*** dims has quit IRC | 00:21 | |
*** dims has joined #kolla | 00:22 | |
*** dims has quit IRC | 00:26 | |
*** dolpher has joined #kolla | 00:28 | |
*** diogogmt has joined #kolla | 00:29 | |
*** dolpher1 has joined #kolla | 00:32 | |
*** davidjbc has joined #kolla | 00:32 | |
*** dolpher has quit IRC | 00:33 | |
*** davidjc has quit IRC | 00:35 | |
*** sdake has joined #kolla | 00:57 | |
*** jasonsb has quit IRC | 00:58 | |
*** jasonsb has joined #kolla | 00:59 | |
*** davidjbc has quit IRC | 01:00 | |
*** sdake_ has quit IRC | 01:01 | |
*** jasonsb has quit IRC | 01:03 | |
*** sdake_ has joined #kolla | 01:19 | |
*** sdake has quit IRC | 01:21 | |
*** erkules_ has joined #kolla | 01:23 | |
*** vinkman has quit IRC | 01:24 | |
*** erkules has quit IRC | 01:26 | |
*** davidjc has joined #kolla | 01:31 | |
*** davidjbc has joined #kolla | 01:34 | |
*** davidjc has quit IRC | 01:37 | |
sdake_ | bmace still out? | 01:40 |
---|---|---|
sdake_ | i'm bummed, the name I voted #1 on came in 13th place in the voting :( | 01:44 |
*** bitblt has quit IRC | 01:45 | |
*** erkules_ has quit IRC | 01:51 | |
*** erkules_ has joined #kolla | 01:52 | |
*** sdake_ has quit IRC | 02:07 | |
*** sdake has joined #kolla | 02:15 | |
*** loth1 has quit IRC | 02:36 | |
*** jasonsb has joined #kolla | 03:23 | |
*** diogogmt has quit IRC | 03:40 | |
*** dims has joined #kolla | 03:45 | |
*** vinkman has joined #kolla | 03:46 | |
*** dims_ has joined #kolla | 03:46 | |
*** dims_ has quit IRC | 03:46 | |
*** dims_ has joined #kolla | 03:46 | |
*** dims has quit IRC | 03:49 | |
*** vinkman has quit IRC | 03:55 | |
*** dims_ has quit IRC | 04:14 | |
*** dims has joined #kolla | 04:15 | |
bmace | sdake around? i'm back :) | 04:15 |
sdake | hey bmace | 04:16 |
bmace | greetings | 04:16 |
*** dims has quit IRC | 04:20 | |
sdake | bmace i think there wa a gate failure this morning but then saw it was a user rror not a get problem | 04:26 |
bmace | looks like infra still hasn't touched the review i re-posted. nice | 04:28 |
sdake | they will get to it eventually | 04:29 |
sdake | their queue is ike 10 pages long ;) | 04:29 |
bmace | there have been like 30 commits that have been merged in. my change is pretty simple. | 04:30 |
bmace | but we'll see | 04:30 |
*** sdake has quit IRC | 04:45 | |
*** dims has joined #kolla | 05:01 | |
*** dims_ has joined #kolla | 05:02 | |
*** dims has quit IRC | 05:06 | |
*** jruano has quit IRC | 05:06 | |
*** davidjbc has quit IRC | 05:12 | |
*** dims_ has quit IRC | 05:14 | |
*** akscram has quit IRC | 05:41 | |
*** akscram has joined #kolla | 05:42 | |
*** sdake has joined #kolla | 06:03 | |
*** dims has joined #kolla | 06:15 | |
*** dims has quit IRC | 06:20 | |
SamYaple | yo yo yo | 06:34 |
*** erkules_ is now known as erkules | 06:39 | |
*** erkules has joined #kolla | 06:39 | |
*** inc0 has joined #kolla | 06:42 | |
inc0 | morning | 06:43 |
SamYaple | morning | 06:47 |
*** gfidente has joined #kolla | 06:54 | |
vincent_vdk | morning | 06:57 |
SamYaple | sdake: ping | 06:57 |
*** nihilifer has joined #kolla | 06:58 | |
*** sdake has quit IRC | 07:03 | |
*** tobe has joined #kolla | 07:18 | |
inc0 | SamYaple, soo...HAProxy | 07:26 |
inc0 | we'll need to discuss multihost. | 07:26 |
SamYaple | MAKE IT WORK DO THE THINGS | 07:27 |
SamYaple | inc0: im cleaning up https://ask.openstack.org/en/questions/ at the moment | 07:27 |
SamYaple | big spam thing happenning | 07:27 |
inc0 | yup, seems so | 07:27 |
SamYaple | out of the 17,500 questions, 5000+ appear to be spam at this point | 07:28 |
SamYaple | inc0: feel free to talk. i will respond | 07:30 |
inc0 | scrum -.- in 15 minutes prepare... | 07:31 |
SamYaple | i totally know what scrum is and totally dont think of video games | 07:31 |
*** shardy has joined #kolla | 07:42 | |
SamYaple | jeez that was stupid | 07:43 |
SamYaple | ok inc0 im available to talk. will be hammering on keystone modules | 07:43 |
SamYaple | harmw: are you around? | 07:43 |
*** jmccarthy has quit IRC | 07:52 | |
inc0 | sooo.... | 08:05 |
inc0 | SamYaple, did you make any ansible multihost yet? | 08:06 |
inc0 | because I'm thinking about ditching confing_inside for haproxy alltogether | 08:06 |
*** athomas has joined #kolla | 08:08 | |
*** jmccarthy has joined #kolla | 08:14 | |
*** dims has joined #kolla | 08:16 | |
*** dims has quit IRC | 08:21 | |
SamYaple | inc0: ansible has always been mltihost | 08:26 |
inc0 | yeah, I know, I'm reading the code now | 08:26 |
inc0 | what do you think about ditching config_inside? | 08:27 |
inc0 | (I won't -1 if someone writes it, I just don't want to write it because it will have to be dirty and soon-not-so-supported) | 08:27 |
SamYaple | I do not use config_internal for the reason i dont think it can work well mltihost | 08:27 |
SamYaple | so do what you want there, but i need config-external. you dont have to write it, but i need it | 08:27 |
inc0 | nah, I'll write config external; | 08:28 |
inc0 | and I won't write internal | 08:28 |
*** gfidente has quit IRC | 08:28 | |
SamYaple | my plan is have ansible lay down a template in a conf.d _for_ _each_ _service_ | 08:28 |
SamYaple | then i believe i can HUP the process without restarting the container to pull in the changes | 08:28 |
SamYaple | one massive config doesn't work well | 08:28 |
inc0 | agree | 08:29 |
SamYaple | i did one big config for yaodu, it was usable, but i dont want to do that long term | 08:29 |
inc0 | things in kolla/etc/kolla/config are empty...why? | 08:31 |
inc0 | this is intentional or it will be filled later on? | 08:32 |
SamYaple | you mean the files? | 08:33 |
*** dolpher1 has quit IRC | 08:33 | |
SamYaple | they are the "augment" files. they wont have any data by default | 08:33 |
SamYaple | i need a chance to explain how the work more fully and get feedback, but havent had the oppurtunity yet | 08:33 |
*** dolpher has joined #kolla | 08:34 | |
inc0 | I see | 08:35 |
inc0 | care to say word or two? | 08:35 |
inc0 | now? | 08:36 |
inc0 | I'm trying to grasp how to best approach external config for haproxy | 08:36 |
SamYaple | that needs to be handled by ansible templates | 08:37 |
inc0 | and where do we put things like custom keystone endpoint port? | 08:38 |
SamYaple | "template" | 08:38 |
SamYaple | it doesnt matter if its "custom" or not | 08:39 |
inc0 | what I mean is where this value will come from? | 08:39 |
inc0 | what end user will fill? | 08:39 |
SamYaple | globals.conf | 08:39 |
SamYaple | er globals.yml | 08:39 |
inc0 | ah ok | 08:39 |
inc0 | and Ansible will know when it's set up glance or keystone | 08:40 |
inc0 | ? | 08:40 |
inc0 | what I mean, if we'd set up haproxy first | 08:41 |
inc0 | there would be no endpoint whatsoever there | 08:41 |
inc0 | and then if we set up ks, haproxy conf should be updated and reread inside | 08:41 |
SamYaple | yep | 08:42 |
inc0 | in yaodu I've seen you've hardcoded these, where could I put hook like that? | 08:42 |
SamYaple | we may have to use supervisord. *maybe* thats only if the HUP kills the process | 08:42 |
inc0 | in keystone for example? | 08:42 |
SamYaple | what do you mean? | 08:42 |
inc0 | I mean we should put hook on ansible deploy | 08:43 |
inc0 | so whenever ansible deploys keystone, it updates haproxy as well | 08:43 |
inc0 | or to be more precise, whenever it adds endpoint to ks | 08:43 |
SamYaple | yes, thats the templates that i was saying | 08:43 |
SamYaple | no thats not correct | 08:43 |
SamYaple | just at the end of the role | 08:43 |
inc0 | but templates are filled once or we can force that at some point of time? | 08:44 |
SamYaple | templates are processed each time the playbooks run | 08:44 |
inc0 | well, ok, I think I'll just try it out;) | 08:46 |
inc0 | not overthink it as of now | 08:46 |
SamYaple | yea just build the container right now, that is enough | 08:46 |
harmw | SamYaple: I'm here now, try to keep it painless | 09:01 |
SamYaple | merge all the things! | 09:01 |
harmw | argh! PAIN | 09:01 |
SamYaple | https://review.openstack.org/#/c/198494/ | 09:01 |
SamYaple | youve already +2 that | 09:01 |
harmw | no no, can't do shit now mate | 09:02 |
SamYaple | DO IT | 09:02 |
harmw | can't login from $work | 09:02 |
SamYaple | JUST DO IT | 09:02 |
SamYaple | harmw: https://www.youtube.com/watch?v=os6U77Hhm_s | 09:02 |
harmw | hehe | 09:02 |
harmw | Ill ack it tonight, if noone else has done it by then | 09:03 |
SamYaple | yesterday you said today. JUTS DO IT | 09:04 |
harmw | yesterday I didn't say shit | 09:04 |
harmw | I could also do it tomorrow you know | 09:04 |
harmw | :P | 09:04 |
SamYaple | JUST DO IT | 09:04 |
harmw | NIKE | 09:06 |
inc0 | I gave you +1 Sam, I know it's poor but I hope it will make you happy, or at least less sad | 09:13 |
*** nihilifer has quit IRC | 09:13 | |
harmw | it won't | 09:13 |
harmw | :p | 09:13 |
SamYaple | /kick harmw | 09:14 |
pbourke | /slap | 09:14 |
* harmw left #kolla | 09:14 | |
openstackgerrit | Swapnil Kulkarni proposed stackforge/kolla: Add config-internal/config-external for Ceilometer https://review.openstack.org/198241 | 09:14 |
*** coolsvap|away is now known as coolsvap | 09:15 | |
inc0 | btw we need to make our own, better kolla midcycle in Europe | 09:16 |
inc0 | I suggest on top of mount Kolla in Norway | 09:16 |
harmw | lol | 09:16 |
SamYaple | i want to move to Norway so bad | 09:16 |
harmw | or in Greece :> | 09:16 |
pbourke | :) | 09:16 |
SamYaple | i could live like a king! | 09:17 |
inc0 | SamYaple, you don't like days? | 09:17 |
SamYaple | no | 09:17 |
inc0 | them you'll be miserable for 6 months | 09:17 |
SamYaple | ill just work night shift over there too | 09:17 |
inc0 | you don't understant...there is no night for 6 months | 09:18 |
inc0 | whatsoever | 09:18 |
SamYaple | lies! | 09:18 |
SamYaple | night is when people go to sleep | 09:18 |
SamYaple | i dont like people, the sun is fine | 09:18 |
inc0 | part of Norway is behind arctic circle | 09:18 |
SamYaple | i also love the cold | 09:19 |
inc0 | SamYaple, then rural Canada seems like good place as well | 09:19 |
SamYaple | norway is the place for me | 09:19 |
SamYaple | yea but... canada | 09:19 |
harmw | canada.. lol | 09:19 |
inc0 | ah, these US jokes...I find Canada appealing | 09:19 |
SamYaple | my wife is canadian | 09:19 |
harmw | http://new1.fjcdn.com/comments/4685688+_3887d3e3550ae26ef3039577c5ac8eec.jpg | 09:20 |
inc0 | Norway is cool place. Visit Alesund, lovely city | 09:21 |
*** gfidente has joined #kolla | 09:21 | |
SamYaple | i want to retire to Norway | 09:22 |
inc0 | it's better to live there and retire there rly | 09:23 |
inc0 | Norway is very expensive, but when you have their pensinon, it's cool | 09:23 |
inc0 | high taxes, strong social security and so on | 09:23 |
SamYaple | ill be rich by then | 09:24 |
SamYaple | right | 09:24 |
harmw | aren't we all | 09:25 |
vincent_vdk | move to .be. we have nice beers :) | 09:28 |
harmw | thats about the only thing you ppl have ;) | 09:28 |
vincent_vdk | thats what everybody thinks | 09:28 |
harmw | you're from the dutch part vincent_vdk ? | 09:29 |
vincent_vdk | yes | 09:29 |
harmw | good | 09:29 |
harmw | :p | 09:29 |
vincent_vdk | :) | 09:29 |
inc0 | Europe is funny. | 09:30 |
inc0 | number of antagonsims on our little continent;) | 09:31 |
vincent_vdk | where do you live inc0 | 09:31 |
inc0 | Poland | 09:31 |
vincent_vdk | ah | 09:32 |
SamYaple | oh boy | 09:32 |
inc0 | surprised Sam?;) | 09:32 |
SamYaple | no i could totally tell that | 09:32 |
vincent_vdk | would like to visit that some day | 09:32 |
inc0 | come, we have beers;) and vodla | 09:32 |
inc0 | vodka | 09:32 |
SamYaple | i think vodla is more accurate | 09:32 |
vincent_vdk | i've heard :) | 09:32 |
harmw | and girls with moustaches | 09:32 |
harmw | or was that Turkey | 09:33 |
vincent_vdk | lol | 09:33 |
*** shardy_ has joined #kolla | 09:33 | |
harmw | well anyway, isn't Europe great :) | 09:33 |
vincent_vdk | it is | 09:33 |
*** shardy_ has quit IRC | 09:33 | |
*** shardy_ has joined #kolla | 09:33 | |
inc0 | its crowded and diverse | 09:34 |
vincent_vdk | but indeed, the Nordics have nice places | 09:34 |
harmw | and nice girls, without moustaches :P | 09:34 |
*** shardy has quit IRC | 09:34 | |
inc0 | yeah, blondes are cool;) | 09:35 |
harmw | vincent_vdk: unrelated question, but you know or even use wehkamp.be? | 09:37 |
openstackgerrit | Swapnil Kulkarni proposed stackforge/kolla: Add config-internal/config-external to Magnum containers https://review.openstack.org/198213 | 09:37 |
vincent_vdk | harmw: never heard of it | 09:38 |
harmw | hehe | 09:38 |
harmw | ok | 09:38 |
vincent_vdk | it's like ibood | 09:38 |
*** shardy_ has quit IRC | 09:38 | |
harmw | not realy | 09:38 |
harmw | just the biggest online retailer in NL | 09:39 |
vincent_vdk | didn't know them | 09:39 |
*** shardy has joined #kolla | 09:39 | |
harmw | we've ventured into .BE last year, so still new I guess | 09:39 |
vincent_vdk | you work for them? | 09:39 |
harmw | yep | 09:40 |
vincent_vdk | nice | 09:40 |
vincent_vdk | running on openstack :) | 09:40 |
SamYaple | whats openstack? | 09:40 |
harmw | no vincent_vdk | 09:40 |
SamYaple | http://www.dopenstack.com/ | 09:40 |
harmw | we don't care about hardware, we just need some piece of infra to run our mesos platform | 09:41 |
harmw | my interests in openstack are just of personal interest:) | 09:42 |
vincent_vdk | ah | 09:42 |
vincent_vdk | i see | 09:42 |
inc0 | SamYaple, dafaq I just seen oO | 09:42 |
vincent_vdk | so you are operating in true devops style :) | 09:42 |
harmw | yes vincent_vdk | 09:43 |
openstackgerrit | Swapnil Kulkarni proposed stackforge/kolla: Add config-internal/config-external to Zaqar containers https://review.openstack.org/198284 | 09:43 |
SamYaple | no one told you to look inc0 | 09:43 |
inc0 | I should know better to klick on links from you.. | 09:43 |
vincent_vdk | NSFW? | 09:44 |
harmw | SamYaple: you know a way to get the ip from inventory_hostname in Ansible? | 09:44 |
harmw | no vincent_vdk :) | 09:44 |
SamYaple | 09:40 < vincent_vdk> nice | 09:44 |
SamYaple | https://www.youtube.com/watch?v=o0u4M6vppCI | 09:44 |
inc0 | vincent_vdk, not rly...just...weird | 09:44 |
SamYaple | harmw: kinda? but probably not how you think | 09:44 |
SamYaple | what are you trying to do | 09:44 |
harmw | i have multiple hostnames in my inventory, but some map the same host | 09:45 |
harmw | just that hosts runs multiple ip's | 09:45 |
harmw | and I want the ip that belongs (resolves) to the used hostname in inventory | 09:45 |
harmw | to put in some .j2 template | 09:45 |
harmw | (again, not Kolla-related) | 09:45 |
harmw | I could pipe inventory_hostname to/through something | 09:46 |
SamYaple | ip that resolves to hostname is realtive to the host it runes on | 09:46 |
vincent_vdk | use the facts from the one you use? | 09:46 |
harmw | yes SamYaple but how do I get that, which var would that be? | 09:46 |
harmw | which var belongs to the inventory_hostname's resolved IP | 09:47 |
SamYaple | {{ hostvars[inventory_hostname]['ansible_' + api_interface]['ipv4']['address'] }} | 09:47 |
SamYaple | where api_interface is the name of an interface ('eth0') | 09:47 |
harmw | thats to much hassle, plus, I have just eth0 with some aliases on it | 09:47 |
SamYaple | if you dont need to address the interface programtically you can do it a bit easier | 09:48 |
*** coolsvap is now known as coolsvap|away | 09:48 | |
harmw | this server has 3 ips, mapped as bla1.domain bla2.domain and bla3.domain and as such they are listed in my inventory | 09:48 |
harmw | but this cfg file I'm editting (or: templating) needs an IP, not that hostname | 09:49 |
SamYaple | if these are static ips you can just set them as variables in your inventory file | 09:49 |
harmw | they aren't, the hostnames are different per environment/setup | 09:49 |
harmw | guess Ill just have to resolve 'em on the spot | 09:49 |
SamYaple | there is no way to do what you are asking dynamically | 09:50 |
*** nihilifer has joined #kolla | 09:55 | |
*** dolpher has quit IRC | 09:56 | |
openstackgerrit | Swapnil Kulkarni proposed stackforge/kolla: Add config-internal/config-external to Magnum containers https://review.openstack.org/198213 | 09:58 |
openstackgerrit | Swapnil Kulkarni proposed stackforge/kolla: Add config-internal/config-external to Zaqar containers https://review.openstack.org/198284 | 09:59 |
*** shardy_ has joined #kolla | 10:03 | |
*** shardy has quit IRC | 10:05 | |
*** shardy_ has quit IRC | 10:09 | |
*** shardy has joined #kolla | 10:10 | |
inc0 | SamYaple, correct me if I'm wrong, current multi-node doesn't have notion of compute node vs controller node right? | 10:10 |
SamYaple | you are wrong and have been corrected! | 10:10 |
openstackgerrit | Swapnil Kulkarni proposed stackforge/kolla: Add config-internal/config-external for Ceilometer https://review.openstack.org/198241 | 10:10 |
inc0 | where do you differentiate? | 10:11 |
SamYaple | inventory file | 10:11 |
vincent_vdk | i'm wondering if you run on top of coreos. would it be possible to leverage fleet to make containers available over multiple locations | 10:11 |
*** coolsvap|away is now known as coolsvap | 10:11 | |
harmw | vincent_vdk: we don't want to have dependencies | 10:11 |
harmw | is fleet specific to coreos? | 10:12 |
SamYaple | yes | 10:12 |
openstackgerrit | Swapnil Kulkarni proposed stackforge/kolla: Add config-internal/config-external for Ceilometer https://review.openstack.org/198241 | 10:12 |
inc0 | SamYaple, it should be in inventory, but right now it isn't right? | 10:12 |
vincent_vdk | i'm not talking about integrating in kolla | 10:12 |
harmw | pipes is something in that area as well perhaps | 10:12 |
vincent_vdk | we are looking for a way to make openstack "HA" accross multiple DCs | 10:12 |
SamYaple | inc0: https://github.com/SamYaple/yaodu/blob/master/ansible/inventories/production#L67-68 | 10:12 |
SamYaple | inc0: thats going to give you an idea when its all done | 10:12 |
vincent_vdk | fleet is specific to coreos | 10:13 |
vincent_vdk | basically distributed systemd | 10:13 |
harmw | no vincent_vdk , but if we start on that people not using CoreOS can't use it | 10:13 |
inc0 | yeah, it's not done *yet*, that's what I was asking;) | 10:13 |
SamYaple | inc0: that sounds right then | 10:13 |
harmw | vincent_vdk: where do you work anyway :) | 10:13 |
SamYaple | in the end, every single container can be set to run on a specific host, with no overlap (except where contaienrs are dependant like neutron_agent+nova-compute) | 10:14 |
SamYaple | *techinically* you can run nova-compute on a diffent host than libvirt and neutron-plugin-agent | 10:14 |
SamYaple | just saying | 10:14 |
*** dims has joined #kolla | 10:17 | |
*** dims has quit IRC | 10:21 | |
harmw | SamYaple: {'msg': "AnsibleError: Failed to template {{ lookup('dig', '@' + nameservers + ' ' + inventory_hostname + './A') }}: ('dns.resolver unhandled exception', NoNameservers())", 'failed': True} | 10:24 |
harmw | interesting, just removed that . and it worked :| | 10:25 |
*** shardy_ has joined #kolla | 10:25 | |
*** coolsvap is now known as coolsvap|away | 10:25 | |
*** shardy has quit IRC | 10:27 | |
*** shardy_ has quit IRC | 10:31 | |
*** shardy has joined #kolla | 10:32 | |
*** sdake has joined #kolla | 10:45 | |
*** sdake_ has joined #kolla | 10:46 | |
sdake_ | morning | 10:47 |
sdake_ | samyaple shoot | 10:48 |
SamYaple | bang bang | 10:48 |
sdake_ | im wounded | 10:48 |
sdake_ | you pung me earlier | 10:48 |
SamYaple | i dont remember why | 10:48 |
SamYaple | it was important at the time | 10:48 |
*** sdake has quit IRC | 10:50 | |
openstackgerrit | Paul Bourke proposed stackforge/kolla: Update centos-source-base from symlink to a real layer https://review.openstack.org/197897 | 11:06 |
openstackgerrit | Paul Bourke proposed stackforge/kolla: Keystone installation from source https://review.openstack.org/193166 | 11:06 |
openstackgerrit | Paul Bourke proposed stackforge/kolla: Add source base image for oraclelinux https://review.openstack.org/200013 | 11:13 |
sdake_ | yaya go pbourke ! :) | 11:14 |
pbourke | :) | 11:14 |
pbourke | +2 all the things | 11:14 |
openstackgerrit | Paul Bourke proposed stackforge/kolla: Add oraclelinux source base image https://review.openstack.org/200013 | 11:15 |
openstackgerrit | Paul Bourke proposed stackforge/kolla: Add source keystone image for oraclelinux https://review.openstack.org/191833 | 11:15 |
openstackgerrit | Paul Bourke proposed stackforge/kolla: Add source base image for oraclelinux Partially implements: blueprint add-optional-oraclelinux-images https://review.openstack.org/200013 | 11:16 |
openstackgerrit | Paul Bourke proposed stackforge/kolla: Add source base image for oraclelinux https://review.openstack.org/200013 | 11:17 |
openstackgerrit | Paul Bourke proposed stackforge/kolla: Add source keystone image for oraclelinux https://review.openstack.org/191833 | 11:18 |
*** coolsvap|away is now known as coolsvap | 11:22 | |
openstackgerrit | Swapnil Kulkarni proposed stackforge/kolla: Add config-internal/config-external to Zaqar containers https://review.openstack.org/198284 | 11:23 |
*** shardy_ has joined #kolla | 11:25 | |
openstackgerrit | Swapnil Kulkarni proposed stackforge/kolla: Add config-internal/config-external to Magnum containers https://review.openstack.org/198213 | 11:26 |
*** shardy has quit IRC | 11:26 | |
vincent_vdk | it seems the ceilometer container tries to install systemd which fails | 11:29 |
*** shardy_ has quit IRC | 11:30 | |
*** shardy has joined #kolla | 11:31 | |
coolsvap | SamYaple, ping, have couple of mins ? | 11:31 |
SamYaple | sure | 11:32 |
*** sdake_ has quit IRC | 11:33 | |
coolsvap | regarding patch https://review.openstack.org/#/c/198242 | 11:33 |
coolsvap | https://review.openstack.org/#/c/198242/1/docker/common/openvswitch/ovs-db-server/start.sh | 11:33 |
coolsvap | /usr/sbin/ovsdb-server seems to be the correct location | 11:34 |
coolsvap | $ which ovsdb-server | 11:34 |
coolsvap | /usr/sbin/ovsdb-server | 11:34 |
SamYaple | ok thats fine | 11:36 |
SamYaple | that wasnt the eason for the -1 | 11:36 |
coolsvap | also regarding the arguments that are missing | 11:37 |
coolsvap | does it remain same in both config-internal & start.sh? | 11:38 |
SamYaple | yes | 11:38 |
coolsvap | alright, so we need the additional variables in both places, correct? | 11:39 |
SamYaple | yes | 11:39 |
coolsvap | thx! updating the patchset | 11:40 |
*** sdake has joined #kolla | 11:42 | |
sdake | vincent_vdk ceilometer doesn't work | 11:43 |
sdake | it needs someone to implement it | 11:43 |
vincent_vdk | ok | 11:43 |
sdake | at last assessment, coolsvap is woring on it | 11:43 |
sdake | but it has been through about 5 hands so ymmv ;) | 11:43 |
sdake | ceiloemter swift are our gaps | 11:43 |
sdake | cinder is close | 11:43 |
sdake | (to working) | 11:43 |
sdake | cinder was woring for a bit then some change broke it | 11:44 |
sdake | sme upstream schagen | 11:44 |
vincent_vdk | i'm hitting the rpm unpacking bug running docker on Ubuntu | 11:44 |
vincent_vdk | that's why i saw the container build failing | 11:44 |
sdake | ya that is because ubuntu runs with something other then devicemapper | 11:45 |
sdake | probably overlayfs | 11:45 |
vincent_vdk | yeah | 11:45 |
vincent_vdk | no, AUFS | 11:45 |
inc0 | anyone uses ceilometer on prod | 11:45 |
inc0 | ? | 11:45 |
sdake | i thin aufs shoudl work | 11:45 |
vincent_vdk | thin aufs? | 11:45 |
sdake | inc0 from what ia hear yes | 11:45 |
vincent_vdk | ah, thin/think | 11:45 |
SamYaple | overlayfs is broke for building | 11:45 |
vincent_vdk | :) | 11:45 |
inc0 | (this is tricky question, I've asked it on summit...there was silence) | 11:45 |
sdake | vicient_vdk not certain on details - if rpm was lowing up yo uneed something else for a storage driver ;) | 11:45 |
vincent_vdk | it's known bug | 11:45 |
SamYaple | inc0: i know a few | 11:46 |
SamYaple | its bad | 11:46 |
vincent_vdk | https://github.com/docker/docker/issues/6980 | 11:46 |
*** tobe has quit IRC | 11:48 | |
sdake | overlyyfs hs same bug vincient_vdk, if your using 1.7.0 your using overlayfs | 11:50 |
sdake | except in overlayfs its not fixed | 11:50 |
vincent_vdk | ah | 11:50 |
sdake | you have you use -s devicemapper to build kolla unfortunately | 11:50 |
vincent_vdk | ubuntu 14.04 has aufs afaik | 11:50 |
sdake | I thin zfs also works | 11:50 |
sdake | ok well depends how you installed it i guess ;) | 11:51 |
vincent_vdk | yeah, it's an image on an openstack instance | 11:51 |
vincent_vdk | so no lvm either :) | 11:51 |
vincent_vdk | uploading centos7 image | 11:51 |
sdake | you can run lvm in a vm | 11:52 |
vincent_vdk | i know | 11:52 |
vincent_vdk | but the ubuntu images come without lvm | 11:53 |
sdake | oh thats a bummer | 11:53 |
vincent_vdk | i could add a volume | 11:53 |
sdake | zfs may be your only option then | 11:53 |
vincent_vdk | but easier to just deploy a centos7 vm | 11:53 |
sdake | agree that solves problem :) | 11:53 |
vincent_vdk | i've been using centos for the last 8 years or so | 11:54 |
sdake | its a shame the builds dont work cross platform but there is little i can do about it | 11:54 |
sdake | other then document it - which I havne't done ;) | 11:54 |
vincent_vdk | but currently working in a "Ubuntu shop" | 11:54 |
sdake | pbourke question about from source | 11:55 |
sdake | pbourke areyou basing your work ontop of centos to start | 11:55 |
sdake | then symlinking from there? | 11:55 |
*** jruano has joined #kolla | 11:56 | |
*** sdake_ has joined #kolla | 11:57 | |
sdake_ | pbourke if you asnwered i was reconnecting | 11:57 |
inc0 | SamYaple, soo...ansible and database | 12:00 |
*** sdake has quit IRC | 12:00 | |
inc0 | https://github.com/stackforge/kolla/blob/master/ansible/roles/database/tasks/bootstrap.yml#L39 you set up data here right? | 12:01 |
inc0 | but use galera image? | 12:01 |
openstackgerrit | Merged stackforge/kolla: Update centos-source-base from symlink to a real layer https://review.openstack.org/197897 | 12:01 |
*** rhallisey has joined #kolla | 12:03 | |
sdake_ | morning rhallisey | 12:03 |
rhallisey | morning | 12:03 |
sdake_ | recinder , ppls keep sking so I am curious what is busted atm with cinder that was workingprior | 12:04 |
sdake_ | at one point it was working | 12:04 |
sdake_ | re cinder that is | 12:04 |
SamYaple | inc0: that ssetups the "data" container for the database yes | 12:05 |
SamYaple | morning rhallisey | 12:06 |
inc0 | then why it uses image for galera itself? bug? | 12:06 |
SamYaple | inc0: there is no reason to use a different image | 12:06 |
rhallisey | sdake_, right, I think only Slower saw it working, unless someone else tried it | 12:06 |
sdake_ | oh i thgouth you had it working as ell | 12:06 |
sdake_ | i thought you fixed a bunch of bugs and tested against the workingversion :) | 12:07 |
sdake_ | apparently I was wrong - so where are yous tuck atm on it? | 12:07 |
inc0 | SamYaple, well, since we have dedicated container for data, there is no reason to use this image I'd say... | 12:07 |
rhallisey | sdake_, it broke about a week before the release | 12:08 |
openstackgerrit | Merged stackforge/kolla: Keystone installation from source https://review.openstack.org/193166 | 12:08 |
sdake_ | broke for slower as well? | 12:08 |
rhallisey | ya | 12:08 |
sdake_ | broke how | 12:08 |
rhallisey | so i furiously went through trying to fix it | 12:08 |
rhallisey | it was an iscsi issue then went away when I upgraded to docker 1.7 | 12:08 |
rhallisey | no session found | 12:09 |
sdake_ | then you upgraded to docke r1.7 release and it came back? | 12:09 |
sdake_ | docker 1.7 reverted some feature iirc - samyaple wasn't that correct? | 12:09 |
rhallisey | ya it did, but I never changed my version of docker | 12:10 |
rhallisey | I thought the same thing | 12:10 |
rhallisey | I've tried using old patch sets of cinder no success yet | 12:11 |
sdake_ | are you actively working on this problem or some other issue? | 12:11 |
sdake_ | just want to make sure its working by july 30 :) | 12:11 |
sdake_ | if it is posisible to do so | 12:11 |
inc0 | guys, https://blueprints.launchpad.net/kolla/+spec/add-proxy-to-dockerfiles this bp. Mind if I do it asap? I'm getting fed up with local proxy problems | 12:12 |
SamYaple | inc0: we cant use that container since it exits | 12:12 |
rhallisey | I haven't been actively working on it this week, but I'll pick it up again later today | 12:12 |
inc0 | SamYaple, by "exist" I mean its written and in tree | 12:12 |
SamYaple | inc0: addionally, we dont need an extra container | 12:12 |
SamYaple | inc0: i meant exits | 12:13 |
rhallisey | I think it I get it in | 12:13 |
rhallisey | I think I can get it in* | 12:13 |
SamYaple | inc0: docker as a whole can do HTTP_PROXY | 12:14 |
sdake_ | rhallisey did you see https://ask.openstack.org/en/question/8942/cinder-broken-iscsiadm-no-session-found/ | 12:14 |
SamYaple | i dont think we need it in the container | 12:14 |
sdake_ | rogue tgtd process | 12:14 |
inc0 | it doesn't seem to work inside container tho | 12:14 |
inc0 | for yum and stuff | 12:14 |
SamYaple | yum stuff does alot of https | 12:15 |
sdake_ | inc0 reading blueprint moment | 12:15 |
inc0 | it does, and there is https_proxy as well | 12:15 |
rhallisey | sdake_, I probably have at one point | 12:15 |
SamYaple | inc0: which is supported by docker itself, outside the build | 12:15 |
rhallisey | I'll set up the environment again and get some more info | 12:16 |
sdake_ | rhallisey ya if you dont have tim eto debug let me knwo and i'll tae a look | 12:17 |
*** dims has joined #kolla | 12:17 | |
sdake_ | we need working storage ;) | 12:17 |
sdake_ | just dont let me know on the 27th "hey cinder is busted need help" | 12:18 |
rhallisey | sdake_, I started setting up everything yesterday I'll just pick that up. By I expect I'll need help debugging this | 12:18 |
sdake_ | rhallisey I try to get it rolling today if I can get my env setup for it | 12:18 |
inc0 | let's make ceph support **flies away** | 12:19 |
sdake_ | is somethingspecial needed onthe host such as an lvm group? | 12:19 |
sdake_ | rhallisey ^^ | 12:20 |
rhallisey | the way I do it is I have a backing file that's in the data container that I use to create the volume group | 12:20 |
openstackgerrit | Swapnil Kulkarni proposed stackforge/kolla: Add config-internal/config-external to OpenvSwitch containers https://review.openstack.org/198242 | 12:21 |
inc0 | SamYaple, with network-host default docker http_proxy setting doesn't seem to work | 12:21 |
inc0 | from inside container | 12:21 |
rhallisey | the issue seems to be that cinder is failing on iscsi then nova can't find it | 12:21 |
SamYaple | inc0: youll have basic ceph by monday | 12:21 |
rhallisey | sdake_, I'll get my env going and we can both look at it if you like | 12:21 |
SamYaple | inc0: its what i use, HTTP_PROXY | 12:21 |
SamYaple | ill look into it again | 12:21 |
inc0 | there is no http_proxy in env inside container | 12:22 |
sdake_ | rhallisey do I need to run any vg commands on the host or in the containers to get things rolling | 12:22 |
sdake_ | or should cinder "just work | 12:22 |
sdake_ | " | 12:22 |
inc0 | at least for me, I might be not be setting it up correctly | 12:22 |
*** dims has quit IRC | 12:22 | |
SamYaple | inc0: the docker build exists behind the docker-proxy, so itll catch it there i bleieve | 12:22 |
rhallisey | I do the vg commands in a script | 12:22 |
SamYaple | inc0: but ill verify im using my proxy cache | 12:22 |
inc0 | thank you | 12:23 |
sdake_ | rhallisey how aboutsubmittinga doc change - or alternately mailing to me and I'll submit a doc change | 12:23 |
sdake_ | that way its documented ;) | 12:23 |
inc0 | also, even if docker build uses http_proxy | 12:23 |
sdake_ | vs sitting on your hard disk in a script somewhere | 12:23 |
inc0 | host-net won't this way, so we might need proxy anyway | 12:23 |
SamYaple | inc0: i am really against setting this in the container for our default dockerfiles | 12:23 |
SamYaple | but lets get all the facts on the table and make a decision | 12:23 |
sdake_ | inc0 sampyaple pls use blueprint discussion so its tracked thanks :) | 12:24 |
rhallisey | sdake_, ya sure, I got distracted when it started breaking so I never got around to it.. I have a patch for it somewhere.. | 12:24 |
rhallisey | sdake_, oh what I meant with the vg stuff is that it's done in a script in the patch | 12:25 |
inc0 | SamYaple, why? | 12:25 |
sdake_ | rhallisey ok so basically i can ust run the cinder containers and they should just work | 12:25 |
sdake_ | di you fix the restart thing? | 12:25 |
rhallisey | yes | 12:25 |
sdake_ | and did that merge yet | 12:25 |
inc0 | is there any way to inject ENV to docker without modyfing Dockerfile? | 12:25 |
rhallisey | I haven't fixed the restart thing yet.. there's a new issue | 12:25 |
rhallisey | well I fixed the old isue | 12:26 |
SamYaple | aside from adding another layer, i dont like the %%%% replace and i dont think we should be using it like we are. now once we start templating things, this will be easier to do | 12:26 |
sdake_ | is the old issue merged? | 12:26 |
vincent_vdk | inc0: i believe not | 12:26 |
rhallisey | sdake_, the config_external stuff is causing ti to break now | 12:26 |
rhallisey | config-internal* | 12:26 |
sdake_ | rhallisey that hsould not be the case - but you need to set genenv appropriately | 12:26 |
rhallisey | sdake_, ya I added the oslo package to the base image | 12:26 |
sdake_ | rhallisey did that patch merge tho? | 12:26 |
rhallisey | I believe so, double checking | 12:27 |
sdake_ | brb disconnecting for a moment | 12:27 |
rhallisey | ya it did | 12:27 |
*** sdake has joined #kolla | 12:28 | |
rhallisey | sdake, it did merge | 12:29 |
sdake | thanks rhallisey | 12:29 |
rhallisey | /opt/kolla/kolla-common.sh: line 175: /opt/kolla/config-internal.sh: Permission denied | 12:30 |
rhallisey | ^ is that a genenv issue? | 12:30 |
sdake | grr | 12:30 |
sdake | no | 12:30 |
SamYaple | rhallisey: sounds like these new files dont all have the proper perms | 12:30 |
rhallisey | yerp | 12:30 |
SamYaple | chmod +x config-internal.sh | 12:30 |
rhallisey | ya I'll patch that | 12:30 |
sdake | rhallisey can you fie bug plz | 12:31 |
SamYaple | rhallisey: thanks. ill check in review patches to ensure they are all good | 12:31 |
rhallisey | it looks like just cinder | 12:31 |
*** sdake_ has quit IRC | 12:31 | |
*** dwalsh has joined #kolla | 12:32 | |
sdake | rhallisey which commands do you run on cinder to get it rolling | 12:35 |
sdake | do you have a test script or something? | 12:35 |
rhallisey | cinder create 1 | 12:36 |
rhallisey | cinder list | 12:36 |
rhallisey | make sure it's available | 12:36 |
rhallisey | boot a nova instance | 12:36 |
rhallisey | nova volume attach | 12:36 |
pbourke | sdake: yes I'm doing centos first then following up with ol as they're mostly symlinks | 12:39 |
sdake | pbourke cool thanks i saw that in the recentpatch revies i just did | 12:39 |
*** dims has joined #kolla | 12:40 | |
*** shardy_ has joined #kolla | 12:41 | |
*** shardy has quit IRC | 12:42 | |
sdake | /start.sh: line 12: set_configs: command not found | 12:43 |
pbourke | hmm in fact the ol keystone should probably just symlink the entire directory as it's identical to centos | 12:43 |
SamYaple | sdake: make sure you rebuild all of your images | 12:43 |
sdake | rhallisey i'll file bugs for these cinder issues and fix them | 12:43 |
sdake | samyaple i rebuilt 16 hours ago | 12:44 |
SamYaple | not finding set_configs is because you haven't rebuilt the base with the latest kolla-common.sh (or that start.sh doesnt source kolla-common.sh) | 12:44 |
rhallisey | sdake, sounds good. I'll have a look again, maybe I'll figure it out now that I stepped away from it for a week | 12:44 |
sdake | only cinder has this problem samyaple ? | 12:44 |
sdake | the resto f hte containers work | 12:44 |
vincent_vdk | i'm currently building all containers | 12:45 |
vincent_vdk | need to check Unknown instruction: CONFIG-SWIFT.SH | 12:45 |
SamYaple | vincent_vdk: that is because swift is broken | 12:45 |
vincent_vdk | ah :) | 12:45 |
pbourke | vincent_vdk: jmccarthy will have a patch for that soon | 12:45 |
pbourke | jmccarthy: no pressure ;) | 12:46 |
SamYaple | we could through a COPY in from of that command, but eh | 12:46 |
jmccarthy | ;) | 12:46 |
*** shardy_ has quit IRC | 12:46 | |
pbourke | not sure why the build-all would be failing though as I thought it was blacklisted | 12:47 |
*** shardy has joined #kolla | 12:47 | |
SamYaple | pbourke: its not failing in the gate, its throwing that error locally (which is to be expected) pbourke | 12:47 |
pbourke | yup | 12:47 |
openstackgerrit | Steven Dake proposed stackforge/kolla: Change permissions on cinder config-internal/external https://review.openstack.org/200055 | 12:48 |
sdake | build-all should fail | 12:49 |
sdake | the gate will ignore it since its whitelisted as an acceptable failure | 12:50 |
sdake | rahter no fial should warn that it went wrong | 12:50 |
openstackgerrit | Paul Bourke proposed stackforge/kolla: Add source keystone image for oraclelinux https://review.openstack.org/191833 | 12:50 |
pbourke | smallest review ever ^ | 12:50 |
inc0 | pbourke, can I -1 it with "add commit message please"?:P | 12:54 |
inc0 | just to -1 review with 1 line and that is an symlink;) | 12:54 |
*** shardy_ has joined #kolla | 12:55 | |
*** shardy has quit IRC | 12:57 | |
pbourke | dont make me come over there ;) | 12:57 |
*** shardy_ has quit IRC | 13:00 | |
*** shardy has joined #kolla | 13:01 | |
sdake | samyaple did you push the liberty 2 tagged images to docker registry? | 13:02 |
*** diogogmt has joined #kolla | 13:02 | |
SamYaple | sdake: i did not. havent had time | 13:02 |
SamYaple | been cleaning up ask.openstack.org | 13:02 |
sdake | that sound slike a dentist visit in terms of pain | 13:03 |
sdake | i ran build.py about 15 hours ago | 13:03 |
inc0 | moving keepalived to ansible - bp or not bp? | 13:03 |
sdake | it didn't appear to build centos-rdo-base | 13:03 |
sdake | something to keep in mind samyaple | 13:03 |
sdake | inc0 you dont need a bp there is already a catchall called 'ansible-service" | 13:03 |
sdake | that way we can see in one blueprint record all the services that have been addressed | 13:04 |
sdake | consoleauth is busted | 13:07 |
sdake | anyone noticed that? | 13:07 |
sdake | https://bugs.launchpad.net/kolla/+bug/1473062 | 13:10 |
openstack | Launchpad bug 1473062 in kolla "consoleauth is broken" [Critical,Triaged] | 13:10 |
sdake | wackamole | 13:11 |
sdake | so many upstreams makes it difficult to have somethingstable | 13:11 |
sdake | rhallisey cinder create 1 and list works | 13:13 |
sdake | it fails at the attach stage? | 13:13 |
rhallisey | yup volume creation is fine, the issue attaching | 13:13 |
sdake | how do you get list of tgtd volumes in the system | 13:14 |
sdake | like in raw format not cinder format | 13:14 |
*** kevsi has quit IRC | 13:16 | |
sdake | rhallisey which process is running tgtd | 13:17 |
sdake | when I run # tgtadm --lld iscsi --mode target --op show | 13:17 |
sdake | i get a failure | 13:17 |
sdake | i would expect this to list the cinder volume i just created | 13:17 |
vincent_vdk | what does iscsiadm --mode discovery --type sendtargets --portal <serverip> do? | 13:17 |
rhallisey | cinder volumes.. let me see | 13:18 |
*** kevsi has joined #kolla | 13:18 | |
sdake | [sdake@bigiron kolla]$ sudo docker exec 41 tgtadm --lld iscsi --mode target --op show | 13:19 |
coolsvap | sdake, most most of the kolla-standard-start is mostly complete, i think only haproxy and mongodb is remaining | 13:19 |
sdake | coolsvap is mongodb dooable? | 13:19 |
rhallisey | I'm reverting to old patches to see | 13:19 |
coolsvap | sdake, i didnt get what to update in it | 13:19 |
coolsvap | sdake, there is no common start.sh for it, if you can give me a hint, i can do | 13:20 |
SamYaple | sdake: build.py i think defaults to binary, not rdo. not sure | 13:20 |
sdake | samyaple oi see | 13:20 |
SamYaple | `./build.py --type rdo` would work | 13:21 |
sdake | tgtd is not running inside the cinder-volume container rhallisey | 13:21 |
sdake | shouldnt it be? | 13:21 |
SamYaple | sdake: i asked that before, rhallisey said no it should be | 13:22 |
sdake | well there is the problem :) | 13:22 |
sdake | or one of them | 13:22 |
rhallisey | I think I was running an old build | 13:22 |
rhallisey | for a while then I rebuilt to my patch after it merged without the tgt stuff | 13:23 |
* rhallisey reverts to old work | 13:23 | |
rhallisey | I figured it was something like that I just haven't had a change to try it this week | 13:23 |
sdake | i manually ran tgtd inside the container with no cli options | 13:24 |
sdake | and created a new cinder container | 13:24 |
sdake | and dont see the container via tgtadmin but atleast the iscsi target is available now | 13:24 |
sdake | rhallisey how do i display the volume group ifformation | 13:27 |
sdake | either inside or otusdide container | 13:27 |
rhallisey | sudo vgs | 13:27 |
rhallisey | you can't see it outside the container | 13:28 |
sdake | 2015-07-09 13:05:34.806 1 WARNING cinder.volume.drivers.lvm [req-dc5fd3af-706a-4e0f-9015-ce8494d541d0 - - - - -] LVMISCSIDriver is deprecated, you should now just use LVMVolumeDriver and specify iscsi_helper for the target driver you wish to use.^M | 13:28 |
rhallisey | /etc/lvm needs to be mounted | 13:28 |
sdake | deprecation warnings concern me becuase the deprecated code is no longer tested ;( | 13:29 |
rhallisey | that's fine for now, I always saw that | 13:29 |
rhallisey | yup I agree | 13:29 |
rhallisey | this patch was 'get cinder working' | 13:29 |
sdake | yes but we saw with keystone it break right during reelease | 13:29 |
sdake | a depreccation warning was the cause | 13:30 |
sdake | noentheless i dont see tgtd in here, expect to see that | 13:30 |
sdake | to mount /etc/lvm i need to bindmount it you mean? | 13:30 |
sdake | in the volume container? | 13:30 |
*** diogogmt has quit IRC | 13:31 | |
rhallisey | I forgot what I did, I think the data container | 13:31 |
sdake | vgs works without bindmounting in the volume container via docker exec | 13:31 |
sdake | how do i list out the details of the volume group | 13:31 |
sdake | [sdake@bigiron kolla]$ docker exec 41b vgs | 13:31 |
sdake | VG #PV #LV #SN Attr VSize VFree | 13:31 |
sdake | cinder-volumes 1 2 0 wz--n- 4.00g 1020.00m | 13:31 |
sdake | 2 logical volumes, what I would expect | 13:32 |
rhallisey | vgdisplay cinder-volumes | 13:32 |
rhallisey | ya you're out of space | 13:32 |
rhallisey | the backing file defaults to 4G | 13:32 |
sdake | http://ur1.ca/n2uic -> http://paste.fedoraproject.org/242260/44877314 | 13:33 |
sdake | i am not out of space yet ;) | 13:33 |
rhallisey | well if you create a new one | 13:33 |
rhallisey | I'm not sure what you're doing lol | 13:33 |
sdake | looks like lvm seems to think it has volumes | 13:33 |
sdake | i did create 1 folllowed by create 2 | 13:33 |
rhallisey | the reason I wanted to mount /etc/lvm is so the host can delete cinder-volumes | 13:33 |
sdake | but the volume belongs to cinder, f you really want to do something like that can't you just use docker exec? | 13:34 |
sdake | backing store size should be configurable btw ;-) | 13:34 |
rhallisey | sure, but I figured the work flow would be better if the volume deletion was done by something not involving docker | 13:35 |
SamYaple | wait the backig storage is a loopback file for cinder? | 13:35 |
sdake | [sdake@bigiron kolla]$ docker exec 41b cat /dev/cinder-volumes/volume-e044f890-4bb6-4ffb-9796-1037169d9013 | 13:35 |
rhallisey | ya, I don't have a physcial device setup to test that, daneyon was going to do it | 13:36 |
sdake | so that locks up | 13:36 |
SamYaple | ah ok that makes more sense | 13:36 |
sdake | meaning lvm gets stuck waiting on tgtd i assume | 13:36 |
rhallisey | sdake, I think tgt will fix it | 13:36 |
sdake | samyaple what is the concern there | 13:37 |
rhallisey | when I was scraping away stuff I must've forgot to rebuuld onetime during testing | 13:37 |
rhallisey | tricked my self into thinking I didn't need it | 13:37 |
sdake | cool ship happens have a patch? :) | 13:37 |
*** jtriley has joined #kolla | 13:39 | |
rhallisey | ya I do, give me a minute I have to build another package real quick | 13:40 |
rhallisey | not related | 13:40 |
sdake | t -2 weeks http://www.feickert.de/index.php?id=10&L=0 | 13:41 |
sdake | samyaple what is the concern with the loopback backing store? | 13:50 |
sdake | will that not work with the revert in docker 1.7? | 13:50 |
SamYaple | itll be super slow. i was just confirming that that wasnt the only option of backing storage | 13:51 |
sdake | it probably is the only option atm | 13:51 |
rhallisey | ya I think it is atm | 13:51 |
sdake | but ya performance will suck | 13:51 |
sdake | we havent tackled performance very much yet | 13:51 |
sdake | whta we really ened are natie volumes in docker :) | 13:51 |
rhallisey | let's call it 1.0. I'll ask daneyon to get on 2.0 with physical support | 13:51 |
rhallisey | or maybe I'll setup my env properly | 13:52 |
sdake | i think daneyon is going to be spending more time fousing on networing in magnum in the future | 13:52 |
sdake | so its unclear how much he will be able to assist here post liberty | 13:52 |
rhallisey | ok | 13:53 |
SamYaple | i want to focus on magnum networking :( | 13:53 |
SamYaple | i like networking | 13:53 |
sdake | anyway physical support is easy enough to tackle | 13:53 |
SamYaple | stupid ansible | 13:53 |
sdake | ya me too samyaple, i'd much rather be woring on magnum full time ;) | 13:53 |
SamYaple | to be fair, once Kolla gets settled, maintenance will be pretty easy | 13:54 |
sdake | i split my time betwent he communities but magnum gets the short end of the stick there | 13:54 |
rhallisey | SamYaple, yup | 13:54 |
sdake | rhallisey eta on patch? :) | 14:03 |
rhallisey | like 15min. Building selinux package. Just a moment | 14:04 |
* sdake rhallisey here is the bug you can use https://bugs.launchpad.net/kolla/+bug/1473090 | 14:05 | |
openstack | Launchpad bug 1473090 in kolla "tgtd is not running inside cinder-volume container" [Critical,Confirmed] - Assigned to Ryan Hallisey (rthall14) | 14:05 |
rhallisey | thx | 14:05 |
sdake | how do you run tgtd, just start it in the background? | 14:05 |
rhallisey | https://review.openstack.org/#/c/170965/21/docker/centos/binary/cinder-app/cinder-volume/start-scripts/tgt.sh | 14:06 |
sdake | or doesit take command line options? | 14:06 |
vincent_vdk | -f | 14:06 |
vincent_vdk | for foreground | 14:06 |
*** gfidente has quit IRC | 14:08 | |
sdake | rhallisey how doy ou delete the volume group so I can get to a fresh state | 14:11 |
inc0 | I'm going off, bye | 14:12 |
rhallisey | inc0, see ya | 14:12 |
rhallisey | sdake, you need to go into the container and do a vgremove cinder-volumes | 14:12 |
rhallisey | or docker exec | 14:12 |
*** inc0 has quit IRC | 14:12 | |
rhallisey | acutally you cant docker exec | 14:12 |
rhallisey | no I take that back | 14:12 |
rhallisey | exec into cinder volumes | 14:13 |
sdake | i got it | 14:13 |
sdake | docker exc works | 14:13 |
rhallisey | ya that works | 14:13 |
vincent_vdk | sdake: not when it's restarting i think | 14:13 |
vincent_vdk | just tried to do the same | 14:13 |
rhallisey | ya you cant' exec into a restarting container | 14:14 |
vincent_vdk | i'm getting ImportError: No module named oslo_reports | 14:19 |
vincent_vdk | in the cinder-volume | 14:19 |
SamYaple | ugh | 14:19 |
SamYaple | bad packages are bad | 14:19 |
SamYaple | source installs are da best installs | 14:19 |
vincent_vdk | did build with --release though | 14:19 |
rhallisey | vincent_vdk, ya I'll fix that too | 14:20 |
rhallisey | wait you see it in cinder volumes | 14:20 |
*** pbourke has quit IRC | 14:20 | |
vincent_vdk | yes | 14:20 |
rhallisey | weird I don't | 14:20 |
rhallisey | I see it in like heat-api-cfn | 14:20 |
*** gfidente has joined #kolla | 14:20 | |
* rhallisey rebuilds | 14:20 | |
*** pbourke has joined #kolla | 14:20 | |
*** davidjbc has joined #kolla | 14:21 | |
jmccarthy | I think I was asking about data volumes before - is that what would be preferred to backup the swift storage ? Any opinions on that (as opposed to data volume container I believe it was) | 14:22 |
pbourke | for the aio setup I'm thinking two instances of each storage service, each with a backing data vol | 14:25 |
pbourke | don't know how much sense that makes | 14:25 |
SamYaple | AIO can still have multiple copies of the data, even on a single disk | 14:27 |
sdake | rhallisey now cinder returns error on creating a cinder container | 14:27 |
rhallisey | so that's with tgt? What's the error | 14:27 |
sdake | http://ur1.ca/n2uzf -> http://paste.fedoraproject.org/242291/45210814 | 14:28 |
rhallisey | what does sudo vgs say | 14:29 |
sdake | it says nothing | 14:29 |
sdake | on the host | 14:29 |
sdake | i deleted the containers | 14:29 |
sdake | so it would have nothing in them as well | 14:29 |
sdake | [sdake@bigiron tools]$ sudo docker exec 04 vgs | 14:30 |
sdake | VG #PV #LV #SN Attr VSize VFree | 14:30 |
sdake | cinder-volumes 1 0 0 wz--n- 4.00g 4.00g | 14:30 |
pbourke | SamYaple: agree, just thought it might be more concrete to have two data vols. I'll see how it works out | 14:30 |
rhallisey | sdake, anything interesting in cinder-volume's logs? | 14:30 |
sdake | ya i'm reubilding - user error | 14:31 |
SamYaple | really need this merged guys https://review.openstack.org/#/c/198494/ | 14:34 |
rhallisey | SamYaple, let me put up some cinder stuff and I'll have look | 14:34 |
SamYaple | thanks rhallisey | 14:34 |
vincent_vdk | where are the oslo_reports installed | 14:35 |
vincent_vdk | can't seem to find it | 14:35 |
rhallisey | sdake, lol I found an old not on my desk that says 'doc for cinder' | 14:37 |
rhallisey | I moved desks recently so it got thrown around | 14:38 |
sdake | rhallisey can you review samyaple's request plz | 14:39 |
*** nihilifer has quit IRC | 14:39 | |
sdake | or jpeeler could do the job too if still around | 14:39 |
* jpeeler is around, just quiet | 14:39 | |
jpeeler | i realized one other reason i like mailing lists is because i can't keep up with IRC sometimes! | 14:39 |
jpeeler | and for the record, i'll be around now and all of next week. then i'll be gone for the rest of the month | 14:40 |
SamYaple | jpeeler: i agree we need to do more on the ML, but ML is slower too | 14:43 |
openstackgerrit | Ryan Hallisey proposed stackforge/kolla: Config-internal.sh and config-external.sh needs permissions https://review.openstack.org/200131 | 14:43 |
SamYaple | i think as Kolla settles the ML will be used alot more | 14:43 |
jpeeler | i know, tradeoffs | 14:43 |
* jpeeler looks at review | 14:43 | |
jpeeler | ansible... still don't know it heh | 14:43 |
*** dims has quit IRC | 14:45 | |
SamYaple | lurn | 14:45 |
sdake | rhallisey i already proposed that patch | 14:46 |
rhallisey | oh you did | 14:46 |
rhallisey | where is it? | 14:46 |
sdake | in the revie queue | 14:47 |
sdake | or in the irc log further up | 14:47 |
rhallisey | oh I se | 14:47 |
rhallisey | ya missed it | 14:47 |
coolsvap | sdake, do we need update to other kolla-standard start patches for permissions? | 14:47 |
jpeeler | SamYaple: after this review you want to look at non-voting new image job? | 14:47 |
sdake | coolsvap these appear to be the only ones broken that we start | 14:47 |
sdake | but we dont start all the containers we make iirc | 14:47 |
SamYaple | jpeeler: sure! | 14:48 |
rhallisey | sdake, update your patch and ill +2 | 14:48 |
sdake | what is wrong with it | 14:49 |
*** dims has joined #kolla | 14:50 | |
*** davidjc has joined #kolla | 14:50 | |
rhallisey | I guess config-external don't need +x | 14:50 |
coolsvap | sdake, we do not need permission change for config-external.sh | 14:50 |
coolsvap | sdake, regarding mongodb, do we need to add new scripts in common? | 14:51 |
*** davidjbc has quit IRC | 14:53 | |
SamYaple | rhallisey: correct | 14:54 |
openstackgerrit | Steven Dake proposed stackforge/kolla: Change permissions on cinder config-internal/external https://review.openstack.org/200055 | 14:56 |
openstackgerrit | Steven Dake proposed stackforge/kolla: tgtd fixup https://review.openstack.org/200139 | 14:56 |
sdake | i dont nkow if that cinder change works btw | 14:56 |
sdake | its wip | 14:56 |
rhallisey | sdake, kk | 14:56 |
sdake | it gets tgtd running in the system and can create a volume | 14:56 |
sdake | haven't tried booting one yet | 14:56 |
rhallisey | ya let me know if that works | 14:57 |
rhallisey | if not, may need to dive into some old patches | 14:57 |
sdake | i have hard time testing because of neutron | 14:57 |
sdake | can you test req like | 14:57 |
rhallisey | k | 14:57 |
jpeeler | i feel like gerrit has slowed down in the past few weeks | 15:01 |
*** sdake_ has joined #kolla | 15:01 | |
SamYaple | just in time for our build script! | 15:02 |
sdake_ | rhallisey you have a non-neutron env setup you can set right? | 15:04 |
sdake_ | test | 15:04 |
rhallisey | ya I do | 15:04 |
sdake_ | it is definately starting tgtd | 15:05 |
sdake_ | but tgtd isn't showing any volumes | 15:05 |
*** sdake has quit IRC | 15:05 | |
sdake_ | tgtdadmin | 15:05 |
*** absubram has joined #kolla | 15:05 | |
SamYaple | sdake_: feedback on the build script? | 15:07 |
sdake_ | you mean did i have any? | 15:07 |
SamYaple | yes | 15:07 |
sdake_ | it wont work with our gate as is | 15:07 |
sdake_ | needs an option to support extra text output | 15:07 |
sdake_ | it was missing --no-cche iirc | 15:08 |
sdake_ | --no-cache | 15:08 |
SamYaple | sdake_: no its there | 15:08 |
sdake_ | cool | 15:08 |
SamYaple | i think its '-c false' | 15:08 |
SamYaple | but yea it wont be a drop in replacement | 15:09 |
SamYaple | itll need its own gate that we promote to voting eventually | 15:09 |
*** sdake_ has quit IRC | 15:09 | |
*** diogogmt has joined #kolla | 15:09 | |
*** coolsvap is now known as coolsvap|away | 15:09 | |
*** sdake has joined #kolla | 15:10 | |
sdake | i'd like to dump our shell script as soon as feasible | 15:10 |
SamYaple | well we need a test gate first, jpeeler said we were going to look into that today. then once that is solid we can move | 15:12 |
jpeeler | yeah i'm going to look at what has changed now | 15:12 |
sdake | samyaple you will need to add the test output mode to build.py beforey ou can do a gate | 15:15 |
sdake | because the test cases look for it and the gate would fail | 15:16 |
SamYaple | sdake since it is all in python, i was hoping to get a bit more integrated | 15:16 |
jpeeler | yeah maybe it's time for a new test script | 15:16 |
* jpeeler is looking! | 15:16 | |
jpeeler | or sam can do it all heh | 15:16 |
sdake | i see well whatever works | 15:16 |
SamYaple | jpeeler: i would love to help, but i tried to understand project-config, I do not | 15:17 |
SamYaple | help would be appreciated | 15:17 |
sdake | project-config = sam's immovable object ;) | 15:17 |
sdake | or mayb eindustructible | 15:17 |
sdake | who nows | 15:17 |
sdake | back to bed for me :) | 15:17 |
sdake | damn k key | 15:17 |
sdake | typing kolla all day | 15:18 |
sdake | really damages the k key | 15:18 |
bmace | nite sdake :) | 15:18 |
sdake | bmace i was just joking about the back to bed - i just got up ;) | 15:18 |
bmace | the sun is up, time for you to go to bed. ;) | 15:18 |
sdake | my left eye is starting to twitch which is a sign of fatigue says my doctor so he wants me to keep normal hours for 1-2 weeks | 15:19 |
bmace | right when I get up it one of the times I most want to go back to bed. | 15:19 |
bmace | yeah, i have had that happen on quite a few occasions. stress / fatigue | 15:19 |
*** shardy_ has joined #kolla | 15:19 | |
sdake | he said its not typically stress but fatigue or eye strain | 15:19 |
sdake | he asked how long i stare ata computer screen | 15:19 |
sdake | I lied and he was still angry :) | 15:19 |
* jpeeler wonders what the lied amount was | 15:20 | |
*** dwalsh has quit IRC | 15:20 | |
sdake | jpeeler 12 hours | 15:20 |
SamYaple | sdake doesnt have 20/10 vision jpeeler | 15:20 |
*** shardy has quit IRC | 15:21 | |
sdake | he said 9 hours tops for next 1-2 weks | 15:21 |
sdake | i'm not particuarly stressed but i do feel tired | 15:22 |
sdake | so i think its definately fatgue in my case bmace | 15:22 |
sdake | he said to get my script checked out on my eyes too | 15:23 |
sdake | i'msure i'll get around to that - never | 15:23 |
*** shardy_ has quit IRC | 15:24 | |
*** shardy has joined #kolla | 15:25 | |
*** dims has quit IRC | 15:30 | |
*** dims has joined #kolla | 15:32 | |
*** dims has quit IRC | 15:32 | |
*** dims has joined #kolla | 15:33 | |
openstackgerrit | Steven Dake proposed stackforge/kolla: WIP: tgtd fixup https://review.openstack.org/200139 | 15:33 |
*** nihilifer has joined #kolla | 15:34 | |
*** nihilifer has quit IRC | 15:34 | |
*** nihilifer has joined #kolla | 15:35 | |
*** diogogmt has quit IRC | 15:35 | |
openstackgerrit | Sam Yaple proposed stackforge/kolla: Ansible Glance support https://review.openstack.org/198494 | 15:36 |
jpeeler | SamYaple: is there a way to build just one image? | 15:37 |
sdake | samyaple this needs a rebase as well https://review.openstack.org/#/c/199463/ | 15:37 |
sdake | jpeeler can you ac sam's rebave for glance there | 15:37 |
*** diogogmt has joined #kolla | 15:37 | |
pbourke | is there a way to diff patchsets in gerrit if the latest one has been rebased? | 15:38 |
sdake | pbourke not that i know of but if you find out how let me know :) | 15:38 |
pbourke | another one, is there a way to share data containers between compose files/ | 15:43 |
SamYaple | jpeeler: not at present | 15:43 |
SamYaple | it is in the TODO | 15:43 |
rhallisey | ugh... why can't I delete cinder-volumes | 15:43 |
SamYaple | jpeeler: it will probably be a regex, so glance would match all images with glance in it | 15:43 |
sdake | pbourke ther eare like 4 pull requests for sharing data in compose files none of them merged so no :) | 15:44 |
pbourke | :( | 15:45 |
sdake | they dont want to deal with versioning so they want the format to be perfect | 15:46 |
sdake | whatever that is... | 15:46 |
rhallisey | wow my docker is hosed | 15:46 |
sdake | rm -rf /var/lib/docker followed by reboot fixes it ;) | 15:46 |
sdake | samyaple a full rdo rebuild with no cache took 8 mins 35 seconds | 15:47 |
sdake | comapred to 45 minutes | 15:47 |
sdake | yay ;) | 15:47 |
SamYaple | yw | 15:47 |
SamYaple | try bumping up the threads too | 15:47 |
SamYaple | i use ./build.py -T1000 | 15:47 |
SamYaple | you know | 15:48 |
SamYaple | because of the Terminator | 15:48 |
sdake | hah | 15:48 |
*** daneyon has joined #kolla | 15:48 | |
sdake | that would probably terminate my machine, and i have 128gb ram | 15:48 |
SamYaple | it really helps on that final tier since it is 26 images and they dont actually do any processing | 15:48 |
SamYaple | my disk is the bottleneck for me | 15:48 |
SamYaple | sdake: its not _real_ threads, just how many builds itll do in parallel | 15:49 |
SamYaple | limited by the number per tier | 15:49 |
SamYaple | basically max ~30 at once | 15:49 |
sdake | i see | 15:49 |
sdake | i just ran -t 32 | 15:49 |
sdake | i'll let you now what the time looks like | 15:49 |
SamYaple | yup thll do ti | 15:49 |
sdake | -T 32 that is | 15:50 |
SamYaple | yea | 15:50 |
jmccarthy | samyaple: -T1000 lol | 15:52 |
sdake | 3 minute builds!! | 15:52 |
sdake | but i have 1+0 ssd plus xeon 16 core + 128gb rams :) | 15:53 |
sdake | so ymmv | 15:53 |
*** coolsvap|away is now known as coolsvap | 15:54 | |
SamYaple | sdake: I have just rebuilt my storage to check my speeds ;) | 15:54 |
SamYaple | thats what ive been doing all night | 15:55 |
SamYaple | sdake: that sounds like youre using cache though | 15:55 |
SamYaple | './build -T 32 -c False' | 15:56 |
SamYaple | ? | 15:56 |
coolsvap | SamYaple, ping | 15:56 |
SamYaple | pong | 15:56 |
rhallisey | brb | 15:56 |
sdake | -c false is cmd line option | 15:56 |
*** rhallisey has quit IRC | 15:56 | |
sdake | oh let me try capital False | 15:56 |
coolsvap | regarding the patchsets for magnum and zaqar for standard start | 15:56 |
SamYaple | idk if that matters sdake | 15:56 |
coolsvap | the pep8 issues are due to variables not present in enironment file | 15:57 |
sdake | coolsvap run min-env.sh | 15:57 |
sdake | like I told you already yesterday :) | 15:57 |
coolsvap | sdake, i ran the same | 15:57 |
sdake | it will create the env files | 15:57 |
coolsvap | and it updates the env file | 15:57 |
coolsvap | SamYaple has -1 on the env file change | 15:57 |
*** dwalsh has joined #kolla | 15:58 | |
sdake | coolsvap leave comments in the review | 15:58 |
*** rhallisey has joined #kolla | 15:58 | |
sdake | 1 min 58 seconds that time samyaple | 15:58 |
SamYaple | sdake: check out the env file it generates | 15:59 |
sdake | maybe -c Falle doesn't wor | 15:59 |
sdake | SamYaple I already saw it | 15:59 |
SamYaple | sdake: im pretty sure your using cache. that might be busted | 15:59 |
SamYaple | ill remove the -1, my concern was the env being modifed | 15:59 |
sdake | i think it is becuase the external environment variables are not being parsed | 15:59 |
sdake | by the shell scripts | 15:59 |
SamYaple | or because its parsing start.sh not config-internal.sh | 15:59 |
sdake | because of errors in config-intiernal | 15:59 |
coolsvap | SamYaple, do we need to udpate the min-env then? | 16:00 |
SamYaple | coolsvap: if this is our autosciprt generating this im fine with the change | 16:00 |
SamYaple | we can fix the script later if we find its broke | 16:00 |
coolsvap | SamYaple, regarding openvswitch, do we need to keep the file blank for config-external? | 16:01 |
SamYaple | coolsvap: yea, with just a little comment saying "intentially blank" or soemthing | 16:02 |
coolsvap | SamYaple, alright | 16:02 |
SamYaple | sdake: it looks like './build.py -cFalse' does the trick | 16:02 |
SamYaple | still figuring out argparser | 16:03 |
openstackgerrit | Swapnil Kulkarni proposed stackforge/kolla: Add config-internal/config-external to OpenvSwitch containers https://review.openstack.org/198242 | 16:03 |
*** bitblt has joined #kolla | 16:03 | |
SamYaple | sdake: i have an equally powerfull machine, 8 minutes sounds like a real build | 16:04 |
coolsvap | SamYaple, done | 16:05 |
coolsvap | sdake, regarding mongodb, do we need to add new scripts in common? | 16:05 |
sdake | samyaple still2 2 minutes | 16:05 |
*** dims_ has joined #kolla | 16:05 | |
SamYaple | sdake: gotta be using the cache. dump all of your images and try again | 16:06 |
harmw | dims: ping | 16:07 |
harmw | no, diga_ | 16:07 |
harmw | sry | 16:07 |
harmw | diga_: https://review.openstack.org/#/c/198242/ you can't be author and co-author at the same time :)\ | 16:07 |
SamYaple | he can if he wants to! | 16:08 |
harmw | only if he's scizophrenic | 16:08 |
harmw | and even then | 16:08 |
SamYaple | diga_: you can do anything you set your minds too | 16:08 |
harmw | aw jeez | 16:08 |
harmw | here we go with the Disney-talk | 16:08 |
diga_ | harmw: hi | 16:09 |
openstackgerrit | Swapnil Kulkarni proposed stackforge/kolla: Add config-internal/config-external for barbican https://review.openstack.org/199728 | 16:09 |
*** dims has quit IRC | 16:09 | |
*** vinkman has joined #kolla | 16:09 | |
harmw | diga_: you mind removing that or should I just +2 it? | 16:09 |
coolsvap | harmw, i updated the change-set | 16:10 |
coolsvap | seems all standard-start patchsets failing for rebase | 16:10 |
diga_ | harmw: you can take a call | 16:10 |
coolsvap | :-/ | 16:10 |
harmw | coolsvap: which one? | 16:10 |
coolsvap | earlier barbican | 16:11 |
coolsvap | now ceilometer | 16:11 |
harmw | ah | 16:11 |
coolsvap | same goes for magnum | 16:11 |
coolsvap | zaqar | 16:11 |
harmw | Ididn't notice Jenkins hadn't check it | 16:11 |
*** bmace has quit IRC | 16:11 | |
harmw | diga_: you mind rebasing first, since that apparently seems neccesary | 16:11 |
harmw | right coolsvap ? | 16:12 |
coolsvap | harmw, i will do, i have all the branches | 16:12 |
harmw | cool, and fix the co-author while at it as well :P | 16:12 |
diga_ | coolsvap: please go ahead | 16:12 |
coolsvap | harmw, sure will do :) | 16:12 |
openstackgerrit | Swapnil Kulkarni proposed stackforge/kolla: Add config-internal/config-external for Ceilometer https://review.openstack.org/198241 | 16:13 |
openstackgerrit | Swapnil Kulkarni proposed stackforge/kolla: Add config-internal/config-external to Magnum containers https://review.openstack.org/198213 | 16:14 |
harmw | why isn't Jenkins marking https://review.openstack.org/#/c/198241/ as failed? | 16:14 |
openstackgerrit | Swapnil Kulkarni proposed stackforge/kolla: Add config-internal/config-external to Zaqar containers https://review.openstack.org/198284 | 16:15 |
openstackgerrit | Swapnil Kulkarni proposed stackforge/kolla: Add config-internal/config-external to OpenvSwitch containers https://review.openstack.org/198242 | 16:15 |
coolsvap | harmw, yes it should mark | 16:16 |
harmw | only because you submitted a new patch | 16:16 |
harmw | SamYaple: https://review.openstack.org/#/q/status:open+project:stackforge/kolla+branch:master+topic:bp/standard-start,n,z | 16:16 |
coolsvap | but it gave error | 16:16 |
harmw | if you could ack those | 16:16 |
SamYaple | sdake: proposed idea for the builder was to use tmux to tail in real time the build for all ongoing builds. If it is successfully close the window, if it fails keep it open | 16:16 |
harmw | coolsvap: yes, but nonetheless Jenkins added it's checkmark | 16:17 |
openstackgerrit | Swapnil Kulkarni proposed stackforge/kolla: horizon start scripts need update for cmd & args https://review.openstack.org/199411 | 16:17 |
coolsvap | harmw, yes, but earlier it used to mark it failed for merge i think | 16:17 |
openstackgerrit | Merged stackforge/kolla: Ansible Glance support https://review.openstack.org/198494 | 16:19 |
*** bmace has joined #kolla | 16:21 | |
harmw | pbourke: ping | 16:22 |
*** davidjc has quit IRC | 16:28 | |
*** bmace has quit IRC | 16:33 | |
pbourke | hi | 16:34 |
pbourke | harmw: o/ | 16:34 |
harmw | yo | 16:34 |
harmw | just commented on something Oracle | 16:34 |
pbourke | a review? | 16:35 |
harmw | yup | 16:35 |
pbourke | ah I see now | 16:35 |
pbourke | ill reply on the review | 16:35 |
harmw | great | 16:35 |
*** coolsvap is now known as coolsvap|away | 16:36 | |
harmw | sdake: https://review.openstack.org/#/c/198520/ could you ack on that please! | 16:38 |
*** cbaesema has joined #kolla | 16:41 | |
*** bmace has joined #kolla | 16:44 | |
openstackgerrit | Paul Bourke proposed stackforge/kolla: Add source base image for oraclelinux https://review.openstack.org/200013 | 16:44 |
openstackgerrit | Paul Bourke proposed stackforge/kolla: Add source base image for oraclelinux https://review.openstack.org/200013 | 16:45 |
*** shardy has quit IRC | 16:45 | |
*** bitblt has quit IRC | 16:45 | |
harmw | pbourke: what I meant was starting the list with Mysql-bla, but thats fine | 16:46 |
pbourke | doh :) | 16:46 |
pbourke | harmw: I use !sort in vim usually for that kind of thing | 16:46 |
harmw | you could've added a comment in there about this beeing an Oracle 'issue' | 16:46 |
pbourke | I'll do that now, np | 16:46 |
harmw | nah its fine | 16:46 |
*** jruano has quit IRC | 16:46 | |
pbourke | no I aim to please ;) | 16:46 |
harmw | hehe | 16:47 |
openstackgerrit | Paul Bourke proposed stackforge/kolla: Add source base image for oraclelinux https://review.openstack.org/200013 | 16:48 |
openstackgerrit | Paul Bourke proposed stackforge/kolla: Add source keystone image for oraclelinux https://review.openstack.org/191833 | 16:50 |
*** jmccarthy has quit IRC | 17:02 | |
*** coolsvap|away is now known as coolsvap | 17:08 | |
*** sdake has quit IRC | 17:16 | |
openstackgerrit | Michal Rostecki proposed stackforge/kolla: Add RabbitMQ image for Ubuntu https://review.openstack.org/200183 | 17:20 |
*** sdake has joined #kolla | 17:20 | |
openstackgerrit | Michal Rostecki proposed stackforge/kolla: Add RabbitMQ image for Ubuntu https://review.openstack.org/200183 | 17:23 |
*** sdake_ has joined #kolla | 17:26 | |
openstackgerrit | Michal Rostecki proposed stackforge/kolla: Add Keystone image for Ubuntu https://review.openstack.org/196424 | 17:27 |
*** nihilifer has quit IRC | 17:29 | |
*** sdake has quit IRC | 17:29 | |
*** sdake has joined #kolla | 17:30 | |
*** sdake_ has quit IRC | 17:34 | |
*** dims_ has quit IRC | 17:35 | |
*** dims has joined #kolla | 17:37 | |
*** loth has joined #kolla | 17:40 | |
*** Slower has quit IRC | 18:05 | |
sdake | rhallisey i took a midmorning nap, anything come out of testing of cinder? | 18:12 |
rhallisey | sdake, well I found that you cant rm the container when it's running | 18:13 |
sdake | wont rm or system implodes? | 18:14 |
rhallisey | I've been doing a bunch of other stuff so I'm slowing picking away at it | 18:14 |
sdake | note i had a new update up with working tgtd.sh file | 18:14 |
rhallisey | ya I had already added that | 18:14 |
rhallisey | I'm working off an old pathc | 18:14 |
rhallisey | patch | 18:15 |
rhallisey | once I revert back it should work again | 18:16 |
rhallisey | I'm pretty confident that's the issue | 18:16 |
rhallisey | so I'm not worried | 18:16 |
rhallisey | but I'll have it up soon | 18:16 |
*** bmace has quit IRC | 18:17 | |
*** sdake is now known as sdae | 18:19 | |
*** sdae is now known as sdake | 18:19 | |
*** bmace has joined #kolla | 18:20 | |
*** dims has quit IRC | 18:21 | |
*** jasonsb has quit IRC | 18:26 | |
*** jasonsb has joined #kolla | 18:26 | |
*** jasonsb has quit IRC | 18:31 | |
*** sdake_ has joined #kolla | 18:33 | |
*** gfidente has quit IRC | 18:34 | |
openstackgerrit | Merged stackforge/kolla: Add source base image for oraclelinux https://review.openstack.org/200013 | 18:35 |
*** sdake has quit IRC | 18:36 | |
*** sdake has joined #kolla | 18:37 | |
*** sdake_ has quit IRC | 18:40 | |
sdake | samyaple around? | 18:57 |
openstackgerrit | Merged stackforge/kolla: Add config-internal/config-external for Ceilometer https://review.openstack.org/198241 | 18:59 |
*** absubram has quit IRC | 19:00 | |
*** coolsvap is now known as coolsvap|away | 19:04 | |
openstackgerrit | Merged stackforge/kolla: horizon start scripts need update for cmd & args https://review.openstack.org/199411 | 19:09 |
*** dims has joined #kolla | 19:10 | |
*** ogelbukh1 has left #kolla | 19:15 | |
openstackgerrit | Steven Dake proposed stackforge/kolla: Change permissions on cinder config-internal/external https://review.openstack.org/200055 | 19:16 |
openstackgerrit | Steven Dake proposed stackforge/kolla: WIP: tgtd fixup https://review.openstack.org/200139 | 19:16 |
*** jasonsb has joined #kolla | 19:19 | |
*** sdake has quit IRC | 19:25 | |
*** sdake has joined #kolla | 19:29 | |
*** gfidente has joined #kolla | 19:33 | |
*** gfidente has quit IRC | 19:33 | |
*** gfidente has joined #kolla | 19:33 | |
*** davidjc has joined #kolla | 20:00 | |
*** daneyon has quit IRC | 20:14 | |
*** sdake_ has joined #kolla | 20:16 | |
*** sdake has quit IRC | 20:19 | |
*** athomas has quit IRC | 20:25 | |
*** sdake_ is now known as sdake | 20:29 | |
openstackgerrit | Imran Hayder Malik proposed stackforge/kolla: Fix grammatical errors in asible docs https://review.openstack.org/200255 | 20:45 |
*** jtriley has quit IRC | 20:49 | |
*** bmace has quit IRC | 20:56 | |
*** jtriley has joined #kolla | 20:57 | |
sdake | harmw comment in there for you | 21:02 |
sdake | answer is no - not required | 21:02 |
*** dims has quit IRC | 21:05 | |
*** dims has joined #kolla | 21:07 | |
*** jtriley has quit IRC | 21:10 | |
*** rhallisey has quit IRC | 21:19 | |
openstackgerrit | Merged stackforge/kolla: Fix grammatical errors in asible docs https://review.openstack.org/200255 | 21:27 |
sdake | yay new contributor gets code into oepnstac ;) | 21:28 |
*** jruano has joined #kolla | 21:33 | |
*** slagle has quit IRC | 21:36 | |
*** shadower has quit IRC | 21:36 | |
*** sdake_ has joined #kolla | 21:39 | |
sdake_ | coolsvap|afk when your not afk ping me | 21:41 |
*** sdake has quit IRC | 21:42 | |
*** daneyon has joined #kolla | 21:46 | |
openstackgerrit | Steven Dake proposed stackforge/kolla: WIP: Start tgtd in cinder-volume container https://review.openstack.org/200139 | 21:48 |
*** jpeeler has quit IRC | 21:48 | |
*** daneyon_ has joined #kolla | 21:49 | |
*** daneyon has quit IRC | 21:52 | |
*** daneyon has joined #kolla | 21:56 | |
*** daneyon_ has quit IRC | 22:00 | |
*** daneyon_ has joined #kolla | 22:01 | |
*** jruano has quit IRC | 22:02 | |
*** dims has quit IRC | 22:03 | |
*** dims has joined #kolla | 22:03 | |
*** daneyon has quit IRC | 22:04 | |
*** coolsvap|away is now known as coolsvap | 22:05 | |
*** dwalsh has quit IRC | 22:05 | |
*** jtriley has joined #kolla | 22:07 | |
coolsvap | sdake_, ping | 22:07 |
sdake_ | say I have an easy project for you if interested | 22:07 |
sdake_ | kolla and cleanup-containers neither require root to run | 22:07 |
sdake_ | but they both wont run without root | 22:08 |
sdake_ | (they check on start) | 22:08 |
sdake_ | need bugs filed to fix these and remove the checks | 22:08 |
sdake_ | interested ? | 22:08 |
sdake_ | i'll sort out our future plans with ansible this weekend | 22:08 |
sdake_ | I'm not quite sure wht the next step there is, we are waiting on sam a bit and reviews of the primary services he isworking on | 22:08 |
sdake_ | and then I want to distribute that work among the team | 22:09 |
coolsvap | sure, I can do that | 22:09 |
sdake_ | cool file bugs first and link them so i can set them propely in the release | 22:09 |
sdake_ | a bug for each plz | 22:10 |
coolsvap | alright, will do that | 22:11 |
*** daneyon_ has quit IRC | 22:11 | |
sdake_ | a non-cached build takes 23 minutes samyaple | 22:14 |
sdake_ | I tested by setting cache to false inside the code | 22:15 |
*** jpeeler has joined #kolla | 22:16 | |
*** prad has joined #kolla | 22:24 | |
*** jasonsb has quit IRC | 22:25 | |
*** jasonsb has joined #kolla | 22:26 | |
*** gfidente has quit IRC | 22:32 | |
*** achanda has joined #kolla | 22:33 | |
*** gfidente has joined #kolla | 22:34 | |
coolsvap | sdake_, can you give workflow+ for standard-start patches, i had to rebase all today since the directory changes in repo | 22:39 |
sdake_ | i can +2 but not w+, requires two reviewers | 22:40 |
sdake_ | if you have links that will speed up the process | 22:40 |
coolsvap | https://review.openstack.org/#/q/status:open+project:stackforge/kolla+branch:master+topic:bp/standard-start,n,z | 22:41 |
coolsvap | all already have +2 | 22:41 |
openstackgerrit | Steven Dake proposed stackforge/kolla: WIP: Start tgtd in cinder-volume container https://review.openstack.org/200139 | 22:42 |
openstackgerrit | Steven Dake proposed stackforge/kolla: Add oslo.reports as a global dependency https://review.openstack.org/200298 | 22:42 |
sdake_ | oh ya i can give workflow on thosse | 22:43 |
openstackgerrit | Merged stackforge/kolla: Add config-internal/config-external to OpenvSwitch containers https://review.openstack.org/198242 | 22:44 |
openstackgerrit | Merged stackforge/kolla: Add config-internal/config-external to Zaqar containers https://review.openstack.org/198284 | 22:44 |
openstackgerrit | Merged stackforge/kolla: Add config-internal/config-external to Magnum containers https://review.openstack.org/198213 | 22:45 |
*** jtriley has quit IRC | 22:45 | |
openstackgerrit | Merged stackforge/kolla: Add config-internal/config-external for barbican https://review.openstack.org/199728 | 22:46 |
*** bmace has joined #kolla | 22:46 | |
*** diogogmt has quit IRC | 22:58 | |
*** jasonsb has quit IRC | 23:04 | |
*** jasonsb has joined #kolla | 23:05 | |
*** prad has quit IRC | 23:08 | |
sdake_ | anyone have ideas on this bug: | 23:12 |
sdake_ | https://bugs.launchpad.net/kolla/+bug/1473242 | 23:12 |
openstack | Launchpad bug 1473242 in kolla "nova-api can't find KS-ODM/services" [Critical,Triaged] | 23:12 |
*** coolsvap is now known as coolsvap|afk | 23:14 | |
*** sdake has joined #kolla | 23:24 | |
*** sdake__ has joined #kolla | 23:26 | |
openstackgerrit | Steven Dake proposed stackforge/kolla: Add oslo.reports as a global dependency https://review.openstack.org/200298 | 23:27 |
openstackgerrit | Steven Dake proposed stackforge/kolla: WIP: Start tgtd in cinder-volume container https://review.openstack.org/200139 | 23:27 |
*** sdake_ has quit IRC | 23:27 | |
*** sdake has quit IRC | 23:29 | |
*** sdake__ is now known as sdake | 23:30 | |
*** vinkman1 has joined #kolla | 23:37 | |
*** vinkman has quit IRC | 23:37 | |
*** openstack has joined #kolla | 23:40 | |
*** diga has joined #kolla | 23:43 | |
*** vinkman1 has left #kolla | 23:46 | |
*** achanda has quit IRC | 23:49 | |
*** davidjc has quit IRC | 23:54 |
Generated by irclog2html.py 2.14.0 by Marius Gedminas - find it at mg.pov.lt!