*** dimsum__ has quit IRC | 00:03 | |
openstackgerrit | Merged stackforge/kolla: Add missing node types to Vagrantfile https://review.openstack.org/220955 | 00:06 |
---|---|---|
*** sdake has joined #kolla | 00:18 | |
sdake | coolsvap ping re ceilometer | 00:19 |
*** weiyu has joined #kolla | 00:22 | |
*** banix has joined #kolla | 00:23 | |
*** banix has quit IRC | 00:27 | |
*** jtriley has joined #kolla | 00:30 | |
*** jtriley has quit IRC | 00:35 | |
*** weiyu has quit IRC | 00:37 | |
*** weiyu has joined #kolla | 00:39 | |
*** akwasnie has quit IRC | 00:39 | |
*** dimsum__ has joined #kolla | 00:48 | |
*** weiyu has quit IRC | 00:59 | |
*** weiyu has joined #kolla | 01:06 | |
*** weiyu has quit IRC | 01:15 | |
openstackgerrit | Merged stackforge/kolla: Add rsyncd image and related Ansible for Swift https://review.openstack.org/220172 | 01:15 |
*** alisonh has quit IRC | 01:21 | |
*** zhiwei has joined #kolla | 01:25 | |
*** erkules_ has joined #kolla | 01:30 | |
*** dolpher has joined #kolla | 01:30 | |
kjelly | Hi guys, I use `tools/build.py -n 172.22.2.81:5000/kollaglue --base ubuntu --type source --push` to build images. And the deployment failed again when create glance endpoint. The reason why it can't create glance endpoint is that there are no table in keystone database. I try many ways and always get the same error. What's wrong for me? | 01:31 |
kjelly | I use the command to deploy, "ansible-playbook -i inventory/all-in-one -e @/etc/kolla/globals.yml -e @/etc/kolla/passwords.yml site.yml -vvvv" | 01:32 |
kjelly | using root. | 01:32 |
*** erkules has quit IRC | 01:33 | |
*** banix has joined #kolla | 01:39 | |
*** alisonh has joined #kolla | 01:39 | |
*** dimsum__ has quit IRC | 01:40 | |
sdake | kjelly what ip are you deployign to? | 01:40 |
sdake | if your deploying to 172.22.2.81, that is where keystone is running | 01:41 |
sdake | your also running your registry there | 01:41 |
sdake | this will cause the ports to collide | 01:41 |
sdake | that would be my best guess as to why keystone dbs are not found | 01:41 |
sdake | in other words is your registry running on the host your deploying from? | 01:42 |
sdake | if the answer is yes, chagne your registry port | 01:43 |
*** weiyu has joined #kolla | 01:46 | |
*** dolpher1 has joined #kolla | 01:46 | |
*** dolpher has quit IRC | 01:48 | |
kjelly | sdake: The ip which is used for deploy openstack is different from docker registry machine. | 01:50 |
kjelly | kjelly: I deploy openstack to virtual machine. | 01:50 |
sdake | what ip do you use for the internal kolla address? | 01:51 |
sdake | the one in globals.yml | 01:51 |
*** alisonh has quit IRC | 01:51 | |
sdake | it has to be an unused ip address - it becomes a virtual ip addrss | 01:52 |
kjelly | kolla_internal_address: "10.10.10.254" | 01:52 |
sdake | from your host try telnet 10.10.10.254 3306 | 01:52 |
sdake | see if that connects you to the database | 01:53 |
kjelly | my virtual machine is 10.0.2.15 | 01:53 |
sdake | your internal address i think needs to be in the same subnet as your vm | 01:53 |
sdake | set yur internal address to 10.0.2.254 | 01:54 |
kjelly | I connect to 3306 and display 5.5.5-10.0.21-MariaDB-1~trusty-wsrep-logxq3FEh[o?�.b)@H)4e&+BLmysql_native_password | 01:54 |
sdake | ok that looks good | 01:54 |
kjelly | How to make sure the code is executed ? https://github.com/stackforge/kolla/blob/master/docker/keystone/start.sh#L26 | 01:55 |
kjelly | It seems that the code is not executed. | 01:55 |
sdake | that is one explination or another is your database cannot be accessed | 01:56 |
*** alisonh has joined #kolla | 01:56 | |
sdake | why not just deploy to your bare meatl machine? | 01:56 |
sdake | i would still try setting your internal address to 10.0.2.254 | 01:56 |
sdake | and delete everything and start fresh in the vm | 01:57 |
kjelly | I have no bare meatl machine | 01:57 |
sdake | you can use cleanup-containers -fv in the vm to do the job | 01:57 |
kjelly | Ok. | 01:58 |
kjelly | Try it again. | 01:58 |
sdake | note if your bootstrap failed previosuly kolla isn't smart enough to detect this and run it again | 01:59 |
sdake | so if you never cleaned up your vm prior and were running into deployment problems you may experience this problem | 01:59 |
kjelly | I see. | 02:00 |
sdake | cleanup-containers is in the tools directory | 02:00 |
sdake | please paste your globals.yaml | 02:01 |
*** jtriley has joined #kolla | 02:01 | |
sdake | kjelly ^^ | 02:03 |
kjelly | sdake: wait a moment. I close the virtual machine. | 02:03 |
kjelly | sdake: sorry | 02:04 |
kjelly | sdake: here is my gloal.yml https://gist.github.com/ya790206/873d67ef7decf902e48f | 02:06 |
*** jtriley has quit IRC | 02:06 | |
sdake | i think your configisn't correct | 02:07 |
sdake | let me past emine | 02:07 |
sdake | http://ur1.ca/npj5n -> http://paste.fedoraproject.org/264629/14416780 | 02:07 |
sdake | also docker on your vm must be set to use an inseucure registry | 02:07 |
sdake | http://ur1.ca/npj6h -> http://paste.fedoraproject.org/264630/16781381 | 02:09 |
sdake | my etc/sysconfig/docker file | 02:09 |
kjelly | my docker config in /etc/default/docker is DOCKER_OPTS="--dns 8.8.8.8 --dns 8.8.4.4 --insecure-registry 172.22.2.81:5000" | 02:10 |
sdake | the default for insecure registry is false | 02:12 |
sdake | you need to override it to true iirc | 02:12 |
sdake | see line 40 in my globals.yml paste | 02:12 |
kjelly | Ok, I change my settings. | 02:14 |
kjelly | By the way, what is the ip of p1p1, em1 in your machine ? | 02:17 |
sdake | em1 is set to 192.168.1.100 | 02:19 |
sdake | on the bar emetal machines its 192.168.1.101-103 | 02:19 |
sdake | p1p1 is unset - and should be in your vm | 02:20 |
kjelly | Ok. | 02:23 |
sdake | kjelly did that get you rollilng? | 02:32 |
kjelly | sdake: The problem still exists. | 02:34 |
sdake | and you ran cleanup-containers fv on the vm? | 02:35 |
sdake | -fv ? | 02:35 |
kjelly | I tried -f. Now I try -fv | 02:35 |
sdake | that won't make much difference | 02:36 |
sdake | why not deploy to your machine your deploying from? | 02:36 |
sdake | if yur using the allinone did you modify ansible/inventory/allinone? | 02:36 |
kjelly | No, I never change the kolla code. | 02:37 |
sdake | ok well your not deploying to your vm then | 02:37 |
sdake | your deploying to where you are running ansible | 02:37 |
sdake | or is that also your vm? | 02:38 |
*** klint has joined #kolla | 02:38 | |
kjelly | sdake: The machine which is used for deploy openstack is the same machine which is used for openstack | 02:38 |
sdake | and that is different from where your running your registry? | 02:39 |
kjelly | sdake: yes. | 02:40 |
sdake | is adatabase already running on your vm? | 02:40 |
*** dimsum__ has joined #kolla | 02:40 | |
sdake | or any other server software? | 02:40 |
kjelly | there are no other service. just ssh, cupsd, dnsmsq | 02:42 |
sdake | how much disk space? | 02:44 |
kjelly | sdake: I added ansible deployment message to here https://gist.github.com/ya790206/873d67ef7decf902e48f | 02:44 |
kjelly | disk: /dev/sda1 105G 7.7G 92G 8% / | 02:45 |
*** dimsum__ has quit IRC | 02:46 | |
SamYaple | hey sdake inc0 is looking for a response on te rsyslog ansible stuff | 02:48 |
sdake | kjelly run openstack endpoint list | 02:48 |
SamYaple | im not fighting about it anymore but he seems to think you are ok with bindmounting dev onto the host filesystem | 02:48 |
sdake | after sourcing openrc | 02:48 |
sdake | see if that shows anything | 02:48 |
SamYaple | sdake: if they keystone database is empty that will do nothing | 02:48 |
SamYaple | if the database is empty its not bootstrapping correctly | 02:49 |
sdake | yes sam i know i want to verify the database is actually empty | 02:49 |
kjelly | sdake: Ok. I try it. | 02:51 |
sdake | the openrc can be retrieved from tools directory openrc-example | 02:51 |
kjelly | wait for a moment because I delete all container. | 02:53 |
sdake | kjelly one problem could be - is a fiewall enabled on the vm? | 02:57 |
kjelly | sudo ufw status | 02:58 |
kjelly | Status: inactive | 02:58 |
kjelly | [+1][root:~/kolla/tools] # keystone endpoint-list | 02:58 |
kjelly | Authorization Failed: The resource could not be found. (HTTP 404) | 02:58 |
sdake | no, openstack endpoint list | 02:58 |
sdake | keystone client doesn't work at all with kolla | 02:58 |
kjelly | I don't have openstack command, or run it in container ? | 02:59 |
sdake | sudo pip install ptyhon-openstackclient | 02:59 |
kjelly | [+1]⌂91% [root:~/kolla/tools] 2 # openstack endpoint list | 03:04 |
kjelly | ERROR: openstack An unexpected error prevented the server from fulfilling your request. (HTTP 500) (Request-ID: req-bcfb82da-853b-4d9c-95ae-b43acbf872c6) | 03:04 |
sdake | docker logs keystone | 03:04 |
SamYaple | sdake: the keystone logs are in apache in the container | 03:04 |
sdake | oh right | 03:05 |
sdake | sam do you know the file locations for the logs on ubuntu | 03:05 |
SamYaple | /var/log/apache2/keystone.log | 03:05 |
sdake | kjelly docker exec keystone cat /var/log/apache2/keysotne.log > log | 03:05 |
sdake | paste the log | 03:06 |
sdake | also pip show ansible | 03:06 |
sdake | tell me which version your running | 03:06 |
kjelly | sdake: here https://gist.github.com/ya790206/873d67ef7decf902e48f . the latest one. | 03:06 |
kjelly | ansible --version | 03:07 |
kjelly | ansible 1.9.2 | 03:07 |
kjelly | configured module search path = None | 03:07 |
sdake | ya keystone database didn't initialize | 03:08 |
sdake | right there in the logs | 03:08 |
kjelly | It seems that the code is not execued. How to make sure the code is executed? https://github.com/stackforge/kolla/blob/master/docker/keystone/start.sh#L26 | 03:09 |
sdake | kjelly lets get the logs from the bootstrap container | 03:10 |
sdake | https://github.com/stackforge/kolla/blob/master/ansible/roles/keystone/tasks/bootstrap.yml#L68 | 03:11 |
sdake | delete lines 68-73 | 03:12 |
sdake | that should leave the bootstrap container intact | 03:12 |
sdake | then after a run do docker logs bootstrap_keystone | 03:13 |
SamYaple | do not delete lines | 03:14 |
SamYaple | thats awful | 03:14 |
SamYaple | run with --step | 03:14 |
SamYaple | it will promt you before each task | 03:14 |
sdake | kjelly do whatevero u like each approach will work - what we are after is the bootstrap logs | 03:15 |
kjelly | stop before deleting bootstrap container? | 03:16 |
sdake | yes | 03:16 |
SamYaple | sdake please dont give advice to modify the playbooks. thats going to render all results past that point suspect | 03:16 |
sdake | "cleaning up bootstrap container" I think it is | 03:16 |
sdake | for keystone | 03:16 |
sdake | all results past that point are suspet anyway | 03:17 |
sdake | but i was not aware of the --step command | 03:17 |
SamYaple | you are because youve used it | 03:17 |
sdake | never have | 03:18 |
SamYaple | i have irc logs that say otherwise | 03:18 |
kjelly | [+1][root:~/kolla] 130 # docker logs bootstrap_keystone | 03:19 |
kjelly | su: System error | 03:19 |
sdake | cool waste some more time and pull em up then | 03:19 |
SamYaple | kjelly: you have a bad kernel | 03:19 |
SamYaple | kjelly: some kernels have trouble with su inside a container | 03:19 |
SamYaple | kjelly: whats your kernel version? | 03:19 |
kjelly | Do yo | 03:20 |
kjelly | uname -a | 03:20 |
kjelly | Linux user-VirtualBox 3.13.0-24-generic #47-Ubuntu SMP Fri May 2 23:30:00 UTC 2014 x86_64 x86_64 x86_64 GNU/Linux | 03:20 |
SamYaple | thats a really old kernel | 03:21 |
SamYaple | try upgrading | 03:21 |
SamYaple | the latest is like 54 or something | 03:21 |
SamYaple | latest trusty kernel is linux-image-3.13.0-63-generic | 03:21 |
kjelly | Ok, try it . | 03:21 |
sdake | is -24 the standard after installing ubuntu? | 03:22 |
sdake | kjelly if that works, would you mind putting in the docs pls? | 03:22 |
kjelly | sdake: I will. If I deploy successfully, I will update the document. | 03:23 |
sdake | tx | 03:23 |
SamYaple | its not the standard, the standard is installing the latest kernel | 03:24 |
SamYaple | its a really old image thats being used | 03:24 |
SamYaple | probably hasnt been rebuilt since trusty initially was released based on that build time | 03:24 |
sdake | sam is there any way to make the bootstrap failure return in asnible a failure ? | 03:25 |
sdake | rather then continuing on like nothing went wrong ;) | 03:25 |
SamYaple | we set errexit | 03:26 |
SamYaple | the problem is this is with the kernel | 03:26 |
SamYaple | so no | 03:26 |
*** dayou has joined #kolla | 03:27 | |
*** banix has quit IRC | 03:31 | |
sdake | SamYaple nice job on the network thing | 03:32 |
sdake | samyaple deployment works great now :) | 03:33 |
SamYaple | its a bit annoying to me to have to do that, but this way we dont have to do ansible config changes | 03:33 |
SamYaple | fact caching solves the issue | 03:33 |
SamYaple | sdake: is anyone working on the kolla-ansible util? | 03:34 |
sdake | there is a guy who contaced me who wnts to work on it | 03:34 |
SamYaple | it needs to be python'd up in like 18 days | 03:34 |
sdake | he is assigned to it | 03:34 |
sdake | i dont know if he is making any progress with it or not | 03:34 |
kjelly | sdake, SamYaple: The problem is the same. | 03:34 |
sdake | run uname -a gain pzl | 03:35 |
SamYaple | kjelly: whats teh docker version? | 03:35 |
kjelly | docker --version | 03:35 |
kjelly | Docker version 1.8.1, build d12ea79 | 03:35 |
sdake | uname -a returns what? | 03:35 |
kjelly | uname -a | 03:35 |
kjelly | Linux user-VirtualBox 3.13.0-63-generic #103-Ubuntu SMP Fri Aug 14 21:42:59 UTC 2015 x86_64 x86_64 x86_64 GNU/Linux | 03:35 |
sdake | bummer if trusty kernel is busted out of the box | 03:36 |
SamYaple | its not, its what i use | 03:36 |
SamYaple | kjelly: you are hitting this bug fyi https://github.com/docker/docker/issues/5899 | 03:36 |
SamYaple | sdake: it happens to centos too | 03:36 |
sdake | never seen it on centos 7.1 SamYaple | 03:37 |
sdake | kjelly ar eyou running with apparmor on? | 03:37 |
SamYaple | ive never seen it on ubuntu and yet here we are | 03:37 |
sdake | SamYaple do you run with apparmor on? | 03:38 |
SamYaple | no, but thats not what this error is | 03:38 |
SamYaple | apparmor==selinux and cant be used with kolla anyway | 03:38 |
sdake | right, so if kjelly has apparmor on he is boundto run into trubles ;) | 03:39 |
SamYaple | sure but thats not related to this issue | 03:39 |
kjelly | sudo apparmor_status | 03:39 |
kjelly | apparmor module is loaded. | 03:39 |
kjelly | 21 profiles are loaded. | 03:39 |
kjelly | 21 profiles are in enforce mode. | 03:39 |
sdake | well lets turn that off and try again | 03:39 |
kjelly | Do I need to disable it ? | 03:39 |
sdake | since we know it wont work with it | 03:40 |
SamYaple | kjelly: yea go ahead and disable it, it will require a reboot | 03:40 |
SamYaple | docker should be on one of those exlcude filters, but lets disable it an move on | 03:40 |
sdake | at one point turning off apparmor wias in the docs ;) | 03:40 |
SamYaple | i think you removed that doc sdake | 03:40 |
sdake | who knows | 03:41 |
sdake | looks like there is no resolution int he docker bug report | 03:42 |
SamYaple | cpuguy83 makes it pretty clear the issue isn't dockers | 03:43 |
SamYaple | and ill have to agree with him on that | 03:43 |
sdake | ya pam is busted in ubuntu - this is why i had that pam noaudit hack in centos images | 03:43 |
SamYaple | if kjelly is running the 63 kernel, it _might_ be apparmor since i can run the 63 with no issue | 03:43 |
sdake | it was fixedi n 3.15 kernel version | 03:44 |
sdake | hope this umcka stuff works | 03:44 |
sdake | dr recommended it for flu and colds | 03:44 |
sdake | kjelly which command did you run to build your images | 03:48 |
kjelly | tools/build.py -n 172.22.2.81:5000/kollaglue --base ubuntu --type source --push | 03:49 |
sdake | ok thanks | 03:49 |
sdake | did that apparmor thing fix it up? | 03:49 |
kjelly | How to disable it? I run the command `sudo update-rc.d -f apparmor remove`. But it seems active. | 03:51 |
sdake | no idea | 03:51 |
sdake | perhaps SamYaple can provide the proper command | 03:51 |
sdake | i run fedora/centos enitrely | 03:51 |
sdake | i can help on those distros ;-) | 03:51 |
SamYaple | https://help.ubuntu.com/community/AppArmor | 03:52 |
SamYaple | that will disable it | 03:52 |
SamYaple | sdake: im doing keystone v3 stuff now | 03:53 |
SamYaple | what should the default domain be named | 03:53 |
kjelly | I remove apparmor package, reboot, and deploy again, then failed. | 03:58 |
SamYaple | kjelly: yea thats what I expected | 03:58 |
SamYaple | my only suggestion would be to run the newer 3.16 or 3.19 kernel | 03:59 |
SamYaple | thats what I use | 03:59 |
SamYaple | (3.19) | 03:59 |
SamYaple | kjelly: the 3.19 kernel package name is 'linux-image-generic-lts-vivid' if you are interested | 04:00 |
kjelly | I use `docker run -i -t 172.22.2.81:5000/kollaglue/ubuntu-source-keystone /bin/bash`. And run su command. No error happened. | 04:01 |
SamYaple | --net=host | 04:01 |
kjelly | Yes, I add the flag "--net=host" and use su command, it display "su: System error" | 04:02 |
kjelly | Nice. After I upgrade kernel to "linux-image-generic-lts-vivid", it works. Thanks, sdake and SamYaple. | 04:08 |
SamYaple | kjelly: you can also use as low as utopic (3.16) for the kernel and it will work | 04:08 |
sdake | cool so ubuntu 14.04 kernel is busted with kolla - should go in the docs | 04:08 |
SamYaple | 14.04 kernel is not busted | 04:08 |
sdake | i think that is what we concluded via this debug sesson right? | 04:08 |
SamYaple | neither is 3.13 | 04:08 |
SamYaple | 3.13 works for me | 04:08 |
SamYaple | there is other things going on here | 04:09 |
sdake | interesitng wonder what the delta is between yuor system and kjelly's | 04:09 |
SamYaple | besides the docs already say 3.19 for ubuntu kernel because of AUFS bug | 04:09 |
SamYaple | he is running a VM | 04:09 |
sdake | ya that could be it | 04:10 |
SamYaple | also not using btrfs | 04:10 |
SamYaple | both of those things could do it | 04:10 |
SamYaple | im just going to name the default domain 'openstack' since you ahvent commented sdake | 04:11 |
sdake | commented on which? | 04:11 |
SamYaple | 03:54 < SamYaple> sdake: im doing keystone v3 stuff now | 04:11 |
SamYaple | 03:54 < SamYaple> what should the default domain be named | 04:11 |
sdake | that seems perfectly reasonable | 04:11 |
sdake | i just came in from outside -hadnt read scrollback | 04:11 |
SamYaple | no problem | 04:11 |
sdake | apologie | 04:11 |
sdake | s | 04:11 |
SamYaple | just letting you know | 04:12 |
sdake | ya wfm | 04:12 |
sdake | kjelly was it worth the effort? :-) | 04:12 |
*** jtriley has joined #kolla | 04:13 | |
*** jtriley has quit IRC | 04:19 | |
*** weiyu has quit IRC | 04:28 | |
*** unicell1 has joined #kolla | 04:33 | |
*** unicell has quit IRC | 04:34 | |
SamYaple | we got quite a few sessions for the summit | 04:36 |
sdake | you mean the 3 fishbowl plus 5 regular sessions plus half day contributor meetup? | 04:37 |
SamYaple | yea | 04:37 |
sdake | yup - was plased we almost got as many as asked for | 04:37 |
sdake | magnum had twice the ask and received same session count as kolla | 04:43 |
*** tfukushima has joined #kolla | 04:51 | |
*** jtriley has joined #kolla | 05:14 | |
*** jtriley has quit IRC | 05:19 | |
*** weiyu has joined #kolla | 05:34 | |
kjelly | sdake: Sorry. I go out for lunch. I am not sure what you said, " kjelly was it worth the effort? :-)" | 05:36 |
*** inc0 has joined #kolla | 05:39 | |
inc0 | good morning | 05:39 |
sdake | kjelly what i mean was what you ahd to go thorugh to get kolla to deploy openstack worth it to you ? | 05:39 |
sdake | yo inc0 | 05:39 |
*** kjelly has quit IRC | 05:39 | |
inc0 | sdake, got a moment? I have questions about https://review.openstack.org/#/c/215062/ | 05:40 |
sdake | inc0 shoot | 05:40 |
inc0 | you've said that you don't want container to write to disk | 05:40 |
inc0 | my question is - we can't do this without container creating shared log device somewhere | 05:41 |
inc0 | so is it acceptable to add /var/lib/kolla/dev/log? | 05:41 |
sdake | ok with a shared device | 05:41 |
inc0 | cool, that was my issue, SamYaple understood that you were against it;) | 05:41 |
sdake | we have alot of bind mounts that write to the host os | 05:41 |
sdake | but i'd rther prefer we dind't do so | 05:42 |
inc0 | that what I thought, like mysql | 05:42 |
sdake | i don'twant the logs file being written | 05:42 |
SamYaple | sdake: you caused a big fight about this about a week ago | 05:42 |
SamYaple | nothing has changed | 05:42 |
sdake | about for example /var/log/mysql.log | 05:42 |
SamYaple | stop causing problems | 05:42 |
inc0 | SamYaple, respectfully, you tend to generate fights a fair bit youself;) | 05:42 |
inc0 | just as I said - there was misunderstading obviously | 05:43 |
*** kjelly has joined #kolla | 05:43 | |
inc0 | cool, that's solved | 05:43 |
inc0 | I'll make bindmounting /var/log/kolla optional and we're done | 05:44 |
sdake | wfm | 05:44 |
SamYaple | inc0: sure but at least i know what im fighting about | 05:44 |
inc0 | also I'll make new patch with keystone logging to work out per-service | 05:44 |
inc0 | SamYaple, question was whether or not person you fighting with knows that;) | 05:44 |
SamYaple | inc0: im not merging that patch until we have another one showing how it will work perservice | 05:44 |
sdake | SamYaple you gotta calm down dude or your going to have a cornoary | 05:45 |
inc0 | yeah being *too* passionate about project isn't very good as well | 05:45 |
sdake | kjelly what i meant was are you happy with the result of being able to deploy openstack with kolla - was it worth the effort? | 05:46 |
kjelly | sdake: yes. I success to deploy openstack finally. | 05:47 |
kjelly | sdake, SamYaple: thanks | 05:47 |
SamYaple | sdake: you could help alot with my calmness by actually listening. mostly its you saying something thats not true and you ignoring me when i say 'thats not true' | 05:47 |
sdake | kjelly cool thanks ;) | 05:48 |
*** weiyu has quit IRC | 05:48 | |
*** kjelly has quit IRC | 05:49 | |
inc0 | guys do we have optional bindmounting anywhere now? | 05:51 |
SamYaple | no inc0, you can leave out the optional bit now. as i said yesterday we need it in alot of places | 05:51 |
SamYaple | its just a quick change in the defaults.yml | 05:52 |
inc0 | ok, I'll file a but for it then and jump to implementation of per-service ok? | 05:52 |
inc0 | bug* | 05:52 |
SamYaple | i think there may already be a bug (old one) but maybe not | 05:53 |
sdake | all bugs are either in rc1 or closed as invalid | 05:53 |
sdake | so if its not in rc1, its not there | 05:53 |
SamYaple | old as in age of bug | 05:54 |
*** kjelly has joined #kolla | 05:57 | |
*** kjelly has quit IRC | 06:01 | |
*** chandankumar has joined #kolla | 06:02 | |
inc0 | SamYaple, I can't find it | 06:04 |
inc0 | I'll file a new one and close one as duplicate if needed | 06:04 |
SamYaple | inc0: yea i dont see one either. a new one is cool | 06:04 |
SamYaple | yea | 06:04 |
inc0 | https://bugs.launchpad.net/kolla/+bug/1493213 | 06:06 |
openstack | Launchpad bug 1493213 in kolla "bindmounts have to be optional" [Undecided,New] | 06:06 |
SamYaple | lgtm | 06:07 |
sdake | which bindmounts do we that could be optional inc0? | 06:08 |
inc0 | well I'd say /var/lib/mysql | 06:08 |
inc0 | for on | 06:08 |
inc0 | one | 06:08 |
sdake | we don't bindmount that do we? | 06:08 |
sdake | i was pretty sure that was a data container | 06:08 |
inc0 | data container has volume | 06:08 |
sdake | ya those are ok | 06:09 |
inc0 | do we consider that a bindmount or not? | 06:09 |
SamYaple | rabbitmq mariadb nova glance logs all of those are getting optional bindmounts | 06:09 |
sdake | what i am talking abut is host bindmounting | 06:09 |
*** smeisner has quit IRC | 06:09 | |
sdake | data volume bindmounts are fine with me | 06:09 |
SamYaple | bindmount is refering to the host | 06:09 |
inc0 | well, host will end up with /var/lib/mysql won't it? | 06:09 |
sdake | inc0 no it ddoes not | 06:09 |
sdake | the data container ends up stroing the data | 06:10 |
SamYaple | inc0: sdake is confusing terminology | 06:10 |
SamYaple | bindmount == host | 06:10 |
sdake | if the data container ends ups troing the data, its not a security problem | 06:10 |
sdake | its when the bindmount connects to hte host os | 06:10 |
SamYaple | thats why its optional | 06:11 |
*** smeisner has joined #kolla | 06:11 | |
sdake | i'm off to bed enjoy | 06:12 |
sdake | night all | 06:12 |
inc0 | good night | 06:13 |
SamYaple | night | 06:13 |
inc0 | see SamYaple all it took was to ask clear and straight question with no room for doubt;) | 06:16 |
inc0 | I suspected this will be the case | 06:16 |
SamYaple | inc0: this exact question was asked last time this conversation came up | 06:16 |
SamYaple | sdake said now | 06:16 |
SamYaple | no* | 06:17 |
SamYaple | i wasnt assuming anything | 06:17 |
SamYaple | i was relaying _what_ _was_ _said_ | 06:17 |
inc0 | well, maybe it's just you thenl) | 06:17 |
SamYaple | im sure it is. he doesnt like to listen to me | 06:17 |
*** kjelly has joined #kolla | 06:18 | |
inc0 | I'd say "noone does" but that would look rude;) | 06:18 |
SamYaple | everyone likes being wrong it seems | 06:18 |
inc0 | just kidding ofcourse | 06:18 |
SamYaple | all jokes have a grain of truth | 06:18 |
inc0 | ignorance is a bliss | 06:18 |
SamYaple | it is a bit iratating because I keep thinking of what Yaodu would be right now if i spent my time on that. | 06:19 |
SamYaple | I had all of this code in Juno working (including ceph) | 06:19 |
inc0 | well, it's quite often that community doesn't really accelerate things | 06:20 |
SamYaple | i know, and i still know community is better than not | 06:20 |
SamYaple | which is why im here | 06:20 |
inc0 | on the other hand, there is little chance yaodu would land in openstack namespace, got design sessions in tokyo and such | 06:20 |
inc0 | brb | 06:20 |
SamYaple | KOlla wouldnt ahve either if i wasnt here ;) | 06:21 |
SamYaple | yoadu may very well have | 06:21 |
*** weiyu_ has joined #kolla | 06:26 | |
inc0 | back | 06:28 |
*** Haomeng has quit IRC | 06:34 | |
*** Haomeng|2 has joined #kolla | 06:34 | |
*** diga has joined #kolla | 06:36 | |
*** jtriley has joined #kolla | 07:32 | |
*** inc0 has quit IRC | 07:34 | |
*** inc0 has joined #kolla | 07:34 | |
*** weiyu_ has quit IRC | 07:36 | |
*** weiyu has joined #kolla | 07:39 | |
*** jtriley has quit IRC | 07:42 | |
*** achanda has joined #kolla | 07:43 | |
*** shardy has joined #kolla | 07:44 | |
*** kjelly has quit IRC | 07:56 | |
*** kjelly has joined #kolla | 07:59 | |
*** mbound has joined #kolla | 08:06 | |
*** SourabhP has joined #kolla | 08:07 | |
*** mbound has quit IRC | 08:07 | |
*** SourabhP_ has joined #kolla | 08:09 | |
*** SourabhP has quit IRC | 08:11 | |
*** SourabhP_ is now known as SourabhP | 08:11 | |
kjelly | Is the file, "ansible-deployment.rst" good place to put some information about disable apparmor or docker bug about su ? | 08:20 |
*** athomas has joined #kolla | 08:20 | |
*** gfidente has joined #kolla | 08:24 | |
*** achanda has quit IRC | 08:25 | |
*** achanda has joined #kolla | 08:28 | |
*** SourabhP has quit IRC | 08:35 | |
*** jmccarthy has quit IRC | 08:36 | |
*** jmccarthy has joined #kolla | 08:37 | |
*** weiyu has quit IRC | 08:45 | |
*** achanda has quit IRC | 08:52 | |
*** weiyu has joined #kolla | 08:55 | |
kjelly | I found that I can deploy openstack using kolla if apparmor is enabled. | 08:56 |
*** dayou has quit IRC | 08:59 | |
*** pbourke has quit IRC | 09:05 | |
*** pbourke has joined #kolla | 09:06 | |
*** diga has quit IRC | 09:13 | |
*** [Leeloo] has joined #kolla | 09:22 | |
[Leeloo] | Hello | 09:22 |
[Leeloo] | I use this guide to deploy Kolla (Fedora 21 Core) --> https://github.com/stackforge/kolla/blob/master/doc/dev-quickstart.rst . Build the images localy (took several hours). Tag/Push the images in a local registry. Then start the deployment, but got this error --> http://picpaste.com/SNAG-7135-PknEA45Z.jpg . Thanks for the help | 09:23 |
*** weiyu has quit IRC | 09:26 | |
kjelly | I guess you need to change the value, kolla_internal_address, in /etc/kolla/globals.yml. | 09:26 |
*** weiyu has joined #kolla | 09:27 | |
[Leeloo] | kjelly: the address shoud be ? i make the deployement on a physical machine (OVH serveur). i should put the Ip addres of the server ? | 09:30 |
inc0 | [Leeloo], you're doing all-in-one? | 09:31 |
[Leeloo] | "./tools/kolla-ansible deploy" | 09:32 |
[Leeloo] | inc0: I would like a all-in-one | 09:33 |
*** shardy_ has joined #kolla | 09:33 | |
kjelly | In all-in-one, If you ip is 10.0.2.15, try to set the value to 10.0.2.254. | 09:33 |
inc0 | in /etc/kolla/globals.yml set up kolla_internal_address to some address in your network but not yet taken | 09:34 |
inc0 | this will be address keepalived will create as floating ip | 09:34 |
inc0 | also make sure that there is no mysql server running on host itself | 09:34 |
*** shardy has quit IRC | 09:34 | |
[Leeloo] | The server is at my provider. Got these addresse 5.39.xx.xx. for the server itself + a block of address routed to this server, ie: 204.23.xx.2--> 204.23.xx.20. What adress shout I use for "kolla_internal_address" | 09:38 |
*** shardy_ has quit IRC | 09:38 | |
*** shardy has joined #kolla | 09:39 | |
inc0 | [Leeloo], you know what, just add enable_haproxy=False to your /etc/kolla/globals.yml :) | 09:39 |
inc0 | and set up kolla_internal_address to your hosts IP | 09:40 |
inc0 | this will remove HAProxy+Keepalived and you won't need to add new address | 09:40 |
inc0 | haproxy and keepalived really makes sense only in multinode ha setup | 09:41 |
[Leeloo] | inc0: thanks I try :) | 09:42 |
*** sdake has quit IRC | 09:42 | |
inc0 | no problem, let us know if that'll help, because otherwise it might be a bug | 09:42 |
[Leeloo] | I tell you in 2 mins :) | 09:42 |
inc0 | remember to remove running containers | 09:43 |
inc0 | because ansible won't redeploy it | 09:43 |
*** dimsum__ has joined #kolla | 09:46 | |
*** dims_ has joined #kolla | 09:50 | |
[Leeloo] | Same message for "TASK: [mariadb | Creating haproxy mysql user] *********************************" --> "...heck login_user and login_password are correct or ~/.my.cnf has the credentials. E..."" | 09:51 |
[Leeloo] | I put " enable_haproxy: "False" " at the end of "/etc/kolla/globals.yml" | 09:51 |
[Leeloo] | and the ip of the host for kolla_internal_address | 09:51 |
inc0 | try to login from host to mysql | 09:53 |
inc0 | do mysql -u root -h <<your ip>> -p | 09:53 |
inc0 | and default password is "password"? | 09:53 |
inc0 | you can check up passwords in /etc/kolla/password.yml | 09:54 |
*** dimsum__ has quit IRC | 09:55 | |
[Leeloo] | e214f0d414f0 localhost:5000/kollaglue/centos-binary-mariadb:latest "/start.sh" 4 minutes ago Up 3 minutes mariadb | 10:01 |
[Leeloo] | ERROR 1045 (28000): Access denied for user 'root'@'ns2cccc02.ip-5-ccccc.eu' (using password: YES) | 10:02 |
inc0 | hmm...that's strange | 10:03 |
*** dolpher1 has quit IRC | 10:05 | |
*** tfukushima has quit IRC | 10:05 | |
[Leeloo] | i will see this, this afternoon for me... thanks :) | 10:06 |
*** jtriley has joined #kolla | 10:06 | |
inc0 | sorry I couldn't be more help | 10:10 |
*** [Leeloo] has quit IRC | 10:10 | |
*** zhiwei has quit IRC | 10:10 | |
*** [Leeloo] has joined #kolla | 10:10 | |
*** jtriley has quit IRC | 10:13 | |
*** weiyu has quit IRC | 10:14 | |
*** weiyu_ has joined #kolla | 10:18 | |
*** akwasnie has joined #kolla | 10:22 | |
inc0 | SamYaple, around? | 10:35 |
*** weiyu_ has quit IRC | 10:42 | |
*** weiyu_ has joined #kolla | 10:45 | |
openstackgerrit | Michal Jastrzebski (inc0) proposed stackforge/kolla: Ansible bits to rsyslog https://review.openstack.org/215062 | 11:02 |
*** kjelly_ has joined #kolla | 11:10 | |
*** weiyu_ has quit IRC | 11:18 | |
*** chandankumar has quit IRC | 11:46 | |
*** shardy_ has joined #kolla | 11:49 | |
*** shardy has quit IRC | 11:50 | |
*** shardy_ has quit IRC | 11:55 | |
*** shardy has joined #kolla | 11:55 | |
*** inc0 has quit IRC | 12:01 | |
[Leeloo] | inc0 : I remove in file site.yml the proxy line.... but stil get the TASK: [mariadb | Creating haproxy mysql user] ********************************* and the the error with the mysql password :( | 12:02 |
*** inc0 has joined #kolla | 12:04 | |
*** mbound has joined #kolla | 12:05 | |
pbourke | [Leeloo]: I dont think inc0 meant to remove that line, I think he meant you should append "enable_haproxy: false" to /etc/kolla/globals.yml | 12:06 |
*** mbound has quit IRC | 12:07 | |
[Leeloo] | pbourke: I do, with no result, so I try to remove the line to see if any effect | 12:07 |
pbourke | [Leeloo]: are you attempting a single node setup? | 12:07 |
*** mbound has joined #kolla | 12:07 | |
inc0 | well, what enable_haproxy: false does is it makes ansible skip deployment of haproxy+keepalived | 12:08 |
inc0 | therefore not messing with IPs | 12:08 |
inc0 | however, in this case it seems it's wrong password or something.. | 12:09 |
pbourke | yeah I think you should check you can access the mysql server directly | 12:09 |
pbourke | mysql -uroot -ppassword -h<kolla_internal_ip> | 12:09 |
[Leeloo] | If I want to use the docker hub to be use during the deployement do I have to change something in the config file ? | 12:11 |
[Leeloo] | inc0: with an empty password, I can connect to the mariadb | 12:20 |
inc0 | thats strange | 12:20 |
[Leeloo] | inc0: do you want me to show you files in /etc/kolla/... ? | 12:22 |
[Leeloo] | inc0: a re-deplys, and now I am at step 'ASK: [rabbitmq | Starting RabbitMQ bootstrap container] ************'...and it continue | 12:23 |
*** chandankumar has joined #kolla | 12:24 | |
[Leeloo] | inc0: get this error --> http://pastebin.com/VagMBdjp | 12:24 |
inc0 | [Leeloo], did you removed old containers before deploy? | 12:29 |
[Leeloo] | yes | 12:30 |
[Leeloo] | docker stop $(docker ps -a -q | grep -v '0253ca131173'); docker rm $(docker ps -a -q | grep -v '0253ca131173') | 12:30 |
[Leeloo] | "0253ca131173" is the registry | 12:30 |
*** jtriley has joined #kolla | 12:31 | |
inc0 | won't that just remove registry? | 12:33 |
inc0 | do docker ps -a and it's empty? | 12:33 |
[Leeloo] | http://pastebin.com/k5dSCBf1 | 12:38 |
inc0 | docker logs bootstrap_keystone please | 12:39 |
[Leeloo] | No handlers could be found for logger "oslo_config.cfg" | 12:41 |
*** jtriley has quit IRC | 12:43 | |
*** diga has joined #kolla | 12:46 | |
*** klint has quit IRC | 12:56 | |
*** jpeeler has joined #kolla | 12:59 | |
inc0 | that seems like image build error...although afair that was not a critical error | 13:03 |
*** slagle_ is now known as slagle | 13:03 | |
[Leeloo] | If I want to use the docker hub to be use during the deployement do I have to change something in the config file ? | 13:05 |
*** mbound has quit IRC | 13:06 | |
inc0 | well you should remove namespace alltogether from global.yml | 13:06 |
inc0 | (I think, I never used pulled images;)) | 13:07 |
pbourke | dockerhub images wont work | 13:07 |
pbourke | they're out of date | 13:07 |
pbourke | if maria works with empty pass then there's a config problem | 13:07 |
[Leeloo] | pbourke: I do the deplyoment on a fresh Fedora 21 Core... | 13:09 |
[Leeloo] | centos 6.5 / 7 is recommanded ? | 13:10 |
inc0 | that shouldn't matter too much | 13:10 |
inc0 | if you do centos images | 13:10 |
inc0 | try to build your own images tho - tools/build.py | 13:11 |
[Leeloo] | inc0 : this is waht I do. Build the images localy (took several hours). Tag/Push the images in a local registry. Then start the deployment | 13:12 |
inc0 | hmm, I don't know :/ | 13:13 |
*** mbound has joined #kolla | 13:13 | |
[Leeloo] | inc0: :( :) | 13:15 |
*** mbound has quit IRC | 13:15 | |
pbourke | [Leeloo]: suggest focusing on the mariadb issue first | 13:15 |
pbourke | [Leeloo]: exec into the container, compare the confs against the ansible template | 13:15 |
pbourke | [Leeloo]: the fact you can access it with an empty password sounds vaguely familiar but I cant remember the exact cause | 13:16 |
pbourke | [Leeloo]: but its a definite issue as the other services will try to access it with -ppassword | 13:16 |
*** diga has quit IRC | 13:16 | |
[Leeloo] | pbourke: i focus on this :) | 13:17 |
pbourke | [Leeloo]: stop and rm all containers, then re-run ansible with --tags=mariadb | 13:17 |
pbourke | and -vv | 13:17 |
pbourke | to get extra output on what ansible is doing | 13:17 |
*** stvnoyes has joined #kolla | 13:19 | |
*** [Leeloo] is now known as [Leeloo]-afk | 13:20 | |
*** kjelly_ has quit IRC | 13:20 | |
*** mbound has joined #kolla | 13:23 | |
*** banix has joined #kolla | 13:25 | |
*** Esai has joined #kolla | 13:32 | |
*** inc0 has quit IRC | 13:35 | |
*** Esai has quit IRC | 13:35 | |
*** Esai has joined #kolla | 13:35 | |
*** [Leeloo]-afk is now known as [Leeloo | 13:39 | |
*** [Leeloo is now known as [Leeloo] | 13:39 | |
[Leeloo] | pbourke: what will be the eaxct command for "re-run ansible with --tags=mariadb -vv" ? | 13:39 |
*** jtriley has joined #kolla | 13:40 | |
pbourke | [Leeloo]: ansible-playbook -i inventory/all-in-one -e @/etc/kolla/globals.yml -e @/etc/kolla/passwords.yml site.yml --tags mariadb -vv | 13:40 |
[Leeloo] | http://pastebin.com/VseiRVwN | 13:46 |
*** diga has joined #kolla | 13:47 | |
*** dwalsh has joined #kolla | 13:48 | |
*** Esai has quit IRC | 13:49 | |
pbourke | [Leeloo]: did you 'docker rm' everything before running the above | 13:49 |
pbourke | ? | 13:49 |
[Leeloo] | yes :) | 13:51 |
[Leeloo] | I connect to mysql witout password. It's "database_password: "password"" in /etc/kolla/passwords.yml | 13:52 |
pbourke | what exact command did you use to rm the containers? | 13:57 |
[Leeloo] | # docker stop $(docker ps -a -q | grep -v '0253ca131173'); docker rm $(docker ps -a -q | grep -v '0253ca131173'); | 13:57 |
[Leeloo] | 0253ca131173 registry:2 "/bin/registry /etc/d" 6 hours ago Up 6 hours 0.0.0.0:5000->5000/tcp registry | 13:58 |
*** dwalsh has quit IRC | 13:58 | |
*** shardy_ has joined #kolla | 14:01 | |
*** shardy has quit IRC | 14:03 | |
*** shardy_ has quit IRC | 14:07 | |
*** shardy has joined #kolla | 14:07 | |
pbourke | [Leeloo]: ok, try this instead: 'docker stop $(docker ps -q); docker rm -f $(docker ps -aq)' | 14:08 |
pbourke | [Leeloo]: then rerun ansible | 14:08 |
[Leeloo] | same :( | 14:16 |
[Leeloo] | If someone want acces to the server, I give the access ? it will be re-install after | 14:17 |
*** jasonsb has quit IRC | 14:24 | |
*** dims_ has quit IRC | 14:26 | |
*** dimsum__ has joined #kolla | 14:27 | |
*** vbel1 has quit IRC | 14:27 | |
*** inc0 has joined #kolla | 14:33 | |
*** unicell has joined #kolla | 14:33 | |
inc0 | afternoon | 14:33 |
*** unicell1 has quit IRC | 14:33 | |
[Leeloo] | I try to rebuild keystone image (tools/build.py keystone), but get this message -> http://pastebin.com/XeExpECQ | 14:44 |
*** dwalsh has joined #kolla | 14:46 | |
*** banix_ has joined #kolla | 14:58 | |
*** banix has quit IRC | 14:58 | |
*** banix_ is now known as banix | 14:58 | |
*** banix has quit IRC | 15:02 | |
*** banix has joined #kolla | 15:04 | |
*** chandankumar has quit IRC | 15:06 | |
*** blahRus has joined #kolla | 15:07 | |
*** dwalsh has quit IRC | 15:21 | |
*** dims_ has joined #kolla | 15:25 | |
*** bradjones has joined #kolla | 15:28 | |
*** bradjones has quit IRC | 15:28 | |
*** bradjones has joined #kolla | 15:28 | |
*** dimsum__ has quit IRC | 15:29 | |
*** dwalsh has joined #kolla | 15:39 | |
*** devlaps has joined #kolla | 15:43 | |
*** chandankumar has joined #kolla | 15:54 | |
*** jasonsb has joined #kolla | 15:56 | |
inc0 | Slower, around? | 16:01 |
*** sdake_ has joined #kolla | 16:06 | |
*** stvnoyes has quit IRC | 16:08 | |
*** unicell has quit IRC | 16:12 | |
inc0 | sdake_, https://review.openstack.org/#/c/215062/ mind taking a look if that's ok with you? | 16:16 |
sdake_ | inc0 i just woe up dude | 16:18 |
devlaps | morning kola folks. i'm new to this project and have a question. can anyone tell me that the stubbed out .conf files in /etc/kolla/config are used for? | 16:18 |
inc0 | well, you logged in to irc means you're exposed to reviews;) | 16:18 |
inc0 | no worries | 16:19 |
inc0 | devlaps, when you do ansible deployment, ansible copies configs from there to container | 16:19 |
inc0 | and container uses these configs | 16:19 |
devlaps | inc0 thanks.. can you point me to where those files get populated? | 16:20 |
inc0 | or actually no, it just bindmounts this directory | 16:20 |
inc0 | yeah, its in ansible/roles/tasks/<< service >>/config.yml | 16:21 |
inc0 | ansible/roles/<<service>>/tasks ** | 16:21 |
inc0 | you have templates of configs in << service >>/templates | 16:22 |
devlaps | thanks.. i saw those, but was confused how those templates interacted with the /etc ones.. | 16:22 |
devlaps | i can see the top level anaislbe config.yml | 16:23 |
inc0 | yeah, this is supposedly only file you need to edit to get barebones openstack running | 16:24 |
inc0 | that and inventory file if you're not doing aio | 16:24 |
devlaps | ok.. i can also see the mege_configs plugin to generate final service config file.. is there an intermediate step that populates the top level etc/kolla/config/*.conf? | 16:25 |
inc0 | soo...I don't think it's used any more, but I'm not 100% sure | 16:28 |
devlaps | ah.. ok.. thanks for the info! | 16:30 |
inc0 | yeah, it seems it's not used, I'll ask around and remove it if so | 16:30 |
inc0 | basically it works like that | 16:31 |
*** dims_ has quit IRC | 16:31 | |
inc0 | you https://github.com/stackforge/kolla/blob/master/ansible/group_vars/all.yml#L14 here you define where configs lands | 16:32 |
*** dimsum__ has joined #kolla | 16:32 | |
*** ccrouch has joined #kolla | 16:32 | |
*** achanda has joined #kolla | 16:32 | |
inc0 | this is example of source config template https://github.com/stackforge/kolla/blob/master/ansible/roles/keystone/templates/wsgi-keystone.conf.j2 | 16:32 |
inc0 | https://github.com/stackforge/kolla/blob/master/ansible/roles/keystone/tasks/config.yml#L19 and this is what copies config and fills up template | 16:33 |
*** stvnoyes has joined #kolla | 16:35 | |
*** mbound has quit IRC | 16:39 | |
*** jmccarthy has quit IRC | 16:41 | |
*** inc0 has quit IRC | 16:46 | |
*** dwalsh has quit IRC | 16:52 | |
*** achanda has quit IRC | 16:56 | |
*** achanda_ has joined #kolla | 16:56 | |
*** dwalsh has joined #kolla | 16:56 | |
*** akwasnie has quit IRC | 16:59 | |
*** unicell has joined #kolla | 17:02 | |
*** tfukushima has joined #kolla | 17:03 | |
devlaps | thanks inc0! | 17:06 |
*** unicell has quit IRC | 17:06 | |
*** unicell has joined #kolla | 17:06 | |
*** achanda_ has quit IRC | 17:08 | |
*** athomas has quit IRC | 17:09 | |
*** jtriley_ has joined #kolla | 17:14 | |
*** jtriley has quit IRC | 17:17 | |
*** sdake_ has quit IRC | 17:19 | |
*** achanda has joined #kolla | 17:21 | |
*** diga has quit IRC | 17:21 | |
*** vilobhmm_11 has joined #kolla | 17:33 | |
*** sdake has joined #kolla | 17:44 | |
*** loth has joined #kolla | 17:45 | |
*** banix has quit IRC | 18:00 | |
*** banix has joined #kolla | 18:02 | |
*** SourabhP has joined #kolla | 18:03 | |
*** inc0 has joined #kolla | 18:07 | |
*** inc0 has quit IRC | 18:12 | |
*** banix has quit IRC | 18:24 | |
*** banix has joined #kolla | 18:24 | |
*** sdake_ has joined #kolla | 18:25 | |
*** sdake has quit IRC | 18:29 | |
*** openstackgerrit has quit IRC | 18:31 | |
*** openstackgerrit has joined #kolla | 18:31 | |
sdake_ | devlaps if your still around i can answer your questions rethe etc dirfiles | 18:33 |
*** chandankumar has quit IRC | 18:35 | |
*** gfidente has quit IRC | 18:38 | |
devlaps | hi sdake_ thanks. i did some more reading, and as far as i can tell, it looks like those configuration files are copied over and merged with the ansible template files to create the final service config file. so, if you want to add in any additional default configuration settings you can do that in etc/kolla/config. is that correct? | 18:38 |
sdake_ | that is correct | 18:38 |
sdake_ | kolla is opinionated unless the operator has opinions | 18:38 |
sdake_ | then it honors those | 18:38 |
sdake_ | one extra detail, it ovrrides any settings as well not just adds | 18:39 |
*** sdake_ is now known as sdake | 18:40 | |
*** achanda has quit IRC | 18:40 | |
*** achanda has joined #kolla | 18:41 | |
*** sdake_ has joined #kolla | 18:46 | |
*** banix has quit IRC | 18:47 | |
*** sdake has quit IRC | 18:49 | |
*** jmccarthy has joined #kolla | 18:51 | |
*** banix has joined #kolla | 18:55 | |
devlaps | thanks sdake_. yes, i saw that happening in the ansible plugin merge_configs.py. | 18:56 |
devlaps | the ordering of /etc/kolla/config/ files in service config.yml is definitely important.. | 18:57 |
sdake_ | cool if you have further questoins feel free to ask | 18:57 |
devlaps | will do. thanks! | 18:57 |
*** achanda has quit IRC | 18:57 | |
*** jmccarthy has quit IRC | 19:09 | |
*** tfukushima has quit IRC | 19:26 | |
*** achanda has joined #kolla | 19:38 | |
*** inc0 has joined #kolla | 19:43 | |
*** achanda has quit IRC | 19:45 | |
*** sdake_ is now known as sdake | 19:49 | |
*** banix has quit IRC | 19:51 | |
*** inc0 has quit IRC | 19:55 | |
*** ccrouch has quit IRC | 19:57 | |
*** ccrouch has joined #kolla | 19:59 | |
*** ccrouch has quit IRC | 19:59 | |
*** ccrouch has joined #kolla | 19:59 | |
*** jtriley_ has quit IRC | 20:01 | |
*** dimsum__ is now known as dims | 20:03 | |
*** dims is now known as Guest80071 | 20:04 | |
*** ccrouch has quit IRC | 20:07 | |
*** ccrouch1 has joined #kolla | 20:07 | |
*** Guest80071 is now known as dims__ | 20:08 | |
*** unicell has quit IRC | 20:11 | |
*** unicell has joined #kolla | 20:12 | |
*** jasonsb has quit IRC | 20:27 | |
*** banix has joined #kolla | 20:29 | |
*** jasonsb has joined #kolla | 20:30 | |
*** shardy_ has joined #kolla | 20:41 | |
*** jmccarthy has joined #kolla | 20:41 | |
*** achanda has joined #kolla | 20:41 | |
*** shardy has quit IRC | 20:42 | |
*** shardy_ has quit IRC | 20:46 | |
*** shardy has joined #kolla | 20:47 | |
*** achanda has quit IRC | 20:51 | |
*** achanda has joined #kolla | 20:54 | |
*** banix has quit IRC | 21:02 | |
*** banix has joined #kolla | 21:07 | |
*** jmccarthy has quit IRC | 21:12 | |
*** jmccarthy1 has joined #kolla | 21:12 | |
*** sdake_ has joined #kolla | 21:16 | |
*** sdake has quit IRC | 21:20 | |
*** achanda has quit IRC | 21:35 | |
*** achanda has joined #kolla | 21:35 | |
*** devlaps has quit IRC | 21:38 | |
*** banix has quit IRC | 21:40 | |
*** jmccarthy1 has quit IRC | 21:40 | |
*** achanda_ has joined #kolla | 21:44 | |
*** achanda has quit IRC | 21:44 | |
*** tummy has joined #kolla | 21:47 | |
sdake_ | jpeeler is rhallisey on pto? | 21:55 |
*** dwalsh has quit IRC | 21:55 | |
*** sdake_ is now known as sdake | 21:55 | |
jpeeler | don't know, haven't seen him online today | 21:56 |
*** ccrouch1 has quit IRC | 21:57 | |
*** ccrouch has joined #kolla | 21:57 | |
*** achanda_ has quit IRC | 21:58 | |
*** achanda has joined #kolla | 22:02 | |
*** sdake_ has joined #kolla | 22:02 | |
*** sdake has quit IRC | 22:05 | |
*** achanda has quit IRC | 22:09 | |
*** achanda has joined #kolla | 22:09 | |
*** achanda_ has joined #kolla | 22:13 | |
*** achanda has quit IRC | 22:13 | |
*** jasonsb has quit IRC | 22:20 | |
*** jasonsb has joined #kolla | 22:24 | |
*** achanda has joined #kolla | 22:55 | |
*** achanda_ has quit IRC | 22:55 | |
*** tummy has quit IRC | 22:58 | |
*** achanda_ has joined #kolla | 23:01 | |
*** achanda has quit IRC | 23:01 | |
*** weiyu has joined #kolla | 23:10 | |
*** weiyu has quit IRC | 23:18 | |
*** achanda has joined #kolla | 23:23 | |
*** achanda_ has quit IRC | 23:23 | |
*** banix has joined #kolla | 23:25 | |
*** jasonsb has quit IRC | 23:41 | |
*** blahRus has quit IRC | 23:56 | |
*** stvnoyes has quit IRC | 23:58 | |
*** weiyu has joined #kolla | 23:59 |
Generated by irclog2html.py 2.14.0 by Marius Gedminas - find it at mg.pov.lt!