Friday, 2015-11-20

projectodd-ciProject wildfly-swarm build #315: ABORTED in 1 hr 11 min:
projectodd-ci* Bob McWhirter: Clean up the test.00:08
projectodd-ci* Lance Ball: Expose topology changes as an EventSource that emits topologyChange events.00:08
*** jerr <jerr!> has joined #wildfly-swarm00:16
*** bobmcw <bobmcw!~bobmcw@redhat/jboss/bobmcw> has joined #wildfly-swarm00:42
*** ChanServ sets mode: +o bobmcw00:42
*** bobmcw <bobmcw!~bobmcw@redhat/jboss/bobmcw> has quit IRC (Ping timeout: 276 seconds)00:47
*** BenT15 <BenT15!> has quit IRC (Ping timeout: 255 seconds)01:01
*** BenT15 <BenT15!> has joined #wildfly-swarm01:03
*** BenT15 <BenT15!> has quit IRC (Ping timeout: 240 seconds)01:23
*** BenT15 <BenT15!> has joined #wildfly-swarm01:26
*** proddbot <proddbot!~irclj@> has quit IRC (Ping timeout: 240 seconds)02:24
*** BenT15 <BenT15!> has quit IRC (Ping timeout: 255 seconds)02:24
*** BenT15 <BenT15!> has joined #wildfly-swarm02:28
*** BenT15 <BenT15!> has quit IRC (Ping timeout: 244 seconds)02:33
*** BenT15 <BenT15!> has joined #wildfly-swarm02:52
*** BenT15 <BenT15!> has quit IRC (Ping timeout: 272 seconds)02:57
*** jerr <jerr!> has quit IRC (Ping timeout: 264 seconds)03:47
*** bobmcw <bobmcw!~bobmcw@redhat/jboss/bobmcw> has joined #wildfly-swarm04:45
*** ChanServ sets mode: +o bobmcw04:45
*** bobmcw <bobmcw!~bobmcw@redhat/jboss/bobmcw> has quit IRC (Ping timeout: 250 seconds)04:50
*** BenT15 <BenT15!> has joined #wildfly-swarm05:14
*** BenT15 <BenT15!> has quit IRC (Ping timeout: 260 seconds)05:18
*** BenT15 <BenT15!> has joined #wildfly-swarm06:14
*** BenT15 <BenT15!> has quit IRC (Ping timeout: 276 seconds)06:21
*** BenT15 <BenT15!> has joined #wildfly-swarm07:27
*** rbenevides <rbenevides!~rbenevid@redhat/jboss/RafaelBenevides> has joined #wildfly-swarm07:47
*** BenT15 <BenT15!> has quit IRC (Ping timeout: 250 seconds)07:49
*** rbenevid_ <rbenevid_!~rbenevid@redhat/jboss/RafaelBenevides> has joined #wildfly-swarm08:00
*** rbenevides <rbenevides!~rbenevid@redhat/jboss/RafaelBenevides> has quit IRC (Ping timeout: 276 seconds)08:00
*** hbraun <hbraun!~hbraun@pdpc/supporter/professional/hbraun> has joined #wildfly-swarm08:14
*** hbraun <hbraun!~hbraun@pdpc/supporter/professional/hbraun> has quit IRC (Client Quit)08:15
*** os890 <os890!> has joined #wildfly-swarm08:30
*** BenT15 <BenT15!> has joined #wildfly-swarm08:46
*** BenT15 <BenT15!> has quit IRC (Ping timeout: 244 seconds)08:50
*** rbenevid_ <rbenevid_!~rbenevid@redhat/jboss/RafaelBenevides> has quit IRC (Ping timeout: 272 seconds)09:04
*** rbenevides <rbenevides!~rbenevid@redhat/jboss/RafaelBenevides> has joined #wildfly-swarm09:04
*** rbenevides <rbenevides!~rbenevid@redhat/jboss/RafaelBenevides> has quit IRC (Quit: Computer has gone to sleep.)10:32
*** rbenevides <rbenevides!~rbenevid@redhat/jboss/RafaelBenevides> has joined #wildfly-swarm10:34
*** BenT15 <BenT15!> has joined #wildfly-swarm10:46
*** bobmcw <bobmcw!> has joined #wildfly-swarm10:48
*** bobmcw <bobmcw!~bobmcw@redhat/jboss/bobmcw> has joined #wildfly-swarm10:48
*** ChanServ sets mode: +o bobmcw10:48
*** BenT15 <BenT15!> has quit IRC (Ping timeout: 260 seconds)10:52
*** bobmcw <bobmcw!~bobmcw@redhat/jboss/bobmcw> has quit IRC (Ping timeout: 244 seconds)10:53
*** rbenevides <rbenevides!~rbenevid@redhat/jboss/RafaelBenevides> has quit IRC (Quit: Computer has gone to sleep.)10:58
*** agoncal <agoncal!> has joined #wildfly-swarm11:23
*** tcrawley-away is now known as tcrawley12:05
*** agoncal <agoncal!> has quit IRC (Ping timeout: 250 seconds)12:38
*** BenT15 <BenT15!> has joined #wildfly-swarm12:49
*** BenT15 <BenT15!> has quit IRC (Ping timeout: 240 seconds)12:53
*** agoncal <agoncal!> has joined #wildfly-swarm13:01
*** rbenevides <rbenevides!~rbenevid@redhat/jboss/RafaelBenevides> has joined #wildfly-swarm13:02
*** bbrowning_away is now known as bbrowning13:16
*** agoncal <agoncal!> has quit IRC (Ping timeout: 250 seconds)13:17
*** lance|afk is now known as lanceball13:25
*** kenfinnigan <kenfinnigan!~ken@redhat/jboss/kenfinnigan> has joined #wildfly-swarm13:39
*** agoncal <agoncal!> has joined #wildfly-swarm13:48
*** bobmcw <bobmcw!~bobmcw@redhat/jboss/bobmcw> has joined #wildfly-swarm13:51
*** ChanServ sets mode: +o bobmcw13:51
*** rbenevides <rbenevides!~rbenevid@redhat/jboss/RafaelBenevides> has quit IRC (Quit: Computer has gone to sleep.)13:53
*** bobmcw <bobmcw!~bobmcw@redhat/jboss/bobmcw> has quit IRC (Remote host closed the connection)13:56
*** rbenevides <rbenevides!~rbenevid@redhat/jboss/RafaelBenevides> has joined #wildfly-swarm13:56
*** bobmcw <bobmcw!> has joined #wildfly-swarm13:56
*** bobmcw <bobmcw!~bobmcw@redhat/jboss/bobmcw> has joined #wildfly-swarm13:56
*** ChanServ sets mode: +o bobmcw13:56
bbrowningbobmcw: it does appear the new swarm:start has a nasty habit of hanging if the app doesn't actually start13:57
bbrowningif I'm looking at the CI logs correctly13:57
bobmcwyah, there's that13:57
bobmcwwhich I can fix13:57
bobmcwand then figure out why it's not working13:57
bobmcwImma assume it's maven.repo.local in CI that's buggering us13:59
bobmcwmvn is weird14:00
bobmcwif I clean my org.wildfly.swarm.** from my M2_REPO and mvn clean14:00
bobmcwinteg-jaxrs cleans fine14:00
bobmcwby integ-messaging complains about missing the plugin14:00
bobmcwa) clean, damnit, who cares14:00
bobmcwb) why not jaxrs?14:00
bbrowningI actually have a local swarm build going with a maven.repo.local set14:00
bbrowningit's taking forever, of course14:01
bobmcwah, lemme know14:01
bobmcwI'm adding a 2-minute startup timeout14:02
bbrowningI love it when maven screws up and picks instead of central to download something14:02
bobmcw(just like the arq container; we can probably do some refactoring between it and the plugin14:02
bbrowningI get like 1.1KB/sec14:02
bobmcwbbrowning: I *think* that's related to <repository> in a pom it's resolving from14:02
bobmcwif it fetches a.pom from, and it mentions and needs b.pom, it only looks in jborg14:03
* bobmcw could be wrong; often is14:03
bbrowningmaybe google will mirror our repo too ;)14:03
bobmcwor looks in jborg first, finds it, and doesn't try central first14:03
bobmcwor something14:03
jbossbotgit [wildfly-swarm] push master dca5d8e.. Bob McWhirter Add a start timeout, so CI won't hang.14:05
jbossbotgit [wildfly-swarm] push master URL:
projectodd-ciStarting build #316 for job wildfly-swarm (previous build: ABORTED -- last SUCCESS #313 18 hr ago)14:05
projectodd-ciProject wildfly-swarm build #316: ABORTED in 45 sec:
projectodd-ciBob McWhirter: Add a start timeout, so CI won't hang.14:06
bobmcwImma abort that and add some debug14:06
jbossbotgit [wildfly-swarm] push master 7690699.. Bob McWhirter Add debug for CI.14:07
jbossbotgit [wildfly-swarm] push master URL:
projectodd-ciStarting build #317 for job wildfly-swarm (previous build: ABORTED -- last SUCCESS #313 18 hr ago)14:07
bobmcwbbrowning: fwiw, may be useful to copy M2_REPO to another dir, rm org.wildfly.swarm.** and then use that are maven.repo.local14:16
bobmcwso you don't have to download the entire world14:16
bobmcwand/or run a local nexus14:16
bbrowningyeah a local nexus would be awesome in this case14:16
bobmcwI had one, but it stopped working reliably14:17
bbrowningI'll let it continue going while I eat breakfast14:17
*** kenfinnigan <kenfinnigan!~ken@redhat/jboss/kenfinnigan> has quit IRC (Quit: Leaving.)14:18
*** jclingan <jclingan!~jclingan@redhat/jclingan> has joined #wildfly-swarm14:42
*** BenT15 <BenT15!> has joined #wildfly-swarm14:49
bbrowningbobmcw: everything passed with a brand new empty maven.repo.local14:54
bobmcwI've added debug so we'll see what CI actually is trying to do14:54
bbrowninga 364MB repo I might add14:55
bobmcwthat's fantastic!14:55
*** enebo <enebo!> has joined #wildfly-swarm14:56
*** BenT15 <BenT15!> has quit IRC (Ping timeout: 250 seconds)14:56
bobmcwI fired up my nexus and trying a fresh build also, but apparently my nexus is pert near empty14:56
*** BenT15 <BenT15!> has joined #wildfly-swarm14:56
*** agoncal <agoncal!> has quit IRC (Ping timeout: 250 seconds)15:00
bobmcwbbrowning: since we're deploying -SNAPSHOTs, until we decomission -examples, do you think going back to the normal M2_REPO would suffice?15:04
bobmcwand having -example use the bees snapshot repo?15:04
bbrowningnow that the timestamp vs snapshot problem is fixed probably so15:04
*** agoncal <agoncal!> has joined #wildfly-swarm15:08
lanceballso this kanban thing - I guess I should read up on how it works15:09
*** BenT15 <BenT15!> has quit IRC (Ping timeout: 264 seconds)15:34
projectodd-ciProject wildfly-swarm build #317: FAILURE in 1 hr 34 min:
projectodd-ciBob McWhirter: Add debug for CI.15:41
bobmcwat least it didn't hang!15:43
bobmcwnow, to debug!15:43
bbrowningmy local test was invalid - I need both a maven.repo.local specified AND a ~/.m2/repository that doesn't have any swarm bits in it15:43
bobmcwso, that's not using maven.repo.local15:44
bbrowningbut shouldn't it be finding that in the jar anyway?15:45
bobmcwthat's my new integs, which just wildfly-swarm:start15:46
bbrowningor is this not building a -swarm.jar?15:46
bobmcwwhich doesn't uberjar (yet)15:46
bbrowningso yeah you'll have to propagate jvm properties to the start15:46
bbrowninglike you did for arquillian tests or whatever15:46
bobmcwI've reproduced locally15:46
bobmcwwell, this is from within the core maven plugin launching it15:46
bobmcwnot having to propogate to any forked test process15:47
bobmcwwell, maybe15:47
bobmcwah, damnit.15:47
* bobmcw fixes15:47
*** BenT15 <BenT15!> has joined #wildfly-swarm15:49
bobmcwalso, -DskipTests doesn't seem to skip failsafe plugin tests15:49
bobmcwor at least the pre/post integration bits15:49
jbossbotgit [wildfly-swarm] push master 37ce373.. Bob McWhirter Propagate some more properties to :start based on sysprops15:51
jbossbotgit [wildfly-swarm] push master URL:
projectodd-ciStarting build #318 for job wildfly-swarm (previous build: FAILURE -- last SUCCESS #313 20 hr ago)15:51
*** BenT15 <BenT15!> has quit IRC (Ping timeout: 252 seconds)15:53
bobmcwif this passes, I'll continue moving -examples over to the integs/15:55
bobmcwwell, first I'll get -swarm.jar version of integs running via a profile15:55
bobmcwand refactor some bits from arq container to start etc15:55
*** jclingan <jclingan!~jclingan@redhat/jclingan> has quit IRC (Quit: jclingan)15:56
*** BenT15 <BenT15!> has joined #wildfly-swarm15:58
*** rbenevides <rbenevides!~rbenevid@redhat/jboss/RafaelBenevides> has quit IRC (Quit: Computer has gone to sleep.)16:00
*** jclingan <jclingan!~jclingan@redhat/jclingan> has joined #wildfly-swarm16:28
jbossbotgit [wildfly-swarm] push master 677b98b.. Toby Crawley Make FractionList a bit easier to use....16:41
jbossbotgit [wildfly-swarm] push master 37f002a.. Toby Crawley Display the full list of included fractions [SWARM-2]16:41
jbossbotgit [wildfly-swarm] push master URL:
jbossbotjira [SWARM-2] swarmtool should list all included fractions [In Progress (Unresolved) Task, Major, Toby Crawley]
*** BenT15 <BenT15!> has quit IRC (Ping timeout: 272 seconds)16:58
*** BenT15 <BenT15!> has joined #wildfly-swarm17:02
*** agoncal <agoncal!> has quit IRC (Ping timeout: 250 seconds)17:05
lanceballany reason not to merge
jbossbotgit pull req [wildfly-swarm] (open) emag Add Undertow.js fraction
lanceballlooks pretty straightforward to me17:12
lanceballalso, I'm just confused about how we're supposed to be using jira now and what should happen with the existing issues (done and/or in progress)17:17
projectodd-ciYippee, build fixed!17:20
projectodd-ciProject wildfly-swarm build #318: FIXED in 1 hr 29 min:
projectodd-ciBob McWhirter: Propagate some more properties to :start based on sysprops17:20
*** BenT15 <BenT15!> has quit IRC (Ping timeout: 272 seconds)17:23
projectodd-ciStarting build #319 for job wildfly-swarm (previous build: FIXED)17:23
bobmcwlanceball: no reason not to, just haven't gotten to it17:23
bobmcwytanabe would appreciate it if we did, in fact17:23
bobmcwI've also not really looked at jira yet, so we'd need to ask jclingan or kenfinnigan17:23
bobmcwor tcrawley17:23
tcrawleylanceball: I just drag jiras around and magic happens17:24
*** BenT15 <BenT15!> has joined #wildfly-swarm17:30
lanceballok, I'll do that then17:31
lanceballbobmcw: I'll merge, j=k?17:31
* bobmcw was starting to solve for j17:32
tcrawleylanceball: I think we drop things in done, where they remain until a release happens?17:32
jbossbotgit [wildfly-swarm] push master c4f38e9.. emag Add Undertow.js fraction17:32
jbossbotgit [wildfly-swarm] push master 98cfcbb.. Lance Ball Merge pull request #161 from emag/add-undertow-js-fraction...17:32
jbossbotgit [wildfly-swarm] push master URL:
jclinganbobmcw: lanceball: What to do about existing github issues hasn’t been defined, to be honest. We should start using Jira for new issues.17:40
jclinganMy personal preference would be to move all existing Open github issues over to jira.17:40
jclinganHowever,IIRC, the imported “closed” github issues are open on jira. We need to fix that. I think the issue owners should do that. ~100 issues, so not a lot of work IMHO.17:42
lanceballI think they all have been17:42
jclinganAh. I haven’t checked.17:42
jclinganI think we should also be using kanban as a dashboard for tracking the project. That this essentially means (and I’m new to this) is that we should proactively keep the status field current17:44
jclinganIf I look at KanBan now, I see that only one issue is being worked on.17:45
*** BenT15 <BenT15!> has quit IRC (Ping timeout: 260 seconds)17:45
jclinganOf course, I’m guilty too because I have to start prioritizing work from a PM perspective.17:45
*** agoncal <agoncal!> has joined #wildfly-swarm17:48
*** BenT15 <BenT15!> has joined #wildfly-swarm17:50
*** lanceball is now known as lance|afk18:04
*** jclingan <jclingan!~jclingan@redhat/jclingan> has quit IRC (Quit: jclingan)18:10
*** agoncal <agoncal!> has quit IRC (Ping timeout: 250 seconds)18:15
*** agoncal <agoncal!> has joined #wildfly-swarm18:27
*** jclingan <jclingan!~jclingan@redhat/jclingan> has joined #wildfly-swarm18:36
*** jclingan <jclingan!~jclingan@redhat/jclingan> has quit IRC (Client Quit)18:37
*** jclingan <jclingan!~jclingan@redhat/jclingan> has joined #wildfly-swarm18:37
*** bbrowning is now known as bbrowning_away18:39
*** BenT15 <BenT15!> has quit IRC (Ping timeout: 260 seconds)18:42
*** BenT15 <BenT15!> has joined #wildfly-swarm18:44
*** rbenevides <rbenevides!~rbenevid@redhat/jboss/RafaelBenevides> has joined #wildfly-swarm18:52
projectodd-ciStarting build #320 for job wildfly-swarm (previous build: SUCCESS)18:55
*** jclingan <jclingan!~jclingan@redhat/jclingan> has quit IRC (Ping timeout: 244 seconds)18:56
*** jclingan <jclingan!~jclingan@redhat/jclingan> has joined #wildfly-swarm18:59
jclinganbobmcw: BTW, as we move from ALPHA5 to ALPHA6, we should have a gate that says booker must work and the examples must work with each alpha. Is that already the case?18:59
bobmcwjclingan: yes, at least wrt -examples, and moving the -examples to wildfly-swarm.git as integration tests should ensure that19:01
bobmcwbooker would be nice, but...19:02
jclinganIs there already a plan to move the examples to wildfly-swarm.git? I have mixed feelings about that. Basically, I want to make sure that developers always have something the evaluate that works. Perhaps, once I figure out bootstrap a bit more, I can add a page with builds of everything that are known to work together. Once we have a 1.0, it’s all clear. But until then, clear as mud :-)19:03
bobmcwyah, work is underway on moving the -examples19:04
bobmcwso far only 2:
bobmcwand ultimately includes testing the -examples, not just building them19:04
bobmcwhopefully as both wildfly-swarm:run and as -swarm.jar executions19:04
bobmcwso far, aside from pom.xml changes and adding src/it/java/**, the examples are un-changed19:05
bobmcwe.g., src/main/java is still useful example19:05
jclinganSo, going forward, how would a generic developer download samples to learn from?19:05
bobmcw(a) checkout the repo or (b) just copy/paste from github.com19:06
jclinganfrom which project?19:06
bobmcwthe main project19:06
jclinganwhich repo?19:06
jclinganHmmm ....19:06
bobmcwthey could clone that, checkout a tag, and cd integs/ and play19:06
bobmcww/o having to build the whole thing19:06
bobmcwand we could rename integs/ to examples/ even19:07
jclinganFrom a “community” perspective, I really like having a separate repo. It’s very clear.19:07
bobmcwI think the benefit of actually tested and known to work is a major thing19:07
*** agoncal <agoncal!> has quit IRC (Ping timeout: 250 seconds)19:07
bobmcweven our separate repo has been -SNAPSHOT centric, aside from the tagging19:07
bobmcwso they'd still have to clone, checkout a tag19:08
bobmcwI think wildfly-swarm.git!/examples/ would be no more difficult, or less clear19:08
bobmcwwith the aforementioned upside of Actually Known To Work19:08
jclinganSo, let’s say we’re at 1.0. How would we provide a .zip (for example) that a developer could download and “mvn package”?19:08
bobmcwof the examples?19:09
bobmcwwe could probably arrange something with maven to produce an -examples.zip19:09
bobmcwalternatively, once we get this working, maybe we could git-modules in the separate -examples repo19:09
bobmcwthe only issue I see in either case is the parent-pom between root and integs/pom.xml19:10
bobmcwbut we could break that linkage19:10
bobmcwhowsabout this:19:10
jclinganYeah. that was my concern. Don’t want to dowload the whole repo just to get examples.19:10
bobmcw1) we get the examples into the core repo, and tested19:10
bobmcw2) we look to then break them back out and linked via git-modules19:10
jclinganI’m not smart on git-modules (I’ll look it up), but I trust you :-)19:11
bobmcwright now, we have no idea if the examples actually still work, are up to date, etc19:11
bobmcwI think that should be the highest priority, because otherwise, the examples are pretty worthless19:11
jclinganYeah, I see where you’re coming from.19:11
bobmcwafter that, then we can figure out how to handily distribute them19:12
jclinganIs there a jira issue on this?19:12
bobmcwof course not :)19:12
*** BenT15 <BenT15!> has quit IRC (Ping timeout: 255 seconds)19:12
* bobmcw goes to create a jira19:12
jclinganDude, I’m still LOL19:12
jclinganI’ve started re-orging the priority in the backlog.19:13
jbossbotjira [SWARM-179] Move -examples into primary repo as tested integration tests [In Progress (Unresolved) Story, Major, Bob McWhirter]
*** BenT15 <BenT15!> has joined #wildfly-swarm19:14
jclinganThere are PM issues, and then there are engineering issues that may be dependencies on the PM issues, so I suspect the ordering will be F’d for a while until we figure it all out.19:14
jclinganFor example, I may move something high in the backlog, but an implementation requirement for it may be lower in the backlog.19:15
jclinganNot sure if Kanban catches that.19:15
bobmcwsure thing19:15
jclingan(if they are linked)19:15
jclinganWe should link issues like this.19:15
bobmcwif there's a dependency relationship, or a top-level to children, definitely19:16
bobmcw"It should make whipped cream" -> ( "Milk a cow", "buy nitrous oxide" )19:17
bobmcwPM issues will tend to be epics/stories, while eng issues will probably be several per PM issue, on details19:17
jclinganok. I can do epics/stories then.19:18
*** jclingan <jclingan!~jclingan@redhat/jclingan> has quit IRC (Read error: Connection reset by peer)19:21
*** jclingan <jclingan!~jclingan@redhat/jclingan> has joined #wildfly-swarm19:21
*** agoncal <agoncal!> has joined #wildfly-swarm19:21
bobmcwagoncal: hey! glad you're okay19:21
jclinganagoncal: So, have you done any hands-on with -swarm yet?19:22
agoncalHi guys19:22
bobmcwhe was one of our early supporters19:22
agoncalThanks, we are ok19:22
jclinganOh, yeah, forgot you were in Paris!19:22
jclinganGlad to hear it.19:22
agoncalI live on 2 rue de Charonne, my daughters' school is at 41 rue de Charonne, and one shooting was at 92 rue de Charonne.... scary19:22
bobmcwjclingan: I think we do need, separate from JIRA, some punchlist for releases (possibly not alphas though): things like Forge support, updating keycloak/hawkular/etc19:23
agoncalGuys, BTW, I might be asking a few questions in the near future19:23
bobmcwagoncal: jeepers19:23
bobmcwsure, we're always happy to help, unless it's after 5pm localtime and I have a beer19:23
agoncalI'm writing a web app invoking 2 external REST services19:23
agoncalWhat I want to do is deploy these 3 war files in a WildFly, check the resources it uses (memory footprint and so on), port them to Swarm and do a comparaison19:24
*** bbrowning_away is now known as bbrowning19:24
agoncalI still struggle a bit with the JSF part in Swarm... but I need to investigate a bit more before I ask for your help guys ;o)19:25
bobmcwI'm going to guess that 3 -swarms are not better than a single WF, in terms of memory usage19:25
bobmcwagoncal: kenfinnigan (not present) is probably the best support for JSF19:25
jclinganagoncal: Is this a revamp of :
jbossbotTitle: O Java EE 6 Application Servers, Where Art Thou? | Antonio's Blog19:26
bobmcw3 -swarms might be marginally better than 3 full WFs, but since WF lazily starts stuff, it's pretty memory efficient19:26
agoncalbobmcw That's exactly those things I want to hightlight. I also want to try "shrinking" WildFly with the minimum set of services and compare it with swarm :19:26
agoncalShrinking the app server, or inflating the jar19:26
bobmcwsounds good19:26
*** BenT15 <BenT15!> has quit IRC (Read error: Connection reset by peer)19:27
agoncaljclingan Yes, it's a revant ;o)  That's why I'm struggling with Weblogic at the moment ;o)19:27
jbossbotgit [wildfly-swarm] push master 9bf6f0e.. Bob McWhirter Make the execution of a swarm .war/.jar project more official-like.19:27
jbossbotgit [wildfly-swarm] push master cc3df2b.. Bob McWhirter Throw interrupted exceptions from stop()....19:27
jbossbotgit [wildfly-swarm] push master URL:
*** BenT15 <BenT15!> has joined #wildfly-swarm19:30
*** jclingan <jclingan!~jclingan@redhat/jclingan> has quit IRC (Read error: Connection reset by peer)19:32
jbossbotgit [wildfly-swarm] push master 9116ee4.. Bob McWhirter Remove un-used code.19:32
jbossbotgit [wildfly-swarm] push master URL:
*** jclingan <jclingan!~jclingan@redhat/jclingan> has joined #wildfly-swarm19:33
*** BenT15 <BenT15!> has quit IRC (Ping timeout: 250 seconds)19:38
*** kenfinnigan <kenfinnigan!~ken@redhat/jboss/kenfinnigan> has joined #wildfly-swarm19:38
*** BenT15 <BenT15!> has joined #wildfly-swarm19:39
*** lance|afk is now known as lanceball19:49
*** bbrowning is now known as bbrowning_away19:53
*** agoncal <agoncal!> has quit IRC (Ping timeout: 250 seconds)19:53
BenT15Hi all. Quick question. I am looking to deploy a simple EJB with a @Schedule method annotation. Eventually will need datasource and transactions, but for now just bootstraping. Any examples which show how to do this? I see lots of JAXRS, but pretty sure that isn't what I need.20:02
bobmcwhrm, we don't seem to have an EJB example20:03
bobmcwkenfinnigan: do you know of an ejb example hiding anywhere?20:04
kenfinniganbobmcw: I think some of the examples might use @Stateless20:04
BenT15My guess is that one would just add an EJB deployment, like JAXWS.20:04
kenfinnigandon't think we have a @Schedule one though20:04
jclinganI just grep’d the examples, didn’t see one.20:05
BenT15They do have @Stateless, but it seems to deploy under another type of deployment, like JAXWS ... just guessing.20:05
kenfinniganBenT15: it would be more like Transactions example20:05
kenfinniganas you're essentially deploying a JAR with no web side to it20:05
jclinganmy bad, forgot grep -i :-)20:05
BenT15Right.  I'll have a look at that.20:06
kenfinniganBenT15: actually, I think there was a closed issue about remote ejb that someone had put into a repo20:06
BenT15hmm.. Transactions uses JAXRSArchive. Not what I expected to see.20:06
bobmcwbbrowning_away: tcrawley: you'll be happy to know that...20:07
bobmcwa) by default integs don't run now20:07
bobmcwb) but add -Pintegs-plugin to run them via wildfly-swarm:run20:07
bobmcwor (c) add -Pintegs-uberjar to run them as their -swarm.jar's built from maven20:07
kenfinniganBenT15: try doing a plain ShrinkWrap JarArchive type20:07
BenT15K, will check it out. Thanks.20:08
kenfinniganBenT15: if it breaks, we need to fix that as that is a major use case for micro service style20:08
BenT15Right, happy to help.20:08
bobmcwwe recommend JARArchive.class20:08
kenfinniganbobmcw: is that ours or SW?20:08
bobmcwnot that it's much different from JavaArchive20:08
bobmcwit's ours20:08
kenfinniganah ok, right type, wrong place it lives20:08
bobmcwbasically it's a JavaArchive + support for addModule/jboss-deployment-structure.xml20:09
BenT15okay, would I call addResource for each EJB class, or is there some kind of package scan?  Forgive my ignorance of EJB, been avoided it till now.20:09
bobmcwor you can addPackage()20:10
BenT15got it - should have checked the API.  Pretty darn clear. :)20:10
bobmcwImma adjust CI to run integs with both profiles after doing the full build20:11
kenfinniganbobmcw: sounds good20:12
bobmcwkenfinnigan: fwiw, there's integs/ now, with just jaxrs and messaging examples, but the plumbing is setup to start moving other -examples or integration-tests/ over20:12
bobmcwand given prior discussion with jclingan, maybe rename integs/ to examples/ some time in the near future.20:13
bobmcwbut still treat them as integration tests20:13
jclingan+1, examples preferred.20:13
kenfinniganbobmcw: so the examples were moved into main repo, or just testing against them?20:13
bobmcwkenfinnigan: I'm moving them, one by one, and adding Drone tests or whatnot20:14
bobmcwwith the idea of decomissioning -examples.git, and ensuring the examples are actually testing/working continuously20:14
bobmcwthen we may git-module them back into a separate repo, to make jclingan happy :)20:14
jclinganPersonal Milestone - I just wrote my first WildFly Swarm user story:
jbossbotjira [SWARM-180] Declarative Static Configuration [Backlog (Unresolved) Story, Major, Bob McWhirter]
bobmcwbut my #1 goal is to get'em all tested20:14
BenT15It's working beautifully. Thanks, all.20:14
bobmcwjclingan: good job!20:15
jclinganBob, when creating stories, I’ll default them to you.20:15
bobmcwcan we not do un-assigned?20:15
jclinganSniff Sniff. I’m so proud of myself.20:15
kenfinniganbobmcw: i'm confused as to the benefit of them being inside over outside the main repo20:15
BenT15I'll try to drop in a example project with a PR.20:15
kenfinniganjclingan: well done20:15
kenfinniganBenT15: awesome! thanks20:15
bobmcwkenfinnigan: inside, they can be part of our build, and share dependency-management bits and project.version etc20:15
bobmcwand also, inside, they can act as actual tests, and intellij can pick up changes to APIs etc20:16
kenfinniganbobmcw: ok, but doesn't that then make it harder for someone to take an example and build on it?20:16
bobmcwyes, that's jclingan's argument also20:16
kenfinniganguess then they also need to do all that themselves20:16
*** mikesir87 <mikesir87!c6521024@gateway/web/cgi-irc/> has joined #wildfly-swarm20:16
bobmcwso I proposed (a) get them working then (b) figure out a better arrangement20:16
bobmcwmikesir87: didja figure out ApplicationRealm bits?20:16
kenfinnigangranted the API thing would be a benefit20:16
bobmcwa git-submodule would probably suffice, but that's more work than I wanted to get into today :)20:17
mikesir87bobmcw: haha... quick timing!  Was just about to give it a try, but can't find where the SNAPSHOTs are available.  Do I just need to build locally?  Don't see them deployed anywhere20:17
bobmcwone moment20:17
mikesir87(Alpha6 anyways)20:17
jbossbotTitle: Index of /snapshot20:17
bobmcwthat's built from CI, so very much a moving target :)20:18
bobmcwbut should be within about 90min of HEAD, assuming no failure20:18
mikesir87Awesome. I was looking at the repos listed in the pom.  The CI is fine for now.  I'm already living on the bleeding edge anyways ;)20:18
bobmcwthat's mostly used for our own downstream builds, but you're welcome to use it20:19
jbossbotgit [wildfly-swarm] push master 5690259.. Bob McWhirter Stuff integration tests under either -Pintegs-plugin or -Pintegs-uberjar.20:22
jbossbotgit [wildfly-swarm] push master URL:
jclinganNew Story: WildFly Swarm Dynamic Configuration Changes:
jbossbotjira [SWARM-181] WildFly Swarm Dynamic Configuration Changes [Backlog (Unresolved) Story, Major, Bob McWhirter]
bobmcwokay, adjusted CI, will trigger a build after this completes20:24
bobmcwwhich should then run integs both ways20:24
bobmcwbut after the `mvn deploy` so that is sub-optimal20:24
tcrawleybobmcw: what constitutes an "integ" for swarm? I'm going to figure out tests for the ispan fraction, and want to do an in-vm test there where I fire up the container and muck with a cache20:25
tcrawleyshould that be in the fraction or an integ?20:25
bobmcwstuff under integs/20:25
bobmcwstuff executed as an -example would20:25
bobmcwwildfly-swarm:run or java -jar -swarm.jar20:25
bobmcwnot arquillian20:25
bobmcwat least at this point20:26
bobmcwmy goal was to get -examples tested and usable as tests for -swarm20:26
bobmcwyour bits could be under infinispan/test/ for Arq or InVm20:26
bobmcwthose still run regardless20:27
* tcrawley needs to figure out how to talk across the container boundary. that should be easy for invm, maybe harder for arq? dunno20:28
* tcrawley knows nutall about arq20:28
*** bbrowning_away is now known as bbrowning20:29
projectodd-ciStarting build #321 for job wildfly-swarm (previous build: SUCCESS)20:29
jclinganjclingan needs to go for lunch20:33
bbrowningtcrawley: for arquillian tests you may need to take an approach like we did in torquebox and have a messaging backchannel or something similar20:35
bbrowningor put a tiny web interface on it20:35
tcrawleybut yeah, I may need to do that20:35
bbrowningarquillian can run tests inside the server or as a client talking to the web interface of the server20:35
tcrawleyor write them in clojure and use a repl channel :)20:35
bbrowningbobmcw: yeah for CI you may want to do mvn install separate from mvn deploy and -DskipTests on deploy20:36
*** rbenevides <rbenevides!~rbenevid@redhat/jboss/RafaelBenevides> has quit IRC (Quit: Quit)20:37
jbossbotgit [wildfly-swarm] push master 522f207.. Lance Ball Make ribbon-webapp's context path configurable.20:37
jbossbotgit [wildfly-swarm] push master URL:
*** BenT15 <BenT15!> has quit IRC (Ping timeout: 265 seconds)20:44
*** BenT15 <BenT15!> has joined #wildfly-swarm20:46
*** agoncal <agoncal!> has joined #wildfly-swarm20:50
mikesir87Ran into an error? Any pointers? (occurring when simply doing Container container = new Container())20:53
mikesir87org.jboss.modules.xml.XmlPullParserException: Failed to resolve artifact 'org.wildfly.swarm:wildfly-swarm-container-runtime:1.0.0.Alpha6-SNAPSHOT' (position: END_TAG seen"/>... @10:95)20:53
bbrowningmikesir87: interesting! our CI server was just seeing that same error20:54
bbrowningthis is on a simple app you're making20:54
mikesir87Yeah. I'm pulling the SNAPSHOTs from the CI server, hence why I'm getting it too then :)20:54
bbrowningfor the CI server the issue was related to using a custom maven repo - do you have -Dmaven.repo.local set?20:55
*** agoncal <agoncal!> has quit IRC (Ping timeout: 250 seconds)20:55
mikesir87i dont. my pom has a repo set for the CIs snapshot repo and one for our local nexus20:55
mikesir87do i need to set that then?20:55
bbrowningno - the CI bug was only present when it was set20:56
mikesir87ah. ok20:56
mikesir87hmmm... any other ideas?21:00
*** BenT15 <BenT15!> has quit IRC (Ping timeout: 240 seconds)21:04
*** jclingan <jclingan!~jclingan@redhat/jclingan> has quit IRC (Quit: jclingan)21:08
bbrowningmikesir87: is there a full stacktrace after that error?21:08
bbrowningbobmcw: that extra output you added to debug CI - is there a way mikesir87 can enable it locally to see what's failing to resolve?21:08
lanceballbobmcw: the ribbon example says "NOTE: This example will not work yet, as it relies upon a patch in the upstream of WildFly not yet available in a released version." in the README - is this still expected to be the case?21:09
bbrowningmikesir87: also, see if you have a non-empty directory of ~/.m2/repository/org/wildfly/swarm/wildfly-swarm-container-runtime/1.0.0.Alpha6-SNAPSHOT21:09
bobmcwbbrowning: no easy way, just add printlns to MavenArtifactUtil21:13
mikesir87yes. i have a stacktrace. just dropped it here -
jbossbotTitle: org.jboss.modules.ModuleLoadException: Error loading module from modules/org/wil - Pastebin.com21:13
bobmcwwe should add a logger or something property-enabled21:13
bbrowningbobmcw: yeah I think especially while we're still alpha logging would be good21:14
bbrowningor hell -Dwildfly.swarm.debug just to turn on some helpful printlns21:14
mikesir87and interesting... yeah. no ~/.m2/repository/org/wildfly/swarm/wildfly-swarm-container-runtime/1.0.0.Alpha6-SNAPSHOT directory. i have a 1.0.0.Alpha5 folder, but no Alpha621:14
bbrowningok so that explains why it can't find it21:14
bbrowningmikesir87: what swarm alpha6 dependencies do you have in your app's pom.xml, assuming maven?21:15
bbrowningor if gradle whatever gradle uses21:15
mikesir87wildfly-swarm-jaxrs, wildfly-swarm-messaging, and wildfly-swarm-msc21:16
mikesir87(using maven)21:16
bbrowningso we're probably missing some maven deps in the swarm pom.xmls somewhere21:17
mikesir87mk. should i go ahead and just add a container-runtime dependency directly for now?21:17
bbrowningas a workaround, add wildfly-swarm-container-runtime to your pom.xml21:17
mikesir87sweet. thanks :)21:17
bbrowningwell thank me if it works ;)21:18
*** BenT15 <BenT15!> has joined #wildfly-swarm21:18
bbrowningbobmcw: so what's supposed to have maven deps on *-runtime artifacts?21:19
bbrowningfor example wildfly-swarm-jaxrs has no dependency on wildfly-swarm-jaxrs-runtime21:19
mikesir87haha... looks like there's a few others then too. got the same thing for wildfly-swarm-jaxrs21:19
bbrowningif a user sticks wildfly-swarm-jaxrs in their app, what has the smarts to download the -runtime bits since there is no maven dep there?21:19
mikesir87yup. just saw that too21:19
bbrowningI think our maven deps aren't setup appropriately based on how we expect users to use them21:20
bbrowningand it means we need some integration tests that start with a completely clean m2_repo to suss this out21:20
bbrowningor some kind of test21:20
bbrowningbobmcw: if -examples stayed in a separate repo as a separate downstream CI build and didn't copy over the m2repo from the upstream build then we would have discovered this problem long ago21:21
bbrowningor they can be in the same repo but a separate CI build21:21
mikesir87wildfly-swarm-undertow-runtime is another one21:22
bbrowningmikesir87: wildfly-swarm-jaxrs-runtime should bring in undertow-runtime21:23
*** jclingan <jclingan!~jclingan@redhat/jclingan> has joined #wildfly-swarm21:23
bbrowningahh wait! no it won't <scope>provided</scope>21:23
bbrowningthanks for being an alpha tester :)21:23
bbrowningmikesir87: an @issue for this problem would be awesome - we definitely have to fix it21:23
bbrowningwill require some renaming and/or fixing of poms21:24
bbrowningtcrawley: proddbot hates me21:24
tcrawleyit probably hates us all21:24
bbrowningbot's not in the room21:24
mikesir87haha... i'll take care of it ;)21:25
* tcrawley really needs to clean up and update that bot21:26
*** proddbot <proddbot!~irclj@> has joined #wildfly-swarm21:26
tcrawley@issue bbrowning21:26
proddbotbbrowning: tcrawley would be happy as the quivering of tall grass on the plains if you would file an issue at
bbrowningthere we go :)21:26
mikesir87haha... that's awesome. thanks bot21:27
jbossbotjira [SWARM-183] Runtime dependencies not listed as artifact dependencies [Backlog (Unresolved) Bug, Major, Unassigned]
bbrowningmikesir87: thank you!21:30
mikesir87bbrowning: you're welcome! i'd be happy to help add dependencies if you want them added in to each project. i'd be happy to send a PR as i find them (unless you want to go a different direction with them)21:31
bbrowningmikesir87: well, right now our api modules are wildfly-swarm-foo and the runtime modules are wildfly-swarm-foo-runtime21:33
bbrowningI'm wondering if that's wrong and we should call api wildfly-swarm-foo-api and runtime wildfly-swarm-foo21:33
bbrowningthen everything would just work, I think21:34
bbrowningalternatively we have to tell users to use wildfly-swarm-foo-runtime in their pom instead of wildfly-swarm-foo but that seems silly21:34
lanceballbobmcw: the ribbon example says "NOTE: This example will not work yet, as it relies upon a patch in the upstream of WildFly not yet available in a released version." in the README - is this still expected to be the case?21:34
mikesir87makes sense. thanks21:35
*** BenT15 <BenT15!> has quit IRC (Ping timeout: 240 seconds)21:41
mikesir87Alrighty... was finally able to get it to start. When starting with dependencies on jaxrs, messaging, and msc, I changed those to runtime dependencies, but then also needed to add runtime dependencies for container, logging, io, naming, security, ee, connector, jca, and transactions.21:49
mikesir87Want me to add it as a comment on the issue?21:49
projectodd-ciProject wildfly-swarm build #321: FAILURE in 1 hr 26 min:
projectodd-ci* Bob McWhirter: Make the execution of a swarm .war/.jar project more official-like.21:56
projectodd-ci* Bob McWhirter: Throw interrupted exceptions from stop().21:56
projectodd-ci* Bob McWhirter: Remove un-used code.21:56
projectodd-ci* Bob McWhirter: Stuff integration tests under either -Pintegs-plugin or -Pintegs-uberjar.21:56
projectodd-ciStarting build #322 for job wildfly-swarm (previous build: FAILURE -- last SUCCESS #320 3 hr 0 min ago)21:56
bbrowningmikesir87: yes please22:00
*** bbrowning is now known as bbrowning_away22:00
mikesir87Sounds good. And with that, I'm out. I'll probably be playing with Swarm some more next week, as we're not sprinting and the office will be mostly empty. I'd be happy to help out more. Have a great weekend all!22:01
*** BenT15 <BenT15!> has joined #wildfly-swarm22:04
*** mikesir87 <mikesir87!c6521024@gateway/web/cgi-irc/> has quit IRC (Quit: - A hand crafted IRC client)22:09
*** agoncal <agoncal!> has joined #wildfly-swarm22:12
*** agoncal <agoncal!> has quit IRC (Ping timeout: 250 seconds)22:16
projectodd-ciProject wildfly-swarm build #322: STILL FAILING in 22 min:
projectodd-ciLance Ball: Make ribbon-webapp's context path configurable.22:18
*** agoncal <agoncal!> has joined #wildfly-swarm22:21
*** BenT15 <BenT15!> has quit IRC (Ping timeout: 246 seconds)22:21
*** enebo <enebo!> has quit IRC (Quit: enebo)22:40
*** jclingan <jclingan!~jclingan@redhat/jclingan> has quit IRC (Quit: jclingan)22:40
*** tcrawley is now known as tcrawley-away22:42
*** agoncal <agoncal!> has quit IRC (Ping timeout: 250 seconds)23:12
*** os890_ <os890_!> has joined #wildfly-swarm23:17
*** os890 <os890!> has quit IRC (Ping timeout: 240 seconds)23:19
*** os890_ is now known as os89023:19
*** bobmcw <bobmcw!~bobmcw@redhat/jboss/bobmcw> has quit IRC (Remote host closed the connection)23:25
*** jclingan <jclingan!~jclingan@redhat/jclingan> has joined #wildfly-swarm23:52

Generated by 2.9.2 by Marius Gedminas - find it at!