| *** pshige has quit IRC | 01:18 | |
| *** dims_ has joined #kolla | 01:18 | |
| *** dims has quit IRC | 01:21 | |
| *** tobe has joined #kolla | 01:24 | |
| *** erkules has joined #kolla | 01:28 | |
| *** erkules_ has quit IRC | 01:30 | |
| *** jasonsb has quit IRC | 02:08 | |
| *** jasonsb has joined #kolla | 02:08 | |
| *** jasonsb_ has joined #kolla | 02:09 | |
| *** jasonsb_ has quit IRC | 02:12 | |
| *** rhallisey|PTO has quit IRC | 02:13 | |
| *** jasonsb has quit IRC | 02:13 | |
| *** rhallisey|PTO has joined #kolla | 02:23 | |
| *** juggler has quit IRC | 02:33 | |
| *** dims_ has quit IRC | 02:33 | |
| *** juggler has joined #kolla | 02:33 | |
| *** tobe has quit IRC | 02:46 | |
| *** tobe has joined #kolla | 02:47 | |
| *** tobe has quit IRC | 02:50 | |
| *** tobe has joined #kolla | 02:51 | |
| *** jasonsb has joined #kolla | 03:13 | |
| *** dims has joined #kolla | 03:33 | |
| *** tobe has quit IRC | 04:04 | |
| *** ccrouch has quit IRC | 04:11 | |
| *** tobe has joined #kolla | 04:13 | |
| *** dolpher has joined #kolla | 04:23 | |
| *** dolpher has quit IRC | 04:40 | |
| *** tobe has quit IRC | 04:56 | |
| *** sdake has joined #kolla | 04:57 | |
| sdake | evening folks | 04:57 |
|---|---|---|
| vinkman | Evening… | 04:58 |
| vinkman | mine is pretty much almost done :( | 04:58 |
| sdake | hey vinkman whats your tz | 04:58 |
| vinkman | PDT | 04:58 |
| sdake | up late ;) | 04:58 |
| sdake | trying to learn everyones timezones since we have quite a variety | 04:59 |
| sdake | i am gmt-7 year round | 04:59 |
| vinkman | heh, yeah, there seem to be quite a variety | 04:59 |
| vinkman | Isn't that PDT? | 05:00 |
| vinkman | O, no, wait that is MST | 05:00 |
| *** tobe has joined #kolla | 05:01 | |
| sdake | its pdt atm | 05:01 |
| sdake | it changes to mst | 05:01 |
| sdake | in arizona there is no DST | 05:01 |
| vinkman | Ahh.. Arizona :D | 05:01 |
| sdake | or pst/mdt | 05:01 |
| sdake | I really have no idea | 05:01 |
| vinkman | the only state where database have no holes :D | 05:01 |
| sdake | gmt is how I communicate it :) | 05:01 |
| sdake | or anything that has a timer built in :) | 05:02 |
| sdake | it was a big job to add monotonic clock support to corosync to fix clusters going down every two years | 05:02 |
| sdake | field never reported the problem | 05:02 |
| vinkman | lol… sounds about right… | 05:03 |
| sdake | when we fixed it, atleast 30 customers contacted me personally and said "that explains why our clusters went down twice a year" | 05:03 |
| sdake | I was like "groan" | 05:03 |
| sdake | how hard of a pattern is that to spot for the field ? :) | 05:03 |
| vinkman | lol… well, you know it only happened twice a year… | 05:03 |
| sdake | "my cluster busted on dst switchvoer" | 05:04 |
| sdake | first quetion out of anyone's mouth when debugging a problem technical or not | 05:04 |
| sdake | "what changed" | 05:04 |
| sdake | uhh dst changed ;) | 05:05 |
| vinkman | heh… Well, in my experience, nothing, I have never have a customer ever told me anything ever changed ever… | 05:05 |
| vinkman | it just happened :D | 05:05 |
| vinkman | especially if it happens at 2am/3am the day the time change that should really be big red flag… | 05:08 |
| *** vinkman has quit IRC | 05:10 | |
| *** vinkman has joined #kolla | 05:13 | |
| sdake | it just hapepned | 05:29 |
| sdake | my dad says that shit alot | 05:29 |
| sdake | he usually cracks after 45-1:30 hours of grilling and remembers "oh I did do this." | 05:29 |
| sdake | just gotta grill harder ;) | 05:30 |
| sdake | for folks unfamiliar with interactive rebasing, please watch | 05:30 |
| sdake | https://cisco.webex.com/ciscosales/lsr.php?RCID=a4d55cc120ed4b1390a496c1197717d1 | 05:30 |
| bmace | link busted? service unavailable. | 05:32 |
| vinkman | Yeah, not workee for me either... | 05:38 |
| *** nihilifer has joined #kolla | 05:38 | |
| sdake | maybe doesn't work externally | 05:42 |
| sdake | https://cisco.webex.com/cisco/ldr.php?RCID=94ff9ae15df13273affcce2f1c600f7f | 05:44 |
| sdake | can someone try that? | 05:44 |
| bmace | nope | 05:48 |
| sdake | what kind of error | 05:49 |
| bmace | same, service unavailable | 05:49 |
| sdake | https://cisco.webex.com/cisco/ldr.php?RCID=94ff9ae15df13273affcce2f1c600f7f | 05:51 |
| sdake | same link? | 05:51 |
| sdake | wierd must ony work inside firewall | 05:51 |
| mandre | works for me | 05:51 |
| sdake | thats a bunch of fail | 05:51 |
| mandre | i have webex installed, it opened the link in "network recording player" | 05:52 |
| bmace | hrm, i'm trying to open it in chrome under ubuntu 15.04.. no webex junk installed. | 05:53 |
| sdake | oh ya ilnux and webex together are aball of fail | 05:53 |
| bmace | yeah, most conf tools... so obnoxious. | 05:54 |
| bmace | i know vinkman is on a mac though.. you in windoze mandre? | 05:54 |
| sdake | i would apologize for the fail there, but I don't apologize for others actions :) | 05:54 |
| mandre | nope, I'm on a mac | 05:54 |
| sdake | ya macc webex works | 05:55 |
| vinkman | I did it OSX 10.10 | 05:55 |
| sdake | thats what I roll on some version of macos | 05:55 |
| vinkman | and it gave an error, something like page not found or something | 05:55 |
| sdake | whatever version cisco deems i should have | 05:55 |
| sdake | try the second link vinkman | 05:55 |
| vinkman | kk will do | 05:56 |
| vinkman | k second link works like a charm | 06:00 |
| sdake | cool | 06:01 |
| sdake | its not perfect but its unedited ;) | 06:01 |
| sdake | enjoy :) | 06:01 |
| sdake | if you know how to use git rebase -i probably dont want to waste your time | 06:01 |
| *** dasm|afk is now known as dasm | 06:03 | |
| *** inc0 has joined #kolla | 06:04 | |
| *** inc0 has quit IRC | 06:04 | |
| *** vinkman has quit IRC | 06:04 | |
| *** inc0 has joined #kolla | 06:06 | |
| inc0 | good morning | 06:06 |
| inc0 | SamYaple, did you get to any agreement yesterday? | 06:06 |
| mandre | next thing you'll have to teach is "git add -p" and "git reset -p" | 06:06 |
| *** sdake has quit IRC | 06:08 | |
| *** zhiwei has joined #kolla | 06:23 | |
| openstackgerrit | Michal Rostecki proposed stackforge/kolla: [WIP] Add designate-sink service https://review.openstack.org/189393 | 06:25 |
| *** absubram has quit IRC | 06:28 | |
| *** absubram has joined #kolla | 06:39 | |
| *** jmccarthy has joined #kolla | 06:39 | |
| harmw | michal, I'm missing changes in start.sh :) | 06:43 |
| nihilifer | you mean the comments from patchset 1? | 06:46 |
| inc0 | btw...there are 3 guys from Poland in Kolla | 07:00 |
| inc0 | all are named Michał | 07:00 |
| nihilifer | :D | 07:00 |
| nihilifer | I see also Dariusz here | 07:01 |
| nihilifer | so 4 guys from Poland | 07:01 |
| inc0 | Darek is more in tripleo | 07:01 |
| openstackgerrit | Michal Rostecki proposed stackforge/kolla: [WIP] Add designate-sink service https://review.openstack.org/189393 | 07:05 |
| *** shardy has joined #kolla | 07:14 | |
| *** juggler has quit IRC | 07:17 | |
| *** bradjones has quit IRC | 07:27 | |
| *** bradjones has joined #kolla | 07:29 | |
| *** bradjones has joined #kolla | 07:29 | |
| *** juggler_ has joined #kolla | 07:29 | |
| openstackgerrit | Michal Rostecki proposed stackforge/kolla: [WIP] Add designate-sink service https://review.openstack.org/189393 | 07:30 |
| *** alisonholloway has quit IRC | 07:42 | |
| *** alisonholloway has joined #kolla | 08:01 | |
| *** mstachow has joined #kolla | 08:07 | |
| mstachow | o/ | 08:14 |
| mstachow | sdake can you assign me into rabbitmq-ha ? | 08:14 |
| *** athomas has joined #kolla | 08:24 | |
| inc0 | Steven's can be deep asleep now;) | 08:26 |
| mstachow | sadly ! | 08:26 |
| mstachow | :D | 08:26 |
| openstackgerrit | Paul Bourke proposed stackforge/kolla: Minor doc fixups in dev-quickstart.md https://review.openstack.org/190543 | 08:34 |
| openstackgerrit | Paul Bourke proposed stackforge/kolla: Move ansible/README.md under docs/ https://review.openstack.org/190544 | 08:34 |
| openstackgerrit | Paul Bourke proposed stackforge/kolla: Minor doc fixups in ansible-deployment.md https://review.openstack.org/190546 | 08:38 |
| *** fangfenghua has joined #kolla | 08:46 | |
| *** shardy_ has joined #kolla | 08:54 | |
| *** shardy has quit IRC | 08:54 | |
| *** shardy_ has quit IRC | 08:59 | |
| *** shardy has joined #kolla | 08:59 | |
| *** shardy has quit IRC | 09:00 | |
| *** shardy has joined #kolla | 09:00 | |
| SamYaple | inc0: yes, someewhere | 09:30 |
| *** alisonholloway has quit IRC | 09:32 | |
| inc0 | SamYaple, please, do tell what did you came out with | 09:39 |
| *** pdb has quit IRC | 09:40 | |
| SamYaple | basically the config management will be very flexible. The default behaviour will be as described in the spec to maintain immutability. There will also be an optional mode which will be a bind mount | 09:40 |
| *** pdb has joined #kolla | 09:40 | |
| SamYaple | that bind mounted config can be semi managed by ansible, or ansible can not manage it at all | 09:41 |
| SamYaple | semi-managed would be ansible lays down the config, user can modify it and restart the container, when ansible runs again the config gets stomped | 09:41 |
| inc0 | yeah, I thought bind mount would make ansible semi-happy as well | 09:42 |
| inc0 | and we'll have env var which will tell either to make your own config, or just bind mount and that's it right? | 09:42 |
| SamYaple | i prefer the bind method, and the user managed config is a toggle option so it could be switched on and off between playbook runs with no ill effects | 09:42 |
| SamYaple | inc0: yea, probably. i dont think its a hard requirement | 09:43 |
| *** dims_ has joined #kolla | 09:44 | |
| inc0 | do we have bp for bindmount configs? | 09:44 |
| inc0 | I guess we'll need a bp for that and trigger | 09:45 |
| inc0 | also, do we keep crudini? | 09:45 |
| SamYaple | yes, that is always staying until downstream doesnt want to consume/maintain it | 09:45 |
| inc0 | so we'll keep crudini, bindmount...and what ansible will use? | 09:46 |
| *** jmccarthy has quit IRC | 09:46 | |
| SamYaple | as far as BP, im not sure since the bind and env methods will be very intertwined | 09:46 |
| SamYaple | there will be no duplicate code | 09:46 |
| *** jmccarthy has joined #kolla | 09:46 | |
| inc0 | explain please "env method" | 09:46 |
| inc0 | what do you mean by that? | 09:46 |
| SamYaple | the method described in the spec | 09:46 |
| SamYaple | we can up with a few names, bind, env, and crudini | 09:47 |
| *** dims has quit IRC | 09:47 | |
| inc0 | eh, env is filling an template made by ansible with env vars right? | 09:47 |
| SamYaple | yes | 09:48 |
| SamYaple | kinda, but mostly yes | 09:48 |
| inc0 | so there are 2 options really, internal config and external config | 09:48 |
| inc0 | internal is crudini and ansible+env | 09:49 |
| inc0 | external is manual and ansible+env | 09:49 |
| inc0 | so it really boils down to if container mount hosts /etc/ or not | 09:49 |
| inc0 | and then if container runs crudini or ansible | 09:49 |
| SamYaple | no | 09:50 |
| SamYaple | ansible will not be using crudini ever | 09:50 |
| inc0 | I know, there are 2 choices in Dockerfile | 09:50 |
| SamYaple | https://review.openstack.org/#/c/182168/ | 09:50 |
| inc0 | that's what I mean | 09:50 |
| SamYaple | yea i suppose thats fair, but it will probably change slightly | 09:51 |
| inc0 | mount/don't mount, if mount -> stop script, if don't mount -> run either ansible.sh or crudini.sh | 09:51 |
| inc0 | and if don't mount, we'll leave user to run ansible manually, or provide custom config alltogether | 09:52 |
| SamYaple | to avoid duplicate code the bind mount will exist in every container created by ansible, it just may not be used | 09:52 |
| inc0 | but bindmount is just one line in Dockerfile right? | 09:52 |
| SamYaple | mountpoints dont need to be specified in the dockerfile | 09:53 |
| inc0 | so we'll mount stuff from within container? | 09:53 |
| inc0 | can we do that? that seems even better in terms of code maintenance | 09:54 |
| SamYaple | what? | 09:54 |
| inc0 | can we mount /etc/ from host after container is deployed? | 09:54 |
| inc0 | and from within container? | 09:54 |
| SamYaple | no | 09:55 |
| inc0 | then how "mountpoints don't need to be specified in the dockerfile"? | 09:55 |
| SamYaple | they are specified at runtime, when specified in the docker file they are only used in "--volumes-from" | 09:56 |
| inc0 | hmm...I'm closing in to an idea:) | 09:56 |
| inc0 | let me write it down in spec, and I'd be happy if you'd review it | 09:57 |
| SamYaple | i will indeed | 09:58 |
| *** zhiwei has quit IRC | 10:09 | |
| *** athomas has quit IRC | 10:23 | |
| *** athomas has joined #kolla | 10:23 | |
| inc0 | SamYaple, https://review.openstack.org/#/c/189157/ go for it ;) | 10:33 |
| *** alisonholloway has joined #kolla | 10:35 | |
| SamYaple | wer arent bindmounting in anythign from /etc | 10:36 |
| *** vbel has quit IRC | 10:37 | |
| *** vbel has joined #kolla | 10:38 | |
| inc0 | path is there just for proving the concept, don't pay too much attention to it | 10:38 |
| *** shardy_ has joined #kolla | 10:39 | |
| SamYaple | this bind/env talk is really only for ansible/other automation tools. and really it only describes _how_ we create the container | 10:39 |
| SamYaple | to avoid duplicatnig code, the bindmount will happen each time, but the config will only be used if you are using the bind method | 10:40 |
| *** shardy has quit IRC | 10:41 | |
| inc0 | this solution doesn't duplicate any code | 10:41 |
| inc0 | and avoid having several configs at the same time | 10:41 |
| SamYaple | mount vs not mount | 10:41 |
| inc0 | this is single if statement | 10:42 |
| SamYaple | you cant do if statements like that in ansible, or most languagues | 10:43 |
| SamYaple | you would be duplicating a block of docker | 10:43 |
| inc0 | it won't be in ansible, it would be in docker runtime | 10:43 |
| inc0 | as you've said yourself | 10:43 |
| inc0 | --volume-from or not --volume-from | 10:43 |
| SamYaple | you cant do stuff like that inside the container | 10:44 |
| SamYaple | thats all outside the cotnaienr before its setup | 10:44 |
| inc0 | that's why this one conditional will not be inside container | 10:44 |
| inc0 | there are 2 steps here | 10:44 |
| SamYaple | what youre proposing is a script that actually does the docker launching? | 10:44 |
| inc0 | yup, unless you can put conditional to Dockerfile | 10:45 |
| SamYaple | where as currently ansible interacts with the docker-py api directly? | 10:45 |
| *** shardy_ has quit IRC | 10:45 | |
| *** shardy has joined #kolla | 10:45 | |
| inc0 | at worst we can have 2 Dockerfiles for bind and no binds | 10:46 |
| SamYaple | no | 10:46 |
| inc0 | but that wouldn't be good | 10:46 |
| inc0 | hmm..so you're saying there is no way to use conditional like that in ansible: "if env==bind then use --volume=from else don't use volume-from"? | 10:50 |
| SamYaple | not like that no | 10:50 |
| inc0 | so you will mount it all the time? | 10:51 |
| SamYaple | not to /etc/keytone/keystone.conf, not | 10:51 |
| SamYaple | it will be to /opt/kolla/keystone/keystone.conf | 10:51 |
| inc0 | allright, but it will be mounted | 10:51 |
| SamYaple | i see no downside to it | 10:52 |
| inc0 | and depending on env, it will use it or not | 10:52 |
| SamYaple | yea | 10:52 |
| inc0 | well, one downside would be to specify service runtime to use custom file path | 10:52 |
| SamYaple | what are you talking about | 10:52 |
| inc0 | if you want actually to use this file | 10:53 |
| SamYaple | no | 10:53 |
| SamYaple | it would be cp from /opt/kolla/keystone/keystone.conf /etc/keystone/keystone.conf | 10:53 |
| SamYaple | for permission reasons | 10:53 |
| inc0 | uhh, I have issues with that | 10:53 |
| SamYaple | why? | 10:54 |
| inc0 | because change of file won't change config run inside container | 10:54 |
| SamYaple | it most certainly will | 10:54 |
| inc0 | unless we'll do symlink rather than cp | 10:54 |
| SamYaple | on contaienr restart | 10:54 |
| inc0 | let's just use ln -s and that's all | 10:55 |
| SamYaple | no | 10:55 |
| SamYaple | thats still got permission issues | 10:55 |
| *** athomas has quit IRC | 10:55 | |
| SamYaple | cp will overwrite the file in /etc/keystone on cotnainer restart | 10:55 |
| inc0 | I'm still thinking about restartless reread | 10:55 |
| SamYaple | this also prevents the file from changing underneatch a service while its running | 10:55 |
| SamYaple | openstack doesnt do that | 10:55 |
| inc0 | we'll work on that in several projects | 10:56 |
| inc0 | restarts are painful in neutron or heat | 10:56 |
| inc0 | really painful | 10:56 |
| inc0 | and I'd hate to have this doors closed | 10:56 |
| SamYaple | but a reread of a conf will not be in liberty. i doubt it will be in M or even N | 10:56 |
| SamYaple | there are many reasons not to do it | 10:56 |
| inc0 | there aren't any reasons why not make this possible tho | 10:57 |
| SamYaple | no ones closing a door, but we arent going to do something that is bad practice for a potential future feature | 10:57 |
| SamYaple | if that feature rolls around then we can revisit the issue | 10:57 |
| SamYaple | besides its not like the functionality is removed. you could always issue a `docker exec cp /opt/keystone/keystone.conf /etc/keystone/keystone.conf` | 10:58 |
| inc0 | well I guess, or use sync | 10:58 |
| SamYaple | but im not bindmounting directly to /etc/keystone/keystone.conf for permission reasons | 10:59 |
| inc0 | well, fair enough | 11:04 |
| SamYaple | if you want a reason why its bad, look at yaodu | 11:04 |
| inc0 | no, I know why its bad | 11:04 |
| SamYaple | i had to create a user on the host and change perms in the container and static UID and GIDs | 11:05 |
| inc0 | and I guess there are ways to achieve both goals | 11:05 |
| SamYaple | Personally I just don't see how this limits anything | 11:05 |
| inc0 | I guess it doesn't really, I was just panicking without reason, I do that sometimes | 11:06 |
| SamYaple | its cool. i want this project to be as flexible as possible while remaining simple and small for maintanability | 11:07 |
| *** mstachow has quit IRC | 11:08 | |
| *** athomas has joined #kolla | 11:11 | |
| *** openstackgerrit has quit IRC | 11:13 | |
| *** openstackgerrit has joined #kolla | 11:14 | |
| *** dwalsh has joined #kolla | 11:44 | |
| *** sdake has joined #kolla | 12:33 | |
| *** shardy_ has joined #kolla | 12:34 | |
| *** shardy has quit IRC | 12:35 | |
| sdake | morning | 12:37 |
| *** absubram has quit IRC | 12:38 | |
| *** fangfenghua has quit IRC | 12:38 | |
| *** shardy_ has quit IRC | 12:39 | |
| *** shardy has joined #kolla | 12:40 | |
| inc0 | morning sdake | 12:43 |
| sdake | morning inc0 | 12:43 |
| inc0 | look at ansible rev plz, I've got an idea;) after a fashon | 12:43 |
| *** nihilifer has quit IRC | 12:43 | |
| sdake | lots o people had ideas in that spec ;0 | 12:44 |
| sdake | ;) that is :) | 12:44 |
| sdake | i'll take a look after i wake up a bit, its 5:44 am | 12:44 |
| inc0 | problem with software devs is that they usually have ideas, and their ideas are usually best possible ideas | 12:44 |
| inc0 | and I'm no exception | 12:45 |
| sdake | 13k arwpa on ruwasy my legs are beat | 12:45 |
| sdake | 13k steps that is | 12:45 |
| sdake | on tuesday | 12:45 |
| inc0 | day off in mountains? | 12:46 |
| dasm | inc0: yeah.... ideas... like standards: https://xkcd.com/927/ | 12:46 |
| sdake | at cisco live | 12:46 |
| dasm | good morning sdake. sorry, didn't have time to look at spec, but a lot had happened there. | 12:46 |
| sdake | i've done a few 20 mile hikes before which is about 45-55k steps | 12:46 |
| sdake | but i was younger dumber and stronger back then :) | 12:47 |
| sdake | dasm good comic :) | 12:47 |
| inc0 | brave new wheel that is | 12:48 |
| sdake | interactive rebase video -> https://cisco.webex.com/cisco/ldr.php?RCID=94ff9ae15df13273affcce2f1c600f7f | 12:48 |
| inc0 | when I was developing heat versioned objects | 12:49 |
| dasm | sdake: i think, there is at least one xkcd strip for every possible developers complaining :) | 12:50 |
| inc0 | I had patch series of about 10 patches which touched literally everything in heat codebase | 12:50 |
| sdake | there is one xkcd let me see if i can find it | 12:50 |
| inc0 | I've learned rebase then. I've learned it in way too intimite way than I'd like to | 12:51 |
| SamYaple | yo | 12:51 |
| sdake | dasm here ya go :) one of my favorites | 12:53 |
| sdake | http://dilbert.com/strip/2010-10-26 | 12:53 |
| dasm | ROTFL | 12:53 |
| sdake | inc0 if you know interactive rebase then that video would be a waste of time | 12:54 |
| SamYaple | if its so simple that everyone understands it and it works how will i keep my job? | 12:54 |
| sdake | dasm funny story, kolla started in that "need it in 6 weeks for a trade show" type of model :) | 12:55 |
| dasm | Oo | 12:55 |
| *** tobe has quit IRC | 12:55 | |
| dasm | didn't know :D but look at this. it's even usable :D | 12:56 |
| dasm | somehow | 12:56 |
| inc0 | SamYaple, https://thc.org/root/phun/unmaintain.html | 12:56 |
| openstackgerrit | Fang Fenghua proposed stackforge/kolla: WIP Add openvswitch contianer https://review.openstack.org/181721 | 12:57 |
| *** tobe has joined #kolla | 12:57 | |
| dasm | inc0: that's why i know, why i cannot maintain your code. you're following these rules. | 12:58 |
| sdake | You don't want to overdo this. Your code should not look hopelessly unmaintainable, just be that way. Otherwise it stands the risk of being rewritten or refactored. | 12:58 |
| inc0 | dasm, that's why you're one maintaining and I'm one writing:P | 12:58 |
| dasm | xD | 12:58 |
| sdake | my advice for job security is be the best in your field so nobody would ever desire to replace you | 12:58 |
| SamYaple | got that covered | 12:59 |
| sdake | http://www.amazon.com/Linchpin-Are-Indispensable-Seth-Godin/dp/1591844096 | 12:59 |
| dasm | sdake: in that way, we should focus on creating small fields in different areas. and then no one will be replaceable, because no one else will know about it ;) | 12:59 |
| sdake | i dont worry about job security | 13:00 |
| sdake | my job security is my skill set ;) | 13:00 |
| inc0 | right now knowing anything about openstack is pretty much job security itself | 13:00 |
| SamYaple | inc0: aint that the truth | 13:01 |
| sdake | ya I heard openstack devs described as "special unicorns" | 13:01 |
| inc0 | and the hype grows | 13:02 |
| inc0 | around openstack, or docker in that instance | 13:02 |
| *** tobe has quit IRC | 13:02 | |
| SamYaple | it is indeed hype | 13:02 |
| SamYaple | have you see Openstack cascading? | 13:02 |
| SamYaple | oh my | 13:02 |
| inc0 | cascading? | 13:03 |
| SamYaple | https://wiki.openstack.org/wiki/OpenStack_cascading_solution | 13:03 |
| SamYaple | "add Nova as the hypervisor backend of Nova, Cinder as the block storage backend of Cinder, Neutron as the backend of Neutron, Glance as one image location of Glance, Ceilometer as the store of Ceilometer." | 13:03 |
| inc0 | I've seen openstack fall in many different, creative, ways if that's what you mean;) | 13:03 |
| SamYaple | we must go deeper | 13:03 |
| inc0 | fifthipleO? | 13:04 |
| inc0 | sixtipleO? | 13:04 |
| inc0 | my English ends here.. | 13:04 |
| SamYaple | it is glorious | 13:04 |
| SamYaple | https://wiki.openstack.org/w/images/b/bb/Cascading02.png | 13:04 |
| SamYaple | that should tell you all you need to know | 13:04 |
| dasm | SamYaple: it's so evil. | 13:05 |
| dasm | if something doesn't work... deal with it. | 13:05 |
| SamYaple | dasm: you know you want it | 13:05 |
| dasm | yeah... i've even thought to buy some servers... and run it at home. | 13:05 |
| SamYaple | i, for one, thought the 4 proxies that existed in the openstack realm werent enough. I want to proxy everything, including my proxies | 13:05 |
| dasm | ofc with cascading openstack on it | 13:05 |
| inc0 | https://www.youtube.com/watch?v=QxJrjV4PNXA | 13:06 |
| inc0 | and put mongodb out there as db backend | 13:06 |
| dasm | "yo dawg, I herd you like openstack, so I put an openstack in your openstack so you can openstack while you openstack" | 13:07 |
| inc0 | thats tripleo | 13:07 |
| inc0 | and I mean, literally, thats tripleo. | 13:07 |
| dasm | pretty much | 13:07 |
| SamYaple | cascading is openstack on openstackssssss | 13:07 |
| SamYaple | as in multiple | 13:08 |
| inc0 | on heat meetings we get recurring theme called template dsl | 13:08 |
| SamYaple | so all the openstack underservices share resources | 13:08 |
| inc0 | with if, loops and so forth | 13:08 |
| inc0 | so in effect, we'll have turing-complete templating language | 13:08 |
| inc0 | so technically we can write heat using heat | 13:08 |
| *** jruano has joined #kolla | 13:10 | |
| inc0 | guys, do we define cidr for my_ip somewhere? | 13:16 |
| inc0 | or we just assume its /24 netmask? | 13:16 |
| sdake | inc0 that video is slightly disturbing and amusing at the same time :) | 13:23 |
| inc0 | that's the best kind | 13:24 |
| *** jtriley has joined #kolla | 13:28 | |
| *** jtriley has quit IRC | 13:40 | |
| *** ccrouch has joined #kolla | 13:42 | |
| *** jmccarthy has quit IRC | 13:48 | |
| *** jtriley has joined #kolla | 14:01 | |
| *** jmccarthy has joined #kolla | 14:07 | |
| *** bmace has quit IRC | 14:08 | |
| sdake | i am altering the deal. pray I don't alter it further. | 14:09 |
| sdake | darth vadar destiny decidedr ftw | 14:10 |
| pdb | sdake + others, could you have a look at https://review.openstack.org/#/c/190117/ when free? | 14:10 |
| pdb | also, when adding a new base distro. would people prefer all images to be submitted at once or merged in one by one? | 14:16 |
| SamYaple | pdb: i dont think all images at once is realistic | 14:16 |
| SamYaple | and that review owuld be a nightmare | 14:16 |
| sdake | one patch at a time in a patch stream - see interfact rebase | 14:17 |
| SamYaple | i think one by one | 14:17 |
| sdake | its helpful if its a stream so folks can understand the full implication of the changes :) | 14:17 |
| pdb | makes sense | 14:17 |
| SamYaple | even in a patch stream all images merging at once isnt realistic | 14:17 |
| SamYaple | maybe requiring the core images all at one (keystone glance neutron nvoa maraidb etc) | 14:17 |
| pdb | i have a base image for oraclelinux and keystone currently, using source installs | 14:18 |
| sdake | off to a tired start this mroning going to hit the rack for a couple hours | 14:18 |
| sdake | those need separate patches ldb | 14:18 |
| pdb | sure | 14:18 |
| pdb | they sort of depend on the above review going through though. otherwise I end up copying a lot of code | 14:19 |
| SamYaple | pdb: that review looks good to me. i dont htink i can aprove the blueprint though | 14:20 |
| *** fabiand has joined #kolla | 14:20 | |
| pdb | SamYaple: thanks, no worries just looking to get some feedback | 14:21 |
| pdb | in case I base too much on it and turns out we need a different approach | 14:22 |
| SamYaple | pdb: you wont get an argument from me. if it violates DRY, it isnt a good idea | 14:22 |
| pdb | well, the thing I like about it compared to some of the other approaches is it allows for a complete override if needed | 14:23 |
| *** mickt has quit IRC | 14:24 | |
| inc0 | uhh I hate networking stuff | 14:24 |
| pdb | also forgot to say at the meeting but let me know if there are priority stuff that needs a review. I think some was mentinoed but no specific links | 14:25 |
| inc0 | cya guys, I'm off for today | 14:31 |
| SamYaple | o/ | 14:31 |
| *** absubram has joined #kolla | 14:31 | |
| *** inc0 has quit IRC | 14:31 | |
| vbel | pdb: does it work with simlinks? | 14:33 |
| pdb | vbel: does what? | 14:33 |
| vbel | pdb: https://review.openstack.org/#/c/190117/ | 14:33 |
| pdb | yes | 14:34 |
| vbel | usually when you run docker build everything is copied over to docker temp place | 14:34 |
| vbel | so symlinks seem to be ok | 14:35 |
| SamYaple | vbel: the build follows symlinks and copies actual files | 14:38 |
| vbel | SamYaple: that's great :) I wonder if we can create base image from scratch that only has those files and then inherit it in os-specific images | 14:39 |
| vbel | like adding stuff before adding os :) | 14:40 |
| *** dwalsh has quit IRC | 14:42 | |
| pdb | SamYaple: smart | 14:45 |
| pdb | I was double checking this for vbel and was worried to find a quick test image didn't seem to work. but the kolla ones did | 14:46 |
| SamYaple | https://github.com/SamYaple/encrypted-zfs-ubuntu-bootstrap | 14:48 |
| SamYaple | ^ in case anyone else wants LUKS + multiple volume ZFS | 14:48 |
| SamYaple | for root | 14:48 |
| *** juggler_ has quit IRC | 14:51 | |
| *** dims_ has quit IRC | 15:02 | |
| *** dims has joined #kolla | 15:02 | |
| *** blahRus has joined #kolla | 15:08 | |
| *** sdake has quit IRC | 15:21 | |
| *** jtriley has quit IRC | 15:53 | |
| *** bmace has joined #kolla | 15:55 | |
| *** jtriley has joined #kolla | 15:57 | |
| *** vinkman has joined #kolla | 16:10 | |
| pdb | question: in order to build alternate images I currently need to place an empty .buildconf in the root of the repo to prevent the default centos-base- being used. This seems fine, but they're in the gitignore so can't be part of the repo | 16:14 |
| pdb | would this be confusing for users who aren't aware of this mechanism? or should we just update the docs to describe this | 16:14 |
| *** dwalsh has joined #kolla | 16:55 | |
| *** diogogmt has joined #kolla | 16:57 | |
| diogogmt | just finished watching the vancouver os summit video on kolla, awesome work being done here! | 16:58 |
| bmace | diogogmt: agreed. | 17:01 |
| *** Slower has quit IRC | 17:02 | |
| *** Slower has joined #kolla | 17:08 | |
| *** athomas has quit IRC | 17:36 | |
| *** sdake has joined #kolla | 17:46 | |
| sdake | hey guys | 17:47 |
| Slower | howdy | 17:48 |
| *** sdake_ has joined #kolla | 17:54 | |
| *** jasonsb has quit IRC | 17:55 | |
| sdake_ | hey slower | 17:55 |
| sdake_ | if your going to make it to san jose plz sign up for the kolla midcycle | 17:55 |
| *** jasonsb has joined #kolla | 17:55 | |
| bmace | good day sdake_ | 17:55 |
| sdake_ | hey bmace | 17:56 |
| sdake_ | decided to start my day over ;) | 17:56 |
| sdake_ | feeling much better as a result | 17:56 |
| bmace | glad to hear it! you know, as a coder, groundhog day would be super frustrating.. all your changes would get nuked at the end of each day :/ | 17:57 |
| *** sdake has quit IRC | 17:58 | |
| Slower | sdake_: I'd love to come but I doubt they'll send me | 17:58 |
| Slower | I'd just drink the whole time anyway ;-) | 17:58 |
| sdake_ | slower doen't hurt to ask :) | 17:59 |
| sdake_ | whats the worst that could happen | 17:59 |
| Slower | yeah I should | 17:59 |
| sdake_ | "damn that slower guy asked again to travel"? :) | 17:59 |
| Slower | hehe | 17:59 |
| Slower | be good to see you again actually, that was fun in van | 17:59 |
| *** jasonsb has quit IRC | 18:00 | |
| *** jtriley has quit IRC | 18:00 | |
| sdake_ | my cubans arrived that i ordered at that smoke shop | 18:04 |
| sdake_ | enjoying one right now as a matter of fact :) | 18:04 |
| *** jtriley has joined #kolla | 18:04 | |
| Slower | sdake_: nice! | 18:17 |
| *** fabiand has quit IRC | 18:26 | |
| *** juggler has joined #kolla | 18:29 | |
| *** vinkman has quit IRC | 18:32 | |
| *** jasonsb has joined #kolla | 18:44 | |
| *** prad_ has joined #kolla | 18:45 | |
| *** jasonsb_ has joined #kolla | 18:45 | |
| *** jasonsb has quit IRC | 18:46 | |
| *** shardy has quit IRC | 19:01 | |
| *** jtriley has quit IRC | 19:23 | |
| *** jtriley has joined #kolla | 19:26 | |
| openstackgerrit | Merged stackforge/kolla: Change documentation to recommend Docker 1.7.0-dev https://review.openstack.org/186013 | 19:33 |
| *** vinkman has joined #kolla | 19:43 | |
| *** sdake_ has quit IRC | 19:53 | |
| *** sdake has joined #kolla | 19:56 | |
| *** dwalsh has quit IRC | 20:24 | |
| sdake | samyaple i'm back awake if you wanted to finish our discussion :) | 20:35 |
| openstackgerrit | Daneyon Hansen proposed stackforge/kolla: Spec to Add Support for High Availability https://review.openstack.org/181983 | 20:39 |
| -openstackstatus- NOTICE: Gerrit has been restarted to terminate a persistent looping third-party CI bot | 21:06 | |
| *** dasm is now known as dasm|afk | 21:08 | |
| *** jasonsb_ has quit IRC | 21:17 | |
| *** jasonsb has joined #kolla | 21:18 | |
| *** jtriley has quit IRC | 21:19 | |
| *** jasonsb has quit IRC | 21:22 | |
| *** jasonsb has joined #kolla | 21:22 | |
| *** Slower has quit IRC | 21:34 | |
| *** Slower has joined #kolla | 21:41 | |
| *** dims_ has joined #kolla | 21:44 | |
| *** dims has quit IRC | 21:46 | |
| *** diogogmt has quit IRC | 21:59 | |
| *** vinkman has quit IRC | 22:10 | |
| *** absubram has quit IRC | 22:22 | |
| sdake | inviting people to review the ha specification - need more input if you are interested in ha: https://review.openstack.org/#/c/181983/ | 22:35 |
| *** vinkman has joined #kolla | 22:43 | |
| *** jtriley has joined #kolla | 22:46 | |
| *** diogogmt has joined #kolla | 22:52 | |
| *** jtriley has quit IRC | 23:02 | |
| *** jruano has quit IRC | 23:12 | |
| *** ccrouch has quit IRC | 23:22 | |
| *** jruano has joined #kolla | 23:40 | |
| *** blahRus has quit IRC | 23:44 | |
| *** jruano has quit IRC | 23:53 | |
| *** vinkman has quit IRC | 23:58 | |
Generated by irclog2html.py 2.14.0 by Marius Gedminas - find it at mg.pov.lt!
