Tuesday, 2015-12-01

*** helio-frota <helio-frota!~hf@> has joined #wildfly-swarm00:00
*** enebo <enebo!~enebo@c-75-73-8-169.hsd1.mn.comcast.net> has quit IRC (Quit: enebo)00:02
helio-frotathe SWARM-80 is 'working'00:02
helio-frotaneed to organize the things because changed/run some things00:04
helio-frota80% > I have no idea what is (rake,gem, leafy stuff)00:05
*** helio-frota <helio-frota!~hf@> has quit IRC (Read error: Connection reset by peer)00:29
*** rbenevides <rbenevides!~rbenevid@redhat/jboss/RafaelBenevides> has quit IRC (Quit: Quit)00:33
*** bobmcw <bobmcw!~bobmcw@redhat/jboss/bobmcw> has joined #wildfly-swarm01:46
*** ChanServ sets mode: +o bobmcw01:46
*** bobmcw <bobmcw!~bobmcw@redhat/jboss/bobmcw> has quit IRC (Ping timeout: 260 seconds)01:51
*** jcrossley3 is now known as jcrossley3-away02:12
*** bgeorges <bgeorges!bgeorges@nat/redhat/x-nylnjdtoccibmzbc> has joined #wildfly-swarm02:18
*** bgeorges <bgeorges!bgeorges@nat/redhat/x-nylnjdtoccibmzbc> has quit IRC (Remote host closed the connection)03:00
*** bgeorges <bgeorges!bgeorges@nat/redhat/x-awoveujkpaujuate> has joined #wildfly-swarm03:00
*** bgeorges_ <bgeorges_!bgeorges@nat/redhat/x-innruuqcwzebdmgm> has joined #wildfly-swarm03:04
*** bgeorges <bgeorges!bgeorges@nat/redhat/x-awoveujkpaujuate> has quit IRC (Read error: Connection reset by peer)03:04
*** agoncal <agoncal!~agoncal@2a01cb080056980099b21fc7405b5b9e.ipv6.abo.wanadoo.fr> has joined #wildfly-swarm03:16
*** agoncal <agoncal!~agoncal@2a01cb080056980099b21fc7405b5b9e.ipv6.abo.wanadoo.fr> has quit IRC (Ping timeout: 264 seconds)03:21
*** agoncal <agoncal!~agoncal@2a01cb080056980099b21fc7405b5b9e.ipv6.abo.wanadoo.fr> has joined #wildfly-swarm04:18
*** agoncal <agoncal!~agoncal@2a01cb080056980099b21fc7405b5b9e.ipv6.abo.wanadoo.fr> has quit IRC (Ping timeout: 264 seconds)04:22
*** bgeorges_ <bgeorges_!bgeorges@nat/redhat/x-innruuqcwzebdmgm> has quit IRC (Remote host closed the connection)05:12
*** bgeorges <bgeorges!bgeorges@nat/redhat/x-pvqdsvtewercknfz> has joined #wildfly-swarm05:24
*** agoncal <agoncal!~agoncal@2a01cb080056980099b21fc7405b5b9e.ipv6.abo.wanadoo.fr> has joined #wildfly-swarm06:20
*** agoncal <agoncal!~agoncal@2a01cb080056980099b21fc7405b5b9e.ipv6.abo.wanadoo.fr> has quit IRC (Ping timeout: 264 seconds)06:24
*** bgeorges <bgeorges!bgeorges@nat/redhat/x-pvqdsvtewercknfz> has quit IRC (Remote host closed the connection)06:49
*** bgeorges <bgeorges!bgeorges@nat/redhat/x-qhwjniibbexwjyaz> has joined #wildfly-swarm06:50
*** bgeorges <bgeorges!bgeorges@nat/redhat/x-qhwjniibbexwjyaz> has quit IRC (Ping timeout: 260 seconds)06:54
*** bgeorges <bgeorges!~bgeorges@59-100-149-90.cust.static-ipl.aapt.com.au> has joined #wildfly-swarm08:31
*** hbraun <hbraun!~hbraun@pdpc/supporter/professional/hbraun> has joined #wildfly-swarm09:01
hbraungood morning09:01
*** agoncal_ <agoncal_!~agoncal@2a01cb080056980099b21fc7405b5b9e.ipv6.abo.wanadoo.fr> has joined #wildfly-swarm09:21
*** agoncal_ <agoncal_!~agoncal@2a01cb080056980099b21fc7405b5b9e.ipv6.abo.wanadoo.fr> has quit IRC (Ping timeout: 264 seconds)09:26
*** helio-frota <helio-frota!~hf@> has joined #wildfly-swarm10:01
*** helio-frota <helio-frota!~hf@> has quit IRC (Ping timeout: 260 seconds)10:06
*** helio-frota <helio-frota!~hf@> has joined #wildfly-swarm10:06
*** bbrowning_away is now known as bbrowning11:13
*** agoncal <agoncal!~agoncal@2a01cb080056980099b21fc7405b5b9e.ipv6.abo.wanadoo.fr> has joined #wildfly-swarm11:23
helio-frotabbrowning, ping11:25
helio-frota50% solved, the web app is running11:25
helio-frotabut has an error i have no idea about11:25
helio-frotaCould not index class jnr/x86asm/SEGMENT.class at /wildfly-swarm-jruby.war/WEB-INF/lib/jnr-x86asm-1.0.2.jar: java.lang.NullPointerException11:26
bbrowninghmm I'm not exactly sure what that error's about11:27
*** agoncal <agoncal!~agoncal@2a01cb080056980099b21fc7405b5b9e.ipv6.abo.wanadoo.fr> has quit IRC (Ping timeout: 264 seconds)11:28
bbrowningyou'd have to get a stacktrace showing what's throwing the NPE11:28
helio-frotathe full stack is on the issue11:28
helio-frotabbrowning, is from jruby11:32
jbossbotTitle: JRuby: Pushing the Java Platform Further11:32
helio-frotaslide 2011:32
helio-frotaso, going to investigate a bit more11:32
bbrowninghelio-frota: yep I know what JNR is. the stacktrace is jandex (a tool that indexes jar files) complaining though11:34
helio-frotabbrowning, ah thanks i never used/know this before11:34
*** bbrowning is now known as bbrowning_away11:42
*** helio-frota <helio-frota!~hf@> has quit IRC (Quit: Leaving)13:17
*** tcrawley-away is now known as tcrawley13:20
*** agoncal <agoncal!~agoncal@2a01cb080056980099b21fc7405b5b9e.ipv6.abo.wanadoo.fr> has joined #wildfly-swarm13:25
*** agoncal <agoncal!~agoncal@2a01cb080056980099b21fc7405b5b9e.ipv6.abo.wanadoo.fr> has quit IRC (Ping timeout: 264 seconds)13:30
*** lance|afk is now known as lanceball13:50
*** enebo <enebo!~enebo@c-75-73-8-169.hsd1.mn.comcast.net> has joined #wildfly-swarm13:57
tcrawleydmlloyd: is there a way to get access to the module for a deployment from another module? If I attempt to get the module "deployment.foo.jar" from either Module.getBootModuleLoader() or Module.getCallerModule(), it's not visible13:59
dmlloydit is possible13:59
dmlloydthe "deployment." thing is a dumb hack13:59
dmlloydyou have to use the right module loader to locate the deployment13:59
tcrawleyhow would I find that module loader?13:59
dmlloydit's called the "service module loader", though in the future we're going to split that into one for deployments and one for filesystem JARs etc.14:00
dmlloydright now you can get this service module loader from MSC14:00
dmlloydthere are a few other ways too, depending on what you're doing at the time that you need it14:00
tcrawleyI'm in a ServiceActivator, so MSC will work, thanks!14:01
tcrawleyperfect, thanks14:01
*** jcrossley3-away is now known as jcrossley314:02
*** bobmcw <bobmcw!~bobmcw@va-67-233-86-120.dhcp.embarqhsd.net> has joined #wildfly-swarm14:13
*** bobmcw <bobmcw!~bobmcw@va-67-233-86-120.dhcp.embarqhsd.net> has quit IRC (Changing host)14:13
*** bobmcw <bobmcw!~bobmcw@redhat/jboss/bobmcw> has joined #wildfly-swarm14:13
*** ChanServ sets mode: +o bobmcw14:13
jbossbotgit [wildfly-swarm] push master 2c51679.. Bob McWhirter Enable datasources to see shrinkwrap.14:15
jbossbotgit [wildfly-swarm] push master URL: http://github.com/wildfly-swarm/wildfly-swarm/commit/2c51679c214:15
*** kenfinnigan <kenfinnigan!~ken@redhat/jboss/kenfinnigan> has joined #wildfly-swarm14:22
kenfinniganhey bgeorges, what's the latest dates for F2F?14:32
hbraunkenfinnigan good question14:37
hbraunkenfinnigan did you see my last email response?14:37
kenfinniganhbraun: yeah, I appreciate the thought, but I'm fine with us both submitting a budget and seeing what happens14:37
kenfinnigancould always see if there's room for us to present jointly14:38
hbraunkenfinnigan that would be even better. I didn't thoroughly look at the cfp criteria, but do they also have hands on sessions?14:38
hbrauni.e. we could submit for session and hands on in that case14:39
kenfinniganhbraun: there are several formats, conf session, quickies (10 mins during lunch), and there are hands on labs for 3 hrs14:39
kenfinnigancould definitely submit for a few sessions and a lab14:39
hbraunwell, then maybe we should aim for both session and lab.14:41
*** bbrowning_away is now known as bbrowning14:42
*** helio-frota <helio-frota!~hf@> has joined #wildfly-swarm14:46
bbrowninghelio-frota: as you progress on that issue, let me know if you run into any questions around ruby / jruby usage14:47
bbrowningI have a lot of experience with those so can probably help :)14:47
helio-frotabobmcw, thanks14:47
helio-frotabbrowning, i know nothing about ruby my first app i did yesterday14:47
kenfinniganhelio-frota: would be great if you could get ruby working, tried unsuccessfully with asciidoctorj just before JavaOne14:48
helio-frotabbrowning, the app is running14:48
helio-frotakenfinnigan, it is running14:48
bbrowninghelio-frota: yeah - I saw that you removed bundler usage, which you'll probably want to get working long-term14:48
bbrowningmost ruby apps use bundler14:48
helio-frotakenfinnigan, but has 1 error jandex jnr (i don't know what is it)14:49
helio-frotakenfinnigan, http://picpaste.com/pics/hail-S9D9A6sZ.1448981392.png14:49
helio-frotaalso attached on SWARM-80 issue14:50
helio-frotabbrowning, ok i don't know what you talking about but i trust you14:51
helio-frotabbrowning, i have some questions14:53
helio-frotais required to have gemfile, mavenfile, rakefile, config.ru lots of config files to get up and running ?14:54
helio-frota4 config files14:54
bbrowninghelio-frota: none of those are required for a basic ruby app and they all serve different purposes14:54
*** rbenevides <rbenevides!~rbenevid@redhat/jboss/RafaelBenevides> has joined #wildfly-swarm14:54
bbrowninggemfile is for bundler14:54
bbrowningmavenfile is for that 'rmvn' command the app is using - it's not something used often14:54
bbrowningrakefile is for 'rake', a ruby build tool that is fairly commonly used14:55
bbrowningconfig.ru is the entrypoint for the web application14:55
bbrowningbut none are required for a basic ruby app14:55
helio-frotabbrowning, thanks a lot !14:55
helio-frotahm...config.ru is the entrypoint for WEB (war)  the final package is a .war14:59
helio-frotagood to have a .jar too14:59
helio-frotawildfly-swarm-jruby.war is the generated final package here15:00
helio-frotabtw i need to learn those things rake, bundle genfile etc15:00
bbrowningconfig.ru gets called by jruby-rack, which is configured in WEB-INF/web.xml15:01
bbrowningright :)15:03
helio-frotaalso the guy is using web-app_2_3.dtd (old stuff)15:03
helio-frotamaybe is good to try a simple baby-step15:03
bbrowningyeah that's probably just because it doesn't need a newer servlet container to run15:03
helio-frotadue leafy-health and others leafy (wtf... is this..)15:04
helio-frotaok thanks bbrowning !15:04
bbrowningthe leafy stuff is from https://github.com/lookout/leafy15:05
helio-frota'The green way to build Ruby-based web services'15:05
helio-frotaruby is possible todo something like undertow-js fraction ?15:06
helio-frotawithout frameworks/libs ?15:06
helio-frotalike this: https://github.com/emag/wildfly-swarm-quickstarts/blob/master/undertowjs/src/main/webapp/hello.js15:07
bbrowningsure it would be possible to have a ruby example like that15:07
helio-frotagood so i have 1 idea to play around15:07
helio-frotatry to create a clojure, sacala, jruby, java, groovy and javascript simple hello15:08
helio-frotajava,groovy,javascript ( is working fine )15:08
helio-frotajruby 50%15:09
helio-frotascala and clojure should work, but the problem (maybe) is the build tools15:09
helio-frotawithout db access and others frameworks/libs as possible15:10
bobmcwfwiw Christian Meier is mkristian, right?  so he's be doing weird rmvn jruby stuff by default15:11
bobmcwnot exactly "traditional" ruby15:11
helio-frotabobmcw, thanks for the warning15:13
*** hbraun <hbraun!~hbraun@pdpc/supporter/professional/hbraun> has quit IRC (Quit: afk)15:18
bbrowningwhee! swarm services are communicating with each other via kubeping, although clustering isn't working yet15:19
bbrowningbut I can at least see service A send a discovery request to service B and get a response15:19
kenfinniganbbrowning: awesome!15:21
helio-frotaThank you all also feel happy to be here in this channel with very experienced professionals who make history in software development15:23
bbrowningkenfinnigan: baby steps - now I have to dig into jgroups itself a bit more to see why that communication isn't resulting in working clustering15:24
kenfinniganbbrowning: sure, but you're further than you were, which is always nice15:25
bobmcwso the ribbon topology isn't actually topologying?15:26
bobmcwif they are PINGing, they should be clustering, per se15:26
bobmcwsince we just use jgroups.sendToAllOtherNodes( request )15:26
bbrowningbobmcw: well ribbon isn't to blame - I think there are some missing pieces in the kubeping impl15:27
*** agoncal <agoncal!~agoncal@2a01cb080056980099b21fc7405b5b9e.ipv6.abo.wanadoo.fr> has joined #wildfly-swarm15:27
bbrowningbut if so that means this never worked or I'm doing it wrong15:27
bobmcwI'd check with mgoldmann15:27
jbossbotgit [wildfly-swarm-examples] push master 01c939e.. Bob McWhirter Add and re-arrange messaging into the tested examples.15:28
jbossbotgit [wildfly-swarm-examples] push master URL: http://github.com/wildfly-swarm/wildfly-swarm-examples/commit/01c939ec715:28
bbrowningahh looks like my current issue is fixed upstream in openshift-ping already perhaps :)15:29
*** mgoldmann <mgoldmann!~mgoldmann@redhat/jboss/mgoldmann> has joined #wildfly-swarm15:29
bbrowningmgoldmann: morning!15:30
bbrowningmgoldmann: do you know if jboss-openshift/openshift-ping is actually used somewhere?15:30
mgoldmannbbrowning: yes15:30
bbrowningie is it tested and known to work for some version of eap or wildfly?15:30
mgoldmannin our product images15:30
mgoldmannEAP 6.4.415:30
mgoldmannthis is what we test it with15:31
mgoldmannor better: use15:31
bbrowningok - so perhaps the one upstream change newer than what I have is what I'm missing then15:31
bbrowningI forked the repo a few days too early ;)15:31
bbrowningI have a fork of it I'm using with swarm and thus wildfly 1015:31
mgoldmannI think we use beta7?15:31
*** agoncal <agoncal!~agoncal@2a01cb080056980099b21fc7405b5b9e.ipv6.abo.wanadoo.fr> has quit IRC (Ping timeout: 264 seconds)15:31
mgoldmannyah, beta 7 is in our latest images15:32
bbrowningyeah I forked from beta5 but there's only 1 commit (other than version #s) between beta5 and beta715:32
mgoldmannwhich are targetting EAP 6.4.4, and few other products15:32
mgoldmannyah, it rearely changes15:32
bbrowningright that's one reason I forked - I need to work with a newer undertow and newer jgroups15:32
mgoldmannand it seems it just works15:32
mgoldmannI do not remember any reports lately15:33
bobmcwkenfinnigan: hey..we might need to adjust perms on our JIRA as to who can move issues around on the kanban board15:33
mgoldmannbbrowning: if you need something, open an issue and let me know15:33
bobmcwI think helio-frota moved something into (and then back out of) "in development" :)15:33
mgoldmannI see youopened one already :)15:33
mgoldmannbbrowning: we build this internally15:34
mgoldmannusing brew15:34
bbrowningmgoldmann: yeah - I'm publishing actual releases to maven central under a different groupId15:34
helio-frotabobmcw, yes15:34
helio-frotamgoldmann, are you from openshift ?15:34
kenfinniganbobmcw: don't know if we can control issues being moved through kanban stages15:35
helio-frotacan i create a free account to publish examples of wf-swarm ?15:35
mgoldmannhelio-frota: nope :)15:35
kenfinniganwere you wanting to tighten or loosen?15:35
bbrowningmgoldmann: org.projectodd.openshift.ping:openshift-ping-kube:1.0.0.Beta5-swarm-5 as an example of my published GAVs15:35
mgoldmannhelio-frota: JBoss here, but a bit closer to OS :)15:35
bobmcwkenfinnigan: tighten... arbitrary JIRA users shouldn't be able to push things around15:35
kenfinniganbobmcw: let me check15:35
bbrowninghelio-frota: anyone can create a free openshift account - sign up button at https://www.openshift.com/15:36
jbossbotTitle: OpenShift by Red Hat15:36
bobmcw(no offense helio-frota :)15:36
bbrowningthat's openshift online and not the latest version of openshift enterprise or openshift origin15:36
bbrowningit's a version behind the on-premise / open-source ones15:36
helio-frotabobmcw, i don't understand the offense yet haha15:37
helio-frotabbrowning, thanks15:37
helio-frotabobmcw, i'm app dev for years(14) so things looks like dumb when trying to not write apps and write tools/whatever :P15:40
bobmcwah, we all learn new things, that's no worries15:40
bobmcwI meant by calling you "an arbitrary JIRA user" moving issues around15:40
bobmcwwe like you, but you found where our tool is too open apparently :)15:40
helio-frotabobmcw, got it15:41
helio-frotabobmcw, booker is good but is complex for users because them have no time to learn in 8hours of day-by-day job15:42
bobmcwyah, for sure15:43
helio-frotai like booker (but it need to learn other things from)15:43
bbrowningmgoldmann: since you're here, how am I supposed to get the exposed docker port 8888 labeled as 'ping' in kubernetes?15:46
bbrowningI could never figure that out short of manually 'oc edit ...' my service15:46
bbrowningfor now I added code to fallback to the default port number if the port wasn't found by name to workaround it15:47
mgoldmannbbrowning: one sec, I'll invite someone here15:47
jbossbotgit [wildfly-swarm-examples] push master ea7e554.. Bob McWhirter Rearrange and test servlet-based examples.15:47
jbossbotgit [wildfly-swarm-examples] push master URL: http://github.com/wildfly-swarm/wildfly-swarm-examples/commit/ea7e554e415:47
* helio-frota lunch time15:52
kenfinniganbobmcw: restrict transition to project lead and assignee?15:53
kenfinniganbobmcw: transition restrictions should now be in place15:56
mgoldmannbbrowning: still working on the guy :)15:57
bbrowningmgoldmann: it's not a big deal to answer right now - but at some point I'd love to know how :)15:57
bobmcwkenfinnigan: thanks!15:58
*** errantepiphany <errantepiphany!~errantepi@> has joined #wildfly-swarm16:01
jbossbotgit [wildfly-swarm] push master f6317bd.. Ben Browning Bump to openshift-ping 1.0.0.Beta7-swarm-116:03
jbossbotgit [wildfly-swarm] push master URL: http://github.com/wildfly-swarm/wildfly-swarm/commit/f6317bdc116:03
mgoldmannbbrowning: david is getting some fodd and will join you in a bit :)16:05
mgoldmannand it's time for me now, laters!16:05
*** mgoldmann is now known as mgoldmann|away16:05
bbrowningmgoldmann|away: thanks - later!16:05
errantepiphanybbrowning: What can I do for you? (ps, I'm David Ward, Red Hat employee on the Cloud Enablement team.)16:21
bbrowningerrantepiphany: morning! I'm using a forked version of openshift-ping to try and get wildfly swarm clustering via kubernetes16:23
bbrowningand I was wondering how I get the ping port of 8888 used by kubeping to get exposed with a name of 'ping' in the kubernetes API16:23
errantepiphanybbrowning: ah, yes - I'll show you. one sec.16:24
errantepiphanybbrowning: You need to specify the port in your app template like so: https://github.com/jboss-openshift/application-templates/blob/master/eap/eap64-basic-s2i.json#L28816:25
*** meastman <meastman!~meastman@> has joined #wildfly-swarm16:26
bbrowningerrantepiphany: ahh - right now I'm not using an app template and instead doing oc new-app ... wildflyswarm-10-centos7~https://github.com/foo/bar16:26
bbrowningI have a source-to-image swarm and create a new app from source with it16:27
bbrowningthe goal is that a user can take any existing swarm app and just run it on openshift without having to modifying their source repo16:27
bbrowningat least for basic cases16:27
errantepiphanygotcha. Well, there's a few things in that template that are important. First, the ping port. But then, also the OPENSHIFT_KUBE_PING_NAMESPACE so the code can do an appropriate kubernetes call, and _LABELS so the code can filter to just the pods that should be clustered together.16:28
bbrowningyeah for now I'm setting OPENSHIFT_KUBE_PING_NAMESPACE to OPENSHIFT_BUILD_NAMESPACE in my source-to-image run script16:28
errantepiphanybbrowning: afaik the "new-app" methodology is not something we exercise much yet on our team. We rely pretty heavily on things that are configured in the app templates.16:29
bbrowningIt's good to have this template to look at so I know what to expect. I also added logic to fallback to the default ping port if it can't find a port of name 'ping' for now.16:29
bbrowningOnce I get things working I may look at creating some template(s) to simplify the deployment of our booker sample app16:30
errantepiphanybbrowning: the OPENSHIFT_BUILD_NAMESPACE probably shouldn't be depended on if you're ever going to use this for non-s2i cases. Which is why you can see here how we always get it dynamically:16:30
bbrowningahh yeah using the downward API16:30
bbrowninggood to know :)16:30
*** meastman <meastman!~meastman@> has left #wildfly-swarm16:31
bbrowningI haven't needed labels yet - in my case I have microservices and each service type is a different app in my project16:32
bbrowningwhich can be scaled independently16:32
bbrowningso for now I've assumed everything with a ping port gets clustered together as one16:33
bbrowninggood to know labels is how I can influence that :)16:33
errantepiphanybbrowning: +116:34
*** os890 <os890!~chatzilla@178-190-77-97.adsl.highway.telekom.at> has joined #wildfly-swarm16:34
*** jclingan <jclingan!~jclingan@redhat/jclingan> has joined #wildfly-swarm17:25
*** agoncal <agoncal!~agoncal@> has joined #wildfly-swarm17:38
*** agoncal_ <agoncal_!~agoncal@> has joined #wildfly-swarm17:42
*** agoncal <agoncal!~agoncal@> has quit IRC (Ping timeout: 250 seconds)17:43
agoncal_Hi guys17:45
agoncal_I'm trying to package the "smallest possible" microservice ever : a servlet17:45
agoncal_I use the wildfly-swarm-undertow  part... but I find it still too big (fat jar is 30Mb)17:46
agoncal_Can I shrink it even more to just focus on servlet ?17:46
bbrowningagoncal_: if you look inside the jar, are there things that stand out as being unnecessary?17:49
bbrowningit's definitely possible to hand craft a fat jar far smaller than that to just run a servlet with undertow17:50
bbrowningso hopefully there's some room to slim down the swarm-produced jar as well17:50
agoncal_jboss-javax-sql, jboss-websocket-api, wildfly-swarm-ee*, and also wildfly-swarm-naming* (is that JNDI ?)17:53
*** agoncal <agoncal!~agoncal@2a01cb080056980099b21fc7405b5b9e.ipv6.abo.wanadoo.fr> has joined #wildfly-swarm17:54
agoncal_(sorry, connection is very bad, keeps connecting/unconnecting)17:57
bbrowningnaming is jndi, yes17:58
*** agoncal <agoncal!~agoncal@2a01cb080056980099b21fc7405b5b9e.ipv6.abo.wanadoo.fr> has quit IRC (Ping timeout: 264 seconds)17:58
jclinganagoncal_: I was going to try a similar experiment, comparing Spring Boot jar with Swarm Jar to see what is different.17:58
agoncal_jclingan I was going to try a similar experiment with KulumuzEE (http://ee.kumuluz.com/) ;o)18:01
jbossbotStatus 40318:01
agoncal_It's actually the topic of my coming blog posts : writing a microservice, a nanoservice, a picoservice ;o)18:01
*** bbrowning is now known as bbrowning_away18:01
jclinganagoncal_: LOL. Yeah, we’ve talked about slimming down the jar, but there were othing topics we wanted to get done first.18:01
jclinganThe question is, what’s small enough? How important is small? “small” is subjective on a developer-by-developer basis.18:02
agoncal_In the jars I've mentioned for a servlet, any one optional ? (jboss-javax-sql, jboss-websocket-api, wildfly-swarm-ee*, and also wildfly-swarm-naming* )18:04
kenfinniganagoncal_: websocket is probably optional, unless you're using websockets18:04
kenfinniganjavax sql is needed because it's part of the JDK bits, I think18:05
agoncal_jclingan I completelly agree. To be honest, sometimes I think the entire "microservice" thing, is just a trend : how small is small ?18:05
jclinganIncreasing granularity can also means increasing the number of maven dependencies (adding dev-time complexity) unless we alter our approach a bit.18:05
agoncal_In jboss-javax-sql I have the JDBC API... couldn't that be an optional part ?18:06
kenfinniganagoncal_: unfortunately not. the javax.api module from WildFly has a module dependency on javax.sql.api18:07
kenfinniganso the former won't start without the latter being present18:07
agoncal_Yes, like for example splitting shrinkwrap-descriptors-api-javaee  (this one is nearly 1Mb) into the appropriate sub-specs (ex.  shrinkwrap-descriptors-api-servlet, ...)18:08
kenfinniganit would likely be possibly to mung the module.xml to remove the reference to sql and it would still work18:08
kenfinnigani don't believe shrinkwrap has separate jars for each sub spec, so that limits what we can do18:08
agoncal_And what about naming* ? Related to JNDI ?18:10
bobmcwerrantepiphany: are you the David Ward full of tattoos and farms chickens?18:10
kenfinniganthe problem stems from WildFly itself having the interdependencies between systems18:11
bobmcwyah, websocket is optional, but also small, so we had to decide if it was worth wildfly-swarm-websockets or not, as a separate dependency18:11
bobmcwnaming, I'm not sure if that's optional or required internally18:12
errantepiphanybobmcw: yes re: the tattoos, not any more re: the chickens. :)18:12
bobmcwthe sql spec, I dunno why, but it seems very required, hence our inclusion at the -container level18:12
bobmcwit baffles me.18:12
bobmcwerrantepiphany: ah, we met when you first joined as an SA, I think, during JavaOne via Ploski18:12
kenfinniganyeah, i don't get it either18:13
errantepiphanybobmcw: yup, man that feels like a *long* time ago!18:13
bobmcwthanks for answering our openshift questions :)18:13
bobmcwand for putting up with mgoldmann|away18:13
errantepiphanybobmcw: lol np :)18:14
errantepiphanybobmcw: Ales Justin made the first version of openshift-ping, I refactored it a bit and added some features, then Rob Cernich stepped in and added AMQ support and made some other bug fixes. It's evolved quite a bit, but honestly could also be cleaned up since there's been several hands in it.18:15
*** errantepiphany is now known as ee_gym18:15
bobmcwagoncal_: kenfinnigan: wrt the non-breakable things that should be optional, but aren't... we can work upstream to get WF less intertwingly, long-term18:20
bobmcwand tcrawley's work to get module.xml shrinkwrap descriptors (which I've been remiss about using correctly) should allow us to rewrite upstream module.xmls in our build, if that's suitable18:21
agoncal_bobmcw I'm trying Kumulzee, TomEE and Spring Boot... I will compare it with Swarm (so we get an idea)18:21
bobmcwagoncal_: sounds good18:21
kenfinniganbobmcw: that would be a good way to experiment and find out what can/can't be stripped and still make it work18:21
bobmcwthere's always room for improvement, of course18:21
bobmcwkenfinnigan: right, w/o having to muck with upstream initially18:22
bobmcwit's on my list18:22
*** lanceball is now known as lance|afk18:25
jbossbotgit [wildfly-swarm-examples] push master ff9af8a.. Bob McWhirter Rearrange and add tests for the jpa-jaxrs-cdi examples.18:28
jbossbotgit [wildfly-swarm-examples] push master URL: http://github.com/wildfly-swarm/wildfly-swarm-examples/commit/ff9af8a8818:28
bobmcwtcrawley: TESTS18:28
*** agoncal_ <agoncal_!~agoncal@> has quit IRC (Ping timeout: 245 seconds)18:36
*** bbrowning_away is now known as bbrowning18:41
bbrowningkenfinnigan: bobmcw: this discussion is partly why in torquebox 4 / immutant 2 we decided to not build on top of any embedded wildfly bits and instead embed undertow, infinispan, etc directly18:44
bbrowningbecause wildfly seemed to have everything so intermingled that it wasn't possible to get everything small enough18:44
bbrowningif you want a target to hit, the TB4 equal a combined total of 10705 KB if you use all its features of messaging, scheduling, web, caching18:46
bbrowningso our fatjars are just that size (if you use all features, less if not) plus your app plus your app's dependencies and jruby if you choose to bundle those in18:47
bbrowningit would be useful to know how much of agoncal_'s 30MB servlet fatjar is specific to swarm and its deps18:48
bbrowningvs the app and m2repo stuff the app needs18:48
tcrawleyanyone with arquillian experience - any idea about this exception? https://gist.github.com/a5b16c603f24227a944b18:49
tcrawleyarq is loading the test class inside swarm, then puking with that when it tries to execute the test18:49
tcrawleyand it doesn't actually execute the test18:50
tcrawleythat error means the result of the test run was null18:50
kenfinnigantcrawley: not seen an error like that before18:50
tcrawleygoogling around, I found that alesj had a similar error, which was a missing dep in his app18:51
tcrawleybut I have the dep he was missing18:51
tcrawleythe arq code is a little obtuse18:51
*** lance|afk is now known as lanceball18:56
bobmcwtcrawley: no idea19:07
bobmcwyou should probably write a test19:07
tcrawleyor become a plumber or something19:08
bobmcwdoughnut-maker is my fallback19:08
tcrawleyyou have to get up too early for that19:08
bobmcwgluten-free homeopathic doughnuts19:08
tcrawleyin that case, you won't have to get up early. one donut will go a long way19:09
bobmcwtcrawley: I'll cater to the late-night doughnut aficionado19:09
bobmcwbbrowning: seems your static content fixes break if wildfly-swarm:run but PWD doesn't match your expectations19:12
bbrowningI have no expectations.19:12
bobmcwsomething does19:12
bobmcwif I run the test from the project dir, it works19:12
bobmcwif I run from a root of a multi-module project, it doesn't19:13
bbrowningis the test bundling the static content into the jar? or trying to serve assploded off disk?19:13
bobmcwmaybe wildfly-swarm:start needs to set a PWD better19:13
bbrowningso off-disk how can we know what directory is right if not using pwd?19:13
bbrowninghow did that work before?19:13
jbossbotgit [wildfly-swarm-examples] push master 8a44f3b.. Bob McWhirter Add transactions example into the test....19:14
jbossbotgit [wildfly-swarm-examples] push master URL: http://github.com/wildfly-swarm/wildfly-swarm-examples/commit/8a44f3b9219:14
bobmcwmaybe it didn't19:14
bobmcwI didn't have the tests actually failing the build, it seems19:14
bobmcwI'll investigate our SwarmExecutor and see how it's dealing with PWD (or not)19:14
bbrowningthere may be some other magic we can do somehow to figure out the right static path in cases like that19:15
bbrowningmaybe maven sets a property for the current module's root dir we can look at19:16
*** jbossbot <jbossbot!~JBossBot@redhat/jbossbot> has quit IRC (Ping timeout: 244 seconds)19:18
*** jclingan <jclingan!~jclingan@redhat/jclingan> has quit IRC (Quit: jclingan)19:19
*** jclingan <jclingan!~jclingan@redhat/jclingan> has joined #wildfly-swarm19:21
*** jclingan <jclingan!~jclingan@redhat/jclingan> has quit IRC (Quit: jclingan)19:48
*** jclingan <jclingan!~jclingan@redhat/jclingan> has joined #wildfly-swarm19:50
*** agoncal <agoncal!~agoncal@2a01cb080056980099b21fc7405b5b9e.ipv6.abo.wanadoo.fr> has joined #wildfly-swarm19:51
*** jclingan <jclingan!~jclingan@redhat/jclingan> has quit IRC (Ping timeout: 260 seconds)19:57
*** jclingan <jclingan!~jclingan@redhat/jclingan> has joined #wildfly-swarm19:57
*** ee_gym is now known as errantepiphany19:57
*** errantepiphany <errantepiphany!~errantepi@> has left #wildfly-swarm19:58
*** jclingan <jclingan!~jclingan@redhat/jclingan> has quit IRC (Ping timeout: 260 seconds)20:01
*** jclingan <jclingan!~jclingan@redhat/jclingan> has joined #wildfly-swarm20:02
*** agoncal <agoncal!~agoncal@2a01cb080056980099b21fc7405b5b9e.ipv6.abo.wanadoo.fr> has quit IRC (Ping timeout: 264 seconds)20:05
*** bbrowning <bbrowning!~bbrowning@redhat/jboss/bbrowning> has quit IRC (Quit: Leaving)20:08
*** jclingan <jclingan!~jclingan@redhat/jclingan> has quit IRC (Quit: jclingan)20:12
*** bbrowning <bbrowning!~bbrowning@redhat/jboss/bbrowning> has joined #wildfly-swarm20:12
*** jclingan <jclingan!~jclingan@redhat/jclingan> has joined #wildfly-swarm20:13
*** jclingan <jclingan!~jclingan@redhat/jclingan> has quit IRC (Quit: jclingan)20:23
*** jclingan <jclingan!~jclingan@redhat/jclingan> has joined #wildfly-swarm20:25
*** jclingan <jclingan!~jclingan@redhat/jclingan> has quit IRC (Ping timeout: 260 seconds)20:30
*** kenfinnigan <kenfinnigan!~ken@redhat/jboss/kenfinnigan> has quit IRC (Quit: Leaving.)20:37
*** agoncal <agoncal!~agoncal@2a01cb080056980099b21fc7405b5b9e.ipv6.abo.wanadoo.fr> has joined #wildfly-swarm20:59
*** agoncal_ <agoncal_!~agoncal@2a01cb08005698001c18eba9040d3a4c.ipv6.abo.wanadoo.fr> has joined #wildfly-swarm21:05
*** agoncal <agoncal!~agoncal@2a01cb080056980099b21fc7405b5b9e.ipv6.abo.wanadoo.fr> has quit IRC (Ping timeout: 264 seconds)21:06
*** os890 <os890!~chatzilla@178-190-77-97.adsl.highway.telekom.at> has quit IRC (Ping timeout: 260 seconds)21:19
*** agoncal_ <agoncal_!~agoncal@2a01cb08005698001c18eba9040d3a4c.ipv6.abo.wanadoo.fr> has quit IRC (Ping timeout: 264 seconds)21:36
*** jclingan <jclingan!~jclingan@redhat/jclingan> has joined #wildfly-swarm21:54
helio-frotai like it , first time i saw test transaction via web21:55
*** tcrawley is now known as tcrawley-away22:05
*** jclingan <jclingan!~jclingan@redhat/jclingan> has quit IRC (Read error: Connection reset by peer)22:17
*** jclingan <jclingan!~jclingan@redhat/jclingan> has joined #wildfly-swarm22:17
*** lanceball is now known as lance|afk22:20
*** bbrowning is now known as bbrowning_away22:23
*** enebo <enebo!~enebo@c-75-73-8-169.hsd1.mn.comcast.net> has quit IRC (Quit: enebo)22:39
*** jclingan <jclingan!~jclingan@redhat/jclingan> has quit IRC (Quit: jclingan)22:41
*** bgeorges_ <bgeorges_!~bgeorges@59-100-149-90.cust.static-ipl.aapt.com.au> has joined #wildfly-swarm22:43
*** jclingan <jclingan!~jclingan@redhat/jclingan> has joined #wildfly-swarm22:44
*** bgeorges <bgeorges!~bgeorges@59-100-149-90.cust.static-ipl.aapt.com.au> has quit IRC (Ping timeout: 260 seconds)22:44
*** jclingan <jclingan!~jclingan@redhat/jclingan> has quit IRC (Quit: jclingan)22:55
*** jclingan <jclingan!~jclingan@redhat/jclingan> has joined #wildfly-swarm22:55
*** rbenevides <rbenevides!~rbenevid@redhat/jboss/RafaelBenevides> has quit IRC (Quit: Computer has gone to sleep.)23:03
*** bobmcw <bobmcw!~bobmcw@redhat/jboss/bobmcw> has quit IRC (Remote host closed the connection)23:06
*** rbenevides <rbenevides!rbenevid@redhat/jboss/RafaelBenevides> has joined #wildfly-swarm23:45
*** rbenevid_ <rbenevid_!rbenevid@redhat/jboss/RafaelBenevides> has joined #wildfly-swarm23:55
*** rbenevides <rbenevides!rbenevid@redhat/jboss/RafaelBenevides> has quit IRC (Ping timeout: 260 seconds)23:58

Generated by irclog2html.py 2.9.2 by Marius Gedminas - find it at mg.pov.lt!