Wednesday, 2015-11-25

*** BenT15 <BenT15!~btomasini@75-146-60-229-Washington.hfc.comcastbusiness.net> has joined #wildfly-swarm00:07
*** enebo <enebo!~enebo@c-75-73-8-169.hsd1.mn.comcast.net> has quit IRC (Quit: enebo)00:13
*** BenT15 <BenT15!~btomasini@75-146-60-229-Washington.hfc.comcastbusiness.net> has quit IRC (Ping timeout: 260 seconds)00:15
*** enebo <enebo!~enebo@c-75-73-8-169.hsd1.mn.comcast.net> has joined #wildfly-swarm00:34
projectodd-ciStarting build #348 for job wildfly-swarm (previous build: SUCCESS)00:47
*** enebo <enebo!~enebo@c-75-73-8-169.hsd1.mn.comcast.net> has quit IRC (Quit: enebo)00:52
*** agoncal <agoncal!~agoncal@2a01cb08005698002df5151E1D925Ab8.ipv6.abo.wanadoo.fr> has joined #wildfly-swarm01:08
*** agoncal <agoncal!~agoncal@2a01cb08005698002df5151E1D925Ab8.ipv6.abo.wanadoo.fr> has quit IRC (Ping timeout: 250 seconds)01:13
*** wangc <wangc!~wangc@redhat/jboss/wangc> has joined #wildfly-swarm02:47
*** rbenevides <rbenevides!~rbenevid@redhat/jboss/RafaelBenevides> has quit IRC (Quit: Quit)03:05
*** agoncal <agoncal!~agoncal@2a01cb08005698002DF5151e1D925Ab8.ipv6.abo.wanadoo.fr> has joined #wildfly-swarm03:09
*** jerr <jerr!~jerr@modemcable227.115-83-70.mc.videotron.ca> has joined #wildfly-swarm03:11
*** agoncal <agoncal!~agoncal@2a01cb08005698002DF5151e1D925Ab8.ipv6.abo.wanadoo.fr> has quit IRC (Ping timeout: 250 seconds)03:14
*** jerr <jerr!~jerr@modemcable227.115-83-70.mc.videotron.ca> has quit IRC (Ping timeout: 265 seconds)03:33
*** hbraun <hbraun!~hbraun@pdpc/supporter/professional/hbraun> has joined #wildfly-swarm06:24
*** hbraun <hbraun!~hbraun@pdpc/supporter/professional/hbraun> has quit IRC (Client Quit)06:26
*** BenT15 <BenT15!~btomasini@c-73-19-28-3.hsd1.wa.comcast.net> has joined #wildfly-swarm06:36
*** BenT15 <BenT15!~btomasini@c-73-19-28-3.hsd1.wa.comcast.net> has quit IRC (Ping timeout: 252 seconds)07:08
*** agoncal <agoncal!~agoncal@2a01cb08005698002DF5151E1d925ab8.ipv6.abo.wanadoo.fr> has joined #wildfly-swarm07:10
*** agoncal <agoncal!~agoncal@2a01cb08005698002DF5151E1d925ab8.ipv6.abo.wanadoo.fr> has quit IRC (Ping timeout: 250 seconds)07:33
*** hbraun <hbraun!~hbraun@pdpc/supporter/professional/hbraun> has joined #wildfly-swarm07:39
*** agoncal <agoncal!~agoncal@2a01cb08005698002DF5151E1d925ab8.ipv6.abo.wanadoo.fr> has joined #wildfly-swarm08:29
*** agoncal <agoncal!~agoncal@2a01cb08005698002DF5151E1d925ab8.ipv6.abo.wanadoo.fr> has quit IRC (Ping timeout: 250 seconds)08:34
*** agoncal_ <agoncal_!~agoncal@2a01cb08005698002DF5151E1d925ab8.ipv6.abo.wanadoo.fr> has joined #wildfly-swarm09:30
*** agoncal_ <agoncal_!~agoncal@2a01cb08005698002DF5151E1d925ab8.ipv6.abo.wanadoo.fr> has quit IRC (Ping timeout: 250 seconds)09:35
*** BenT15 <BenT15!~btomasini@c-73-19-28-3.hsd1.wa.comcast.net> has joined #wildfly-swarm10:05
*** BenT15 <BenT15!~btomasini@c-73-19-28-3.hsd1.wa.comcast.net> has quit IRC (Ping timeout: 272 seconds)10:10
*** myfear <myfear!Markus@nat/redhat/x-gpdpptfhbbgtkbhy> has joined #wildfly-swarm10:18
*** qmx <qmx!~qmx@redhat/jboss/qmx> has quit IRC (Quit: ZNC - http://znc.in)10:25
*** wangc <wangc!~wangc@redhat/jboss/wangc> has quit IRC (Quit: Leaving)10:28
*** qmx <qmx!~qmx@v01.codepizza.org> has joined #wildfly-swarm10:30
*** qmx <qmx!~qmx@redhat/jboss/qmx> has joined #wildfly-swarm10:30
*** myfear <myfear!Markus@nat/redhat/x-gpdpptfhbbgtkbhy> has quit IRC (Quit: Verlassend)11:47
*** bobmcw <bobmcw!~bobmcw@va-67-233-86-120.dhcp.embarqhsd.net> has joined #wildfly-swarm13:01
*** bobmcw <bobmcw!~bobmcw@redhat/jboss/bobmcw> has joined #wildfly-swarm13:01
*** ChanServ sets mode: +o bobmcw13:01
*** rbenevides <rbenevides!~rbenevid@redhat/jboss/RafaelBenevides> has joined #wildfly-swarm13:08
*** tcrawley-away is now known as tcrawley13:28
*** rbenevides is now known as rbenevides|afk13:29
*** jclingan <jclingan!~jclingan@redhat/jclingan> has quit IRC (Quit: jclingan)13:46
hbraunbobmcw good morning13:48
hbraundo you need a heads up regarding yesterdays discussion with brian?13:49
*** lanceball is now known as lance|afk13:51
*** lance|afk is now known as lanceball13:51
lanceballhbraun: I think bobmcw is out most of the day today, fwiw13:58
hbraunlanceball ah, that's good to know13:59
lanceballholiday tomorrow which often involves lots of food/cooking/effort/etc if one is hosting (which I think bobmcw is)14:00
*** jclingan <jclingan!~jclingan@redhat/jclingan> has joined #wildfly-swarm14:01
bobmcwI'm sorta around14:04
bobmcwhbraun: brian summarized14:04
bobmcwI am smoking turkey tomorrow, half-day-ish today14:05
*** BenT15 <BenT15!~btomasini@c-73-19-28-3.hsd1.wa.comcast.net> has joined #wildfly-swarm14:05
projectodd-ciStarting build #349 for job wildfly-swarm (previous build: SUCCESS)14:08
hbraunbobmcw well maybe a few comments from my side, while you are around14:09
bobmcwsure thing!14:09
bobmcwbrian noted that he's expecting a JIRA for the WF side of things?14:09
*** kenfinnigan <kenfinnigan!~ken@redhat/jboss/kenfinnigan> has joined #wildfly-swarm14:09
bobmcwmight be a good place to track all of it14:10
hbraunyes, I'll create that an send a PR for the initial API change, which is fairly small14:10
*** BenT15 <BenT15!~btomasini@c-73-19-28-3.hsd1.wa.comcast.net> has quit IRC (Ping timeout: 264 seconds)14:10
hbraunin the discussion we balanced the ideas of reuse of what's there, versus flexibility on the swarm side of things14:11
hbraunbobmcw when users come with a standalone.xml, what do you think where it comes from?14:12
bobmcwstandalone/configuration/standalone.xml14:13
bobmcworiginally14:13
bobmcwwhen used with -swarm, I don't really know14:13
hbraunto what degree has is been modified?14:13
hbraunor adopted for swarm?14:13
bobmcwprobably a bit, as time progresses14:13
bobmcwe.g., folks who use WF, but want a smaller WF and get started in microservices14:14
hbraunyou mean they've added config's for ds, jms, logging, etc14:14
bobmcwso they might be deleting entire swaths14:14
bobmcwand yah, jms/ds primarily14:14
bobmcwand removing stuff like... infinispan or whatnot14:14
hbraunok, that's where i wanted to get to14:14
hbrauncustomisation of the server profile14:14
hbrauni.e. removing subsystem and such14:14
bobmcwI think it'd be useful, in -swarm, to look for META-INF/standalone.xml in their .jar (or somewhere), overridable via argv[]14:15
bobmcwe.g., if they run: java -jar myapp-swarm.jar # should work14:15
bobmcwor java -jar myapp-swarm.jar --config modified-standalone.xml # should also work14:15
bobmcwand maybe ask tcrawley to make swarmtool eventually be able to replace an internal standalone.xml in a -swarm.jar w/o a full rebuild14:16
bobmcw./bin/swarmtool update config myapp-swarm.jar new-config.xml14:16
hbraunyes, we get to that in a minute14:16
hbrauni wanted to ask something else first14:16
hbraunso people customise the server profile14:16
bobmcwright14:17
hbraunthe problem is see is when the runtime bits (modules) don't match what's in the xml14:17
bobmcwwhat's your current strategy for that?14:17
hbraunyesterday i've outlined two cases:14:18
bobmcw"I saw <subsystem=infinispan> but you don't have the infinispan support" log?14:18
hbrauna) xml for mere configuration, but the modules are determined by fractions the regular maven way14:18
hbraunb) xml for both configuration and resolution of fractions (aka modules)14:18
bobmcwbeing the <extension> bits?14:18
bobmcwor at run-time, bringing various wildfly-swarm-* down the network to the process based on the .xml?14:19
*** kenfinnigan <kenfinnigan!~ken@redhat/jboss/kenfinnigan> has quit IRC (Quit: Leaving.)14:19
hbrauni think we should expect both cases14:19
*** kenfinnigan <kenfinnigan!~ken@redhat/jboss/kenfinnigan> has joined #wildfly-swarm14:19
bobmcwand maybe INFO or some log saying "by the way, add <dependency>x</dependency> to your pom.xml?"14:20
hbraunthe point is however the runtime is assembled (with or without xml) requires us to be more or less lenient about the parsing14:20
hbraunin WF parsing is very strict14:20
bobmcwrecognizing and ignoring or reacting to unknown <elements> basically14:20
bobmcwnot just failing14:21
hbraunexactly14:21
bobmcwsounds good14:21
bobmcwI think you've got a great handle on it all14:21
bobmcwcertainly more than I do :)14:21
hbrauni did argue for flexibility14:21
hbraunsince we don't not all the ways in which we use standalone.xml14:21
hbrauncurrently n the prototype, it's very lenient about missing things14:22
tcrawleybobmcw: can you request perms for me to push to org.wildfly.swarm on oss.sonatype? I want to publish arq-daemon there14:22
hbraunif the fractions is not declared (maven) then the parser skips the subsystem14:22
bobmcwand even if we can't Just Make Things Work, if we can at least catch errors and report -swarm-centric errors instead of StandaloneXmlParser-centric errors14:22
tcrawleyor I can pub to org.projectodd, it latters not14:22
tcrawleymatters*14:23
bobmcwtcrawley: sure, lemme figure out how to ask again14:23
hbraunso the mechanic for determining the runtime modules is maven deps, like we started of with14:23
tcrawleyI think it's filing a jira, or commenting on the setup jira14:23
bobmcwwhat's your username?14:23
tcrawleytcrawley14:23
bobmcwthis is closed, so I'll open a new14:24
bobmcwhttps://issues.sonatype.org/browse/OSSRH-1830814:24
hbraunrefering to our previous discussion however its likely change, or at least alternative approaches are being anticipated14:24
tcrawleydonkey14:24
hbraunlike resolving the fractions from the xml14:24
hbraunthis is where I see the overlap with tcrawley 's work14:24
hbraun(kind of)14:24
*** enebo <enebo!~enebo@c-75-73-8-169.hsd1.mn.comcast.net> has joined #wildfly-swarm14:24
bobmcwtcrawley: https://issues.sonatype.org/browse/OSSRH-1906814:25
kenfinniganbobmcw: can I get access too?14:25
hbraunbobmcw long story short: what we plan to introduce initially is a Factory that produces parsers for the main (top level) xml elements in standalone.xml14:25
bobmcwkenfinnigan: sure.. what's your oss.sonatype.org username?14:25
kenfinniganbobmcw: might need to create one first ;-)14:25
bobmcwor issues.sonatype.org14:25
hbraunso we can parse intertaces, sockets, management, and subsystems indepently or wrap a custom root parser around that14:26
bobmcwhbraun: likewise... do you have, or want to setup an issues.sonatype.org account, so I can request maven push perms for ya?14:26
hbraunwhich is what I did in the prototye14:26
hbraunbobmcw i think I have one, hang on14:27
bobmcwhbraun: sounds good, wrt parsering14:27
kenfinniganbobmcw: kenfinni14:27
hbraunthe login hbraun, sdo you need the email as well?14:27
hbraunbobmcw haha, do you want me to stop talking?14:28
bobmcwhbraun: not at all, it's good to understand it :)14:28
bobmcwhopefully username is sufficient14:28
bobmcwhttps://issues.sonatype.org/browse/OSSRH-1906814:29
bobmcwhbraun: so this will require an upstream change in WF, and a release before we can benefit?14:29
kenfinniganbobmcw: thanks14:29
bobmcwor is this totally -swarm usage of existing WF stuff?14:29
hbrauncurrently we can parse all subsystems, but the core element are missing14:30
hbraunlike sockets14:30
hbrauni cannot get hold of these parser with out an upstream change14:30
hbraunhowever without sockets, some fractions don't work14:31
hbraunin particular the ones that add socket bindings as part of their default config, like tx14:31
hbraunat this point we ca either wait for upstream, or temporarily duplicate some code in our repo14:32
hbraunwhich is to be removed when upstream releases14:32
hbraunbut since WF is on a maintenance cycle, this can take some time14:32
*** bobmcw_ <bobmcw_!~bobmcw@va-67-233-86-120.dhcp.embarqhsd.net> has joined #wildfly-swarm14:33
hbrauni'll make sure the changes get on their way soon, but I would still prefer to duplicate some code to finish the first iteration and we have something to look at in swarm14:33
bobmcw_huh, transformer blew a few streets over14:33
bobmcw_loud *pop* and then it was dark14:34
bobmcw_I'm back!14:34
bobmcw_last I saw was 09:31 hbraun: however without sockets, some fractions don't work14:34
bobmcw_hbraun: cool.  one thing we used to do with torquebox was have a CI job build nightlies (non-SNAPSHOT) of wildfly14:34
hbraunhbraun:in particular the ones that add socket bindings as part of their default config, like tx14:34
hbraun[15:32:08] hbraun:at this point we ca either wait for upstream, or temporarily duplicate some code in our repo14:34
hbraun[15:32:24] hbraun:which is to be removed when upstream releases14:34
hbraun[15:32:45] hbraun:but since WF is on a maintenance cycle, this can take some time14:34
hbraun[15:33:05] bobmcw_ (~bobmcw@va-67-233-86-120.dhcp.embarqhsd.net) joined the channel14:34
hbraun[15:33:32] hbraun:i'll make sure the changes get on their way soon, but I would still prefer to duplicate some code to finish the first iteration and we have something to look at in swarm14:34
bobmcw_that we could then consume by torquebox (or now -swarm)14:34
bobmcw_we'd use the maven-versions-plugin, and build things like wildfly-11.incremental.2314:34
hbraunthat's probably a good idea anyway14:35
bobmcw_then we can rebase torquebox (-swarm) against non-released versions14:35
hbrauni think we'll have dep's on upstream WF every once in a while14:35
bobmcw_and not have to screw with -SNAPSHOTs or WF's release schedule14:35
bobmcw_right, we have in the past, around the SelfContainedServer buts14:35
hbraunbobmcw_ +1 i am all for it14:35
bobmcw_but that was during their .AlphaX cycle, so wasn't too long of a wait ever14:35
*** bobmcw <bobmcw!~bobmcw@redhat/jboss/bobmcw> has quit IRC (Ping timeout: 250 seconds)14:35
*** bobmcw_ is now known as bobmc14:35
*** bobmc is now known as bobmcw14:35
*** bobmcw <bobmcw!~bobmcw@redhat/jboss/bobmcw> has joined #wildfly-swarm14:35
*** ChanServ sets mode: +o bobmcw14:35
bobmcwhbraun: okie dokie... probably not until Monday, but I'll get that CI job going14:36
bobmcwas long as they merge the PR, we can then use it w/o a bonafide release14:36
hbraunno hurry, I still need to do the changes to upsteam anyway14:36
hbraunbut according to brian, the impact is minimal and I don't expect to much resistance to merge the PR14:37
bobmcwexcellent14:37
hbraunit's basically just working around their package protection14:37
bobmcwdmlloyd stopped whinging about public API guarantees?14:37
hbraunand pull some existing API up14:37
bobmcw(hi dmlloyd)14:38
hbraunwell, the stability expectations are an ongoing discussion14:38
hbrauni can see their point14:38
bobmcwat least in the near term, I'm okay with stability non-expectations14:38
hbraunfor WF it's a long term commitment14:38
hbraunyes, that's what I told them as well14:39
hbraunbut maybe we should put it on the F2F agenda14:39
bobmcwwe need a f2f agenda, if there's not already a doc14:40
kenfinniganhaven't seen a doc linked, so probably need one14:40
dmlloydmy problem is that we've over-promised our API contracts14:40
hbraunbobmcw i think this will evolve in WF anyway, as there is much demand for opening up the server internals and we havn't found a satsifying policy for that14:41
hbraundamn dmlloyd was listening all the time14:41
dmlloydbasically set it up so that we can't remain compatible and make substantial progress at the same time14:41
dmlloyd:)14:41
bobmcwdmlloyd: there was discussion, in the torquebox days, of a "public guarantee API" and an "internal Bob's team guarantee" of differing SLAs14:41
dmlloydideally we'd have a clean and minimal set of supported APIs, and then if people need more functionality, they'd request it and we'd add it (or not)14:41
hbraunyes, I agree that the compatibility guarantees are very limiting at the moment14:42
dmlloydyeah I expect those ideas are just as relevant now14:42
bobmcwand ideally I'd like to think of -swarm as really "internal to WildFly itself"14:42
hbraunbobmcw +114:42
bobmcwe.g., don't really consider us downstream, but rather side-car of the same project14:42
dmlloydideally wildfly would have provided all the -swarm capabilities to begin with :)14:43
bobmcwbut SOMEBODY never did it.14:43
* bobmcw isn't pointing fingers14:43
dmlloydheh well the reasons are social and boring14:43
dmlloydI should have just done it though, damn the consequences14:43
dmlloydwe'd be in better shape now if I had14:43
* bobmcw isn't sure if he's being insulted14:43
kenfinniganlol14:43
bobmcwI think he's saying "dmlloyd's swarm would be better that our swarm"14:44
dmlloydhaha14:44
dmlloydthink of it more that you'd spend less time working on the boring parts and more time working on the useful parts :)14:44
* kenfinnigan likes that idea14:45
dmlloydone way or another it'll still happen14:45
dmlloydit's just, I can't really get any resources so it's just my spare time project14:45
dmlloydlike jboss modules was, or MSC :)14:45
bobmcwat least we have the wildfly-config-api stuff now14:45
hbrauni think swarm and wf will co-evolve14:46
hbraunswarm is healthy for wf, because it introduces a fresh look14:46
dmlloydwell, reasonably fresh :)14:47
* bobmcw isn't sure if he's being insulted14:47
hbraunbut the influence is bi-directional14:47
hbraun(in an ideal world)14:47
* bobmcw needs more coffee14:47
hbraunwhenever it get's emotional bobmcw leaves14:48
dmlloydmy goals aren't exactly the same though - I want to see wildfly retain its current functionality but with a fraction of the code, and with the numerous duct-taped parts of the codebase cleaned up14:48
dmlloydgetting the nice swarmish API is just a side benefit to me :)14:48
hbraundmlloyd  which parts do you have in  mind? what goals do you pursue?14:49
hbraun(in an ideal world)14:49
*** rbenevides|afk is now known as rbenevides14:54
* hbraun is obviously asking the wrong questions14:55
dmlloydall the operation management stuff14:56
dmlloydit should be properly transactional14:56
dmlloydthe way we locate and instantiate extensions is completely backwards14:56
dmlloydthe way we start and stop the server is very bug-prone14:57
dmlloydservices don't meet the service contract14:57
dmlloydbasically the whole thing is taped together with hacks and wishes14:57
hbraundmlloyd i understand most of it, except for "services don't meet the service contract"15:00
dmlloydwe have a lot of MSC services in wildfly15:00
dmlloydfor many of them (maybe even most), if you stop the service, then start it again, everything enters a broken state15:01
kenfinnigandmlloyd: how would you fix that though? rewrite all the services to handle it better?15:01
dmlloydyes15:01
dmlloydpreferably using MSC 215:02
kenfinniganwhat does WF use now? MSC 1?15:02
dmlloydyes15:02
kenfinnigandoes MSC 2 exist? or its a plan of yours15:02
dmlloydMSC 1 is fine but it was never designed for transactions, so we have a variety of hacks in place to track services and get some idea of when the container is "stable"15:02
dmlloydMSC 2 exists15:03
kenfinniganwhat prevents it being used in WF now? or is it planned for WF 11?15:03
dmlloydit's not directly compatible with MSC 1 though, which brings complications :)15:03
kenfinniganah ok15:03
dmlloydit requires many things to be rewritten15:03
kenfinniganso basically moving to MSC 2 involves rewriting all services in WF today15:03
dmlloydyeah15:04
dmlloydgranted for many of those services that'll be simple15:04
dmlloydit's mostly just thunking code into a different shape15:04
kenfinnigani imagine that would be enough man hrs that mgmt is not keen on "downing tools" to make the change15:05
dmlloydyeah for this kind of change, it is impossible to convince anyone to invest15:05
dmlloydyou can only make the big change if you can show the cost is low and the benefit is high15:06
hbraununless there is significant value15:06
dmlloydbut, that's never stopped me before :)15:06
hbrauni.e. in terms of satisfying demand15:06
hbraunoutside demand, not an engineering point of view. it requires a justification15:07
kenfinniganis it something that Jason would back without mgmt ok?15:07
dmlloydyeah I hope it will be like jboss-modules, where after it was introduced everyone realized they really needed it all along :)15:07
dmlloydno Jason is not on board with this15:07
kenfinniganhe doesn't like MSC 2, or the work to rewrite?15:07
dmlloydhe's become skeptical of big changes in general15:07
*** hbraun <hbraun!~hbraun@pdpc/supporter/professional/hbraun> has quit IRC (Quit: afk)15:08
dmlloydI don't blame him though, he has a lot to worry about15:08
kenfinnigancan understand that view, but there is need for revolution on occasion to propel evolution15:08
dmlloydyeah15:08
dmlloydI've always been a bit of a revolutionary :)15:08
kenfinniganhah15:09
dmlloydJason was too when the status quo was so bad that it was the only rational position15:09
dmlloydbut now there's a good argument that the status quo is generally "good enough"15:09
dmlloydso the counter argument has to be that changing things results in such a better world that the (ideally minimal) cost is worth it15:10
dmlloydonly way to accomplish that right now though is, on my own time :)15:10
kenfinniganis it a situation where WF 20 needs to be built with MSC 2, even if it takes longer and in spare time, to then show the benefits?15:10
kenfinnigandmlloyd: you mentioned that MSC 2 isn't backward compatible with MSC 1, but is there any way you could "wrap" existing MSC 1 services so a conversion could be done piecemeal? or that would just be a can of worms?15:12
dmlloydit's probably a can of worms to an extent, but the worms can probably be contained within a somewhat larger can :)15:12
dmlloydin any event we'd have no choice, since (against my recommendation) we made MSC 1 a public API with backwards compatibility guarantees15:13
kenfinniganouch15:14
dmlloydJason is a bit schizophrenic on backwards compatibility - while he agrees that we can deprecate, and ultimately remove, things on a schedule, I think he'd like us to retain backwards compatibility on everything pretty much forever15:14
bobmcwI'M BACK!15:14
dmlloydso convincing him to deprecate and remove things can be tough :)15:15
bobmcwis the emotionalism over?15:15
* dmlloyd (((hugs))) bobmcw15:15
kenfinniganthat's a shame, prevents advancement to a degree15:15
dmlloydit does15:15
dmlloydit's understandable though15:15
kenfinnigantotally, trick is always finding the right balance between the two15:15
kenfinniganfeels like we may have gone from too much churn to not enough15:15
dmlloydat the end of the day though, my vision/dream for WildFly is pretty much Jason's nightmare :) so I have to find a balance that keeps everyone satisfied15:16
dmlloydyeah it's a bit reactionary in a way15:16
dmlloydthe EAP 6 lifecycle can only be described as "traumatic"15:16
kenfinniganwhy his nightmare? what have you got planned?15:16
dmlloydhaha, I just mean that I like changing things15:16
dmlloydadding things15:16
kenfinnigandon't we all15:16
dmlloydbut most importantly, making the adding of things be more sustainable than it is today15:16
dmlloydwildfly is still so monolithic - even the -core split is proving to be difficult from a maintenance perspective15:17
bobmcwwrt msc2, is there any easy way at least to introduce msc2, and wrap Service<T> with msc2.ServiceWrapper<T> for the short-term?15:17
dmlloydprobably not bobmcw15:17
bobmcwwhile not fixing anything, at least handling the simple cases with low effort?15:17
dmlloydWildFly itself depends on specific weird MSC behaviors15:17
bobmcwah15:18
dmlloydwe can do that to wrap end user services, but not for the container itself probably15:18
kenfinniganwhich I presume is something you'd also like to remove? MSC behavior bit15:18
dmlloydyeah, the MSC 2 contracts are simpler and cleaner15:18
dmlloydit is a huge improvement performance-wise15:18
dmlloydbut it's hard to imagine an "increment" of change smaller than "the core code and all existing subsystems rewritten"15:19
kenfinniganreally? what kind of perf improvement?15:19
dmlloydwhich makes it a difficult sell :)15:19
kenfinniganindeed15:19
bobmcwdmlloyd: a tub of coffee and a long weekend, I'm sure you could do it15:19
kenfinniganintravenous drip of coffee more like ;-)15:20
dmlloydbobmcw: part of my strategy is to generate all of the management code for all subsystems using a one-off inverse DMR-to-Java tool15:20
bobmcwso #cloudbees offered a recommendation on our slow builds; hopefully in the next 30 min, when I fire build #2, we'll see if it worked15:20
dmlloydno doubt finding all manner of horrific DMR errors in the process :)15:20
bobmcwdifferent than the config-api, which is java to ModelNode?15:21
bobmcw(and back, if I recall, but we've never done that)15:21
bobmcws/done/tested/15:21
bobmcws/tested/used/15:21
bobmcwhttp://wildfly-swarm.github.io/wildfly-config-api/0.3.18-SNAPSHOT/apidocs/index.html15:22
jbossbotTitle: config-api 0.3.18-SNAPSHOT API15:22
dmlloydyeah basically similar15:25
*** agoncal_ <agoncal_!~agoncal@2a01cb08005698002dF5151e1d925AB8.ipv6.abo.wanadoo.fr> has joined #wildfly-swarm15:32
projectodd-ciStarting build #3 for job wildfly-swarm-examples (previous build: SUCCESS)15:33
projectodd-ciProject wildfly-swarm-examples build #3: FAILURE in 2.3 sec: https://projectodd.ci.cloudbees.com/job/wildfly-swarm-examples/3/15:33
*** agoncal_ <agoncal_!~agoncal@2a01cb08005698002dF5151e1d925AB8.ipv6.abo.wanadoo.fr> has quit IRC (Ping timeout: 250 seconds)15:36
bobmcwdamnit, -examples15:38
projectodd-ciStarting build #350 for job wildfly-swarm (previous build: SUCCESS)15:38
bobmcwyay, it's downloading so much less!15:38
kenfinniganbobmcw: what was the fix?15:43
bobmcw"Invoke top-level maven target" and "use private repository" instead of "Invoke Maven3 target" in the build steps15:44
kenfinniganweird15:44
kenfinniganwould seem the same to me15:44
bobmcwapparently "Invoke Maven3 target" is some JFrog plugin target, but dunno why anyone would want to use it15:45
kenfinniganvery weird15:45
bobmcwso it's churning right along now15:45
bobmcwonce this build finishes, I'll try to figure out what's up with -examples15:45
projectodd-ciProject wildfly-swarm build #350: SUCCESS in 15 min: https://projectodd.ci.cloudbees.com/job/wildfly-swarm/350/15:53
projectodd-ciStarting build #4 for job wildfly-swarm-examples (previous build: FAILURE -- last SUCCESS #2 13 hr ago)15:53
projectodd-ciProject wildfly-swarm-examples build #4: STILL FAILING in 1.7 sec: https://projectodd.ci.cloudbees.com/job/wildfly-swarm-examples/4/15:53
bobmcw15minutes!15:54
bobmcwtcrawley: didja get your @Inject/lookup working?15:55
tcrawleynot yet, no15:56
tcrawleyI don't see any mention of an initial factory in sysprops. where would we be setting that?15:56
tcrawleysomewhere in the naming fraction?15:56
bobmcwtcrawley: WF should be doing that internally15:58
bobmcwwhen it sees the subsystem=naming stuffs15:58
bobmcwnothing we have to do from our end, assuming we're enabling the naming subsystem15:58
bobmcwtcrawley: kenfinnigan: you should have push privs to oss.sonatype for org.wildfly.swarm now15:58
tcrawleythanks!15:58
projectodd-ciStarting build #5 for job wildfly-swarm-examples (previous build: STILL FAILING -- last SUCCESS #2 13 hr ago)15:59
*** BenT15 <BenT15!~btomasini@c-73-19-28-3.hsd1.wa.comcast.net> has joined #wildfly-swarm16:06
jbossbotgit [wildfly-swarm-examples] push master 7bed774.. Bob McWhirter Add our snapshot repo to the list of pluginRepositories16:06
jbossbotgit [wildfly-swarm-examples] push master URL: http://github.com/wildfly-swarm/wildfly-swarm-examples/commit/7bed7745b16:06
bobmcw-= THIS MESSAGE NOT LOGGED =-16:07
*** BenT15 <BenT15!~btomasini@c-73-19-28-3.hsd1.wa.comcast.net> has quit IRC (Ping timeout: 260 seconds)16:10
*** jclingan <jclingan!~jclingan@redhat/jclingan> has quit IRC (Quit: jclingan)16:11
*** tcrawley is now known as tcrawley-away16:21
projectodd-ciProject wildfly-swarm-examples build #5: STILL FAILING in 46 min: https://projectodd.ci.cloudbees.com/job/wildfly-swarm-examples/5/16:45
projectodd-ciStarting build #6 for job wildfly-swarm-examples (previous build: STILL FAILING -- last SUCCESS #2 14 hr ago)16:48
projectodd-ciProject wildfly-swarm-examples build #6: ABORTED in 22 sec: https://projectodd.ci.cloudbees.com/job/wildfly-swarm-examples/6/16:48
projectodd-ciBob McWhirter: Add our snapshot repo to the list of pluginRepositories16:48
projectodd-ciStarting build #7 for job wildfly-swarm-examples (previous build: ABORTED -- last SUCCESS #2 14 hr ago)16:50
projectodd-ciProject wildfly-swarm-examples build #7: ABORTED in 26 sec: https://projectodd.ci.cloudbees.com/job/wildfly-swarm-examples/7/16:51
projectodd-ciStarting build #351 for job wildfly-swarm (previous build: SUCCESS)16:51
projectodd-ciProject wildfly-swarm build #351: ABORTED in 42 sec: https://projectodd.ci.cloudbees.com/job/wildfly-swarm/351/16:52
projectodd-ciStarting build #8 for job wildfly-swarm-examples (previous build: ABORTED -- last SUCCESS #2 14 hr ago)16:52
projectodd-ciStarting build #352 for job wildfly-swarm (previous build: ABORTED -- last SUCCESS #350 1 hr 16 min ago)16:54
bobmcwI give up16:55
bobmcwhave a great holiday everyone16:55
bobmcwespecially enebo16:55
enebobobmcw: eat stuff and be merry16:55
enebonot too merry and not too much stuff16:56
*** jclingan <jclingan!~jclingan@redhat/jclingan> has joined #wildfly-swarm17:11
*** BenT15 <BenT15!~btomasini@c-73-19-28-3.hsd1.wa.comcast.net> has joined #wildfly-swarm17:11
*** agoncal <agoncal!~agoncal@2a01cb08005698002DF5151E1d925Ab8.ipv6.abo.wanadoo.fr> has joined #wildfly-swarm17:17
*** BenT15 <BenT15!~btomasini@c-73-19-28-3.hsd1.wa.comcast.net> has quit IRC (Ping timeout: 246 seconds)17:25
*** agoncal <agoncal!~agoncal@2a01cb08005698002DF5151E1d925Ab8.ipv6.abo.wanadoo.fr> has quit IRC (Ping timeout: 250 seconds)17:30
*** lanceball is now known as lance|afk17:30
*** agoncal <agoncal!~agoncal@2a01cb08005698002DF5151E1d925Ab8.ipv6.abo.wanadoo.fr> has joined #wildfly-swarm17:30
projectodd-ciYippee, build fixed!17:40
projectodd-ciProject wildfly-swarm-examples build #8: FIXED in 47 min: https://projectodd.ci.cloudbees.com/job/wildfly-swarm-examples/8/17:40
projectodd-ciProject wildfly-swarm build #352: SUCCESS in 1 hr 12 min: https://projectodd.ci.cloudbees.com/job/wildfly-swarm/352/18:07
projectodd-ciStarting build #9 for job wildfly-swarm-examples (previous build: FIXED)18:09
kenfinniganwow, over an hr for build18:10
*** tcrawley-away is now known as tcrawley18:22
*** lance|afk is now known as lanceball18:34
*** jclingan <jclingan!~jclingan@redhat/jclingan> has quit IRC (Quit: jclingan)18:41
tcrawleybobmcw: ar you still around?18:41
tcrawleyare*18:41
tcrawleydo we have have any example/test code that actually creates an InitialContext?18:41
*** jclingan <jclingan!~jclingan@redhat/jclingan> has joined #wildfly-swarm18:56
projectodd-ciProject wildfly-swarm-examples build #9: SUCCESS in 49 min: https://projectodd.ci.cloudbees.com/job/wildfly-swarm-examples/9/18:58
*** hbraun <hbraun!~hbraun@pdpc/supporter/professional/hbraun> has joined #wildfly-swarm19:17
*** hbraun <hbraun!~hbraun@pdpc/supporter/professional/hbraun> has quit IRC (Quit: afk)19:33
*** agoncal <agoncal!~agoncal@2a01cb08005698002DF5151E1d925Ab8.ipv6.abo.wanadoo.fr> has quit IRC (Ping timeout: 250 seconds)20:01
*** kenfinnigan <kenfinnigan!~ken@redhat/jboss/kenfinnigan> has quit IRC (Quit: Leaving.)20:08
*** helio-frota <helio-frota!~hf@8.80.249.186.araujosat.com.br> has joined #wildfly-swarm20:23
*** agoncal <agoncal!~agoncal@2a01cb08005698002DF5151e1D925Ab8.ipv6.abo.wanadoo.fr> has joined #wildfly-swarm20:43
*** BenT15 <BenT15!~btomasini@c-73-19-28-3.hsd1.wa.comcast.net> has joined #wildfly-swarm20:44
*** tcrawley is now known as tcrawley-away20:46
*** tcrawley-away is now known as tcrawley20:47
*** BenT15 <BenT15!~btomasini@c-73-19-28-3.hsd1.wa.comcast.net> has quit IRC (Client Quit)20:49
*** jclingan <jclingan!~jclingan@redhat/jclingan> has quit IRC (Quit: jclingan)21:12
*** agoncal <agoncal!~agoncal@2a01cb08005698002DF5151e1D925Ab8.ipv6.abo.wanadoo.fr> has quit IRC (Ping timeout: 250 seconds)21:26
*** agoncal <agoncal!~agoncal@2a01cb08005698007cca8083e5363c8d.ipv6.abo.wanadoo.fr> has joined #wildfly-swarm21:44
jbossbotgit [wildfly-swarm-examples] push master 186fbe3.. Lance Ball Add ribbon-webapp and hack out the beginning of an example UI21:51
jbossbotgit [wildfly-swarm-examples] push master URL: http://github.com/wildfly-swarm/wildfly-swarm-examples/commit/186fbe3f921:51
*** lanceball is now known as lance|afk21:52
*** jclingan <jclingan!~jclingan@redhat/jclingan> has joined #wildfly-swarm21:53
*** enebo <enebo!~enebo@c-75-73-8-169.hsd1.mn.comcast.net> has quit IRC (Quit: enebo)22:03
*** agoncal <agoncal!~agoncal@2a01cb08005698007cca8083e5363c8d.ipv6.abo.wanadoo.fr> has quit IRC (Ping timeout: 240 seconds)22:16
*** tcrawley is now known as tcrawley-away22:23
*** enebo <enebo!~enebo@c-75-73-8-169.hsd1.mn.comcast.net> has joined #wildfly-swarm22:52
*** helio-frota <helio-frota!~hf@8.80.249.186.araujosat.com.br> has quit IRC (Quit: Leaving)23:11
*** bobmcw <bobmcw!~bobmcw@redhat/jboss/bobmcw> has quit IRC (Remote host closed the connection)23:29
*** jclingan <jclingan!~jclingan@redhat/jclingan> has quit IRC (Quit: jclingan)23:47

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