Friday, 2015-11-13

*** enebo <enebo!> has quit IRC (Quit: enebo)00:19
*** helio-frota <helio-frota!> has quit IRC (Quit: Leaving)00:31
*** jcrossley3 is now known as jcrossley3-away00:57
*** qmx <qmx!~qmx@redhat/jboss/qmx> has joined #wildfly-swarm01:12
*** silentsky <silentsky!~silentsky@> has joined #wildfly-swarm01:56
*** silentsk_ <silentsk_!~silentsky@> has joined #wildfly-swarm02:02
*** silentsky <silentsky!~silentsky@> has quit IRC (Read error: Connection reset by peer)02:02
*** bgeorges <bgeorges!~bgeorges@> has quit IRC (Remote host closed the connection)02:12
*** bgeorges <bgeorges!~bgeorges@> has joined #wildfly-swarm02:13
*** bgeorges <bgeorges!~bgeorges@> has quit IRC (Ping timeout: 264 seconds)02:20
*** bobmcw <bobmcw!~bobmcw@redhat/jboss/bobmcw> has joined #wildfly-swarm02:29
*** ChanServ sets mode: +o bobmcw02:29
*** bobmcw <bobmcw!~bobmcw@redhat/jboss/bobmcw> has quit IRC (Ping timeout: 240 seconds)02:34
*** wangc <wangc!~wangc@redhat/jboss/wangc> has joined #wildfly-swarm02:45
*** silentsk_ <silentsk_!~silentsky@> has quit IRC ()02:52
*** bgeorges <bgeorges!~bgeorges@> has joined #wildfly-swarm03:36
bgeorgeshi, where can I find Ken's J1 slides ?03:37
*** bobmcw <bobmcw!~bobmcw@redhat/jboss/bobmcw> has joined #wildfly-swarm06:32
*** ChanServ sets mode: +o bobmcw06:32
*** bobmcw <bobmcw!~bobmcw@redhat/jboss/bobmcw> has quit IRC (Ping timeout: 244 seconds)06:38
*** bgeorges <bgeorges!~bgeorges@> has quit IRC (Remote host closed the connection)06:57
*** bgeorges <bgeorges!~bgeorges@> has joined #wildfly-swarm06:58
*** bgeorges <bgeorges!~bgeorges@> has quit IRC (Remote host closed the connection)07:59
*** bgeorges <bgeorges!~bgeorges@> has joined #wildfly-swarm07:59
*** hbraun <hbraun!~hbraun@pdpc/supporter/professional/hbraun> has joined #wildfly-swarm08:10
*** bobmcw <bobmcw!~bobmcw@redhat/jboss/bobmcw> has joined #wildfly-swarm08:34
*** ChanServ sets mode: +o bobmcw08:34
*** bobmcw <bobmcw!~bobmcw@redhat/jboss/bobmcw> has quit IRC (Ping timeout: 240 seconds)08:39
*** bgeorges <bgeorges!~bgeorges@> has quit IRC (Remote host closed the connection)09:12
*** bgeorges <bgeorges!~bgeorges@> has joined #wildfly-swarm09:12
*** bgeorges_ <bgeorges_!~bgeorges@> has joined #wildfly-swarm09:28
*** bgeorges_ <bgeorges_!~bgeorges@> has quit IRC (Read error: Connection reset by peer)09:29
*** bgeorges_ <bgeorges_!~bgeorges@> has joined #wildfly-swarm09:30
*** bgeorges <bgeorges!~bgeorges@> has quit IRC (Ping timeout: 255 seconds)09:32
*** bgeorges_ <bgeorges_!~bgeorges@> has quit IRC (Remote host closed the connection)10:13
*** bgeorges <bgeorges!~bgeorges@> has joined #wildfly-swarm10:14
*** bgeorges <bgeorges!~bgeorges@> has quit IRC (Ping timeout: 246 seconds)10:18
*** helio-frota <helio-frota!> has joined #wildfly-swarm10:20
*** bgeorges <bgeorges!~bgeorges@> has joined #wildfly-swarm10:21
*** hbraun <hbraun!~hbraun@pdpc/supporter/professional/hbraun> has quit IRC (Quit: afk)11:27
*** bbrowning_away <bbrowning_away!~bbrowning@redhat/jboss/bbrowning> has quit IRC (Remote host closed the connection)11:38
*** bbrowning_away <bbrowning_away!~bbrowning@redhat/jboss/bbrowning> has joined #wildfly-swarm11:38
*** tcrawley-away is now known as tcrawley12:29
*** bobmcw <bobmcw!~bobmcw@redhat/jboss/bobmcw> has joined #wildfly-swarm12:36
*** ChanServ sets mode: +o bobmcw12:36
*** bobmcw <bobmcw!~bobmcw@redhat/jboss/bobmcw> has quit IRC (Ping timeout: 240 seconds)12:42
*** bbrowning_away <bbrowning_away!~bbrowning@redhat/jboss/bbrowning> has quit IRC (Ping timeout: 240 seconds)12:44
*** lance|afk is now known as lanceball12:49
*** bbrowning <bbrowning!~bbrowning@redhat/jboss/bbrowning> has joined #wildfly-swarm12:49
bgeorgeslanceball: Thank you12:51
bgeorgesWe are doing our second JBug aroind MSA and Swarm in Beijing right now :)12:51
lanceballthere are a number of presentations in that repo, but I think ken's is the most comprehensive12:53
*** jcrossley3-away is now known as jcrossley312:53
bgeorgesWe did our owns for now based on local demand and language13:01
bgeorgesReusing your work13:01
*** hbraun <hbraun!~hbraun@pdpc/supporter/professional/hbraun> has joined #wildfly-swarm13:36
*** bgeorges <bgeorges!~bgeorges@> has quit IRC (Remote host closed the connection)13:46
*** bgeorges <bgeorges!~bgeorges@> has joined #wildfly-swarm13:46
*** bgeorges <bgeorges!~bgeorges@> has quit IRC (Ping timeout: 250 seconds)13:51
*** hbraun <hbraun!~hbraun@pdpc/supporter/professional/hbraun> has quit IRC (Quit: afk)13:55
*** kenfinnigan <kenfinnigan!~ken@redhat/jboss/kenfinnigan> has joined #wildfly-swarm14:04
*** bobmcw <bobmcw!~bobmcw@redhat/jboss/bobmcw> has joined #wildfly-swarm14:07
*** ChanServ sets mode: +o bobmcw14:07
*** enebo <enebo!> has joined #wildfly-swarm14:09
*** wangc <wangc!~wangc@redhat/jboss/wangc> has quit IRC (Quit: Leaving)14:12
*** helio-frota <helio-frota!> has quit IRC (Quit: Leaving)14:14
*** helio-frota <helio-frota!> has joined #wildfly-swarm15:02
*** hbraun <hbraun!~hbraun@pdpc/supporter/professional/hbraun> has joined #wildfly-swarm15:45
*** johnament <johnament!6cb33342@gateway/web/freenode/ip.> has joined #wildfly-swarm15:58
*** hbraun <hbraun!~hbraun@pdpc/supporter/professional/hbraun> has quit IRC (Quit: afk)16:04
tcrawleydo we have any tests that exercise a fraction other than just creating it?16:59
tcrawleybobmcw: is there a way to detect if fraction X is loaded from fraction Y?16:59
bobmcwtcrawley: yes!16:59
bobmcwat least in postInitialize()17:00
tcrawleyyou seem exicted17:00
bobmcwsince we don't know the order they'll initialize, we initialize them all, then we call postInitialize() where you can query17:00
bobmcwand then finally we getList()17:00
bobmcwsee postInitialize(PostInitContext)17:00
bobmcwyou can even get the FractionX instance so we can do cross-fraction frobnostication17:01
bobmcwtcrawley: just excited that my answer doesn't have to be "dunno" or "nope"17:01
tcrawleyI need to know from IspanFraction at createDefaultFraction() time if JGroupsFraction is loaded17:02
tcrawleybut I may be able to defer that decision17:02
tcrawleybobmcw: hey, since you are full of answers: what about my test question above?17:05
bobmcwyou mean at runtime, to see if it produces the right List<>?17:05
bobmcwor some integration tests, which don't test much?17:05
tcrawleyI want to actually put something in an ispan cache and confirm it's there17:06
tcrawleywhich would be an integration test, I reckon17:07
tcrawleyhmm, the gitbook shows InfinispanFraction fraction = new InfinispanFraction(); as an example, but InfinispanFraction() is private17:09
tcrawleyshould it be private?17:09
bobmcwyah, we're slowly moving to FooFraction.createDefault() or whatnot, I think17:15
tcrawleywell, what if you don't want the default?17:16
tcrawleythe default ispan fraction has the standard cache-containers17:17
tcrawleybut if you want different ones, it's currently only additive17:17
jbossbotgit [wildfly-swarm] push master 26a6c99.. Toby Crawley Provide separate ispan defaults for clustered & local...17:18
jbossbotgit [wildfly-swarm] push master 62ada1e.. Toby Crawley Apply default cache-container settings if not set...17:18
jbossbotgit [wildfly-swarm] push master 0c437b3.. Toby Crawley Infinispan needs jmx, so we depend on the fraction.17:18
jbossbotgit [wildfly-swarm] push master 4dcb04b.. Toby Crawley Detect jgroups to decide what defaults to apply.17:18
jbossbotgit [wildfly-swarm] push master URL:
projectodd-ciStarting build #298 for job wildfly-swarm (previous build: SUCCESS)17:18
bobmcwholy carp... reworking BuildTool to make it testable, and now a full build takes 50% of the time as previous17:19
bobmcwand also knocked the messaging-swarm.jar example from 89m to 5m17:21
bobmcwsurely I've broken something, but it still functions.17:21
tcrawleymeaning the tests still pass?17:23
tcrawleythat's not saying much, since they just create the fraction17:23
bobmcwyah, and the example runs, but I think it's resolving most stuff from M2_REPO17:23
tcrawleyah, example17:23
tcrawleykenfinnigan: can you give me admin rights to SWARM? I want to add a swarmtool component17:24
tcrawleyor bobmcw, if you have the rights to give rights17:24
kenfinnigantcrawley: sure, for a second you had me confused as to what SWARM was!17:24
kenfinnigannow I realise its jira17:25
bobmcwlemme see17:25
bobmcwyou got it kenfinnigan?17:25
kenfinniganbobmcw: getting in there17:25
kenfinniganbobmcw, tcrawley: should now be project admin17:26
tcrawleykenfinnigan: thanks, my friend!17:26
*** jclingan <jclingan!> has joined #wildfly-swarm17:29
*** johnament <johnament!6cb33342@gateway/web/freenode/ip.> has quit IRC (Ping timeout: 246 seconds)17:44
bobmcwyah, it was faster because we weren't injecting all of wildfly into the -swarm.jar17:51
bobmcwthat copying takes time, apparently17:51
bobmcwwho'd have thunk17:51
*** bbrowning is now known as bbrowning_away18:13
*** rbenevides <rbenevides!~rbenevid@redhat/jboss/RafaelBenevides> has joined #wildfly-swarm18:29
*** jcrossley3 is now known as jcrossley3-away18:29
*** jclingan <jclingan!> has quit IRC (Ping timeout: 244 seconds)18:40
*** bbrowning_away is now known as bbrowning19:04
jbossbotgit [wildfly-swarm] push master 9476c4a.. Bob McWhirter Break out the dependency-scrobbling bits from BuildTool and possibly...19:07
jbossbotgit [wildfly-swarm] push master URL:
projectodd-ciStarting build #299 for job wildfly-swarm (previous build: SUCCESS)19:07
bobmcwtcrawley: good job on all the @issues19:08
proddbotWe'd be happy as a kitten in a yarn shop if you would file an issue at
tcrawleyhey, I'm just filing them19:08
tcrawleydon't expect me to implement them19:09
* bobmcw expects you to19:09
bobmcwcheck it
tcrawleya test?!?19:09
bobmcwyes, around BuildTool stuffs19:10
bobmcwmuch of BuildTool has been moved to
* tcrawley is too embarrassed to add a jira titled "swarmtool needs tests"19:10
tcrawleyI'll see if swarmtool needs to adjust19:10
bobmcwBuildTool's public API is unchanged19:11
bobmcwbut it uses DependencyManager now to muck about the deps19:11
bobmcwmostly so I could test it w/o doing too much BuildTool'ing19:11
bbrowningdo we currently muck with the maven-generated .war or .jar before creating our -swarm.jar?19:20
bbrowningor we always take it as-is from the user's existing build?19:20
bbrowningactually let me rephrase I think - is it fair to say the user's artifact type in their pom.xml needs to be 'war' if generating a web app? or should 'jar' be valid too?19:21
bobmcwwe don't muck with the user-provided project19:32
bobmcwthey can build a .jar, but then deploy() would just deploy the jar I think19:32
bobmcwand WF can't do nothing with that19:32
bobmcwif their artifact is a .jar, they pretty much have to have a main() to build a synthetic .war19:33
*** jcrossley3-away is now known as jcrossley319:34
*** kenfinnigan <kenfinnigan!~ken@redhat/jboss/kenfinnigan> has quit IRC (Quit: Leaving.)19:35
bbrowningbobmcw: right, and booker does that19:40
bbrowningbut the problem is then nothing in src/main/webapp gets put in the user's .jar that we wrap in our -swarm.jar19:40
bbrowningthis is why src/main/resources was a valid static resource option at one time, I guess?19:41
bobmcwI think that might've been why I had src/main/resources/... for the .jar use-case19:41
bobmcwyou can use the maven-resources-plugin to add src/main/webapp to the .jar19:41
bobmcwbut that's sorta ugly19:41
bbrowningless ugly if the presence of wildfly-swarm-plugin in their pom can do that automatically19:42
bobmcwI've not found a way for our plugin to affect other parts of the build reliably, but maybe you can19:42
bobmcwiff you can, it'd be nice to also remove wildfly-swarm-* deps from WEB-INF/lib/ when packaging the .war if packaging==war19:43
bobmcwmaybe try another mojo bound to phase=initialize or somesuch?19:43
bobmcwI haven't tried that19:44
bbrowningyeah I have no idea yet :)19:44
bobmcwI'll be happy to take a look, since I'm mucking about with dependencies anyhoo19:46
*** rbenevides <rbenevides!~rbenevid@redhat/jboss/RafaelBenevides> has quit IRC (Quit: Computer has gone to sleep.)20:07
helio-frotatrying to do functional tests with 'groovy-geb' but until now without success20:10
bobmcwso, a) I can't get a mojo to execute at all w/o an execution20:18
bobmcwb) one that executes at phase=initialize and mucks with dependencies still doesn't succeed20:19
bobmcwc) we can probably add a resource path via plugin, but our current PackageMojo fires too late to affect .war building etc20:19
bobmcwbbrowning: ^20:19
bobmcwbbrowning: feel free to try hard than I did20:20
* bbrowning tries harder20:20
jbossbotgit [wildfly-swarm] push master dc1bfd7.. Bob McWhirter Touch up all the provided-dependencies.txt files.20:23
jbossbotgit [wildfly-swarm] push master URL:
bobmcwfwiw, I added an InitializeMojo, and had this.project.getArtifacts().clear()20:24
bobmcwtaking it from 50 deps to 020:24
projectodd-ciStarting build #300 for job wildfly-swarm (previous build: SUCCESS)20:24
bobmcwbut in PackageMojo, it was back to 5020:24
bobmcwwhen invoked `mvn wildfly-swarm:initialize package`20:24
bobmcwidentityHashCode of this.project was identical between the two mojos20:24
bobmcwso dunno what is re-blessing dependencies into it20:25
bobmcwand could not figure out how to get it implicitly executed anyhow20:25
bobmcwwe *could* do more magic, I think if we made a new <packaging> type/lifecycle20:27
bobmcwbut that seems super involved and dunno if it breaks .war and .jar building in general or whatnot20:28
bobmcwbut akin to <packaging>rubygem</packaging>20:28
bobmcwtcrawley: any luck with testing your cache?20:28
tcrawleybobmcw: I haven't yet tried20:29
*** mikesir87 <mikesir87!c65269d7@gateway/web/cgi-irc/> has joined #wildfly-swarm20:49
mikesir87Hey guys! Anyone out there? Got a question for ya.20:50
bobmcwmikesir87: sure, what's up?20:50
mikesir87Awesome! So... I've recently migrated our development's QA environment to deploy our application using a custom Docker image that has a configured Wildfly server (datasources, JMS, application users, etc.).  Was finally able to spend some time watching presentations about Wildfly Swarm.  Can I say... awesome?  :)  So, I've been able to figure out h20:52
mikesir87ow to configure JMS, datasources, etc.  But, haven't been able to figure out how to configure application users/roles?  Of course, in production, we wouldn't do it this way. But, for our qa env, we have set users/roles for testing. Any pointers?20:52
bobmcwhmm, good question!20:54
bobmcwhow do you normally do that in regular WildFly?20:55
bobmcwso far we've mostly focused on an external Keycloak server to do authn/authz20:55
bobmcwI guess normally application-(users,roles).properties?20:56
mikesir87When building the qa docker image (pre-swarm), we have RUN commands that use the jboss/ to add a user.  Which, yes, is putting them in the properties file20:56
mikesir87For production, we've built (and open-sourced) a Wildfly extension that provides CAS support for our applications (
mikesir87(Will probably end up making a fraction to add swarm support to it too btw)20:57
bobmcwso those .properties are loaded via this portion of the standalone.xml normally:
bobmcwand I don't think we have anything that maps to the <management> stanzas yet20:58
bobmcwif you wouldn't mind filing an @issue, we could probably take a look early next week20:58
proddbotWe would finally be Turing Complete if you would file an issue at
mikesir87Lol. Thanks bot ;)20:58
mikesir87Sure thing. I'll go ahead and make an issue request for it.20:59
bobmcwlanceball: do you know if config-api supports anything in <management> yet, or if those bits are exposed via the DMR at all?20:59
bobmcwor jamezp, do you know?20:59
bobmcwI think we only have support for <subsystem> bits from our config java API20:59
* jamezp reads back21:00
bobmcware <management> bits from standalone.xml exposed as DMR, therefore generatable in config-api?21:00
jamezpNo they're not. They're currently in .properties files, but I think that might change when elytron is introduced.21:01
bobmcwokie dokie... mikesir87 we can probably figure something out nonetheless for ya21:01
mikesir87Ok. Cool. Want me to still go ahead and file an issue then?21:02
bobmcwyes please; we'll forget by monday otherwise21:02
jamezpProbably not that usable in WF Swarm, but here's what I did in the maven plugin
mikesir87When we've got it figured out, I'd be happy to contribute an example or whatever. I'm happy to contribute21:03
mikesir87And shall do. Issue incoming in a moment21:03
bobmcwjamezp: yah, but we also need to do wtf <management> stanza in the standalone.xml does21:03
bobmcwwrt setting up security-realms etc21:03
bobmcw bits21:03
bobmcwwe have no API yet for those things on Container21:04
jamezpAh so that's kind of a pain. I was never able to sort out the correct CLI commands to make that work.21:04
bobmcwto point to the properties, etc21:04
bobmcw'k, I'll poke at config-api monday and see what I can dredge up21:04
jamezpThey have weird addresses and some of the path values are hard-coded.21:05
bobmcwthis is all picketlink crap still, until elytron?21:05
* bobmcw hates picketlink21:05
jamezpYes and the subsystem config is very strange.21:06
* jamezp doesn't blame you21:06
* bobmcw blames Anil21:06
bobmcwhave a great weekend; beer o'clock here21:06
mikesir87And to you all as well. Thanks for your responsiveness! :)21:07
jamezpNot quite that time here yet... ...well it could be, but I should probably work a little more first :)21:08
mikesir87Yeah... not yet here too. Here's the issue, in case anyone cares for now - Feel free to let me know if you want any more details/comments in it21:10
*** jclingan <jclingan!> has joined #wildfly-swarm21:13
bbrowningbobmcw: so I have to go for the weekend but I think we can use the shrinkwrap APIs to muck with the user's .jar / .war before adding it to our Archive21:15
bbrowningbasically import their .jar using ZipImporter so we now have an Archive from it21:15
bbrowningand then add/remove things from there before passing that Archive to our -swarm.jar Archive21:15
bbrowningtheir original on-disk foo.jar isn't changed but the foo.jar inside _bootstrap would be our modified version21:16
bbrowningthat still doesn't solve my exact issue, but will help in removing extraneous crap from WEB-INF/lib21:17
bbrowninga rough hacked up version I'm testing w/ locally in -
*** bbrowning is now known as bbrowning_away21:18
*** mikesir87 <mikesir87!c65269d7@gateway/web/cgi-irc/> has left #wildfly-swarm21:29
lanceballbobmcw: we are only working with subsytem in config-api21:39
*** jclingan <jclingan!> has quit IRC (Ping timeout: 260 seconds)21:44
*** dmlloyd <dmlloyd!~dmlloyd@redhat/jboss/dmlloyd> has quit IRC (Remote host closed the connection)22:07
*** helio-frota <helio-frota!> has quit IRC (Quit: Leaving)22:25
*** lanceball is now known as lance|afk22:37
*** dmlloyd <dmlloyd!~dmlloyd@redhat/jboss/dmlloyd> has joined #wildfly-swarm22:47
*** rbenevides <rbenevides!~rbenevid@redhat/jboss/RafaelBenevides> has joined #wildfly-swarm22:52
*** tcrawley is now known as tcrawley-away22:56
*** rbenevides <rbenevides!~rbenevid@redhat/jboss/RafaelBenevides> has quit IRC (Quit: Computer has gone to sleep.)23:04
*** jclingan <jclingan!~jclingan@> has joined #wildfly-swarm23:06
*** ibothwell <ibothwell!> has joined #wildfly-swarm23:31
*** jcrossley3 is now known as jcrossley3-away23:32
*** ibothwell <ibothwell!> has quit IRC (Client Quit)23:34

Generated by 2.9.2 by Marius Gedminas - find it at!