Wednesday, 2015-05-06

« Tuesday, 2015-05-05 Index Thursday, 2015-05-07 »
*** Aish has quit IRC00:13
*** mlavalle has quit IRC00:30
*** mwang2 has quit IRC00:55
*** madhu_ak has quit IRC00:59
*** bharath_ has quit IRC00:59
*** SumitNaiksatam has joined #openstack-lbaas01:04
*** Tiancheng has joined #openstack-lbaas01:22
*** haigang has joined #openstack-lbaas01:53
*** fnaval has quit IRC01:54
*** ajmiller has joined #openstack-lbaas02:01
*** haigang has quit IRC02:05
*** fnaval has joined #openstack-lbaas02:07
*** Tiancheng has quit IRC02:09
*** mestery has joined #openstack-lbaas02:23
openstackgerritMerged openstack/neutron-lbaas: Updated from global requirements  https://review.openstack.org/17990802:24
*** Tiancheng has joined #openstack-lbaas02:41
*** kiran-r has joined #openstack-lbaas02:57
openstackgerritDoug Wiegley proposed openstack/neutron-lbaas: WIP - Octavia driver  https://review.openstack.org/17411403:04
openstackgerritDoug Wiegley proposed openstack/neutron-lbaas: WIP - Octavia driver  https://review.openstack.org/17411403:24
*** kiran-r has quit IRC03:34
*** bharath has joined #openstack-lbaas04:00
*** bharath has quit IRC04:05
*** blogan_ has joined #openstack-lbaas04:54
*** madhu_ak has joined #openstack-lbaas04:57
*** Tiancheng has quit IRC05:01
*** kiran-r has joined #openstack-lbaas05:09
*** madhu_ak has quit IRC05:11
*** crc32 has quit IRC05:17
*** SumitNaiksatam has quit IRC05:19
*** SumitNaiksatam has joined #openstack-lbaas05:20
*** blogan_ has quit IRC05:32
*** fnaval has quit IRC05:36
*** SumitNaiksatam has quit IRC05:57
*** SumitNaiksatam has joined #openstack-lbaas06:01
*** evgenyf has joined #openstack-lbaas06:09
openstackgerritBrandon Logan proposed stackforge/octavia: Fix queue producer handler to send ids  https://review.openstack.org/18041706:52
*** kobis has joined #openstack-lbaas07:20
*** woodster_ has quit IRC07:20
*** nmagnezi has joined #openstack-lbaas07:35
*** apuimedo has joined #openstack-lbaas07:45
openstackgerritAishwarya Thangappa proposed openstack/neutron-lbaas: Tempest tests for Members using testscenarios.  https://review.openstack.org/18043608:11
*** chlong has quit IRC08:22
*** apuimedo has quit IRC08:35
*** apuimedo has joined #openstack-lbaas08:47
openstackgerritAishwarya Thangappa proposed openstack/neutron-lbaas: Tempest tests for Members using testscenarios.  https://review.openstack.org/18043608:47
openstackgerritAishwarya Thangappa proposed openstack/neutron-lbaas: Tempest tests for Listener using testscenarios.  https://review.openstack.org/17981808:50
openstackgerritAishwarya Thangappa proposed openstack/neutron-lbaas: Tempest tests for Listener using testscenarios.  https://review.openstack.org/17981808:52
openstackgerritAishwarya Thangappa proposed openstack/neutron-lbaas: Tempest tests for Listener using testscenarios.  https://review.openstack.org/17981808:57
*** rdekel has joined #openstack-lbaas09:00
*** apuimedo has quit IRC09:13
*** rdekel has quit IRC09:22
*** apuimedo has joined #openstack-lbaas09:30
*** kiran-r has quit IRC09:47
*** evgenyf has quit IRC11:09
*** woodster_ has joined #openstack-lbaas11:37
*** evgenyf has joined #openstack-lbaas11:46
*** kiran-r has joined #openstack-lbaas12:16
*** ajmiller has quit IRC12:38
*** openstackgerrit has quit IRC12:53
*** openstackgerrit has joined #openstack-lbaas12:53
*** openstackgerrit has quit IRC13:38
*** openstackgerrit has joined #openstack-lbaas13:38
*** kiran-r has quit IRC13:41
*** jschwarz has joined #openstack-lbaas14:04
*** shakamunyi has joined #openstack-lbaas14:09
*** shakamunyi has quit IRC14:13
*** shakamunyi has joined #openstack-lbaas14:35
jschwarzHi guys14:36
*** shakamunyi has quit IRC14:36
*** shakamunyi has joined #openstack-lbaas14:36
jschwarzI'm writing a patch for neutron-lbaas (https://review.openstack.org/#/c/174384/) which depends on a yet-unmerged change on neutron14:36
jschwarzI've put in the Depends-On statement on the commit message but the tests still seem to fail because of using the old code and not the unmerged one14:37
jschwarzany advice?14:37
*** nmagnezi has quit IRC14:39
*** Miouge has quit IRC15:04
*** Miouge has joined #openstack-lbaas15:07
*** fnaval has joined #openstack-lbaas15:21
*** mlavalle has joined #openstack-lbaas15:22
*** fnaval has quit IRC15:23
openstackgerritPhillip Toohill proposed stackforge/octavia: Add devstack plugin for octavia  https://review.openstack.org/16779615:24
*** fnaval has joined #openstack-lbaas15:26
*** jschwarz has quit IRC15:31
*** openstackgerrit has quit IRC15:37
*** openstackgerrit has joined #openstack-lbaas15:37
*** madhu_ak has joined #openstack-lbaas15:48
*** nmagnezi has joined #openstack-lbaas15:55
*** shakamunyi has quit IRC15:58
*** ajmiller has joined #openstack-lbaas16:08
*** kiran-r has joined #openstack-lbaas16:11
*** evgenyf has quit IRC16:16
xgermanbrogan should we skip today’s Octavia meeting so we have more time working on the demo>16:17
xgermanblogan16:17
bloganwell i think the main point of the meeting woudl be getting reviews done and hopefully merged, but we can do that throughout today16:17
blogani got the api to queue to controller working last night and pushed up a review to fix some minor issues16:18
xgermanwell, we also need to plan the Octavia sessions at the summit16:18
xgermanmail sent to the ML16:27
dougwighmm, i was going to see if we could finalize a meetup location today.16:36
dougwigcan we do that in channel?16:36
xgermanyes16:36
xgermanpcm said he can’t come :-(16:37
dougwigjschwarz: you've got a unit failure, which doesn't honor Depends-On. the tempest tests pass, though.16:38
dougwigwell, i think seattle was the only thing with a ton of votes. were there any other serious contenders?16:38
xgermansadly no —16:41
xgermanSeattle gets us all the HP people16:41
*** shakamunyi has joined #openstack-lbaas16:49
*** kobis has quit IRC16:50
*** nmagnezi has quit IRC16:53
*** crc32 has joined #openstack-lbaas16:55
*** mwang2 has joined #openstack-lbaas16:58
*** bharath_ has joined #openstack-lbaas17:00
rm_workand I like going to Seattle :P17:05
rm_workI hope the midcycle won't be "HP employees sitting in the HP office, meeting up" though17:05
rm_workgotta work out that travel budget... <_<17:05
*** shakamunyi has quit IRC17:05
*** SumitNaiksatam has quit IRC17:05
rm_workIf I can sleep on xgerman's couch again, I'm prolly good to come up ;)17:06
xgermansure - I can arrange for that17:06
rm_workhehe17:09
rm_workI feel like I'm in college again T_T17:09
rm_workand I totally understand one of my friends from Nokia better now -- he used to come to the Maemo conferences and sleep on my hotel room floor, and I always wondered why that was necessary....17:10
rm_workguessing now that Nokia was too cheap to actually send him, but he wanted to go badly enough that he just winged it :P17:11
*** kiran-r has quit IRC17:14
*** [1]evgenyf has joined #openstack-lbaas17:15
xgermanwe will have  bunch of places to hook people up ;-)17:19
dougwiganyone here object to seattle?17:19
*** shakamunyi has joined #openstack-lbaas17:20
xgermanblogan, ptoohill, ?17:22
xgermanpothole17:22
potholeI probably wont get to go in either case unless it was here. So I have no objections to where it is since no body wants it here17:25
*** SumitNaiksatam has joined #openstack-lbaas17:27
xgermanI hear you…17:41
*** bharath_ has quit IRC17:45
*** madhu_ak has quit IRC17:50
blogani odnt want it here :)18:15
blogani mean i do just to i'd be guaranteed to go, but its not fair to everyone else18:15
xgermanI feel the same about Seattle — I know we will get all of HP + not fair to non-HP people18:16
bloganwell we've already had 2 so that makes it even more unfair :)18:17
xgermanI think we had only one :-)18:17
bloganbut yeah seattle is fine for me too, adam would be the most likely person to go since he's got family up there18:18
xgermanmaybe you can switch gorge’s arm so pothole can come, too18:19
xgermanjorgem18:19
xgerman(this  spell checker is too smart for me)18:19
bloganjorgem is on vacation right now, like all managers do18:20
bloganamiright dougwig?18:20
xgermanwell, so just book then tickets...18:20
xgermaneverybody gets a trip18:20
bloganha upon booking he has to approve in 24 hours18:21
bloganor they get canceled18:21
xgermanit doesn’t go to the next manager who just clicks ok18:21
xgermannot a good policy18:21
blogangood question, maybe it does18:21
xgermanonly one way to find out18:21
bloganactually there's a few ways to find out18:22
bloganbut you want me to do the most extreme :)18:22
xgermanyou know me by now :-)_18:22
blogansome people just want to see the world burn18:22
xgermanlol18:23
*** SumitNaiksatam_ has joined #openstack-lbaas18:28
*** SumitNaiksatam has quit IRC18:30
*** SumitNaiksatam_ is now known as SumitNaiksatam18:30
openstackgerritPhillip Toohill proposed stackforge/octavia: Preparing for tempest testing  https://review.openstack.org/17219918:32
openstackgerritPhillip Toohill proposed stackforge/octavia: Add devstack plugin for octavia  https://review.openstack.org/16779618:33
*** apuimedo has quit IRC18:43
bloganjust had first successful api call to controller worker to create LB with amphora18:47
*** openstackgerrit has quit IRC18:47
*** ajmiller has quit IRC18:48
*** jkoelker has quit IRC18:48
*** pothole has quit IRC18:48
*** raginbajin has quit IRC18:48
*** rohara has quit IRC18:48
*** ajmiller has joined #openstack-lbaas18:49
*** pothole has joined #openstack-lbaas18:49
*** jkoelker has joined #openstack-lbaas18:49
*** raginbajin has joined #openstack-lbaas18:49
*** rohara has joined #openstack-lbaas18:49
*** openstackgerrit has joined #openstack-lbaas18:49
*** ajmiller has quit IRC18:54
*** ajmiller has joined #openstack-lbaas18:55
xgermanok, restocking and will try that our18:59
xgerman^^ johnsom19:00
*** [1]evgenyf has quit IRC19:00
johnsomYeeHaw!19:01
bloganyou'll have to pull down the latest devstack plugin, then cherry-pick 2 commits19:10
bloganhttps://review.openstack.org/#/c/180417/19:11
bloganhttps://review.openstack.org/#/c/179452/19:11
bloganjohnsom: would you happen to have any ideas why it would take half a minute to be able to ssh into an instance after it became ACTIVE?19:12
johnsomIf you are not using kvm I have seen qemu peg out the CPU and cause problems19:14
johnsomI'm running a pretty new set of patches, could build amp (IP still missing from DB), build LB, but now listener is failing with ssh error19:14
johnsomAlso, strange, 2015-05-06 12:11:16,614 - octavia.amphorae.drivers.haproxy.ssh_driver - DEBUG - Amphora HaproxyManager haproxy, updating listener 80, vip 172.24.4.519:15
johnsom| 51182631-9597-4f28-a2cc-f24c76f702a9 | amphora-0567a4d0-d85a-4e42-881f-1b0a49633ec5 | ACTIVE | -          | Running     | lb-mgmt-net=192.168.0.5; public=172.24.4.6 |19:15
johnsomNot sure why those two addresses are off.19:15
johnsomI think something bad happened with the network plug inside the amp19:17
bloganoh the vip will be a different ip than the amp's ip, something i've been thinking about changing bc i was doing it with the allowed address pairs, but we don't need that right now19:18
*** jkoelker has quit IRC19:18
*** pothole has quit IRC19:18
johnsomOh, I see, the vrrp stuff.19:18
*** pothole has joined #openstack-lbaas19:19
*** jkoelker has joined #openstack-lbaas19:19
bloganyeah19:19
johnsomYeah, something with the vip setup inside the amp went wrong, the lb_mgmt-net IP doesn't work anymore19:19
bloganthough to do that, the allocate_vip will have to just take in the loadbalancer object instad of the port_id, network_id, and ip_address bc it'll need to get a hold of the amphora's ip19:20
bloganoh adn i haven't tested this out with listeners create, i was going to do that next, just wanted to make sure an amp got created and all of taht with the creation of a load baalncer19:20
bloganjohnsom: did you try this through the api?19:23
johnsomNo, not yet19:24
xgermanon my devstack nova fails :-(19:33
xgermanbut I am hoping to catch up soon19:33
blogancan i just throw out "works for me" and go on vacation?19:39
xgermanwe schold at least reach works in devstack ;-)19:40
bloganlol what19:40
bloganyour autocorrect19:40
*** mwang2_ has joined #openstack-lbaas19:45
*** Aish has joined #openstack-lbaas19:45
xgermanit’s the Mac you know — shield make me look smarter19:46
*** mwang2 has quit IRC19:47
*** TrevorV|home has joined #openstack-lbaas19:55
dougwigseattle vs texas is the question again?  :)20:16
bloganha no20:16
bloganim all for boise!20:16
bloganbut no one else is20:16
dougwigheh. i'm all for a location that gets us more coders.20:16
dougwigwas there a decision?20:16
bloganill code wherever i'm at20:17
bloganbut it is nice to whiteboard in person and not have any meetings20:17
xgermanI will try to get budget for any location20:21
xgermanbut Seattle is easiest for HP20:21
bloganany towyard in america is easiest for me20:22
xgermanis there any place you guys can drive?20:22
bloganthe plane tickets wouldn't be the problem i suspect, i dont know20:23
xgermanok, we can put you up on copuches20:23
xgerman+ we have a nursing mothers room on site ;-)20:24
xgermanbut Boise we all can camp in dougwig’s backyard and shoot things20:25
openstackgerritmin wang proposed openstack/neutron-lbaas: Admin_state_up tempest test for health monitor  https://review.openstack.org/17882720:25
*** barclaac|2 has joined #openstack-lbaas20:32
rm_workI mean, yeah, I will go wherever, since it will probably be similar for me in any case (maybe get airfare covered, sleep on a couch)20:34
rm_workSeattle is just easy since I know people there20:34
dougwigalright, let's draw this to a close.  does anyone *object* to the location being seattle this time?20:34
*** barclaac has quit IRC20:35
crc32 didn't we have the last meet up in seattle?20:36
crc32oh wait never mind.20:37
xgermantechnically yes BUT we had one in San Antonio afterwards for LBaaS20:37
xgermansince we call that L4-L7 technically it’s the first metope of it’s kind ;-)20:37
bloganwe've had 2 in san antonio, one in seattle20:37
xgermanyep20:38
crc32I say go for it then.20:39
xgermanok, Seattle it is20:39
sballeyeah!!120:39
crc32or New york.20:39
rm_worklolol20:40
sballeNew York... Hotels are expensive :-(20:40
rm_workI mean, I want to go to new york :P20:40
rm_workbut yeah20:41
sballeyeah we'll all have to sleep in the subway20:41
xgermanairbnb?20:41
xgermanyouth hostel?20:41
blogandumpster?20:41
sballelol20:42
crc32oh sorry I was kidding about new york20:42
bloganmight get some free food out of that one20:42
sballecrc32 we know but this is fun anyway20:42
xgermanI would love to do it once in DC we can use the free Wifi from the Smithsonian and sit next to dinosaurs...20:44
sballe+120:44
blogani'd love to do it on the moon one day20:46
bloganhotels are cheap there20:46
sballebut there are few of them20:47
xgermanif you believe Apollo was a conspiracy we would be the first humans there20:48
bloganand you should tell buzz aldrin that to his face20:48
*** nuxxer has joined #openstack-lbaas20:49
*** TrevorV|home has quit IRC20:50
dougwigOk, Seattle.  Can someone update the ether pad?20:53
bloganxgerman can you put the HP office address up there too20:55
bloganor johnsom20:55
bloganor sballe20:55
nuxxer@xgerman Hi xgerman.. where can I locate lbaas v2 CLI ?20:55
sballe@blo20:56
sballeblogan: I believe xgerman20:56
sballeis doingit20:56
openstackgerritmin wang proposed openstack/neutron-lbaas: Admin_state_up tempest test for health monitor  https://review.openstack.org/17882720:57
xgermannuxxer lbaas v2 is part of the official neutron cli20:59
xgermanthe commands start with labs-* instead of lb20:59
xgermanlbaas20:59
nuxxerxgerman which is available on kilo time frame ?20:59
xgermanpart of kilo release20:59
nuxxerxgerman I could not find openstack page that gives full lbaas v2 CLIs.21:00
nuxxerxgerman https://wiki.openstack.org/wiki/Neutron/LBaaS/CLI21:01
nuxxerwhich is v1.21:01
xgermanI rmemeber vividly that we worked with the doc team on that21:01
nuxxerxgerman only reference on v2 CLI was .. devstack example.21:01
xgermanAish - you recall where our V2 CLI docs ended up?21:02
xgermannm - look at http://docs.openstack.org/cli-reference/content/neutronclient_commands.html21:03
xgermanhas all the v2 commands21:03
Aish http://docs.openstack.org/cli-reference/content/neutronclient_commands.html21:03
Aishyeah the same oone21:03
xgermanthanks Aish!21:04
xgermancheck that nuxxer21:04
Aishnuxxer everything that starts with “neutron lbaas “21:04
nuxxerone more dumb question.. still learning lbaas.21:06
Aishxgerman: np. you found that before I did ;)21:06
nuxxerwhen horizon can be there for v2 ?21:06
xgermanthat is planned for Liberty21:07
nuxxerThanks . xgerman Aish blogan21:07
*** ajmiller has quit IRC21:07
Aishnuxxer: :)21:08
*** nuxxer is now known as nuxxer-away21:11
*** nuxxer-away is now known as nuxxer21:18
xgermanhttps://www.irccloud.com/pastebin/lUV0c0jg21:21
xgermanblogan ^^ what’s that?21:21
bloganxgerman: that's just bad path handling, or lack of, tack on /v1/loadbalancers21:22
xgermanok, I would expect it to return a version string...21:23
bloganit should and it can! but doesn't now obviously21:23
xgermanyep, but v1/loadbalancers works21:24
xgermanso I am happy21:24
xgermanjust need to avoid that URL in the demo ;-)21:24
nuxxerquestion/answer here is limited to octavia .. correct ?21:34
rm_worknuxxer: this is the channel for neutron-lbaas and octavia both21:34
rm_worknuxxer: octavia is essentially developed by a subset of the neutron-lbaas team, very few that work on only octavia21:35
nuxxerFor v1. I found no good scheduler.. only one was ChanceScheduler... just randomly pick lbaas agent when there are multiple on net.21:35
blogana demo in front of everyone has to have some hiccups!21:35
rm_worknuxxer: is that for the haproxy driver?21:36
bloganrm_work: yeah it is21:36
rm_worknuxxer: not many of us here now really worked with v1 very much21:36
bloganits the only scheduler we have21:37
rm_workblogan or dougwig might be able to answer21:37
rm_workah, kk21:37
rm_workthere we go :P21:37
bloganbut another can be created and pointed at in the config21:37
rm_worknuxxer: V1, and especially the haproxy driver for V1, are not very robust21:37
rm_worknuxxer: well, the haproxy driver at least isn't (and is NOT recommended for production use)21:37
nuxxerso I added a few lines to make haproxy distributed over multiple agents.21:37
nuxxer@rm_work. I thought that octavia is also using haproxy for now.21:38
dougwigThere is only one stock v1 scheduler, and it sucks.21:38
rm_worknuxxer: the neutron-lbaas haproxy driver is "haproxy namespace agent driver"21:38
rm_worknuxxer: octavia will use haproxy, but not at all in the same manner21:39
bloganwhats wrong with random? it is perfect all the time!21:39
rm_workhaproxy != haproxy-driver21:39
nuxxer@rm_work. :- ) I see. yep. I am dying to see octavia in action. LOL21:39
rm_worknuxxer: us too :)21:39
rm_worknuxxer: there will be a demo soon -- that is what xgerman was discussing21:39
rm_workmaybe you can get in on that21:40
bloganit'll die when you see it in action!21:40
rm_workblogan:  ... possibly T_T21:40
rm_workwe're in pre-alpha :P21:40
nuxxer@blogan .. the problem with random is ...  one agent gets 10 proxies and the other might get 3 haproxies. :- )21:40
rm_workclosing in on alpha21:40
blogannuxxer: i was joking21:40
bloganlol21:40
rm_worknuxxer: your dataset is too small21:40
nuxxer@blogan .. not scaling well. :- )21:40
johnsomblogan It looks like we are making good progress.  I am currently stuck with the post vip plug call to the amp taking out lb_mgmt connectivity.21:40
rm_worknuxxer: get more LBs :P21:40
rm_workI am sure it should even out around 10,000 or so21:41
rm_worklol21:41
johnsomI have been trying to get a console working to debug21:41
* rm_work is also kidding21:41
bloganjohnsom: hmm, you able to see the logs and if it is able to ssh into it?21:41
rm_workbut seriously, it would21:41
xgermanblogan what did you use for port-id?21:42
johnsomYeah, the LB ssh for the post-vip-plug completed, but dead to the world after from what I could see in the logs21:42
bloganxgerman: i'm just passing network_id in right now so it creates a new port21:42
xgermanok21:42
nuxxer@rm_work I added about 5 lines. now the scheduler can even out haproxy processes over multiple agents.21:42
bloganjohnsom: so you can't ssh into that machine when it completes?21:43
johnsomYep, none of the IPs respond anymore21:43
bloganhmm21:44
*** ajmiller has joined #openstack-lbaas21:44
johnsomCan you get a listener fired up on your stack?21:44
bloganjohnsom: haven't gotten a chacne to try that yet21:47
johnsomblogan Ok.  I'm good through LB create.  Listener create fails as SSH doesn't connect21:48
bloganjohnsom: ahh, okay thats good then21:49
bloganthink we can get those reviews merged today then?21:49
johnsoma devstack issue with keystone and oslo module slowed me down, had to restack again21:49
bloganat least -1'ed21:49
xgermanhttps://www.irccloud.com/pastebin/cgRan6gZ21:50
xgermanI got to  post the create LB to the API but then it runs that course21:50
johnsomThese? https://review.openstack.org/179452 https://review.openstack.org/180417 https://review.openstack.org/16779621:50
bloganjohnsom: yep21:51
bloganxgerman: did it only try to ssh once?21:51
blogani mean twice21:51
xgermanI might add that nova list does;t show any vms21:52
bloganif you're doing this through the octavia api, then yeah that would be a problem21:52
johnsomThe ssh driver tries ten times to connect for me21:52
xgermanyep, through the Octavia API21:52
bloganjohnsom: yeah you can change that inthe config21:53
xgermanand devstack install 4721:53
xgermanand the two patches you mentioned21:53
bloganxgerman: are you nova listing under admin user?21:53
xgermanyes21:53
bloganhmm21:53
xgermanmight be me though… other tests never yield a vm but always said success21:53
bloganit woudln't have gotten to that point without a compute_id21:53
xgermanyeah, crazy21:54
bloganoh wait that started reverting before21:54
blogangot any logs before that?21:54
bloganwait, lol thats just a weird log21:55
johnsomblogan Let me take a quick pass on those and we can push for merge to make life easier.  The devstack seems like it is still evolving a bit, but that can be additional patch sets21:56
johnsomxgerman you paste bin indicates it could not build a vm21:57
bloganjohnsom: yeah, the devstack one would be great to go in bc then we can just git review -d things instead of cherry-picking on top21:57
*** ajmiller_ has joined #openstack-lbaas21:58
xgermanhttps://usercontent.irccloud-cdn.com/file/A60eeg45/o-cw.log21:58
johnsomnot sure why it would be calling the ssh driver after that though21:58
xgermanhere is the complete log21:58
johnsomThat is a really strange log22:00
xgermanyep, I am hitting all the strange stuff22:01
blogani would say that is an exceptional log22:01
*** ajmiller has quit IRC22:02
johnsomRun with debug true in the octavia config.  I don't understand how you got to postvipplug without an amp22:07
bloganjohnsom: good point, its not showing any INFO or DEBUG logs22:07
xgermanhttps://usercontent.irccloud-cdn.com/file/st3oz1YF/o-cw.log22:08
potholexgerman, just seen the -1 and message. Is the api still not running for you?22:09
xgermanI had to start it manually - worker runs22:09
potholeAre you specifiying o-api as an enabled service?22:10
xgermangood question, I used the sample config22:10
potholefrom readme?22:10
johnsomAllocated Amphora with id 1d831ec6-3961-4631-a11d-39375ffccba0 for load balancer with id b0f13f41-1f76-4fba-9e35-c340a1d253d2 execute /opt/stack/octavia/octavia/controller/worker/tasks/database_tasks.py:24622:11
xgermanfrom samples/local.conf22:11
potholeoh22:11
xgermanlike brogan I don;t know how to open Readmes22:11
potholeyea, then not sure how other things are running also22:11
pothole:P22:11
potholehttps://review.openstack.org/#/c/167796/47/devstack/README.md22:11
potholeExample of the enabled services. though, hk/hw arent used right now22:12
xgermanwhat I copied https://review.openstack.org/#/c/167796/47/devstack/samples/local.conf22:12
bloganxgerman: for some reason it cannot ssh into the amphora, key issue22:12
potholeYea, thats a generic devstack sample22:12
xgermanyeah, because there is no amphora nova list shows nada22:13
johnsomYeah, key issue22:13
xgermanpothole ok, I will make a patch and add o-api to that file ;-)22:13
potholeI never updated that, didnt know it was a thing. Ill need to add it.22:13
potholeThat'd be awesome!22:13
potholeThank you22:13
xgermanI basically copy that file and stack.sh22:13
xgermanisn’t that what I am supposed to do?22:14
potholeWell yea, I just use my own and specifically add the services22:14
potholeI overlooked this file22:14
potholeMy apologies22:16
johnsomblogan I think the rename to controllerworker.py is confusing given controllerworker is the library that other processes will be calling into22:17
bloganjohnsom: i agree honestly, but i wasn't thinking straight or something, for some reason naming it queue-consumer is just odd to me bc its so generic22:18
bloganbut that is really what it is22:18
johnsomHaha, we can go controller-queue-consumer or CQC22:19
bloganlol yikes22:19
bloganthe binary in setup.cfg should start with octavia- though22:20
johnsomOCQC22:20
johnsomoctavia-cqc22:20
bloganthat'd be fine, these are all just entry points, im terribel with naming things anyway but quick to hate names as well :)22:21
johnsomoctavia-croquet is all I think of22:21
bloganlol22:21
bloganwell that kind of makes sense22:22
blogancorquet balls being packets22:22
johnsomIt does actually22:22
johnsomwickets are the queue22:22
blogannow what is latin for croquet and wickets22:23
johnsomgoogle translate says croquet and wicket22:23
bloganlol22:24
johnsomgreek is κροκέ and γκισές22:25
johnsomnot so good22:25
bloganthats not confusing at all22:25
bloganlets just go with queue-consumer for now, easy to rename it in the future22:25
blogani'd still keep the binary created from setup as octavia-worker though, so we don't have to change the devstack plugin as well22:26
bloganor we can just wait till they both merge to change the name22:26
johnsomI think pothole is rolling a plugin patchset right now for the o-api issue22:28
johnsomIf you two could coordinate we can land those two changes and merge them both22:29
johnsomAs I don't have any other issues with either22:29
bloganpothole: are you doing another ps?22:29
bloganjohnsom: you fine with queue-consumer though?22:31
johnsomSure22:31
openstackgerritGerman Eichberger proposed stackforge/octavia: Add devstack plugin for octavia  https://review.openstack.org/16779622:31
johnsomblogan queue-consumer and octavia-worker are fine with me.  controller-worker is confusing.  German just landed the plugin change so let's stick with octavia-worker22:33
bloganokay im fine with taht then22:33
xgermanI fixed the o-api22:34
johnsomblogan https://review.openstack.org/#/c/167796/22:35
blogandone22:35
johnsomblogan are you rolling https://review.openstack.org/#/c/180417/?22:36
bloganyep22:36
johnsomOk, cool.  Looking over fixes patch22:36
openstackgerritBrandon Logan proposed stackforge/octavia: Fix queue producer handler to send ids  https://review.openstack.org/18041722:37
bloganjohnsom: ^^22:37
rm_worklet me know if you need reviews on things that are quick22:37
johnsomzippy22:37
xgermanour art department wants to AI files of the sticker22:37
bloganAI?22:37
rm_workswamped today but will make time for some reviews if necessary22:37
rm_worklooks like as of next week I might have ACTUAL TIME for reviews and stuff :)22:38
xgermanAdobe Illustrator22:38
openstackgerritMerged stackforge/octavia: Add devstack plugin for octavia  https://review.openstack.org/16779622:38
rm_workhttp://i.imgur.com/tGeke0X.gifv22:38
blogandamn we still need this reviewed and merged for the update changes22:38
bloganhttps://review.openstack.org/#/c/170989/22:38
bloganupdates will probably be wonky without, but we can still test create and delete22:39
johnsomLooking22:39
xgermanrm_work not that one22:39
johnsomAh, yes, the code that confused the heck out of me22:40
bloganjohnsom: yea i'd like to test it out, so i propose we hold off on merging just yet22:40
xgerman+122:40
johnsom+122:40
xgermanwho knows in what bizarre problems I might run22:40
xgermanchanneling my inner dummy user22:41
bloganif there are bizarre problems to be found, you will find them!22:41
bloganpower users find many bizarre problems as well22:41
xgermanrm_work: https://docs.google.com/drawings/d/1PvJj_8A6oFRdUN9xAgm9Z8joM33Z7CuTJUzCQn3GoCM/edit  — option D as AI file but I guess this is all we have22:45
xgermanI think we (=HP) ordered stickers and now there is talk about other schwag22:46
johnsomxgerman you can export svg from that google drawing and they can import that in AI22:47
xgermanyeah, I send it to them22:47
xgermanhope they can figure it out22:47
bloganim gonna head home right now, ill be on later tonight im sure, ill eventually get to creating a listener22:48
bloganof course tomorrow is literally full of meetings so i won't be around until after 522:48
johnsomOk, reviewing "fixed" will keep trying it out22:49
bloganyeah please do22:49
bloganbut hey at least we've got the api hooked up to controller now22:49
openstackgerritMerged stackforge/octavia: Fix queue producer handler to send ids  https://review.openstack.org/18041722:49
johnsomThat is a great step22:50
bloganand devstack plguin to spin it allup for us, and it works for 3/4's of us right now22:50
xgermanwell, I am hoping I can join that club soon22:50
xgermanhttps://www.irccloud.com/pastebin/YZ16FLuh22:52
xgermannot sure what I did this time — that is what my o-cw spits put22:53
johnsomDo you have this change? https://review.openstack.org/#/c/180417/2/octavia/cmd/octavia_worker.py22:54
xgermanok, I might be doing something wrong then22:55
*** SumitNaiksatam has quit IRC22:58
*** nuxxer has quit IRC23:01
rm_workxgerman: yeah TrevorV made those23:11
rm_workxgerman: but he did it in that document directly, so no other source files23:11
rm_worki trust the HP artists are capable :P23:11
*** chlong has joined #openstack-lbaas23:16
xgermanI hope so, too23:28
openstackgerritMerged openstack/neutron-lbaas: Fix a pid file name error  https://review.openstack.org/17906823:44
*** barclaac has joined #openstack-lbaas23:58
« Tuesday, 2015-05-05 Index Thursday, 2015-05-07 »

Generated by irclog2html.py 2.14.0 by Marius Gedminas - find it at mg.pov.lt!