| *** jamielennox|away is now known as jamielennox | 00:02 | |
| *** pmannidi has quit IRC | 00:03 | |
| *** markvoelker has quit IRC | 00:11 | |
| *** thorst_afk has joined #openstack-ansible | 00:19 | |
| *** thorst_afk has quit IRC | 00:24 | |
| *** pmannidi has joined #openstack-ansible | 00:25 | |
| *** pmannidi is now known as pmannidi|Trg | 00:25 | |
| *** pmannidi|Trg is now known as pmannidi | 00:25 | |
| *** yifei has joined #openstack-ansible | 00:28 | |
| *** cuongnv has joined #openstack-ansible | 00:56 | |
| *** adreznec has quit IRC | 00:59 | |
| *** hw_wutianwei has joined #openstack-ansible | 01:00 | |
| *** adreznec has joined #openstack-ansible | 01:01 | |
| *** thorst_afk has joined #openstack-ansible | 01:02 | |
| *** thorst_afk has quit IRC | 01:04 | |
| *** SerenaFeng has joined #openstack-ansible | 01:24 | |
| *** cNilesh has joined #openstack-ansible | 01:30 | |
| *** zul has quit IRC | 01:44 | |
| *** thorst_afk has joined #openstack-ansible | 01:58 | |
| *** thorst_afk has quit IRC | 01:58 | |
| *** jamielennox is now known as jamielennox|away | 02:00 | |
| *** winggundamth has joined #openstack-ansible | 02:18 | |
| *** thorst_afk has joined #openstack-ansible | 02:57 | |
| *** thorst_afk has quit IRC | 02:57 | |
| *** jhesketh_ is now known as jhesketh | 03:08 | |
| *** thorst_afk has joined #openstack-ansible | 03:28 | |
| *** fergus has quit IRC | 03:42 | |
| *** fergus has joined #openstack-ansible | 03:42 | |
| *** woodard has quit IRC | 03:43 | |
| *** woodard has joined #openstack-ansible | 03:44 | |
| *** thorst_afk has quit IRC | 03:46 | |
| *** woodard has quit IRC | 03:48 | |
| *** jamielennox|away is now known as jamielennox | 03:53 | |
| *** SerenaFeng has quit IRC | 04:15 | |
| *** hybridpollo has quit IRC | 04:37 | |
| *** fergus has quit IRC | 04:46 | |
| *** nyloc has joined #openstack-ansible | 04:46 | |
| *** fergus has joined #openstack-ansible | 04:46 | |
| *** udesale has joined #openstack-ansible | 04:47 | |
| *** shausy has joined #openstack-ansible | 05:12 | |
| *** shausy has quit IRC | 05:12 | |
| *** weezS has quit IRC | 05:17 | |
| *** pranavb has joined #openstack-ansible | 05:19 | |
| *** cuongnv has quit IRC | 05:25 | |
| *** winggundamth has quit IRC | 05:28 | |
| *** SerenaFeng has joined #openstack-ansible | 05:32 | |
| *** cuongnv has joined #openstack-ansible | 05:38 | |
| *** thorst_afk has joined #openstack-ansible | 05:43 | |
| *** jvidal has joined #openstack-ansible | 05:45 | |
| *** thorst_afk has quit IRC | 05:48 | |
| *** weezS has joined #openstack-ansible | 05:50 | |
| evrardjp | good morning everyone | 05:55 |
|---|---|---|
| *** tobberydberg has joined #openstack-ansible | 05:56 | |
| openstackgerrit | Jean-Philippe Evrard proposed openstack/openstack-ansible master: Move to a all/ group var folder https://review.openstack.org/466278 | 06:07 |
| evrardjp | if someone could vote on https://review.openstack.org/#/c/463659/ that would be great | 06:09 |
| *** Oku_OS-away is now known as Oku_OS | 06:11 | |
| *** chas has joined #openstack-ansible | 06:19 | |
| *** markus_z has joined #openstack-ansible | 06:21 | |
| *** armaan has joined #openstack-ansible | 06:32 | |
| prometheanfire | evrardjp: hi | 06:34 |
| *** os-slack1 has quit IRC | 06:34 | |
| prometheanfire | evrardjp: commit message title could be better, I assume this is for leapfrog? | 06:35 |
| prometheanfire | and I didn't know we were targeting juno as a start | 06:35 |
| *** chas has quit IRC | 06:37 | |
| *** chas has joined #openstack-ansible | 06:37 | |
| openstackgerrit | Kevin Lefevre proposed openstack/openstack-ansible-os_designate stable/ocata: Standardize the testing for Designate https://review.openstack.org/465663 | 06:39 |
| *** chas has quit IRC | 06:41 | |
| *** chas has joined #openstack-ansible | 06:41 | |
| *** thorst_afk has joined #openstack-ansible | 06:44 | |
| *** thorst_afk has quit IRC | 06:49 | |
| *** vnogin has joined #openstack-ansible | 06:52 | |
| *** chas has quit IRC | 06:55 | |
| *** shardy has joined #openstack-ansible | 06:55 | |
| *** chas has joined #openstack-ansible | 06:55 | |
| *** vnogin has quit IRC | 06:56 | |
| *** chas has quit IRC | 06:57 | |
| *** chas has joined #openstack-ansible | 06:57 | |
| *** SerenaFeng has quit IRC | 06:57 | |
| *** tobberyd_ has joined #openstack-ansible | 07:03 | |
| evrardjp | prometheanfire: the code is OSA is already there to leapfrom from J. We are now adding the feature to leapfrog from K | 07:03 |
| evrardjp | to N | 07:03 |
| *** SerenaFeng has joined #openstack-ansible | 07:06 | |
| *** chas has quit IRC | 07:06 | |
| *** chas has joined #openstack-ansible | 07:06 | |
| *** tobberydberg has quit IRC | 07:06 | |
| *** chas has quit IRC | 07:11 | |
| prometheanfire | ah | 07:11 |
| *** pcaruana has joined #openstack-ansible | 07:11 | |
| *** pcaruana has quit IRC | 07:12 | |
| *** pcaruana has joined #openstack-ansible | 07:13 | |
| *** drifterza has joined #openstack-ansible | 07:17 | |
| *** vnogin has joined #openstack-ansible | 07:30 | |
| *** david-lyle has quit IRC | 07:36 | |
| *** mpranjic has quit IRC | 07:39 | |
| *** mpranjic has joined #openstack-ansible | 07:42 | |
| *** david-lyle has joined #openstack-ansible | 07:42 | |
| *** thorst_afk has joined #openstack-ansible | 07:45 | |
| *** david-lyle has quit IRC | 07:46 | |
| *** chas has joined #openstack-ansible | 07:47 | |
| *** thorst_afk has quit IRC | 07:50 | |
| *** gouthamr has joined #openstack-ansible | 07:51 | |
| *** manheim_ has joined #openstack-ansible | 07:58 | |
| *** david-lyle has joined #openstack-ansible | 08:03 | |
| *** fergus has quit IRC | 08:04 | |
| *** fergus has joined #openstack-ansible | 08:04 | |
| hwoarang | good morning | 08:14 |
| *** tobberyd_ has quit IRC | 08:14 | |
| hwoarang | does anyone make any sense of this http://paste.openstack.org/show/610124/ ? node is set to rabbitmq@container1 but rabbitmqctl is executed with -n ' ' (ie empty node) | 08:15 |
| * hwoarang grabs some coffee | 08:15 | |
| *** pbandark has joined #openstack-ansible | 08:17 | |
| *** gouthamr has quit IRC | 08:19 | |
| evrardjp | good morning hwoarang | 08:25 |
| evrardjp | it looks like your rabbitmq isn't properly started/bootstrapped. | 08:26 |
| evrardjp | which version are you running? Do you have HOSTNAME= set in your env vars somewhere? | 08:27 |
| hwoarang | evrardjp: let me check. but isn't it suspicous that the module does not take the 'node' param into consideration? | 08:30 |
| evrardjp | that's highly suspicious | 08:30 |
| hwoarang | i checked the source, it seems that -n parameter is only used if you passed a node parameter but it doesn't seem to work like this... | 08:30 |
| evrardjp | that's why I am talking about this hostname thing | 08:30 |
| evrardjp | could you also manually check the playbook to see what's going on for your variables? | 08:31 |
| evrardjp | and I still don't know which version you are running | 08:31 |
| evrardjp | if stable/ocata, you're running ansible < 2.3 | 08:32 |
| hwoarang | HOSTNAME is set in /etc/profile and ansible is 2.3.0.0 | 08:32 |
| evrardjp | oh. | 08:32 |
| evrardjp | /etc/profile on your physical host? | 08:33 |
| hwoarang | yes | 08:33 |
| evrardjp | could you unset that? | 08:33 |
| evrardjp | I thought I fixed this. | 08:33 |
| evrardjp | please unset the HOSTNAME from the host, and re-try to start your playbook | 08:33 |
| hwoarang | where is this fix supposed to be? in the rabbitmq_server role or somewhere else? | 08:34 |
| evrardjp | ansible | 08:34 |
| hwoarang | ah | 08:34 |
| evrardjp | ahah, no I get it. | 08:34 |
| evrardjp | well could you try without it, to see if it works better? | 08:34 |
| evrardjp | But I think this one is a different one. | 08:34 |
| pjm6 | good morning all | 08:35 |
| evrardjp | hwoarang: please test that the module is receing the proper arguments too. | 08:42 |
| hwoarang | yeah it does :( | 08:42 |
| hwoarang | but let me check... | 08:42 |
| evrardjp | so you did add debugging on the play level | 08:42 |
| drifterza | hello | 08:42 |
| evrardjp | drifterza: hello! | 08:42 |
| hwoarang | i did add some printfs at the rabbitmq_user.py file | 08:42 |
| evrardjp | it's been a while! | 08:42 |
| evrardjp | pjm6: good morning | 08:43 |
| drifterza | indeed, hows things ? | 08:43 |
| evrardjp | good and you? | 08:43 |
| evrardjp | hwoarang: good. | 08:43 |
| drifterza | Good thanks. | 08:43 |
| evrardjp | hwoarang: do you see something weird about hostname? | 08:43 |
| *** karimb has joined #openstack-ansible | 08:44 | |
| evrardjp | oh yeah I forgot to tell you why I'm obsessed with the hostname hwoarang | 08:44 |
| hwoarang | checking because i need to see how to run only this specific task | 08:44 |
| evrardjp | it's because if it's in the environment when running a rabbitmq command, it overrides whatever you give in the cli | 08:44 |
| hwoarang | and i am not sure why the container has the same env with the host | 08:44 |
| hwoarang | because HOSTNAME in the container is the same with host | 08:45 |
| hwoarang | something is fishy | 08:45 |
| evrardjp | that part is correct | 08:45 |
| evrardjp | when creating a container we copy the nev | 08:45 |
| evrardjp | env | 08:45 |
| evrardjp | like profile | 08:45 |
| hwoarang | hmm | 08:45 |
| evrardjp | for example if there are proxy set up etc. | 08:45 |
| evrardjp | I still think we should change that, but hey, I'm not the only one :p | 08:46 |
| *** thorst_afk has joined #openstack-ansible | 08:46 | |
| evrardjp | assuming we can't change that, if the container + host doesn't have the HOSTNAME it should work. | 08:46 |
| hwoarang | let me gather some more output | 08:47 |
| evrardjp | In your case my fix doesn't apply, which makes 100% sense. | 08:47 |
| hwoarang | because it does not make much sense | 08:47 |
| evrardjp | what doesn't make sense? | 08:47 |
| evrardjp | well IMO rabbitmqctl sucks. | 08:48 |
| evrardjp | it shouldn't behave like that | 08:48 |
| hwoarang | the module says that it executes [/usr/sbin/rabbitmqctl', '-q', '-n', 'rabbit@container1'] | 08:48 |
| hwoarang | yet the ansible output says it's -n ' ' instead | 08:48 |
| evrardjp | yeah | 08:48 |
| hwoarang | and segfaults.... | 08:49 |
| evrardjp | that's what I'm telling you | 08:49 |
| evrardjp | input is not the same as output | 08:49 |
| evrardjp | [/usr/sbin/rabbitmqctl', '-q', '-n', 'rabbit@container1'] is your input + ENV | 08:49 |
| hwoarang | so ansible is not reporting the input to the _exec method? | 08:49 |
| hwoarang | oh that SUCKS! | 08:49 |
| evrardjp | wait | 08:49 |
| evrardjp | the ansible input is right | 08:50 |
| *** tobberydberg has joined #openstack-ansible | 08:50 | |
| evrardjp | it's the output that tricks you | 08:50 |
| evrardjp | if you do | 08:50 |
| evrardjp | (at least IIRC) | 08:50 |
| hwoarang | i thought that the output was the same with input :( | 08:50 |
| evrardjp | HOSTNAME=babar /usr/sbin/rabbitmqctl', '-q', '-n', 'rabbit@container1' | 08:50 |
| evrardjp | the output will give your | 08:50 |
| evrardjp | you | 08:50 |
| evrardjp | (without the ,_ | 08:51 |
| evrardjp | ) | 08:51 |
| andymccr | wasnt babr the elephant character? :P | 08:51 |
| evrardjp | /usr/sbin/rabbitmqctl -q -n '' | 08:51 |
| hwoarang | that sucks... | 08:51 |
| evrardjp | andymccr: yes indeed. | 08:51 |
| evrardjp | Because foo bar isn't fun enough. | 08:51 |
| *** SerenaFeng has quit IRC | 08:52 | |
| andymccr | haha ba bar instead of foo bar :P | 08:52 |
| evrardjp | well when it's about finding interesting examples, I don't rely on foo/bar. I prefer babar, tintin, haddock. | 08:52 |
| andymccr | haddock is the best. | 08:52 |
| *** SerenaFeng has joined #openstack-ansible | 08:52 | |
| andymccr | what about asterix/dogmatix etc? :D | 08:52 |
| evrardjp | sometimes MILOU comes | 08:52 |
| evrardjp | but you don't know milou ofc | 08:53 |
| andymccr | milou is snowy - according to google :D | 08:53 |
| evrardjp | you name it snowy or something like that | 08:53 |
| evrardjp | yeah | 08:53 |
| evrardjp | I kinda don't use spirou that much | 08:53 |
| andymccr | dont recognize that one | 08:53 |
| evrardjp | I guess too long | 08:53 |
| evrardjp | interesting one by Franquin. | 08:54 |
| *** tobberydberg has quit IRC | 08:54 | |
| *** chhavi has joined #openstack-ansible | 08:55 | |
| evrardjp | The question is why these are all comic book chars. I don't know yet. | 08:56 |
| evrardjp | hwoarang: to come back to your issue I'm not 100% certain this is your issue | 08:57 |
| hwoarang | well good thing is that centos7 works so i do have an issue on my end :) | 08:57 |
| evrardjp | please start by unsetting any reference of HOSTNAME just in case. | 08:58 |
| hwoarang | yeah | 08:58 |
| evrardjp | everywhere :p, containers, physical hosts. | 08:58 |
| *** tobberydberg has joined #openstack-ansible | 09:02 | |
| *** thorst_afk has quit IRC | 09:05 | |
| *** electrofelix has joined #openstack-ansible | 09:17 | |
| *** pranavb has quit IRC | 09:27 | |
| *** stuartgr has joined #openstack-ansible | 09:30 | |
| *** pranavb has joined #openstack-ansible | 09:40 | |
| *** yifei has quit IRC | 09:40 | |
| *** SerenaFeng has quit IRC | 09:45 | |
| hwoarang | evrardjp: unsetting HOSTNAME before the rabbitmq_user call seems to do something different | 09:47 |
| hwoarang | so that's a good start | 09:47 |
| hwoarang | fails a bit after that because the physical hostname is used inside the container but one step at a time | 09:48 |
| *** cuongnv has quit IRC | 09:49 | |
| *** cNilesh has quit IRC | 09:53 | |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible-os_designate stable/ocata: Reduce init restart/kill times https://review.openstack.org/460717 | 09:53 |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible-os_designate stable/ocata: Use internalURL by default for Neutron https://review.openstack.org/465064 | 09:54 |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible-os_designate stable/ocata: Optimise apt cache update task https://review.openstack.org/462850 | 09:54 |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible-os_designate stable/ocata: Ensure the components are isolated from the system https://review.openstack.org/460155 | 09:54 |
| *** ivve has quit IRC | 09:54 | |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible-os_designate stable/ocata: Fix for lookup and get_url occuring in different places https://review.openstack.org/463785 | 09:54 |
| *** ivve has joined #openstack-ansible | 09:55 | |
| evrardjp | hwoarang: yes that's what I thought. | 09:59 |
| *** kjz1 has joined #openstack-ansible | 09:59 | |
| *** kjz has quit IRC | 10:00 | |
| *** thorst_afk has joined #openstack-ansible | 10:02 | |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible stable/newton: Add gnocchiclient constraints to global pins https://review.openstack.org/466660 | 10:06 |
| odyssey4me | joelgriffiths ^ I don't know why I didn't think of doing that before, but that should do away with the user vars workaround | 10:06 |
| *** thorst_afk has quit IRC | 10:07 | |
| *** chas has quit IRC | 10:08 | |
| *** chas has joined #openstack-ansible | 10:09 | |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible-os_ceilometer stable/newton: Revert "Pin gnocchiclient to comply with optional deps" https://review.openstack.org/466663 | 10:10 |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible-os_ceilometer stable/newton: Revert "Pin gnocchiclient to comply with optional deps" https://review.openstack.org/466663 | 10:10 |
| hwoarang | evrardjp: so is a sensible approach to unset this variable when we have a task with the rabbitmq_user module? | 10:10 |
| hwoarang | not sure which layer needs to be "fixed" | 10:11 |
| *** chas has quit IRC | 10:13 | |
| openstackgerrit | Merged openstack/openstack-ansible-os_designate stable/ocata: Standardize the testing for Designate https://review.openstack.org/465663 | 10:20 |
| evrardjp | hwoarang: let me think a little, but yes it seems sensible | 10:25 |
| evrardjp | I'll have a look at the module itself | 10:25 |
| evrardjp | alternatively we could patch the connection plugin to do it | 10:27 |
| evrardjp | or | 10:28 |
| evrardjp | we just stop wiring this env file | 10:28 |
| evrardjp | and only write what's relevant to us | 10:28 |
| evrardjp | and make sure it's possible for a deployer to override it | 10:28 |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible stable/newton: Add missing tempest_venv_bin group_var https://review.openstack.org/466670 | 10:32 |
| hwoarang | i am curious why centos does not have this issue | 10:33 |
| hwoarang | i am going to investigate | 10:33 |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible stable/newton: Add missing tempest_venv_bin group_var https://review.openstack.org/466670 | 10:36 |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible stable/newton: Add missing tempest_venv_bin group_var https://review.openstack.org/466670 | 10:36 |
| evrardjp | hwoarang: https://github.com/ansible/ansible/blob/devel/lib/ansible/module_utils/basic.py#L2361-L2364 | 10:38 |
| hwoarang | hmm | 10:38 |
| evrardjp | if we want we could change the rabbitmq | 10:38 |
| evrardjp | to use the environ_update | 10:39 |
| evrardjp | the rabbitmq modules* | 10:39 |
| hwoarang | interesting... | 10:39 |
| evrardjp | well I guess for the centos/suse difference, maybe some has HOSTNAME, some don't, at provisioning times | 10:39 |
| evrardjp | hwoarang: not worth time right now I'd say. Too much pain to get these merged | 10:40 |
| evrardjp | I'll at least file a bug, see what they say. | 10:41 |
| *** chas has joined #openstack-ansible | 10:43 | |
| hwoarang | i need to check if i am doing anything wrong in the lxc roles | 10:44 |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible stable/newton: Add missing tempest_venv_bin group_var https://review.openstack.org/466670 | 10:45 |
| drifterza | how do we currently do autoscalling ? | 10:45 |
| *** tobberydberg has quit IRC | 10:45 | |
| *** chas has quit IRC | 10:47 | |
| *** chas has joined #openstack-ansible | 10:47 | |
| evrardjp | andymccr: is there someone already fixing the nova issues? | 10:53 |
| andymccr | evrardjp: which issue specifically? | 10:53 |
| andymccr | im working on that upgrade bug now | 10:53 |
| odyssey4me | drifterza do you mean autoscaling as in the service provided by a combination of openstack services? if so, then it's up to the deployer to ensure that the config is set and the architecture is right for their needs | 10:53 |
| *** toddnni has quit IRC | 10:54 | |
| drifterza | odyssey4me, I'm thinking more on the lines of, backend services needing to scale to more containers for example. | 10:54 |
| drifterza | without manual intervention | 10:55 |
| odyssey4me | we do not currently implement any sort of automatic scaling | 10:55 |
| drifterza | that would require metric monitoring of some sort? | 10:55 |
| odyssey4me | you can scale by adding more hosts or adding more containers to hosts, depending on what you're looking to do | 10:56 |
| drifterza | I'm thinking that 2 am call that requires something to be added | 10:56 |
| drifterza | etc.. | 10:56 |
| odyssey4me | yeah, it'd need metric monitoring and quite a few changes to how we implement the inventory | 10:56 |
| drifterza | would be kinda cool, sort of like k8s | 10:56 |
| odyssey4me | We've seen a 3 infra node environment scale to 500 compute nodes though... so while it'd be cool, there isn't really a need for it | 10:57 |
| drifterza | has there been any considerations on how to monitor the infra ? | 10:58 |
| drifterza | not perscribed monitoring, but rather a means. | 10:58 |
| evrardjp | andymccr: latest patch of nova_role in master changed the code that did the nova_virt_type during detection | 10:59 |
| evrardjp | I think there is a problem somewhere there, because it's sometime skipping the task and not setting anything. | 10:59 |
| evrardjp | I will probably have a patch then | 10:59 |
| andymccr | evrardjp: ok cool - how did it pass gates? | 10:59 |
| odyssey4me | drifterza there's been some work going into putting together a generic set of monitoring scripts in the repo openstack/monitorstack | 11:00 |
| odyssey4me | https://github.com/openstack/monitorstack | 11:00 |
| *** toddnni has joined #openstack-ansible | 11:00 | |
| odyssey4me | those can be used for zabbix, nagios or whatever | 11:00 |
| *** fergus has quit IRC | 11:01 | |
| *** smatzek has joined #openstack-ansible | 11:01 | |
| evrardjp | andymccr: I haven't a patch yet to fix this yet. | 11:02 |
| evrardjp | But it should be done soon | 11:02 |
| evrardjp | I guess I know the issue. | 11:03 |
| evrardjp | or I will soon when I'll read the patch that broke it :p | 11:03 |
| andymccr | hahaha | 11:04 |
| andymccr | evrardjp: ok cool let me know if you need more info | 11:04 |
| andymccr | or rather | 11:04 |
| andymccr | need help :P | 11:04 |
| andymccr | dunno why i said info | 11:04 |
| *** vnogin has quit IRC | 11:04 | |
| asettle | Whhy not both! | 11:05 |
| *** thorst_afk has joined #openstack-ansible | 11:18 | |
| *** chhavi has quit IRC | 11:20 | |
| openstackgerrit | Jean-Philippe Evrard proposed openstack/openstack-ansible-os_nova master: Ensure tasks are not skipped https://review.openstack.org/466687 | 11:21 |
| odyssey4me | evrardjp ah, I think I see where the os_nova role is failing to get the virt type | 11:21 |
| evrardjp | andymccr: odyssey4me ^ | 11:21 |
| evrardjp | odyssey4me: see my commit message :) | 11:21 |
| *** toddnni has quit IRC | 11:22 | |
| odyssey4me | evrardjp hehe, yeah - you got it exactly | 11:22 |
| *** armaan has quit IRC | 11:23 | |
| evrardjp | odyssey4me: at the same time I have realised this needs a refactor :p | 11:24 |
| evrardjp | what can run on itself is definitely a fact :p | 11:24 |
| *** armaan has joined #openstack-ansible | 11:24 | |
| evrardjp | but why do we need it ephemeral? | 11:24 |
| evrardjp | :p | 11:24 |
| odyssey4me | what do you mean? | 11:24 |
| odyssey4me | thinking about it, we could actually just set qemu as the default, then let set_fact override it | 11:25 |
| evrardjp | set_fact: nova_virt_type: kvm blablabla set_fact nova_virt_type powervm ... | 11:25 |
| evrardjp | yes ! | 11:25 |
| evrardjp | next commit | 11:25 |
| evrardjp | and using host facts to not look it up over and over again. | 11:25 |
| odyssey4me | heh, well actually a local fact would be better | 11:26 |
| evrardjp | exactly | 11:26 |
| evrardjp | something simple as a shell script we would drop | 11:26 |
| evrardjp | that returns "kvm" "powervm" or "qemu" | 11:26 |
| evrardjp | :p | 11:26 |
| odyssey4me | yep, using this mechanism: http://serverascode.com/2015/01/27/ansible-custom-facts.html | 11:27 |
| *** shardy has quit IRC | 11:27 | |
| evrardjp | yes, exactly what I had in mind. :p | 11:27 |
| odyssey4me | perhaps you should register a bug for it to remind yourself ;) | 11:27 |
| drifterza | what does rackspace call your team odyssey4me ? | 11:27 |
| drifterza | is it devops | 11:27 |
| hwoarang | evrardjp: another reason why centos and suse behave differently is that centos7 uses 3.6.X and suse 3.5.X so maybe rabbitmqctl was fixed in the meantime... | 11:28 |
| odyssey4me | drifterza heh, no | 11:28 |
| odyssey4me | drifterza we're part of a software engineering team | 11:28 |
| *** armaan has quit IRC | 11:28 | |
| evrardjp | hwoarang: interesting. You mean both had HOSTNAME set, and one didn't had the failure? Interesting | 11:29 |
| evrardjp | very interesting | 11:29 |
| hwoarang | yeah | 11:29 |
| hwoarang | centos has HOSTNAME set. suse also has HOST set as well | 11:29 |
| hwoarang | both have the same value | 11:29 |
| * hwoarang context switches to galera_server | 11:30 | |
| evrardjp | afk lunch | 11:32 |
| evrardjp | woops. | 11:32 |
| evrardjp | well I guess you all know that I am starving. | 11:32 |
| *** karimb has quit IRC | 11:32 | |
| *** toddnni has joined #openstack-ansible | 11:33 | |
| ioni | is it ok that now gnocchi client is not available in gnocchi container? | 11:34 |
| ioni | in utility container gnocchi returns that it cannot connect on localhost | 11:34 |
| ioni | so it doesn't get its endpoint for some reason | 11:35 |
| ioni | i'm on ocata | 11:35 |
| *** udesale has quit IRC | 11:36 | |
| *** zul has joined #openstack-ansible | 11:36 | |
| *** armaan has joined #openstack-ansible | 11:38 | |
| evrardjp | mmm | 11:40 |
| evrardjp | I am no expert and the only thing I can remember about endpoints change was for master (keystone versionless) | 11:43 |
| evrardjp | anyone else? | 11:43 |
| *** admin0 has joined #openstack-ansible | 11:44 | |
| *** askb has quit IRC | 11:44 | |
| admin0 | \o | 11:44 |
| admin0 | from june, i will be doing OSA full time \o/ | 11:45 |
| *** jvidal has quit IRC | 11:54 | |
| evrardjp | admin0: lucky you. We expect your commits then! | 11:54 |
| admin0 | i think that is a yes :) | 11:54 |
| openstackgerrit | Merged openstack/openstack-ansible-security master: [Docs] Fix missing hyphen in status https://review.openstack.org/466492 | 11:56 |
| openstackgerrit | Merged openstack/openstack-ansible-security master: [Docs] Put FAQ after getting started docs https://review.openstack.org/466493 | 11:56 |
| openstackgerrit | Merged openstack/openstack-ansible-security stable/ocata: [Docs] Fix deprecation docs for RHEL 6 content https://review.openstack.org/466491 | 11:56 |
| *** yifei has joined #openstack-ansible | 11:56 | |
| odyssey4me | andymccr we should probably get https://review.openstack.org/466099 merged asap, and port that one back to newton | 11:56 |
| jrosser_ | odyssey4me: https://github.com/openstack/openstack-ansible-plugins/commit/67d9ed2491fb01a6904df39fc826e4f4db897abe <- that seems good on ocata, just got our lab redeployed with that patch | 11:57 |
| odyssey4me | jrosser_ oh good, would you like to propose the patch then? | 11:57 |
| jrosser_ | and you are right there are still a few bits to address, db connectivity check etc | 11:58 |
| openstackgerrit | Jonathan Rosser proposed openstack/openstack-ansible-plugins stable/ocata: Support delegation in strategy plugin https://review.openstack.org/466703 | 11:59 |
| *** yifei has quit IRC | 12:00 | |
| odyssey4me | thanks | 12:01 |
| *** fabg has joined #openstack-ansible | 12:05 | |
| *** pranavb has quit IRC | 12:06 | |
| *** jvidal has joined #openstack-ansible | 12:09 | |
| fabg | ioni: for gnocchi client in utility container, after source openrc, export OS_AUTH_TYPE=password and then do a "gnocchi status" | 12:09 |
| *** armaan has quit IRC | 12:16 | |
| openstackgerrit | Merged openstack/openstack-ansible-security stable/ocata: Do not update grub if grub not used https://review.openstack.org/463797 | 12:16 |
| *** armaan has joined #openstack-ansible | 12:16 | |
| *** admin0 has quit IRC | 12:19 | |
| *** armaan has quit IRC | 12:20 | |
| *** jgriffiths has quit IRC | 12:22 | |
| openstackgerrit | Andy McCrae proposed openstack/openstack-ansible-os_nova master: Fix upgrades to map existing nova instances to cell1 https://review.openstack.org/466717 | 12:28 |
| *** foutatoro has joined #openstack-ansible | 12:30 | |
| ioni | fabg, thanks for the hint | 12:32 |
| ioni | fabg, is that a bug? | 12:32 |
| *** admin0 has joined #openstack-ansible | 12:33 | |
| *** armaan has joined #openstack-ansible | 12:34 | |
| fabg | ioni: don't really know ? I got this info from the irc… Just know about this issue… ;-) | 12:39 |
| *** chhavi has joined #openstack-ansible | 12:39 | |
| *** retreved has joined #openstack-ansible | 12:41 | |
| *** admin0 has quit IRC | 12:46 | |
| openstackgerrit | Alexandra Settle proposed openstack/openstack-ansible master: [ops-guide] Adds links to networking info in ops and networking guide https://review.openstack.org/466721 | 12:48 |
| evrardjp | weirdly I don't have the same behavior on a PoC on my machine and in the gates. Same version of ansible. Not the same env. | 12:55 |
| evrardjp | http://logs.openstack.org/87/466687/1/check/gate-openstack-ansible-os_nova-ansible-func-ubuntu-xenial/91ba86d/console.html#_2017-05-22_11_44_02_396493 | 12:55 |
| evrardjp | if anyone knows why it tries to load common/drivers/kvm/nova_compute_kvm_install.yml instead of tasks/drivers/kvm/nova_compute_kvm_install.yml I'm all ears (probably an env var) | 12:56 |
| *** karimb has joined #openstack-ansible | 12:56 | |
| *** admin0 has joined #openstack-ansible | 12:56 | |
| *** hogwash has joined #openstack-ansible | 12:59 | |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible-os_keystone master: Add venv_tag local fact https://review.openstack.org/466723 | 13:00 |
| *** manheim_ has quit IRC | 13:01 | |
| *** esberglu has joined #openstack-ansible | 13:01 | |
| *** jheroux has joined #openstack-ansible | 13:01 | |
| odyssey4me | palendae evrardjp ^ I intend to apply that as a pattern, so any thoughts are welcome. | 13:01 |
| *** manheim_ has joined #openstack-ansible | 13:02 | |
| *** retreved has quit IRC | 13:02 | |
| *** esberglu has quit IRC | 13:03 | |
| *** udesale has joined #openstack-ansible | 13:05 | |
| *** esberglu has joined #openstack-ansible | 13:05 | |
| *** coolkil has joined #openstack-ansible | 13:05 | |
| *** zul has quit IRC | 13:05 | |
| *** esberglu_ has joined #openstack-ansible | 13:06 | |
| *** klamath has joined #openstack-ansible | 13:06 | |
| evrardjp | odyssey4me: you want to ensure that a host x has a version y of code, right? | 13:06 |
| *** manheim_ has quit IRC | 13:06 | |
| *** klamath has quit IRC | 13:06 | |
| evrardjp | this way we can later fetch info on a smaller matrix | 13:06 |
| evrardjp | right? | 13:06 |
| *** retreved has joined #openstack-ansible | 13:06 | |
| odyssey4me | I just want to record the version of software implemented as a fact. | 13:07 |
| *** klamath has joined #openstack-ansible | 13:07 | |
| odyssey4me | This will allows us to serialise the playbook based on the facts of the targets. | 13:07 |
| odyssey4me | ie for a new deployment, don't serialise - for an existing deployment, serialise | 13:07 |
| odyssey4me | so if you're doing a version upgrade, it'll serialise the playbook | 13:08 |
| logan- | nice | 13:08 |
| odyssey4me | so this is to try and implement rolling upgrades | 13:08 |
| odyssey4me | initially I'm just focusing on minor upgrades, but major upgrades will come later | 13:08 |
| evrardjp | yes as we discussed, it's an approach I like. | 13:09 |
| evrardjp | :D | 13:09 |
| evrardjp | not sure how you're gonna serialize but that's for later :) | 13:09 |
| *** esberglu has quit IRC | 13:09 | |
| evrardjp | at least you can skip what you don't want | 13:09 |
| odyssey4me | yes, this should also allow us to skip the entire set of install tasks if the tag deployed is the same as the tag intended for deployment | 13:10 |
| odyssey4me | ie just a config change | 13:10 |
| *** zul has joined #openstack-ansible | 13:11 | |
| *** rcarrillocruz has quit IRC | 13:11 | |
| odyssey4me | of course using tags is more efficient, but this opens the opportunity to optimise for people who don't know ansible all that well | 13:11 |
| logan- | how does this fact look when it is loaded? ansible_local.venv_tag? or ansible_local.keystone.general.venv_tag? | 13:11 |
| odyssey4me | logan- the latter | 13:12 |
| *** woodard has joined #openstack-ansible | 13:12 | |
| *** cpuga has joined #openstack-ansible | 13:12 | |
| odyssey4me | https://pastebin.com/rrNA40ev | 13:12 |
| *** vnogin has joined #openstack-ansible | 13:13 | |
| *** woodard has quit IRC | 13:13 | |
| *** woodard has joined #openstack-ansible | 13:14 | |
| *** cpuga_ has joined #openstack-ansible | 13:15 | |
| odyssey4me | for those in the UK, a new channel has been started: #openstack-uk - please join if you're interested git-harry mancdaz andymccr asettle boxrick1 jrosser_ hughsaunders robcresswell and any others who I may not know are in the UK | 13:16 |
| evrardjp | I'll join anyway. | 13:16 |
| *** armaan has quit IRC | 13:16 | |
| mancdaz | odyssey4me why a new channel? | 13:17 |
| asettle | Looks like it might be for openstack days UK | 13:17 |
| asettle | In Sept | 13:17 |
| asettle | Community stuff? | 13:17 |
| odyssey4me | mancdaz not my initiative, ask dabukalam :) I'm just passing the message on. | 13:17 |
| mancdaz | fair enough | 13:17 |
| *** cpuga has quit IRC | 13:18 | |
| *** admin0_ has joined #openstack-ansible | 13:18 | |
| dabukalam | mancdaz: just somewhere to collect UK people, make people aware of days, and local UG meetups | 13:18 |
| dabukalam | like this one https://www.meetup.com/Manchester-OpenStack-Meetup/events/239507652/ | 13:18 |
| dabukalam | :D | 13:18 |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible-lxc_container_create master: Add variant local fact https://review.openstack.org/466727 | 13:21 |
| odyssey4me | dabukalam you're such a spammer :p | 13:21 |
| *** cathrich_ has joined #openstack-ansible | 13:21 | |
| *** cathrichardson has quit IRC | 13:22 | |
| odyssey4me | logan- you raise a good point re local facts - perhaps we should collect all of OSA's static local facts in an 'openstack-ansible' fact file and use a section for each role? | 13:22 |
| odyssey4me | or perhaps just do that for the openstack stuff at least | 13:22 |
| odyssey4me | the infra stuff could perhaps be seperate | 13:22 |
| *** markvoelker has joined #openstack-ansible | 13:22 | |
| odyssey4me | evrardjp ^ thoughts? | 13:23 |
| evrardjp | Still thinking | 13:23 |
| evrardjp | we'll have multiple files if we are loading things dynamically | 13:24 |
| *** jwitko has joined #openstack-ansible | 13:25 | |
| evrardjp | but I don't see a problem of having multiple files, or having a few files next to an "openstack" one | 13:25 |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible-lxc_container_create master: Add variant local fact https://review.openstack.org/466727 | 13:25 |
| odyssey4me | yes, obviously dynamic fact generation will have to be excluded from the pattern, that's why I suggested this only for 'static local facts' | 13:26 |
| odyssey4me | evrardjp I think I like using 'openstack_ansible.fact' and having a section for each role - so [keystone] for keystone, etc | 13:28 |
| evrardjp | yup, but you have to remember you're gonna have supported_virt_types.fact for example | 13:29 |
| evrardjp | I am fine with openstack_ansible.fact | 13:29 |
| evrardjp | because it's still predictable | 13:30 |
| *** kylek3h has quit IRC | 13:30 | |
| evrardjp | I'd be against using ansible_hostname as a filename for example | 13:30 |
| odyssey4me | hmm, I see that the file name is a parent | 13:31 |
| evrardjp | predictable and consistent accross hosts of a group I mean | 13:31 |
| *** admin0_ has quit IRC | 13:32 | |
| *** manheim_ has joined #openstack-ansible | 13:32 | |
| *** fabg has quit IRC | 13:34 | |
| *** smatzek has quit IRC | 13:34 | |
| *** markvoelker has quit IRC | 13:35 | |
| *** fabg has joined #openstack-ansible | 13:35 | |
| evrardjp | openstack_ansible [keystone] lxc_container_variant: default is fine for me | 13:35 |
| *** markvoelker has joined #openstack-ansible | 13:35 | |
| evrardjp | but I'm not sure we need to track that inside the container | 13:36 |
| logan- | yeah that sounds good | 13:36 |
| evrardjp | that doesn't make sense to me, but the idea is there | 13:36 |
| evrardjp | for me it's the host that has a knowledge of running containers. Containers don't need to be "self aware" of their version | 13:37 |
| evrardjp | but that's maybe details | 13:37 |
| evrardjp | containers are implementation details for the hosts, where venvs version are implentation details of the deployed code (therefore on the containers) | 13:38 |
| evrardjp | but any hack is good at our point :p | 13:39 |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible-os_keystone master: Add venv_tag local fact https://review.openstack.org/466723 | 13:40 |
| odyssey4me | evrardjp logan- ^ | 13:41 |
| odyssey4me | evrardjp the lxc-container-create role targets containers, so if we were to do, for example, a rolling replacement, we'd be reacting to the facts for the container not the host | 13:42 |
| *** drifterza has quit IRC | 13:42 | |
| odyssey4me | It's also the easiest way to track per-container variants used - otherwise the host's fact file would have to contain an entry for every container. | 13:43 |
| odyssey4me | IT'll be somewhat messy, and I'd rather tackle that if/when we move towards making the containers a service/property of the host instead of a host in its own right, | 13:43 |
| evrardjp | odyssey4me: it started with a good idea. | 13:43 |
| evrardjp | odyssey4me: I agree with that part | 13:44 |
| evrardjp | I mean the reading fact of linked to the inventory | 13:44 |
| odyssey4me | evrardjp ok, back to thekeystone patch | 13:45 |
| odyssey4me | I made it a var for maintenance purposes, not to allow changes - that's why it's in vars/main.yml | 13:45 |
| evrardjp | please not the var :p | 13:45 |
| evrardjp | yes | 13:45 |
| odyssey4me | if you think that's too tempting, then I can change it | 13:45 |
| evrardjp | oh yes | 13:45 |
| evrardjp | ansible_local[whatwasthefilenameagain?DamnIhavetoreadthevar] | 13:46 |
| evrardjp | from outside the role | 13:46 |
| *** udesale has quit IRC | 13:46 | |
| evrardjp | that's a good example of what I think we should avoid (same as inventory_hostname) | 13:47 |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible-os_keystone master: Add venv_tag local fact https://review.openstack.org/466723 | 13:48 |
| evrardjp | if we keep one single file, we could even have the file created in lxc_container_create. | 13:50 |
| odyssey4me | well, evrardjp logan- what do you think - should we store the infrastructure facts in the same file, or do something different? | 13:51 |
| odyssey4me | same file seems fine to me | 13:52 |
| logan- | same file | 13:52 |
| logan- | osa managed infra seems appropriate there | 13:52 |
| *** schwicht has joined #openstack-ansible | 13:52 | |
| odyssey4me | yep | 13:52 |
| palendae | Infra facts like mariadb versions? | 13:53 |
| odyssey4me | palendae well, I'll be pushing up a patch for that shortly - as it'll be a dyanmic fact it'll be a different file | 13:53 |
| evrardjp | hwoarang: I got an idea | 13:53 |
| *** hogwash has quit IRC | 13:53 | |
| palendae | odyssey4me: Ah; what would be in the same file here then? | 13:54 |
| evrardjp | hwoarang: what if on the tasks themselves we directly do environment: hostname: {{ ansible_hostname }} | 13:54 |
| *** ivveh has joined #openstack-ansible | 13:54 | |
| odyssey4me | this file will be for static facts only - for for example the variant used when creating the container | 13:54 |
| palendae | Sure | 13:54 |
| admin0 | if every bridge in the nodes are ovs bridges, does OSA still create containers on top of it, or does it need to be a LinuxBridge at some point ? | 13:55 |
| palendae | I'm not 100% sure what you mean by 'infrastructure' here, but fewer files is fine by me :p | 13:55 |
| odyssey4me | in our artifacted deployment the variant is <service name>-<release tag>, so that'll allow us to flip the recreate switch :) | 13:55 |
| evrardjp | odyssey4me: don't forget the group_by :) | 13:56 |
| *** fabg has quit IRC | 13:56 | |
| *** fabg has joined #openstack-ansible | 13:56 | |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible-lxc_container_create master: Add variant local fact https://review.openstack.org/466727 | 13:56 |
| *** kylek3h has joined #openstack-ansible | 13:56 | |
| palendae | evrardjp: Replied on https://review.openstack.org/#/c/466108/ | 13:57 |
| palendae | Also going to need at least one more patch on https://review.openstack.org/#/c/461884/, evidently I messed up the conditonal I spent an hour trying to formulate, heh | 13:57 |
| *** smatzek has joined #openstack-ansible | 13:58 | |
| *** cjloader_ has joined #openstack-ansible | 14:00 | |
| evrardjp | palendae: yes | 14:02 |
| evrardjp | "'mac' in ansible_local" | 14:02 |
| *** smatzek_ has joined #openstack-ansible | 14:02 | |
| evrardjp | "inventory_hostname in ansible_local['mac']" | 14:02 |
| evrardjp | IIRC | 14:02 |
| palendae | Ahh | 14:02 |
| palendae | Wouldn't that be around the whole line? | 14:02 |
| palendae | Since there's an "or"? | 14:03 |
| evrardjp | yes, that's what I just did? | 14:03 |
| odyssey4me | do we really need to use local facts for the static mac recreation? | 14:03 |
| evrardjp | well it makes sense to me | 14:03 |
| odyssey4me | surely we can just gather the network facts for the existing container, then use those to recreate it? | 14:03 |
| palendae | evrardjp: They were separate lines, just wondering :p | 14:03 |
| evrardjp | wait | 14:03 |
| odyssey4me | unless you're trying to allow a time period between destroy and create for some reason? | 14:04 |
| evrardjp | odyssey4me: if you have a better approach we can iterate :) | 14:04 |
| palendae | odyssey4me: It's possible, yes; I had another patch that tried this but it was much messier than this. 1s | 14:04 |
| *** lostRhino has joined #openstack-ansible | 14:04 | |
| palendae | https://review.openstack.org/#/c/453333/7 | 14:05 |
| evrardjp | but it's kinda simple right now: You have a fact? Use it, You don't? Generate and use directly + create the fact | 14:05 |
| *** smatzek has quit IRC | 14:06 | |
| odyssey4me | sure, alright - I don't want to throw a spanner in the works... it was just a thought that popped into my head | 14:07 |
| evrardjp | well the thing is, for those who exist already, we don't want to destroy their mac address | 14:07 |
| palendae | odyssey4me: I don't think you're wrong, but the weird hoops for accessing the facts as they were wasn't desirable in my mind | 14:07 |
| evrardjp | else it would be simpler :p | 14:07 |
| palendae | My first pass did gather them into local facts just before deletion | 14:08 |
| palendae | Then try to access them, but the fact name used is wonky | 14:08 |
| evrardjp | palendae: I think current one is quite elegant | 14:08 |
| palendae | evrardjp: I'm not sure I would call it that, but I do think it's a bit safer | 14:09 |
| palendae | Just in case someone does something "wrong" and deletes containers manually | 14:09 |
| evrardjp | I think it could have been cleaner if we started from a blank sheet, but we aren't | 14:09 |
| palendae | Right | 14:09 |
| *** fabg has quit IRC | 14:09 | |
| evrardjp | as a reminder, from the first iteration of mac storage: I personally find this disgusting :p | 14:10 |
| palendae | evrardjp: Same | 14:10 |
| evrardjp | but we are where we are :p | 14:10 |
| palendae | We're doing a lot of the work that docker/lxd would do but in a templating language | 14:11 |
| evrardjp | we just need bgp | 14:11 |
| evrardjp | mp-bgp | 14:11 |
| evrardjp | that would be good enough | 14:12 |
| *** fabg has joined #openstack-ansible | 14:12 | |
| evrardjp | but then it becomes too complex | 14:12 |
| evrardjp | for some pp | 14:13 |
| evrardjp | ppl* | 14:13 |
| openstackgerrit | Merged openstack/openstack-ansible-os_nova stable/ocata: Remove default for nova_console_keymap https://review.openstack.org/466099 | 14:16 |
| *** jgriffiths has joined #openstack-ansible | 14:18 | |
| *** piliman974 has joined #openstack-ansible | 14:18 | |
| neith | hello all | 14:19 |
| *** redaudit has joined #openstack-ansible | 14:20 | |
| *** jvidal has quit IRC | 14:21 | |
| *** fabg has quit IRC | 14:23 | |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible-os_keystone master: Add venv_tag local fact https://review.openstack.org/466723 | 14:24 |
| *** cjloader_ has quit IRC | 14:24 | |
| odyssey4me | logan- evrardjp ^ forgot to change the fact conditional too | 14:25 |
| klamath | cloudnull odyssey4me found the error with https://gist.github.com/odyssey4me/4af6a759b7ce1a4df9b36df412f57f0a horizon_keystone_endpoint: "{{ keystone_service_publicurl }}/v3" hits the public keystone endpoint that wont allow admin to login to horizon | 14:25 |
| evrardjp | odyssey4me: and fixed my laziness of . moved properly to [' '] | 14:26 |
| odyssey4me | yes, it's beautiful again :p | 14:27 |
| *** fabg has joined #openstack-ansible | 14:27 | |
| evrardjp | . is so easy to type and python is too lenient | 14:29 |
| evrardjp | jinja* | 14:29 |
| *** markvoelker_ has joined #openstack-ansible | 14:29 | |
| openstackgerrit | Logan V proposed openstack/openstack-ansible-plugins master: Add container-aware connection plugin tests https://review.openstack.org/466743 | 14:29 |
| *** sliver has joined #openstack-ansible | 14:30 | |
| mhayden | mpranjic: good comment on Debian 9 vs 8 | 14:30 |
| mhayden | i need to take a looksee there | 14:30 |
| *** mcrafty has quit IRC | 14:32 | |
| *** markvoelker has quit IRC | 14:33 | |
| *** woodard has quit IRC | 14:33 | |
| *** david-lyle has quit IRC | 14:34 | |
| *** vnogin has quit IRC | 14:34 | |
| evrardjp | could someone help me on something? | 14:34 |
| *** david-lyle has joined #openstack-ansible | 14:35 | |
| evrardjp | http://logs.openstack.org/87/466687/1/check/gate-openstack-ansible-os_nova-ansible-func-ubuntu-xenial/91ba86d/logs/ara/file/a33dc9f6-0c6a-49a9-a362-8e38a3cacea6/#line-16 | 14:35 |
| evrardjp | http://logs.openstack.org/87/466687/1/check/gate-openstack-ansible-os_nova-ansible-func-ubuntu-xenial/91ba86d/logs/ara/result/0c57bde2-e451-401f-8896-1e32ceb9c970/ | 14:35 |
| openstackgerrit | Jean-Philippe Evrard proposed openstack/openstack-ansible-os_nova master: [WIP] Ensure tasks are not skipped https://review.openstack.org/466687 | 14:36 |
| evrardjp | This is really weird. | 14:36 |
| odyssey4me | yeah, saw that earlier - did a little poking around but gave up | 14:37 |
| odyssey4me | without actually building a test environment, that's a hard one to figure out | 14:37 |
| *** vnogin has joined #openstack-ansible | 14:37 | |
| evrardjp | oh yes that remind me! | 14:37 |
| evrardjp | that what I was starting up. | 14:37 |
| andymccr | hmm | 14:38 |
| evrardjp | LIFO work | 14:38 |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible-os_cinder master: Add venv_tag local fact https://review.openstack.org/466750 | 14:38 |
| andymccr | seems wonky - common isnt even listed in the role structure for ansible best practices. | 14:38 |
| *** admin0 has quit IRC | 14:39 | |
| logan- | playbook_dir is probably a low precedence fallback | 14:39 |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible-os_cinder master: Add venv_tag local fact https://review.openstack.org/466750 | 14:40 |
| odyssey4me | I think that logan- may just be right on the money there | 14:40 |
| *** weezS has quit IRC | 14:41 | |
| logan- | since the include is occurring in tasks/drivers/qemu/nova_compute_qemu.yml, maybe try "../kvm/nova_compute_kvm_install.yml" as the task file path | 14:41 |
| *** armaan has joined #openstack-ansible | 14:42 | |
| *** fabg has quit IRC | 14:42 | |
| *** smatzek_ is now known as smatzek | 14:42 | |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible-os_glance master: Add venv_tag local fact https://review.openstack.org/466754 | 14:43 |
| openstackgerrit | Merged openstack/openstack-ansible-ops master: Make possible to start leap from K https://review.openstack.org/463659 | 14:43 |
| openstackgerrit | Merged openstack/openstack-ansible-ops master: Implement CLI test picker https://review.openstack.org/466108 | 14:44 |
| openstackgerrit | Logan V proposed openstack/openstack-ansible-plugins master: Add container-aware connection plugin tests https://review.openstack.org/466743 | 14:44 |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible-os_heat master: Add venv_tag local fact https://review.openstack.org/466755 | 14:45 |
| cloudnull | mornings | 14:45 |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible-os_neutron master: Add venv_tag local fact https://review.openstack.org/466757 | 14:47 |
| openstackgerrit | Merged openstack/openstack-ansible-plugins stable/ocata: Support delegation in strategy plugin https://review.openstack.org/466703 | 14:47 |
| *** admin0 has joined #openstack-ansible | 14:48 | |
| *** schwicht has quit IRC | 14:48 | |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible-os_nova master: Add venv_tag local fact https://review.openstack.org/466758 | 14:48 |
| openstackgerrit | Merged openstack/openstack-ansible stable/newton: Add gnocchiclient constraints to global pins https://review.openstack.org/466660 | 14:48 |
| *** fabg has joined #openstack-ansible | 14:48 | |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible-os_swift master: Add venv_tag local fact https://review.openstack.org/466759 | 14:51 |
| *** david-lyle has quit IRC | 14:52 | |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible-os_tempest master: Add venv_tag local fact https://review.openstack.org/466761 | 14:55 |
| *** lucasxu has joined #openstack-ansible | 14:56 | |
| *** piliman974 has quit IRC | 14:56 | |
| *** piliman974 has joined #openstack-ansible | 14:57 | |
| *** fabg has quit IRC | 14:58 | |
| mpranjic | mhayden: thnx | 14:59 |
| mpranjic | mhayden: I am testing migrations from deb8 to deb9. it is really stable | 14:59 |
| neith | would you used a RAID0 of 6 drives or a lvm stripped volume with 6 drivers to host /var/lib/nova? | 14:59 |
| mpranjic | it is in freeze for quite some time now. looks ready to me | 14:59 |
| mhayden | mpranjic: seems to run fine on debian testing | 15:02 |
| mhayden | i just ran it right quick | 15:02 |
| mpranjic | nice | 15:04 |
| *** woodard has joined #openstack-ansible | 15:06 | |
| openstackgerrit | Nolan Brubaker proposed openstack/openstack-ansible-lxc_container_create master: Write MAC addresses to local facts folder https://review.openstack.org/461884 | 15:07 |
| openstackgerrit | Nolan Brubaker proposed openstack/openstack-ansible-lxc_container_create master: Add option to recreate containers https://review.openstack.org/463403 | 15:08 |
| *** woodard has quit IRC | 15:08 | |
| *** woodard has joined #openstack-ansible | 15:09 | |
| openstackgerrit | Merged openstack/openstack-ansible-tests master: test-prepare-host: Remove useless filter for RedHat network ifcfg files https://review.openstack.org/466421 | 15:09 |
| andymccr | https://review.openstack.org/#/c/466717/ - reviews por favor :) fixes the upgrade issues on nova, will need a backport to stable/ocata. | 15:09 |
| openstackgerrit | Merged openstack/openstack-ansible-tests master: SUSE: Remove post-up/post-down scripts from default ifcfg files https://review.openstack.org/466422 | 15:09 |
| openstackgerrit | Merged openstack/openstack-ansible-os_ceilometer stable/newton: Revert "Pin gnocchiclient to comply with optional deps" https://review.openstack.org/466663 | 15:10 |
| openstackgerrit | Merged openstack/openstack-ansible-lxc_container_create master: Add variant local fact https://review.openstack.org/466727 | 15:11 |
| *** Oku_OS is now known as Oku_OS-away | 15:18 | |
| openstackgerrit | Merged openstack/openstack-ansible-os_designate stable/ocata: Fix for lookup and get_url occuring in different places https://review.openstack.org/463785 | 15:19 |
| *** coolkil has quit IRC | 15:24 | |
| *** fabg has joined #openstack-ansible | 15:24 | |
| *** cjloader_ has joined #openstack-ansible | 15:24 | |
| *** galstrom_zzz is now known as galstrom | 15:25 | |
| openstackgerrit | Nolan Brubaker proposed openstack/openstack-ansible-ops master: Make usage of exceptions more uniform https://review.openstack.org/466769 | 15:25 |
| palendae | evrardjp: ^ added the keystone exception catching | 15:25 |
| *** weezS has joined #openstack-ansible | 15:26 | |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible-os_aodh master: Add venv_tag local fact https://review.openstack.org/466770 | 15:28 |
| *** jvidal has joined #openstack-ansible | 15:28 | |
| *** armaan has quit IRC | 15:28 | |
| *** armaan has joined #openstack-ansible | 15:29 | |
| *** hogwash has joined #openstack-ansible | 15:29 | |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible-os_barbican master: Add venv_tag local fact https://review.openstack.org/466771 | 15:29 |
| openstackgerrit | Merged openstack/openstack-ansible-os_glance master: Add venv_tag local fact https://review.openstack.org/466754 | 15:34 |
| *** fabg has quit IRC | 15:34 | |
| *** markvoelker_ has quit IRC | 15:35 | |
| *** jvidal has quit IRC | 15:36 | |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible-os_ceilometer master: Add venv_tag local fact https://review.openstack.org/466779 | 15:36 |
| openstackgerrit | Nolan Brubaker proposed openstack/openstack-ansible-ops master: Allow graceful exiting with SIGINT and SIGTERM https://review.openstack.org/466780 | 15:38 |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible-os_designate master: Add venv_tag local fact https://review.openstack.org/466781 | 15:38 |
| openstackgerrit | Merged openstack/openstack-ansible-os_heat master: Add venv_tag local fact https://review.openstack.org/466755 | 15:39 |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible-os_gnocchi master: Add venv_tag local fact https://review.openstack.org/466783 | 15:40 |
| xdfil | stupid question: where do I submit a bug? | 15:41 |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible-os_horizon master: Add venv_tag local fact https://review.openstack.org/466784 | 15:41 |
| *** piliman974 has quit IRC | 15:42 | |
| *** hogwash has quit IRC | 15:42 | |
| andymccr | xdfil: https://bugs.launchpad.net/openstack-ansible/+filebug | 15:42 |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible-os_ironic master: Add venv_tag local fact https://review.openstack.org/466786 | 15:42 |
| *** piliman974 has joined #openstack-ansible | 15:43 | |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible-os_magnum master: Add venv_tag local fact https://review.openstack.org/466788 | 15:43 |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible-os_molteniron master: Add venv_tag local fact https://review.openstack.org/466789 | 15:45 |
| openstackgerrit | Merged openstack/openstack-ansible-os_keystone master: Add venv_tag local fact https://review.openstack.org/466723 | 15:45 |
| openstackgerrit | Nolan Brubaker proposed openstack/openstack-ansible-ops master: Allow graceful exiting with SIGINT and SIGTERM https://review.openstack.org/466780 | 15:45 |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible-os_octavia master: Add venv_tag local fact https://review.openstack.org/466790 | 15:46 |
| *** fabg has joined #openstack-ansible | 15:47 | |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible-os_rally master: Add venv_tag local fact https://review.openstack.org/466791 | 15:47 |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible-os_sahara master: Add venv_tag local fact https://review.openstack.org/466792 | 15:48 |
| *** jvidal has joined #openstack-ansible | 15:49 | |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible-os_trove master: Add venv_tag local fact https://review.openstack.org/466793 | 15:49 |
| *** admin0 has quit IRC | 15:50 | |
| *** admin0 has joined #openstack-ansible | 15:50 | |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible-os_trove master: Add venv_tag local fact https://review.openstack.org/466793 | 15:51 |
| *** admin0 has quit IRC | 15:51 | |
| openstackgerrit | Nolan Brubaker proposed openstack/openstack-ansible-ops master: Allow graceful exiting with SIGINT and SIGTERM https://review.openstack.org/466780 | 15:51 |
| palendae | odyssey4me: ^ should be the last general framework thing we need, unless we want an aggregated log file | 15:51 |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible-os_magnum master: Add venv_tag local fact https://review.openstack.org/466788 | 15:51 |
| odyssey4me | palendae if we could log to syslog, then it'll aggregate for us | 15:52 |
| palendae | odyssey4me: I mean like aggregate in terms of x failures of y attempts | 15:53 |
| palendae | I guess we can shove there and parse later | 15:53 |
| odyssey4me | palendae to make it useful in the gate we will need some sort of parsable summary so that we can make it fail if it hits a threshold | 15:53 |
| palendae | Right | 15:54 |
| palendae | Ok, I'll make it go to syslog and output some sort of summary | 15:54 |
| odyssey4me | it could do log to a file/syslog, but output a summary to stdout. | 15:55 |
| palendae | Sure | 15:55 |
| palendae | I'll run with that that assumption | 15:55 |
| openstackgerrit | Merged openstack/openstack-ansible-os_molteniron master: Add venv_tag local fact https://review.openstack.org/466789 | 15:55 |
| odyssey4me | cool, thanks | 15:55 |
| *** woodard has quit IRC | 15:55 | |
| *** woodard has joined #openstack-ansible | 15:56 | |
| *** jvidal has quit IRC | 15:56 | |
| *** manheim_ has quit IRC | 15:57 | |
| odyssey4me | palendae small spelling error in https://review.openstack.org/466769 | 15:58 |
| *** lbragstad has joined #openstack-ansible | 15:58 | |
| *** pcaruana has quit IRC | 15:59 | |
| openstackgerrit | Nolan Brubaker proposed openstack/openstack-ansible-ops master: Allow graceful exiting with SIGINT and SIGTERM https://review.openstack.org/466780 | 15:59 |
| palendae | odyssey4me: fixed | 15:59 |
| openstackgerrit | Merged openstack/openstack-ansible-ops master: Make usage of exceptions more uniform https://review.openstack.org/466769 | 16:02 |
| *** david-lyle has joined #openstack-ansible | 16:03 | |
| *** lucasxu has quit IRC | 16:04 | |
| *** udesale has joined #openstack-ansible | 16:05 | |
| openstackgerrit | Merged openstack/openstack-ansible-os_barbican master: Add venv_tag local fact https://review.openstack.org/466771 | 16:05 |
| logan- | odyssey4me: https://review.openstack.org/#/c/466791/ in case you didnt see it | 16:06 |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible-os_rally master: Add venv_tag local fact https://review.openstack.org/466791 | 16:08 |
| odyssey4me | ah, thanks logan- - missed it | 16:08 |
| odyssey4me | mhayden is clearly a terrible reviewer, or trusts me way too much | 16:08 |
| mhayden | odyssey4me: they followed the pattern! | 16:08 |
| odyssey4me | haha, perhaps just a little too closely ;) | 16:08 |
| mhayden | haha, yeah, i missed one of those | 16:09 |
| *** chas has quit IRC | 16:09 | |
| *** jvidal has joined #openstack-ansible | 16:09 | |
| openstackgerrit | Merged openstack/openstack-ansible-os_neutron master: Add venv_tag local fact https://review.openstack.org/466757 | 16:10 |
| openstackgerrit | Merged openstack/openstack-ansible-os_cinder master: Add venv_tag local fact https://review.openstack.org/466750 | 16:10 |
| openstackgerrit | Merged openstack/openstack-ansible-ops master: Allow graceful exiting with SIGINT and SIGTERM https://review.openstack.org/466780 | 16:12 |
| openstackgerrit | Merged openstack/openstack-ansible-os_tempest master: Add venv_tag local fact https://review.openstack.org/466761 | 16:13 |
| *** mikal has quit IRC | 16:14 | |
| *** mikal has joined #openstack-ansible | 16:16 | |
| *** lucasxu has joined #openstack-ansible | 16:18 | |
| *** pbandark has quit IRC | 16:19 | |
| openstackgerrit | Merged openstack/openstack-ansible-os_nova master: Fix upgrades to map existing nova instances to cell1 https://review.openstack.org/466717 | 16:20 |
| openstackgerrit | Merged openstack/openstack-ansible-os_aodh master: Add venv_tag local fact https://review.openstack.org/466770 | 16:22 |
| *** udesale has quit IRC | 16:22 | |
| *** fabg has quit IRC | 16:26 | |
| *** drifterza has joined #openstack-ansible | 16:29 | |
| *** armaan has quit IRC | 16:31 | |
| *** armaan has joined #openstack-ansible | 16:32 | |
| *** vnogin has quit IRC | 16:33 | |
| *** jvidal has quit IRC | 16:35 | |
| openstackgerrit | Merged openstack/openstack-ansible-os_nova master: Add venv_tag local fact https://review.openstack.org/466758 | 16:36 |
| openstackgerrit | Merged openstack/openstack-ansible-os_trove master: Add venv_tag local fact https://review.openstack.org/466793 | 16:37 |
| *** kjw3 has joined #openstack-ansible | 16:37 | |
| openstackgerrit | Merged openstack/openstack-ansible-os_rally master: Add venv_tag local fact https://review.openstack.org/466791 | 16:44 |
| openstackgerrit | Merged openstack/openstack-ansible-os_sahara master: Add venv_tag local fact https://review.openstack.org/466792 | 16:44 |
| *** markus_z has quit IRC | 16:45 | |
| *** lostRhino has left #openstack-ansible | 16:45 | |
| openstackgerrit | Merged openstack/openstack-ansible-os_gnocchi master: Add venv_tag local fact https://review.openstack.org/466783 | 16:49 |
| *** jvidal has joined #openstack-ansible | 16:49 | |
| *** weezS has quit IRC | 16:49 | |
| xdfil | Is memcached syncronized between the memcached containers? is it active/active ? | 16:55 |
| *** jvidal has quit IRC | 16:56 | |
| *** galstrom is now known as galstrom_zzz | 16:58 | |
| xdfil | nevermind I found the info I was looking for | 17:00 |
| openstackgerrit | Merged openstack/openstack-ansible-os_designate master: Add venv_tag local fact https://review.openstack.org/466781 | 17:03 |
| *** galstrom_zzz is now known as galstrom | 17:04 | |
| odyssey4me | xdfil memcache has no concept of a cluster | 17:12 |
| *** marst_ has joined #openstack-ansible | 17:13 | |
| *** piliman974 has quit IRC | 17:13 | |
| *** lucasxu has quit IRC | 17:14 | |
| openstackgerrit | Merged openstack/openstack-ansible-os_horizon master: Add venv_tag local fact https://review.openstack.org/466784 | 17:16 |
| *** admin0 has joined #openstack-ansible | 17:16 | |
| *** jmccrory_away is now known as jmccrory | 17:18 | |
| *** piliman974 has joined #openstack-ansible | 17:25 | |
| openstackgerrit | Merged openstack/openstack-ansible-os_ironic master: Add venv_tag local fact https://review.openstack.org/466786 | 17:32 |
| openstackgerrit | Merged openstack/openstack-ansible-os_octavia master: Add venv_tag local fact https://review.openstack.org/466790 | 17:32 |
| *** armaan has quit IRC | 17:32 | |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible-galera_server master: [WIP] Use local facts for MariaDB status https://review.openstack.org/466827 | 17:34 |
| odyssey4me | palendae ^ I'd appreciate your review of the python script there - it's an adaptation of the library that git-harry did some time ago | 17:35 |
| palendae | Ok | 17:36 |
| palendae | Wait | 17:37 |
| palendae | It's just deleted | 17:37 |
| palendae | Oh nm | 17:37 |
| palendae | I get it | 17:37 |
| palendae | Read too fast :) | 17:37 |
| openstackgerrit | Miguel Alex Cantu (alextricity25) proposed openstack/openstack-ansible master: DO NOT MERGE - testing integrated gate w/ceilo https://review.openstack.org/453788 | 17:38 |
| odyssey4me | Thanks palendae - I'll work on its placement and the replacement of all the existing complicated stuff that uses the existing module and various CLI things | 17:38 |
| odyssey4me | tomorrow | 17:39 |
| palendae | Sure thing | 17:39 |
| *** rpittau_ has joined #openstack-ansible | 17:42 | |
| *** gouthamr has joined #openstack-ansible | 17:44 | |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible-os_ironic stable/ocata: Add venv_tag local fact https://review.openstack.org/466831 | 17:45 |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible-os_horizon stable/ocata: Add venv_tag local fact https://review.openstack.org/466832 | 17:46 |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible-os_horizon stable/newton: Add venv_tag local fact https://review.openstack.org/466833 | 17:46 |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible-os_designate stable/ocata: Add venv_tag local fact https://review.openstack.org/466834 | 17:46 |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible-os_gnocchi stable/ocata: Add venv_tag local fact https://review.openstack.org/466835 | 17:46 |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible-os_sahara stable/ocata: Add venv_tag local fact https://review.openstack.org/466836 | 17:47 |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible-os_rally stable/ocata: Add venv_tag local fact https://review.openstack.org/466837 | 17:47 |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible-os_trove stable/ocata: Add venv_tag local fact https://review.openstack.org/466838 | 17:47 |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible-os_nova stable/ocata: Add venv_tag local fact https://review.openstack.org/466839 | 17:47 |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible-os_aodh stable/ocata: Add venv_tag local fact https://review.openstack.org/466840 | 17:47 |
| openstackgerrit | Merged openstack/openstack-ansible-os_magnum master: Add venv_tag local fact https://review.openstack.org/466788 | 17:48 |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible-os_tempest stable/ocata: Add venv_tag local fact https://review.openstack.org/466841 | 17:48 |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible-os_tempest stable/newton: Add venv_tag local fact https://review.openstack.org/466842 | 17:48 |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible-os_cinder stable/ocata: Add venv_tag local fact https://review.openstack.org/466843 | 17:48 |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible-os_neutron stable/ocata: Add venv_tag local fact https://review.openstack.org/466844 | 17:48 |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible-os_neutron stable/newton: Add venv_tag local fact https://review.openstack.org/466845 | 17:48 |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible-os_barbican stable/ocata: Add venv_tag local fact https://review.openstack.org/466846 | 17:49 |
| *** gouthamr has quit IRC | 17:49 | |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible-os_keystone stable/ocata: Add venv_tag local fact https://review.openstack.org/466847 | 17:49 |
| *** rpittau_ has quit IRC | 17:49 | |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible-os_heat stable/ocata: Add venv_tag local fact https://review.openstack.org/466848 | 17:49 |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible-os_glance stable/ocata: Add venv_tag local fact https://review.openstack.org/466849 | 17:49 |
| *** kjw3 has quit IRC | 17:49 | |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible-lxc_container_create stable/ocata: Add variant local fact https://review.openstack.org/466850 | 17:50 |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible-lxc_container_create stable/newton: Add variant local fact https://review.openstack.org/466851 | 17:50 |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible-os_magnum stable/ocata: Add venv_tag local fact https://review.openstack.org/466853 | 17:50 |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible-os_magnum stable/newton: Add venv_tag local fact https://review.openstack.org/466855 | 17:53 |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible-os_ironic stable/newton: Add venv_tag local fact https://review.openstack.org/466856 | 17:57 |
| *** poopcat has joined #openstack-ansible | 18:00 | |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible-os_gnocchi stable/newton: Add venv_tag local fact https://review.openstack.org/466857 | 18:00 |
| *** admin0 has quit IRC | 18:01 | |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible-os_sahara stable/newton: Add venv_tag local fact https://review.openstack.org/466858 | 18:02 |
| *** kjw3 has joined #openstack-ansible | 18:02 | |
| *** armaan has joined #openstack-ansible | 18:02 | |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible-os_rally stable/newton: Add venv_tag local fact https://review.openstack.org/466859 | 18:04 |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible-os_nova stable/newton: Add venv_tag local fact https://review.openstack.org/466860 | 18:05 |
| *** weezS has joined #openstack-ansible | 18:05 | |
| *** albertcard1 has joined #openstack-ansible | 18:05 | |
| *** poopcat has quit IRC | 18:06 | |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible-os_aodh stable/newton: Add venv_tag local fact https://review.openstack.org/466861 | 18:07 |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible-os_cinder stable/newton: Add venv_tag local fact https://review.openstack.org/466862 | 18:08 |
| cloudnull | ^ odyssey4me on the rampage ! | 18:09 |
| mpranjic | again :) | 18:09 |
| *** chas has joined #openstack-ansible | 18:09 | |
| cloudnull | always ! | 18:09 |
| cloudnull | :D | 18:09 |
| mpranjic | odyssey4me: no more coffee for you! :) | 18:10 |
| mpranjic | lemesee some of those ... | 18:10 |
| mhayden | scotch + coffee | 18:10 |
| mpranjic | :D | 18:10 |
| mhayden | so on the centos side, i'm wondering if we can get some speedups from re-enabling fastestmirror and setting a more aggressive timeout | 18:10 |
| mhayden | we've still got that damned tempest/libvirt bug | 18:11 |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible-os_keystone stable/newton: Add venv_tag local fact https://review.openstack.org/466863 | 18:11 |
| *** jwitko_ has joined #openstack-ansible | 18:11 | |
| palendae | mhayden: Well that sure sounds like a good setting to enable :D | 18:11 |
| mpranjic | mhayden: how much is centos slower than same setup on ubuntu? | 18:11 |
| *** rcarrillocruz has joined #openstack-ansible | 18:11 | |
| mhayden | something about how tempest asks for a VM makes libvirtd upset | 18:11 |
| *** admin0 has joined #openstack-ansible | 18:11 | |
| mhayden | mpranjic: well there are still some issues with python 2.7.5 on centos w/ansible | 18:11 |
| mhayden | for some reason, it gets quicker if you install a newer version of python | 18:11 |
| mhayden | and gets close to ubuntu's usual speed | 18:11 |
| palendae | mhayden: And that's separate from the Xenial issue? | 18:12 |
| mhayden | haha, right | 18:12 |
| mhayden | :| | 18:12 |
| * mhayden collapses | 18:12 | |
| mhayden | i thought centos7 would be a more fun place, but now i have 2 problems | 18:12 |
| mpranjic | that is weird | 18:12 |
| mhayden | kinda like doing kubernetes the first time | 18:12 |
| *** lucasxu has joined #openstack-ansible | 18:12 | |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible-os_heat stable/newton: Add venv_tag local fact https://review.openstack.org/466864 | 18:12 |
| *** retreved_ has joined #openstack-ansible | 18:13 | |
| palendae | mhayden: Is it a specific version of Ansible? | 18:13 |
| mhayden | nah | 18:13 |
| mhayden | not specific to ansible | 18:13 |
| palendae | Ah | 18:13 |
| mhayden | 2.7.5 on CentOS 7 has some peculiar slowdowns w/ansible when it comes to running pip and yum | 18:13 |
| palendae | Welp, time to rewrite it all in Go | 18:13 |
| mhayden | most other things are fairly similar in timings | 18:14 |
| mhayden | haha | 18:14 |
| palendae | yum or dnf? | 18:14 |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible-os_glance stable/newton: Add venv_tag local fact https://review.openstack.org/466865 | 18:14 |
| palendae | did not finish | 18:14 |
| mhayden | yum | 18:14 |
| mhayden | dnf is going to be in RHEL 8, The Ocho | 18:14 |
| palendae | Ah, ok | 18:14 |
| *** retreved has quit IRC | 18:14 | |
| *** chas has quit IRC | 18:14 | |
| *** jwitko has quit IRC | 18:14 | |
| mpranjic | yeah, it is in fedora atm | 18:15 |
| palendae | Right | 18:15 |
| mhayden | can't wait til it comes out so i can call it "The Ocho" | 18:15 |
| mhayden | i'm weird | 18:15 |
| mhayden | very well aware | 18:15 |
| mpranjic | :) | 18:15 |
| palendae | I'd call it that too | 18:15 |
| odyssey4me | mhayden well, there are local mirrors or reverse proxies for RDO which I don't think we're using | 18:15 |
| odyssey4me | the centos and epel mirrors should already be set to use the infra mirrors | 18:15 |
| openstackgerrit | Merged openstack/openstack-ansible-security master: Move tasks to 'accounts' file https://review.openstack.org/466495 | 18:15 |
| mpranjic | toobad red hat is restricting rhel so much | 18:16 |
| odyssey4me | if you could spend some time mucking with https://review.openstack.org/462872 you might get some improved speed | 18:16 |
| odyssey4me | I haven't had the time to get back to it | 18:16 |
| mpranjic | they would benefit more to take ubuntu's approach: os and patches free but you can pay for support | 18:16 |
| palendae | mpranjic: Probably makes their life easier | 18:16 |
| mpranjic | atm I have no valid rhel subscription to try stuff | 18:16 |
| *** stuartgr has quit IRC | 18:16 | |
| odyssey4me | it's easy enough to get an eval subscription | 18:17 |
| mhayden | mpranjic: well, read a few articles and you'll find that canonical isn't always friendly about ubuntu | 18:17 |
| *** retreved_ has quit IRC | 18:17 | |
| mhayden | using it for free, that is | 18:17 |
| odyssey4me | mhayden I expect you should probably port https://review.openstack.org/466099 back to newton too? | 18:17 |
| mhayden | both approaches have their issues | 18:17 |
| * mhayden ganders | 18:17 | |
| mpranjic | mhayden: really? will check it out... | 18:17 |
| mhayden | odyssey4me: i can't verify that the bug exists there | 18:18 |
| mhayden | but if you want me to bring it back there, i can | 18:18 |
| odyssey4me | mhayden ah ok, I guess we'll do it reactively if someone asks for it | 18:18 |
| * mhayden reacts | 18:18 | |
| openstackgerrit | Merged openstack/openstack-ansible-os_ceilometer master: Add venv_tag local fact https://review.openstack.org/466779 | 18:19 |
| mpranjic | mhayden: we use debian for entire stack at work. have Ubuntu for Tower since it is not supportted on Debian | 18:19 |
| mpranjic | and developers mostly have Ubuntu on laptops | 18:19 |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible-os_ceilometer stable/ocata: Add venv_tag local fact https://review.openstack.org/466866 | 18:19 |
| mpranjic | but I didn't stumble upon canonical issues you mentioned.. will google a bit ... | 18:20 |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible-os_ceilometer stable/newton: Add venv_tag local fact https://review.openstack.org/466867 | 18:21 |
| *** admin0 has quit IRC | 18:21 | |
| openstackgerrit | Nolan Brubaker proposed openstack/openstack-ansible-ops master: Add syslog logging handler https://review.openstack.org/466868 | 18:22 |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible-os_nova stable/ocata: Fix upgrades to map existing nova instances to cell1 https://review.openstack.org/466869 | 18:23 |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible-os_designate stable/ocata: Perform an atomic policy file change https://review.openstack.org/466870 | 18:23 |
| openstackgerrit | Nolan Brubaker proposed openstack/openstack-ansible-ops master: Add syslog logging handler https://review.openstack.org/466868 | 18:24 |
| *** gouthamr has joined #openstack-ansible | 18:24 | |
| *** armaan has quit IRC | 18:30 | |
| *** armaan_ has joined #openstack-ansible | 18:30 | |
| *** chas has joined #openstack-ansible | 18:31 | |
| *** chas has quit IRC | 18:35 | |
| openstackgerrit | Merged openstack/openstack-ansible-os_aodh stable/ocata: Add venv_tag local fact https://review.openstack.org/466840 | 18:38 |
| *** vnogin has joined #openstack-ansible | 18:39 | |
| *** lucasxu has quit IRC | 18:41 | |
| openstackgerrit | Kevin Carter (cloudnull) proposed openstack/openstack-ansible master: Added an ALL is_metal scenario https://review.openstack.org/449372 | 18:41 |
| *** albertcard1 is now known as poopcat | 18:41 | |
| *** esberglu_ has quit IRC | 18:42 | |
| hwoarang | evrardjp: sorry for the delay | 18:42 |
| hwoarang | yeah good idea but i want to understand the issue first :/ | 18:42 |
| hwoarang | so my plan is to use the centos rpm as provided by upstream and see if the 3.6.X version fixes it | 18:42 |
| *** electrofelix has quit IRC | 18:43 | |
| *** lucasxu has joined #openstack-ansible | 18:43 | |
| *** esberglu has joined #openstack-ansible | 18:44 | |
| openstackgerrit | Kevin Carter (cloudnull) proposed openstack/openstack-ansible-os_nova master: [WIP] Testing using a common SystemD init role https://review.openstack.org/466179 | 18:45 |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible-os_ceilometer stable/ocata: Source template files from git or deploy host https://review.openstack.org/466874 | 18:48 |
| *** armaan_ has quit IRC | 18:51 | |
| *** armaan has joined #openstack-ansible | 18:51 | |
| *** manheim_ has joined #openstack-ansible | 18:51 | |
| *** gouthamr has quit IRC | 18:54 | |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible stable/ocata: Gnocchi: Update URL used to source default configs https://review.openstack.org/464985 | 18:55 |
| *** manheim_ has quit IRC | 18:55 | |
| openstackgerrit | Merged openstack/openstack-ansible-lxc_container_create stable/ocata: Add variant local fact https://review.openstack.org/466850 | 18:55 |
| openstackgerrit | Merged openstack/openstack-ansible-lxc_container_create stable/newton: Add variant local fact https://review.openstack.org/466851 | 18:56 |
| *** retreved has joined #openstack-ansible | 18:56 | |
| *** lostRhino has joined #openstack-ansible | 18:57 | |
| palendae | Does anyone know where rsyslog puts the 'user' facility logfiles on Ubuntu? | 18:57 |
| palendae | I'd think /var/log/user.log, but not seeing it | 18:57 |
| *** Mannheim has joined #openstack-ansible | 18:57 | |
| openstackgerrit | Merged openstack/openstack-ansible-os_rally stable/ocata: Add venv_tag local fact https://review.openstack.org/466837 | 18:57 |
| *** lucasxu has quit IRC | 18:58 | |
| *** admin0 has joined #openstack-ansible | 18:59 | |
| *** retreved_ has joined #openstack-ansible | 18:59 | |
| *** hybridpollo has joined #openstack-ansible | 19:00 | |
| openstackgerrit | Merged openstack/openstack-ansible-os_aodh stable/newton: Add venv_tag local fact https://review.openstack.org/466861 | 19:00 |
| *** retreved has quit IRC | 19:01 | |
| *** smatzek has quit IRC | 19:02 | |
| *** smatzek has joined #openstack-ansible | 19:02 | |
| *** lucasxu has joined #openstack-ansible | 19:03 | |
| openstackgerrit | Merged openstack/openstack-ansible-os_rally stable/newton: Add venv_tag local fact https://review.openstack.org/466859 | 19:03 |
| *** hybridpolio has joined #openstack-ansible | 19:04 | |
| *** hybridpollo has quit IRC | 19:04 | |
| openstackgerrit | Merged openstack/openstack-ansible-os_barbican stable/ocata: Add venv_tag local fact https://review.openstack.org/466846 | 19:05 |
| openstackgerrit | Merged openstack/openstack-ansible-os_glance stable/ocata: Add venv_tag local fact https://review.openstack.org/466849 | 19:06 |
| openstackgerrit | Merged openstack/openstack-ansible-os_heat stable/ocata: Add venv_tag local fact https://review.openstack.org/466848 | 19:06 |
| openstackgerrit | Merged openstack/openstack-ansible-os_heat stable/newton: Add venv_tag local fact https://review.openstack.org/466864 | 19:06 |
| openstackgerrit | Merged openstack/openstack-ansible-os_gnocchi stable/ocata: Add venv_tag local fact https://review.openstack.org/466835 | 19:08 |
| openstackgerrit | Merged openstack/openstack-ansible-os_gnocchi stable/newton: Add venv_tag local fact https://review.openstack.org/466857 | 19:08 |
| openstackgerrit | Merged openstack/openstack-ansible-os_horizon stable/newton: Add venv_tag local fact https://review.openstack.org/466833 | 19:08 |
| *** hybridpolio is now known as hybridpollo | 19:08 | |
| openstackgerrit | Merged openstack/openstack-ansible-os_trove stable/ocata: Add venv_tag local fact https://review.openstack.org/466838 | 19:09 |
| openstackgerrit | Merged openstack/openstack-ansible-os_glance stable/newton: Add venv_tag local fact https://review.openstack.org/466865 | 19:10 |
| *** chas has joined #openstack-ansible | 19:12 | |
| lostRhino | can anyone help out - i know its late in the day - but we are getting an error during the “galera_client : Install pip packages” portion of our install of OSAD… its a fairly long error that points to the openstack/upper-constraints (see here: http://paste.openstack.org/show/610269/ ) we know the fix - is to run “ansible -a "systemctl restart keepalived" compute-infra_hosts” | 19:12 |
| lostRhino | is there a reason why this is happening or a way to avoid it - there seems to be a issue with keepalived not starting/restarting, we’re not sure why this started - we were having successful builds before and now it stops… | 19:12 |
| lostRhino | any help would be appreciated | 19:12 |
| mhayden | palendae: 14.04 or 16.04? | 19:12 |
| cloudnull | lostRhino: normally that's due to an problem w/ the load balancer. | 19:13 |
| cloudnull | using haproxy ? | 19:13 |
| palendae | mhayden: 14.04 on the utility containers. Looks like we leave the user and local7 facilities unconfigured | 19:13 |
| lostRhino | yes currently | 19:13 |
| palendae | It's not vital to what I'm doing right now, so I'm gonna punt on that | 19:13 |
| openstackgerrit | Merged openstack/openstack-ansible-os_sahara stable/ocata: Add venv_tag local fact https://review.openstack.org/466836 | 19:13 |
| cloudnull | lostRhino: I'd check to see if the hap is running and that the backends are up | 19:14 |
| openstackgerrit | Nolan Brubaker proposed openstack/openstack-ansible-ops master: Log failure percentage before exiting https://review.openstack.org/466878 | 19:14 |
| lostRhino | well as along as we restart keepalived it will continue - but we have to get on our provisioner and then run openstack-ansible setup-infrastructure | 19:14 |
| lostRhino | to continue with the install - which is a pain in the @** | 19:14 |
| openstackgerrit | Merged openstack/openstack-ansible-os_keystone stable/ocata: Add venv_tag local fact https://review.openstack.org/466847 | 19:16 |
| openstackgerrit | Merged openstack/openstack-ansible-os_keystone stable/newton: Add venv_tag local fact https://review.openstack.org/466863 | 19:17 |
| *** chas has quit IRC | 19:17 | |
| openstackgerrit | Merged openstack/openstack-ansible-os_ceilometer stable/newton: Add venv_tag local fact https://review.openstack.org/466867 | 19:18 |
| openstackgerrit | Nolan Brubaker proposed openstack/openstack-ansible-ops master: Log failure percentage before exiting https://review.openstack.org/466878 | 19:18 |
| openstackgerrit | Miguel Alex Cantu (alextricity25) proposed openstack/openstack-ansible master: DO NOT MERGE - testing integrated gate w/ceilo https://review.openstack.org/453788 | 19:18 |
| openstackgerrit | Merged openstack/openstack-ansible-os_horizon stable/ocata: Add venv_tag local fact https://review.openstack.org/466832 | 19:19 |
| openstackgerrit | Merged openstack/openstack-ansible-os_magnum stable/ocata: Add venv_tag local fact https://review.openstack.org/466853 | 19:20 |
| openstackgerrit | Merged openstack/openstack-ansible-os_neutron stable/newton: Add venv_tag local fact https://review.openstack.org/466845 | 19:20 |
| lostRhino | cloudnull: any other things I should look for? Thanks for your help…. | 19:22 |
| openstackgerrit | Merged openstack/openstack-ansible-os_cinder stable/newton: Add venv_tag local fact https://review.openstack.org/466862 | 19:22 |
| *** lostRhino has left #openstack-ansible | 19:23 | |
| *** lostRhino has joined #openstack-ansible | 19:23 | |
| cloudnull | lostRhino: so you're seeing the vip flip back and forth ? | 19:24 |
| cloudnull | which is causing the issues? | 19:24 |
| *** hogwash has joined #openstack-ansible | 19:24 | |
| cloudnull | and restarting keepalived is fixing it ? | 19:24 |
| lostRhino | how would I check that ? or know the vip is fliping back and forth? | 19:24 |
| lostRhino | yes restarting keepalive is fixing it | 19:24 |
| cloudnull | stop keepalived and hap on the second haproxy node. | 19:25 |
| lostRhino | i just got done doing the setup-infrasturcte | 19:25 |
| cloudnull | get it stable on the one and continue the installation | 19:25 |
| lostRhino | done and done sir | 19:25 |
| cloudnull | you can always add to it | 19:25 |
| cloudnull | without having to redploy | 19:25 |
| openstackgerrit | Merged openstack/openstack-ansible-os_ironic stable/ocata: Add venv_tag local fact https://review.openstack.org/466831 | 19:26 |
| cloudnull | once the cloud is online and stable you can circle back and get multi-node haporxy running correctly | 19:26 |
| cloudnull | no sense in fighting that battle upfront if you don't have to. | 19:26 |
| cloudnull | that said, evrardjp should be around in the am | 19:26 |
| cloudnull | and is our resident keepalived export | 19:26 |
| cloudnull | **expert | 19:26 |
| lostRhino | yeah I just dont get why it worked well a couple weeks again and now its stopping mid build | 19:27 |
| *** drifterza has quit IRC | 19:27 | |
| openstackgerrit | Merged openstack/openstack-ansible-os_tempest stable/newton: Add venv_tag local fact https://review.openstack.org/466842 | 19:28 |
| openstackgerrit | Merged openstack/openstack-ansible-os_tempest stable/ocata: Add venv_tag local fact https://review.openstack.org/466841 | 19:28 |
| lostRhino | on this one service … | 19:28 |
| cloudnull | lostRhino: really hard to say, the logs should show the disconnects. | 19:28 |
| lostRhino | which logs - | 19:29 |
| lostRhino | :eek: | 19:29 |
| cloudnull | I think /var/log/keepalived-notifications.log ? | 19:30 |
| openstackgerrit | Merged openstack/openstack-ansible-os_cinder stable/ocata: Add venv_tag local fact https://review.openstack.org/466843 | 19:31 |
| *** whiteveil has joined #openstack-ansible | 19:33 | |
| lostRhino | 2017-05-22 17:12:16 Trying to restart haproxy to get out of faulty state | 19:33 |
| lostRhino | 2017-05-22 17:12:55 Unknown state | 19:33 |
| lostRhino | 2017-05-22 18:47:08 Unknown state | 19:33 |
| *** chas has joined #openstack-ansible | 19:33 | |
| lostRhino | anyways - I’m outta here cloudnull - I’ll be back tomorrow to follow up - thanks for the pointers/tips/help | 19:34 |
| *** lostRhino has left #openstack-ansible | 19:34 | |
| openstackgerrit | Merged openstack/openstack-ansible-os_designate stable/ocata: Add venv_tag local fact https://review.openstack.org/466834 | 19:35 |
| openstackgerrit | Merged openstack/openstack-ansible-os_designate stable/ocata: Use internalURL by default for Neutron https://review.openstack.org/465064 | 19:35 |
| openstackgerrit | Merged openstack/openstack-ansible-os_designate stable/ocata: Optimise apt cache update task https://review.openstack.org/462850 | 19:35 |
| openstackgerrit | Merged openstack/openstack-ansible-os_designate stable/ocata: Ensure the components are isolated from the system https://review.openstack.org/460155 | 19:35 |
| openstackgerrit | Merged openstack/openstack-ansible-os_designate stable/ocata: Reduce init restart/kill times https://review.openstack.org/460717 | 19:35 |
| cloudnull | anytime . | 19:35 |
| openstackgerrit | Nolan Brubaker proposed openstack/openstack-ansible-ops master: Log failure percentage before exiting https://review.openstack.org/466878 | 19:36 |
| cloudnull | let me know how it goes lostRhino | 19:36 |
| openstackgerrit | Merged openstack/openstack-ansible-os_nova stable/ocata: Add venv_tag local fact https://review.openstack.org/466839 | 19:38 |
| openstackgerrit | Merged openstack/openstack-ansible-os_nova stable/newton: Add venv_tag local fact https://review.openstack.org/466860 | 19:38 |
| *** chas has quit IRC | 19:38 | |
| *** whiteveil_ has joined #openstack-ansible | 19:43 | |
| *** admin0 has quit IRC | 19:43 | |
| openstackgerrit | Nolan Brubaker proposed openstack/openstack-ansible-ops master: Log failure percentage before exiting https://review.openstack.org/466878 | 19:44 |
| openstackgerrit | Merged openstack/openstack-ansible-os_magnum stable/newton: Add venv_tag local fact https://review.openstack.org/466855 | 19:44 |
| *** admin0 has joined #openstack-ansible | 19:45 | |
| *** whiteveil has quit IRC | 19:46 | |
| *** whiteveil_ is now known as whiteveil | 19:46 | |
| openstackgerrit | Merged openstack/openstack-ansible-os_ironic stable/newton: Add venv_tag local fact https://review.openstack.org/466856 | 19:47 |
| *** chhavi has quit IRC | 19:48 | |
| openstackgerrit | Merged openstack/openstack-ansible-os_neutron stable/ocata: Add venv_tag local fact https://review.openstack.org/466844 | 19:49 |
| openstackgerrit | Matthew Thode proposed openstack/openstack-ansible-ops master: make the default base network match variables.sh https://review.openstack.org/466883 | 19:53 |
| *** chas has joined #openstack-ansible | 19:54 | |
| *** smatzek has quit IRC | 19:55 | |
| *** armaan_ has joined #openstack-ansible | 19:56 | |
| *** smatzek has joined #openstack-ansible | 19:58 | |
| *** chas has quit IRC | 19:59 | |
| *** armaan has quit IRC | 19:59 | |
| openstackgerrit | Merged openstack/openstack-ansible-os_ceilometer stable/ocata: Add venv_tag local fact https://review.openstack.org/466866 | 20:03 |
| *** mpranjic_ has joined #openstack-ansible | 20:04 | |
| *** mpranjic_ has quit IRC | 20:04 | |
| *** weezS has quit IRC | 20:05 | |
| *** vnogin has quit IRC | 20:08 | |
| *** admin0 has quit IRC | 20:11 | |
| *** chas has joined #openstack-ansible | 20:15 | |
| *** chas has quit IRC | 20:19 | |
| *** admin0 has joined #openstack-ansible | 20:21 | |
| *** weezS has joined #openstack-ansible | 20:22 | |
| *** lbragstad has quit IRC | 20:26 | |
| *** karimb has quit IRC | 20:33 | |
| *** smatzek has quit IRC | 20:34 | |
| *** weezS has quit IRC | 20:35 | |
| *** chas has joined #openstack-ansible | 20:36 | |
| openstackgerrit | Nolan Brubaker proposed openstack/openstack-ansible-ops master: Catch all errors and provide failure summary https://review.openstack.org/466878 | 20:38 |
| *** hogwash has quit IRC | 20:38 | |
| *** weezS has joined #openstack-ansible | 20:39 | |
| *** chas has quit IRC | 20:41 | |
| *** hogwash has joined #openstack-ansible | 20:49 | |
| *** askb has joined #openstack-ansible | 20:53 | |
| *** hogwash has quit IRC | 20:54 | |
| *** gouthamr has joined #openstack-ansible | 20:54 | |
| *** chas has joined #openstack-ansible | 20:57 | |
| *** thorst_afk has quit IRC | 20:58 | |
| *** thorst_afk has joined #openstack-ansible | 20:59 | |
| *** NeilHanlon_ is now known as NeilHanlon | 20:59 | |
| *** NeilHanlon has quit IRC | 20:59 | |
| *** NeilHanlon has joined #openstack-ansible | 20:59 | |
| *** retreved_ has quit IRC | 21:00 | |
| jgriffiths | Is there are way to sepcify "flat_networks = provider | 21:01 |
| jgriffiths | " and "network_vlan_ranges = provider" in openstack ansible? | 21:01 |
| *** chas has quit IRC | 21:02 | |
| *** thorst_afk has quit IRC | 21:03 | |
| *** whiteveil has quit IRC | 21:05 | |
| *** woodard has quit IRC | 21:09 | |
| openstackgerrit | OpenStack Proposal Bot proposed openstack/openstack-ansible stable/newton: Updated from global requirements https://review.openstack.org/466909 | 21:09 |
| *** aludwar has quit IRC | 21:10 | |
| *** hogwash has joined #openstack-ansible | 21:11 | |
| *** gouthamr has quit IRC | 21:13 | |
| *** hogwash has quit IRC | 21:16 | |
| *** gouthamr has joined #openstack-ansible | 21:16 | |
| *** chas has joined #openstack-ansible | 21:18 | |
| openstackgerrit | Jimmy McCrory proposed openstack/openstack-ansible master: [WIP] Update Ansible to 2.3.1.0-0.1.rc1 https://review.openstack.org/466912 | 21:19 |
| *** aludwar has joined #openstack-ansible | 21:21 | |
| *** chas has quit IRC | 21:22 | |
| *** esberglu has quit IRC | 21:24 | |
| openstackgerrit | Merged openstack/openstack-ansible-os_swift master: Add venv_tag local fact https://review.openstack.org/466759 | 21:30 |
| *** jheroux has quit IRC | 21:30 | |
| *** galstrom is now known as galstrom_zzz | 21:31 | |
| *** thorst_afk has joined #openstack-ansible | 21:36 | |
| *** chas has joined #openstack-ansible | 21:39 | |
| *** cpuga_ has quit IRC | 21:39 | |
| *** esberglu has joined #openstack-ansible | 21:39 | |
| *** thorst_afk has quit IRC | 21:40 | |
| *** gouthamr has quit IRC | 21:41 | |
| *** jwitko has joined #openstack-ansible | 21:41 | |
| *** armaan_ has quit IRC | 21:42 | |
| *** chas has quit IRC | 21:43 | |
| *** jwitko_ has quit IRC | 21:44 | |
| *** cjloade__ has joined #openstack-ansible | 21:50 | |
| *** cjloader_ has quit IRC | 21:54 | |
| *** admin0 has quit IRC | 21:54 | |
| *** cjloade__ has quit IRC | 21:55 | |
| *** lucasxu has quit IRC | 21:58 | |
| *** chas has joined #openstack-ansible | 21:59 | |
| *** chas has quit IRC | 22:04 | |
| *** weezS has quit IRC | 22:10 | |
| *** Mannheim has quit IRC | 22:12 | |
| *** kylek3h has quit IRC | 22:12 | |
| *** manheim_ has joined #openstack-ansible | 22:20 | |
| *** chas has joined #openstack-ansible | 22:20 | |
| *** manheim__ has joined #openstack-ansible | 22:23 | |
| *** chas has quit IRC | 22:25 | |
| *** klamath has quit IRC | 22:25 | |
| *** manheim_ has quit IRC | 22:26 | |
| jgriffiths | Is there a way to delete a single neutron agent container so I can have openstack-ansible recreate it | 22:40 |
| *** manheim__ has quit IRC | 22:40 | |
| cloudnull | jgriffiths: yes. | 22:41 |
| cloudnull | you can target the container | 22:41 |
| *** chas has joined #openstack-ansible | 22:41 | |
| cloudnull | openstack-ansible lxc-container-destroy.yml lxc-container-create.yml --limit $CONTAINER_NAME | 22:41 |
| jgriffiths | I'm not sure I know what you mean. Sorry | 22:41 |
| cloudnull | openstack-ansible os-neutron-install.yml | 22:42 |
| cloudnull | ^ jgriffiths with those two commands you should be able to recreate the specific contianer and redeploy neutron things inside it | 22:42 |
| jgriffiths | Yeah. I do that. I just wanted to delete a container. Can I do that without messing everything up | 22:42 |
| cloudnull | you could even limit that last command with something like . | 22:42 |
| cloudnull | openstack-ansible os-neutron-install.yml --limit all_containers | 22:42 |
| cloudnull | so that it omits the compute nodes. | 22:43 |
| cloudnull | as for "flat_networks = provider", what are you looking for ? | 22:43 |
| jgriffiths | Right now. Everything just took a crap. I found your github about provider networks earlier today | 22:44 |
| jgriffiths | lxc-container-destroy blows away the entire environment, no? | 22:44 |
| jgriffiths | That would be bad. | 22:45 |
| *** chas has quit IRC | 22:46 | |
| jgriffiths | For the networks, we have a bunch of very specific VLANS we need to connect to individual tenants. I'm not a networking person, but I'm beind told they are provider networks. When I looked at an existing envirnment, that's the setting they had in the ml2 plugin configuration | 22:47 |
| jgriffiths | Right now, I have one container that's not responding and I can't get it back up. | 22:48 |
| jgriffiths | I want to blow it away and re-run setup-hosts to recreate it. Does "openstack-ansible lxc-container-destroy.yml --limit "node003_neutron_agents_container-46918615" wirk? | 22:51 |
| jgriffiths | Work - omg | 22:51 |
| jgriffiths | I'm afraid to run that | 22:54 |
| cloudnull | jgriffiths: sorry saw a Squirrel | 22:58 |
| cloudnull | if you run that playbook with the limit it wont kill the entire env. | 22:58 |
| jgriffiths | haha. | 22:58 |
| cloudnull | the limit is important in that context. | 22:58 |
| logan- | run it with --list-hosts first | 22:59 |
| logan- | to verify it is going to do what you think it will do | 22:59 |
| cloudnull | ^ thats a good way to validate it | 22:59 |
| openstackgerrit | Merged openstack/openstack-ansible stable/newton: Updated from global requirements https://review.openstack.org/466909 | 22:59 |
| jgriffiths | Thank you | 22:59 |
| *** thorst_afk has joined #openstack-ansible | 23:00 | |
| cloudnull | in your example, I'd run ``openstack-ansible lxc-container-destroy.yml lxc-container-create.yml --limit "node003_neutron_agents_container-46918615" | 23:00 |
| jgriffiths | I just ran the destroy. No I'm trying to run the create and it's looking for the ond container | 23:01 |
| jgriffiths | Thanks for the help btw | 23:01 |
| logan- | rm /etc/openstack_deploy/ansible_facts/node003_neutron_agents_container-46918615 after destroy | 23:01 |
| cloudnull | np | 23:01 |
| cloudnull | ^ good point. | 23:02 |
| *** chas has joined #openstack-ansible | 23:02 | |
| cloudnull | jgriffiths: which release? | 23:02 |
| jgriffiths | Newton | 23:02 |
| cloudnull | cool. | 23:02 |
| cloudnull | that should do it :) | 23:03 |
| jgriffiths | lawl. Now it's not defined. haha | 23:03 |
| jgriffiths | fatal: [node003_neutron_agents_container-46918615]: FAILED! => {"changed": false, "failed": true, "module_stderr": "Error: container node003_neutron_agents_container-46918615 is not defined\n", "module_stdout": "", "msg": "MODULE FAILURE"} | 23:03 |
| *** thorst_afk has quit IRC | 23:04 | |
| *** thorst_afk has joined #openstack-ansible | 23:05 | |
| logan- | which playbook? | 23:05 |
| logan- | after destroy, delete the cached fact and then run "openstack-ansible lxc-containers-create.yml --limit node003_neutron_agents_container-46918615; openstack-ansible os-neutron-install.yml --limit all_containers;" | 23:05 |
| jgriffiths | os-neutron-install.yml | 23:06 |
| jgriffiths | ls | 23:06 |
| jgriffiths | mt | 23:06 |
| *** chas has quit IRC | 23:06 | |
| *** thorst_afk has quit IRC | 23:09 | |
| jgriffiths | God. I really hate to waste all this help on my inability to do Ansible. I deleted the /etc/openstack_deploy/ansible_facts/node003_neutron_agents_container-46918615' file but os-neutron-install tells me the it's not defined. | 23:11 |
| logan- | did you run the lxc-containers-create.yml first? | 23:12 |
| *** cpuga has joined #openstack-ansible | 23:13 | |
| jgriffiths | haha. Of course I didn't. That's it~ | 23:13 |
| jgriffiths | Thanks! | 23:13 |
| jgriffiths | fg | 23:14 |
| jgriffiths | I've been trying to get VLANs to work in my Ansible config. Everytime I tried to create a instance, it couldn't attach a port. | 23:16 |
| cloudnull | are you seeing a vif type binding failure ? | 23:16 |
| *** jamesdenton has joined #openstack-ansible | 23:16 | |
| *** weezS has joined #openstack-ansible | 23:17 | |
| cloudnull | and do you have https://github.com/openstack/openstack-ansible/blob/master/etc/openstack_deploy/openstack_user_config.yml.aio#L46-L54 ? | 23:17 |
| cloudnull | in your openstack_user_config? | 23:17 |
| jgriffiths | Yes. Both of those things. The flat and the vlan. I was thinking it was the missing veth settings in the br-vlan.cfg file though. I found the settings on something you wrote cloudnull. | 23:18 |
| jgriffiths | I was adding them to my br-vlan, when I lost half my nodes | 23:18 |
| cloudnull | do your compute hosts have eth12 ? | 23:19 |
| jgriffiths | Yes | 23:19 |
| cloudnull | ok. | 23:19 |
| *** jamesden_ has joined #openstack-ansible | 23:20 | |
| jgriffiths | One last Ansible question if you don't mind. | 23:20 |
| cloudnull | a lot of the time we'll see a vif type failure when something can not bind | 23:21 |
| *** jamesdenton has quit IRC | 23:21 | |
| cloudnull | check to make sure your neutron linux-bridge-agents are running | 23:21 |
| cloudnull | maybe stop one on a given compute node, and start it in the foreground | 23:21 |
| cloudnull | so that you can see the direct log output | 23:21 |
| jgriffiths | They were. That's the problem now though. I'm still trying to recover from the broekn node and the facts are trying to use configuration settings from some time ago | 23:22 |
| cloudnull | if it hangs or otherwise does very little (or stacktraces) It's likely that theres a misconfiguration somewhere | 23:22 |
| cloudnull | you are free to destroy all of the facts if that helps . | 23:22 |
| jgriffiths | I wish I had spoken to you before everything blew up | 23:22 |
| *** piliman974 has quit IRC | 23:23 | |
| jgriffiths | That's safe? | 23:23 |
| *** chas has joined #openstack-ansible | 23:23 | |
| cloudnull | rm /etc/openstack_deploy/ansible_facts/* | 23:23 |
| cloudnull | if they don't exist, they'll be recreated | 23:23 |
| *** piliman974 has joined #openstack-ansible | 23:24 | |
| cloudnull | sorry, I've been fairly absent these last couple of weeks. | 23:25 |
| cloudnull | working on too many things at once | 23:25 |
| * cloudnull knows he | 23:25 | |
| jgriffiths | God. No. Thanks! Last week was hell trying to get this thing running. | 23:25 |
| * cloudnull knows he's preaching to the Choir :) | 23:26 | |
| jgriffiths | I'm doing a full lxc-create now. | 23:27 |
| *** chas has quit IRC | 23:27 | |
| cloudnull | while that will iterate through the containers, it will only create things that are missing . | 23:28 |
| *** weezS has quit IRC | 23:28 | |
| cloudnull | it should be fairly quick | 23:28 |
| openstackgerrit | Merged openstack/openstack-ansible-os_designate stable/ocata: Perform an atomic policy file change https://review.openstack.org/466870 | 23:29 |
| *** jamesden_ has quit IRC | 23:30 | |
| jgriffiths | Oh. I have eth12 because of "pre-up ip link add br-vlan-veth type veth peer name eth12 || true" this, don't I? | 23:30 |
| jgriffiths | That wasn't there when I was having the trouble. I added it right before I lost half my nodes. | 23:31 |
| cloudnull | did that half have that stanza? | 23:31 |
| *** pmannidi has quit IRC | 23:31 | |
| jgriffiths | No. I added it to the br-vlan interface. I lost br-mgmt | 23:32 |
| cloudnull | https://github.com/openstack/openstack-ansible/blob/master/etc/network/interfaces.d/aio_interfaces.cfg#L52-L59 | 23:32 |
| cloudnull | so those lines are needed to make the veth bits go | 23:32 |
| jgriffiths | Yeah. That's where I got it from (that and a github repo by somenody named cloudnull) | 23:33 |
| cloudnull | and make sure you dont have a duplicate of this line https://github.com/openstack/openstack-ansible/blob/master/etc/network/interfaces.d/aio_interfaces.cfg#L59 | 23:33 |
| * cloudnull hides | 23:33 | |
| jgriffiths | haha. | 23:33 |
| cloudnull | you can however have multiple ports assigned to bridge_ports, something like """bridge_ports eth2 br-vlan-veth""" | 23:34 |
| cloudnull | did you figure out why mgmt went down ? | 23:34 |
| jgriffiths | I lost two nodes. No idea why yet | 23:35 |
| cloudnull | ok. | 23:35 |
| cloudnull | are the other interfaces up ? | 23:35 |
| jgriffiths | Half. Yes, it's small. It's supposed to get really big | 23:35 |
| jgriffiths | Yeah, actually | 23:35 |
| jgriffiths | They're all up now | 23:35 |
| cloudnull | even mgmt ? | 23:35 |
| jgriffiths | ifdown ifup fixed them all | 23:35 |
| cloudnull | cosmic rays? | 23:35 |
| cloudnull | :) | 23:35 |
| jgriffiths | I rebooted node003 though. That caused a whole host of other problems. Agent still refused to start on any nodes. | 23:36 |
| cloudnull | so ifup/down on infra nodes will break the containers connections to the veth, so if you do that there make sure to run the lxc-veth-check command after | 23:37 |
| jgriffiths | Agent is throwing "MessagingTimeout" on every node, but all the other containers are running just fine | 23:37 |
| jgriffiths | Oh | 23:37 |
| cloudnull | that sounds like rabbitmq has died | 23:37 |
| jgriffiths | Okay | 23:37 |
| jgriffiths | Yeah, you would think so, but everything else is working | 23:38 |
| cloudnull | or maybe not able to communicate due to link breakage | 23:38 |
| cloudnull | ansible -m shell -a 'lxc-veth-check' hosts | 23:38 |
| cloudnull | just to do a blanket check across the environment. | 23:38 |
| cloudnull | if the link is busted the command will attempt to fix it | 23:39 |
| jgriffiths | container node002_neutron_agents_container-7abef153 had a broken veth 7abef153_eth12 not being plugged into br-vlan: this issue is now res | 23:39 |
| cloudnull | if it can't fix it, it will tell you | 23:39 |
| cloudnull | boom! | 23:39 |
| jgriffiths | Oh. I just had a loose wire | 23:41 |
| jgriffiths | Running setup-hosts to see if the agent will come backup. Probably should have limited it | 23:42 |
| jgriffiths | Is lxc-veth-check yours? | 23:44 |
| *** chas has joined #openstack-ansible | 23:44 | |
| jgriffiths | In https://github.com/openstack/openstack-ansible/blob/master/etc/network/interfaces.d/aio_interfaces.cfg#L52-L59 , why is there an IP on the interface? | 23:45 |
| *** chas has quit IRC | 23:48 | |
| cloudnull | yea that tool was written by me but, its part of the lxc-hosts role. | 23:51 |
| cloudnull | the IP is there on that interface for the gate. | 23:51 |
| cloudnull | we create vms using the flat netowrk | 23:51 |
| cloudnull | and ensure we can access and install things within them | 23:51 |
| cloudnull | you wont need that in 90% of prod deployments. | 23:52 |
| jgriffiths | For some reason, a bad setting I made a long time ago is still popping up in openstack_inventory.json and it's causing setup-hosts to fail | 23:52 |
| cloudnull | what settin g? | 23:54 |
| jgriffiths | /etc/openstack_deploy/openstack_inventory.json: "POC_Web_Net-interface_address": { | 23:55 |
| jgriffiths | I was playing with the interfaces this morning, and the change doesn't want to go away | 23:55 |
| cloudnull | is that in the openstack _user_config.yml or any of the files within /etc/openstack_deploy/conf.d | 23:55 |
| cloudnull | ? | 23:55 |
| *** pmannidi has joined #openstack-ansible | 23:55 | |
| jgriffiths | Nowhere. I grepped the entire /etc and /opt directories and only found it in the inventory file. | 23:56 |
| cloudnull | ok, sadly, you have to delete that entry by hand or revert to an earlier inventory version that doesn't have it | 23:57 |
| jgriffiths | Okay. I can do that | 23:57 |
| jgriffiths | I think. | 23:57 |
| jgriffiths | It's pissed at me again though | 23:57 |
| cloudnull | there's the openstack_inventory.tar which contains timestamp'd version of inventory | 23:57 |
Generated by irclog2html.py 2.14.0 by Marius Gedminas - find it at mg.pov.lt!