| *** bradjones|away has joined #kolla | 00:05 | |
| *** vinkman has joined #kolla | 00:05 | |
| *** coolsvap has joined #kolla | 00:05 | |
| *** sdake has joined #kolla | 00:05 | |
| *** cemason has joined #kolla | 00:07 | |
| *** walters has joined #kolla | 00:07 | |
| *** jasonsb__ has joined #kolla | 00:07 | |
| *** loth has joined #kolla | 00:07 | |
| *** mandre has joined #kolla | 00:07 | |
| *** clayton has joined #kolla | 00:07 | |
| *** daneyon has joined #kolla | 00:08 | |
| *** pbourke has joined #kolla | 00:08 | |
| *** slagle has joined #kolla | 00:08 | |
| *** zehicle has joined #kolla | 00:08 | |
| *** akscram has joined #kolla | 00:08 | |
| *** nihilifer has joined #kolla | 00:08 | |
| *** kragniz has joined #kolla | 00:08 | |
| *** ByPasS has joined #kolla | 00:08 | |
| *** bmace has joined #kolla | 00:11 | |
| *** alisonh has joined #kolla | 00:11 | |
| *** bigjools has joined #kolla | 00:11 | |
| *** siwos has joined #kolla | 00:11 | |
| *** mfalatic has joined #kolla | 00:11 | |
| *** cloudnau_ has joined #kolla | 00:13 | |
| *** Ti-mo has joined #kolla | 00:13 | |
| *** larsks has joined #kolla | 00:13 | |
| *** kelkoobenoitr has joined #kolla | 00:13 | |
| *** Windir has joined #kolla | 00:13 | |
| *** harmw has joined #kolla | 00:13 | |
| *** asalkeld has joined #kolla | 00:14 | |
| *** kfox1111 has joined #kolla | 00:15 | |
| *** openstackgerrit has joined #kolla | 00:15 | |
| *** gfidente|afk has joined #kolla | 00:15 | |
| *** apmelton has joined #kolla | 00:15 | |
| *** jmccarthy has joined #kolla | 00:15 | |
| *** thomasem_ has joined #kolla | 00:15 | |
| *** jpeeler has joined #kolla | 00:15 | |
| *** kjelly has joined #kolla | 00:15 | |
| *** anteaya has joined #kolla | 00:15 | |
| *** cloudnau_ has quit IRC | 00:16 | |
| *** diogogmt has joined #kolla | 00:16 | |
| *** prometheanfire has joined #kolla | 00:16 | |
| *** shakamunyi has joined #kolla | 00:16 | |
| *** diogogmt has quit IRC | 00:17 | |
| *** dims_ has joined #kolla | 00:19 | |
| *** unicell has joined #kolla | 00:19 | |
| *** vincent_vdk has joined #kolla | 00:19 | |
| *** SamYaple has joined #kolla | 00:19 | |
| *** Slower has joined #kolla | 00:19 | |
| *** kysse has joined #kolla | 00:19 | |
| *** cloudnull has joined #kolla | 00:19 | |
| *** sdake has quit IRC | 00:20 | |
| *** dims_ has quit IRC | 00:20 | |
| *** dims_ has joined #kolla | 00:26 | |
| *** dims_ has quit IRC | 00:27 | |
| *** dims_ has joined #kolla | 00:27 | |
| *** alisonh has quit IRC | 00:29 | |
| *** achanda has joined #kolla | 00:32 | |
| *** achanda has quit IRC | 00:34 | |
| *** sdake has joined #kolla | 00:44 | |
| *** Kennan has quit IRC | 00:44 | |
| *** Kennan2 has joined #kolla | 00:44 | |
| *** alisonh has joined #kolla | 00:46 | |
| *** achanda has joined #kolla | 00:50 | |
| *** alisonh has quit IRC | 00:52 | |
| *** alisonh has joined #kolla | 00:58 | |
| *** sdake has quit IRC | 00:59 | |
| *** dwalsh has joined #kolla | 01:00 | |
| *** dwalsh has quit IRC | 01:21 | |
| *** jasonsb__ has quit IRC | 01:24 | |
| *** dolpher has joined #kolla | 01:28 | |
| *** daneyon has quit IRC | 01:36 | |
| *** tfukushima has joined #kolla | 01:45 | |
| *** tpot has joined #kolla | 01:49 | |
| *** diogogmt has joined #kolla | 01:54 | |
| openstackgerrit | Kuo-tung Kao proposed openstack/kolla: add "registry" flag to "tools/build.py" https://review.openstack.org/234629 | 02:03 |
|---|---|---|
| *** unicell has quit IRC | 02:04 | |
| *** achanda has quit IRC | 02:09 | |
| *** achanda has joined #kolla | 02:11 | |
| *** dims_ has quit IRC | 02:14 | |
| *** dims_ has joined #kolla | 02:14 | |
| *** achanda has quit IRC | 02:31 | |
| *** CBR09 has joined #kolla | 02:48 | |
| *** dims_ has quit IRC | 02:49 | |
| *** dims_ has joined #kolla | 02:49 | |
| *** dims_ has quit IRC | 02:51 | |
| *** dims_ has joined #kolla | 02:51 | |
| *** achanda has joined #kolla | 02:53 | |
| *** tobe has joined #kolla | 03:11 | |
| *** achanda has quit IRC | 03:16 | |
| *** achanda has joined #kolla | 03:20 | |
| *** achanda has quit IRC | 03:27 | |
| *** tpot has quit IRC | 03:47 | |
| *** achanda has joined #kolla | 04:09 | |
| *** achanda has quit IRC | 04:10 | |
| *** achanda has joined #kolla | 04:11 | |
| *** tfukushima has quit IRC | 04:25 | |
| *** dims_ has quit IRC | 04:29 | |
| *** diga has joined #kolla | 04:32 | |
| *** Kennan2 has quit IRC | 04:43 | |
| *** Kennan has joined #kolla | 04:44 | |
| *** unicell has joined #kolla | 04:45 | |
| *** unicell has quit IRC | 04:49 | |
| *** tpot has joined #kolla | 04:52 | |
| *** unicell has joined #kolla | 04:53 | |
| zehicle | I've got the docker system setup so it can pull Kolla layers; however the first container "docker.io/kollaglue/centos-binary-data" does not seem to exist. Any suggestions? | 04:53 |
| zehicle | I tried the ubuntu one too | 04:53 |
| zehicle | the other containers are OK, but that one seems to be missing | 04:53 |
| *** gfidente|afk is now known as gfidente | 04:56 | |
| *** diga has quit IRC | 04:57 | |
| mandre | zehicle, you should build the images yourself and not rely on the ones on the docker hub | 04:57 |
| mandre | we're going to push all images on the docker.io registry as soon as we tag the release | 04:58 |
| mandre | very soon | 04:58 |
| *** zehicle has quit IRC | 05:08 | |
| *** achanda has quit IRC | 05:17 | |
| *** tfukushima has joined #kolla | 05:26 | |
| *** achanda has joined #kolla | 05:28 | |
| SamYaple | morning | 05:31 |
| *** tfukushima has quit IRC | 05:31 | |
| *** tfukushima has joined #kolla | 05:37 | |
| *** tpot has quit IRC | 05:38 | |
| mandre | hey SamYaple | 05:58 |
| mandre | have you seen this? http://venturebeat.com/2015/10/15/source-red-hat-is-buying-ansible-for-more-than-100m/ | 05:58 |
| SamYaple | welp. time to use a new technology | 05:59 |
| SamYaple | im kidding :P | 05:59 |
| mandre | i couldn't confirm | 05:59 |
| mandre | well, we'll know soon enough if it's true | 05:59 |
| *** sdake has joined #kolla | 06:01 | |
| SamYaple | well thats interesting | 06:02 |
| *** inc0 has joined #kolla | 06:03 | |
| *** achanda has quit IRC | 06:14 | |
| *** exploreshaifali has joined #kolla | 06:15 | |
| *** sdake has quit IRC | 06:16 | |
| *** achanda has joined #kolla | 06:18 | |
| *** inc0 has quit IRC | 06:22 | |
| *** dims_ has joined #kolla | 06:26 | |
| *** achanda has quit IRC | 06:28 | |
| *** dims_ has quit IRC | 06:28 | |
| *** tobe has quit IRC | 06:28 | |
| *** athomas has joined #kolla | 06:35 | |
| *** achanda has joined #kolla | 06:37 | |
| *** tobe has joined #kolla | 06:39 | |
| *** tpot has joined #kolla | 06:50 | |
| *** tpot has quit IRC | 06:54 | |
| *** achanda has quit IRC | 06:58 | |
| *** inc0 has joined #kolla | 06:58 | |
| inc0 | lol, business news time! http://venturebeat.com/2015/10/15/source-red-hat-is-buying-ansible-for-more-than-100m/ | 06:59 |
| SamYaple | yea mandre posted that before | 07:01 |
| *** tobe has quit IRC | 07:01 | |
| openstackgerrit | Sam Yaple proposed openstack/kolla: Download, install, and enable rabbitmq_cluster https://review.openstack.org/235756 | 07:11 |
| openstackgerrit | Sam Yaple proposed openstack/kolla: Use rabbitmq-clusterer when deploying https://review.openstack.org/235757 | 07:11 |
| SamYaple | super impressed with clusterer! | 07:11 |
| SamYaple | way way better than traditional clustering for rabbitmq | 07:11 |
| *** tobe has joined #kolla | 07:18 | |
| *** exploreshaifali has quit IRC | 07:21 | |
| *** dims_ has joined #kolla | 07:28 | |
| *** shardy has joined #kolla | 07:29 | |
| *** jmccarthy has quit IRC | 07:30 | |
| *** jmccarthy has joined #kolla | 07:30 | |
| *** sdake has joined #kolla | 07:33 | |
| *** dims_ has quit IRC | 07:33 | |
| SamYaple | man clusterer is great | 07:34 |
| *** tfukushima has quit IRC | 07:36 | |
| inc0 | really? | 07:41 |
| SamYaple | check out those patches | 07:42 |
| SamYaple | (only tested on centos) | 07:42 |
| SamYaple | its fast | 07:42 |
| SamYaple | hot reloads | 07:42 |
| *** shardy_ has joined #kolla | 07:43 | |
| *** shardy has quit IRC | 07:45 | |
| inc0 | how about scallability/partition tolerance? | 07:48 |
| *** shardy_ has quit IRC | 07:49 | |
| inc0 | I mean it would be hard to be worse thank mirrored queues which simply loses everything | 07:49 |
| *** shardy has joined #kolla | 07:50 | |
| SamYaple | its got alot better support for all of that | 07:52 |
| SamYaple | apparently they have been using it internally at pivotal for 2 years :/ | 07:52 |
| SamYaple | inc0: try it out | 07:52 |
| SamYaple | i need a centos tester anyway | 07:53 |
| inc0 | I don't have centos aruond :( | 07:53 |
| SamYaple | oh well | 07:53 |
| SamYaple | ubuntu tester then | 07:53 |
| SamYaple | youll see its much more awesome | 07:53 |
| inc0 | gates says its doesn't build, but let see for ourself | 07:56 |
| sdake | evening | 07:56 |
| inc0 | hi sdake | 07:56 |
| SamYaple | evening sdake | 07:56 |
| inc0 | do we have session schedule yet | 07:57 |
| inc0 | ? | 07:57 |
| SamYaple | or a liberty tag :p | 07:57 |
| sdake | lazy | 07:57 |
| sdake | http://venturebeat.com/2015/10/15/source-red-hat-is-buying-ansible-for-more-than-100m/ | 07:58 |
| SamYaple | come on ptl | 07:58 |
| SamYaple | PTL some stuff! | 07:58 |
| SamYaple | yea weve been talking about that | 07:58 |
| sdake | wonder if its true | 07:58 |
| SamYaple | if only we knew someone who worked at ansible... or redhat... | 07:59 |
| inc0 | tripleo moving to ansible \o/ | 07:59 |
| sdake | or both | 07:59 |
| SamYaple | ;) | 07:59 |
| sdake | tripelo is moving to ansible? | 07:59 |
| openstackgerrit | Sam Yaple proposed openstack/kolla: Use rabbitmq-clusterer when deploying https://review.openstack.org/235757 | 08:00 |
| openstackgerrit | Sam Yaple proposed openstack/kolla: Download, install, and enable rabbitmq_cluster https://review.openstack.org/235756 | 08:00 |
| openstackgerrit | Sam Yaple proposed openstack/kolla: WIP remove vip for rabbitmq https://review.openstack.org/235777 | 08:00 |
| sdake | this is part of matt hicks escape plan from tirpleo | 08:00 |
| sdake | by ansible, then its ok to kill the 50 million in investment on tripleo | 08:00 |
| sdake | buy that is | 08:00 |
| kjelly | I am working on the doc. Should I preserve old style ? leave old style "-n 127.0.0.1:4000/kollaglue" and new style " --registry 127.0.0.1 -n kollaglue" https://review.openstack.org/#/c/234629/ | 08:02 |
| SamYaple | jesus 50 million | 08:02 |
| SamYaple | kjelly: just put the new style | 08:02 |
| kjelly | Ok | 08:02 |
| SamYaple | the old style isnt broken | 08:02 |
| SamYaple | so those who do it by acciedent it will still work | 08:03 |
| SamYaple | no need to document it | 08:03 |
| sdake | ya thats the word samyaple | 08:03 |
| sdake | 4 years of 50+ engineers is expesnive | 08:03 |
| SamYaple | esh | 08:03 |
| SamYaple | sunk cost much | 08:03 |
| SamYaple | hey sdake ^^ clusterer works fantastic | 08:04 |
| SamYaple | im super happy with it | 08:04 |
| SamYaple | like its crazy the difference | 08:04 |
| sdake | i'm going back to bed soon | 08:07 |
| inc0 | https://aphyr.com/posts/315-call-me-maybe-rabbitmq for good sleep | 08:07 |
| kjelly | SamYaple: Will the patch backport ? | 08:08 |
| inc0 | I'd be mildly against backporting it | 08:09 |
| inc0 | it changes CLI, it's not really a bugfix | 08:09 |
| SamYaple | kjelly: yea im with inc0. unless we run into an issue with liberty i say we leave it as is for liberty | 08:10 |
| SamYaple | i waited on doing it for the reason of stability | 08:10 |
| kjelly | SamYaple: Ok | 08:12 |
| SamYaple | kjelly: if there are issue with rabbitmq we can disuss a backport | 08:12 |
| inc0 | kjelly, did you ask about Sams patchset or yours? | 08:13 |
| inc0 | I'm totally against backporting Sam's work right now...it's backward incompatible | 08:13 |
| kjelly | SamYaple, inc0: I am asking my patch https://review.openstack.org/#/c/234629/ | 08:14 |
| inc0 | then yeah, I'm mildly against backporting | 08:14 |
| SamYaple | oh sorry kjelly i was confused | 08:15 |
| SamYaple | kjelly: we can backport that since it is a non-breaking thing. it doesnt even change/deprecate existing behavoir | 08:15 |
| SamYaple | the wholep oint was to reduce confusion for some and it does that | 08:15 |
| kjelly | SamYaple: That's ok. I am not say it clearly. | 08:16 |
| sdake | i'd prefer to leave liberty alone until after summit | 08:17 |
| inc0 | I'm not too fond of backporting something that changes CLI tho Sam...I know old stuff will keep working but still, if someone install liberty, docs should stay liberty imho | 08:17 |
| inc0 | and way you build stuff should stay the same | 08:17 |
| SamYaple | inc0: again nothing changes | 08:18 |
| SamYaple | thats m point | 08:18 |
| inc0 | new option is introduced...let's do as sdake suggested and wait until summit ok? | 08:18 |
| inc0 | with backporting whatever except from bugfixes | 08:18 |
| *** tfukushima has joined #kolla | 08:18 | |
| inc0 | we don't want to have liberty as sorta "master" | 08:19 |
| inc0 | we have master for that | 08:19 |
| SamYaple | inc0: dont worry, I dont get to single handly merge something | 08:20 |
| SamYaple | thats said i am 100% +2 on that backport | 08:20 |
| SamYaple | it only adds to the cli | 08:20 |
| inc0 | and I'm saying I won't -2 it;) just I need to be swayed a bit more to cast +2...but any other core can do that | 08:21 |
| inc0 | still, let's start with merging it to master | 08:21 |
| SamYaple | kjelly: i would say propose it as a backport after it merges. inc0 i we will make sure it doesnt merge into liberty without your concerns being addressed and us all coming to an agreement | 08:22 |
| SamYaple | fair? | 08:22 |
| inc0 | agreed | 08:22 |
| kjelly | SamYaple: Ok. But if no backport, we should document old style because build.py in liberty doesn't support registry flag. | 08:25 |
| SamYaple | kjelly: well you arent changin the docs in stable/liberty | 08:26 |
| SamYaple | you are chnaning the docs in master | 08:26 |
| inc0 | kjelly, doc for liberty will be still available | 08:26 |
| inc0 | so don't worry about that | 08:26 |
| kjelly | SamYaple, inc0: Ok. Thanks. | 08:27 |
| inc0 | np, thanks for your work! | 08:28 |
| sdake | docsfor master are all that are published fyi | 08:28 |
| kjelly | By the way, Is it always master in the site ? http://docs.openstack.org/developer/kolla/ | 08:28 |
| sdake | yes kjelly | 08:28 |
| SamYaple | sdake: sure, but we have to update docs for master | 08:28 |
| inc0 | yeah, but it's obvious that they relate to master | 08:28 |
| kjelly | Ok. Thank you for answering. | 08:29 |
| *** dims_ has joined #kolla | 08:29 | |
| sdake | wonder how muc hdehann going to end up with out of ansible | 08:31 |
| sdake | 20-30 mil would be my guess | 08:31 |
| inc0 | duh I need to write some cool stuff that people will love and get shitload of money | 08:32 |
| sdake | kolla - its been written | 08:33 |
| inc0 | yeah and it's not for sale | 08:33 |
| inc0 | I can be kollsuntant tho | 08:33 |
| *** dims_ has quit IRC | 08:35 | |
| SamYaple | sdake: i highly doubt that | 08:35 |
| sdake | if i cared i could turn kolla into a 100 mllion aquisition | 08:36 |
| SamYaple | sdake: i highly doubt that | 08:36 |
| inc0 | openstack is too narrow I think for that stuff...unless you'd like to deploy and support | 08:37 |
| sdake | my plans would be beyond openstack | 08:37 |
| SamYaple | you sure you arent starting a startup sdake.... | 08:37 |
| sdake | ya too lazy | 08:38 |
| openstackgerrit | Kuo-tung Kao proposed openstack/kolla: add "registry" flag to "tools/build.py" https://review.openstack.org/234629 | 08:38 |
| SamYaple | no joke | 08:38 |
| SamYaple | ;) | 08:38 |
| sdake | i've got a porsche payment | 08:38 |
| SamYaple | its pronouced porsche | 08:38 |
| sdake | right | 08:38 |
| sdake | doesn't chang e the fact that i have a payment ;) | 08:39 |
| *** pbourke has quit IRC | 08:39 | |
| *** pbourke has joined #kolla | 08:39 | |
| sdake | think installshield 2015 | 08:40 |
| sdake | easy 100 million aquisition | 08:40 |
| sdake | only need a round a, plus 18 months max | 08:41 |
| SamYaple | so easy amiright | 08:41 |
| sdake | no it would be the hardest thing i've ever done | 08:41 |
| sdake | but good odds | 08:41 |
| SamYaple | you do realized its not just teh tech right? | 08:42 |
| SamYaple | you need a business model and customers and employees | 08:42 |
| sdake | its mostly the employees | 08:43 |
| SamYaple | good luck getting funding with no business model or customers | 08:43 |
| sdake | business model is easy | 08:43 |
| sdake | customers harder | 08:43 |
| sdake | i coudl easily get a round a if i had a seed round to build an engineering and sales team | 08:44 |
| SamYaple | im going to doubt you very hard like | 08:44 |
| sdake | the only piece i lack is someone to head up sales | 08:45 |
| sdake | but it doen't matter | 08:45 |
| sdake | i make plenty of money already | 08:45 |
| sdake | do wht i wnt | 08:45 |
| sdake | keep what schedule i want | 08:45 |
| sdake | work on what i want | 08:45 |
| sdake | why disrupt that for a shot at the title? :) | 08:45 |
| SamYaple | this is why i doubt you! | 08:46 |
| sdake | people management is extremely stressful | 08:46 |
| sdake | i do it all day long in #kolla | 08:46 |
| sdake | and that is enough for me ;) | 08:46 |
| *** sdake has quit IRC | 08:48 | |
| *** sdake has joined #kolla | 08:50 | |
| *** inc0_ has joined #kolla | 08:52 | |
| *** inc0 has quit IRC | 08:54 | |
| sdake | one day when i have an idea better then a generic deployment tool using ansible and docker dependencies i'll start a gig | 08:55 |
| sdake | but so far that idea is one of the best that has emerged from my work | 08:55 |
| SamYaple | ansible wasnt your idea :/ | 08:55 |
| sdake | if i had started a business with beekhof around corosync and pacemaker we coudl have made a100 million aquisition easily | 08:56 |
| sdake | we have a deployment tool that is specific to a platform, my idea is a generic deployment tool for all software | 08:56 |
| sdake | easy to understand easy to sell patterns already developed | 08:56 |
| sdake | has almost all the key properties of a strong startup | 08:57 |
| sdake | every company and their brother is trying to sort out what to dow ith ocntainers | 08:58 |
| *** inc0_ has quit IRC | 08:58 | |
| sdake | we have one solid pattern that can be replicated | 08:59 |
| *** inc0 has joined #kolla | 08:59 | |
| sdake | but too fat, dumb, and lazy | 08:59 |
| *** inc0 has quit IRC | 08:59 | |
| SamYaple | eh. i dont actually like kollas model for the average application | 08:59 |
| SamYaple | works great for openstack | 08:59 |
| SamYaple | but i wouldnt use it past that | 08:59 |
| *** inc0 has joined #kolla | 08:59 | |
| sdake | there are hundreds of enterprise applications that would fit the kolla model | 09:00 |
| sdake | SAP oracle etc | 09:01 |
| *** diga has joined #kolla | 09:01 | |
| inc0 | well containers are by far best deployment tool I've seen so far | 09:01 |
| sdake | ya highly repeatable pattern | 09:02 |
| kysse | I think you should change name to kontti which means container in finnish ":)" | 09:02 |
| sdake | could be automated | 09:02 |
| sdake | no name changes ;-) | 09:02 |
| kysse | :-( | 09:02 |
| kysse | h8ter | 09:02 |
| sdake | once a project has a name it hsould stick with it forever | 09:02 |
| sdake | nah its a good name | 09:03 |
| SamYaple | not change 5 times in 1 year? | 09:03 |
| *** dolpher has quit IRC | 09:03 | |
| sdake | haha SamYaple | 09:03 |
| * sdake points | 09:03 | |
| sdake | osad sosad | 09:04 |
| sdake | osadsosad | 09:05 |
| sdake | that woudl be a good name ;-) | 09:05 |
| SamYaple | /kick sdake go to bed youre drunk | 09:05 |
| sdake | have had no liquor in the last few weeks | 09:05 |
| inc0 | thats why! | 09:05 |
| inc0 | drink something! | 09:05 |
| SamYaple | you clearly dont have a problem with alcohol | 09:06 |
| SamYaple | you have a problem without it | 09:06 |
| sdake | its only a problem if you cant afford it ;) | 09:06 |
| sdake | the best part about if that rumor is true about ansible is all of robyn's rsus will be valid again since she hasn't been away from redhat for more then 1 year | 09:09 |
| inc0 | I've seen porsche topic....I guess it's priorities | 09:10 |
| sdake | i've got young chlidren, need to pay for their university etc | 09:10 |
| sdake | startups generally fail | 09:10 |
| sdake | cant take the risk to be honest | 09:10 |
| sdake | once they are out of school, I CAN take the risk | 09:10 |
| inc0 | I was referring to alcohol;) | 09:11 |
| sdake | if i was young and single i'd probbly roll a busines swith kolla | 09:11 |
| sdake | but i'm not either | 09:11 |
| sdake | off to bed night | 09:14 |
| inc0 | good night | 09:15 |
| *** diga has quit IRC | 09:24 | |
| *** inc0 has quit IRC | 09:30 | |
| *** sdake has quit IRC | 09:30 | |
| *** dims_ has joined #kolla | 09:31 | |
| *** inc0 has joined #kolla | 09:31 | |
| *** dims_ has quit IRC | 09:37 | |
| *** dims_ has joined #kolla | 09:48 | |
| *** inc0 has quit IRC | 09:57 | |
| *** inc0 has joined #kolla | 10:06 | |
| *** sdake has joined #kolla | 10:12 | |
| *** bradjones|away is now known as bradjones | 10:15 | |
| *** exploreshaifali has joined #kolla | 10:16 | |
| *** tfukushima has quit IRC | 10:19 | |
| *** sdake has quit IRC | 10:31 | |
| *** bmace has quit IRC | 10:38 | |
| *** bmace has joined #kolla | 10:38 | |
| *** diga has joined #kolla | 10:39 | |
| *** achanda has joined #kolla | 11:02 | |
| *** achanda has quit IRC | 11:21 | |
| *** rhallisey has joined #kolla | 11:23 | |
| *** diga has quit IRC | 11:44 | |
| *** gfidente is now known as gfidente|afk | 11:49 | |
| *** achanda has joined #kolla | 12:14 | |
| *** exploreshaifali has quit IRC | 12:15 | |
| *** inc0_ has joined #kolla | 12:29 | |
| *** dwalsh has joined #kolla | 12:29 | |
| *** inc0 has quit IRC | 12:31 | |
| *** inc0_ has quit IRC | 12:46 | |
| *** inc0 has joined #kolla | 12:46 | |
| *** tobe has quit IRC | 12:50 | |
| *** achanda has quit IRC | 12:51 | |
| *** athomas has quit IRC | 12:55 | |
| *** athomas has joined #kolla | 13:00 | |
| *** inc0 has quit IRC | 13:09 | |
| *** inc0_ has joined #kolla | 13:09 | |
| *** zehicle has joined #kolla | 13:13 | |
| *** dwalsh has quit IRC | 13:14 | |
| *** cemason has quit IRC | 13:25 | |
| *** inc0_ has quit IRC | 13:29 | |
| *** inc0 has joined #kolla | 13:33 | |
| *** CBR09 has quit IRC | 13:41 | |
| *** dwalsh has joined #kolla | 13:58 | |
| *** Hui has joined #kolla | 14:02 | |
| *** mfalatic has quit IRC | 14:18 | |
| *** mfalatic has joined #kolla | 14:20 | |
| *** gfidente|afk is now known as gfidente | 14:32 | |
| *** diogogmt has quit IRC | 14:32 | |
| *** diogogmt_ has joined #kolla | 14:32 | |
| *** sdake has joined #kolla | 14:33 | |
| *** inc0 has quit IRC | 14:34 | |
| *** inc0_ has joined #kolla | 14:34 | |
| sdake | morning | 14:39 |
| *** diogogmt_ has quit IRC | 14:39 | |
| *** inc0_ has quit IRC | 14:40 | |
| *** diogogmt has joined #kolla | 14:41 | |
| *** cemason has joined #kolla | 14:46 | |
| *** athomas has quit IRC | 14:47 | |
| rhallisey | morning | 14:47 |
| *** unicell has quit IRC | 14:48 | |
| *** unicell1 has joined #kolla | 14:48 | |
| Hui | morning, sdake, rhallisey | 15:00 |
| *** athomas has joined #kolla | 15:09 | |
| *** sdake has quit IRC | 15:11 | |
| *** walters has quit IRC | 15:14 | |
| *** shardy_ has joined #kolla | 15:18 | |
| *** cloudnautique has joined #kolla | 15:19 | |
| *** shardy has quit IRC | 15:20 | |
| *** bkosick has joined #kolla | 15:20 | |
| *** shardy_ has quit IRC | 15:24 | |
| *** shardy has joined #kolla | 15:24 | |
| *** cemason has quit IRC | 15:26 | |
| *** blahRus has joined #kolla | 15:50 | |
| *** bkosick has quit IRC | 15:55 | |
| *** walters has joined #kolla | 16:03 | |
| *** daneyon has joined #kolla | 16:19 | |
| *** gfidente is now known as gfidente|afk | 16:20 | |
| *** walters has quit IRC | 16:21 | |
| *** unicell1 has quit IRC | 16:25 | |
| *** bradjones has quit IRC | 16:26 | |
| *** sdake has joined #kolla | 16:33 | |
| *** sdake_ has joined #kolla | 16:36 | |
| *** sdake has quit IRC | 16:38 | |
| *** walters has joined #kolla | 16:47 | |
| *** dims_ has quit IRC | 16:54 | |
| *** unicell has joined #kolla | 16:55 | |
| *** dims_ has joined #kolla | 16:55 | |
| *** dtturner has joined #kolla | 16:58 | |
| *** athomas has quit IRC | 17:00 | |
| *** sdake_ is now known as sdake | 17:04 | |
| *** achanda has joined #kolla | 17:10 | |
| *** inc0 has joined #kolla | 17:12 | |
| *** inc0 has quit IRC | 17:12 | |
| *** jasonsb has joined #kolla | 17:13 | |
| *** shardy is now known as shardy_afk | 17:16 | |
| *** gfidente|afk is now known as gfidente | 17:28 | |
| *** achanda has quit IRC | 17:30 | |
| *** achanda has joined #kolla | 17:32 | |
| *** sdake has quit IRC | 17:43 | |
| *** sdake has joined #kolla | 17:45 | |
| *** jasonsb has quit IRC | 18:09 | |
| *** jasonsb has joined #kolla | 18:10 | |
| *** jasonsb has quit IRC | 18:10 | |
| *** jasonsb has joined #kolla | 18:12 | |
| *** achanda has quit IRC | 18:27 | |
| *** achanda has joined #kolla | 18:29 | |
| *** achanda has quit IRC | 18:35 | |
| *** dims_ is now known as dimsum__ | 18:41 | |
| *** achanda has joined #kolla | 18:56 | |
| *** daneyon has quit IRC | 18:56 | |
| *** daneyon_ has joined #kolla | 18:56 | |
| *** sdake has quit IRC | 19:09 | |
| *** sdake has joined #kolla | 19:12 | |
| *** sdake has quit IRC | 19:14 | |
| *** sdake has joined #kolla | 19:21 | |
| *** sdake has quit IRC | 19:37 | |
| *** achanda has quit IRC | 19:47 | |
| *** dwalsh has quit IRC | 20:00 | |
| *** bkosick has joined #kolla | 20:02 | |
| *** bkosick has quit IRC | 20:16 | |
| *** blahRus has quit IRC | 20:25 | |
| *** sdake has joined #kolla | 20:26 | |
| kfox1111 | does anyone have a galera recovery mechanism for after power outages? | 20:26 |
| kfox1111 | not seeing a way to do that, and its really critical. | 20:26 |
| *** walters has left #kolla | 20:43 | |
| larsks | kfox1111: are you looking for documentation (e.g., "how do I recover from a total cluster failure"), or are you asking about implementation? | 20:47 |
| *** achanda has joined #kolla | 20:47 | |
| kfox1111 | the former. | 20:47 |
| larsks | kfox1111: Okay. This is a decent overview: https://www.linkedin.com/pulse/20140730200902-5707100-galera-mysql-cluster-crash-recovery | 20:47 |
| kfox1111 | yeah, I've read that in the past. | 20:48 |
| kfox1111 | I'm wondering how I do that with kolla. | 20:48 |
| larsks | Ah, okay. I can't help you there. I've had to run through that a few times with a pacemaker cluster... | 20:49 |
| kfox1111 | do I have to manually just start the containers on the hosts via docker, via ansible? manually tweak configs/rerun, etc. | 20:49 |
| kfox1111 | I'm worried without that procedure, we can't use kolla. | 20:50 |
| kfox1111 | I can't just abandon the production db on power outage, or restore from backups. | 20:50 |
| larsks | kfox1111: probably want to wait for some of the more kolla-familiar people to come online (e.g., SamYaple, sdake, etc) | 20:50 |
| larsks | They can probably provide you with a better answer. | 20:50 |
| kfox1111 | k. thanks. | 20:50 |
| sdake | hey kfox1111 | 20:50 |
| sdake | we talked about a lights out failure | 20:51 |
| sdake | and im not sur ewe came to a concluson as to how to handle it | 20:51 |
| sdake | other thene the bringup has to be semi-manual | 20:51 |
| sdake | i dont know the detilas | 20:51 |
| sdake | samyaple is your man there | 20:51 |
| sdake | he is sleeping atm | 20:52 |
| kfox1111 | k. I'll ask him when he gets up. thanks. | 20:53 |
| *** achanda has quit IRC | 20:53 | |
| sdake | the obvious but unsatisifying answer is dont permit a lights out fialure (by power backing your nodes with ups) | 20:53 |
| kfox1111 | yeah, galera bringup's a bit scary to me. :/ | 20:53 |
| kfox1111 | "just pick the newest and bring that one back up first with some special flags".... :/ | 20:53 |
| sdake | i thnk with clusterer in master it is automatic | 20:53 |
| kfox1111 | not a good answer for production. we've seen ups's fail, etc. | 20:54 |
| prometheanfire | anyone going to the summit avail for https://etherpad.openstack.org/p/TYO-ops-infrastructure-containers (wed 11:15-11:55) | 20:54 |
| prometheanfire | kolla was asked about and I don't know anything about it | 20:54 |
| kfox1111 | I've seen a tape library sprayed with water due to a failed chilled water door rack... | 20:54 |
| sdake | kfox1111 i hear you i'll sycn up with sam later and see if we can't document or implementa a solution for that problem | 20:54 |
| kfox1111 | stuff happens. :) | 20:54 |
| kfox1111 | awesome. thanks. | 20:54 |
| sdake | but seems doable | 20:55 |
| sdake | as insomething that wouldn't effect an eval | 20:55 |
| sdake | as in we will make it happen | 20:55 |
| kfox1111 | yeah. totally think its doable. digging through the containers, it was not obvious how to do it. | 20:55 |
| kfox1111 | just need an answer before I can sell it to management, cause they are going to ask too. :) | 20:55 |
| sdake | ya if we didn't fix that, it would be a fatal flaw i suppose | 20:56 |
| sdake | I dont permit fatal flaws in software i develop ;-) | 20:56 |
| sdake | how is the rest of your evailg oing | 20:56 |
| kfox1111 | perfect. thanks. :) | 20:56 |
| sdake | how many nodes do you plan to deploy | 20:56 |
| kfox1111 | pretty good. | 20:56 |
| sdake | all the config bits you need present? | 20:56 |
| kfox1111 | with this one, about a hundred. | 20:56 |
| sdake | sweet well i'll be riding sam to make it happen | 20:57 |
| sdake | i really want a 100 node proof point | 20:57 |
| kfox1111 | a couple of files are missing, but looks exceedinglyl easy to patch the templates to add. | 20:57 |
| sdake | and if he wont i will | 20:57 |
| kfox1111 | the new unopinionated stuff really looks nice. big thanks for that. | 20:57 |
| sdake | the best place to patch in is ansible i think | 20:57 |
| sdake | in the json blobs we pass in | 20:57 |
| sdake | which tell which files to copy | 20:57 |
| kfox1111 | yeah. those. | 20:58 |
| sdake | what files specifcally are missing? | 20:58 |
| kfox1111 | keystone domains have a /etc/keystone/domains.d directory I need to drop files into. | 20:58 |
| sdake | ya we have a gap atm around dropping files directly into the deployment easily | 20:58 |
| sdake | we need to address that | 20:58 |
| kfox1111 | also, need to tweak /etc/neutron/plugins/openvswitch/ovs_neutron_plugin.ini so I can attach multiple networks. | 20:58 |
| sdake | for example it is not possile to replace or modify the json blobs | 20:59 |
| sdake | ike policy.json | 20:59 |
| kfox1111 | hmmm.. yeah. that would be a problem. We're not tweaking them currently, but I know there have been requests at times. | 20:59 |
| sdake | i would think ovs_neutron_plugin.ini is configurable now aleady | 20:59 |
| sdake | but i coudl be wrong | 20:59 |
| kfox1111 | I couldn't find any code for it. | 20:59 |
| kfox1111 | If I have to patch it in, I'd make sure to contribute it back. :) | 21:00 |
| kfox1111 | um, what else... | 21:00 |
| sdake | your right the ovs config file isn't configurable | 21:00 |
| * sdake grunts | 21:00 | |
| kfox1111 | I've heard bad things still about running haproxy in front of rabbit. | 21:00 |
| sdake | well hard to be perfect ouf othe gate | 21:00 |
| kfox1111 | I was thinking I'd just configure the rabbit lines in each ofthe configs manually for now. | 21:01 |
| sdake | yes i'm not conficed that is the right approach but sam seems keen on it | 21:01 |
| kfox1111 | looks flexable enough already o allow that. | 21:01 |
| kfox1111 | I think it may work now that oslo messaging suppoorts heartbeating, but Its very untested I think. :/ | 21:01 |
| sdake | things seem relaible for me at 3 nodes ;-) | 21:02 |
| kfox1111 | I don't want to be that particular guinny pig on this cloud. :) | 21:02 |
| sdake | not sur eabout 100 | 21:02 |
| kfox1111 | I've got to figure out what to do with cinder, since the cinder team claims it still doesn't work ha for volumes. | 21:02 |
| sdake | the matrix is a coputer generated dreamworld | 21:02 |
| sdake | to keep us under control | 21:02 |
| sdake | to change a human being into this | 21:02 |
| kfox1111 | :) | 21:03 |
| sdake | cinder is backed by ceph | 21:03 |
| kfox1111 | yeah. at our site too. but the cinder-volume engine isn't locking safe, so "there can be only one" | 21:03 |
| kfox1111 | :/ | 21:03 |
| sdake | bummer | 21:03 |
| sdake | that is bad design imo ;) | 21:03 |
| sdake | i think there are some other services that don't do nway active ha as well | 21:04 |
| kfox1111 | its on their to fix list, but thats wrapped up in that whole " we need a distributed lock manager" spec thing | 21:04 |
| kfox1111 | " | 21:04 |
| sdake | larsks did you hear robyn will be back at red hat ! | 21:04 |
| larsks | sdake: yes, it was hard to miss the news :) | 21:04 |
| kfox1111 | oh. has anyone tried dvr with kolla? | 21:04 |
| sdake | kfox1111 i h ave not | 21:04 |
| kfox1111 | It looks pretty streight forward, but there may be some missing packages? | 21:04 |
| sdake | sam was talking about it | 21:05 |
| sdake | but i'm not sure if its implemented or not | 21:05 |
| sdake | all these details, hard forme to know the answers | 21:05 |
| dimsum__ | n00b question - vagrant up is giving me an error "Connection activation failed" for command "/sbin/ifup eth2" - any tips? | 21:05 |
| kfox1111 | sure. just going through the list trying to figure all this out. :) | 21:05 |
| sdake | dimsum__ your go to guys there are mandre or pbourke | 21:06 |
| kfox1111 | considering most of it wasn't supported 6 months ago, kolla's made enormous strides! :) | 21:06 |
| dimsum__ | thanks sdake | 21:06 |
| sdake | they come in in about 8-12 hours i think | 21:06 |
| sdake | but not on weekends typically sorry | 21:06 |
| sdake | they are in europe/japan | 21:06 |
| sdake | dimsum__ if you are evaling a single node, i'd just deploy on your bare metal | 21:07 |
| sdake | this is what i do - makes my life alot simler ;) | 21:07 |
| sdake | kfox1111 not e if you use the ha queue sstuff instead of haproxy in front of rabbit, you wont have the ability to hot-add compute nodes | 21:08 |
| dimsum__ | sdake: ah ok makes sense | 21:08 |
| sdake | you may be able to get it to work but it is undocumented and untested | 21:08 |
| dimsum__ | sdake: steps here? https://github.com/openstack/kolla/blob/master/doc/dev-quickstart.rst | 21:09 |
| kfox1111 | sdake: if you go with out haproxy in front of rabbit, I think hot adding compute still works, just not hot adding rabbits. | 21:09 |
| kfox1111 | not having more then 3 rabbits is ok with me. | 21:09 |
| sdake | kfox ya | 21:10 |
| sdake | dimsum__ yup | 21:10 |
| sdake | kfox we have reailly come a long way since kilo huh ;) | 21:10 |
| * kfox1111 nods | 21:10 | |
| *** achanda has joined #kolla | 21:10 | |
| sdake | up to kilo was about building a community | 21:11 |
| sdake | liberty was abuilt building software | 21:11 |
| kfox1111 | that reorg spec was a big deal, but really helped kolla grow. | 21:11 |
| kfox1111 | yeah. | 21:11 |
| sdake | started out with 5 -1's from cor ereviewers ;) | 21:11 |
| kfox1111 | yeah. it was quite the fight. | 21:11 |
| sdake | ended wit h7 +2 | 21:11 |
| sdake | there were over 300 comments on that blueprint | 21:12 |
| kfox1111 | I think if it wasn't for the spec though, I couldn't be considering it today. | 21:12 |
| kfox1111 | I've been kind of negative on using gerrit for architecturey things. leads to way too much bikeshedding. | 21:12 |
| kfox1111 | but, I think this is a case where it really paid off. | 21:12 |
| sdake | yup but you wont see many specs on kolla | 21:13 |
| sdake | only the big battles ;) | 21:13 |
| kfox1111 | +1 :) | 21:13 |
| kfox1111 | hmm... another thought. | 21:14 |
| kfox1111 | you can use an external ceph with kolla? | 21:14 |
| sdake | i think you could make it work but i dont know how | 21:14 |
| kfox1111 | (just override the settings in /etc/kolla/*?) | 21:14 |
| kfox1111 | k. | 21:14 |
| sdake | you should be able to turn off or on any service in kolla including things like the database | 21:15 |
| kfox1111 | we've got an existing one I just want to point to. | 21:15 |
| sdake | and the rest wlil work | 21:15 |
| sdake | if you config is properly | 21:15 |
| kfox1111 | kk. | 21:15 |
| dimsum__ | kfox1111: sdake: i wanted to checkout the rabbitmq deployment stuff if i got this working :) | 21:15 |
| kfox1111 | cool. | 21:16 |
| sdake | kfox what is that ml2 fie clled again - was going to file a bug | 21:16 |
| sdake | alll ini files should be overridable with augmentation files | 21:17 |
| kfox1111 | /etc/neutron/plugins/openvswitch/ovs_neutron_plugin.ini | 21:24 |
| kfox1111 | sdake: another feature request. a config option to use something other then /etc/kolla. | 21:30 |
| kfox1111 | I want to split our cloud into a shared keystone cluster, and a region. | 21:30 |
| kfox1111 | so they would be seperate kolla configs in the end. | 21:30 |
| *** achanda has quit IRC | 21:30 | |
| kfox1111 | I can work around it, but it would be nicer long term to not have to. | 21:31 |
| *** daneyon_ has quit IRC | 21:32 | |
| *** daneyon has joined #kolla | 21:32 | |
| *** achanda has joined #kolla | 21:33 | |
| *** sdake has quit IRC | 21:34 | |
| *** dims_ has joined #kolla | 21:38 | |
| *** jasonsb has quit IRC | 21:38 | |
| *** dimsum__ has quit IRC | 21:40 | |
| *** achanda has quit IRC | 21:44 | |
| *** dims_ has quit IRC | 21:45 | |
| *** achanda has joined #kolla | 21:51 | |
| *** sdake has joined #kolla | 21:57 | |
| sdake | kfox1111 please file a bug for the multi config | 21:58 |
| sdake | my take initially is i don't see the value proposition | 21:58 |
| sdake | but that is what the bug tracker is for | 21:58 |
| kfox1111 | k. yeah, the value proposition is multiregion with one mangement node. | 22:00 |
| *** tpot has joined #kolla | 22:04 | |
| *** bkosick has joined #kolla | 22:08 | |
| *** dimsum__ has joined #kolla | 22:08 | |
| *** sdake has quit IRC | 22:16 | |
| *** dimsum__ has quit IRC | 22:16 | |
| *** dimsum__ has joined #kolla | 22:17 | |
| *** cemason has joined #kolla | 22:43 | |
| *** daneyon_ has joined #kolla | 22:45 | |
| *** daneyon has quit IRC | 22:46 | |
| *** tpot has quit IRC | 22:48 | |
| *** dimsum__ has quit IRC | 22:52 | |
| *** bkosick has quit IRC | 23:00 | |
| *** Hui has quit IRC | 23:05 | |
| *** dimsum__ has joined #kolla | 23:08 | |
| *** loth has quit IRC | 23:21 | |
| *** gfidente has quit IRC | 23:26 | |
| *** achanda has quit IRC | 23:27 | |
| *** tpot has joined #kolla | 23:33 | |
| *** mfalatic has quit IRC | 23:38 | |
| *** achanda has joined #kolla | 23:51 | |
Generated by irclog2html.py 2.14.0 by Marius Gedminas - find it at mg.pov.lt!