| *** achanda_ has joined #kolla | 00:02 | |
| *** achanda has quit IRC | 00:05 | |
| *** achanda_ has quit IRC | 00:07 | |
| *** jasonsb has quit IRC | 00:10 | |
| *** jasonsb has joined #kolla | 00:11 | |
| *** banix has quit IRC | 00:19 | |
| *** rhallisey has quit IRC | 00:43 | |
| *** jasonsb has quit IRC | 00:43 | |
| *** jasonsb has joined #kolla | 00:44 | |
| *** dims has joined #kolla | 00:46 | |
| *** dims_ has quit IRC | 00:47 | |
| *** jasonsb has quit IRC | 00:48 | |
| *** dims has quit IRC | 00:56 | |
| *** dims has joined #kolla | 00:56 | |
| *** banix has joined #kolla | 01:11 | |
| *** weiyu has joined #kolla | 01:14 | |
| *** SourabhP has quit IRC | 01:17 | |
| *** SourabhP has joined #kolla | 01:17 | |
| *** erkules_ has joined #kolla | 01:23 | |
| *** erkules has quit IRC | 01:26 | |
| *** weiyu has quit IRC | 01:27 | |
| *** weiyu has joined #kolla | 01:30 | |
| *** banix has quit IRC | 01:32 | |
| *** Haomeng has joined #kolla | 01:46 | |
| *** britthouser has joined #kolla | 01:48 | |
| *** britthou_ has joined #kolla | 01:50 | |
| *** britthouser has quit IRC | 01:53 | |
| *** loth has quit IRC | 01:58 | |
| *** diogogmt has joined #kolla | 02:11 | |
| *** jasonsb has joined #kolla | 02:28 | |
| *** achanda has joined #kolla | 02:32 | |
| *** unicell has quit IRC | 02:33 | |
| *** Windir has quit IRC | 02:49 | |
| SamYaple | pbourke: how do you mean? each hostvars? | 02:50 |
|---|---|---|
| SamYaple | like a loop | 02:50 |
| openstackgerrit | Sam Yaple proposed stackforge/kolla: Remove start.yml abstraction for kolla-ansible https://review.openstack.org/213600 | 02:53 |
| *** Windir has joined #kolla | 02:53 | |
| *** dims has quit IRC | 02:55 | |
| SamYaple | sdake: ping | 02:56 |
| *** weiyu has quit IRC | 03:18 | |
| openstackgerrit | Sam Yaple proposed stackforge/kolla: Move defaults.yml to group_vars/all.yml https://review.openstack.org/213961 | 03:19 |
| openstackgerrit | Sam Yaple proposed stackforge/kolla: Move lots of global.yml options https://review.openstack.org/214200 | 03:19 |
| *** weiyu has joined #kolla | 03:21 | |
| openstackgerrit | Merged stackforge/kolla: Cinder bootstrap container is not initializing the DB https://review.openstack.org/213845 | 03:22 |
| *** diogogmt has quit IRC | 03:42 | |
| *** shardy_ has joined #kolla | 03:47 | |
| *** unicell has joined #kolla | 03:48 | |
| *** shardy has quit IRC | 03:49 | |
| *** diga has joined #kolla | 03:49 | |
| *** shardy_ has quit IRC | 03:52 | |
| *** shardy has joined #kolla | 03:53 | |
| openstackgerrit | Steven Dake proposed stackforge/kolla: Remove data containers and adjust compose files https://review.openstack.org/212830 | 04:00 |
| openstackgerrit | Steven Dake proposed stackforge/kolla: Use one data container for all data volumes https://review.openstack.org/212804 | 04:00 |
| sdake | samyaple shoot | 04:00 |
| SamYaple | globals.yml and comments | 04:03 |
| SamYaple | just so were clear, those comments are essentially docs | 04:03 |
| SamYaple | that means they will probably get out of date | 04:03 |
| SamYaple | Additionally I dont want to maintain docs in two locations for where options are in both globals.yml and all.yml | 04:04 |
| sdake | rebasing ftl | 04:06 |
| sdake | agree they are docs | 04:08 |
| sdake | hopefully they dont need to change, except to improve | 04:08 |
| SamYaple | i agree, but lets please take that approach | 04:08 |
| sdake | i'm going to try to make sphinx publish our defaults | 04:08 |
| sdake | as in the docs will come right out of etc/kolla/globals.yml | 04:09 |
| sdake | other projects do it with ini files, should be possible with yaml | 04:09 |
| *** weiyu has quit IRC | 04:17 | |
| *** diga has quit IRC | 04:26 | |
| sdake | coolsvap|away priority #1 is gnocchi templating so we can finish the job on the dockerfile-templates | 04:27 |
| sdake | thats the last container set needed | 04:27 |
| sdake | your signed up for it, if you want to pass it along please do so ;) | 04:27 |
| sdake | ok bedtime | 04:27 |
| *** sdake has quit IRC | 04:29 | |
| *** diga has joined #kolla | 04:35 | |
| *** coolsvap|away is now known as coolsvap | 04:46 | |
| *** weiyu has joined #kolla | 04:53 | |
| *** coolsvap is now known as coolsvap|away | 05:08 | |
| *** weiyu has quit IRC | 05:16 | |
| openstackgerrit | Sam Yaple proposed stackforge/kolla: Add Dockerfile template for kolla-ansible https://review.openstack.org/213394 | 05:20 |
| *** coolsvap|away is now known as coolsvap | 05:23 | |
| openstackgerrit | Swapnil Kulkarni (coolsvap) proposed stackforge/kolla: Add Dockerfile template for Gnocchi https://review.openstack.org/214452 | 05:24 |
| coolsvap | SamYaple, can you have a quick look at http://www.slideshare.net/SwapnilKulkarni9/project-kolla-bof-at-containercon | 05:25 |
| SamYaple | coolsvap: mission statement is outdated | 05:27 |
| coolsvap | I picked it up from the etherpad | 05:27 |
| SamYaple | either way its outdated | 05:28 |
| coolsvap | can you point me to the correct one? the review? | 05:29 |
| SamYaple | i dont really agree with the structure of that kolla architecture page but meh | 05:29 |
| SamYaple | https://review.openstack.org/#/c/209343/1/gerrit/projects.yaml | 05:29 |
| SamYaple | multinode and single node are both done and have been for a while | 05:30 |
| openstackgerrit | Sam Yaple proposed stackforge/kolla: Remove barbican https://review.openstack.org/214453 | 05:31 |
| coolsvap | yes, but i we are still testing it, so just added it as in progress | 05:32 |
| SamYaple | i only mention it since you said single node was done | 05:32 |
| SamYaple | there is no difference between the two | 05:32 |
| coolsvap | hmm, makes sense will update it a bit and i forgot to add the vagrant part | 05:33 |
| openstackgerrit | Sam Yaple proposed stackforge/kolla: Add Dockerfile template for nova https://review.openstack.org/213478 | 05:42 |
| *** weiyu has joined #kolla | 06:01 | |
| *** coolsvap is now known as coolsvap|away | 06:10 | |
| *** britthou_ has quit IRC | 06:21 | |
| *** britthouser has joined #kolla | 06:22 | |
| openstackgerrit | Sam Yaple proposed stackforge/kolla: Remove mysql-server from base https://review.openstack.org/214471 | 06:28 |
| *** achanda has quit IRC | 06:34 | |
| *** SourabhP has joined #kolla | 06:37 | |
| *** SourabhP_ has joined #kolla | 06:40 | |
| *** SourabhP has quit IRC | 06:42 | |
| *** SourabhP_ is now known as SourabhP | 06:42 | |
| *** dwalsh has joined #kolla | 07:06 | |
| *** inc0 has joined #kolla | 07:07 | |
| inc0 | good morning | 07:08 |
| inc0 | yay, docker 1.8 | 07:09 |
| SamYaple | hey inc0 you still working on logging? | 07:09 |
| SamYaple | i see the blueprints still in my name | 07:09 |
| inc0 | we were to coop there, you've said you have ideas | 07:09 |
| SamYaple | yea im just swamped | 07:10 |
| SamYaple | basically we just need an rsyslog container I believe | 07:10 |
| inc0 | yeah, that's not a problem, do you want me to take over alltogether? | 07:10 |
| SamYaple | yea sure. i thought that was the plan when we talked about coop, you leading and all | 07:11 |
| inc0 | misunderstaning then | 07:11 |
| SamYaple | no problem | 07:11 |
| inc0 | anyway, I'll take over then | 07:12 |
| SamYaple | im hammering the ubunt ustuff out right now | 07:13 |
| inc0 | ok | 07:14 |
| inc0 | SamYaple, do we want to have syslog on each container? It seems wrong. How else do we want to solve this? | 07:18 |
| inc0 | mount volume and log to volume, then read from volume in rsyslog container? | 07:19 |
| SamYaple | yea | 07:20 |
| SamYaple | thats what we had talked about I believe | 07:20 |
| SamYaple | but no | 07:20 |
| SamYaple | I think we should have each service log directly to a remote syslog | 07:20 |
| SamYaple | or have the /dev/ log device shared | 07:20 |
| SamYaple | lots of options | 07:20 |
| inc0 | yeah, let's pick one | 07:21 |
| SamYaple | sharing the dev device is probably the best choice | 07:21 |
| SamYaple | since it would appear as a log device and things would write straight to it | 07:22 |
| inc0 | ok, so we'll need to add this line to every ansible role right? | 07:22 |
| *** dolpher has joined #kolla | 07:23 | |
| inc0 | anyway, I'll think this through | 07:23 |
| inc0 | will start from rsyslog tho | 07:23 |
| *** weiyu has quit IRC | 07:24 | |
| SamYaple | yea we need an rsyslog container first | 07:24 |
| SamYaple | then sharing the dev will require some thought | 07:24 |
| *** weiyu_ has joined #kolla | 07:26 | |
| inc0 | brb | 07:29 |
| *** SourabhP has quit IRC | 07:33 | |
| *** gfidente has joined #kolla | 07:35 | |
| *** dolpher1 has joined #kolla | 07:45 | |
| *** dolpher has quit IRC | 07:47 | |
| *** dwalsh has quit IRC | 08:05 | |
| *** jmccarthy has joined #kolla | 08:16 | |
| *** inc0 has quit IRC | 08:23 | |
| *** akwasnie has joined #kolla | 08:23 | |
| *** inc0 has joined #kolla | 08:43 | |
| *** achanda has joined #kolla | 08:48 | |
| pbourke | SamYaple: hey. not sure if you saw some of the conversation from earlier, I was basically trying to figure out how we can maintain host specific vars and so far it looks like hostvars would be the place to access that | 08:55 |
| pbourke | the specific host vars I need are things like the list of devices on that host that should be added to the swift ring, the weights for each, etc. | 08:56 |
| SamYaple | pbourke: hmmm | 08:59 |
| SamYaple | i dont think thats such a good way t odo it | 08:59 |
| pbourke | SamYaple: so far it's been a bit tricky but it may well be my lack of experience with ansible | 08:59 |
| SamYaple | thats the way OSAD does it, its... bad | 08:59 |
| pbourke | id really welcome suggestions | 08:59 |
| SamYaple | wel lthe way i did ceph was dynamically | 08:59 |
| pbourke | Im not familiar with ceph at all, did it have a similar situation? | 09:00 |
| SamYaple | yea im looking for the code | 09:00 |
| SamYaple | basically iy has disks that need ot be mounted | 09:00 |
| SamYaple | it* | 09:00 |
| SamYaple | yea so i collect 'osds' as a variable | 09:01 |
| SamYaple | it has all initialized and unintiazed volumes | 09:01 |
| pbourke | I was thinking Id still like to use the data vol for an 'out of the box' kind of setup | 09:02 |
| SamYaple | I wrote this module to do that | 09:02 |
| SamYaple | https://github.com/SamYaple/yaodu/blob/master/ansible/library/ceph_osd_list | 09:02 |
| pbourke | just to help with demos/testing etc | 09:02 |
| SamYaple | i have no opinion on that | 09:02 |
| SamYaple | Swift won't ever be default so it doesnt have to be simple to setup | 09:02 |
| SamYaple | like one click deploy simple i mena | 09:02 |
| pbourke | hmm | 09:04 |
| SamYaple | here is what I dont want to see, manually specified disks for each thing in swift | 09:05 |
| SamYaple | that might work with 5 nodes with 4 disks each, but try 100 nodes with 20 disks | 09:05 |
| SamYaple | its not a good plan | 09:05 |
| pbourke | the thing is, do we need to allow for ops to specify that | 09:07 |
| SamYaple | yea this is a ops project. ops rules here | 09:07 |
| SamYaple | that doesnt mean it can't be dynamic | 09:07 |
| pbourke | yeah we need both | 09:08 |
| SamYaple | what do you mean both? | 09:08 |
| SamYaple | the manually specified is not a good thing at all | 09:08 |
| pbourke | so the values would be done dynamically but the option to override is there | 09:09 |
| pbourke | something like {{ hostvars[host][weight] | default(1) }} | 09:09 |
| SamYaple | If the disks are looked up dynamically im not sure how the override could ever happen | 09:09 |
| *** jmccarthy has quit IRC | 09:09 | |
| *** jmccarthy has joined #kolla | 09:09 | |
| pbourke | the disks are discovered dynamically but the values for each should be overridable | 09:10 |
| pbourke | weight, zone, etc | 09:10 |
| SamYaple | Lets take a step back for a second. Why are we trying to solve swift? | 09:10 |
| SamYaple | Why not make the operator generate the rings and we just distribute? | 09:10 |
| pbourke | that did cross my mind | 09:10 |
| SamYaple | Then this issue goes away and the operator can rely on official documentation | 09:10 |
| SamYaple | I can assure you we are OK with that too | 09:10 |
| pbourke | sdake kind of said he'd like it to work out of the box | 09:11 |
| SamYaple | Im sure he does | 09:11 |
| SamYaple | one click and all | 09:11 |
| SamYaple | but thats not how this stuff works | 09:11 |
| SamYaple | its swift | 09:11 |
| SamYaple | its complicated and requires raw disks | 09:11 |
| SamYaple | it wont work 'out of the box' | 09:11 |
| pbourke | yeah. maybe for the first iteration at least we dont have to solve this | 09:11 |
| pbourke | I'll look again and see what I come up with over the next few hours | 09:12 |
| pbourke | it sucks because you can't test without rings and they're such a manual process | 09:13 |
| SamYaple | but its not on us ot fix that | 09:13 |
| SamYaple | thats a swift thing | 09:13 |
| pbourke | agree | 09:13 |
| SamYaple | what if we do 'fix it' and swift goes a different direction? now we have this wierd non-standard thing | 09:13 |
| *** jmccarthy has quit IRC | 09:17 | |
| *** achanda has quit IRC | 09:17 | |
| SamYaple | you know this problem goes away when you back swift with ceph :P | 09:18 |
| *** dolpher1 has quit IRC | 09:35 | |
| *** jmccarthy has joined #kolla | 09:36 | |
| openstackgerrit | Merged stackforge/kolla: Use one data container for all data volumes https://review.openstack.org/212804 | 09:52 |
| *** tfukushima has joined #kolla | 09:55 | |
| *** apmelton has quit IRC | 10:06 | |
| *** apmelton has joined #kolla | 10:09 | |
| pbourke | SamYaple: wrt https://review.openstack.org/#/c/214471/1 | 10:15 |
| pbourke | I have a feeling some of the source images need the server package to build | 10:15 |
| *** tfukushima has quit IRC | 10:15 | |
| pbourke | might be wrong | 10:15 |
| *** tfukushima has joined #kolla | 10:16 | |
| SamYaple | on the ubuntu side ther need python-mysqldb | 10:17 |
| SamYaple | on the centos side... idk? | 10:17 |
| SamYaple | but either way this is bad since the mariadb repo isnt added until the mariadb container | 10:18 |
| SamYaple | im rewritting the ubuntu stuff so it actually works | 10:18 |
| SamYaple | pbourke: i put it as wip | 10:19 |
| pbourke | kk | 10:19 |
| openstackgerrit | Merged stackforge/kolla: Move defaults.yml to group_vars/all.yml https://review.openstack.org/213961 | 10:25 |
| openstackgerrit | Merged stackforge/kolla: Move lots of global.yml options https://review.openstack.org/214200 | 10:31 |
| openstackgerrit | Sam Yaple proposed stackforge/kolla: Remove defaults.yml from docs https://review.openstack.org/214147 | 10:35 |
| openstackgerrit | Sam Yaple proposed stackforge/kolla: Remove barbican https://review.openstack.org/214453 | 10:43 |
| inc0 | SamYaple, remove barbican...lol | 10:53 |
| SamYaple | im a man of few words | 10:54 |
| inc0 | +1 from me on that | 10:54 |
| inc0 | anyone uses barbican anyway? | 10:54 |
| inc0 | although | 10:56 |
| *** dims has joined #kolla | 10:56 | |
| inc0 | after a thought...that looks like perfect contrib container for me | 10:56 |
| SamYaple | it doesnt work | 10:57 |
| SamYaple | it needs to be split into 4 containers | 10:57 |
| SamYaple | it never worked | 10:57 |
| inc0 | whatever, code is there | 10:57 |
| inc0 | don't throw it away | 10:57 |
| inc0 | its contrib, it's just...there | 10:57 |
| inc0 | use it on your own risk | 10:57 |
| inc0 | maybe someone actually would like to fix that | 10:57 |
| SamYaple | But i guess thats my point, if it never worked then what code is there is nothing like it should be | 10:58 |
| inc0 | then let it go to contrib to die | 10:58 |
| SamYaple | dod you know if you ^C and then hit enter in a docker contaienr it wil lsay 'Eh?' | 10:58 |
| inc0 | oO | 11:00 |
| *** dims_ has joined #kolla | 11:03 | |
| openstackgerrit | Sam Yaple proposed stackforge/kolla: Fix small typo in task name https://review.openstack.org/214559 | 11:05 |
| *** dims has quit IRC | 11:07 | |
| *** britthouser has quit IRC | 11:10 | |
| SamYaple | pbourke: would you mind reviewing https://review.openstack.org/#/c/211442/ | 11:15 |
| *** weiyu_ has quit IRC | 11:17 | |
| SamYaple | thanks man | 11:18 |
| SamYaple | Also if you wouldn't mind changing your vote on https://review.openstack.org/#/c/213480/ since I do have that patch in place for docs | 11:19 |
| openstackgerrit | Ryan Hallisey proposed stackforge/kolla: Libvirt is moving around the wrong config file https://review.openstack.org/214241 | 11:24 |
| *** dims has joined #kolla | 11:29 | |
| *** dims_ has quit IRC | 11:31 | |
| *** dims_ has joined #kolla | 11:31 | |
| pbourke | SamYaple: sure, but I dont see a DocImpact flag there? | 11:34 |
| *** dims has quit IRC | 11:34 | |
| SamYaple | sdake never set it up | 11:35 |
| SamYaple | the patch is already +2+1 | 11:35 |
| SamYaple | jpeeler: ping | 11:40 |
| *** tfukushima has quit IRC | 11:42 | |
| *** diogogmt has joined #kolla | 11:46 | |
| *** diga has quit IRC | 11:55 | |
| *** dwalsh has joined #kolla | 12:01 | |
| *** jmccarthy has quit IRC | 12:07 | |
| *** jmccarthy1 has joined #kolla | 12:07 | |
| *** jpeeler has quit IRC | 12:07 | |
| openstackgerrit | Sam Yaple proposed stackforge/kolla: Remove Ansible abstraction for Neutron https://review.openstack.org/211442 | 12:17 |
| *** diogogmt has quit IRC | 12:20 | |
| *** rhallisey has joined #kolla | 12:21 | |
| *** britthouser has joined #kolla | 12:28 | |
| openstackgerrit | Michal Jastrzebski (inc0) proposed stackforge/kolla: rsyslog container https://review.openstack.org/214589 | 12:33 |
| inc0 | soo...now let's talk - what do with logs once they land there | 12:34 |
| pbourke | inc0: it seems there's additional work before that discussion takes place? | 12:40 |
| pbourke | inc0: as right now they're not landing anyway | 12:40 |
| pbourke | s/anyway/anywhere | 12:40 |
| *** tfukushima has joined #kolla | 12:43 | |
| *** tfukushima has quit IRC | 12:48 | |
| inc0 | well, in containers /var/log/messages | 12:49 |
| openstackgerrit | Paul Bourke proposed stackforge/kolla: Add ansible role for memcached https://review.openstack.org/214593 | 12:50 |
| inc0 | anyway yeah, it's wip | 12:50 |
| inc0 | we need to discuss what to do with logs and how to push logs there in the first place | 12:51 |
| inc0 | also how to setup rest of logging stack (logstash, kibana...) | 12:51 |
| pbourke | inc0: ah yeah. I would think most would just need the correct config settings to log to the remote rsyslogd | 12:51 |
| inc0 | in every container | 12:51 |
| inc0 | and ansible roles to point to correct rsyslog | 12:51 |
| inc0 | this container is supposed to be this "remote rsyslog";) | 12:52 |
| openstackgerrit | Paul Bourke proposed stackforge/kolla: Add ansible role for memcached https://review.openstack.org/214593 | 12:54 |
| *** erkules_ is now known as erkules | 13:01 | |
| *** erkules has joined #kolla | 13:01 | |
| *** banix has joined #kolla | 13:14 | |
| *** dims_ has quit IRC | 13:29 | |
| *** tfukushima has joined #kolla | 13:29 | |
| *** dims has joined #kolla | 13:29 | |
| *** diogogmt has joined #kolla | 13:43 | |
| *** jpeeler has joined #kolla | 13:45 | |
| *** diogogmt has quit IRC | 13:47 | |
| jpeeler | SamYaple: hey | 13:48 |
| *** jtriley has joined #kolla | 13:48 | |
| *** weiyu has joined #kolla | 13:51 | |
| openstackgerrit | Merged stackforge/kolla: Libvirt is moving around the wrong config file https://review.openstack.org/214241 | 13:56 |
| *** sdake has joined #kolla | 13:56 | |
| *** sdake_ has joined #kolla | 13:58 | |
| openstackgerrit | Sam Yaple proposed stackforge/kolla: Fix up base, RabbitMQ, and MariaDB for Ubuntu https://review.openstack.org/214625 | 13:59 |
| sdake_ | oh glorious 9 hours of sleep | 13:59 |
| sdake_ | morning folks | 13:59 |
| SamYaple | sdake_: cool! now get to work | 13:59 |
| * SamYaple cracks the whip | 14:00 | |
| sdake_ | samyaple have you tried rebuilding all your images from sratch | 14:00 |
| sdake_ | i think it doesn't work if you do | 14:00 |
| SamYaple | sdake_: yea i mentioned neutron broke | 14:00 |
| sdake_ | it being kolla | 14:00 |
| SamYaple | only on the new images | 14:00 |
| sdake_ | ya i fixed that neutron bug | 14:00 |
| sdake_ | i am getting nms in the "ERROR" state | 14:00 |
| sdake_ | on nva boot | 14:01 |
| SamYaple | ive been hammering on ubuntu so i havent tried booting recently | 14:01 |
| sdake_ | ok | 14:01 |
| sdake_ | i'll see if i can sort it out today | 14:01 |
| sdake_ | see if its a regressoin in our code or upstream rdo | 14:01 |
| *** sdake has quit IRC | 14:01 | |
| openstackgerrit | Sam Yaple proposed stackforge/kolla: Fix group name in vagrant docs https://review.openstack.org/213950 | 14:02 |
| openstackgerrit | Sam Yaple proposed stackforge/kolla: Update support to control https://review.openstack.org/213480 | 14:03 |
| akwasnie | hi | 14:03 |
| SamYaple | hey akwasnie | 14:03 |
| sdake_ | hey akwasnie | 14:03 |
| akwasnie | sdake: https://bugs.launchpad.net/kolla/+bug/1485742 in this bug you want to replace all /opt/kolla occurences with /var/lib/kolla so that files like config-external will be copied to /var/lib/kolla ? | 14:04 |
| openstack | Launchpad bug 1485742 in kolla "global search and replace /opt/kolla with /var/lib/kolla" [Critical,Confirmed] - Assigned to Alicja Kwasniewska (alicja-kwasniewska) | 14:04 |
| sdake_ | akwasnie right | 14:05 |
| akwasnie | I can do it if u don't mind | 14:05 |
| SamYaple | sdake_: no | 14:05 |
| SamYaple | those are configs | 14:05 |
| SamYaple | those go in /etc | 14:05 |
| SamYaple | the templates can go to /var/lib/kolla | 14:05 |
| sdake_ | the tricky part is wha tto do about kolla-common.sh | 14:06 |
| *** weiyu has quit IRC | 14:06 | |
| sdake_ | kolla-ocmmon.sh and config-external.sh should probably go in /usr/share/kolla | 14:06 |
| SamYaple | oh you mean the container stuff? | 14:06 |
| SamYaple | or the ansible dirs? | 14:06 |
| sdake_ | welll basically we shouldn't be using /opt | 14:06 |
| SamYaple | agreed | 14:07 |
| sdake_ | so we might as well get it right now ;) | 14:07 |
| SamYaple | but what yare you talking about | 14:07 |
| sdake_ | there is a bug i filed to search and replace /opt/kolla with /var/lib/kolla | 14:07 |
| *** inc0 has quit IRC | 14:08 | |
| sdake_ | but i really link shell scripts should end in /usr/share/kolla | 14:08 |
| sdake_ | think | 14:08 |
| SamYaple | Ok, I have the Ansible directory changes (where it lays down configs) handled under the update-configs blueprint | 14:08 |
| SamYaple | this bug is not related to what im talking about | 14:08 |
| SamYaple | I have no strong opinion on this | 14:08 |
| *** bmace has joined #kolla | 14:09 | |
| sdake_ | i just think we should be following lsb | 14:09 |
| sdake_ | otherwise people will think our implementation is "out of place" | 14:09 |
| SamYaple | well that means /etc for configs | 14:09 |
| sdake_ | the generated configs? | 14:09 |
| sdake_ | for each node? | 14:09 |
| SamYaple | yup | 14:09 |
| SamYaple | those arent required to be generated for one (this was in the spec even if we dont have al lthe code in place) | 14:10 |
| sdake_ | those are more like persistent data i think | 14:10 |
| sdake_ | because they are generated | 14:10 |
| SamYaple | read above | 14:10 |
| SamYaple | and for COPY_ALWAYS those configs are designed to change | 14:10 |
| sdake_ | ok let me wake up a bit | 14:10 |
| sdake_ | and then we can discuss | 14:10 |
| SamYaple | For those of you who care about Ubuntu, weigh in now https://review.openstack.org/#/c/214625/ | 14:12 |
| sdake_ | here is the ops meetup etherpad | 14:13 |
| sdake_ | https://etherpad.openstack.org/p/PAO-ops-meetup | 14:13 |
| *** sdake_ is now known as sdake | 14:13 | |
| *** jtriley has quit IRC | 14:13 | |
| sdake | https://etherpad.openstack.org/p/PAO-ops-containers-for-deployment | 14:13 |
| sdake | samyaple that is worth a read | 14:15 |
| sdake | above | 14:15 |
| sdake | requirements in there | 14:15 |
| sdake | it's what the cool kids are doing | 14:15 |
| sdake | lol | 14:15 |
| *** thumpba has joined #kolla | 14:15 | |
| SamYaple | 'data outside the container' | 14:18 |
| SamYaple | told you ops would want that | 14:18 |
| sdake | well duh | 14:18 |
| SamYaple | Ha you say that now and yet I have logs of you fighting me! | 14:18 |
| sdake | i'm pretty sure i prposed that in the ansible-multi spe ;) | 14:18 |
| sdake | not fighting grilling so icould write that spec | 14:19 |
| sdake | iirc | 14:19 |
| SamYaple | after some prodding.... | 14:19 |
| sdake | question | 14:19 |
| SamYaple | not how I recall it but I also dont care so proceed | 14:19 |
| sdake | what nova containers require privilege | 14:19 |
| SamYaple | the ones that have "privileged: True" | 14:19 |
| sdake | yes | 14:19 |
| SamYaple | :) | 14:19 |
| sdake | you mean look at the code | 14:19 |
| sdake | so tired | 14:20 |
| SamYaple | rtfm | 14:20 |
| sdake | i am pretty sure nova-compute doesn't require privileged container | 14:21 |
| SamYaple | you can say that, but youd be wrong! | 14:21 |
| SamYaple | loading modules requires CAPS | 14:21 |
| SamYaple | we can isolate said CAPs, but I would focus on dropping to a user before I do that | 14:22 |
| SamYaple | in M I will be pushing al lthe containers to drop privileges and run as unpriveleged in the containers | 14:22 |
| SamYaple | and hopefullly Isolating caps | 14:22 |
| SamYaple | I spent about a month on yaodu doing just that | 14:22 |
| SamYaple | sdake: https://review.openstack.org/#/c/211442/ please merge that in before I have t orebase AGAIN | 14:23 |
| SamYaple | I have had t orebase 5 patched :( | 14:23 |
| *** absubram has joined #kolla | 14:23 | |
| SamYaple | pbourke: let me know if you need any ansible helps with memcached/swift. I can provide guidance should you need it | 14:27 |
| pbourke | SamYaple: thanks. I think I'm going to submit a WIP in the next hour or two just without the ring generation | 14:27 |
| pbourke | get feedback at least on what's there | 14:27 |
| sdake | ya rebasing sucks | 14:27 |
| SamYaple | Ok guys I am heading out for a bit. be back for the team meeting | 14:29 |
| SamYaple | If anyone has an desire to influence Ubuntu, please look at https://review.openstack.org/#/c/214625/ | 14:29 |
| SamYaple | If not I am going t odo things my way! | 14:29 |
| sdake | what is missing today which would get you on the container path? | 14:31 |
| sdake | Time | 14:31 |
| sdake | kernel bugs | 14:31 |
| sdake | unprivilaged containers | 14:31 |
| sdake | less industry hype, more best practices | 14:31 |
| sdake | we will have to work on the unprivileged c ontainers in mitaka | 14:32 |
| sdake | identify minimum set of capacibiltiies to get the job done | 14:33 |
| sdake | i still dont think nova-compute rquires privileged but my system doesn't run so Ican verify ! | 14:33 |
| openstackgerrit | Merged stackforge/kolla: Remove Ansible abstraction for Neutron https://review.openstack.org/211442 | 14:34 |
| *** jtriley has joined #kolla | 14:39 | |
| *** sdake_ has joined #kolla | 14:41 | |
| sdake_ | damn out of disk space | 14:43 |
| sdake_ | i bet that is why nova isn'tworking | 14:43 |
| *** sdake has quit IRC | 14:44 | |
| sdake_ | samyaple can you confirm docker-py 1.1.0 wont work for us? | 14:49 |
| sdake_ | for building | 14:49 |
| sdake_ | that is what is in the global-requirements.txt file | 14:49 |
| *** diga has joined #kolla | 14:49 | |
| SamYaple | docker 1.1.0 wont work | 14:51 |
| sdake_ | ok | 14:51 |
| SamYaple | 1.2.0 is minimum | 14:51 |
| sdake_ | which featuere is needed (so I can put in commit log) | 14:51 |
| sdake_ | so i can get the global requirements updated | 14:51 |
| SamYaple | sdake_: nova-compute _does_ require privilieged because it needs to load modules | 14:51 |
| sdake_ | so we can get a requirements sync job | 14:52 |
| *** blahRus has joined #kolla | 14:53 | |
| openstackgerrit | Steven Dake proposed stackforge/kolla: The requirements.txt file isn't correct https://review.openstack.org/213398 | 14:53 |
| sdake_ | samyaple do you know which api feature is needed from docker 1.2.0? | 14:53 |
| SamYaple | sdake_: oh sorry | 14:53 |
| SamYaple | i see you say for building | 14:53 |
| SamYaple | i think were fine for building at 1.1.0 | 14:53 |
| SamYaple | i think | 14:53 |
| SamYaple | dunno | 14:54 |
| *** diogogmt has joined #kolla | 14:54 | |
| sdake_ | might as well try to get the job updated | 14:54 |
| *** dwalsh has quit IRC | 14:54 | |
| SamYaple | might need 1.2.0 to support docker 1.7.0 (which we dont need) | 14:54 |
| sdake_ | rather get the requirement updated | 14:54 |
| sdake_ | docker.1.7.0 is needed to bindmount /dev iiuc | 14:54 |
| SamYaple | yea but the reason that supposedly worked was taken out before 1.7.0 | 14:55 |
| SamYaple | But im not pushing it | 14:55 |
| SamYaple | were going to be requiring the newest docker soon enough I think | 14:55 |
| sdake_ | ok so back on topic :) I want to update global rquirements to docker 1.2.0 | 14:55 |
| sdake_ | what is my argument | 14:55 |
| sdake_ | (for the commit log) | 14:55 |
| sdake_ | if you dont know that is a fine answer, i'll make something up | 14:56 |
| sdake_ | just looking for anything to help along the review | 14:56 |
| *** pbourke has quit IRC | 14:57 | |
| *** pbourke has joined #kolla | 14:57 | |
| SamYaple | wait globals has an upperlimit on the docker-py? | 14:58 |
| sdake_ | no a lower limit | 14:58 |
| sdake_ | upper limits have been removed | 14:58 |
| SamYaple | oh thats fine leave that alone | 14:58 |
| openstackgerrit | Merged stackforge/kolla: Add oslo_versionedobjects dependency for neutron https://review.openstack.org/214362 | 14:59 |
| sdake_ | so what will happen when i set up the global sync job | 14:59 |
| sdake_ | is our requirements.txt will be changed by a script to docker-py > 1.1.0 | 14:59 |
| SamYaple | i know | 14:59 |
| SamYaple | thats fine for building | 14:59 |
| sdake_ | ok i'll change our requirements.txt to 11.0 then | 15:00 |
| openstackgerrit | Steven Dake proposed stackforge/kolla: The requirements.txt file isn't correct https://review.openstack.org/213398 | 15:00 |
| SamYaple | wont the update job do that for us? | 15:00 |
| SamYaple | proposalbot | 15:01 |
| sdake_ | once i have it in place | 15:01 |
| sdake_ | its not enabled atm | 15:01 |
| *** dims has quit IRC | 15:01 | |
| sdake_ | only so many hours in the day | 15:01 |
| sdake_ | i also dont know how to enable it | 15:01 |
| *** dims has joined #kolla | 15:01 | |
| SamYaple | har har har | 15:01 |
| openstackgerrit | Swapnil Kulkarni (coolsvap) proposed stackforge/kolla: Add Dockerfile template for Gnocchi https://review.openstack.org/214452 | 15:02 |
| sdake_ | i can figure itout | 15:02 |
| sdake_ | coolsvap|away thanks bro :) | 15:02 |
| openstackgerrit | Steven Dake proposed stackforge/kolla: The requirements.txt file isn't correct https://review.openstack.org/213398 | 15:03 |
| pbourke | has anyone seen issues with mariadb since the data container change? | 15:03 |
| pbourke | sdake_: ^ | 15:03 |
| SamYaple | pbourke: define issues | 15:03 |
| SamYaple | there are some tricky bits surrounding that | 15:04 |
| sdake_ | pbourke you must rebuild all your images | 15:04 |
| SamYaple | permissions wise | 15:04 |
| pbourke | http://paste.openstack.org/show/421959/ | 15:04 |
| pbourke | yeah permissions | 15:04 |
| sdake_ | sorry about the rebuid requirement | 15:04 |
| pbourke | no worries | 15:04 |
| sdake_ | its in the ocmmit log | 15:04 |
| SamYaple | pbourke: yea we rm -rf the /var/lib/mysql folder | 15:04 |
| SamYaple | or at least teh contents | 15:04 |
| sdake_ | no docker mounts /var/lib/mysql as root | 15:04 |
| SamYaple | anyway a rebuild and youll be fine | 15:04 |
| SamYaple | sdake_: not quite | 15:05 |
| *** dims_ has joined #kolla | 15:05 | |
| SamYaple | but end result is the same | 15:05 |
| sdake_ | pbourke the code in ansible is incompatible with the old images | 15:05 |
| sdake_ | it sucks but it is what it is | 15:05 |
| sdake_ | i'd like to stop making these breaking changes | 15:05 |
| sdake_ | and i thiink we are almost there, once we get /run in data containers | 15:05 |
| SamYaple | thats unlikely to happen... | 15:05 |
| sdake_ | samyaple you may be right | 15:06 |
| SamYaple | run in a data container im not so keen on, but I can come around on the subject | 15:06 |
| *** dims has quit IRC | 15:06 | |
| SamYaple | maybe | 15:06 |
| SamYaple | well discuss not in a position to atm | 15:06 |
| sdake_ | samyaple i'm not even sure if it would work | 15:06 |
| sdake_ | its more like a research project for someone | 15:07 |
| SamYaple | it would work for most things | 15:07 |
| sdake_ | probably not for libvirt | 15:07 |
| SamYaple | my issue is the /run default socket kinda blocks other apps from starting | 15:07 |
| sdake_ | i have a bug or blueprint up for it | 15:07 |
| sdake_ | which default socket | 15:07 |
| SamYaple | like you could have two different libvirt or openvswitch things going with different sockets, but the same kernel | 15:07 |
| SamYaple | that would be ba | 15:07 |
| sdake_ | ya libvirt needs to share /run host os because it communicates with systemd | 15:08 |
| sdake_ | there is no way around that from what i can tell | 15:08 |
| SamYaple | There are | 15:08 |
| SamYaple | but i dont know that this is worth it | 15:08 |
| sdake_ | you mean bindmount the specific run stuff needed by libvirt | 15:08 |
| SamYaple | no, but that would work too | 15:09 |
| sdake_ | rather then the whole directory | 15:09 |
| * sdake_ wtb someone to fix the keepalive pid bug | 15:09 | |
| SamYaple | what bug? | 15:09 |
| sdake_ | https://bugs.launchpad.net/kolla/+bug/1485079 | 15:14 |
| openstack | Launchpad bug 1485079 in kolla "if /run/keepalived.pid file exists, keepalive wont start" [Critical,Confirmed] | 15:14 |
| SamYaple | do you not properly stop your keepalived container? | 15:15 |
| SamYaple | if not then youre leaving your vip aroun | 15:15 |
| SamYaple | d* | 15:15 |
| SamYaple | This should be a check to cleanup though | 15:15 |
| SamYaple | still you shoudl stop the container rather than forcibly removing it | 15:16 |
| sdake_ | ok well if the container crashes for example | 15:16 |
| sdake_ | same outcome | 15:16 |
| SamYaple | yea but thats not why you hit that bug ;) | 15:16 |
| sdake_ | keepalived is busted, when you createa pid file yoursupposed to run fcntl to lock it | 15:16 |
| sdake_ | then if the process dies he lock si removed | 15:17 |
| sdake_ | if the file is llocked then you dont start, not if it exists | 15:17 |
| SamYaple | that doesnt work when you sigkill the process | 15:17 |
| SamYaple | and thats what you are doing when you remove the container like that | 15:17 |
| sdake_ | sigkilll doesn't remove the lock on the file? | 15:17 |
| SamYaple | not that I am aware | 15:18 |
| *** coolsvap|away is now known as coolsvap | 15:18 | |
| SamYaple | this is why all the daemons have a check/cleanup | 15:18 |
| sdake_ | how shoudl i cleanup between runs | 15:18 |
| sdake_ | if not container-cleanup -fv/ | 15:18 |
| sdake_ | do we have a script for that? | 15:19 |
| SamYaple | yea i dont use that cleanup script cause it bad | 15:19 |
| SamYaple | but I can fix it up for you | 15:19 |
| sdake_ | something to get me going would be fantastic ;) | 15:19 |
| SamYaple | stop the container, then remove it | 15:19 |
| *** dims_ has quit IRC | 15:19 | |
| *** dims has joined #kolla | 15:19 | |
| coolsvap | sdake_, hi | 15:19 |
| coolsvap | please have a look at http://www.slideshare.net/SwapnilKulkarni9/project-kolla-bof-at-containercon | 15:19 |
| sdake_ | sometimes i forget to rm the /run/keepalived pid file | 15:19 |
| sdake_ | and then all kinds of bad things happen | 15:19 |
| SamYaple | well if yo ustop the cotnainer, the file gets removed properly | 15:19 |
| SamYaple | I run this beast: | 15:20 |
| SamYaple | echo kolla_ansible log_data heat_{api{,_cfn},engine} mariadb{,_data} rabbitmq{,_data} keystone glance_{api,registry,data} nova_{data,scheduler,novncproxy,consoleauth,conductor,api,compute,libvirt} neutron_{server,agents,linuxbridge_agent,openvswitch_agent} openvswitch_{vswitchd,db} haproxy keepalived | xargs -P100 -n1 sh -c 'docker stop $1; docker rm -v -f $1' sh | 15:20 |
| SamYaple | thats one line btw | 15:20 |
| sdake_ | samyaple nice can you put that in clenaup-containers -fv ;) | 15:20 |
| sdake_ | rathre the clenaup-containers script | 15:20 |
| SamYaple | can you file a bug and assign it to me? | 15:20 |
| sdake_ | of course | 15:21 |
| SamYaple | I will do any bugs you assign to me | 15:21 |
| *** coolsvap is now known as coolsvap|away | 15:22 | |
| *** daneyon has joined #kolla | 15:22 | |
| SamYaple | hey sdake_ did you email your docker contact? we still dont have that namespace propgation in master | 15:22 |
| pbourke | still getting permission issues on rabbitmq | 15:23 |
| sdake_ | i talked to him | 15:23 |
| sdake_ | and he was keen to help us out | 15:24 |
| pbourke | I've rebuilt and 'docker rm' ed everything | 15:24 |
| SamYaple | pbourke: I have tested rabbitmq as working | 15:24 |
| sdake_ | pbourke ugh | 15:24 |
| SamYaple | is it folder permissions? | 15:24 |
| pbourke | SamYaple: how did you know it was rabbit :) | 15:24 |
| SamYaple | 15:24 < pbourke> still getting permission issues on rabbitmq | 15:25 |
| sdake_ | i wish i was young and could remember everything everyone says :) | 15:25 |
| SamYaple | i dont remember anything | 15:25 |
| SamYaple | I have logs and grep | 15:25 |
| sdake_ | pbourke try a build with --no-cache | 15:25 |
| sdake_ | pbourke areyou building -t binary? | 15:25 |
| SamYaple | I grep across all my IRC channels when I have an issue ;) | 15:25 |
| pbourke | I tend not to use the build.py as it builds everything | 15:26 |
| pbourke | i'll try no-cache | 15:26 |
| sdake_ | -t binary works | 15:26 |
| *** daneyon_ has quit IRC | 15:26 | |
| SamYaple | I built against the templates fyi | 15:26 |
| sdake_ | not sure about -t source, tbh i didn't test oracle linux with -t source | 15:26 |
| sdake_ | maybe i missed something with -t source | 15:26 |
| SamYaple | rabbitmq source and binary are teh same | 15:27 |
| sdake_ | cool so the templates are actually working? | 15:27 |
| SamYaple | its all ive been using | 15:27 |
| sdake_ | have you been able to nova boot? | 15:27 |
| SamYaple | havent tested recent | 15:27 |
| SamYaple | a couple days ago | 15:27 |
| SamYaple | fyi two main templates left | 15:27 |
| sdake_ | yesterday on master i was unable to nova boot, but i may had an out of disk space problem | 15:28 |
| SamYaple | https://review.openstack.org/#/c/213394/ | 15:28 |
| SamYaple | https://review.openstack.org/#/c/213478/ | 15:28 |
| sdake_ | reviewing | 15:28 |
| jpeeler | should there be a when check for enable_haproxy here? https://github.com/stackforge/kolla/blob/master/ansible/roles/mariadb/tasks/register.yml#L12 | 15:28 |
| sdake_ | git clone ansible.git so evil ;) | 15:28 |
| SamYaple | jpeeler: that is used as a check of sorts | 15:29 |
| SamYaple | to ensure the database is active | 15:29 |
| SamYaple | that user is not given any privleges at all | 15:29 |
| SamYaple | we can change the username to health_check probably | 15:30 |
| sdake_ | gate is maddening - need --retry asap!! | 15:30 |
| SamYaple | btw guys. read https://github.com/docker/docker/blob/c4d45b6a29a91f2fb5d7a51ac36572f2a9b295c6/docs/extend/plugins_volume.md | 15:31 |
| *** inc0 has joined #kolla | 15:35 | |
| *** britthouser has quit IRC | 15:35 | |
| SamYaple | what time is meeting today? | 15:36 |
| inc0 | in 30 mins Sam | 15:36 |
| SamYaple | thought so | 15:37 |
| SamYaple | we still ahvent updated the time to 1630 | 15:37 |
| sdake_ | i am sending out an announcement after this meeting for an a adjust in the scheduling | 15:37 |
| SamYaple | slacker | 15:38 |
| sdake_ | i'll talk more about it at our meeting today | 15:38 |
| inc0 | ah...damn I hoped it wont move before I move;) | 15:38 |
| sdake_ | i want to remove the 3pm pst meeting too | 15:38 |
| SamYaple | yea that seems like a must | 15:38 |
| SamYaple | unfortunately | 15:38 |
| sdake_ | not remove,but move to 9:30 | 15:38 |
| SamYaple | GMT man | 15:38 |
| SamYaple | UTC | 15:38 |
| sdake_ | ya exepcted apac contribs, and mandre was one, but he moved to france | 15:38 |
| sdake_ | everyone that contribs is in europe/us | 15:39 |
| inc0 | cool, I won't miss half of meetings | 15:39 |
| inc0 | guys...about logging | 15:40 |
| inc0 | are you comfortable with having logs on volume? | 15:40 |
| SamYaple | inc0: we talked a bit about this already. Where they spit out to a volume (which can be bind-mounted to the host) | 15:41 |
| SamYaple | so the logs are accessible from the host or a single container | 15:41 |
| SamYaple | setup in a way we can forward them to a logstash type server too | 15:42 |
| SamYaple | but thats not a requirement initially | 15:42 |
| inc0 | we will need them to be accessible from other containers as well | 15:42 |
| inc0 | either way, volume works | 15:42 |
| inc0 | I'll do this in separate, ansiblish patch | 15:42 |
| SamYaple | yea I dont know that I agree with teh other containers part | 15:42 |
| SamYaple | the way we talked about the containers themseleves wouldn't even have the log since its spit to a syslog container | 15:43 |
| *** dwalsh has joined #kolla | 15:43 | |
| SamYaple | right? | 15:43 |
| inc0 | yeah, I meant containers that can parse files | 15:43 |
| SamYaple | oh yea well that would be done via forwarding | 15:43 |
| inc0 | although if we'll push it to logstash, that would be redundant | 15:43 |
| sdake_ | all logs go to data container, logstash mounts data container and pushes to one place | 15:44 |
| inc0 | so we do data container instead of volume? | 15:45 |
| SamYaple | sdake_: no thats not going to work | 15:45 |
| sdake_ | lets discuss in meeting so there is a meeting log heh? :) | 15:45 |
| SamYaple | the logs go from the application to an rsyslog container | 15:45 |
| inc0 | yeah, and rsyslog pushes it to logstash container | 15:45 |
| inc0 | and elasticsearch container uses this logstash stuff | 15:46 |
| sdake_ | rsyslog container is good for me - as long as it ends up with all the files as well :) | 15:46 |
| inc0 | we'll need to make it work ofc | 15:46 |
| sdake_ | lets discuss in meeting | 15:46 |
| inc0 | kk | 15:46 |
| sdake_ | thanks to whoever updated the meeting agenda | 15:49 |
| * sdake_ been slacking on that | 15:49 | |
| *** dims has quit IRC | 15:49 | |
| *** sdake_ is now known as sdake | 15:49 | |
| *** dims has joined #kolla | 15:49 | |
| SamYaple | slacker | 15:51 |
| *** dwalsh has quit IRC | 15:52 | |
| inc0 | hmm there is logstash on dockerhub | 15:52 |
| inc0 | I'll give it a try, and with a bit of luck we will get this stuff for free | 15:53 |
| *** dims_ has joined #kolla | 15:53 | |
| inc0 | and just configure ansible/rsyslog to use it | 15:53 |
| sdake | i'd prefer any dependencies be in our repo (logstash) | 15:54 |
| SamYaple | inc0: well logstash wont be included by default | 15:54 |
| SamYaple | sdake: it depends on if its an official repo | 15:54 |
| sdake | meeting in 5 minutes - lets discuss then :) | 15:54 |
| *** dims has quit IRC | 15:54 | |
| openstackgerrit | Paul Bourke proposed stackforge/kolla: Fix missing line break in memcached template https://review.openstack.org/214683 | 15:54 |
| SamYaple | if logstash is pumping out an official container, I would like to use that | 15:54 |
| inc0 | yeah, let's not create brave new wheel | 15:55 |
| SamYaple | but only if its offical | 15:55 |
| jmccarthy1 | sdake: Hmm I'm bit concerned about swift bugs #1477993 and #1477990 .. I've noticed there is more involved in #1477993 than I first thought .. it's just under two weeks left in L3 is it ? What's up for discussion this meeting ? | 15:55 |
| openstack | bug 1477993 in kolla "kolla bp/swift-container: Fix up supporting swift container functionality" [Critical,Triaged] https://launchpad.net/bugs/1477993 - Assigned to James McCarthy (jmccarthy) | 15:55 |
| openstack | bug 1477990 in kolla "kolla bp/swift-container: Method for bind mounts with swift-data" [Critical,Triaged] https://launchpad.net/bugs/1477990 - Assigned to James McCarthy (jmccarthy) | 15:55 |
| sdake | the problem is distro | 15:55 |
| sdake | if the distro used to build is ubuntu and its running on centos i don't have high confidence it will work ;) | 15:56 |
| sdake | meeting in 4 minutees, we can discuss all these things jmccarthy1 | 15:56 |
| jmccarthy1 | Okie | 15:56 |
| SamYaple | sdake: if its a self contained app that talks over the network, it will work | 15:56 |
| sdake | would like in official log | 15:56 |
| SamYaple | but in this case it sint | 15:56 |
| SamYaple | anyway we dont have to worry about that now | 15:56 |
| pbourke | sdake: did you have a bug logged for oraclelinux template builds? | 15:57 |
| *** jmccarthy1 has quit IRC | 15:57 | |
| sdake | pbourke no | 15:57 |
| sdake | could you file one | 15:57 |
| pbourke | sdake: kk will file now | 15:57 |
| pbourke | have patch ready | 15:57 |
| *** jmccarthy has joined #kolla | 15:57 | |
| sdake | nice | 15:57 |
| pbourke | this is just for binary, will test source next | 15:58 |
| openstackgerrit | Paul Bourke proposed stackforge/kolla: Fix binary builds for oraclelinux templates https://review.openstack.org/214687 | 15:58 |
| openstackgerrit | Merged stackforge/kolla: Add Dockerfile template for kolla-ansible https://review.openstack.org/213394 | 16:01 |
| openstackgerrit | Merged stackforge/kolla: Remove barbican https://review.openstack.org/214453 | 16:01 |
| sdake | meeting time | 16:02 |
| *** stvnoyes has joined #kolla | 16:04 | |
| *** daneyon_ has joined #kolla | 16:05 | |
| *** daneyon has quit IRC | 16:08 | |
| *** jtriley has quit IRC | 16:17 | |
| *** jtriley has joined #kolla | 16:20 | |
| *** unicell has quit IRC | 16:23 | |
| *** thumpba has quit IRC | 16:32 | |
| *** dims_ has quit IRC | 16:39 | |
| *** dims has joined #kolla | 16:39 | |
| openstackgerrit | Ryan Hallisey proposed stackforge/kolla: Ceph container https://review.openstack.org/214706 | 16:43 |
| *** SourabhP has joined #kolla | 16:48 | |
| *** erkules_ has joined #kolla | 16:50 | |
| *** SourabhP_ has joined #kolla | 16:50 | |
| *** SourabhP has quit IRC | 16:52 | |
| *** SourabhP_ is now known as SourabhP | 16:52 | |
| *** erkules has quit IRC | 16:53 | |
| *** erkules has joined #kolla | 16:55 | |
| *** unicell has joined #kolla | 16:55 | |
| *** erkules_ has quit IRC | 16:56 | |
| * rhallisey gets lunch | 17:00 | |
| harmw | o/ | 17:00 |
| inc0 | https://etherpad.openstack.org/p/kolla-logging-stack I'll also put link to bp | 17:00 |
| harmw | great inc0 | 17:00 |
| inc0 | tomorrow I'll write something there | 17:00 |
| *** britthouser has joined #kolla | 17:00 | |
| inc0 | hopefully | 17:00 |
| sdake | inc0 sounds fantastic - appreciate the help here :) | 17:01 |
| inc0 | imho this is critical | 17:01 |
| inc0 | for people to appreciate kolla | 17:01 |
| inc0 | I mean, ops will love this | 17:02 |
| *** diga has quit IRC | 17:02 | |
| harmw | def. inc0 | 17:02 |
| inc0 | I'm out for tonight, please, if you have any ideas feel free to put these to etherpad | 17:02 |
| inc0 | I'll do the same tomorrow | 17:03 |
| inc0 | have a good day guys, cyas | 17:04 |
| *** unicell has quit IRC | 17:04 | |
| harmw | \o/ | 17:04 |
| *** inc0 has quit IRC | 17:04 | |
| *** unicell has joined #kolla | 17:04 | |
| sdake | pbourke rebuilding now with master and more disk sspace, i'll let you know if i can get a deploy | 17:11 |
| pbourke | sdake: I think I was missing building the new data container | 17:11 |
| pbourke | sdake: building now and I reckon it will work | 17:11 |
| sdake | oh that would be a problem ;) | 17:11 |
| pbourke | sdake: strange though that it tried to start at all | 17:11 |
| pbourke | like all came up just the services that used data containers went into a restarting state | 17:12 |
| sdake | that would explaint he rabbitmq permissions problem ;) | 17:12 |
| pbourke | ok better run! btw i was chatting to SamYaple and he doesn't think we should solve the generation of swift rings in ansible | 17:13 |
| pbourke | maybe we could, but it probably shouldn't hold back this blueprint | 17:13 |
| sdake | i htink we need to solve it | 17:14 |
| sdake | but it can be in rc1 | 17:14 |
| sdake | suggest making a new blueprint for swift-ring-generation and putting in rc1 | 17:14 |
| pbourke | i need to take a better look at what ops are currently doing to generate these rings | 17:14 |
| sdake | because without it, swift is unable | 17:14 |
| sdake | unusable | 17:14 |
| pbourke | yeah :( | 17:15 |
| harmw | guys, anyone knows a decent openstack-running cloudprovider? | 17:15 |
| sdake | rax is good | 17:15 |
| sdake | little pricey | 17:15 |
| harmw | to expensive | 17:15 |
| harmw | too, that is | 17:15 |
| sdake | but has best deployment that is public | 17:15 |
| harmw | I was thinking of cloudvps | 17:16 |
| sdake | havne't used them | 17:16 |
| harmw | which happens to be NL-based | 17:16 |
| harmw | but they don't utilize much of the openstack eco | 17:16 |
| jmccarthy | Gotta jet also, catch you guys next time ! | 17:16 |
| harmw | o/ | 17:16 |
| *** sdake_ has joined #kolla | 17:17 | |
| *** SourabhP has quit IRC | 17:17 | |
| *** jmccarthy has quit IRC | 17:17 | |
| harmw | digitalocean is using some homecooked piece, right | 17:18 |
| harmw | hpcloud is also expensive | 17:19 |
| harmw | hm, dreamhost, ofcourse... | 17:20 |
| *** sdake has quit IRC | 17:20 | |
| sdake_ | i only have experience with rackspace | 17:26 |
| sdake_ | and it is expensive :) | 17:26 |
| sdake_ | but otherwise rock solid | 17:26 |
| *** britthouser has quit IRC | 17:27 | |
| *** britthouser has joined #kolla | 17:28 | |
| *** loth has joined #kolla | 17:35 | |
| *** sdake has joined #kolla | 17:37 | |
| *** sdake_ has quit IRC | 17:40 | |
| *** achanda has joined #kolla | 17:42 | |
| *** sdake has quit IRC | 17:47 | |
| *** sdake has joined #kolla | 17:51 | |
| *** SourabhP has joined #kolla | 17:52 | |
| *** achanda_ has joined #kolla | 17:54 | |
| *** dwalsh has joined #kolla | 17:54 | |
| *** achanda has quit IRC | 17:55 | |
| *** dwalsh has quit IRC | 18:02 | |
| harmw | ever been to FOSDEM sdake ? | 18:07 |
| sdake | no harmw | 18:13 |
| harmw | ok | 18:14 |
| *** tfukushima has quit IRC | 18:33 | |
| *** banix has quit IRC | 18:33 | |
| *** tfukushima has joined #kolla | 18:42 | |
| *** jasonsb has quit IRC | 18:43 | |
| *** jasonsb has joined #kolla | 18:44 | |
| *** tfukushima has quit IRC | 18:46 | |
| *** cfarquhar has joined #kolla | 18:47 | |
| *** cfarquhar has quit IRC | 18:47 | |
| *** cfarquhar has joined #kolla | 18:47 | |
| *** jasonsb has quit IRC | 18:48 | |
| *** banix has joined #kolla | 18:53 | |
| *** akwasnie has quit IRC | 19:09 | |
| *** jasonsb has joined #kolla | 19:25 | |
| *** gfidente has quit IRC | 19:34 | |
| *** achanda_ has quit IRC | 19:37 | |
| *** shardy_ has joined #kolla | 19:41 | |
| *** shardy has quit IRC | 19:42 | |
| *** shardy_ has quit IRC | 19:46 | |
| *** shardy has joined #kolla | 19:47 | |
| *** achanda has joined #kolla | 19:51 | |
| vincent_vdk | you should definitely come to FOSDEM | 20:29 |
| *** achanda has quit IRC | 20:47 | |
| *** achanda has joined #kolla | 20:52 | |
| *** achanda has quit IRC | 21:06 | |
| openstackgerrit | imain proposed stackforge/kolla: Add an environment variable to allow args to be set on startup. https://review.openstack.org/214799 | 21:15 |
| *** shardy has quit IRC | 21:31 | |
| *** rhallisey has quit IRC | 21:40 | |
| *** jtriley has quit IRC | 21:44 | |
| *** unicell has quit IRC | 21:45 | |
| openstackgerrit | William Frischemeier proposed stackforge/kolla: Load ip_vs module for keepalived https://review.openstack.org/214811 | 21:45 |
| *** unicell has joined #kolla | 21:45 | |
| *** absubram has quit IRC | 22:03 | |
| *** unicell has quit IRC | 22:12 | |
| *** unicell has joined #kolla | 22:12 | |
| *** loth1 has joined #kolla | 22:46 | |
| *** loth has quit IRC | 22:48 | |
| *** jasonsb has quit IRC | 22:55 | |
| *** britthouser has quit IRC | 22:56 | |
| *** stvnoyes has quit IRC | 23:02 | |
| *** banix has quit IRC | 23:02 | |
| *** diogogmt has quit IRC | 23:03 | |
| *** tfukushima has joined #kolla | 23:03 | |
| *** banix has joined #kolla | 23:31 | |
| *** achanda has joined #kolla | 23:33 | |
| *** dims_ has joined #kolla | 23:33 | |
| *** dims has quit IRC | 23:36 | |
| *** dims_ has quit IRC | 23:39 | |
| *** dims has joined #kolla | 23:39 | |
| *** dims has quit IRC | 23:44 | |
| *** loth1 is now known as loth | 23:46 | |
| sdake | samyaple if you appear | 23:47 |
| sdake | samyaple I can't install from master | 23:47 |
| sdake | glance fails with some wierd ansible spam trace | 23:48 |
| *** blahRus has quit IRC | 23:52 | |
| *** Haomeng|2 has joined #kolla | 23:55 | |
| *** Haomeng has quit IRC | 23:59 | |
| *** jasonsb has joined #kolla | 23:59 | |
Generated by irclog2html.py 2.14.0 by Marius Gedminas - find it at mg.pov.lt!