sdake | wifey bought a macbook new release iwth usbc | 00:09 |
---|---|---|
* sdake groans | 00:09 | |
sdake | i'm not buying a new macbook until something faster then ac comes out or usbc is released in that model with multiple usb ports ;) | 00:10 |
sdake | although I am tempted to get a macbook standard model for travel | 00:10 |
*** achanda has quit IRC | 00:12 | |
*** dims has joined #kolla | 00:42 | |
*** dims has quit IRC | 00:44 | |
*** openstackgerrit has quit IRC | 00:51 | |
*** openstackgerrit has joined #kolla | 00:51 | |
*** dolpher has joined #kolla | 01:06 | |
*** achanda has joined #kolla | 01:12 | |
*** achanda has quit IRC | 01:18 | |
*** erkules has joined #kolla | 01:19 | |
*** erkules_ has quit IRC | 01:21 | |
*** dims has joined #kolla | 01:24 | |
*** dims_ has joined #kolla | 01:28 | |
*** britthouser has quit IRC | 01:30 | |
*** dims has quit IRC | 01:31 | |
*** tobe has joined #kolla | 02:29 | |
*** jasonsb has joined #kolla | 02:35 | |
*** daneyon has joined #kolla | 02:46 | |
*** tobe has quit IRC | 03:01 | |
*** tobe has joined #kolla | 03:03 | |
*** britthouser has joined #kolla | 03:15 | |
*** britthouser has quit IRC | 03:15 | |
*** sdake_ has joined #kolla | 03:17 | |
*** sdake_ has quit IRC | 03:22 | |
*** daneyon has quit IRC | 03:25 | |
SamYaple | sdake: ntp is pretty important, but also not something that is strictly on Kolla's shoulder | 03:31 |
*** jasonsb has quit IRC | 03:31 | |
SamYaple | without ntp the services get out of sync pretty bad | 03:31 |
SamYaple | but lots of companies have internal NTP that we dont want ot override | 03:32 |
sdake | i think that is for an ntp client iirc | 03:34 |
sdake | fang added it | 03:34 |
sdake | not sure on the background | 03:34 |
SamYaple | everything i said applies to server and client | 03:35 |
SamYaple | sdake: i did push the nova bits even though it is 3 commits deep. i guess ill start working on the neutron bits. all this stuff needs to land soon so we can tweak it. i havea feeling everyone will want to modify the structure a good bit before L2 | 03:38 |
sdake | ok well it doesn't seem like #1 priority atm :) | 03:38 |
sdake | (the ntp thing) | 03:38 |
sdake | the nova + neutron is yes | 03:38 |
SamYaple | yea im not even sure the ntp thing is a Kolla issue to solve | 03:38 |
SamYaple | maybe we do an NTP container, maybe we dont. it could go either way in my opinion | 03:39 |
SamYaple | id probably vote no to keep scope creep down | 03:39 |
sdake | well we dont want to put server for sure but client may make sense | 03:39 |
sdake | that came up from harmw asking about it earlier today | 03:40 |
SamYaple | sdake: i disagree. server is needed if we implement it | 03:40 |
sdake | oh ya? | 03:40 |
sdake | shows what I know :) | 03:40 |
sdake | I don't even see why the times shoudl be synched | 03:40 |
SamYaple | client has no gaurantee that you will have a server "locally" (not on a global scale) and ntp over the internet is not close enough for some services to not get out of sync | 03:40 |
sdake | other then having synchronized logs I don't see why it is required technically | 03:40 |
SamYaple | message in the message queue | 03:41 |
SamYaple | services will appear to bounce but its just a time sync issue | 03:41 |
SamYaple | also if we do Ceph that has a 0.05 second warn limit. at 0.5 seconds the cluster is unusable | 03:42 |
sdake | bought a usb c mac today | 03:42 |
sdake | getting it setup atm | 03:42 |
SamYaple | ugh | 03:42 |
sdake | 0.5 pretty aggressive | 03:42 |
SamYaple | 0.05 | 03:42 |
sdake | mainly uisng for travel | 03:42 |
SamYaple | 0.5 is really bad | 03:42 |
SamYaple | yea let me know how you like it | 03:43 |
SamYaple | the USB C single interface mainly | 03:44 |
sdake | keyboard is different then mac pro | 03:44 |
sdake | oh ya i'll lety ou know how annoying that is | 03:44 |
sdake | will definatley be an adjustment | 03:44 |
sdake | although I dont use any other interfaces atm except display port | 03:44 |
SamYaple | hate mac keyboards anyway. I would probably use it more if the keyboard was decent | 03:44 |
SamYaple | have you submitted a talk? | 03:44 |
sdake | i'll do our talk monday morning | 03:45 |
sdake | submission | 03:45 |
sdake | you should get an email note about it i suspect | 03:45 |
SamYaple | yea i gotta fill out the Bio page | 03:45 |
SamYaple | hat ethat stuff | 03:45 |
*** dims_ has quit IRC | 03:48 | |
*** achanda has joined #kolla | 03:48 | |
sdake | mail me your bio as well | 03:49 |
sdake | I may need to enter it separeately | 03:49 |
sdake | I am not certain on this point | 03:49 |
sdake | it used to be the case but I don't recall from last year if it was or not | 03:49 |
SamYaple | im not sure what to put for the bio... | 03:50 |
*** coolsvap|away is now known as coolsvap | 03:50 | |
SamYaple | Speaker's Bio - "Am Human" | 03:50 |
sdake | try to let people know why they would caret o listen to you speak over joe on the street | 03:51 |
sdake | look at other people's bios for inspiration | 03:52 |
sdake | i gotta say one thingabout usbc, no more magsafe = YAY | 03:52 |
sdake | boy magsafe is annoying | 03:52 |
SamYaple | how can i view others bio's? not seeing a big list or anything | 03:53 |
*** diga has joined #kolla | 03:53 | |
*** tobe has quit IRC | 03:55 | |
*** tobe has joined #kolla | 03:56 | |
coolsvap | sdake, hi, can you please do workflow+1 to https://review.openstack.org/#/c/200702/ , the change was in wip when it was done workflow+1 | 04:04 |
sdake | done - not sure if that will fix it | 04:06 |
sdake | if not, slightly modify the commit log and submit a new review | 04:06 |
sdake | or git pull --rebase | 04:06 |
sdake | and submit a rebased vesion | 04:06 |
sdake | version | 04:06 |
*** jasonsb has joined #kolla | 04:07 | |
*** achanda has quit IRC | 04:07 | |
coolsvap | alright, let me see if it fails merge, will update accordingly | 04:07 |
openstackgerrit | Merged stackforge/kolla: Remove requirement for root user in scripts https://review.openstack.org/200702 | 04:08 |
SamYaple | sdake: what is the "speaker bureau"? | 04:23 |
*** jasonsb has quit IRC | 04:26 | |
*** mfalatic has quit IRC | 04:36 | |
*** sdake has quit IRC | 04:36 | |
*** sdake has joined #kolla | 04:37 | |
*** dims has joined #kolla | 04:48 | |
*** dims has quit IRC | 04:55 | |
*** achanda has joined #kolla | 05:01 | |
*** sdake has quit IRC | 05:08 | |
*** nihilifer has joined #kolla | 05:11 | |
*** sdake has joined #kolla | 05:13 | |
sdake | samyaple not sure | 05:14 |
sdake | samyaple maybe past speakers, not sure | 05:15 |
SamYaple | i finished it | 05:15 |
SamYaple | thanks though | 05:15 |
*** sdake_ has joined #kolla | 05:21 | |
*** jasonsb has joined #kolla | 05:34 | |
*** sdake_ has quit IRC | 05:35 | |
*** sdake_ has joined #kolla | 05:35 | |
*** sdake_ has quit IRC | 05:48 | |
*** sdake has quit IRC | 05:56 | |
*** juggler_ has joined #kolla | 06:09 | |
*** juggler has quit IRC | 06:12 | |
openstackgerrit | Swapnil Kulkarni proposed stackforge/kolla: Update docker-py in requirement.txt https://review.openstack.org/200705 | 06:37 |
*** cbaesema_ has joined #kolla | 06:51 | |
*** cbaesema has quit IRC | 06:52 | |
*** cbaesema_ is now known as cbaesema | 06:52 | |
*** achanda has quit IRC | 06:56 | |
openstackgerrit | Swapnil Kulkarni proposed stackforge/kolla: Update docker-py in requirement.txt https://review.openstack.org/200705 | 06:56 |
*** achanda has joined #kolla | 07:03 | |
openstackgerrit | Swapnil Kulkarni proposed stackforge/kolla: Update scripts to check if docker is usable for $USER https://review.openstack.org/201022 | 07:12 |
*** achanda has quit IRC | 07:16 | |
coolsvap | SamYaple, ping | 07:16 |
SamYaple | pong | 07:17 |
coolsvap | reqarding the docker-py version comment you did on review https://review.openstack.org/200705 | 07:17 |
SamYaple | yea | 07:17 |
coolsvap | we need docker-py<1.3,>=1.2.3 | 07:17 |
SamYaple | no, docker-compose does | 07:18 |
coolsvap | yes | 07:18 |
SamYaple | docker-py is only listed because of the requirement around build.py | 07:18 |
SamYaple | that version should stay as it is | 07:18 |
coolsvap | but with docker-py>=1.2.0 | 07:18 |
SamYaple | what you need to do is move docker-py above docker-compose on the list | 07:18 |
coolsvap | we get docker-py 1.3.0 | 07:18 |
coolsvap | which fails with error mentioned in bug https://launchpad.net/bugs/1473856 | 07:19 |
openstack | Launchpad bug 1473856 in kolla "kolla start fails for docker-py=1.2.0" [Undecided,In progress] - Assigned to Swapnil Kulkarni (coolsvap) | 07:19 |
SamYaple | that is because it is processed last. What you need to do is move docker-py above docker-compose in the requirements.txt | 07:19 |
*** akwasnie has joined #kolla | 07:19 | |
coolsvap | i did it | 07:19 |
SamYaple | you do _not_ need ot adjust the versio nnumber | 07:19 |
coolsvap | still the result is same | 07:19 |
SamYaple | remove docker-compose and docker-py | 07:20 |
SamYaple | the result will not be the same | 07:20 |
SamYaple | you did not remove the installs so of course nothing changed | 07:20 |
coolsvap | i did sudo pip uninstall -r requirements.txt | 07:21 |
coolsvap | then update requirements.txt | 07:21 |
coolsvap | then sudo pip install -r requirements.txt | 07:22 |
*** inc0 has joined #kolla | 07:24 | |
inc0 | good mornign | 07:26 |
SamYaple | morning inc0 | 07:28 |
SamYaple | coolsvap: im not sure what the issue is here as I am not able to reproduce your issue even when I am using docker 1.3.0 | 07:29 |
SamYaple | docker-compose 1.3.0 and docker-py 1.3.0 play fune together | 07:29 |
SamYaple | s/fune/fine | 07:29 |
coolsvap | i faced the issue twice for two new setups i did | 07:30 |
coolsvap | and its even reproducible like always | 07:31 |
coolsvap | as soon as i install docker-py1.3.0 i get the error i added in bug with last line pkg_resources.DistributionNotFound: docker-py<1.3,>=1.2.3 | 07:31 |
coolsvap | afk, lunch | 07:32 |
*** coolsvap is now known as coolsvap|afk | 07:32 | |
SamYaple | I do not see the same behviour even on a freshly installed machine. docker-py 1.3.0 works ffine alongside docker-compose 1.3.0 | 07:36 |
SamYaple | inc0: did you confirm HUP would reload haproxy data? | 07:37 |
inc0 | SamYaple, not yet | 07:47 |
inc0 | I'll start working on this one today I guess | 07:47 |
inc0 | any new ideas regarding ansible? | 07:49 |
SamYaple | an haproxy role will work fine. i was playing with it earlier | 07:51 |
SamYaple | i would prefer to lay them down as the services go down, but thats fine too | 07:51 |
inc0 | them? configs? | 07:52 |
SamYaple | yea | 07:52 |
SamYaple | the per service haproxy config | 07:52 |
inc0 | what about not having haproxy on same server? | 07:53 |
SamYaple | thats why its an haproxy role | 07:53 |
SamYaple | it only runs on the haproxy hosts | 07:54 |
*** erkules has quit IRC | 07:54 | |
*** erkules has joined #kolla | 07:54 | |
inc0 | yeah, but where does configs lands? | 07:54 |
SamYaple | haproxy hosts... | 07:55 |
inc0 | so haproxy will generate configs? | 07:55 |
inc0 | haproxy role I mean | 07:55 |
SamYaple | it will generate the haproxy config files, yes | 07:55 |
inc0 | based on what? | 07:56 |
*** coolsvap|afk is now known as coolsvap | 07:56 | |
SamYaple | im very confused. what do you see the issue to be here | 07:56 |
inc0 | soo...haproxy has to know which endpoints are registered (or to be registered) right? | 07:56 |
inc0 | to prep configs for them | 07:56 |
SamYaple | yup | 07:57 |
inc0 | where is the endpoint list? with ports and stuff? | 07:57 |
SamYaple | ports are know since they are set globally. hosts are know since theyare set in the inventory | 07:58 |
inc0 | ok, but do we *automatically* asume that if port is set, we create endpoint for it? | 07:58 |
inc0 | assume* | 07:58 |
SamYaple | we dont create endpoints from the haproxy role, we simple setup the service to listen on that vip/port combo | 07:58 |
SamYaple | the port isn't _assumed_ to be correct, it is correct. that is the port the services will be using | 07:59 |
inc0 | I know, by setting up endpoint I mean create vip for it | 07:59 |
inc0 | there is 1:1 correlation between endpoints and vips, so I assume we should laverage that | 08:00 |
inc0 | brb | 08:00 |
SamYaple | im still not sure what the disconnect is here. but please just make the container since i have to build the role for keepalived and haproxy soon here | 08:00 |
SamYaple | im almost done with neutron | 08:00 |
SamYaple | once all of that is in place then it is functional and we can go back over all the strucutre and tweak it as desired | 08:01 |
*** inc0_ has joined #kolla | 08:01 | |
inc0_ | hmf | 08:03 |
SamYaple | i do not recognize that acroynm | 08:03 |
*** inc0 has quit IRC | 08:04 | |
*** jmccarthy has joined #kolla | 08:05 | |
inc0_ | its not an acronym, its an onomatopoeia shows deep concern about topic | 08:05 |
*** inc0_ is now known as inc0 | 08:05 | |
* SamYaple googles onomatopoeia | 08:05 | |
SamYaple | isnt that normally hmph? | 08:06 |
inc0 | it might, depends on lungs I guess | 08:06 |
*** athomas has joined #kolla | 08:10 | |
*** shardy has joined #kolla | 08:11 | |
mandre | hi all | 08:16 |
mandre | SamYaple: I played a with your python build script | 08:16 |
mandre | it didn't work out of the box for me, with boot2docker | 08:17 |
mandre | you need to pass env variables to docker client | 08:17 |
mandre | details at http://docker-py.readthedocs.org/en/latest/boot2docker/ | 08:18 |
SamYaple | mandre: im confused. im just using the docker api to build. that seems like it bypasses the env variabls... | 08:20 |
SamYaple | boot2docker doesnt use a different API | 08:21 |
SamYaple | also hi mandre! | 08:21 |
mandre | hi :) | 08:21 |
mandre | at least with boot2docker you need to explicitly pass environment variables to the client usign kwargs_from_env otherwise it can't connect with the server | 08:22 |
SamYaple | thats not how i read that article | 08:23 |
SamYaple | let me try out this boot2docker thing | 08:23 |
SamYaple | the build script is getting executed inside the boot2docker vm mandre? | 08:24 |
mandre | I think by default it tries to open unix socket, which doesn't work with boot2docker | 08:24 |
SamYaple | ah yea that could definetely be the case | 08:25 |
mandre | the build script executes on the client, in my case my mac | 08:25 |
SamYaple | but we can configure that. in fact i was just working on that | 08:25 |
SamYaple | thats al lthe remote stuff i need to bake in | 08:25 |
mandre | ah, great minds think alike :) | 08:25 |
SamYaple | and fools seldom differ | 08:26 |
coolsvap | SamYaple, can you please check /usr/lib/python2.7/site-packages/docker_compose-1.3.0-py2.7.egg-info/requires.txt on your environment? | 08:27 |
SamYaple | so mandre, I can go two ways with this. I can make the connection param completely configurable, or I can have "--boot2docker" flag that pulls in the environment var for the client | 08:28 |
SamYaple | coolsvap: sure | 08:28 |
coolsvap | mine has docker-py >= 1.2.3, < 1.3 entry | 08:28 |
mandre | wouldn't injecting env vars break compatibility with linux clients? | 08:29 |
mandre | i don't have a linux machine at hand to test myself | 08:29 |
SamYaple | mandre: no. not if i understand the situation correctly. but injecting env would only happen with the flag anyway | 08:30 |
SamYaple | coolsvap: wait you arent installing via the wheel? | 08:31 |
mandre | yeah, I mean if injecting env doesn't break connection to local docker host, why bother making a flag? | 08:32 |
SamYaple | mandre: it could if that variables were set | 08:32 |
SamYaple | i dont know what all vars it checks to be honest | 08:32 |
coolsvap | SamYaple, no, let me check with wheel | 08:34 |
mandre | SamYaple, neither do i | 08:34 |
SamYaple | coolsvap: ill check not the wheel ;) | 08:34 |
mandre | I believe it's only the DOCKER_* variables but I haven't checked | 08:36 |
SamYaple | mandre: it just checks 3, DOCKER_{HOST,CERT_PATH,TLS_VERIFY} | 08:36 |
SamYaple | yea we can do that. no problem. ill test it locally | 08:36 |
SamYaple | without boot2docker | 08:36 |
mandre | cool, thanks SamYaple | 08:36 |
SamYaple | ill file a bug for it real quick | 08:37 |
mandre | I've also noticed I couldn't interrupt the build script | 08:37 |
SamYaple | yea thats the parallelness | 08:37 |
SamYaple | you _can_ | 08:37 |
SamYaple | with ctrl + \ | 08:37 |
SamYaple | im not sure how to handle the normal KeyboardInterupt | 08:38 |
mandre | ah.. | 08:38 |
SamYaple | since im calling a blocking function ( pool.join() ) it doesnt break for interput | 08:38 |
SamYaple | will probably have to switch to looping :( | 08:38 |
coolsvap | SamYaple, perhaps you are not getting docker-py 1.3.0 with wheel | 08:39 |
coolsvap | because docker-compose needs docker-py >=1.2.3 <=1.3.0 | 08:43 |
coolsvap | https://github.com/docker/compose/commit/ac56ef3d659f04164aa44f1a4f0c991fb5eb6060 | 08:43 |
coolsvap | SamYaple, ^ | 08:43 |
coolsvap | so for all new installation from pypi i think it will fail | 08:48 |
SamYaple | coolsvap: no i am definetely installing form the wheel, you dont appear to be though wit hthe egg-info and all | 08:48 |
coolsvap | SamYaple, didnt get what you exactly meant | 08:50 |
SamYaple | docker_compose-1.3.0-py2.7.egg-info/requires.txt | 08:50 |
SamYaple | ^ thats not a WHEEL | 08:50 |
SamYaple | whell has this stored in METADATA | 08:50 |
coolsvap | no, its not wheel | 08:50 |
SamYaple | my metadata does have the <1.3 requirement, but it isn't breaking where yours is | 08:50 |
coolsvap | i agree, but even after i tried with wheel, i am getting failure | 08:50 |
coolsvap | oh.. | 08:51 |
SamYaple | im not a huge python packaging guy, but we need to figure out why this is not resolving the more restrictive <1.3 dependacy for you when docker-py is above docker-compose | 08:53 |
SamYaple | it is for me | 08:53 |
SamYaple | (docker-py below doent break it either, as long as I am not running with -U) | 08:54 |
SamYaple | my guess is it has to do with the non-use of wheel for you, but trying to confirm | 08:54 |
coolsvap | hmm, i will try with wheel one fresh system, let me check | 08:57 |
*** inc0 has quit IRC | 09:08 | |
*** inc0 has joined #kolla | 09:15 | |
SamYaple | mandre: if you simple change self.db in the WorkerThread class to 'self.dc = docker.Client(**docker.utils.kwargs_from_env())' does it work for you? | 09:15 |
SamYaple | if so that works fine for me as well, no other modifications needed | 09:16 |
SamYaple | self.dc* | 09:16 |
*** athomas has quit IRC | 09:17 | |
mandre | SamYaple: that's the change I did locally, so yes it works | 09:22 |
SamYaple | cool then thats a safe change. will add with the push script im writing | 09:22 |
SamYaple | thanks! | 09:22 |
SamYaple | any other feedback on the script? | 09:23 |
mandre | well, thanks to you | 09:23 |
mandre | no other feedback yet, maybe after using it for a while | 09:24 |
SamYaple | i was considering using tmux (if installed, not a requirement) to generate a tmux window for each build so you can watch each build as it goes. | 09:24 |
SamYaple | killing the windows if it completes successfully, keeping them if it fails | 09:25 |
SamYaple | with the main window being for ongoing status for all builds. thoughts? | 09:25 |
mandre | would it play well if you start the build from inside tmux? | 09:25 |
SamYaple | hell yea! but we need to define behaviour | 09:25 |
SamYaple | tmux on tmux or extend current window | 09:26 |
SamYaple | ot new tmux session | 09:26 |
SamYaple | i tmux everywhere | 09:26 |
mandre | then i'm all good :) | 09:26 |
SamYaple | what is your prefered behaviour out of the list above? or do you want options? | 09:26 |
mandre | it's a cool idea | 09:26 |
mandre | I would start a new session, and print a message explaining how to attach to the new session | 09:27 |
mandre | that might be confusing though for people not used to tmux... | 09:28 |
SamYaple | not required though | 09:28 |
SamYaple | i mean with this threading we need something more usable than what we have for progress | 09:29 |
*** athomas has joined #kolla | 09:29 | |
mandre | right, at the moment you buffer logs until build finishes, right? | 09:29 |
SamYaple | correct | 09:29 |
mandre | not ideal | 09:29 |
SamYaple | its the only way its even slightly readable | 09:30 |
mandre | i like the idea of using tmux for separating build outputs | 09:31 |
mandre | leaving now, bye | 09:32 |
SamYaple | bai | 09:33 |
*** dims has joined #kolla | 09:58 | |
*** dolpher has quit IRC | 10:03 | |
openstackgerrit | Swapnil Kulkarni proposed stackforge/kolla: Update scripts to check if docker is usable for $USER https://review.openstack.org/201022 | 10:15 |
openstackgerrit | Swapnil Kulkarni proposed stackforge/kolla: Update scripts to check if docker is usable for $USER https://review.openstack.org/201022 | 10:17 |
*** coolsvap is now known as coolsvap|away | 10:25 | |
harmw | SamYaple: regarding the NTP container, I was just curious on the why-part :) I too feel that's not neccesarily on our shoulders, and certainly not on short term | 10:28 |
*** coolsvap|away is now known as coolsvap | 10:28 | |
SamYaple | message queue messages timing | 10:28 |
harmw | sure, or mariadb-galera? | 10:28 |
SamYaple | openstack is pretty tolerate, but get 5-10 seconds out of sync and its bad | 10:28 |
harmw | out of sync sucks, always | 10:29 |
SamYaple | galera should be safe with widly different times thans to the GTID | 10:29 |
harmw | pbourke: https://bugs.launchpad.net/kolla/+bug/1473554 | 10:29 |
openstack | Launchpad bug 1473554 in kolla "properly check if docker is usable for $USER" [Undecided,In progress] - Assigned to Swapnil Kulkarni (coolsvap) | 10:29 |
SamYaple | ceph requires <0.05 seconds | 10:29 |
pbourke | harmw: yes? | 10:30 |
harmw | removing the root check was fine at first | 10:30 |
harmw | though I'd rather have a thourough check instead | 10:30 |
harmw | that bugreport has a little motivation for that | 10:31 |
harmw | furthermore, regarding the specific changeset, we can't use kolla-common for such a thing since this targets the build/dev env, and kolla-common is for in-container-usage :) | 10:31 |
SamYaple | hey harmw. i need some patches merged. https://review.openstack.org/#/c/200859/ https://review.openstack.org/#/c/199463/ | 10:32 |
SamYaple | niether touch the dockerfiles | 10:32 |
harmw | SamYaple: how do call that again, where one doesn't copy paste the same function on and on but instead includes it from a central library or something? | 10:32 |
SamYaple | just ansibleneww | 10:32 |
SamYaple | wait what | 10:32 |
harmw | there is some fancy word for that | 10:32 |
SamYaple | source? | 10:33 |
harmw | Ill ack those reviews tonight | 10:33 |
harmw | no no | 10:33 |
harmw | a higher level term | 10:33 |
harmw | programmers lingo | 10:33 |
harmw | *devvers | 10:33 |
SamYaple | harmw: whats tonight for oyu? | 10:34 |
harmw | good one | 10:34 |
harmw | lets keep it around 6hrs from now | 10:34 |
harmw | that's 18:34 | 10:34 |
SamYaple | :) im just giving you are hard time | 10:34 |
harmw | you've looked at single nic lately SamYaple? | 10:35 |
SamYaple | harmw: for OVS.... | 10:35 |
harmw | linuxbridge, sory | 10:35 |
SamYaple | i already said i dont know linuxbridge enough to get it working | 10:35 |
SamYaple | i _know_ ovs works | 10:36 |
harmw | I know that as well :) | 10:37 |
harmw | but we're lacking ovs containers, right? | 10:37 |
harmw | well, they are there but probably just not working | 10:37 |
SamYaple | for config-interal, i dont think anyone wrote support for that | 10:37 |
SamYaple | im doing OVS for ansible config-external stuff | 10:37 |
SamYaple | screw linux bridge | 10:37 |
harmw | yep | 10:38 |
harmw | I thought of getting bridge working using an additional vlan interface on eth0, but it maybe better to focus on OVS instead | 10:38 |
SamYaple | linuxbridge will never support some new neutron features like DVR, so i tend to ignore it when i can | 10:39 |
harmw | yep | 10:39 |
SamYaple | with OVS 2.3.1 it doesnt even have a large performance advantage anymore | 10:39 |
SamYaple | maybe 10% | 10:40 |
harmw | is your OVS ansible work up for review? | 10:41 |
harmw | with that, we can go single-nic and get gating properly:) | 10:42 |
*** coolsvap is now known as coolsvap|away | 10:43 | |
SamYaple | im working on neutron now, but its about 4 commits deep. thats why i need acks | 10:44 |
*** fangfenghua has joined #kolla | 10:44 | |
harmw | Ill get you those tonight ;) | 10:45 |
*** rhallisey has joined #kolla | 10:52 | |
openstackgerrit | Mark Duggan proposed stackforge/kolla: Fixes login failure to Horizon board https://review.openstack.org/201081 | 11:14 |
*** shardy_ has joined #kolla | 11:17 | |
*** shardy has quit IRC | 11:19 | |
*** shardy_ has quit IRC | 11:23 | |
*** shardy has joined #kolla | 11:23 | |
*** tobe has quit IRC | 11:36 | |
*** achanda has joined #kolla | 11:37 | |
*** shardy_ has joined #kolla | 11:41 | |
SamYaple | welp. thats why our pushes take for fucking ever | 11:41 |
*** shardy has quit IRC | 11:42 | |
openstackgerrit | Swapnil Kulkarni proposed stackforge/kolla: Update scripts to check if docker is usable for $USER https://review.openstack.org/201022 | 11:44 |
*** coolsvap|away is now known as coolsvap | 11:44 | |
*** shardy_ has quit IRC | 11:46 | |
*** shardy has joined #kolla | 11:47 | |
*** achanda has quit IRC | 11:50 | |
*** gfidente has joined #kolla | 11:53 | |
*** inc0 has quit IRC | 12:01 | |
*** inc0 has joined #kolla | 12:02 | |
harmw | coolsvap: Ill add my notes when I get home | 12:11 |
coolsvap | harmw, sure | 12:13 |
*** achanda has joined #kolla | 12:15 | |
*** britthouser has joined #kolla | 12:22 | |
*** inc0 has quit IRC | 12:47 | |
*** fangfenghua has quit IRC | 12:48 | |
*** absubram has quit IRC | 12:50 | |
*** inc0 has joined #kolla | 12:54 | |
*** dims has quit IRC | 12:55 | |
*** Guest33195 has joined #kolla | 12:57 | |
*** Guest33195 is now known as dims_ | 12:58 | |
*** jmccarthy has quit IRC | 12:58 | |
*** jmccarthy has joined #kolla | 12:59 | |
*** shardy_ has joined #kolla | 13:03 | |
*** shardy has quit IRC | 13:04 | |
*** inc0 has quit IRC | 13:06 | |
*** shardy_ has quit IRC | 13:08 | |
*** shardy has joined #kolla | 13:09 | |
pbourke | coolsvap: is noop a valid command? | 13:13 |
SamYaple | it is not | 13:17 |
harmw | nope | 13:17 |
harmw | it'd be better to put the logic the other way around | 13:17 |
harmw | if not bla; then exit | 13:17 |
coolsvap | no | 13:17 |
coolsvap | i think i should use not in if | 13:17 |
pbourke | I agree | 13:17 |
harmw | that was pseudo-code:) | 13:17 |
pbourke | I didnt want to give another -1 :( | 13:17 |
harmw | I would've, tonight :P | 13:18 |
harmw | (but there was more, oh well) | 13:18 |
coolsvap | harmw, please let me know the comments | 13:19 |
harmw | I only gave it a quick glance, this if statements was the first that I noticed | 13:20 |
harmw | I had another note, I'll think of it when I properly do my review-task :) | 13:20 |
harmw | will that work for you? | 13:20 |
harmw | since that'll be another 4hrs or so from now | 13:21 |
pbourke | I bet harmw will comment on the double space in the commit msg ;P | 13:21 |
harmw | not if it's after a period ;) | 13:21 |
harmw | since apparently that is proper English | 13:21 |
*** inc0 has joined #kolla | 13:22 | |
pbourke | that's debateable apparently | 13:22 |
harmw | it is? | 13:22 |
harmw | fair enough | 13:22 |
pbourke | https://en.wikipedia.org/wiki/Sentence_spacing | 13:23 |
pbourke | I always used double but recently have tried switching | 13:23 |
inc0 | rhallisey, ping | 13:24 |
harmw | funny, I had never heard of using double spacing in the first place :) | 13:24 |
*** dwalsh has joined #kolla | 13:25 | |
rhallisey | inc0, hey | 13:32 |
SamYaple | morning all. heading home | 13:32 |
SamYaple | morning rhallisey | 13:32 |
inc0 | hi, are we ready to submi talk? | 13:32 |
SamYaple | I just submit | 13:32 |
rhallisey | SamYaple, morning | 13:32 |
SamYaple | MWAHAHA | 13:32 |
inc0 | cyas Sam, have a nice..day? night? sleep? | 13:32 |
SamYaple | dunno. might be back around | 13:32 |
SamYaple | gotta submit the push patch for the build script | 13:32 |
rhallisey | ya I'll submit when I talk to Slower again | 13:33 |
inc0 | kk | 13:33 |
*** jtriley has joined #kolla | 13:33 | |
rhallisey | inc0, I'll add you to it | 13:33 |
inc0 | feel free to change my abstract however you like | 13:33 |
inc0 | or throw it away alltogether;) | 13:33 |
SamYaple | inc0: what was your talk on again? | 13:33 |
inc0 | SamYaple, which one?;) | 13:33 |
SamYaple | yes | 13:33 |
inc0 | it seems I'll have 3 | 13:33 |
inc0 | one we were talking will be upgrades using kolla | 13:34 |
inc0 | besides that I'll have one about "how to migrate your lame non-contenerized openstack to cool kolla-based one without anyone noticing" | 13:34 |
harmw | both sound cool inc0 | 13:35 |
inc0 | and Mesos + Kolla = setting up openstack quite dynamically | 13:35 |
harmw | wicked | 13:35 |
harmw | mesos+kolla | 13:35 |
SamYaple | all centered around triple-o? any config-external/Ansibleness? | 13:35 |
inc0 | so you can shring OS, expand OS, whatever you like to do | 13:35 |
harmw | interesting | 13:35 |
inc0 | SamYaple, that depends really, I hardly can imagine turining non-tripleo openstack to tripleo | 13:35 |
inc0 | so probably I'll use ansible + config external to keep old configs | 13:36 |
inc0 | for migration talk | 13:36 |
inc0 | but I want to test out both | 13:36 |
SamYaple | cool sounds good | 13:36 |
inc0 | I'm a fan of ansible now you knwo... | 13:36 |
*** mstachow has joined #kolla | 13:36 | |
harmw | everyone should be! | 13:37 |
SamYaple | i was checking if we had strong overlap :) | 13:37 |
inc0 | do we? | 13:37 |
SamYaple | ive got one about build/deploying customer openstack form source (tweak openstack source then test and deploy it) | 13:37 |
*** jruano has joined #kolla | 13:37 | |
SamYaple | and then a lab about, operators guid eto everything kolla around the ansible stuff | 13:38 |
inc0 | cool one, I'll be there to correct you | 13:38 |
SamYaple | correct me? | 13:38 |
inc0 | (even if you won't be wrong) | 13:38 |
SamYaple | you cant correct correct | 13:38 |
inc0 | there is nothing that cant be more correct | 13:38 |
inc0 | correcter | 13:38 |
SamYaple | unless I said it then its the correctest | 13:39 |
inc0 | there is nothing that cant be more correct | 13:39 |
inc0 | correcter | 13:39 |
inc0 | thats only rule of world | 13:39 |
SamYaple | i will be correcterer | 13:39 |
SamYaple | im correcter than the correcter | 13:39 |
SamYaple | whos cooler than cool | 13:39 |
inc0 | we'll see about that | 13:40 |
SamYaple | you have failed. the correct answer is "ice cold" | 13:40 |
SamYaple | you are the weakest link. goodbye | 13:41 |
*** shardy_ has joined #kolla | 13:44 | |
*** shardy has quit IRC | 13:44 | |
*** prad has joined #kolla | 13:46 | |
*** shardy_ has quit IRC | 13:49 | |
*** shardy has joined #kolla | 13:50 | |
*** shardy_ has joined #kolla | 13:54 | |
*** shardy has quit IRC | 13:56 | |
*** shardy_ has quit IRC | 14:00 | |
*** shakamunyi has joined #kolla | 14:00 | |
*** barra204 has joined #kolla | 14:00 | |
*** shardy has joined #kolla | 14:01 | |
*** shardy_ has joined #kolla | 14:06 | |
*** absubram has joined #kolla | 14:06 | |
*** shardy has quit IRC | 14:07 | |
*** shardy_ has quit IRC | 14:11 | |
*** bmace has joined #kolla | 14:11 | |
*** shardy has joined #kolla | 14:11 | |
*** achanda has quit IRC | 14:18 | |
*** pbourke has quit IRC | 14:21 | |
*** pbourke has joined #kolla | 14:22 | |
*** barra204 has quit IRC | 14:25 | |
*** shakamunyi has quit IRC | 14:25 | |
*** whydidyoustealmy has joined #kolla | 14:26 | |
*** shakamunyi has joined #kolla | 14:26 | |
*** whydidyoustealmy has quit IRC | 14:34 | |
*** shakamunyi has quit IRC | 14:34 | |
*** jtriley has quit IRC | 14:36 | |
*** blahRus has joined #kolla | 14:48 | |
inc0 | I'm off | 14:50 |
inc0 | have a nice day guys | 14:50 |
inc0 | rhallisey, let me know when you submit talk please. | 14:51 |
rhallisey | inc0, sure | 14:51 |
inc0 | cya tomorrow | 14:51 |
*** inc0 has quit IRC | 14:52 | |
*** kevsi has quit IRC | 14:59 | |
*** mstachow has quit IRC | 15:00 | |
*** daneyon has joined #kolla | 15:05 | |
*** jtriley has joined #kolla | 15:11 | |
SamYaple | so i found out why docker pushes take forever rhallisey | 15:26 |
rhallisey | why | 15:26 |
SamYaple | apparently, and this is great, wit hthe new docker v2 registry you end up pushing the same layer over and over again. you basically push EVERY LAYER FOR EVERY IMAGE | 15:26 |
SamYaple | download is apparently unaffected | 15:27 |
SamYaple | but the compression is breaking the "layer already exists" mess | 15:27 |
SamYaple | a few bugs about it floating around | 15:27 |
SamYaple | ok i need to go to the sleep | 15:29 |
SamYaple | im out see ya guys | 15:29 |
rhallisey | interesting | 15:30 |
rhallisey | SamYaple, night :) | 15:30 |
bmace | nite SamYaple | 15:32 |
*** jpeeler has quit IRC | 15:33 | |
*** akscram has quit IRC | 15:36 | |
*** mordred has quit IRC | 15:36 | |
*** kragniz has quit IRC | 15:36 | |
*** vincent_vdk has quit IRC | 15:36 | |
*** mfalatic has joined #kolla | 15:36 | |
*** kragniz has joined #kolla | 15:37 | |
*** vincent_vdk has joined #kolla | 15:37 | |
*** akscram has joined #kolla | 15:37 | |
*** mordred has joined #kolla | 15:37 | |
*** mfalatic_ has joined #kolla | 15:37 | |
*** jpeeler has joined #kolla | 15:39 | |
*** daneyon_ has joined #kolla | 15:41 | |
*** mfalatic has quit IRC | 15:41 | |
*** daneyon has quit IRC | 15:44 | |
*** vinkman has joined #kolla | 16:07 | |
*** sdake has joined #kolla | 16:15 | |
sdake | morning | 16:18 |
*** diogogmt has joined #kolla | 16:20 | |
*** sdake has quit IRC | 16:23 | |
*** sdake has joined #kolla | 16:24 | |
*** shakamunyi has joined #kolla | 16:25 | |
*** whydidyoustealmy has joined #kolla | 16:26 | |
*** achanda has joined #kolla | 16:31 | |
*** akwasnie has quit IRC | 16:31 | |
*** whydidyoustealmy has quit IRC | 16:37 | |
*** shakamunyi has quit IRC | 16:37 | |
*** mfalatic_ is now known as mfalatic | 16:49 | |
*** jasonsb has quit IRC | 16:51 | |
*** jasonsb has joined #kolla | 16:52 | |
*** shakamunyi has joined #kolla | 16:52 | |
*** whydidyoustealmy has joined #kolla | 16:52 | |
sdake | samyaple around? | 16:52 |
*** jasonsb has quit IRC | 16:53 | |
bmace | hey sdake, he said he was going to sleep about an hour and a half ago. | 16:53 |
sdake | bmace ack thanks | 16:53 |
sdake | http://www.amazon.com/Samson-Stage-XPD1-Digital-Wireless/dp/B00LN0NUHQ/ref=sr_1_2?ie=UTF8&qid=1436806593&sr=8-2&keywords=usb+handheld+mics+wireless | 16:57 |
sdake | do folks think that will work out well for midcycle? | 16:57 |
sdake | wireless and usb | 16:57 |
bmace | make sure it is drop resistant, for those dramatic speech endings :) | 16:59 |
sdake | says it is mac compatible | 16:59 |
sdake | you drop it you bought it ;) | 16:59 |
sdake | we have 11 people signed upf or remote | 16:59 |
sdake | so I was thinking of getting 3 mics and passing around room | 16:59 |
sdake | and 11 people on site | 17:00 |
*** juggler_ is now known as juggler | 17:04 | |
*** athomas has quit IRC | 17:04 | |
*** bmace has quit IRC | 17:06 | |
*** jruano has quit IRC | 17:07 | |
*** bmace has joined #kolla | 17:08 | |
*** shakamunyi has quit IRC | 17:09 | |
*** whydidyoustealmy has quit IRC | 17:09 | |
*** shakamunyi has joined #kolla | 17:09 | |
*** bmace has quit IRC | 17:10 | |
*** athomas has joined #kolla | 17:11 | |
openstackgerrit | Merged stackforge/kolla: Add Keystone image for Ubuntu https://review.openstack.org/196424 | 17:16 |
openstackgerrit | Merged stackforge/kolla: filter ansible/library from tox.ini https://review.openstack.org/200859 | 17:16 |
openstackgerrit | Merged stackforge/kolla: Add temporary Ansible keystone modules https://review.openstack.org/199463 | 17:16 |
*** bmace has joined #kolla | 17:17 | |
* sdake shakes fist at gate | 17:21 | |
sdake | jpeeler thoughts on http://logs.openstack.org/24/196424/6/check/check-kolla-functional-f21/7c8d9ec/console.html.gz#_2015-07-09_17_53_21_395 | 17:21 |
sdake | it is doing 3 retries | 17:21 |
sdake | but still fails | 17:21 |
sdake | because of that ipv6 address, not even sure where that is originating from | 17:21 |
jpeeler | wish i knew ipv6. i'd say let's turn it off in the gate | 17:27 |
jpeeler | sdake: you okay with that? | 17:27 |
sdake | jpeeler I'd be ok with that but I don't think that owuld fix it | 17:27 |
sdake | the problem maybe that ipv6 isn't on in the gate | 17:27 |
sdake | the client is trying to obtain ipv6 addresses for that node | 17:28 |
jpeeler | right, and so turning it off would force ipv4 no? | 17:28 |
harmw | guys | 17:28 |
sdake | but the other images it doesnt seem to happen to or the ipv6 flakiness goes away | 17:28 |
sdake | hey harmw | 17:28 |
harmw | just a IP6 routing issue | 17:28 |
sdake | jpeeler not sure worth a go | 17:28 |
sdake | jpeeler that worst that could happen is we would have to revert such a change | 17:28 |
harmw | either they fix routing IP6 or they remove that default route or V6 addr | 17:28 |
harmw | inet6 2001:4800:7815:102:be76:4eff:fe05:1e60/64 scope global | 17:29 |
harmw | thats the problem | 17:29 |
harmw | having that one | 17:29 |
harmw | and the rest | 17:29 |
harmw | just have the infra boys fix IP6 :) | 17:30 |
sdake | harmw you think its definately an ipv6 routing issue? | 17:30 |
harmw | yea | 17:30 |
sdake | join #openstack-infra and make your case ;) | 17:30 |
harmw | nah, I'm busy with other stuff atm | 17:30 |
* jpeeler rather make a four line change and move on | 17:31 | |
harmw | pbourke: https://review.openstack.org/199077 , you mind rebasing that? | 17:32 |
*** jruano has joined #kolla | 17:35 | |
openstackgerrit | Jeff Peeler proposed stackforge/kolla: Disable ipv6 in gate https://review.openstack.org/201267 | 17:41 |
jpeeler | see if that works | 17:41 |
harmw | jpeeler: can't you just sysctl -w that? or is there a reason for storing/saving it? | 17:43 |
harmw | plus, are those spaces arround =-sign allowed? | 17:43 |
jpeeler | could! | 17:43 |
jpeeler | spaces worked on my test | 17:43 |
jpeeler | um, actually i didn't test the sysctl | 17:43 |
jpeeler | but i have an existing entry in mine with spaces | 17:44 |
harmw | fair enough, might as well be you can't use spaces when issueing sysctl -w | 17:44 |
*** bmace has quit IRC | 17:44 | |
harmw | let me know if this is tested jpeeler, so I can ack it | 17:45 |
jpeeler | ha, well apparently sysctl isn't in the gate | 17:46 |
harmw | yum install :> | 17:46 |
harmw | wwhat kind of env is the gate anyway? | 17:46 |
harmw | qemu vm's per project or something? | 17:47 |
openstackgerrit | Jeff Peeler proposed stackforge/kolla: Disable ipv6 in gate https://review.openstack.org/201267 | 17:47 |
jpeeler | project-config dictates the gate env, this one is a f21 host. no idea where the image used came from | 17:48 |
sdake | jpeeler someone from rht built it - ian sometihng iirc | 17:51 |
sdake | not mccloud a different ian | 17:51 |
jpeeler | it's a one off image though? | 17:52 |
jpeeler | come on gate, this should be a 5 second fix | 17:52 |
sdake | jpeeler it will run all the image builds | 17:53 |
sdake | hence take 45 mins or so | 17:53 |
*** jasonsb has joined #kolla | 17:53 | |
jpeeler | yeah, well i just want to make sure the change works. fairly confident it'll solve the issue! | 17:53 |
sdake | jpeeler ya one off | 17:53 |
sdake | that would be great if it does :) | 17:53 |
* sdake so happy | 17:53 | |
jpeeler | is that funcsigs thing still around? | 17:54 |
sdake | not sure haven't seen it for awhile | 17:54 |
jpeeler | uh | 17:55 |
jpeeler | how is the command not found | 17:55 |
jpeeler | sudo: sysctl: command not found | 17:55 |
sdake | specify full path | 17:55 |
sdake | you have to do that for some reason | 17:55 |
jpeeler | at least sysctl was already installed, i did think that being missing was strange | 17:56 |
openstackgerrit | Jeff Peeler proposed stackforge/kolla: Disable ipv6 in gate https://review.openstack.org/201267 | 17:56 |
sdake | ya /usr/sbin is not in the gate path | 17:57 |
sdake | that is why it wasn't found | 17:57 |
harmw | env sysctl | 17:57 |
jpeeler | is that the proper way? | 17:58 |
harmw | it just looks it up in $PATH | 17:58 |
harmw | /usr/bin/env sysctl | 17:58 |
harmw | or something | 17:58 |
sdake | hardcoded pat hs i fine too | 17:58 |
harmw | yep | 17:58 |
harmw | but this is just a little less hardcoded :) | 17:59 |
jpeeler | why is /usr/sbin not in the path? | 17:59 |
sdake | i think it needs to be passed into the gate env or something - not sure | 18:00 |
sdake | in project-config | 18:00 |
*** jasonsb has quit IRC | 18:00 | |
*** jasonsb has joined #kolla | 18:05 | |
*** jasonsb has quit IRC | 18:09 | |
*** inc0 has joined #kolla | 18:13 | |
inc0 | evening guys | 18:13 |
sdake | hey inc0 | 18:19 |
inc0 | sdake, are you comming to ops midcycle? | 18:19 |
sdake | i am not | 18:19 |
sdake | when is it? | 18:19 |
inc0 | mid august | 18:19 |
inc0 | 18-19 I think | 18:19 |
inc0 | or something like that | 18:19 |
jpeeler | ok sdake that patch seems to be working, guess you can approve it once gate finishes successfully | 18:20 |
inc0 | I was thinking that we should have some presence there, anyone going there? | 18:20 |
sdake | my budget is gone, I have to use personal money to travel to magnum midcycle | 18:20 |
sdake | plus I am out approximately 300 for microphones for our midcycle | 18:21 |
sdake | so needless to say, I have the out of F money error :) | 18:21 |
inc0 | yeah, trust me, I understand | 18:22 |
sdake | jpeeler i acked it | 18:22 |
inc0 | for me even kolla midcycle isn't small undertaking | 18:22 |
sdake | hopefully that tightenes up the gate operation | 18:23 |
sdake | inc0 I hear ya | 18:23 |
sdake | inc0 your over in emea right? | 18:23 |
inc0 | yeah, Poland | 18:23 |
sdake | never been | 18:23 |
sdake | is that near .cz | 18:23 |
inc0 | it is | 18:24 |
inc0 | north of it | 18:24 |
sdake | i've been to the cz republic before | 18:24 |
Slower | ENOMONEY | 18:24 |
sdake | prague to be specific | 18:25 |
sdake | prague rocks | 18:25 |
sdake | we had a 30 person feast for 25 $ USD | 18:25 |
sdake | I couldn't believe it | 18:25 |
inc0 | that had to be long time ago | 18:25 |
harmw | yea, that sounds a lot like eastern europe :p | 18:26 |
sdake | 6 years ago? | 18:26 |
sdake | 8 year sago | 18:26 |
sdake | it was a couple thousand crowns | 18:26 |
inc0 | then wow, I have no idea how did you find spot like that | 18:26 |
sdake | but the exchane rate was pretty good :) | 18:26 |
sdake | it may have been 40$ USD | 18:26 |
sdake | it was really cheap | 18:26 |
sdake | it wasn't super fancy or anything | 18:26 |
sdake | reminded me ofa home cooked meal | 18:27 |
inc0 | normal dinner would be like 15$ per person | 18:27 |
sdake | pretty sweet place- wish I remember where it was | 18:27 |
inc0 | or something like that | 18:27 |
sdake | prague itself was expensive | 18:27 |
sdake | dinners were more then us dinners - couldn't believe it | 18:27 |
sdake | this was outsdie of prague | 18:27 |
sdake | i do remember laing on a board for a bed | 18:27 |
inc0 | still it was quite a find:) | 18:28 |
sdake | they had no mattress, or it was like a sheet or two or something | 18:28 |
sdake | painful | 18:28 |
inc0 | Prague is very beautiful city, it wasn't destroyed during WW2 | 18:28 |
inc0 | so most of their architecture is authentic | 18:28 |
*** slagle_ has joined #kolla | 18:31 | |
*** slagle has quit IRC | 18:33 | |
*** vinkman has left #kolla | 18:36 | |
*** shakamunyi has quit IRC | 18:36 | |
*** gfidente has quit IRC | 18:41 | |
*** sdake has quit IRC | 18:41 | |
*** sdake has joined #kolla | 18:44 | |
*** shakamunyi has joined #kolla | 18:47 | |
inc0 | good night everyone | 18:59 |
*** inc0 has quit IRC | 18:59 | |
*** jasonsb has joined #kolla | 19:05 | |
*** thumpba has joined #kolla | 19:42 | |
*** jasonsb_ has joined #kolla | 20:07 | |
*** jasonsb has quit IRC | 20:07 | |
sdake | daneyon_ ping | 20:09 |
daneyon_ | sdake hey! | 20:09 |
sdake | hey I need your email for registering our talk with openstack foundation | 20:10 |
sdake | is it danehans@gmail.com? | 20:10 |
daneyon_ | daneyonhansen@gmail.com | 20:11 |
sdake | that account doesn't have your bio on it | 20:12 |
sdake | or your picture | 20:13 |
sdake | etc | 20:13 |
daneyon_ | hmm | 20:13 |
daneyon_ | maybe it's associated to danehans@cisco.com | 20:13 |
daneyon_ | let me check | 20:13 |
sdake | let me try that one | 20:14 |
sdake | 1394 is your profile that has your bio and pic | 20:14 |
sdake | its not 1394 | 20:14 |
daneyon_ | looks like danehans@cisco.com is the email | 20:15 |
sdake | can you login to https://www.openstack.org/summit/tokyo-2015/call-for-speakers/ | 20:16 |
sdake | and add yourself as a speaker there , and enter your pic/bio/etc | 20:17 |
sdake | maybe the db's aren't synced | 20:17 |
sdake | what login does danehans@cisco.com give you for id, 1394? | 20:17 |
sdake | your in the registry as 1394 | 20:17 |
sdake | note when I entere dthe cfp, I had to enter my speaker bio as well | 20:18 |
sdake | from scratch iirc | 20:18 |
sdake | daneyon_ I entered a couple details for you, but could you finish it out including pic plz | 20:25 |
sdake | thanks :) | 20:25 |
daneyon_ | sdake, sure | 20:26 |
sdake | cool - that talk should be done then | 20:26 |
sdake | as in submitted done :) | 20:26 |
sdake | not actually done | 20:26 |
sdake | how was nerdcon :) | 20:26 |
sdake | daneyon_ gophercon rather :) | 20:27 |
daneyon_ | sdake gC was good. A lot of stuff over my head, but it was good to block everything out and focus on learning Go. I learned a fair bit the last week... just need to keep plugging away and hopefully soon I can start writing some code | 20:30 |
sdake | need to have something to write first :) | 20:30 |
*** athomas has quit IRC | 20:32 | |
*** athomas has joined #kolla | 20:33 | |
*** diga has quit IRC | 20:37 | |
*** shardy_ has joined #kolla | 21:04 | |
harmw | something involving world domination will do just fine | 21:04 |
*** shardy has quit IRC | 21:05 | |
openstackgerrit | Merged stackforge/kolla: Disable ipv6 in gate https://review.openstack.org/201267 | 21:07 |
sdake | as long as it isn't skynet ;-) | 21:07 |
sdake | google is already on that one | 21:07 |
harmw | yea, like 10yrs ago | 21:08 |
*** shardy_ has quit IRC | 21:09 | |
*** absubram has quit IRC | 21:09 | |
*** shardy has joined #kolla | 21:10 | |
*** dwalsh has quit IRC | 21:19 | |
sdake | jpeeler that gate went through without error yay :) | 21:24 |
*** shadower has joined #kolla | 21:24 | |
*** coolsvap has quit IRC | 21:28 | |
*** dperaza has joined #kolla | 21:33 | |
*** coolsvap|away has joined #kolla | 21:33 | |
dperaza | has any one gotten cinder-volume configured with LVM using LIO instead of TGT | 21:34 |
*** dims___ has joined #kolla | 21:34 | |
dperaza | TGT working good for me but RHEL7 does not include the scsi-target-utils | 21:34 |
*** dims_ has quit IRC | 21:35 | |
dperaza | I keep getting a login error from iscsiadm on the nova-compute side | 21:35 |
sdake | dperaza we have not got tgt working either so no | 21:36 |
dperaza | what are you missing in tgt maybe I can can help | 21:37 |
sdake | not sure what we are missing | 21:37 |
sdake | something apparently :( | 21:38 |
sdake | here is the patch stream to add tgtd support: https://review.openstack.org/#/c/200139/ | 21:38 |
*** thumpba has quit IRC | 21:38 | |
sdake | here is the backtrace we get: https://bugs.launchpad.net/kolla/+bug/1469323 | 21:39 |
openstack | Launchpad bug 1469323 in kolla "cinder nova volume attach not working" [Critical,Triaged] - Assigned to Ryan Hallisey (rthall14) | 21:39 |
sdake | if you have a solution for it, would love to hear it ;) | 21:39 |
dperaza | let me check that patch | 21:39 |
sdake | you really have to check master, that patch just runs tgtd | 21:39 |
sdake | in the cinder-volume container | 21:40 |
dperaza | sure | 21:40 |
sdake | rhallisey said he has had it working somewhere,but he doesn't know where and doesn't have time to look at it at present | 21:40 |
sdake | so alas, we are jammed up | 21:40 |
sdake | have you seen that backtrace before? | 21:40 |
sdake | last):\n', ' File "/usr/lib/python2.7/site-packages/oslo_messaging/rpc/dispatcher.py", line 142, in _dispatch_and_reply\n executor_callback))\n', ' File "/usr/lib/python2.7/site-packages/oslo_messaging/rpc/dispatcher.py", line 186, in _dispatch\n executor_callback)\n', ' File "/usr/lib/python2.7/site-packages/oslo_messaging/rpc/dispatcher.py", line 130, in _do_dispatch\n result = func(ctxt, **new_args)\n', ' File "/usr/lib/python2.7/site-packages/osprofiler/pro | 21:42 |
sdake | filer.py", line 105, in wrapper\n return f(*args, **kwargs)\n', ' File "/usr/lib/python2.7/site-packages/cinder/volume/manager.py", line 1086, in initialize_connection\n raise exception.VolumeBackendAPIException(data=err_msg)\n', 'VolumeBackendAPIException: Bad or unexpected response from the storage volume backend API: Create export for volume failed.\n'] | 21:42 |
*** jtriley has quit IRC | 21:51 | |
dperaza | I added some comments on the patch hopefully it works, if you want I can send a patch too | 21:51 |
dperaza | sdake: looks like an issue with creating the iscsi target, you can try to create manually with cinder-rtstool | 21:54 |
dperaza | and see what the issue is | 21:54 |
harmw | [Errno 14] curl#7 - "Failed connect to 209.132.178.14:80; Connection refused | 21:55 |
harmw | argh! @ delorean repo | 21:55 |
*** bmace has joined #kolla | 21:55 | |
*** dperaza has quit IRC | 22:03 | |
*** dperaza has joined #kolla | 22:04 | |
*** bmace has quit IRC | 22:06 | |
*** bmace has joined #kolla | 22:13 | |
*** shardy has quit IRC | 22:14 | |
*** prad has quit IRC | 22:16 | |
*** britthouser has quit IRC | 22:18 | |
dperaza | sdake: looking at master I think cinder-volume is missing a few things | 22:22 |
dperaza | sending a patch to show | 22:29 |
*** dims_ has joined #kolla | 22:32 | |
*** bradjones has quit IRC | 22:33 | |
*** dims___ has quit IRC | 22:35 | |
*** shakamunyi has quit IRC | 22:45 | |
*** dims__ has joined #kolla | 22:46 | |
*** blahRus has quit IRC | 22:46 | |
*** jtriley has joined #kolla | 22:47 | |
*** sdake has quit IRC | 22:48 | |
*** dims_ has quit IRC | 22:49 | |
openstackgerrit | David Peraza proposed stackforge/kolla: Adding tgt config to cinder-volume container https://review.openstack.org/201361 | 22:52 |
openstackgerrit | David Peraza proposed stackforge/kolla: Adding tgt config to cinder-volume container https://review.openstack.org/201361 | 22:54 |
*** dims__ has quit IRC | 22:56 | |
*** britthouser has joined #kolla | 22:57 | |
*** sdake has joined #kolla | 22:57 | |
sdake | dperaza cool i'll take a patch ;) | 22:57 |
*** dperaza has quit IRC | 23:00 | |
*** sdake_ has joined #kolla | 23:00 | |
*** bmace has quit IRC | 23:00 | |
*** bmace has joined #kolla | 23:02 | |
*** sdake has quit IRC | 23:04 | |
*** diogogmt has quit IRC | 23:08 | |
sdake_ | rhallisey around? | 23:08 |
rhallisey | yo | 23:12 |
rhallisey | sdake_, saw the cinder addition | 23:12 |
rhallisey | also formally in my patch set | 23:12 |
rhallisey | :/ | 23:12 |
sdake_ | ok well if that was in your patch set lets get it merged | 23:13 |
sdake_ | can you ack it pls | 23:13 |
sdake_ | perhaps that is the last thing we require | 23:13 |
sdake_ | https://review.openstack.org/#/c/201361/ | 23:14 |
rhallisey | somewhere.. | 23:15 |
rhallisey | ya wfm | 23:15 |
*** jtriley has quit IRC | 23:18 | |
*** jtriley has joined #kolla | 23:22 | |
sdake_ | i am thinking we need to start iscsid as well | 23:25 |
*** jruano has quit IRC | 23:27 | |
*** achanda has quit IRC | 23:31 | |
*** achanda has joined #kolla | 23:32 | |
*** achanda has quit IRC | 23:36 | |
*** sdake has joined #kolla | 23:37 | |
*** sdake_ has quit IRC | 23:41 | |
*** dims_ has joined #kolla | 23:42 | |
*** jruano has joined #kolla | 23:44 | |
rhallisey | sdake, ya the supervisor.conf in patch ~17 I think is what we need | 23:46 |
sdake | i have no idea what your talking about | 23:46 |
sdake | got a link? | 23:46 |
openstackgerrit | Merged stackforge/kolla: Adding tgt config to cinder-volume container https://review.openstack.org/201361 | 23:46 |
sdake | does that patch 17 have the gt config that just merged? | 23:47 |
sdake | tgt | 23:47 |
*** achanda has joined #kolla | 23:49 | |
*** bradjones has joined #kolla | 23:58 | |
*** bradjones has quit IRC | 23:58 | |
*** bradjones has joined #kolla | 23:58 |
Generated by irclog2html.py 2.14.0 by Marius Gedminas - find it at mg.pov.lt!