarls: try packaging emma in your @Deployment
[10:01:07] <hatchetman82> and yess, embedded has been a giant classloader hell for me, but its still the "cleanest" approach :-)
[10:01:09] <hatchetman82> yes*
[10:02:40] <nickarls> hey, @DataSourceDefinition appears to work. Although it appears to bind to some java:internal/test/test/DB/LTK/NikDS
[10:03:50] <nickarls> with @DataSourceDefinition(name = "java:global/LTK/NikDS" on a stateless EJB
[10:05:42] *** michaelschuetz has joined #jbosstesting
[10:27:31] <nickarls> hatchet: what? classloader hell? the marketing drones told med arquillian fixed all that! ;-)
[10:35:35] *** aslak has quit IRC
[10:41:31] *** vtunka has joined #jbosstesting
[11:01:34] *** mgoldmann|away has quit IRC
[11:20:25] *** michaelschuetz1 has joined #jbosstesting
[11:21:58] *** michaelschuetz has quit IRC
[11:44:35] *** Jaikiran has quit IRC
[11:50:22] *** Jaikiran has joined #jbosstesting
[12:04:35] <nickarls> hmm, am I supposed to be able to do DataSource dataSource = (DataSource) new InitialContext().lookup("java:/DefaultDS"); if my jndi.properties are OK? I'm getting "DefaultDS not bound"
[12:20:12] *** mgoldmann|away has joined #jbosstesting
[12:21:30] *** bgeorges has joined #jbosstesting
[12:25:42] *** pmuir has joined #jbosstesting
[12:25:42] *** pmuir has quit IRC
[12:25:42] *** pmuir has joined #jbosstesting
[12:54:27] <hatchetman82> how can i enumerate what jarsare on the CP if i was invoked with a classpath jar only ?
[12:55:00] <hatchetman82> its a common trick on windows, where if the CP is too long to fit on a command line (win has a limit on that) you create a jar with all the CP jars in its manifest
[12:55:11] <hatchetman82> so "java.class.path" is only a single jar :-)
[12:56:43] <nickarls> that would be nice as the cp runs out pretty quick with incontainer deps
[13:06:51] <hatchetman82> surefire does it for you
[13:07:20] <hatchetman82> you can create one of those yourself pretty easily. its an otherwise empty jar with a manifest
[13:07:34] <hatchetman82> i think jboss use the same trick with their jboss-client-all.jar on windows
[13:07:38] <nickarls> I just notice that running in-container tests from eclipse with "run as" usually gives me the #87
[13:10:06] <hatchetman82> ?
[13:14:48] <nickarls> if I activate e.g. a jbossas-managed-6 maven profile and do "run as junit" test on a win host I get a "createprocess error 87 the parameter is incorrect" which usually means the cp has become too long
[14:15:14] *** jeand has joined #jbosstesting
[14:17:32] *** balunasj_ has joined #jbosstesting
[14:18:08] *** balunasj_ has left #jbosstesting
[14:18:21] *** bgeorges has quit IRC
[14:31:13] *** Jaikiran has quit IRC
[14:32:19] <hatchetman82> nickarls: (apologies for the delayed response, was in a meeting) doesnt eclipse do this automatically ?
[14:32:31] <hatchetman82> intellij does ... join the dark side ;-)
[14:34:17] * hatchetman82 doesnt know eclipse all that well, but this kind of option has to be there
[14:39:17] *** ldimaggi_ has joined #jbosstesting
[14:53:34] *** jharting has quit IRC
[14:59:56] *** michaelschuetz1 has quit IRC
[15:00:39] *** michaelschuetz has joined #jbosstesting
[15:04:00] *** lightguard_jp has joined #jbosstesting
[15:06:26] *** michaelschuetz has quit IRC
[15:07:08] *** bgeorges has joined #jbosstesting
[15:07:10] *** michaelschuetz has joined #jbosstesting
[15:34:03] *** Jaikiran has joined #jbosstesting
[15:51:40] *** jharting has joined #jbosstesting
[15:57:11] *** jharting has quit IRC
[16:05:19] *** ALR has joined #jbosstesting
[16:16:54] *** lightguard_jp has quit IRC
[16:26:27] *** jdlee has quit IRC
[16:33:25] *** pmuir has quit IRC
[16:37:15] *** jdlee has joined #jbosstesting
[16:37:15] *** jdlee has joined #jbosstesting
[16:42:45] *** lightguard_jp has joined #jbosstesting
[16:49:55] *** oskutka has quit IRC
[16:57:51] *** rruss has joined #jbosstesting
[16:58:48] *** pmuir has joined #jbosstesting
[16:58:58] *** pmuir has quit IRC
[16:58:58] *** pmuir has joined #jbosstesting
[17:03:34] *** pmuir has quit IRC
[17:06:13] *** pmuir has joined #jbosstesting
[17:06:13] *** pmuir has quit IRC
[17:06:13] *** pmuir has joined #jbosstesting
[17:22:15] *** jbossbot has quit IRC
[17:23:07] *** jbossbot has joined #jbosstesting
[17:37:46] *** pmuir has quit IRC
[17:40:29] *** Jaikiran has quit IRC
[17:51:44] *** maeste has quit IRC
[17:57:00] *** bgeorges has quit IRC
[18:03:08] *** pmuir has joined #jbosstesting
[18:03:08] *** pmuir has quit IRC
[18:03:08] *** pmuir has joined #jbosstesting
[18:06:54] *** maeste has joined #jbosstesting
[18:07:40] *** pmuir has quit IRC
[18:08:29] *** pmuir has joined #jbosstesting
[18:11:04] *** bgeorges has joined #jbosstesting
[18:19:09] *** kpiwko has quit IRC
[18:32:22] *** michaelschuetz has quit IRC
[18:39:39] *** vtunka has quit IRC
[18:51:41] *** lfryc has quit IRC
[18:52:36] *** lfryc has joined #jbosstesting
[18:53:51] *** lfryc has quit IRC
[19:11:26] *** aslak has joined #jbosstesting
[19:11:26] *** aslak has quit IRC
[19:11:26] *** aslak has joined #jbosstesting
[19:44:31] *** bgeorges has quit IRC
[19:52:13] *** ge0ffrey has quit IRC
[20:11:45] *** jharting has joined #jbosstesting
[20:21:35] *** bgeorges has joined #jbosstesting
[20:58:28] *** oskutka has joined #jbosstesting
[21:02:51] *** oskutka has quit IRC
[21:04:13] *** bgeorges has quit IRC
[21:04:27] *** bgeorges has joined #jbosstesting
[21:05:00] *** rruss has quit IRC
[21:31:06] *** Tashtego has joined #jbosstesting
[21:36:28] <Tashtego> hi
[21:36:44] *** ldimaggi_ has quit IRC
[21:52:48] *** bgeorges has quit IRC
[21:54:36] *** pmuir has quit IRC
[21:56:34] *** pmuir has joined #jbosstesting
[21:56:34] *** pmuir has quit IRC
[21:56:34] *** pmuir has joined #jbosstesting
[21:57:53] *** pmuir has quit IRC
[21:57:58] *** tcunning has joined #jbosstesting
[22:00:00] *** pmuir has joined #jbosstesting
[22:00:39] *** pmuir has quit IRC
[22:01:29] *** pmuir has joined #jbosstesting
[22:01:29] *** pmuir has quit IRC
[22:01:29] *** pmuir has joined #jbosstesting
[22:02:44] *** kpiwko has joined #jbosstesting
[22:03:33] *** pmuir has quit IRC
[22:04:22] *** pmuir has joined #jbosstesting
[22:04:22] *** pmuir has quit IRC
[22:04:22] *** pmuir has joined #jbosstesting
[22:06:07] *** pmuir has quit IRC
[22:09:31] <nickarls> howdy-ho
[22:15:25] *** wolfc has quit IRC
[22:20:24] *** pmuir has joined #jbosstesting
[22:20:24] *** pmuir has quit IRC
[22:20:24] *** pmuir has joined #jbosstesting
[22:21:49] *** pmuir has quit IRC
[22:26:52] <tcunning> is there an example of how to use multiple deployments with Arquillian?  i want to deploy a deployer + a deployment.
[22:29:00] <aslak> tcunning, not supported in release yet
[22:29:26] <tcunning> right - i cloned the trunk at github and built it with the options you gave me
[22:30:06] <aslak> tcunning, trunk is no good, build https://github.com/aslakknutsen/arquillian/tree/the_bigger_picture
[22:38:03] *** jeand has quit IRC
[22:39:00] <tcunning> do i just do multiple createXXXArchive methods?
[22:40:55] *** Tashtego has quit IRC
[22:43:05] <nickarls> aslak: BTW, I'm playing around with dbunit integration like @Data(src="foo.xls", ds="java:/DefaultDS") for test. Shouldn't I technically be able to lookup that DS in my junit @Rule if my jndi.properties are OK? I just get a namenotboundexception...
[22:43:28] <aslak> tcunning, sorry, example.. http://arquillian.pastebin.com/nKcsQtyu
[22:44:09] <tcunning> aslak : thanks!
[22:44:15] <aslak> nickarls, @Rules are not handled currently, so they are executed everywhere.. :)
[22:44:31] <aslak> but yea, in a embedded container it should work
[22:45:29] *** lfryc has joined #jbosstesting
[22:45:32] <nickarls> aslak: yeah, I have one @Rule and then just check for precence of the @Data annotation on the frameworkmethod
[22:46:57] <aslak> nickarls, DataSources aren't remotable tho..
[22:47:15] <nickarls> remotable?
[22:47:45] <aslak> i was hoping you would catch that.. hehe :)
[22:48:02] <aslak> remote look up capable..
[22:48:14] <aslak> so jndi.properties are irrelevant
[22:49:22] <nickarls> but if the DS is on the server and the test runs incontainer, it's not remote. or?
[22:49:31] <aslak> you don't use jndi.properties with a embedded container, and you can't lookup a DataSource from a remote client
[22:49:42] <aslak> @Rule is executed both placed
[22:50:44] <nickarls> this new learning amazes me
[22:51:10] <nickarls> if I add an EJB with a @DataSourcDefinition is that local DS still no-go?
[22:51:54] <nickarls> or do I just have to do DriverManager.getConnection and work from there?
[22:54:08] *** michaelschuetz has joined #jbosstesting
[22:57:26] * nickarls is falling asleep on the keyboard, please speak to the log if you have any insights. goodnight ;-)
[23:06:00] *** lightguard_jp is now known as lightguard_jp|aw
[23:12:01] <kpiwko> evening aslak, do you have a minute?
[23:12:19] <aslak> kpiwko, sure
[23:12:52] <kpiwko> aslak, good news is we integrated Ajocado with Arquillian-Selenium today ;)
[23:13:15] <kpiwko> aslak, bad news is I'm falling into nasty problem with shrinkwrap/arquillian
[23:14:08] <kpiwko> aslak, basically if I deploy an archive with arquillian, container raises an exception, if I deploy the very same archive manually, everything works
[23:14:40] <kpiwko> aslak, exception is similar to https://issues.jboss.org/browse/SEAMJMS-16
[23:14:42] <jbossbot> jira [SEAMJMS-16] Event to JMS to Event mapping behavior [Open (Unresolved) Bug, Major, Unassigned] https://issues.jboss.org/browse/SEAMJMS-16
[23:15:04] <kpiwko> aslak, e.g it seems like a bad classloader in container
[23:16:14] <kpiwko> aslak, here's the output http://pastebin.com/8aYV78Wb
[23:16:21] <aslak> kpiwko, with arquillian you mean arquillian for incontainer testing, or just using arquillina ?
[23:17:00] <kpiwko> aslak, just using arquillian, it is as_client test
[23:17:37] <aslak> hmm, arquillina, arquillians girlfriend
[23:17:43] <aslak> :)
[23:17:48] <kpiwko> :)
[23:18:34] <aslak> but same deployment works if not done via arq?
[23:18:50] <kpiwko> aslak, the strange thing is I though it could be shrinkwrap/package related but if I let sw write archive to a file and then deploy manually, it works
[23:19:06] <kpiwko> aslak, exactly
[23:19:40] <aslak> kpiwko, so.. could you try to extract a ProfileService client, outside of arq and see if that also results in failure?
[23:20:55] <aslak> kpiwko, which container you using btw?
[23:21:00] <kpiwko> aslak, sure, I can try that
[23:21:08] <kpiwko> aslak, jboss-as 6 cr1
[23:21:20] <aslak> cr1?
[23:21:22] <kpiwko> I can't go to final unfortunately
[23:21:28] <aslak> aa ok
[23:21:32] <aslak> remote ?
[23:21:38] <kpiwko> yeah
[23:21:54] <aslak> latest the_bigger_picture ?
[23:22:01] <kpiwko> let me verify
[23:22:20] <kpiwko> yes, it's the latest
[23:23:05] *** mgoldmann|away has quit IRC
[23:23:06] <aslak> if the latest, i just removed the disabling of the HD DeploymentScanner..  so the deployment via ProfileSrevice will be redeployed during test executed and cause a lot of strange issues
[23:23:26] <aslak> kpiwko, you can try to manualy turn of the HDScanner in jmx-console and see if you get the same
[23:24:15] <aslak> i believe this issues was first fixed in .Final
[23:24:44] *** ldimaggi_ has joined #jbosstesting
[23:24:44] *** ldimaggi_ is now known as ldimaggi_home
[23:36:46] <tcunning> anyone ever seen a noclassdeffounderrror on AbstractUnboundContext when using the weld-se-embedded profile?
[23:37:19] <tcunning> http://pastie.org/1550700
[23:39:03] <aslak> tcunning, which weld-se-embedded arq v? 1 or 1.1 and which weld v are you running it on?
[23:40:48] <tcunning> arquillian-weld-se-embedded-1, weld-core-bom 1.0.1-SP1
[23:41:47] <aslak> hmm..
[23:41:54] <aslak> what deps do you have in the profile?
[23:43:11] <aslak> https://github.com/arquillian/arquillian/blob/1.0.0.Alpha4/examples/junit/pom.xml#L84
[23:43:43] *** jharting has quit IRC
[23:46:39] <tcunning> looks like it might be pulling in weld-se-core 1.1.0-SNAP
[23:47:15] <tcunning> that's from my stuff though
[23:47:23] <aslak> aa, right
[23:48:47] <aslak> maybe swap to weld-se-embedded-1.1 then. should support 1.1.0.Final in the_bigger_picture branch
[23:48:59] <aslak> i see it's only at CR1 in the current pom but
[23:51:44] <aslak> tcunning, got any docs / plans for Switchyeard Test ?
[23:52:20] <aslak> SwitchYard even
[23:52:33] *** ALR has quit IRC
[23:53:05] *** maeste has quit IRC
[23:53:23] <tcunning> we've almost got our deployer tied down, so we're hoping to get some arquillian tests up where we deploy a switchyard app and test through arquillian
[23:53:38] <aslak> tcunning, great!
[23:53:56] <aslak> tcunning, I don't need to tell you you should focus this around Arquillian then.. hehe
[23:54:11] <tcunning> heh - it's definitely the plan





top