14:31:24 #startmeeting Karaf Happy Hour 14:31:24 Meeting started Fri Aug 29 14:31:24 2014 UTC. The chair is edwarnicke. Information about MeetBot at http://ci.openstack.org/meetbot.html. 14:31:24 Useful Commands: #action #agreed #help #info #idea #link #topic #startvote. 14:31:24 The meeting name has been set to 'karaf_happy_hour' 14:31:31 #topic Rollcall 14:31:33 #info edwarnicke 14:31:39 Please #info in :) 14:31:43 #info George Zhao 14:31:46 #chair gzhao phrobb 14:31:46 Current chairs: edwarnicke gzhao phrobb 14:31:54 #info lori for lispflowmapping 14:31:58 #info tbachman for Group Based Policy 14:32:55 Rollcall will go till x:35 :) 14:33:07 #info vjanandr for SNBI 14:33:14 #info abhijitkumbhare OpenFlow plugin 14:33:39 radhika for sdniapp 14:33:54 vjanandr: saw the question in your email, hope will have it resolved here. 14:33:57 Welcome radhika 14:34:23 #info radhika for sdniapp 14:35:15 gzhao: made some more changes.. but hard luck 14:35:36 vjanandr: Lets get you past it here :) 14:35:44 #topic Status 14:35:53 OK, please #info in where you are :) 14:35:56 #info gbp still at mvn clean install :( 14:36:10 #info need to resolve openflowplugin extensions 14:37:18 #info snbi still stuck at mvn clean install.. 14:37:31 #info lispflowmapping is committed to integration 14:37:39 build logs.. http://pastebin.com/NqsCMSz2 14:37:57 * tbachman is rebuilding to provide logs and files 14:37:59 Cool... lets move to 14:38:05 #topic Troubleshooting 14:38:15 vjanandr: Looking at your logs :) 14:39:00 vjanandr: OK.. need to get a bit more hands on, can you push a gerrit so I can pull it down and poke at it? 14:39:11 ok will do.. 14:39:58 what is status for lisp, sdni, snbi, sfc? 14:40:46 #info snbi still stuck at mvn clean install.. 14:41:23 gzhao: see above, I infoed in 14:41:55 lori: great, thanks 14:41:59 gzhao: lisp is in integration as of today 14:43:43 https://git.opendaylight.org/gerrit/10485 14:43:48 https://git.opendaylight.org/gerrit/10486 14:43:57 do we have representative from defense4all, opencontrail, snmp4sdn and ttp 14:45:41 Colin (TTP) is on a paternity leave with the newborn 14:46:04 vjanandr: Do I need 10485 or 10486 ? 14:46:08 edwarnicke: current features file: http://pastebin.com/LcNQKVNj 14:46:21 abhijitkumbhare: I texted with colindixon yesterday and he's stil going to try to get it together 14:46:31 edwarnicke: I didn’t work on this much last night 14:46:40 I think the issue is the right reference to the extensions 14:46:52 tbachman: No need to apologize 14:47:01 tbachman: The pink elephant eventually comes for us all :) 14:47:05 lol 14:47:09 * regXboi wanders in 14:47:17 regXboi: interesting timing 14:47:18 edwarnicke: abhijitkumbhare Colin said he is going to work on it this weekend 14:47:24 tbachman: What do you get when you build your features file? 14:47:25 OK 14:47:25 * tbachman wonders if regXboi answers to pink elephant 14:47:41 tbachman: Eventually, we all do ;) 14:47:53 gzhao, edwanicke: just got off phone with colindixon and yes, he will be working on it this weekend 14:48:17 * tbachman hopes regXboi scrolls back to understand my feeble attempt at a 14:48:19 regXboi: thanks 14:48:21 * edwarnicke notices we are an insanely async community 14:48:30 edwarnicke: Tests in error: 14:48:30 The bundle "org.opendaylight.groupbasedpolicy_0.1.0.SNAPSHOT [196]" could not be resolved. Reason: Missing Constraint: Import-Package: org.opendaylight.yang.gen.v1.urn.opendaylight.openflowjava.nx.match.rev140421; version="[0.0.0,1.0.0)" 14:48:35 nx.match 14:48:55 wanted to go into the OF plugin to track down which bundle exports this 14:48:56 tbachman: Next time, use the tag :) (or if you are ambitious, nest them ;) 14:48:56 #info TTP will start Karaf this weekend (Colin) 14:49:00 and add it to my pom 14:49:04 lol 14:49:04 tbachman: no I don't answer to "pink elephant" - I do answer to "hey, you", "help", and various other nom-de-plumes 14:49:12 regXboi: lol 14:49:17 tbachman: Are you trying to pull things in bundle-wise? 14:49:34 edwarnicke: I’m trying to make mvn clean install succeed :) 14:49:49 do I want/need to pull them in bundle-wise? 14:49:57 vjanandr: Do I need to look at 10485 of 10486 ? 14:50:23 10485 14:50:30 it contains features changes.. 14:50:42 tbachman: LOL.. so I think we discussed yesterday, that you have a decision as to whether to be more or less granular, but your best starter bet was less granular 14:50:48 vjanandr: Pulling now :) 14:51:09 edwarnicke: yeah 14:51:24 maybe I misunderstood 14:51:31 we can’t go less granular than a bundle, right? 14:51:37 tbachman: LOL 14:51:39 so…. you’re saying I need to think “bigger” :) 14:51:44 10486 contains the parent pom.xml changes.. which is minimal.. 14:52:08 tbachman: So lets put aside for a moment questions of the right thing and start with getting something working (after which we can optionally call the solution nasty names and replace it :) ) 14:54:22 tbachman: I am saying lets put aside questions of right and wrong and start with working... one moment... gathering a suggestion 14:54:47 mvn:org.opendaylight.ovsdb/features-openflow-nxm/0.0.3-SNAPSHOT/xml/features 14:54:56 odl-openflow-nxm-extensions 14:55:18 so, the first one is in there 14:55:24 https://www.irccloud.com/pastebin/c6o1dAHZ 14:55:32 And that last one was the pom file fragment 14:56:40 * tbachman builds with single new line in features file 14:57:07 I already had the other two changes (presumably from yesterday) 14:57:10 (thx, btw) 14:57:21 vjanandr: When I try to build snbi I get: 14:57:23 https://www.irccloud.com/pastebin/11eH3U1Z 14:57:36 edwarnicke: super-genius :) 14:57:39 that worked 14:57:41 vjanandr: I'm just doing 'mvn clean install' from the top level directory 14:57:44 * tbachman goes back to consult the wiki 14:57:56 tbachman: Take a note to come back and call the solution funny names with me a in bit ;) 14:58:05 lol 14:58:54 edwarnicke: I tried mvn clean install from features dir.. 14:59:21 let me try from the parent dir.. 14:59:28 vjanandr: Cool :) 15:00:57 * tbachman blew away .m2 and is working through failure 15:02:43 tbachman: Death to .m2 ! 15:02:48 :) 15:02:51 * edwarnicke is in a apparently in a marshal mood this morning 15:02:58 lol 15:02:59 * edwarnicke listening to 'Search and Destroy' 15:03:06 ROFL 15:03:14 "Look out honey cuz I'm using technology!' 15:03:21 very up-tempo :) 15:03:32 * tbachman pictures edwarnicke’s fingers flying across the keyboard 15:04:40 tbachman: Magical coffee of dooooooom! https://twitter.com/edwarnicke/status/484920608266797056/photo/1 15:04:50 Madhu: Welcome to the party! :) 15:04:56 vjanandr: How is it going? 15:05:16 still building.. 15:05:19 edwarnicke: thanks :) 15:05:51 Madhu: Just to catch you up... 15:06:22 Madhu: tbachman has tried just adding the odl-openflow-nxm-extensions so we can see if we can get him working, and then we can debate issues of granularity 15:06:41 Madhu: vjanandr has pushed https://git.opendaylight.org/gerrit/#/c/10485/ and https://git.opendaylight.org/gerrit/#/c/10486/ 15:06:45 edwarnicke: thanks sir. 15:07:06 tbachman: please holler if you need any help buddy 15:07:11 Madhu: vjanandr is getting what look to me like very strange errors: http://pastebin.com/NqsCMSz2 , thus the request that he push some code 15:07:22 tbachman: as you see, odl-openflow-nxm-extensions work fine in the ovsdb project 15:07:31 Madhu: what vjanandr pushed didn't build from the top of the directory for me, so he's trying it out and will report back 15:07:43 Madhu: I have every faith it will work for tbachman as well :) 15:07:58 build success with clean .m2 15:08:07 tbachman: awesome brother 15:08:11 :) 15:08:14 * edwarnicke hops out of his chair and does a happy dance! :) 15:08:15 * tbachman goes back to wiki 15:08:39 * Madhu hopes to spend a nice long weekend with the family 15:08:48 Madhu: I wish that for you too :) 15:09:53 edwarnicke: u must take a break too. u deserve it 15:10:04 and don't forget... we have 1 more month of sleepless nights 15:10:05 :) 15:10:11 Madhu: You are both wise and correct 15:10:13 * tbachman hopes all get some R&R 15:10:19 Madhu: The hardest part for me is turning the brain off 15:10:31 edwarnicke: hibernate it. 15:10:33 * tbachman does not appear to share such problems 15:10:45 Madhu: It was 4 days after I took started my vacation after Hydrogen before I could deal with no longer working on it 15:10:53 Madhu: :) 15:11:03 Madhu: I have better biohacks for that now :) 15:14:02 Madhu: Do you know if https://git.opendaylight.org/gerrit/#/c/10472/ is ready for review? It looks like it is to me... but was curious your thoughts 15:14:51 #info paulq sfc 15:14:55 (very late to the party) 15:15:12 paulq_: Welcome... pull up a chair, grab a drink :) 15:15:13 edwarnicke: oh ya... am supposed to review that. thanks edwarnicke for the reminder 15:15:25 paulq_: welcome to the party 15:15:29 edwarnicke: questino for you 15:15:37 Madhu: I wasn't sure if you needed a reminder, or I needed to be informed it wasn't quite ready yet ;) 15:15:38 paulq_: How is SFC now 15:15:44 tbachman: 42 15:15:46 lol 15:15:58 are we supposed to run the distribution archetype in features as well? 15:15:59 sfc is so so to be honest 15:16:08 tbachman: I would recommend against it 15:16:09 paulq_: don't worry sir 15:16:14 we have some time till the M5 15:16:15 :) 15:16:17 ed tried to teach me, but I'm a slow learning 15:16:19 tbachman: the distribution archetype will create a new project for distributions 15:16:24 right now trying to resovle: 15:16:29 tbachman: To my aesthetics, it doesn't belong under features 15:16:33 k 15:16:34 Tests in error: 15:16:34 The bundle "org.opendaylight.sfc.provider_0.0.1.SNAPSHOT [180]" could not be resolved. Reason: Missing Constraint: Import-Package: com.sun.jersey.api.client; version="[1.17.0,2.0.0)" 15:16:34 The bundle "org.opendaylight.sfc.provider_0.0.1.SNAPSHOT [180]" could not be resolved. Reason: Missing Constraint: Import-Package: com.sun.jersey.api.client; version="[1.17.0,2.0.0)" 15:16:40 tbachman: But if there is unclarity there, could you update the wiki to make it clearler 15:16:45 sure 15:16:47 paulq_: Could you pastebin your features file? 15:16:47 tbachman: fwiw, i don't use distribution yet :) 15:16:51 ah 15:16:52 it is not mandatory 15:16:57 k 15:17:05 Madhu: You are correct, not mandatory... but often handy for local testing ;) 15:17:26 features.xml: http://pastebin.com/255sx4JF 15:17:29 paulq_: odl-base-jersey might be missing as feature dependency 15:17:30 Madhu: I know a lot of folks really *like* having a local distro that gets built when they build their stuff... so wanted it to be easy :) 15:17:41 Madhu: Here's what's weird there 15:17:49 edwarnicke: certainly awesome to have one 15:17:51 Madhu: I think we have jersey-client as a dependency already 15:18:11 Madhu: I tried to keep 'Necessary' and 'Optional' clearly distinguishable 15:18:18 edwarnicke: ah i know :) 15:18:25 Madhu: Because as you know... I have all kinds of strong aesthetic opinions :) 15:18:35 edwarnicke: paulq_ this is a tricky thing i struggled with too 15:18:43 so ... please bear with me when i try to explain 15:18:53 of course :) 15:18:58 Madhu: And folks don't have to follow my aesthetics, they are free to be wrong ;) 15:19:30 paulq_: sorry..my issue was with catalina ... the tricky issue 15:19:31 vjanandr: You've been awfully quiet... where are you at? 15:19:32 not jersey. 15:19:42 paulq_: but in any case, it will be good to know :) 15:19:49 Madhu: I think this is way simpler than your catalina issue 15:19:52 so... the odl-parent project has a dependencymanagement section 15:19:59 that exports a version x 15:20:06 edwarnicke: I had cleaned my .m2 folder.. 15:20:06 * edwarnicke he says not knowing the solution yet ;) 15:20:17 vjanandr: Ah... that could make things slow :( 15:20:23 and its taking awefully lot of time .. to build again 15:20:25 yeah 15:20:39 but one of the features in controller has hard dependency on version y for the same artifact. 15:20:47 vjanandr: Did you delete .m2/ or ./m2/repository/org/opendaylight/ ? 15:20:53 I wanted to check if the error goes away with a clean build 15:20:59 now if there is a version conflict between the feature expecting a version and the version defined in a project (and its parent) 15:21:04 :( .m2 15:21:12 maven decides to pick only the version defined in the project / parent 15:21:19 Ah... often ./m2/repository/org/opendaylight/ is just as good 15:21:36 paulq_: can you kindly check if you have any defined version for jersey.client in your project that is not 1.17 ? 15:21:47 looking... 15:21:56 edwarnicke: ^^^ this could be useful for many scenarios :) 15:22:13 my catalina issue was just one of it exposed the version conflict and maven selecting the one it decides to download 15:22:13 edwarnicke: the error I reported was seen with a build from features dir.. 15:22:17 and causing runtime issue 15:22:28 vjanandr: OK.. let me try that 15:23:02 btw paulq_ i would strongly encourage you to use the features-base-jersey 15:23:05 features-base-jackson 15:23:13 because all the dependencies are resolved for you already 15:23:34 sfc-provider pom: 15:23:36 15:23:37 com.sun.jersey 15:23:37 jersey-client 15:23:39 1.17 15:23:40 15:23:50 paulq_: by defining these bundles directly in your feature file... you are going to resolve all of the dependencies by hand 15:24:35 paulq_: u have a draft of the changes you are trying ? 15:24:49 i can take a look at all the related changes (including the features/pom.xml) 15:24:57 should I include base or just pick and choose from within it? 15:25:24 https://git.opendaylight.org/gerrit/#/c/10469/ 15:26:06 paulq_: base has individual features like 15:26:13 base-jackson, base-jersey, etc... 15:26:28 you can pick only those features that you really want 15:26:31 not base-all definitely 15:27:00 paulq_: at least that will help quickly to narrow down the dependency isssue 15:27:13 Madhu: Why is jettison in base-jackson? Was curious about that? 15:27:16 and if you want to define the dependencies by hand, we could do that as well 15:27:39 edwarnicke: thats because if i remember right, :) there was a dependency to it. 15:27:45 edwarnicke: mlemay_away might know that better 15:28:05 Hmm... 15:28:11 edwarnicke: or maybe its up for cleaning 15:28:15 edwarnicke: lemme check 15:28:20 Cool :) 15:29:03 hideyuki: Welcome to the party! :) 15:29:07 so — just curious. after the initial mvn clean install.... 15:29:16 can I *do* anything with that feature yet? 15:29:29 Or do I need to complete the integration step first 15:30:19 tbachman: if you have a local distro as edwarnicke said 15:30:24 k 15:30:25 you can use that to test your feature end to end 15:30:29 vjanandr: This is the error I'm getting building in features/ on 10485 with 10486 cherry-picked in: 15:30:30 [ERROR] Failed to execute goal org.apache.maven.plugins:maven-jar-plugin:2.3.2:jar (default-jar) on project features-snbi: Error assembling JAR: Manifest file: /Users/hagbard/try/snbi/features/target/classes/META-INF/MANIFEST.MF does not exist. -> [Help 1] 15:30:36 tbachman: if not... you have to do the integration changes 15:30:41 k 15:30:44 Madhu: One of the other nice things with the local distro 15:30:48 will try the distro… seems worthwhile :) 15:30:58 tbachman: just a note :) 15:31:06 Madhu: Is that FeatureService is sometimes less helpful with errors than one might hope... and you can poke a bit more interactively there 15:31:08 edwarnicke: yes.. ran into the same 15:31:18 things that work in your local distro doesn't guarantee the successful integration distro build :) 15:31:36 vjanandr: But the feature tests seem to pass... 15:31:37 tbachman: thats because of the hot vs warm vs cold .m2 repository that edwarnicke nicely explains 15:31:47 vjanandr: So it may be that your issue is not your features.xml but something else 15:31:51 but things that don’t work in your local distro are likely indicators that they won’t work in integration ;) 15:31:52 edwarnicke: agreed. 15:31:52 lol 15:32:05 edwarnicke: i was just lazy to include the distro, and I feel i should :) 15:32:23 tbachman: software is funny. never say never :) 15:32:25 edwarnicke: yes it passed for me as well.. 15:32:27 lol 15:32:27 Guys... is everyone in a place I can go to the wiki for 5 minutes and provide a semi-coherent explanation of hot,warm, cold .m2 15:32:45 * edwarnicke suppresses the urge to use the story of the goldilocks and the three bears as a metaphor 15:33:01 edwarnicke: martincasodo beat you to it 15:33:19 s/casodo/casado/ 15:33:32 radhika : how is sndi going? 15:33:34 * edwarnicke but did he do it for .m2/ caches? 15:33:38 lol 15:34:16 vjanandr: OK... so here's what I know... the mdsal feature is also not generating a MANIFEST.MF file... and its happy... so we have some pom.xml weirdness to track down 15:34:44 parent pom ? 15:35:32 tbachman: edwarnicke i heard the same goldilocks story yesterday from a US senator :) 15:35:39 lol 15:35:42 it’s the in thing 15:35:56 tbachman: edwarnicke this is regarding the Ferguson story 15:36:16 * tbachman tries to fathom just how goldilocks could possibly apply to Ferguson, MO 15:36:21 i was seriously thinking.. wow... i just read the goldilocks story to my kid and everyone is talking about it... 15:36:36 tbachman: hey.. as u said. it is the "in" thing now... 15:36:50 * Madhu pictures pixar goldilocks 15:36:50 Madhu: what!!! You weren’t reading him Vol 1 of Knuth? 15:36:55 lol 15:37:05 tbachman: we just finished that for my elder 15:37:10 lol 15:37:12 am talking about my little one ;) 15:37:17 he’s probably already fixing my bugs 15:37:32 tbachman: well... he is annoyed with my code. that is sure 15:37:33 the come out knowing scala 15:37:36 lol 15:37:40 Madhu: ROFL 15:37:41 Madhu: tbachman vjanandr Let me know if this makes the warm vs cold .m2 thing easier to understand: https://wiki.opendaylight.org/view/Karaf:Step_by_Step_Guide#A_bit_more_about_warm_vs_cold_.m2_and_why_it_matters 15:37:44 * tbachman understand 15:37:47 understands 15:38:04 vjanandr: Could be a parent pom thing yes 15:38:17 vjanandr: can you build from top level dir? 15:38:47 tbachman: All children should know Knuth's maxim: "Premature optimization is the root of all evil" 15:39:12 * edwarnicke was taught at six years old that proper keys should be unique, immutable, and dataless 15:39:49 edwarnicke: yes building.. it.. but awfully slow... 15:40:13 edwarnicke: tbachman i was in a PTA yesterday. 15:40:29 and they were talking about introducing keyboard typing for kids in 1st grade 15:40:49 * Madhu shocked. ashamed of human race 15:41:05 * tbachman is on phone 15:42:58 Madhu: Are they going to drop cursive writing? 15:43:34 edwarnicke: possible.... but again that is a US senate discussion :( 15:43:39 * Madhu ashamed again 15:43:52 Madhu: WTF... the US Senate? 15:44:08 edwarnicke: ya.. there was a discussion that happened in summer on dropping cursive writing 15:44:45 edwarnicke: http://www.postandcourier.com/article/20140515/PC1610/140519536 15:44:47 Madhu: I am not personally a fan of cursive writing 15:45:20 vjanandr: OK.. I know the root of your issue... trying to figure out the fix 15:45:26 vjanandr: In your parent pom you have: 15:45:41 https://www.irccloud.com/pastebin/1Lb63TVK 15:45:52 Which is inherted by your features/ 15:46:00 Which is then unhappy when it doesn't have a MANIFEST.MF 15:46:06 My naive attempt to clear the setting failed 15:46:14 So going to ask Dr Google for advice 15:55:13 * regXboi wonders if karaf happy hour is becoming "late night with karaf" 15:56:17 regXboi: its more like "late night with karaf" ends with "Karaf happy hour" :) 15:56:29 * edwarnicke ponders a top 10 list 15:56:41 * tbachman embodies a stupid human trick 15:57:11 vjanandr: This diff fixes your features/ build... but might break other stuff: 15:57:18 https://www.irccloud.com/pastebin/CU6WNy2W 15:57:24 vjanandr: Diff is against the parent pom 15:57:40 vjanandr: still looking for a way to get the features/pom.xml to ignore that stuff from the parent 15:57:44 Madhu: Any ideas on that? 15:57:51 edwarnicke:let me try this.. 15:58:22 vjanandr: Please note, I haven't looked to see if it breaks other stuff you are doing 15:58:25 edwarnicke: overriding the parent/pom ? 15:58:26 It might 15:58:36 Madhu: specifically this case: 15:58:36 edwarnicke: i do that today by just overriding the dependency 15:59:11 Madhu: In pluginManagement the parent pom has: 15:59:14 https://www.irccloud.com/pastebin/4WYPwsaB 15:59:28 edwarnicke: I understand that... will check if it helps the build.. 15:59:44 Madhu: In the features/pom.xml I tried the following in : 15:59:46 https://www.irccloud.com/pastebin/yf97VWFg 16:00:02 Which did not, as I would hope, cause the features/pom.xml to ignore the archive directive from the parent 16:00:09 * edwarnicke is probably doing something dumb :( 16:00:39 edwarnicke: hmm u r right. i would have hoped for the same. 16:00:53 edwarnicke: intact.. i do this to change a dependency version and it works great 16:01:06 * Madhu wonders if the version is the same, it merges other instructions ? 16:01:51 Madhu: I do this lots of places for various things and it works 16:02:01 Maybe if I try an empty configuration... 16:02:56 Madhu: Just tried: 16:02:57 https://www.irccloud.com/pastebin/tnhBLFaZ 16:03:01 It doesn't work either 16:04:57 gzhao: Did we loose the SDNi folks? 16:05:30 edwarnicke: I pinged radhika, no response yet 16:05:43 edwarnicke: in the distribution pom.xml 16:05:45 the TODOs 16:05:49 gzhao: Who do we not have engagement from so far? 16:05:52 tbachman: Yes? 16:06:00 it says: “ Necessary TODO put your features here." 16:06:02 edwarnicke: features build went thru.. with an empty configuration in parent pom.. 16:06:08 these are the features we’re declaring, or depending on? 16:06:24 vjanandr: But does your overall build still work? My suggestion has some potential to break it... 16:06:34 yes its still on :( 16:06:34 i.e. am I putting “group based policy”, or stuff like “md-sal”? 16:06:35 tbachman: this is in the distro pom.xml? 16:06:39 yeah 16:06:51 distribution/karaf/pom.xm 16:06:53 xml 16:06:58 edwarnicke: totally not engaged are: defense4all, snmp4sdn 16:06:58 tbachman: So there are two things that inform this 16:07:13 tbachman: It should be the case that your features file pulls through anything it needs 16:07:24 tbachman: So you definitely want your features file listed here as a dependency 16:07:34 edwarnicke: I don't get anything from oc, but you mentioned that they were here in happy hour 16:07:41 tbachman: And its probably true you want to list your top level feature in the 16:07:57 gzhao: They were two days ago, but have not heard from them since 16:08:30 * tbachman realizes he could probably also go look into a project like l2switch and grok this from their pom.xml 16:08:45 tbachman: So... simple first answer: start with your feature 16:08:49 tbachman: More complicated answer 16:09:07 tbachman: If you want to include odl-restconf in your and thus have odl-restconf start up on boot 16:09:14 tbachman: You will need to also put in the one for md-sal 16:09:20 Because you can only put in boot features 16:09:21 edwarnicke: got it 16:09:23 thx 16:09:25 edwarnicke: ok, lisp is in integration, and you are helping gbp, snbi and sfc right now 16:09:27 Things with an explicite runtime dependency 16:09:31 Once you are running 16:09:39 Karaf will pull in all the referenced repos down the tree 16:09:44 And all their features become available 16:09:54 Does that make sense? 16:09:57 I think so 16:09:59 btw 16:10:05 gzhao: Madhu and I are :) 16:10:06 back to the original scoping issue 16:10:24 is there are smaller granularity I could/should be using for the openflowplugin extensions? 16:10:42 edwarnicke: of cause, Madhu as well 16:10:49 tbachman: Is all of your stuff working? 16:10:55 not yet — trying distro 16:10:59 if that works, will go to integration 16:11:05 but - you got me past the main hurdle 16:11:06 thx 16:11:19 tbachman: So on the granularity thing 16:11:29 The question is 'which extensions do I need' 16:11:33 right — granularity, not scope (sorry) 16:11:41 If you look at the features.xml file Madhu did for extensions 16:11:52 You should be able to pick out and put in just the ones you need 16:11:58 k 16:12:05 I’ll have to see how those are offered 16:12:05 Rather than pulling through the whole kitchen sink in a go by referencing the feature 16:12:21 tbachman: There is however a point of decision there for you as to what is best for GBP 16:12:41 edwarnicke: I also view this as a learning process 16:12:50 tbachman: I'm simply recommending *thought* on this issue... you have to figure out based on that thought what the best decision is for you guys 16:12:59 edwarnicke: got it 16:13:01 tbachman: Its a learning process for all of us 16:13:24 tbachman: Main point is... I don't know enough about your situation to even make a concrete recommendation beyond 'There are these over inclusion issues you probably want to think about' 16:13:32 k 16:13:39 * edwarnicke tries to not make grand pronouncements out of pure ignorance... usually fails :) 16:13:45 lol 16:14:50 Madhu: Did you get a chance to look at: https://git.opendaylight.org/gerrit/#/c/10472/ 16:15:03 Madhu: Its kind of blocking dlux... would like to figure out next step on it 16:19:21 edwarnicke: thanks for looking into this.. I will get my parent pom building as well.. and hopefully try to get this done by tonight :) 16:21:14 vjanandr: OK... let me know if the fix I suggested broke stuff and we have to try something else 16:21:38 edwarnicke: sure.. will do.. will keep you posted.. 16:21:52 edwarnicke: sorry on a call. brb 16:23:34 Madhu: Cool :) 16:25:15 edwarnicke: pushed. but still brb :) 16:27:12 Madhu: Many thanks :) 16:27:45 vjanandr: Thanks for staying up so late... I know its late there 16:28:28 :)... and its festive day as well :) 16:28:51 but thanks to you.. for helping me out :) been struggling with this from morning.. 16:30:00 edwarnicke: am able to build the distro and see my feature installed 16:30:07 will try some functional testing 16:30:11 * edwarnicke does a happy dance 16:30:18 tbachman: Yes, next question... does it work :) 16:30:31 lol 16:31:02 * tbachman apologies for long RTT between tasks… too much context switching 16:32:07 * tbachman sees things working 16:32:46 so… with these changes, I should be able to push this commit, then move on to integration? 16:33:02 (s/merge/create in/) 16:35:59 * edwarnicke is the last one who can complain about context switching 16:36:08 tbachman: Yep... 16:36:26 tbachman: Let me know if those instructions work for you on the integration stuff 16:36:34 will do 16:37:56 Madhu and edwarnicke: know just how busy you all are, so definitely really appreciate your help here 16:38:13 tbachman: This is critical path to getting the release out :) 16:38:16 tbachman: Glad to help :) 16:40:55 tbachman: so GBP passed mvn clean install? -:) 16:41:15 gzhao: yes, and have created a local distro version for testing, which appears to pass 16:41:31 tbachman: terrific 16:41:32 gzhao: am looking to push this commit, and will transition to the integration part 16:41:43 gzhao: thx (well, thx to edwarnicke and Madhu :) ) 16:42:30 tbachman: we all want to thanks Ed and Madhu 16:46:26 I need to go to different meeting 17:09:05 After I execute "log:tail" command on Karaf console, a problems occur. The problem is that I have to a push a key on my keyboard twice to input a character. 17:11:16 Does this problem occurr on my environment? Do you understand what i mean? It may seem strange... 17:11:34 hideyuki_: I've not tried log:tail yet 17:11:50 hideyuki_: i use log:tail it works fine 17:12:14 hideyuki_: log:tail will lock the console and you have to press Ctrl+c to exit 17:12:22 and then you can type your commands 17:12:32 if u want to just check the log 17:12:32 edwarnicke: Madhu Thank you... every time I execute "log:tail", it happens. 17:12:36 u can do log:display 17:12:44 After Ctrl+C, it occurs. 17:13:05 hideyuki_: :) okay. it is not serious issue i hope 17:13:06 So to shutdown the Karaf, I have to type like this. sshhuuttddoowwnn 17:13:16 hideyuki_: ROFL 17:13:25 isn’t it ctrl-d to exit? 17:13:27 hideyuki_: i do Ctrl+d to exit 17:13:32 yes. 17:13:42 hideyuki_: thats strange for sure. and i don't face this issue 17:13:50 hideyuki_: mostly your terminal settings. 17:13:58 hideyuki_: what do u use ? 17:14:16 I'm using teraterm. do you know teraterm? 17:15:00 Madhu: I'll check my terminal settings. 17:15:25 hideyuki_: no :( 17:15:39 i use the default terminal comes in mac & have no issues using karaf console 17:16:02 edwarnicke: fyi. i have a runtime issue with karaf container 17:16:14 edwarnicke: still debugging... but happens only on Karaf and not on regular distribution 17:16:32 edwarnicke: something to do with runtime service hook-ups 17:16:46 Madhu: Let me know when you know more 17:16:47 edwarnicke: these are the issues i was really worried about 17:16:53 edwarnicke: absolutely 17:17:10 edwarnicke: seems like a timing.. works sometimes (i think) 17:17:20 anyways... lets c 17:26:19 ok, got most of it working (thanks Madhu!) I cleaned out my .m2 controller dir (as per the instructions) 17:26:23 and how how the following: 17:27:35 http://pastebin.com/KFn22vA3 17:30:13 paulq_: seems like a yang tools dependency 17:30:21 paulq_: lemme take a closer look 17:31:05 i think you are right 17:31:07 looking here 17:31:12 and applying pathc from ed yesterday 17:31:37 paulq_: https://gist.github.com/ee422d29a82d5936ffc0 17:31:40 can u try this ? 17:34:16 paulq_: hmm... u have included the yang tools features 17:34:26 have to check how the yang tools feature is defined 17:57:43 #endmeeting